Click on one of the following categories:

Microsoft's newer operating systems, Windows Vista and Windows 7, 8, 10 and 11 have increased security, which basically translates into more hassle getting programs installed and running. Once a program is installed, you must run it as the administrator or have administrator privileges. Most all of our current programs will automatically force you to run the program as administrator, so it may be wise to update to the latest version. When you double click on the desktop icon, it will give you the option to "Cancel" running the program or "Allow" the program to run. Therefore, other than choosing the "Allow" option when starting the program, there is nothing special you have to do when running our programs in Windows Vista or Windows 7, 8, 10 and 11.

On rare occasions:

Occasionally you may encounter an error 339 or other "licensing errors" when starting a program in Windows Vista or Windows 7, 8, 10 and 11. This typically happens if the program was not run with administrator privileges.

To fix this situation:

First, try restarting your computer a couple of times to see if the error is fixed.
Second, if the problem is still there and it says something like "file not found" (not error 339), try installing again. This time do a "Complete (typical) Installation" even if the installer recommends just a "Refresh Installation". Important: To eliminate other possible installation problems on Win 10, restart your computer and then go to the installer. DO NOT run the program you are trying to install before installing it again.
Third (and this typically only applies to older programs which are not the latest version, like EA32 when there is EA34 available), right click on the desktop icon and select Properties. Then click on the Compatibility tab at the top of this new screen, then check the box in the lower left for "Run as Administrator". Then Click on OK. Now when you double click on the desktop icon, it will should you the option to "Allow" the program to run.

If after doing this, you still have problems, shut down your computer and restart it. Now try to run the program again and see if the problems have been fixed. In most all cases, these steps will get the program running on Vista and Windows 7, 8, 10 and 11.

If you are still have a compatibility problem with a current Performance Trends' product and Vista or Windows 7, 8, 10 and 11, please email us at: feedback@performancetrends.com We'll get on it as soon as we can, as all our products will have to be compatible with Vista and Windows 7, 8, 10 and 11.

What if I don't know some spec the program is asking me?

Most all programs have a library of example specs loaded for you to start with. For example, in the Engine Analyzer there are over 70 complete engines pre-loaded by Performance Trends. Try to pick an engine close to what you will be building. Then if you don't know some spec, just leave the value which is already loaded for the example engine you started with.

Many programs have some way to "Get Examples" of components. For example, in the Engine Analyzer, there are over 300 example Heads pre-loaded. Some are specific like "Stock Chev Big Block Iron Rect Port", and some are general like "Typical Ported 4 Valve for 3.8" Bore".

Some specs are shown with a "Clc" button to their right which stands for "Calculate". An example is Compression Ratio. You can enter a compression ratio directly by just typing it in if you know it. Or you can calculate it from other specs like Chamber Volume, Gasket Thickness, etc.

In most all programs, when you click on a spec a brief description of the spec is displayed in the "Help" box on that screen. Many times this "Help" includes a page number in the manual if you want more explanation (for programs that come with manuals).

If you are building something completely different than any specs pre-loaded in the program, make a good estimate of the spec and calculate performance. Then make a change to the spec and recalculate performance and see the effect on performance. Lets say you don't know the Frontal Area for your 13 second, 66 Nova when running the Drag Racing Analyzer. You see from the example vehicles that a 69 Mustang is rated at 22 square feet and a 70 Corvette is rated at 19. You calculate performance for your 66 Nova using 22, and then 19 and see only a .018 second change in ET. This tells you that for your particular car, Frontal Area is not that important, and some number like 20 or 21 will work fine.

However, if you car was running in the 10s at 130 MPH or so, Frontal Area would be much more important. Then you may want to come up with a better way to estimate Frontal Area. Also see the "How do I use a simulation program?" question below.

Why is the program predicting more performance than I'm seeing on the dyno (or at the track)?

Simulation programs are predicting how things should work with all systems working properly, carbs metering properly, ignition systems firing perfectly every time, combustion chambers burning efficiently every firing, etc. They can predict a mis-match of parts, but can not predict a improper operation or failure of some component. Therefore, programs are usually predicting "best case" performance.

How do I use a simulation program?

We are always getting phone calls saying that these programs must be "magic" because the results exactly match the "real world" of peak HP on the dyno, or ET at the drag strip. The important thing is not that the program matches some dyno's measurements (because another dyno can easily produce 5-10% different results), but do the programs match the dyno's trends accurately.

For example, if you change header primary length 4 inches, does the program's predicted trends match the dyno's trends. Even if 2 different dynos give 10% different HP numbers, both should show the same trends and both can be used to improve engine performance. If the program can predict the trends accurately, then the program can also be used to improve engine performance. Fortunately, we get just as many phone calls saying the programs are predicting the trends accurately also.

That's why we have more inputs in our programs than most others. Its not to be more accurate at predicting performance, but to let you try more modifications to see performance trends. What do I gain by blocking the crossover? What happens if I can quicken my shifts? What happens if headwind changes? Although the programs are very good at predicting the final performance (like HP and ET), the true power lies in letting you try nearly an unlimited amount of combinations to find the best one.

What is the advantage of using a simulation program?

Computers will never replace track or dyno testing. However, computer simulations do have several advantages:

The difference between running 1st and last in many competitive forms of racing may only be a 1-2% difference in HP, or cornering ability, or reaction time. The way to become more competitive is to make many small improvements (maybe only 1 or 2 HP at a time) which add up to a larger improvement. In track or dyno testing, test to test variability can "cloud" the actual results. Computers repeat exactly each time so trends can be more easily found.
The computer is fast and cheap. Many modifications can be done in minutes for free which would require weeks of testing and thousands of dollars in parts (throw in a few bruised knuckles).
The computer is not limited by current technology. You can try modifications which your competition has never even dreamed of.

    Try a "fast burn" head which only needs 10° of spark at 8000 RPM.

    Try headers or intake manifolds with 100% anti-reversion.

    Install a 4 valve head on a 5 HP Briggs & Stratton.

    Try an extremely aggressive cam with only 180� Duration and .800" valve lift.

    Rev a low friction Chevy (1" diameter rod & crank journals) to 12000 RPM.

When you find a new design with big gains, you will find or develop the required technology. Then use your dyno to verify and fine tune the "computer optimized" designs.
Simulations let you get inside the engine or vehicle to actually see and understand how things work. Watch Camber change as the car rolls. Watch the exhaust blow down the exhaust pipe when the exhaust valve opens.

How can I Email you a data file to show you a problem I'm having with a program?

This example is for our Engine Analyzer v3.0, but is very similar for other programs: The way to e-mail a file is to first save the engine file. Then go into the ENGDAT folder and attached that file to an e-mail. Send the e-mail to: feedback@performancetrends.com Be sure to include some explanation of why you are sending the file even if you talked to use or Emailed before. We talk to and receive many Emails from many people every day.

Can I use your Windows programs on my Apple Mac (Macintosh) ?

Most modern Macs can run Windows programs without any problems. Apple has its own software called Boot Camp to run Windows programs. Third party programs like Parallels, VMware Fusion, and VirtualBox are also available. Please check the specs of these programs, as they can change. Most all Performance Trends programs are 32 bit Windows applications.

We recently had a customer report he could NOT run our Circle Track Analyzer v4.0 on Mac with 12 Core CPU, 30 Core GPU running Parallels Desktop 18, so there can always be special cases.

for one customer's experience running our Shock Dyno program with his Mac Book

What is error 429?

This error message can appear if a software company does not install the system files correctly. If the error message includes the name of a file (commonly Resize32.ocx), it probably means that there are multiple copies of this file on your computer. The correct location for system files is typically in the Windows\System or Windows\System32 folder. (Performance Trends always installs system files to these locations to prevent this problem.) Copies of the file in other locations are probably the problem.

For fix the problem:

Click on Start (lower left corner of your desktop), then Search, then Files and Folders.
Then type in the name of the file causing the problem. In the example above, you would type in "Resize32.ocx" (without the quotes).
For "Look In", be sure to select All Hard Drives.
Then click on Search.
You should at least find 1 occurrence of the file. If not, you did not type in the name correctly or did not look in All Hard Drives. Try again.
If you find more than 1 copy of this file, rename those NOT found in either the Windows\System or Windows\System32 folder as something else. For example, rename Resize32.ocx as Resize32.ocx.bak (add .bak to the end of its name). (Note: The folder where you find this file should give you a clue as to what software company did not install the system file correctly.)
Now restart your computer and see if this error message is gone when you run the Performance Trends program. If not, reinstall the Performance Trends program and then see if the error message is gone.

My program runs very slow and is locking up. The scroll bar in the table at the lower left is flickering.

We've seen this problem on our DataMite, Cam Analyzer and Port Flow Analyzer programs. It could be a problem on other programs also but just shows up differently. This is a rare problem and is NOT repeatable for any particular computer, program or operating system, meaning we can not predict if it will happen. The solution has been to reduce the screen resolution, from say 1440 x 900 to something like 1024 x 768. This is done by clicking on Start (lower left corner), then Control Panel (possibly Settings, then Control Panel), then Display (possibly Personal Settings in Vista, then Display), then Settings tab. Slide the slide bar for Resolution to a lower setting and then keeping these changes.

Note: Since we wrote this FAQ, we have discovered what is causing this problem, and have put fixes in most of our programs. If you experience this problem, check to see if you have the latest version of the software at our downloads page. If you don't have the latest, update your software (there may be a charge depending on your version). If you do have the latest, please let us know you are having the problem and we'll get the fix implemented.

My security (virus checking) software says my program is infected. What should I do?

First, it is very possible it is a "false positive", which means there really is NOT a virus there. This is a common problem with the free virus checkers. Can you check with a different virus protection software to see if it also says there is a virus there? If a 2nd program does NOT find it, assume there is no virus.

If you think a program really is infected...

Install the same version from our website. This is not critical, but when asked, select to do a Refresh installation instead of a Complete (typical) installation. See if the virus warning goes away.

If it does not:

Is the warning saying the main ".exe" file is infected. The reinstall described above should have fixed that. Therefore, it is pointing even more strongly that it is a false positive. There really is NOT a virus present.
If the warning is coming from a small, support file (like a .doc or .txt file), let your virus protection software deal with it (quarantine, etc). If the virus protection software can not fix it, try renaming the file by putting a .bkp at the end of the files name (helpp12.txt changes to helpp12.txt.bkp). Then run the program and see if it runs OK. If it does, delete the file. You may get an error at some time when this file is needed in the future, but perhaps never will depending on what you ask the program to do.

My software runs, but it takes 1-2 minutes for the graphs to come up. What is wrong?

We have seen this when a customer is using Bitdefender (tm) security. Everything works fine except the graphs take VERY long to come up, possibly up to 2 minutes. If you go into the Bitdefenders settings screen, then click on the Protection item on the left, it gives you an option of Advanced Threat Defense (not shown in screen below). Click on the Advanced Threat Defense option to see screen below, click on Settings and slide the Advanced Threat Defense button to the left to turn it off. Close this screen and problem should be fixed.

Performance Trends' software has many messages. How can I get through these messages faster?

When I press the Function keys in the Performance Trends software, nothing happens. Why?

These 2 questions are related, and we've produced a document to explain them.

for an explanation of these questions

Engine Software Questions

Are these (Engine Analyzer) Program’s accurate?

(Background: This was asked on our website and we had several of our users provide "glowing endorsements". Thanks everyone for the endorsements!)

Our registration cards run about 98% having positive to very positive feedback. The 1-2% with negative feedback usually are because they bought a program that was more than they wanted. Like Dave Koehler said, it may be best to start with the standard Engine Analyzer unless you are VERY familiar with computers and engines.

Any of our computer programs are tools, like a torque wrench or camber gauge. They don't automatically make your car faster, but let you work smarter if you use the tool correctly. You working smarter is what ultimately makes your car faster.

No computer simulation replaces dyno or track testing. GM, Ford and Chrysler have invested millions of dollars on simulation programs which can still be off several %. That is why they still use dynos and test tracks. However, they are still using (and increasingly so) simulation programs because it gets them closer before they start building their first prototypes. It also helps them understand what might happen with trade offs with the actual parts and what potential problems they might encounter. For example, if the actual engine idles at 18" vacuum, but the program predicted 20", the program can show that reducing overlap 4 degrees gets them 2" more Idle Vacuum to get the actual engine to idle with 20" vacuum.

(Update Feb, 2010) We should expand on this answer, as this is a very common question. Probably once a month we have customer's call up and say our Engine Analyzer program's are within 1-2% of the engine dyno. We love to hear about this excellent correlation to a dyno, but realize that if they took their engine to a different dyno, the second dyno's results could easily be 5% different from the first dyno. We also know that you could have your first 4 engines you simulate match the dyno within a couple of percent through the entire RPM range, and then the 5th engine could be 10% off.

What these Engine Simulation programs are doing is showing the potential of a certain combination of engine specifications. They will help you chose a cam, header, carb CFM rating, etc for a particular engine size and RPM range, idle vacuum requirements, etc. They can not anticipate problems, like a combustion chamber which does not want to burn efficiently, an A/F mal-distribution problem, an oil windage problem in the oil pan causing very high losses, etc.

For more explanation on Accuracy for these programs, open up our user manuals and check the Appendix in each titled "Accuracy and Assumptions". Click on the links below:

Std Engine Analyzer User's Manual Engine Analyzer Pro User's Manual

The Seat (advertised) Timing for my cam does not match my cam's actual specs. How can I adjust this?

The Ramp Rating feature in Engine Analyzer Pro and Engine Analyzer Plus lets you fine tune cam profiles. You can enter either .050" and seat timing duration, OR .050" and .200" duration and calculate a Ramp Rating to produce a cam profile which matches those 2 duration inputs. We don't have a feature to match all 3 at this time (April 2012). However, in Engine Analyzer Pro we do have an additional input to better define the cam profile, the Designed Valve Lash. We let you say a cam is designed for, say, .010" lash but specify that you will set the lash at .020". This allows the program to "cut off" some of the more gentle part (low lift part) of the opening and closing ramps.

The graph below shows how we kept the Actual Valve Lash at .020" for 3 different Designed Valve Lash settings. You can see that the Duration at Seat Timing (advertised duration) changed significantly, from the high of 320 degrees at .020" Designed Valve Lash down to 296 degrees at .005" Designed Valve Lash.

The program's Seat (advertised) Timing and Duration for my cam does not match my cam's actual specs. Can the program still be accurate?

This is a common question, that Seat Timing events, especially the Intake Closing event, is critical for predicting engine performance. One reason for this is this number is used in the formula for Dynamic Compression Ratio (compression calculated based on cylinder volume at intake closing instead of total cylinder volume).

This would be somewhat true if the valve immediately snapped to full open at the opening event and then snapped back close at the closing event. However, that is not what happens. The valve is gradually opened and closed to keep the valve train from self destructing. In addition, because of a real valve train bending and flexing, the valve does not even lift when the cam "tells" it to. You may have a cam which is suppose to close at 84 deg after BDC, but once the engine starts running, the valve may actually close at 80 deg, 76 deg, or even earlier. This is affected by RPM, mass of valve train components, valve spring loads, valve train rigidity, etc.

Look at the figure below. It shows how the actual valve lift profile is significantly smaller than what you would think because of these factors. The table shows that the intake closing event has dropped from 84 degrees (as the cam is designed) down by 16 deg to 68 deg at 7000 RPM. It is not as though these ramps are not important. They have a huge effect on how quickly the valve opens once it does open, and on the stress put upon the valve when it closes.

(click picture to enlarge)

The picture below shows the detail of the closing ramp.

(click picture to enlarge)

This table shows how the Intake Closing Event changes with RPM.

Engine RPM	Intake Closing Event
As cam is designed, with a perfectly stiff, zero- mass valve train	84 deg ABDC
1000	80 deg ABDC
3000	76 deg ABDC
5000	72 deg ABDC
7000	68 deg ABDC

So, how do I use this information? Use the .050" (or 1 mm (.040") lift for metric or .053 for Harley Davidson) specs for entering cam data. For calculating a Ramp Rating (for Plus and Pro version), use the .200" lift duration if you have it. Then if you want to further refine the profile in the Engine Analyzer Pro, view the FAQ above (click on blue link).

The Roots Supercharger I Installed Shows No Boost. Why?

Roots superchargers make boost because they pump more air CFM than the engine normally would pump without a supercharger. If you install too small a supercharger or have too low a belt ratio, a roots supercharger will actually be a restriction and produce manifold VACUUM instead of boost

You've chosen a supercharger which is TOO SMALL for this big an engine. Either put on a bigger supercharger or increase the belt ratio, either one will pump more air to the engine.

Click on the blue link to our Supercharger Boost on line Calculator to better understand how this works.

I just made a small change to the turbo specs and it made a HUGE difference in power. Why?

A turbocharger makes power from the engine's exhaust being force through the exhaust turbine. This turbine is connected to a compressor on the inlet side to create boost pressure. If this turbine is too big, then no pressure will be produced and no boost will be produced on the inlet side. If you start to reduce the size of the exhaust turbine, then pressure will be created, which will produce power to compress the inlet air producing more boost. This more boost will produce more exhaust flow which produces more boost. This affect will "snowball" (feedback on itself) until very large levels of boost will be produced rather quickly.

This is why a small change in some engine specification, RPM, Turbocharger Specs, and especially Exhaust Turbine Nozzle Diameter can have a HUGE effect on the performance. The graph below shows a Porsche 2.5L with a Turbine Nozzle Diameter at 1.7" diameter (too big) and the factory size of 1.3" (sized correctly). Just this relatively small change bumped up the boost from about 2 psi boost up to about 20 psi, and the power almost doubled from less than 200 to close to 400 HP.

In the Engine Analyzer programs, what is the advantages and disadvantages of going higher than 75% on the Total Exh/Int %?

The 75% number has been developed from experience over the last 60 years or so. Taylor from MIT developing aircraft engines in WWII was one the first to come up with this # that I know of.

If you have more than 75%, it does NOT means you should restrict the exhaust. It does mean that there is probably more benefit to enlarging the intake valve at the expense of the exhaust valve. Since a bore only allows an intake and exhaust valve combo so big, there usually must be a tradeoff. If you make the intake bigger, the exhaust must get smaller.

This 75% (85% for Nitrous Oxide or supercharged) is VERY general, and reported for info only. It is not used in the program to predict performance, as those calculations are much more detailed.

Back to Top

How do I simulate a restrictor on an EFI engine with a plenum and individual runner throttle plates?

The best way is to pretend the vehicle is using a single plenum EFI Type manifold with appropriate length runners, a plenum matching your air box, and then a restrictor ahead of that To calculate the Carb/Throttle Body CFM rating, click on the Clc button for Carb/Throttle Body CFM Rating. Then just use the dimensions of the restrictor. (The restriction of the other throttle plates in the runners is negligible compared to the restrictor.) For a "perfectly designed" venturi type of restrictor, the "% Improved" spec in the Clc menu should be about 150%.

Back to Top

Why does the spark advance value only increase by maybe 10 degs from idle to 3000 (even on stock setups) when actual distributors have 15-24 degs of advance?

Required spark advance is a complicated subject, especially when you the program tries to predict when detonation will occur. A/F distribution, thermal gradients (differences) across the heads, pistons, etc, combustion chamber design, are just a couple of things the program must make assumptions about. Then we have the factory, which has many different things to consider than just detonation and best power, things like: lighting off the catalysts fast, max catalyst temperatures, transient response, knock sensor calibration, cold start strategies, warranty, transmission durability, etc, etc. That the program does not match the factory calibration is not surprising.

Just a note, idle at 700 RPM is a very different condition than full throttle at 700 RPM, both requiring very different spark advance. Idle or some part throttle conditions may want up to 50 degrees of spark advance. Because of the light load and possibly EGR it may not detonate. But the same RPM at full throttle may only be able to run 10 degrees before it starts to detonate.

Back to Top

Which cylinder heads have a "Compact Wedge" combustion chamber?

Compact Wedge is a chamber which has had considerable race development to improve Burn Rate and efficiency (power for a given amount of fuel & air intake). This includes "cutting edge" Winston Cup, Formula 1, and NHRA Pro Stock type of engines. The "chamber" is actually a combination of the chamber in the head and the Piston Dome design, and how closely they come together in the quench zone.

Most any combination that simply takes "this head" and puts it with "that piston", without careful dyno study, would NOT be considered a "compact wedge".

Back to Top

How much harder is Engine Analyzer Pro compared to Engine Analyzer?

When people ask your question, we usually recommend the std EA3.0 which we sell for $109.95. Should you get comfortable with it and want to go to the Pro version later, you qualify for a $75 discount. Therefore, you are only out $35 for getting the EA3.0 first.

The Pro's more inputs and many more features (menus, commands, etc) can be overwhelming at first. More inputs can also mean more chances to input a wrong measurement, which will produce more inaccurate results than accurate results.

Back to Top

Why can’t I get the Engine Analyzer to match the factory torque and HP ratings for 1960s muscle cars?

Unless you have the actual curves and know how the engine was set up, it is difficult to know exactly what the factory was doing. In the "old days", the factory would put on headers, block the crossover, change the spark and fuel curves, run open exhaust, etc, etc to get "factory ratings". This changed around 1972 when power ratings suddenly dropped. You've seen the "heads up" comparisons between old and new production cars in recent magazines. How else is it possible for modern 300HP Corvette to "blow the doors off" a 1968 425HP Corvette.

Also, if the factory marketing department wanted to "push" low RPM torque, they would simply pick a low RPM point off the torque curve, even if it was not the peak. In addition, dual plane intake manifold tuning is difficult to simulate accurately. Some of these engines could have made more low end torque than the Engine Analyzer program estimates.

Back to Top

How do I enter or calculate the "Corrected Compression Ratio" for my supercharged engine?

The Compression Ratio stays the same, whether supercharged or not. Compression Ratio is a volume ratio and does not change with pressure. To try to explain that supercharging increases cylinder pressures and the tendency to knock (detonate or ping), people have come up with this idea of an "effective" or "corrected" compression ratio based on boost pressure and static compression ratio. However, if everything that effected cylinder pressures and the tendency to knock would have an "effective compression ratio", then we would have one for cam timing, carb size, port restrictions, type of fuel, nitrous oxide, RPM, etc., etc. This term is rather "old fashioned".

The Engine Analyzer program is considering all these aspects of supercharging "behind the scenes". You will notice that torque (an indication of peak cylinder pressures) goes up and the tendency to spark knock goes up when you start increasing boost. (The Pro version would actually graph you cylinder pressures with and without boost for comparison.)

At present, we don't feel a need to display a "Corrected Compression Ratio" in the Engine Analyzer Results. That is not to say we won't do it in a future version if we get other requests.

Is the Engine Analyzer doing ECA (Engine Cycle Analysis)?

In general, ECA means to analyze the Cylinder Pressure over the 720 deg, 4 stroke cycle. Usually this is by measuring cylinder pressure and crank degree very precisely and quickly. Much can be learned about performance by doing this. EA30 does NOT display this type of simulated data, however the EA Pro does. EA Pro also displays many other "cycle" details like port pressures and velocities, piston acceleration or thrust on the cyl. wall, pushrod force, valve opening, etc. If this is the type of detail you want, I'd recommend the EA Pro. Download the demo to see yourself. If you already own EA30, you get a discount upgrading to EA Pro.

Why does the Engine Analyzer Pro show that some Thermal Barrier Coatings reduce performance?

Coatings which reduce friction should always help performance. Coatings which reduce temps on the intake port/runner should always help performance by keeping the intake charge cool and dense. However, coatings which reduce thermal transfer to the piston or cylinder head do 2 things:

1) They reduce heat transfer (loss) which should help performance.
2) They usually increase the surface temp of the piston which should hurt performance.

The total effect of both things determines if there is a net gain or loss. Engine Analyzer Pro v2.1 "C" or later has an alternate coating choice which reduces heat loss but doesn't increase the surface temp as much. Coating manufacturers. say this is more typical of "modern coatings". Unfortunately, we don't know which description of "modern" or "typical" is more accurate.

Does the Engine Analyzer or Engine Analyzer Pro consider Port Size (volume) when estimating Performance?

Yes, the EA Pro (and even the standard EA v3.0) will show the effects of large ports at high and low RPM. Both programs ask for port volume (or average port diameter) for both the head and the intake manifold runner. Most other programs don't ask for these details, but they are required for reasonably accurate predictions. You will definitely see low RPM performance suffer when ports get too big.

My Turbo catalog only lists lb/min for airflow not cfm as Engine Analyzer uses. Is there a formula to convert lb/min to cfm??

Multiply lb/min by 13 for an approximate CFM.

Can the Engine Analyzer and Engine Analyzer Pro be used for small engines, like a 5 HP Briggs?

Yes, we have checked both programs against dyno runs from 5 HP Briggs engines, and other small 4 stroke engines. The agreement is quite good, but we haven’t checked it as much as we have with typical V-8 dyno runs. Its just that we’ve got a lot more V-8 dyno experience.

When HP and torque numbers are reported at very low numbers, they are reported to the nearest .01 ft lbs and HP. Inputs like Bore and Stroke can be specified to .001 ". The displacement is calculated with no rounding for performance calculations. When displacement is displayed, it is rounded to the nearest 0.1 CC.

The Spark Curve reported by the Standard Engine Analyzer starts high, then drops down, then increases again. Is this possible?

The standard Engine Analyzer tries to make the spark curve somewhat automatic. It gives the engine as much spark as the program thinks the engine wants, but not so much as to let the engine detonate (spark knock) for the given conditions (intake air temp, humidity, fuel octane, CR, etc.) The EA Pro is not so automatic, but does give you much more freedom to try different things with spark advance.

In the actual engine, spark advance curves have been developed from what a simple centrifugal advance could do in an "old fashioned" distributor: start at the "initial advance", then at some RPM start advancing to a higher "full advance". This approximately follows what most engines "want".

Most spark requirements increase gradually with RPM. However, if your engine has a big cam, for example, it may produce poor low RPM torque (poor combustion due to poor mixture) and may benefit from more advance until the engine gets "up on tune", where spare requirements then fall, then pick up gradually as RPM increases. Most of the time you really aren't interested in max performance until the engine does get "up on tune", anyway, so don't be too worried about spark advance at low RPM. Just be sure not to let the engine detonate.

Another possibility is if the engine is Supercharged or Turbocharged, at low RPM there is little boost so the engine can run a normal amount of spark. Then, when boost starts to come on, the program "pulls out" spark advance to avoid detonation. Then as RPM increase, spark can be put back in as detonation is not as likely at higher RPM.

Please note that retarding the spark advance a few degrees from optimum timing will not show a large loss in performance, but will give you a better safety margin for avoiding detonation.

Do the Engine Analyzer’s consider Turbo A/R in their calculations? Do I alter the Turbo CFM rating to simulate different exhaust turbine sizes?

Our Engine Analyzer v3.0 and Pro v2.1 both have specs that describe the turbine (exh) side of the turbo separately from the compressor side. A flow map describes the compressor, not the turbine at all. Turbo A/R does NOT affect the flow map, but the turbine size, and how quickly (at what RPM) the turbo starts to develop boost.

I assume you mean that you have timing specs when the Valve Lash of .020 to .025 has been taken up and the valve just starts to open. That is what the program calls seat timing. Just specify seat timing as the method of rating events. However, seat timing is not nearly as accurate as timing at .050", as explained below.

What you want is to have the program simulate the most important and largest amount of the cam profile accurately. That would be duration at .050", .100", .200" etc., the higher lifts. If you use .050" for rating the cam events, this is done much more accurately, because the gradual part of the ramps is over. You are already at the higher lifts.

Cam lift changes very gradually at the very beginning or end of the cam lobe where seat timing is measured. For example duration at .001" could be 320 degrees. Duration at .002" for a cam with mild ramps could be 305 degrees, but for a cam with aggressive lobes could be 316 degrees. This allows minor differences in cam lift aggressiveness to produce very different timing figures. Using the same seat timing events for 2 cams, minor differences in design can produce totally different durations at .050", .100", .200", etc. That is why the cam grinders adopted the .050" cam lift point for comparing cams. It is much more accurate than seat timing.

Also, remember, the Engine Analyzer graphs shows valve lift:

valve lift = tappet lift x Rocker Ratio - Valve Lash

tappet lift = valve lift / rocker ratio + lash

Assuming .020" lash and a 1:1 rocker ratio, if you look for .025 lift on the graph, you are getting cam or tappet lift at .025" + .020 , which would be .045 inches cam lift. With a rocker ratio of 1.5, this would be a cam lift of .025 / 1.5 + .020 = .037 cam lift.

How do I enter the weight of my crankshaft, pistons or rods so I can see the effect on HP?

These weights do affect the rotating inertia of the engine. Rotating inertia is an input in the Short Block specs screen, however currently there is no way in the program to estimate the change in rotating inertia for rod or piston weight changes. This is addressed in our Rotating Inertia Calculator program, and may be included in Engine Analyzer Pro updates.

Rotating inertia only affects power when the engine is accelerating, like in a vehicle when it is accelerating or on a dyno's accelerating ("sweep") testing, say at 300 RPM/sec. Then some of the engine's power must be used (lost) to accelerating the engine's own internal inertia. The lower the inertia, then the less power lost.

People often think that because a piston is lighter, it takes less of the engine's power to accelerate it down from TDC. That is true, but what they forget is that energy is recovered when the piston is decelerated when it again approaches TDC. Therefore there is no net loss when running steady state dyno tests. (Note that there could always be some secondary effects on performance that are very difficult to model. These could be things like lighter pistons don't stretch the rod as much, therefore things like compression ratio or ring seal is maintained, or lighter pistons "rock" in the bore differently as they approach and accelerate from TDC, thus sealing or friction could be slightly different, etc.)

Then you may ask "Why do engine builders strive to use lighter pistons and rods?" That's because lighter pistons and rods put lower stresses on the rods and rod bolts. Lower stresses mean longer durability and/or you can rev the engine higher. With proper choice of cam, Head Flow, etc., most anything that lets rev the engine higher will produce more HP.

What is the formula for determining Intake Valve Closing angle in the Compression Ratio Calculator?

The "Formula" is quite complex (many formulas actually) and has to do with our Cam Profile Designer routines. At .050 tappet lift, the valve may still be .100 " off the seat (not closed). For Dynamic Compression ratio you want to know when the valve is ON the seat so pressure will start to build in the chamber, which changes with profile type, Hyd vs Solid Lifters, Lash, etc. for a particular intake closing at .050" tappet lift. If you know the intake closing at .000 valve lift (seated timing), you should enter it directly. (The .006" lash point is used for hydraulic lifters because they are assumed to have about .006" effective lash.)

Why does the Engine Analyzer and the Engine Analyzer Pro give different results for detonation (spark knock)?

The 2 programs work differently:

EA30 retards spark to not allow detonation to occur. EA30 does not report detonation level, just spark advance.

EA Pro runs the spark curve you give it or one it believes will produce best power and just reports the detonation level.

Behind the scenes, the 2 program's calculations are quite different, especially how they determine detonation levels. As we say for both, you can not use the absolute levels of detonation (or spark curve in EA30) to tune your engine, just use them to see trends. This cam or CR will produce about this much more or less detonation.

Back to Top

Why is there a dip in the torque and HP curve at 3500 RPM?

All highly tuned engines with high overlap cams will encounter good and bad tuning peaks through their RPM range. See Appendix 3 in user's manual "Intake and Exhaust Tuning". For header primaries from 30-40 inches long, a "bad peak" for exhaust tuning comes around 3000-5000 RPM, where you will see a "dip" in the power curves. This is NOT an indication of bad performance, just a "necessary evil" to get good tuning and performance at other RPMs. If you try to eliminate this dip (close up overlap), you will hurt performance at other RPMs. You can move where this dip occurs by changing header primary length. For example, going to longer primaries will move the dip to lower RPMs.

How do I build an engine to run a certain quarter mile ET and MPH?

You would need both an Engine Analyzer (std, Plus or Pro) and Drag Race Analyzer (std or Pro). The Engine Analyzer builds the engine and shows the torque and HP curves. The Drag Race Analyzer takes the power curves and couples it with vehicle specs to come up with ET, MPH, etc. You can Auto-link the 2 programs so that each time you make an engine mod in the Engine Analyzer, the ET and MPH (and 60 ft, finish RPM, etc) from the Drag Race Analyzer are also displayed in the Engine Analyzer results.

How can I speed up the Calculations in Engine Analyzer Pro?

The Engine Analyzer Pro's calculations are VERY detailed and do require a significant amount of time to perform accurately. Some things which will speed up the calculations include:

Use a fast processor computer. Note: At this time we can not take advantage of Dual Core Processor computers.
Run as few RPMs as possible. This means using large RPM increments (500 RPM or larger). Also, start at a high RPM. Low RPMs are VERY time consuming, especially if the engine is unstable at those RPMs (high overlap cams).
Click on Preferences, then Calculations tab. Set the Calculation Repeatability to "Standard".
Use "Simplified Collector" in the Exhaust Specs as the "Detailed Collector" takes more time.
Turn off Valve Train Dynamics in the Cam/Valve Train Specs screen.
Use the "Chain Calculation" feature to run through several calculations unattended (like overnight). Then investigate the best 5 or 10 combinations for further refinement.

Once you find some promising combinations, you can go back to smaller RPM increments, turning ON valve train dynamics, etc. and do more refinement.

I've run the same exact conditions and the Engine Analyzer Pro does not give the exact same results. Why?

The first thing you must understand is there is NO exact answer. Each torque, HP, spark advance, etc result at each RPM is an approximation. The calculation method for complex systems like an engine is an iterative approach to problem solving, and it is the only way it can be done. This involves assuming certain conditions going in, see what the answer is and if the results confirm your assumptions were close enough. See Appendix 2 in the Engine Analyzer Pro user's manual for more explanation.

If you watch the Tuning Pressures graphs during the calculations, you see each cycle's pressure waves. These waves are NOT exactly the same from cycle to cycle for the same RPM and conditions. The program must decide when to stop the calculations after the cycles appear to repeat "good enough", within a tolerance, maybe within 0.5%. This tolerance can NOT be zero difference for an exactly repeatable answer, or you will NEVER get an answer.

The graph below shows the Intake Port Pressure waves for the exact same engine at 8000 RPM. If the pressures were EXACTLY the same, you would only see the green line. Because they are slightly different, you can see the blue line also underneath the green line. The test with the blue line showed 908 HP at 8000, the green line showed 910 HP. (Remember that there is more affecting these results than just Intake Port Pressure. Exhaust Port Pressure, burn rate, heat transfer, etc all have an effect on the end result of 908 vs 910 HP.) However, in both cases, the "tolerance" for an acceptable answer was met. The difference of 2 HP our of 910 HP is only 0.2%.

You should also understand that each cycle (cylinder firing) in the real engine is not exactly the same as the last. Cycle to cycle variation of IMEP (torque on top of the piston), can easily be 5.0 % or more. If you just took, say 3 cylinder firings to get the torque at each RPM, you would get VERY non-repeatable dyno runs. But because good dyno software averages out hundreds of firings together, you can get repeatability of 1.0 % or better, which is still not nearly as repeatable as the Engine Analyzer Pro.

Even our smaller Engine Analyzer programs use this iterative approach. However, their calculations are not nearly as complex so you do get almost exactly the same answer for each particular engine condition.

Notes on Repeatability:

Engine's which are less repeatable are those with high overlap cams, high specific output (high HP/cubic inch) and turbochargers.

To improve repeatability, increase the "Calculation Repeatability" under the "Calculations" tab in the Preferences section. This will make the calculations take more time, but will improve repeatability.

What should I adjust to get the Engine Analyzer results (std, Plus or Pro) to match my real dyno results?

The Users Manuals for both the standard, Plus and Pro versions talk about this in Appendix 4 in each. Click on the blue links below for either manual:

Do not adjust some input spec which you know for sure. For example, you may find that reducing the head flow 20% will make the program match the dyno results. But now the program will show a bigger improvement in performance from a bigger cam than the real engine would show on a dyno. That's because the program sees the heads as being more restrictive, and a bigger cam improves the overall flow significantly. The real difference could be that the dyno calibration is low or high, the dyno has a different amount of accessories it's driving on the front end, etc.

You should enter all the numbers you know for sure, accurately. Then just accept the difference in the results. This way the program can better predict the change in performance from an engine change (modification). Remember, dynos vary in the results they report, especially chassis dynos which can be 15-25% lower than an engine dyno. The Engine Analyzer Pro lets you enter a chassis dyno correction factor to reduce the output to match a particular dyno. Also, the Engine Analyzer programs predict the results you would expect to see on an engine dyno doing a step test, unless you have specified otherwise in the Engine Analyzer Pro.

I'm a little confused about measuring Port Length. Is it measured along the short side of the port?

Port length is measured from the center of the port at the mating surface to the manifold (or header on the exhaust side) down the middle of the port to the valve. This picture may explain it best.

How can I export the cam profiles and cylinder flows and pressures from Engine Analyzer Pro to Excel?

The Engine Analyzer Pro produces what we call RPM Data like torque, HP, fuel flow, BSFC, etc at each RPM of the run. It also produces Cycle Data which are things like cylinder temperature, cylinder pressure, piston thrust on the wall, intake valve lift, intake valve flow, etc at every 4 degrees of crankshaft rotation. (Behind the scenes the data is being calculated at something more like every .1 degrees, but we only report it at every 4 degrees.)

This Cycle Data is one thing which really sets the Engine Analyzer Pro apart from other simulation programs. You can graph it and gain tremendous insight into the internal workings of your engine. However some people want to export it into other programs like Excel (tm) to do their own analysis. The Engine Analyzer Pro provides several methods to let you analyze and export this Cycle Data.

Shown below are some tips for producing Cycle Data graphs. Choose Cycle Data graphs by clicking on the "rpm-CYC" menu option. If you are currently doing RPM data, this menu option is shown as "RPM-cyc". Then you have the choice of doing Cycle Data graphs of "MIXED" data (from one to 4 different cycle data types on one graph like in the graph below of Int and Exh Valve Lift) or just "SINGLE" data graphs (where you pick just one data type from a list as shown below).

(click image to enlarge it)

You can also use the "See-Engine" option at the top of the tabular results. Click on "See-Engine" to bring up the Piston-to-Valve Clearance screen. There you can watch the interaction of the valve lifts, Piston Position, cylinder and port pressures and flows to truly better understand what goes on inside the engine. There you can also click on Options to bring up the See-Engine Options screen, where you can choose to export Tabular Data of the data used on this screen to Notepad.

(click image to enlarge it)

You can also use the "ASCII File" option at the top of the tabular results. Click on "ASCII File" to bring up the Export ASCII File screen. There you can chose most any RPM Data or Cycle Data to export to an ASCII (text) file in various formats. This file can be imported to Excel for most any analysis you wish.

(click image to enlarge it)

How can I create a cam file to be read by Engine Analyzer Pro or other Performance Trends software?

The Engine Analyzer Pro and many of our other engine programs can read a detailed cam file, which typically contains a tappet lift for every cam degree (every 2 crank degrees). These formats include Cam Dr (.c1), Cam Pro Plus (.cpp), S96, Andrews, Comp Cams, to name a few. When you want to create a cam file from generic data, like possibly Excel spread sheet data, you need to create one of these files. Missing just one line of data, or missing a comment line is enough to throw everything off. Therefore we will try to explain the best course of action to produce a cam file.

First, the Comp Cams and S96 formats are the easiest to produce from Excel or other text data.

Comp Cams format consists of 42 lines of comments followed by 360 tappet lift data points. for an example file.

S96 format consists of 360 tappet lift data points followed by an additional 36 lift points. These last 36 points are typically the same as the first 36 points, but are used by some software to confirm the reliability of the measurement. If these last 36 do not match the first 36 very well, something is wrong with the measurements. for an example file.

Cam Dr is a complicated and definitely not recommended. for an example file to show you the complexity.

Our Cam Analyzer has many options for producing many different cam file formats from various types of cam measurements, besides doing all sorts of cam analysis of its own. Here's an example of how to import Excel or text data copied to the Windows clipboard, and create a cam file.

Create a new file and be sure to have selected "Type of Cam Data" as "Measured by Hand". To check this, click on the Test/Cam Setup option at the top of the main screen. This type of data allows for the most manual modification, including entering data by hand from degree wheel and dial indicator readings.

click image to enlarge it

If your file just has lift data, you need to generate some degree wheel readings to match up to the lift data points. Click on the "Generated Degree Wheel Readings" to generate several degree readings. If your lift data is every cam degree, you will select "Crank Deg Steps" of 2 degrees (1 cam degree). The program limits you to 500 degrees, but that should be plenty as the base circle should be 0 lift and that is what the program will assume for the missing data from 502 to 720 degrees. If you are clever, you will enter the proper "First Degree Wheel Reading" point to coincide with the timing of the first lift reading. However, most all Performance Trends products let you specify where you want the centerline of the cam profile. So in the process of creating this file, if you are off in timing, you can either use the Edit commands in Cam Analyzer to advance or retard this data to get the proper timing, or specify the correct Centerline in another program (like Engine Analyzer Pro) to shift the profile into the proper timing.

click image to enlarge it

Here's an example of S96 data read into Notepad that we will use as an example of tappet lift data. Obviously if we already had the S96 file, we wouldn't have to go through all this. Copy all this data using the Ctrl-A (for selecting all lines of text) and then the Ctrl-C (for copying all the selected text to the invisible clipboard). The Edit commands in Notepad also let you do this.

click image to enlarge it

Click the top of the Tappet Lift column (on the title "Tappet Lift") and the Edit Text Data screen appears. Click on the big white field to place the cursor there, and do a Ctrl-V (paste from clipboard) to paste the tappet lifts from Notepad into that field. Then click the "Use This Data" button to fill the Tappet Lift column with these readings. You will notice lots of other Edit options on this screen, which are used if your data has 2 columns of data, of both degree and lift readings.

click image to enlarge it

Once you have pasted the tappet lifts, a cam profile is graphed, as shown below. Other Edit options of Advance/Retard let you change this timing if you want to. Click on File, then "Export as Cam Dr File (and other formats)" for a new screen letting you create several different file formats.

click image to enlarge it

I don't know much about my engine. How can I build an engine file for it?

Obviously we can not have an Example Engine file for every engine ever produced produced. You can start with ANY engine to create your engine. Click on File, then Open Example Engine file and pick an engine file close to your engine. Starting with an Example Engine file close to your engine will be the easiest since you will not have to make as many changes. And, for those things you don't know, leave the settings already in the Example Engine file. Then save your engine file to a name of your choosing, like "My Mustang" by clicking on File, then Save As and entering a name of your choosing.

Read Example 4.1 "I Don't Know my Engine's Specs" starting on page 151 in the Engine Analyzer's User Manual at link below:

https://performancetrends.com/PDFs/EAMAN4.pdf

Since that section of the manual was written, we've added thousands more example parts. So, it should be much easier to get a file close to your engine than even written in Example 4.1.

Also, get used to using Google or other search engines to find info. If you need head flow data for, say, a 1990 1.6L Alfa Romeo engine, enter terms into the google search field like:

1990 Alfa Romeo 1.6L CFM or Alfa Romeo 1.6L head flow CFM

If you need cam specs for the same engine, enter terms into the google search like:

1990 Alfa Romeo 1.6L cam centerline or Alfa Romeo 1.6L cam duration .050

Think of the words or terms you would see in an article describing the info you are looking for. Google will open up a wealth of information for building your engine models.

When calculating Average Port Diameter, what do I use for the Port Length?

Our Engine Analyzer programs (and some others like Port Flow Analyzer) let you calculate Avg Port Diameter from Port Length and Port Volume. The best estimate of the Port Length is using a wire down the center of the port to the top of the center of the valve if there were no valve stem there. There is no way you can do this super accurate; an estimate within 0.25 inches (5 mm) is sufficiently close. See picture below.

How do I use a .jpg file of a turbo compressor map for entering a detailed map into Engine Analyzer Pro Enterprise Edition?

We call this feature "translating" a compressor map.

for a 3 page description of how this is done.

Why does the program show air flow of 600 CFM, but the carb CFM rating is 850 CFM?

People mistake at carb CFM rating with carb CFM flow. A typical performance carb is rated at 1.5" mercury (20.4" water) pressure drop. If you put more pressure drop across it, the carb will flow more, less pressure drop it will flow less. When that 850 CFM carb is setting on a bench to get worked on, it is NOT flowing 850 CFM. It is flowing 0 CFM because there is no pressure drop across it. If you watch the Intake Vacuum calculated in the program for non-boosted applications, you should see the engine flowing close to the carb's CFM rating when the Intake Vacuum is close to 1.5".

People are often surprised that a carb which is rated at a higher CFM flow than what the program needs will produce more power. Like in the example above, if you put on a 600 CFM carb, the power will likely drop even though the engine is only flowing 600 CFM. The reason is because the 600 CFM produces 1.5" intake vacuum and the 850 CFM carb likely only produces 1.0" intake vacuum. The engine has 0.5" less intake vacuum. That's like having a free supercharger providing 0.5" mercury boost, or about 0.25 psi boost.

Now, at some point if you go too big, the metering signal to pull fuel through the jets gets weak, and atomization gets poor, and air fuel distribution between cylinders can get very bad. This will result in a power loss. It is very difficult to predict when a carb starts getting too big with a simulation program. The Engine Analyzer programs typically assume good distribution and atomization unless things are obviously really bad. Then the MxtrQlty number starts going down from 100%, typically just at lower RPMs.

So, to summarize, the Engine Analyzer programs are calculating the proper effect for carb CFM ratings until the CFM rating gets so high that A/F mixture quality can start to suffer.

How do I measure the Intake Runner Length and Diameter for my Individual Runner Weber carbs?

The Engine Analyzer programs use Intake Runner Length and Diameter to determine the quality of the tuning pulses, which have a huge effect on engine performance. Without these pulses, it would be near impossible to get a naturally aspirated engine to produce more than about 90% volumetric efficiency. With these runners properly sized on free flowing heads, you can get up to 130% volumetric efficiency or higher.

Click on the picture below to see the difference between measuring an intake manifold with a plenum shared by more than 1 cylinder, or an individual runner system. As the definition in the manual states, it is the distance from the head back toward the air cleaner to the first "abrupt enlargement". With individual runner carbs or throttle bodies, the runner basically extends from the port through carbs or throttle body and any velocity stack to the "abrupt enlargement" of the opening to atmosphere or a large air cleaner.

click on picture to enlarge it

For Diameter, you want to obtain the average diameter over the entire length. The most precise way is to cc the manifold, carb and velocity stack. With the intake off the engine, block off the mating surface for the head with a plate. Pour in a measured volume of water until you fill the manifold, carb and velocity stack. With this cc volume and the Intake Runner Length, use the "Clc" utility in the program to calculate an exact average diameter. For example, the length could be 13 inches and the volume could be 335 ccs to produce an Average Intake Diameter of 1.41", as shown below.

Typically cc'ing is a lot of work. If you know the carb's diameter where it bolts to the manifold, you can approximately adjust this number up or down slightly knowing the rest of the components. Like in the picture above, if the carb diameter is 1.50", the manifold tapers down slightly to the head, the venturi in the carb also reduces the diameter some, but the velocity stack increases the diameter. With these trade-offs, the 1.50" measurement at the carb base is a very reasonable estimate of the Average Intake Runner Diameter. If one component is especially long, like the velocity stack or manifold, then it's diameter would have a much bigger effect on the Average Intake Runner Diameter.

What is the difference between Designed Valve Lash and Actual Valve Lash?

Some of our more advanced programs which create cam profiles have 2 inputs for the valve lash for solid lifter cams, Designed Valve Lash and Actual Valve Lash. This gives you the ability to run different valve lash settings on the same profile, which could have been designed for a different valve lash than what you are running.

The picture below shows a cam lobe designed to run .026" of valve lash. The Designed Valve Lash is .026". However, we ran it in our Engine Analyzer Pro with 3 different Actual Valve Lashes of .010", .026" (what is was designed for) and .040". You will see on the opening and closing ramps at 0 valve lift, when the Designed and Actual lashes match, the ramp is quite smooth. At a lower lash your follower will "see" some of the typical constant velocity ramp which is designed to take up the lash before opening the valve. At higher lash, the follower starts on a much steeper part of the ramp. In both cases of not having the Designed and Actual valve lashes match, it puts higher stresses on the parts including possible higher side loading on the follower.

click image to enlarge

EZ Flow and Port Flow Analyzer Questions

If I build an EZ Flow flow bench just using a shop vacuum as the flow source, how will that affect my data?

A SuperFlow ^tm 600 has 8 vacuum motors about the size of the motor in a typical shop vacuum cleaner. You will not get the same flow from 1 motor as you do from 8 motors.

We estimate these motors can deliver from 60-80 CFM once fitted into a flow bench or one of our EZ flow systems. This is less than the motor specs may show, but that is because you need flow to overcome the restriction of the measurement orifice (typically 10" water in our EZ Flow System, or more in larger SuperFlow benches) and other losses. Each of these motors will require about 10 amps at 110 VAC or 5 amps at 220-240 VAC if you wire 2 in series. There are similar motors designed specifically for 220-240 VAC.

Vacuum motors also come with different number of stages, 1, 2 or 3. The more stages, the more pressure drop you get, but likely slightly less flow. Typical shop vacuums are 1 or 2 stages, but Superflow motors are all 2 stage motors. A 1 stage vacuum motor is unlikely to produce more than 20" water pressure when fitted in a flow bench.

For example, if you have a typical, "healthy" small block Chevy head which can flow about 300 CFM at 28" pressure drop. At only 10" test pressure, it will flow about 170 CFM. One single stage shop vacuum is likely to produce about 20" pressure when the valve is at low lift and drop off to only 5" to 7" water at high lift for this head.

The Port Flow Analyzer will ask you what test pressure you want to test at, like 10 or 28" water. This is designed to make small corrections, like you were actually at 27.85" and the program corrects to what it would be at exactly 28". If you are at only 5" to 7", the correction done by the software can not be as accurate. For example, if you measure at 5" and correct the flow to what is should have been at 28", it could very likely be 20 CFM or more different than if you actually measured it at 28".

The more important problem could be that a port modification which produced a small improvement in flow found when flowing at 5", could actually hurt flow if you were flowing at 28".

Now, there is nothing magic about 28" water. Smokey Yunick did some significant work on air flow many years ago, and picked this pressure (about 1 psi) as what he thought was reasonable, and it stuck. It is now the industry standard. It is not like a running engine produces 1 psi of pressure across the valves. When the exhaust valve opens, there can be over 1000 psi pressure, and both valves can have reverse pressure so the port actually flows in the reverse direction.

for info on some of Smokey's work on why we flow at 28" water. Source: Allen Roberts, Starting Line Products, Idaho Falls, ID Many thanks Allen.

for our blog entry discussing why we flow at 28" water.

Pressure across intake valve in running engine (with more graphs in blog entry)

So, what does all this mean?

If you want to duplicate what most head porters are doing, you want to actually measure flow at 28" water. Many motorcycle head porters flow at 10" water because of the popularity of the older SF110 bench in the motorcycle market.
One shop vacuum will produce very low test pressures at higher flows. The Port Flow software can correct this to be close to what the head would flow at 28".
The more vacuum motors you use to obtain higher flows, the more amps or electrical power you will need.

My EZ Flow System shows I have significant amounts of leakage when I have my valve closed. Why?

The EZ Flow system is a single range flow bench. It uses an orifice to measure flow, the same as a SuperFlow ^tm bench like the SF600 ^tm, the industry standard. A SuperFlow 600 ^tm has several ranges for measuring flow, which has advantages, especially when measuring very low flows.

The picture below shows the flow measuring inclined manometer of a SuperFlow ^tm bench. You will see that it is very non-linear. This means that a certain change in pressure could be a 10% change in flow at very low flows, but only a 0.5% change in flow at the higher end of the scale. For this reason, the best accuracy is obtained when the flow pressure is above 30% of its full scale flow.

click image to enlarge it

When you leak check, any small amount of electrical variation in the flow pressure sensor will indicate some flow. Because of the very non-linear nature of the relationship between pressure and flow when close to zero flow, the flow it indicates can be very high, much higher than is truly happening. For example, a .01 volt change (like .40 to .41 volts with .40 volts indicating zero flow) could indicate a 3% flow reading. If your EZ Flow is fitted with a 3" diameter flow orifice (400 CFM range), that 3% flow could be 12 CFM flow reading. At the more accurate higher end of the flow range, .01 volt change (like 4.50 to 4.51 volts) would only be 0.2 CFM change.

For this reason, it is not recommended you try to measure leakage flow with the EZ Flow system. Be very careful with your test setup and always assume zero leakage. To do leakage accurately you would need to use a much smaller flow orifice, but in the process of changing flow orifices you could easily change the leakage value.

Does my flow bench require corrections for weather changes?

Many people think flow bench data needs to be corrected for weather changes because engine dyno data is corrected for weather. Here's some things to know:

You typically do not need weather sensors or weather corrections because flow benches deal with volume flow, not mass flow. Engine performance on dynos is affected by air mass flow, so they need corrections.

With our EZ Flow system, we are basically comparing 1 unknown hole (the cylinder head or test piece) with a known hole (the orifice inside bench) with the same quality air. Because of this method of measuring air flow is pressure drop across a restriction, no weather corrections are required. In simple terms, we are comparing apples to apples.

Some flow benches place the blower motor between the test piece and the measurement orifice, we call a "blower center bench" on page 12 of the Port Flow Analyzer user's manual. In these cases, the air temperature can be different flowing through the test piece and the flow orifice, having been heated by the blower motor. Then you need to do a correction for the temperature difference.

Some benches measure air flow with a different method than pressure drop across an orifice, like a hot wire anemometer, Laminar Flow Element, pitot tube, etc. These type of benches sometimes do require some correction for weather, because in these benches you are comparing apples to oranges.

How do I copy my program from an old computer to a new one?

There are several ways to do this. The most universal way is to copy the complete folder from the old computer to a CD or memory stick (zip drive). This is done without the Port Flow Analyzer running. Right click on Start in the lower left corner, then select Explore. On the left side of the new screen, look for the C: drive (sometimes called Local Disk C:) and double click it. On the left side, now look for Program Files folder and double click it. (For Win 7 and Win 10 this may be Program Files (x86).)

For v3.5 Port Flow Analyzer, look for Performance Trends folder and double click on it.

For v3.5 Port Flow Analyzer, look for Port Flow Analyzer v3.5 folder and right click on it and select Copy.

Now find your memory stick or CD drive on the left side of the screen, right click on it and select Paste. You have now copied everything from your old computer to your memory stick or CD. This is an excellent way of making a backup of all you data also.

Now go to your new computer and insert your memory stick or CD. Again, this is done without the Port Flow Analyzer running. A message may come up automatically and you want to "Browse" the memory stick or CD. If not, Right click on Start in the lower left corner, then select Explore. On the left side of the new screen, look for the yellow folder Port Flow Analyzer v3.5. Right click on it and select Copy.

Now look for the C: drive through the same screen from where you Copied the folder. You may have to go up a few levels. When you find the C: drive (sometimes called Local Disk C:), double click it. On the left side, now look for Program Files folder and double click it. (For Win 7 and Win 10 this may be Program Files (x86).)

For v3.5 Port Flow Analyzer, look for Performance Trends folder and double click on it.

For v3.5 Port Flow Analyzer, right click on Performance Trends folder and select Paste. It may ask you if you want to overwrite some files and say Yes.

Now install the Port Flow Analyzer program to your new computer to the default location (where it automatically wants to go). When asked, select to do a Refresh installation instead of a Complete (typical) installation. Now you should be able to start your new Port Flow Analyzer installation on the new computer by double clicking on the desktop icon and all your old test files should be there.

Resetting your Master Flow Bench Specs: To get your flow bench settings and calibrations back, you need to reload the Master Flow Bench Specs. In the Port Flow Analyzer program, open one of your old tests which you know represents how your flow bench is configured today, a recent good test. Click on Flow Bench at the top of the main screen. Once inside Flow Bench Specs screen, click on File (at top) then Save as Master Flow Bench Specs. Click on Back to return to the main screen. Now you should be ready to start testing again.

-----------------------------------------------------------

An alternate way (if you have v3.5 or newer on both computers) is to click on File (upper left corner of main screen), then Backup Flow Tests and follow program instructions.

Install the Port Flow Analyzer program to your new computer, and start the program. Click on File (upper left corner of main screen), then Restore from a Backup, then Restore All Backed Up Tests and follow program instructions.

How do I share data files on a hub (network with shared folder on server and clients)?

for more updated info on this topic for newer computers, like Windows 10

Some of our more advanced programs have a preference where you can store your data files in a different location. The example of how to do this is given below for our Port Flow Analyzer software.

Assume you want to share your Port Flow Analyzer files on 2 different computers, one called the Shop Computer and one called the Office Computer. We will assume that the Office computer is where you want all the new files you create to be stored. This is the procedure to follow:

1. Install and activate the Port Flow Analyzer on both computers.

2. Using Windows Explorer, set up the Port Flow Analyzer folder ( PFA30 for v3.0 and Port Flow Analyzer v3.5 for v3.5 ) on the Office computer to be shared. Usually this is done by right clicking on the PFA 30 or Port Flow Analyzer v3.5 folder (usually under the PerfTrns.pti folder on the C drive for v3.0, or Performance Trends folder under Program Files or Program Files(x86) for v3.5) and selecting the Sharing option. If you don't see Sharing, click on Start, then Help and look for help on Sharing Resources.

3. Map the Shared PFA folder on the Office Computer to the Shop Computer. On Win Me, this is done by right clicking on My Computer and selecting Mapping Network Drive. You should be able to select a drive letter and match it up with the shared resource from the Office Computer. It could be the K: drive for the PFA folder. You will probably have an option to allow this to always be mapped on startup, something like "Reconnect at Logon". This is the easiest, but it can slow down your Shop Computer's system. If you don't see Mapping..., click on Start, then Help and look for help on Mapping or Network Drives, etc.

4. In the PFA program in the Shop Computer, set the Preference to Yes for Use Alternate File Location. Then type in the path to the Mapped drive from the Office Computer, perhaps K:\ PFA . You can use Windows Explorer to see this location, by expanding the Drive of Network Places or Network Neighborhood. Later versions of Port Flow Analyzer have a Browse button to do this. NOTE: It is important that there is no slash "\" at the end of the path. Later versions of Port Flow Analyzer get rid of a slash automatically.

5. Now, when both computers are running, all your flow files, engine and head spec files, etc are being opened from and saved to the Office Computer, both for the Port Flow Analyzer program on either the Office Computer or the Shop Computer.

Update: Newer versions of Windows, like Windows 10 can present some additional problems. Instead of the letter for the drive, like the "Z:" drive shown below, you may have to enter the shared "name" of the drive, like shown in the 2nd picture below. The 3rd picture below shows how this same drive looks when identified in Windows File Explorer.

for pic of "Path to Data Files" as a letter drive

for pic of "Path to Data Files" as a drive name

for pic of how this drive looks in Windows File Explorer

for more info on this topic (possible problems sharing files) on the web

for more updated info on this topic for newer computers, like Windows 10

I lost the power supply for my Black Box II. Can I get one at Radio Shack?

You can get a replacement power supply from us at a nominal charge. If you need something right away, here are the specs for most all our Black Box II, DataMite III USB, and DataMite Mini USB loggers: 12 volt DC, 300 mAmp minimum, 2.1mm center pin diameter, 5.5 mm barrel OD, and most importantly: center positive. If you are not sure what these specs mean, purchase one from us.

My Black Box II readings do not exactly match my manometer readings at some points. What can I do?

For most users, a flow bench is a comparator to see if the flow is improving or not. So, if the electronics reads, say, 1% low at some flows it always reads the same 1% low. If you make a 5% improvement in flow in that flow range, the bench still reads it as a 5% improvement.

For various reasons, 2 sensors reading the same thing (pressure, temperature, etc) do not always read the same. If you assume your flow manometer is correct (and that can be a big assumption), you may want the electronic reading to exactly match the manometer at all readings. The Head Porter version of the Port Flow software has a feature called "Fine Tuning" that can do this. It is somewhat involved and too long for this FAQ page, but is explained in the link below:

to read how to "Fine Tune" your Flow Sensor.

My Flow Bench readings seem erratic or inaccurate, even based on my manometer readings. How can I troubleshoot this?

Many times erratic flow bench readings can be caused by problems with the hoses which provide the pressure readings to the fluid manometers and/or pressure sensors for the flow bench electronics like Performance Trends Black Box II or the SuperFlow Flowcom.

to read how to "Troubleshooting Flow Bench Hoses"

Besides CFM Flow, what can I measure on a flow bench?

Most people think flow benches can only measure head flow, and produce graphs of CFM vs lift. But with the proper tools from Performance Trends, you can learn a lot more about your heads.

to read the blog post "Besides CFM Flow, what can you measure on a Flow Bench?"

Why won't my automatic valve opener find the tip of my valve?

One of the most common problems encountered with the automatic valve opener is that the square body of the valve opener is not grounded to the valve you are testing through the head. The way to check this is with an ohm meter. Check the ohms of resistance between the square body and the tip of the valve, when the opener is mounted on the head. Place 1 probe on a screw head on the square body, and the other on, say, the intake valve tip (or stem) when the valve opener is retracted off the valve tip. If it is more than about 50 ohms or so, you probably need a better way to ground the valve opener to the head. The picture below with the black wire shows 1 method of correcting this problem.

for a PDF with more info for a movie

What can I check on my Electronic Velocity Probe to see if it has a fault?

Disconnect the Electronic Velocity Probe's 4 pin connector. Set a digital volt meter to read ohms of resistance. With the locking tab at the top, measure the resistance between the 2 male pins at the bottom of the connector. It should read within about 4 ohms of 100 ohms.

click image to enlarge

To test the remote tip, measure the resistance between the 2 female sockets in the connector. Again, it should read within about 4 ohms of 100 ohms.

click image to enlarge

If it is not within this range, contact Performance Trends with this information.

Another thing to check: The Electronic Velocity Pro requires 12 volts to operate. Be sure you have the 12 volt DC power supply plugged in and it is producing at least 11 volts DC. This power could plug directly into the probe's cable (on FlowComs), or the USB Logger for Performance Trends' USB Black Box.

Are the Electronic Velocity Probes as accurate as the traditional pitot tubes?

No velocity probe is exactly repeatable from day to day. Weather and temperature conditions, how you hold the probe, the exact position of the probe, etc all effect the readings. That being said, the standard pressure style pitot tubes are more repeatable from day to day. The electronic velocity probes can drift some from day-to-day, but their small size advantage greatly make up for this drift.

For example, say the electronic velocity probe reads 10% higher today than yesterday, which is more "drift" than is likely to occur. It still will show that position A in the port is reading 50% higher velocity than position B. That is really what you are interested in, is where are the high and low velocity areas in the port or around the valve.

Check the graph below, where the one on the right has readings 10% higher than the one on the left. Both graphs show that the velocity at the 12:00 position is MUCH higher than the rest of the valve, about 210+ ft/sec, and the flow drops quickly at both the 3:00 and 9:00 positions, to about 60 ft/second. Then at the 7:00 position it increases back to about 120 ft/second.

click image to enlarge it

Considering that this type of detailed testing is not even possible with conventional pressure style pitot tubes, the electronic velocity probes are the clear winner even if they can drift some.

I'm having trouble understanding how to get my Valve Opener to work with my SF1020 to automatically do a flow test. Can you help?

The Automatic Valve Opener works best when the flow ranges you will use for your test are already set up in your Port Flow Analyzer. Then the software can do the range changes as necessary and set the correct valve lifts you have scheduled.

for a PDF with more info

We flow components like air cleaners at different test pressures. Can the Port Flow Analyzer do this?

Yes, there is a Preference setting in the Pro or Head Porter versions to turn on this feature in your Port Flow Analyzer.

click image to enlarge

and check pages 171-172 for details in the Port Flow Analyzer manual.

Cam Analyzer / Cam Test Stand Questions

How does the software know my cam timing when I measure the cam on the Cam Test Stand?

When you measure a cam on a stand, you don't know how the cam will be installed in the engine. Typically you will tell the software something about the cam, like #1 Intake centerline is at 110 deg. Then you measure that lobe and that info indexes the rotary encoder. The #1 intake will have exactly a 110 deg centerline. All other lobes are indexed from that, like #1 exhaust centerline measures to be 106.4 deg, or a 108.2 deg lobe separation. When you measure #2 intake, it will likely be slightly different than #1 due to manufacturing tolerance, like 109.63 degrees.

Here's an example of the Lobe Comparison report when #1 intake centerline was indexed to 102 degrees.

Now, if you KNOW where a keyway or dowel pin is located, you can measure that location with the universal flat and Plus version of software. Then after measuring that you tell the software this dowel pin is exactly at, say, 44 deg after TDC, and that will index the rotary encoder. Then you can measure the lobes based on that info.

It is important to note that unless you KNOW this keyway or dowel pin location, do NOT make any assumptions. Aftermarket cams rarely have predictable timing locations. That is why they recommend you check cam timing after the cam is installed. If you are measuring production cams, like checking cams from a factory, or doing tech inspection, you probably DO KNOW where this location is and can use this method to index the rotary encoder.

How do I convert a cam file (CPP, S96, Cam Dr, etc) to a manufacturing type file (.p, .igs, .iges, X/Y, etc) for a CNC cam grinder.

First, you need the Cam Analyzer Plus and Cam Grinder versions of the Cam Analyzer to do this. However, you do not need our Cam Test Stand if you have the file in some other type of computer file format, like Cam Dr, S96, Cam Pro Plus CPP, etc).

for a step-by-step procedure starting with a CPP file and creating an X/Y file and an .iges file which can be read by various CNC software.

for a newer step-by-step procedure starting with an S96 file and creating a solid, 3D .iges file by 2 different methods which, can be read by various CNC software.

I have a cam profile in Excel and I want to use it in my Cam Grinding machine that has large grinding wheel. Can Cam Analyzer make that conversion?

First, you need the Can Analyzer Plus and Cam Grinder versions of the Cam Analyzer to do this. However, you do not need our Cam Test Stand if you have the file in some other type of computer file format. In addition to the computer file, you will need to know:

Base circle of the cam
Radius of the probe or roller follower used for measuring the lift file you have
Diameter of the grinding wheel on the cam grinding machine, typically 10-20 inches (200-500 mm) or so.

for a step-by-step procedure (including picture below) starting with an Excel file and creating a .csv file which can be read with Excel after the conversion.

How do I create a manufacturing file to exactly copy of a cam lobe I just measured?

You need the Cam Grinder version of the Cam Analyzer to do this. First you must measure the cam using the Virtual Follower option, available via the "See Virtual Follower Details" button in the Test/Cam Layout screen, as shown in the screen below.

click image to enlarge

You can use either the .750" diameter Universal Roller or the linear encoder's standard tip directly on the cam lobe. The critical specs to set are the Base Circle you will measure from the cam you are measuring, specifying one of the Virtual Follower "Follower Types" (recommend Virtual Roller), and the Probe Radius you are using.

click image to enlarge

Then measure the lobe you want to duplicate with the Cam Test Stand. When you have the lobe measured, click on File at upper left of main screen, the Export Manufacturing Style File from the list.

click image to enlarge

In the Export screen which comes up, choose IGS as the file type at top (or one of the others if you are sure you can use that type), the file name, file destination and which Lobe to Use (the one you measured). Then click on the Make File button. IGS is the Format recommended because many CAD/CAM programs like SolidWorks and Mastercam will import it directly.

click image to enlarge

A portion of the resulting IGS file is shown below in Notepad below.

click image to enlarge

to view the entire file.

Can I figure out how much to reduce the base circle for regrinding an existing cam for larger lobes and lobe separation?

to review a related FAQ: Can I create an XY or polar angle/length file for roughing out a cam on my CNC from .050" duration and max lift data only? Many of the steps in that procedure will be used for this FAQ.

We will assume you have a customer with an existing cam similar to the cam in the FAQ below, with 110 deg lobe separation, .380 lobe lift, and 235 deg duration at .050" lift. You can either create a generic profile from simple .050" specs (like FAQ below, and simpler) or by measuring the actual cam (more time consuming but slightly more accurate). However, the critical aspect of the profile is that it includes Absolute Lift, so that the lift on base circle is half the base circle diameter. For this FAQ, we will assume you've created the original cam file from .050" specs like in the FAQ below.

Next we will assume the customer wants you to regrind this camshaft to specs of 114 deg intake centerline, 110 deg exhaust centerline (which equates to 112 deg lobe separation and 2 deg retard), .410" lobe lift, and 245 deg duration at .050" lift. During this process, the program asks what to use for base circle. The smaller cam has a 1.250" base circle. You know this larger cam will need less, so you specify 1.200" base circle as a "first cut". Now you make an overlay graph of the actual cam profile for the 2 files.

Screen showing Graph Specs for comparison graph
click image to enlarge

Graph showing initial comparison between small cam with larger base circle and bigger cam with smaller base circle. Note: All graphs for this FAQ are shown drawn with thick lines so features show up better. Several line thicknesses are available for graphing.
click image to enlarge

Advancing the bigger cam profile 2 degrees on the graph shows a better comparison
click image to enlarge

Using the cursor you can find the exact amount of lift between the 2 profiles. You want all points of the new, larger cam to be below (less than) the points of the smaller profile. This ensures there is enough material to be ground off the cam to create the new, larger profile. The "first cut" of using a 1.200" base circle is not small enough. It appears that the bigger cam profile is at least .007" higher than the smaller cam at some points.
click image to enlarge

Go back into the Virtual Follower specs and reduce the base circle to reduce lift by .008" lift (.016 inches diameter). This would be a new base circle of 1.184 diameter base circle. Make the graph again, advance the bigger lobe 2 degrees for better comparison, and check for any points of the bigger profile being above the smaller profile. There are still some points higher by .002", so try again with a 1.176" base circle. This should keep the larger cam below the smaller cam by .002".
click image to enlarge

Virtual Follower specs for creating the bigger cam profiles with 1.176" base circle. Note that we are also assuming a 1.000" roller diameter as was the case in the FAQ below.
click image to enlarge

Graph again with 1.176 base circle and check that all bigger cam lifts are less than the smaller cam lifts. By using the cursor at various points, you verify that the bigger cam is below the smaller cam by about .002" at the closest point.
click image to enlarge

Set the Graph Setting of "Show End View of Cam" to Yes. This is the same data as in the graph above, just graphed a different way. NOTE: Before you can do this you must advance the bigger cam by 2 degrees, as has been done in all other comparison graphs. When graphing with the Graph Setting of "Show End View of Cam" to Yes, you can not advance on the graph screen as the math to do this is very time consuming. So you must advance it in the Test/Cam Setup screen as shown in the figure below.
click image to enlarge

Graph with "Show End View of Cam" set to Yes. This also confirms that the larger cam is below the smaller cam at all points. This means there is enough material on the smaller cam for the larger cam profile to be ground.
click image to enlarge

To summarize: A 1.176 base circle should provide at least .002" of material on the smaller profile camshaft (1.250" base circle) for this particular larger cam profile to be ground.

Note: If the intake and exhaust profiles are the same for each cam, then this process could have been easier by keeping the cam timing the same for the bigger cam. In this case, the bigger cam had 2 deg of retard in the timing, which required the extra step of advancing the cam 2 deg for all comparisons.

Can I create an XY or polar angle/length file for roughing out a cam on my CNC from .050" duration and max lift data only?

(Note: The profiles created by this simple method do not consider all the required design features to be optimum in a race engine.)

Yes you can, but you need the Cam Grinder version. The process consists of the following:

Start a new Cam Profile with the "Type of Cam Data" set to "Create from .050" Cam Specs".

Enter specs you want. Say you want to create a profile for a solid flat tappet cam with .380 lobe lift and 235 duration at .050" lift with a 110 lobe separation and .020" valve lash. You would enter them in the screen as shown below, with this slight modification. Because the Cam Analyzer requires at least .004" lash, you add .004 to the Max Lobe Lift and Designed Vlv Lash, and enter a .004 Actual Valve Lash. This way you get the full .020" ramp for taking up valve lash before the actual profile. (If you don't understand this, just do as suggested above by adding .004 to those specs and specifying .004 Actual Valve Lash.)

click image to enlarge

This creates a file for these cam lobes, intake and exhaust.

Click on Settings at top of main screen, then Preferences. Click on Electronics tab at top, and be sure to set "Show More Options in Setup" to Yes to enable some needed options. Click on OK to keep this change.

Click on Test/Cam Setup at top of main screen. In this screen, change the "Type of Cam Data" to "Measured with Electronics". (This "Type of Cam Data" format allows for much more advanced analysis than any other type.) The program will explain that this is a 2 step process. During this process, it will ask in you want to include a Base Circle Measurement. Say Yes and enter your desired base circle for this cam, say, 1.25". This flags the software that you want to do Absolute Measurements, which includes the Base Circle value in lift. It also flags the software that you want to deal with all 400 data points which make up a "Measured with Electronics" file, 360 plus 40 additional to check repeatability, similar to the S96 file format.

Once created, click on "Back" at upper left corner to back out of this screen to main screen so this conversion is finalized.

Now, click on Test/Cam Setup at top of main screen again. Click on the Virtual Follower button and in the Virtual Follower screen. You should see your Base Circle entry in this screen. Set the Roller Diameter to the diameter roller you want to simulate, 1.00". Then also set the Probe Radius to exactly half of that, .50". This tells the program that the profile created is the exact profile I want this cam profile to produce. Click on Recalculate Results to make sure the program recalculates everything, which produces the Actual Profile data in addition to the follower profile data. Click on Back (OK + save) and you will see this conversion being done.

click image to enlarge

Now click on Graphs, upper left of main screen. For the "Type" select "Pick from List" and you will activate the more advanced graphing types. Highlight just the "Actual Cam Profile" option from this list. Also set "Graph Absolute Lift" to Yes. In addition, if you want to get a file of XY coordinates, set "Show End View of Cam" to Yes, or No for Polar Coordinates. Then click Make Graph in lower left and a graph of your data will be produced.

click image to enlarge

click image to enlarge

To Export as a file, click on File (upper left), then Export and choose either XY or Polar as the 2 export options. It will ask for a file name for saving the data.

for an example of XY data.

for an example of polar coordinates data.

I have an OHC engine, with 2 cams, an Intake and an Exhaust. How do I set this up?

There are several Example Cam Designs preloaded for various OHC engines. See Figure below.
click image to enlarge

Even if you don't have one of these engines, the cam design may still match engines by Dodge, GM, Nissan, etc.

If you don't think one of these choices will work for your cam, click on the following link: Multi Lobe Instructions.pdf

I measured my diesel cam with a .750" and .850" roller and show very little difference. Is this correct?

First, let's see what the difference "SHOULD BE". This is most easily done using the Virtual Follower feature in the Plus version of Cam Analyzer. So you will measure the cam with the linear sensor directly on the cam lobe. (The v4.0 lets you use a larger diameter follower like our .750" Universal Roller, pn CTS-UR750.) Then you will fill in the Virtual Follower specs screen with the information needed for the simulation of what a .750" and .850" would do "riding on" this profile.

Surprisingly, the difference was less than 1 deg at .050" lift, but this big diesel cam had a 1.65" base circle. Perhaps that is why the difference is less than expected. The simulation was run again, this time using a 1.10" base circle, which showed almost a 2 deg difference. The table below summarizes the results:

Base Circle	.750" Roller Duration at .050"	.850" Roller Duration at .050"	Difference, Duration at .050"
1.65"	249.42	250.34	0.92
1.10	256.44	258.22	1.78

This graph shows the results:

The conclusions which can be drawn are:

The difference in duration produce by a .750" and .850" roller may not be as large as you imagine.
The difference in duration produce by a .750" and .850" roller increases as the cam's base circle decreases.

I measured the cam with the pointer directly on the lobe and the results don't look anything like the Cam Catalog's Specs. Why?

Measuring a cam profile with a pointer (not the actual follower) directly on the cam lobe DOES NOT give the true follower lift as it would in the engine. You have to either measured the cam profile with the actual follower, OR use the Virtual Follower feature, part of the Plus version of Cam Analyzer.

Click here to see what the Plus version adds, paying close attention to the Virtual Follower feature.

Check the graph below of typical test results "as measured" (pink line is lift, yellow line is accel) and after simulating a roller follower using Virtual Follower (blue line is lift, green is accel). You can see they are totally different, which explains why your measurements do not match the catalog's Cam Specs. You can see the Virtual results, simulating a follower will give more lift at TDC, larger duration numbers, much lower acceleration over the nose of the cam.

My Results do not repeat when I measure the same lobe again. What could be happening?

Repeatability should easily be within .0005 for lift and within .25 deg for duration and events. If not, check some of these testing tips:

For roller lifters, the roller bearings could change slightly as the roller rolls. If you have a roller lifter to waste, force some shim stock between the roller and the lifter to prevent rolling. Also, how the lifter roller axis aligns with the cam axis (they should be exactly parallel) will affect test-to-test repeatability. We have the Universal Roller with a ball which eliminates the 2 potential problems above.
Rotate the cam relatively slowly and smoothly. The software is designed to be very forgiving for erratic movement. Stopping and starting should not be a problem. However, going too fast over the lobe will produce errors. Going faster on the base circle is typically not a problem. The "warnings" the program gives about rotating the cam too fast are overly cautious. Most users turn off these warnings in the Preferences section.
Be sure the linear encoder's arm holding the lifter is always perpendicular (90 degrees) to the cam axis. This means it is as close to the cam as possible, and the lifter is always directly over the centerline of the cam. This is especially critical for roller lifters and when using the linear encoder's pointer directly on the cam lobe for the Virtual Follower feature.
Be sure to tighten up the linear encoder stand when you move to the next lobe.
Be sure the linear encoder does not top out or bottom out as it goes through the total range of motion.
Be sure the lifter moves smoothly in the bore so its weight can keep it in contact with the lobe at all times.
The cam should turn easily and smoothly in the "V" blocks. You may have to clean any coatings off the cam lobes for this to occur.
Be sure the cam lobe is free from dirt, coatings and debris. New cams often come with a "break in" coating. This should be cleaned off before you can get accurate measurements. If all you need are approximate measurements, you may choose to leave this coating on the lobe.
Be sure the linear encoder's point is setting on a flat spot on top of the lifter. We have magnets which you can place on top of lifters to provide a flat top. If the pointer is on the edge of the pushrod's oil hole, it may move to different depths as you measure the lobe.
The rotary encoder can not slip during testing, as then you lose your degrees index. Things to check for slippage include:
- When you are done testing a cam, back out to the main screen and note the centerline of the 2nd lobe you measured. Typically this would be the Exhaust lobe on Cylinder #1. Click on File, then Save to save your results. Then go back into the Record screen without powering down the Black Box II. If the box is not powered down and you don't re-zero it, it will retain its rotary index. Re-measure this 2nd lobe and see if the centerline comes up within .25 degrees of the first measurement. If it is way off (2 deg or more), you've had slippage or the rotary encoder has lost its index for some reason.
- Another thing that could look like slippage. It is critical that the body of the rotary encoder be held stationary. This is done by making sure the rotary encoder's arm is held firmly to the magnet. If the magnet the rotary encoder's arm is attached to has a rough edge, and the cam is moved front to back, the arm may be in a slight notch for one lobe, and on a slight peak for the next lobe. This is not good. Also, if the encoder's cable is not helping to hold the encoder arm to the magnet, the encoder's body may be moving around.
- Some cams do not have a large face for the rotary encoder magnet to attach to. We use a split collar to lightly clamp on the end of these cams to provide a larger face.
- Be sure the rotary encoder is centered on the end of the cam. Slide and hold the magnet back from the encoder's index point. Place the point in the end of the cam, then slowly release the magnet. This ensures the index point is well seated into the centering hole in the end of the cam.

Click here for the PDF we send with the Cam Test Stand with these and possibly more Accuracy Tips

I'm using the Virtual Follower feature, but my results do not look like what the cam should be?

Here's some things to check for when using the Virtual Follower feature:

Be sure your measurements are accurate. When you put the very small radius pointer from the linear encoder directly on the lobe, errors which are minor when using the actual follower on the lobe now are very critical.

Be sure the linear encoder pointer is directly on top of the lobe. This means that when the pointer is on base circle, it is on the highest part of the lobe. See Figure Below.
Be sure that all parts of the Cam Stand are tight and can not move during the measurement. This includes the V blocks, the tip of the linear encoder, the linear encoder bracket (especially the thumb screw for the slide of this bracket), etc.
Be sure the linear encoder is not binding or catching from side loading on steep ramps. This is especially common on new cams where the lobe is still rough.

Be sure your calculations are accurate. This means you have accurately filled in all the Virtual Follower specs correctly.

Cam Lobe Base Circle Diameter has a much larger effect on this than most people would think, and can easily change from cam to cam for the same engine design. Do NOT just assume our default is correct for your cam.
Obviously, Roller Diameter is critical for roller cams.
For the OHC Rocker Arm designs, most all measurements are critical. Also, be sure you have correctly identified the rotation direction of the cam. Once you have measured a couple of lobes, use the "Watch Cam/Follower" feature to watch the valve train move to verify this is correct. Also notice on the left side of this screen that the view is correct, either from the Front or Rear of the engine.

What is the advantage of using the Universal Flat Tappet (or TDC Checking Bridge) to find TDC?

Without bridge, you must tell program where #1 Intake Lobe is timed (like 108 centerline, .045" lift at TDC, etc). Then the rotary encoder is timed from this for all other lobes on the cam.

With the TDC checking bridge, you find dowel pin or keyway location. Then with info Performance Trends has on where these are suppose to be located (timed) for about 100 common engines, we can say how the cam was designed to be timed in the engine. This assumes you will use standard timing gears, timing marks, etc. However, if Performance Trends does NOT have information on your particular engine, and you can not obtain it, you will have to measure a few cams to obtain your own number.

Will the software let you extrapolate from a .750 inch diameter roller lifter to a .800 inch diameter roller?

V3.8 and earlier required you to measure the cam with encoder's pointer to do Virtual Follower simulation, not a follower. With v4.0, the program now lets you specify the diameter of the "pointer" you are using, which could be the standard tiny Ono Sokki pointer on the tip of the linear encoder, or our Performance Trends .750" diameter Universal Roller, or some other diameter. This other diameter could be for a roller follower you used. However, it is not recommended you go very much larger than .750" diameter.

Do I need a different follower for every type of cam I measure?

If you are measuring a flat tappet cam, then you can use the flat tappet of the same size or larger for all flat tappet cams. Note: There can be very slight difference due to the differences in "crown radius" between different flat tappet cams, but this typically would produce less than 0.1 deg error. Also, if a cam is designed wrong and runs on the edge of the flat tappet (tappet is too small for the cam) you can also get small errors.

If you are measuring a roller tappet cam, then you can use any roller with the same roller diameter for that cam.

You can also use the "Virtual Follower" simulation in the Plus version of the program to simulate most any flat or roller follower and several OHC rocker arm geometries. Then you don't need any followers to ride on the cam lobe when you measure it. However, the "simulation" may be slightly different than using the actual follower because the setup is more critical to have exactly correct. See FAQ above by clicking here.

Do I need to zero out the linear encoder (lift measuring device) before I measure a lobe?

Rezeroing the lift indicator (linear encoder) before measuring has NO affect on the results. After the full profile is recorded, the program figures out what should be called zero lift, based on either of 2 criteria which can be set in Preferences:

1) The program takes the AVERAGE lift it finds on base circle and calls this zero lift. This can be a little difficult because it can be difficult to determine where base circle ends and a fairly gentle ramp begins. However, this is the number used by several cam grinders. For this situation, you will see both + and - lift values on base circle.

2) The program takes the MINIMUM lift it finds on base circle and calls this zero lift. This is much easier to determine, and you will see only + lift values for the whole profile.

I've got a cam with multiple lobes for each cylinder. How can I export a Cam Dr file to the Engine Analyzer Pro?

I recommend you do NOT export the file as a Cam Dr file. The Cam Dr format consists of 1 Intake and 1 Exhaust lobe, and was developed many years ago for cams with 1 intake lobe AND 1 exhaust lobe for all cylinders. Many cams with multiple lobes for each cylinder are all intake or all exhaust lobes, and do not have both intake and exhaust lobes. Also, you may want to export the "B" lobe for the intake with the "A" lobe for exhaust lobe. The combinations makes a Cam Dr format very difficult to deal with.

Instead, export as a "S96" format, which is just for 1 lobe. This file can be used in the Engine Analyzer Pro for either the Intake or Exhaust lobe. This is done by clicking on "File" (upper left corner of main screen), then "Export as Cam Dr (and other formats)" for the screen below. The "Browse" button should quickly find the "CamFiles" folder in your Engine Analyzer Pro, the default location the Engine Analyzer Pro uses for cam files. Choose the S96 format as the "Format" and enter the other info required, including picking the "Lobe to Use" from the drop down.

If I want to check a cam in the engine (or an OHC cam in the head), what do I do different?

There are 2 options in the Record Screen which change how the sensors read lift and rotation. If you start a new test using the EZ Start Wizard, the Wizard will prompt you for these 2 options if you say 'On Engine' at the 'Cam Test Stand' screen. See below.

(click image to enlarge)

Even if you don't use the EZ Start Wizard, you can make these changes under Options in the Recording screen. See below:

(click images to enlarge)

The 'Reversed' setting for the Linear Sensor means the sensor records extension (as the valve goes down) as lift.

The 'On Crankshaft' lets the program know to expect 2 complete revolutions to be equal to 1 camshaft revolution.

These options should allow you to make most any cam measurements, either on the stand or in the engine or head.

Filtering seems to have a large effect on the acceleration readings. What is the correct Filtering setting to use?

When you have a very high "spike" in data (like the accelerations of this cam) over a VERY short period of time, the filtering level has a large effect the height of the spike. Going from None to Some to Medium to Heavy filtering reduces the magnitude of the acceleration spike significantly each step. On the closing ramp, it drops from .035 to .012, or to much less than half of the unfiltered value. However, over the nose of the cam (the negative acceleration), there is almost no effect from Some to Medium to Heavy, where it remains constant at -.006 to -.005. That's because the acceleration over the nose occurs over a long period of time, and is not changing quickly.

Therefore, there is a certain amount of user judgment which must be used for the proper Filtering setting. With filtering = None, the acceleration looks way too jagged to be useful. At Filtering = Some, it looks fairly well behaved. Filtering beyond this amount is NOT improving the data quality. Judgment at this point would say Some filtering would be correct.

Here's some guidelines for the proper filtering.

The basic principle for filtering is use the smallest amount which smoothes the data just enough for your use.
If the values you are looking at go up at one point, then down, then up again, then down again, you probably need more filtering.
For acceleration measurements, you most all the time need at least Light filtering. When the program produces a number for reporting PkOpen, PkNose, and PkClose acceleration numbers, it is using Medium filtering.
At Light filtering for acceleration, if you measure a cam lobe again, and a "spike" or "drop" occurs at the same exact point each time you measure a lobe, then the effect could be real. Do not increase the Filtering to Medium or Heavy to smooth out this real affect. It may not necessarily be something wrong with the cam lobe. There could be a problem with your measurement setup, like dirt on the lobe, or a loose tip on the linear encoder, or a lifter which is getting cocked in the lifter bore from side loading. With filtering, you want to improve the data quality, but NOT cover up real effects.

for a FAQ on recommendations for Jerk filtering

I'm a cam grinder, and I want to see all lobes spread out like when viewed from the end of the cam. Is that possible?

Yes, but you need the more advanced Cam Grinder version of the software. Most cam grinders will also want to locate the lobes with respect to the "timing point", be it a dowel pin or keyway.

click image to enlarge

for a 4 page description of how this is done.

When I graph the Jerk for a profile, it just seems to be a bunch of jumpy lines. Why?

Valve or Tappet Jerk is a calculation which really looks at a cam profile in great detail. It is technically called the third derivative of Lift, or the derivative of acceleration. If you are driving on a new highway, it feels quite smooth. However, if you get out a magnifying glass, you would see little stones and holes all over. Jerk is like looking at the cam profile with a strong magnifying glass. The graph below shows the definition of jerk (red line), as the rate of change in acceleration (light blue line).

click image to enlarge

Here's what jerk looks like with no smoothing or filtering to the data for 5 different measurements on the same lobe, manually turning the cam on a non-motorized Cam Test Stand with V Blocks. Some measurements were purposely done fast (in about 10 seconds) and some were done slow (taking over 20 seconds) to see if it affected the data. Testing showed almost no advantage to taking a great deal of time doing the measurements.

click image to enlarge

Here's the Graph Type screen which lets you apply some smoothing to the data. For Jerk, you often want to apply as much filtering as the program allows. We applied filtering both to the tappet lift measurements, and then also applied Heavy filtering to the Jerk calculations.

click image to enlarge

When you apply these heaviest filtering levels, you can see that the Jerk graphs all compare very well with each other. So, in conclusion, for evaluating Jerk calculations it is best to apply the heaviest filtering levels.

click image to enlarge

for a FAQ on recommendations for Acceleration filtering

Can I find TDC with the linear encoder on the top of the piston?

Yes, it can be done if you set the "TDC Method" in Test/Cam Setup screen to "User Finds TDC with Sensor". This is typically meant to find a dowel pin or keyway location, but will work equally well with the piston going up and down. Put sensor on piston top about 0.2" from TDC, rotate crank through TDC and back down to about 0.2" after TDC (program will tell you when it is enough), and then tell program this was 0 deg from TDC. This should be the TDC during overlap, not during compression..

How can I tell if a cam is hydraulic or solid, and if solid, what lash it is designed for?

It is not always obvious if a cam profile is hydraulic or solid, but generally a solid has a different type of ramp up to the point to where all the mechanical lash is taken up. There is typically a "constant velocity" ramp, which means the acceleration rate is zero. After this ramp, the acceleration jumps way up to get the valve moving off the seat. If the solid cam graph below, you will see a small acceleration bump at the beginning of the opening ramp, then the constant velocity portion with zero velocity, then the strong opening acceleration portion of the opening profile. The hydraulic cam shows more of one big acceleration ramp on the opening.

click image to show lift and acceleration graph for solid cam

click image to show lift and acceleration graph for hydraulic cam for comparison

The Plus version of the Cam Analyzer has some routines built in it to automatically find the lash point for both the opening and closing profile. We do both so that if you see good agreement, within a few .001", you know that the lash point found looks reliable. If there is more than .005" difference, the lash point may not be that accurate. To do this, you first select the "Seating Velocity" report type shown below:

click image to show Report Specs

Click on the More Report Settings to obtain more inputs to define the Seating Velocity report. Be sure to turn on "Estimate Lash on Cam Profile".

click image to show More Report Settings

Click OK on More Report Settings, and Make Report button on Report Specs to create a report like the one below. Here you see this particular profile is estimated to have a .0123 lash point on the opening ramp and .0129 lash point on the closing ramp. Since they agree quite closely, it can be assumed this profile was designed for about .012" lash.

Why does the Radius of Curvature jump around so much?

On base circle, the radius of curvature (ROC) is the base circle radius. Over the nose of the cam is should be less than base circle ROC. If there is some dwell over the nose of the cam, the minimum positive radius of curvature may be before or after the exact nose of the cam. On ramps where the radius of curvature goes negative, it should start at base circle ROC, then to get to a negative ROC it must go through a portion of the ramp where it is almost flat. Well, flat means a very large ROC. So it should jump up to a high positive ROC like +30, then drop to a very low ROC (flat with slight negative ROC) like -30, then come back to the minimum negative ROC as shown in the graph.

In other words, to go from positive to negative, it has to make big jumps.

(click image to enlarge it)

I'm measuring in millimeters. Why do my lobes look so "rough", like stair steps?

If you are measuring in millimeters, are you using a computer where the operating system uses a comma "," for the decimal point? An example would be the number in the United States would appear as 55,332.11, but as 55.332,11 in say Germany. This is very common outside the USA and will cause problems as shown in the pictures below:

(click image to enlarge it)

To fix the problem, go into Control Panel in your computer and find Regional Settings or Regional and Language Options. Here you can just select "English (United States)" or use the "Customize" option to make the changes needed. Selecting "English (United States)" is usually the most reliable.

How do I check if the rotary and linear sensors are good?

Encoders are NOT likely to go out of calibration. They are completely digital.

Linear Sensor: Remove the cam from the stand. Lower linear sensor so it is very close to the stand. Lay a piece of steel (ideally a 1-2-3 rectangular block) on the stand and lift sensor and slide under sensor and read result. Then do again with a different side up on the block and read result. The difference in the result should be difference between 2 thicknesses of the steel block. If the result is close to what it should be, I would strongly trust the linear encoder's measurements more than what you can measure on the rectangular block.

Rotary Sensor: Put cam on the stand and attach rotary sensor. Put linear sensor directly on base circle of cam. Mark base circle with a "shapie" pen where pointer is touching the cam. Note the linear and rotary sensor readings. Rotate cam exactly 1 rev based on sharpie mark. Are both encoders back to their original readings? Did rotary sensor just show 1 revolution? Now rotate cam 9 more revs. Did it return to the same reading? Now try 10 revs in the reverse direction.

Again, if the linear sensor (.002 inches) is off slightly, my guess it is due to some 'sticktion' in the system or lateral movement on the cam on the stand.

If the rotary sensor is off slightly (5 deg in 10 revs), my guess it is due to some slippage of the magnet on the end of the cam.

More than that could be a problem.

Can I import a lobe from 1 file into another file? or Can I combine 2 files together?

Yes you can, but the process requires a few steps. The picture below shows the option to "Add this lobe to the existing file".

for a detailed explanation.

I'm getting a "bump" in my profile when trying to do virtual follower on a cam file I've imported from a different software. Can I fix it?

Minor measurement errors can get greatly magnified with the Virtual Follower simulation, depending on where they occur on the lobe and the severity. Cam Analyzer v4.3 A.004 and later will watch for these types of problems and give you warnings and options. The picture below shows what a "bump" could look like.

for a detailed explanation of your options.

I'm measuring my OHC rocker arm cam with Intake and Exhaust lobes on the stand (not in the engine). How do I set this up in the software?

My cam is for an OHC with center pivot rocker arms with the intakes on one side and the exhausts on the other. How do I describe this layout to the Cam Analyzer software so I can measure it on the Cam Test Stand and have the lobe separation done correctly.

for a detailed explanation.

Engine Log Book Questions

What measurement gauges will load info directly into the Engine Log Book?

We are often asked about what measurement tools will communicate with the Engine Log Book, for automatic data entry. The only tool we have made special provisions for is reading the TraceBoss (tm) software files for recording surface finish. Click the TraceBoss (tm) link to see the many gauges it works with.

For Measuring Bores: QMP Racing Engines uses a Sunnen GR-2121E for the Cylinder Bores (2.0 to 6.0 inches) and a Mitutoyo 511-751 with a 543-312B Indicator (0.70 to 1.40 inches) for the Lifter Bores and other smaller bores. You would need a cable # 06AFM380F that will work for both gauges.

For Measuring Lengths, Widths, etc:

for youtube movies on other gauges inputting data directly to the Log Book.

My Mitutoyo bore gauge will not communicate with the Engine Build Log Book. Why?

A customer was using a Mitutoyo "Absolute" bore gauge with Mitutoyo U-Wave-Pak blue tooth. It did not seem to communicate with the Engine Build Log Book. If he used Windows blue tooth, it did communicate. Mitutoyo recommends also using the Mitutoyo U-Wave-R to fix this issue. Performance Trends can not verify this will always fix this issue.

Valve Spring Tester Questions

After testing my valve springs, can the software tell me the required install height to produce a certain seated force?

The Valve Spring Tester software has setting called "Find Ht at This Force". See the Test Setup Specs screen below. Note that this force can be different for the intake and the exhaust springs.

(click image to enlarge it)

After you test your springs, you can do a Report as shown below. The header section shows the "Ht for:" and lists the force for both the Intake and Exhaust. Then the Installed Height to produce this force is shown in the rightmost "Ht for Force" column. Also included in the report is the current "Seated Force" for the "Seated Height" you set in the Test Setup Specs screen.

So, for the first spring "Int 1", the actual Seated Force measured was 41.7 lbs at the specified Seated Height of 1.640 inches. The "Ht for Force" column shows that you would have to set an Installed Height of 1.649 inches to produce a Seated Force of 40 lbs.

(click image to enlarge it)

Drag Racing Software Questions

How can your Drag Racing Analyzer accurately predict Dial Ins and Throttle Stops if it doesn't let me enter several past ETs and weather conditions about my car.

Good question. It is correct that you can build up a data base of how your car performs in different weather conditions. Then the data base program will predict how your particular car will respond to different weather conditions. However, this is much easier to do on paper than it is in reality for these 4 critical reasons:

When you first get the program, you have no data base. It will take many passes to build up enough different conditions so the program can predict things accurately.
You must encounter many very different weather conditions for your data base. For example, to accurately predict the effect of headwind or tailwind, you must have passes in the data base of all ranges of wind conditions. Ideally, these different wind conditions would be at all different levels of barometer, temperature and humidity.
The data you enter must be extremely accurate. You have all made back-to-back runs where your ET changes, perhaps increased .100 seconds. Lets say the weather was exactly the same except the air temp fell 1 degree. A data base program would then assume 1 degree drop in temperature (better air) produces .100 seconds increase in ET (slower), or 30 degree drop produces 3.0 second slower ET. We all know this is totally wrong. (Also, you must never make mistakes just entering all the data., For example, entering 29.50 when you should have entered 29.05 could completely distort the predictions.)
The data you enter must only change for the reason you say it changed. If you make a change to your car (shock settings, tire pressure, engine modifications, etc), then the data base is "screwed up". The data base only sees weather conditions as producing effects in ET. Like in the example above, if you change tire pressure which slows you down .100 seconds, and the air temp dropped 1 degree, a data base program would again see a 1 degree drop producing .100 second increase in ET. Again, this is totally wrong.

To summarize, a data basing program can work if the data you input is perfectly accurate, your car is extremely consistent, and you never modify your car or driving style.

The Drag Racing Analyzer is much easier and foolproof. It uses the same sophisticated techniques used by automotive engineers (GM, Ford, etc) to predict the effects of weather conditions. These techniques are based on the physics and aerodynamics of a broad range of cars, fine tuned over decades of experience.

When you first get the Drag Racing Analyzer, you enter your car's specs into the program and save them. You can check these inputs by seeing if the program can predict your ETs, RPM range, 60 foot times, etc. This fine tunes the program's predictions for your car. For example, a pickup truck will be affected by wind conditions differently than a Corvette, and the Drag Racing Analyzer knows how differently.

To predict a Dial In, you make your first pass (time trial or practice run) and tell the program how you ran and the weather conditions. This could be with a different cam, tire pressure, converter, etc. than last week. (Many of these modifications you can enter into the program so it will be close after you make these modifications.) Entering the exact ET and weather for a pass for how your car is performing right now makes the upcoming predictions extremely precise.

Now the program is ready to predict your ET for the next pass. Just enter the weather conditions for the next pass and the program predicts the ET based only on the change in weather. As long as you don't modify the vehicle between passes, the prediction will be as precise as is mathematically possible. (Throttle Stop prediction is a little more involved and requires 2 practice runs.)

In addition to Dial In and Throttle Stop prediction, the Drag Racing Analyzer lets you try all sorts of "what if's" with vehicle, engine, transmission, converter/clutch, weather and driving style modifications. It is a complete drag racing analysis package. Say, that's a good name for it: Drag Racing Analyzer!

Do I always have to make 2 runs to "calibrate" my Drag Racing Analyzer for predicting Throttle Stops?

If you read Example 4.4 in the Drag Racing Analyzer v3.0's user's manual on page 138, it explains how to let the Drag Racing Analyzer know how your particular throttle stop, engine, vehicle, etc responds to a certain change in throttle stop setting.

A new feature added in v3.2 is the ability to predict throttle stops based the "Adjustment Factor" obtained from these 2 runs. This means you do only have to run the 2 "calibration runs" once for your car. Once you obtain the "Adjustment Factor", you simply supply it for the next series of runs at various tracks and events. Click here for the Drag Racing Analyzer v3.2's Supplement of changes which describes how this is done. Also check pictures below.

If you have the time, it is a good idea to check this "Adjustment Factor" if you have made a big change to something, different gearing, engine, major altitude change, etc. This feature just gives you an additional option.

Throttle Stop Screen from 2 "calibration runs" to obtain the "Adj. Factor

Throttle Stop Screen using this "Adj. Factor" for predicting a Throttle Stop Setting in "new air"

How do I import Drag Race DataMite data into my Drag Race Pro 'Team Engineer' simulation program?

To send data to Drag Race Pro from Drag Race DataMite:

In Drag Race DataMite, make a time Report of chosen channels you want to export in some short time interval, like .1 second.
Click on File at top to open up the ASCII file options.
Choose just "Include Text" and "Comma Separated" options at the top, nothing else.
Enter or browse to create file name, then hit Save File.
In Drag Race Pro, select File, then Import Data Logger File.
Browse to the DataMite ASCII file you created and select Open.
In the lower left corner, click on a data type from the File Data Column of available ASCII data channels from the DataMite. Then click the associated Drag Race Pro data type which would match it from the Graph Data Name section.
You will now see these data types and Data Names paired up in the Data Column Assignments at the top.
Set the other inputs on the left side, then click on OK (import file) at the upper left corner.
Now this file will appear in the History Log for you to graph with other data. Calculate ET, then click on Graph at the top of the performance results.
In the Graph Screen, click on History Log at the top, then find this imported data logger file name and put a Yes in the "Graph?" column to choose it for inclusion on the graph. Then click on "Graph Tests marked Yes" at the upper left of the History Log. You should see this DataMite data included on the graph.

How do I measure the car's Center of Gravity Height, or CG Height?

The vehicle's Center of Gravity (or CG) is the point at which you can assume all the vehicle's mass is located to do calculations. The height of the CG primarily affects weight transfer under accelerating conditions, and therefore is critical for predicting vehicle performance. For Drag Racing, typically a high CG is good to transfer weight to the rear (on the driving tires) during launch, unless you have a front drive car. For Road Racing, typically a low CG is good to minimize the transfer weight to the outside tires (and off the inside tires) during a turn, which keeps all tires more evenly loaded for good traction.

However, even though the height of the CG is critical for performance, it is not necessarily critical to enter your vehicle's exact CG height to do a good analysis. That is because for the programs, you are looking at trends. Say you enter the CG height as 20" in our 4 Link Calculator, and frame mount A produces 88% anti-squat and frame mount B produces 101% anti-squat. If you change the CG height to 19", then frame mount A produces 92% anti-squat and frame mount B produces 105% anti-squat. The anti-squat numbers change, but B still has about 13% more anti-squat for either CG height. The trend is the same.

For that reason, and because most of the vehicle's weight is concentrated in the engine and transmission, a rule of thumb has evolved over the years that the CG height can be estimated as the camshaft height in a V-8 powered car. We at Performance Trends say 5" above crankshaft height to cover all engine types and camshaft layouts.

Most all the programs which have CG Height as an input also include a "Clc" screen to let you calculate it from a test. The test is simple "on paper". You simply record the change in vehicle weight on one of the car when you raise the other end of the car a certain distance, as shown below from page 52 of the Suspension Analyzer user manual.

However, here are some critical points when running this test:

The springs can not compress or extend. Ideally you should replace all springs with solid struts adjusted to ride height lengths.
Ideally, the brakes should not be on. Chock the tires so the vehicle can not roll off the scales.
You must place the raised tires on blocks to take the scale readings. For example, you can not just raise the back of the car at the rear bumper to obtain the change in weight of the front tires.
It is probably a good idea to have scales at all 4 corners at ride height and raised, so you can confirm the total weight of the car does not change when you have the one end raised.
The more you raise the one end of the car, the more accurate the test. However, you must be very careful doing this test as the tires can easily roll off the scales.
You should "jostle" the car to eliminate any "sticktion" in the scale readings before taking them. A reading error of just a few pounds can create totally different results as indicated below.

Here's the effect of just a 5 pound difference (from 1356 to 1351 lbs, or 0.4%) in the weight change, producing a 1.3" CG height difference (from 21.5 to 20.2 inches).

I think you can see that for all the hassle and potential safety issues, a reasonable estimation of 5" above crankshaft height is much easier, faster and safer.

Why can't I just type in my converter's stall RPM and have the program always use it for "stall RPM"?

You have all seen the ads for "4500 RPM stall converters". They certainly make you think a converter has a certain stall RPM built in. Well they don't.

First an explanation of stall RPM: Stall RPM is the RPM of the engine (input RPM to the converter) when the output of the converter is stalled (not turning). Think of any automatic transmission car with the wheels locked by pressing the brakes. At idle in gear, the "stall RPM" may be 700 RPM for a production car, where the engine is producing nearly 0 torque. If you step on the gas some, the engine RPM and therefore stall RPM goes up to, say, 1000 RPM. Step on the gas some more, and the stall RPM goes up to, 1500 RPM. This is all with the same converter, but different amounts of engine torque going into the converter. This trend does not stop, but continues as you put more and more torque into the converter. Stall RPM increases as shown by the graph below.

So what is the difference between different converters, because they do change the "stall RPM" if you install a "4500 RPM stall converter" compared to a stock converter. Well, what is changing between converters is something called "converter capacity". The higher the capacity, the higher the stall RPM for a given amount of engine torque. The graph below shows how stall RPM changes for 3 very different converter capacities of 100, 200 and 300, as engine torque changes from 10 to 1000 ft lbs.

Stall RPM vs Engine Torque for different Converter Capacity factors

To be accurate and realistic, the Drag Racing Analyzer programs will ask for a converter's capacity factor. The Drag Racing Analyzer programs all have methods of entering a stall RPM for the particular engine's torque curve, and then the program will estimate the Capacity Factor. This makes it easy for you to estimate Converter Capacity. But if engine torque changes (because of engine modifications, weather, etc) the stall RPM will also realistically change just like a real converter.

for a more in depth explanation in our Motorsports Blog.

How do I adjust my ladder bars in 4 Link Calculator to prevent wheel stands?

Ladder bars have significantly less adjustability than a full 4 Link. This particular customer had mapped out his ladder bar's anti-squat settings and they varied from 138% to 188%, all quite high. With a ladder bar, the instant center is at the attach point of the bar on the frame. So the instant center length has little adjustability, but it is quite far back, not toward the front of the car which can increase the tendency of wheel stands.

So, as far as ladder bar setup goes, nothing stands out as a problem. So I pulled in a favor from our "hands on" drag race specialist Burgess Coleman and he had the following thoughts:

It sounds like his car hooks up good. What I would do is work to control the lifting rate. Here's some of the things I would do to help control the rate. I use that sensor that I got from you to help me significantly with controlling the lift rate.
Front shocks-Tighten the extension knob so to slow down the lift rate. Go to the point where it starts to affect traction.
Timing kill-I use that a lot to limit that initial surge of power and it won't hit the converter so hard. I activate mine for 0.8 seconds and up to 14 degree retard. This is really briefly and affects performance only slightly.
Weights-Place a ten pound or so weight at the front most part of the car. That small weight will generate 100 or more pound of torque to resist lift. Ten pounds won' t really affect performance much. Couple thousands.
This would be my first steps. After that I would put a switch on the front suspension and go into Two Step mode or some other power reduction mode. (At a certain height)

Again, many thanks to Burgess for great ideas. You'll see his expertise in other FAQs on this page.

My 4 Link Calculator does not draw my ladder bars correctly. Does that matter?

Ladder bars are assumed to be welded (or attached) solid to the axle. Because of that, the bar's shape has no affect on its function. For example, in the picture below, it shows the 4 Link Calculator's standard drawing of a ladder bar at a particular hole on the frame, the 2 blue bars which meet at a single point. Bars shaped like the red bar or green "J" bar are exactly the same as far as suspension performance is concerned. All that matters is where the bar attaches to the frame.

Suspension Software Questions

I don't know which suspension program to get. Can you help? How about Dirt vs Asphalt?

We have 2 "families" of suspension analysis programs, Circle Track Analyzer and the 3D Suspension Analyzers. for a side-by-side comparison of features in these programs.

Our Roll Center and Roll Center Plus are all based on the Circle Track Analyzer with portions of the Circle Track Analyzer turned off. Circle Track Analyzer with all features removed but the Front Suspension screen is the Roll Center Calculator. Circle Track Analyzer with all features removed but the Front Suspension and Rear Suspension screens is the Roll Center Calculator Plus, which also does simple handling analysis (how tight or loose a particular setup is). These 3 programs do the analysis with just simple 2D inputs, which treat the front suspension as if it was drawn on a flat sheet of paper. Because of that, these programs will NOT do bump steer, Ackerman, caster, toe in changes and roll steer and anti-squat in the rear, and many other calculations.

But, the full Circle Track Analyzer can actually "drive" the car around an asphalt track you specify and estimate lap times and how lap times are likely to change with vehicle modifications. It is our only suspension program which does this type of lap time simulation and analysis.

Our other family of programs is the much more detailed 3D Suspension Analyzer. It comes in 3 versions, Standard (front end only), Full Vehicle (front and rear suspensions), and Full Vehicle with Data Logger features (with extra features including ability to read and "play back" data logger data). Because is it 3D, it requires 3 measurements (out from centerline, up from the ground like the 2D programs, but also "depth" into the car) for each suspension point and uses many more points than the simpler 2D Circle Track Analyzer. The Suspension Analyzers WILL do bump steer, Ackerman, caster, toe in changes and roll steer and anti-squat in the rear. However, it also requires you to make more measurements to use it.

If you know what you are looking for, like "I want my Camber Gain to be 2 deg per inch of dive", or "I want more rear roll steer", you need a version of the Suspension Analyzer program. If you are an ASPHALT circle track racer, not sure of what a program can do for you, and want a program to make suggestions of things to try, we recommend the Circle Track Analyzer. If you are an DIRT circle track racer and fairly new to computers, we recommend the Circle Track Analyzer or maybe Roll Center Plus.

Unless you know exactly what you are looking for, we typically do not recommend a program which just deals with the front suspension only. That is because a program can not analyze a setup (give a "handling rating") without knowing something about both ends of the car.

For Dirt vs Asphalt

Again, we have 2 "families" of suspension programs, Circle Track Analyzer and 3D Suspension Analyzer.

The Circle Track Analyzer actually tries to predict lap times, dynamic corner weights, dive and roll, and much more. For it, all assumptions for lap times, cornering traction, etc are based on asphalt. We have many users using the program for dirt, and reduce the "Traction Factor" in the Vehicle specs to numbers like 80% or so and can simulate their dirt times quite well.

Circle Track Analyzer makes for doing it's lap time simulation and handling analysis based on these 2 assumptions:

- The car is fastest if you do NOT break the tires loose.

- The car is fastest if you most evenly load all 4 corners of the car, to not overload any particular tire.

If you agree with these 2 assumptions, then the Circle Track Analyzer can do a good analysis of your dirt car. But we have all seen winning dirt racers with 1 tire off the ground, sliding the rear end out, using the tire's "side bite", etc. So on dirt, there are many ways to win on a particular track and surface.

The Suspension Analyzer just calculates all the suspension geometry in great detail, like roll center, camber gain, roll stiffness, roll steer, bump steer, etc as the suspension goes through dive, roll and steer. Those calculations have nothing to do with the track surface. However, what kind of geometry you WANT (roll center, camber gain, roll stiffness, etc), CAN change from asphalt to dirt. You have to decide if your modifications will change the suspension geometry the way you want.

for another FAQ about Dirt vs Asphalt.

In the Roll Center Calculator and Circle Track Analyzer, how do I measure where the Frame Mounts for the A Arms are located?

We often get this question. This is a limitation of doing a front suspension layout in 2 dimensions (like a flat sheet of paper) compared to doing it in full 3 dimensions (3D) like done in our Suspension Analyzer. This is best explained with the picture below.

Draw a line connecting the left and right ball joints. We call this the "axle line". Draw a line connecting the front and rear frame mounts. Where these 2 lines intersect, that is what the 2 D programs want for a frame mount point. Both the X Right/Left (distance out from the car centerline), and Y Height (distance up from the ground) are measured at this point.

(click image to enlarge v3.6 "More Details" screen shown here)

Click on blue link to view the "More Details" screen available in v3.6 of Circle Track Analyzer and Roll Center Calculator to precisely locate this single point based on the 2 actual mounting points.

Do you have software which deals with Rear Panhard Bar settings?

Our Circle Track Analyzer v3.5 or newer will automatically pick a panhard bar height (actually rear roll center height) to produce certain changes in handling to produce a "balanced" condition. This is done by knowing other vehicle characteristics like front roll center height, front roll bar specs and front and rear suspension geometry. From this it calculates the Front Lateral Load Distribution or FLLD. This number along with front to rear weight distribution is useful to determining how tight or loose the car will feel at the apex of the turn.

These features are found by clicking on the Details and Find buttons on the main screen in the " 'Transition' Handling Rating " section.

Why doesn't Circle Track Analyzer showing a change in lap times when I change suspension settings?

Which suspension settings are optimum and how they affect lap times depend on many things, many of which we just don't know. These include the track surface, driver preference, tire performance curves, etc. We are not able to predict changes from suspension settings ACCURATELY enough at this time. Therefore, we believe it is better not to try and to show NO effect at all. You will see the demo or program make this VERY apparent in the Suspension Specs menus. We would like to be able to predict all effects accurately for anything you can change in the program, we're just not there yet.

When the Roll Center gets close to ground level, it moves left or right dramatically for small changes in dive and/or roll. Is this correct?

The program IS calculating the Roll Center position correctly MATHEMATICALLY. When both instant centers are at approx. ground level (0.0 height), minor changes will move the Roll Center left or right considerably, but not change the height much. Most circle track cars are set up so the instant centers and Roll Center stay well above ground, so this does not become an issue.

This brings up a debate about Roll Centers. How critical is the position left or right. Some "authorities" only use the Roll Center HEIGHT to determine handling characteristics. Most experts agree that asymmetries in spring and wheel rates left and right have a large effect on how the body rolls. This contradicts the simple Roll Center theory that says the body rolls about the Roll Center, and spring rates have no effect. Performance Trends is simply showing you the mathematical roll center and letting the user decide how to use it.

Can the Circle Track Analyzer be used for dirt track racing?

Many aspects of the Circle Track Analyzer are equally applicable to asphalt or dirt. For example, Roll Center, Camber Gain, etc only depend on the front suspension layout, not the track surface. However, what Roll Center or Camber Gain you decide to run is your decision.

The program's lap times are based on the basic theory that you get your best traction (fastest lap times) right before the tires break loose and start spinning. This is definitely true on asphalt, and on many types of dirt tracks also. However, the program does not simulate any conditions where you are breaking the tires loose and "steering with the throttle". There are also some recommendations (springs, stagger, etc) the program can calculate and these ARE for asphalt.

for another FAQ about Dirt vs Asphalt.

The programs let me put in Dive and Roll to watch what happens when the suspension moves. How much should I tell the program?

This is a common question. First, this is an additional feature we offer. You don't have to use it. Suspension programs from other companies only do the static calculations at ride height, because putting things in motion is much more complicated.

Typically what you want is the roll center height not to move around too much. Ideally the roll center goes up and down with dive. For example if you go into 1" of dive, the roll center height will also move down 1". This not an exact rule, but a trend to look for. So you put the suspension though various amounts of dive and roll and watch the general trends in roll center position.

The only real way to know the exact dive and roll of your vehicle is with shock sensors and a data logger, like our DataMites, but that is expensive. Reading commercial shock travel indicators can give you some idea of what's happening on the track because they indicate maximum travel. (Using tie wraps tied to the shock stems is another method, but that only shows you max compression.) However, if the right side is down 1" and the left side is up 1" at the same time, that is a lot of roll and not much dive. If the right side and left side are both down 1" at the same time, that is a lot of dive and not much roll.

If all you have is shock travel indicators, I would recommend moving the suspension through all extremes of those travels, making sure you put the car through the most roll possible with max down on one side with max rise on the other. If you have coil-overs, watch the shock/spring compression in the program to match your travel indicators. In the picture below, with separate springs and shocks, watch the shock travel as shown by the blue arrows, negative means compression.

Realistically, the actual roll is about half of going to these extremes. If max rise on left is 1" and max dive on right is .75", and the roll from this condition is 3 degrees, realistically the car probably goes through only 1.5 deg roll.

If you know nothing about your car, use these suggestions for roll:

Circle Track Big Bar, Soft Spring (BBSSS) setup on flat track: 1.0 deg roll
Circle Track Conventional setup on flat track: 2.0 deg roll
On high bank tracks, you will likely see less roll.

For dive estimates, most circle track cars are setup to let the car dive a specific amount, either to the bump rubbers, or to within a certain distance to the track. Use this difference in ride height to be the amount of dive you enter.

Can you explain the different labels for the 5 Link rear suspension points?

The 5 Link rear suspension type has 10 linkage points, so it can be quite confusing what each point is called. The 5 points mounted on the spindle (which the wheel bolts to) are called "Ball Joints", except for the toe link which is called "Tie Rod on Spindle". The other 5 points are called out as "on Frame" because they mount to the frame or chassis. Figure A 6.17 on page 153 in the manual shows this suspension type with some explanation. Here is a portion of that Figure with the 10 labels used for each point.

click image to enlarge it

What should I use for the vehicle's centerline when I measure my suspension?

When you measure you suspension, the X measurement is measured out from the centerline of the car. Many users ask "What should the centerline be?" It could be the engine and driveline's centerline, midway between the 2 frame rails, midway between the 2 tire patchs, and many more.

The most proper answer is midway between the 2 tire patches. However, if you use this, and change wheel details like offset, tire size, camber, etc. this centerline will change. Luckily, our suspension programs have edit features to move all the suspension points left or right with one command. They also have a handy "Center Car based on Tire Tracks". This one command lets you recenter all the measurements if the track from left and right are different.

Now for some reality to make your life easier. If you move the points, say, 1" to the right, the only calculated result which changes is the roll center location left/right (not height). Camber gain, ackerman, bump steer, roll steer in the rear, and most everything else does not change. And, a 1" or 2" change on roll center location left or right is not a significant effect for handling. The height is much more important as it affects the jacking forces.

So, what this means is we recommend you use something which is very stable in the car, like the midpoint between the frame rails, or the driveline centerline. Ideally the car is square so that either of these center points at the front of the car and close to being the centerline at the rear of the car. Then if you change camber or tire offset, you do not have to change any other measurements, just the Track measurement. But, should you ever want to change the measurements, to center them on the tire patches, the single Edit command should allow you to do this.

In the picture below you will see the right track is 36.63" and the left track is 30.63 inches (bottom of pic), so the front of the car is not centered on the tire patches. If you click on Edit, then Move Groups of Measurements in the new screen, you will see what the program will do to center the vehicle on the tire patches. Click "Move Points" button to have this modification performed. Note: The program will not automatically center the other end of the car. You will have to make a note of what happened at the front of the car, then move the points the same direction and amount at the rear of the car.

click image to enlarge it

How can I graph data from my front suspension with that from a rear suspension?

The program is designed to graph data from the end of the car you are working with on the main screen, before you click on the 'Graph' feature in the upper left. So it will not graph rear suspension data with front suspension data.

However, you can use a powerful feature to do a "Selective Open" of the rear suspension file and open the Rear Suspension of this file as a Front Suspension, and then do the graphing. First, create a new file name that will hold the Rear Suspension data as a Front Suspension. Click on File, then New.

click image to enlarge it

In the "Start a New Suspension" screen, enter a name for this file.

click image to enlarge it

Back at the main screen, you will see this is the new name of the file you are working with. Click on File, then "Open (from all saved suspensions)". Find the file with the Rear Suspension data you want to graph with the other file's front suspension. Check the "Yes" box for "Selective Open" and follow directions in picture below.

click image to enlarge it

Now this rear suspension data is on the front end of the car for this new file. Now you can graph the "front" suspension of this file, with the actual front suspension of the other file, as shown below.

click image to enlarge it

Note: The "Selective Open" feature is very powerful. This FAQ just shows one handy thing you can do with it.

Dyno DatateMite Data Logger Questions

I'm building an inertia dyno myself. How do I figure out how big to make the flywheel, or the roller for a chassis dyno?

Performance Trends has created 2 simple programs which let you approximate the size of flywheel for an engine inertia dyno, or roller(s) and any additional flywheel for a chassis dyno. This is the best method to get you close to the approximate sizes and weights.

for the program to estimate a flywheel for an "engine only" inertia dyno.

for the program to estimate a roller size (and any additional flywheel) for a chassis inertia dyno.

Diagram of some of the measurements used by these programs.
click image to enlarge it.

Some of the assumptions used by these programs include:

The engine accelerates through 50% of its RPM range for its test. For example, if you run your test up to 6000 engine RPM, it assumes you started at 3000 RPM.
The programs assume no losses in the driveline, engine inertia effects, etc. This means the acceleration times will likely be somewhat longer than what is calculated.
You need an acceleration time of 5 to 10 seconds for good accuracy.
If you tell these programs you will run the dyno to 2000 RPM or 140 MPH, then you must do that to obtain the acceleration times calculated. If you run the test to a lower RPM or MPH, the acceleration times will be significantly shorter, producing less accurate results.

----------------------------------------------------------------------------------------

If you want to go into more detail, then you need to do the following 4 steps using the Dyno DataMite 10 day demo program from the link below:

to install the Dyno DataMite demo program

1) Run DataMite program and open an example data file with an inertia dyno. Click on File (upper left corner), then Open from All Saved Tests. Click on the Examples folder in the lower right corner, then Black Box II in upper left corner, then Open button in lower left corner.

2) Click on Dyno at top of main screen.

Make sure "Sections in Main Wheel" in upper left is set to 1

Make sure "# of Main Wheels" in upper right corner is set to 1

For "Main Wheel, section 1", enter 0 for Inside Diameter, Material to Steel. Then set a Outside Diameter and Width (len) for a flywheel you would consider building. Program will calculate Weight and Inertia for this
flywheel. Especially note the Inertia.

(For chassis dynos, be sure to set one of the Chassis Dyno types in the lower left corner.)

3) Click on "Est. Required Inertia" at the top.

Enter the Average Torque for the highest HP engine you are likely to test. Estimate the Average Torque as follows:

    Average Torque = Peak HP x 5252 / Peak HP RPM

For example, if your engine makes 75 HP at 8000 RPM, it would be:

    Average Torque = 75 x 5252 / 8000
    Average Torque = 49 ft lbs

Enter the Starting RPM and Finish RPM for the proposed dyno run, and the Desired Accel Time in seconds. For you highest HP engine, you should probably enter 6 seconds, so that for your smaller engines the accel time
would be longer than 6 seconds. Ideally, you want the run to be around 8-10 seconds.

Enter possible Total Gear Reduction to keep the Maximum RPM (Calculated at the top) low and safe, and the Safety Factor high (ideally 8 or higher).

(For chassis dynos, a calculate vehicle speed will be displayed also. Adjust the Total Gear Reduction so that you get the approx MPH/KPH for the gear you will be testing at for the highest RPM for the test. The faster you can safely test at, the lower will be the required inertia.)

In the top section you will see the accel time for your current flywheel, and now much inertia (not weight) is required to get the accel time you want.

4) You may have to go back and forth several times to come up with combinations that work for you.

Back to Top

I'm having trouble getting my USB DataMite to "talk" to my computer. How can you help me?

Getting a USB device to communicate typically requires a USB driver to be installed. This driver is on the Performance Trends blue CD.

Click on this blue link to DataMite III Installation Tips.pdf for detailed info with pictures.
Click on this blue link to DataMite Software 1.wmv for a narrated, movie about getting the DataMite USB to "talk".

When I start my engine (or go to full power), I lose communications with my data logger. How can I troubleshoot this?

If you have good communications without the engine running, but lose communications when the motor starts or goes to full power, the reasons are likely:

Vibration is causing intermittent connections in the cables, most likely the USB or serial cable connecting the logger to the computer. This could be the case if the logger box is mounted on the dyno stand itself, which is not a good practice if it can be avoided. Exposing the data logger to excess vibration is just asking for trouble eventually.
Electrical noise from the ignition system is creating "confusion" in the data logger or the PC. Here's a list of things to try to troubleshoot electrical noise.
When troubleshooting, it is important that we are trying to temporarily find the source of the problem, and not the final fix yet. You may think what we suggest is not a good permanent fix, but lets first diagnose the problem The ways to troubleshoot the source of the electrical noise include:

- See if you can watch the current readings screen for 2 minutes or more with no loss of communications. This is to ensure the problem is actually caused by the motor running.

- Disconnect some or all the sensor cabling from the data logger box and see if you can start the engine now and not lose communications. Be sure to mark the connections so you plug them back into the proper connector. For example, on a DataMite III or 4, you can disconnect Analogs B breakout cable and eliminate up to 5 sensors at once. If the data logger keeps communications now, plug in the connectors 1 or a few at a time to narrow down the problem connector or sensor. Any sensor or cabling which goes close to an ignition component like a coil, spark plug, ignition wire, distributor, etc could pick up noise spikes and "confuse" the data logger.

- Check the Power Volts in Current Readings to ensure it is around 11-17 VDC. If it is not, investigate why.

- Move the PC farther away from the engine, especially if the PC gives an error message that seems unrelated to the DataMite program. The PC should be a minimum of 5 ft away from the engine, but 10 or more ft is better.

- Move the data logger farther away from the engine. The data logger should be a minimum of 5 ft away from the engine, but 10 or more ft is better.

- Reduce the size of the Current Readings screen if it is maximized to fill the entire screen. The larger the display, the more computer resources it takes to run everything. for info on a related FAQ.

- Click on Options at the top of the Current Readings screen and reduce the Update Rate (probably at 10) to something slower. for info on a related FAQ.

If you still are having trouble, call or email Performance Trends for assistance. Note: If this is a system more than 1 year old, there may be a nominal tech help charge.

for info on a related FAQ.

Do you have classes to teach user's all of the DataMite software's features?

We don't have classes. However, we do have about 20 narrated, movies for you to watch, 6-12 minutes each. Click on the blue links to:

How do I hang weights on my dyno for calibrating the load cell for torque?

Appendix 5 in the Dyno DataMite Users Manual (click on blue link) provides good information on calibrating your load cell to read torque, or calibrating most any sensor. It involves hanging a known amount of weight a known distance out from the center of the dyno. However, there are some details the manual overlooks.

1) How do I account for the torque put on the dyno by just bolting the calibration arm in place?

The manual talks about using a balanced arm which puts no torque on the dyno when installed, or an arm which is permanently attached (which puts an offset on the torque, but the calibration procedure corrects for the arm). However, most people bolt their calibration arm to the dyno only for calibration, and then remove it. If this arm is very long, this amount of torque can be very high, 50 to 100 ft lbs or more. A long arm has the advantage that you do not need a lot of weight to generate a lot of torque.

To determine the amount of torque put on the dyno from the arm, you must first determine the center of gravity (CG) of the arm. This is done by finding the approximate location where the arm will balance. See figure below. This does not need to be done exactly, as that is quite difficult and time consuming. Accuracy within a quarter inch is typically sufficient. Mark the CG location on the arm. Then weigh the arm.

(click image to enlarge it

The amount of torque from the arm is the distance the CG is out from the dyno in feet multiplied by the weight of the arm. In the figure below, with the arm mounted to the dyno, the CG measures 34 inches out from the center of the dyno, and the arm weights 79 lbs.

34 inches / 12 = 2.83 feet (12 inches in a foot)

Torque from Arm = 2.83 feet x 79 lbs = 223.8 ft lbs

(click image to enlarge it

Then for the higher torque, you hang 320 lbs (four 80 lb plates) out at a distance 55.5 inches from the center of the dyno

55.5 inches / 12 = 4.625 feet (12 inches in a foot)

Torque from Weights = 4.625 feet x 320 lbs = 1480 ft lbs

For this higher torque, you must then add the torque from the weights to the torque from the arm for the total calibration torque

Higher calibration torque = 1480 + 223.8 = 1703.8

These would be the torques you would enter into the Dyno DataMite's sensor calibration screen for torque, as shown in figure below. The voltages shown on the calibration screen are obtained by clicking on the Read button for the 2 conditions, arm by itself producing 223.8 ft lbs, then arm with weights producing 1703.8 ft lbs.

(click image to enlarge it

Because it may be difficult to fabricate a long arm, or you may not have room for it, or you do not have much weight to hang, you may be tempted to only produce a small amount of torque during the calibration. For example, you test engines in the 500 ft lb range, but you only calibrate by hanging 60 lbs on the dyno, 1 ft out. See figure below. In this situation, the arm by itself produced 20 ft lbs. Then the arm plus 60 ft lbs produced 80 ft lbs. This is much lower than the 500 ft lbs you test at.

In a perfect world, with no errors, this would work fine. The calibration in the 80 ft lb range would extend out to the 500 ft lb range with no problem. However, in the real world, every data point has some error in its reading. The figure below shows the calibration points of 20 and 80 ft lbs with some error, and how that error gets multiplied many times by the time you get to 500 ft lbs. However, when the higher torque calibration point is in the 500 ft lb range, the calibration error in the 500 ft lb range is much less.

This is the same reason the sights on a rifle are separated by as much distance as possible. If the sights were only separated by 3", the aim and accuracy for hitting a target at 100 yards would be 10 times worse than with the sights separated by 30".

(click image to enlarge it

Notes:

There is nothing "magic" about having the weights hang exactly 1 ft out. Any distance will do as long as you know it exactly.

There is nothing "magic" about using something called a weight for calibrating. A cylinder head which has been weighed on an accurate scale works just as well. However, make sure your weight can not change. A head with valves may weight 60 lbs today, but if someone takes out the valves and springs, may weigh 57 lbs the next time you use it.

Pay attention to the volts generated by the 2 calibration points. If they are only separated by 0.5 volts or less, your calibration is not as accurate as could be. Also, if the volts at the higher reading is 4.5 volts or higher and you test at a torque in the higher torque point range, you may be starting to max out the signal. Due to the pulsing nature of the torque signal, 500 ft lbs can be pulsing from 400 to 600 ft lbs. If 550 ft lbs is equal to 5 volts, all torque values greater than 550 ft lbs are being lost and your torque reading is reading too low.

My torque reading is not reading zero with the engine not running. How can I rezero the torque (or most any analog channel) reading?

First, was your torque reading zero correctly before and now it is not? If yes, check the following:

With the engine not running, pull up on the load cell arm and let go and check torque reading. Then push down on the load cell arm and let go and check torque reading. Do they repeat? If not, there could just be "sticktion" in the system, which typically goes away once engine is running (lots of vibration). The actual zero is likely half way between these 2 readings. To see what the actual zero reading is with the engine not running, start the engine with the dyno disconnected so the dyno sees no torque and see what zero torque reads. for FAQ below describing how to calibrate with sticktion in the system.
If you go back and hang a weight, is the upscale torque off by the same amount as zero? For example, if zero is reading +10 ft lbs when it should read 0, and you hang weights to produce 450 ft lbs, does it also read 10 ft lbs high, or 460 ft lbs. If yes, then it makes sense to rezero per the attached procedure?

to read how to "rezero" the sensor through the software.

If the torque reading is not reading zero correctly after you have recalibrated (hung weights), check the following:

Did you calibrate your system (hanging weights), as described in Appendix 5?
Did you calibrate your system (hanging weights) with a load arm installed, and have now removed the arm? If yes, this would change the calibration you arrived at with the arm installed.
When you calibrated, did you click the "Read" button with both zero torque (or some low torque number) and with a high torque? You need to read 2 torque settings to calibrate, one very low (typically zero) and one high.

for details on how to correctly hang weights to calibrate your dyno.

When hanging weights to calibrate torque, the numbers can change by how I hang the weights. How can I be more accurate?

To calibrate the load cell on a brake dyno, you will hang weights as described in the FAQ above. If you find the voltage the computer reads can change depending on the "sticktion" mentioned in the FAQ above (push down, let go and read voltage, pull up and let go and read voltage), you should first see if you can find the source. for FAQ above describing sticktion. With a hydraulic load cell, the seals in the hydraulic cylinder can cause sticktion because the piston must move slightly to produce pressure. With an electronics load cell, there is typically only a few thousandths of an inch of movement with load so sticktion should be less of a problem. See if you can find the source, maybe binding in the mounting ball joints, perhaps the cable, maybe movement in a rubber mount, etc.

If you can not find and eliminate the sticktion, the next best thing is to try to calibrate in the middle of the sticktion range, which is most likely the signal produced when the engine is running and vibrating the load cell. This is done as follows:

With zero torque on the dyno, push down on the load cell arm and let go. Click the Read button for "1st Value, volts" and write down the voltage. Then pull up on the load cell arm and let go. Click the Read button for "1st Value, volts" and write down the voltage. Do this pattern 2 more times for a total of 6 readings. If there is sticktion, the 3 pull up voltages should be very close together and the the 3 push down voltages should be very close together, something like: pull up readings of .288, .286, .289 and push down readings of .298, .297, .300. Average all 6 readings together and manually type in the resulting voltage of .293. See picture below.
If you got a set of 3 voltage readings for pulling up of .288, .286, .308, you would see that the .308 was way different than the other 2 and was even higher than the 3 pushing down readings. This is probably a bad reading and you should take another pulling up reading to replace it.
Then hang the weight for the "up scale" torque reading described above and go through the same process. Manually write down the 6 readings, check that the pattern repeats and the readings are very close together, average all 6 readings and manually type in the voltage.

How do I know what A/F to shoot for when I change fuels?

Air to Fuel Ratio (A/F) is a ratio of the mass of air burned in the engine to the mass of fuel. Different fuels require a different A/F because of the chemical composition of the fuel. For example, methanol engines require about 2.5 times as much fuel as a gasoline engine requires.

Oxygen sensors don't actually measure the A/F ratio, they measure the richness of the mixture. In a gas engine, 15% rich (a good rule of thumb for best power) is about 12.5 A/F. In a methanol engine, this is about 5.5 A/F. Both engines are going to make good power at about 15% rich. Because different fuels require different A/F, and especially with the advent of the ethanol blends with gasoline, E10 (10% ethanol at the pump), E85 (85% ethanol at the pump), etc., a different measure of richness was developed. It is called Lambda, or the excess air factor. A lambda of 1.1 means there is 1.1 times as much air as required, or 10% more. A lambda of .85 means there is 15% less air than required, or there is about 15% excess fuel. A lambda of .85 or 15% rich is also a good number to shoot for for best power.

See the chart below for an explanation of different A/F for various fuels.

(click image to enlarge it)

So, what do you use for the A/F calibration for your sensor for various fuels? If you want to keep it simple, keep the A/F calibration set to the gasoline calibration and shoot for about 12.5 A/F even if running different fuels. When you are running methanol, the "true" A/F will be more like 5.5 when the gasoline calibration says 12.5, but 12.5 using the gasoline calibration is still 15% rich.

Think of A/F as room temperature. A room temp of 35 deg C sounds very cold to us Americans, until you convert it to its equivalent of 95 deg F. Both feel equally hot, just measured on different scales.

Now if you want to be more correct, and if a Lambda calibration is available, start reading A/F in lambda. This will be easier to understand because it is constant over all A/Fs, you want to shoot for about .85 lambda. And you don't have to worry about the "know it all" looking over your shoulder saying your 12.5 A/F on the gasoline scale is all wrong for methanol fuel.

If you always run the same fuel, pick the calibration for your fuel and keep it there and shoot for the best power A/F for that fuel from the chart above. If you are very familiar with the proper A/Fs for different fuels, then change the A/F calibration for each new fuel you run and shoot for the best power A/F for that fuel from the chart above.

The Dyno DataMite v4.1 has the correct calibrations for different fuels, and lambda, pre-programmed for the various oxygen sensors we use for easy choosing.

Can the DataMite software use vehicle coastdown data on a chassis dyno to estimate driveline losses, and therefore produce flywheel torque and HP numbers?

The Enterprise Edition of the Dyno DataMite v4.1 vA.040 or later provides an additional test Type of "Measure Tq/HP from accel/decel". This test Type is only for chassis dynos with a test run consisting of an accel followed immediately by a decel with no braking of the vehicle. This coastdown info after the accel is use to estimate vehicle and dyno losses and better estimate flywheel torque and HP.

The graph below shows what this type of run looks like, with Dyno RPM vs Time. Note that the accel took only about 6 seconds but the coastdown portion took over 60 seconds.

When you identify a test like this and make a run as described, there are 4 new types of Graph Data types available:

Coastdown Torque
Coastdown HP
Total Corrected Torque
Total Correct HP

Total Corrected Torque and Total Correct HP are the estimated values at the engine flywheel. Coastdown Torque and Coastdown HP are the losses calculated from the vehicle coastdown portion of the test.

Total Corrected Torque = Corr. Flywheel Torque + Coastdown Torque

Total Corrected HP = Corr. Flywheel HP + Coastdown HP

where Corr. Flywheel Torque and Total Flywheel HP are the numbers typically reported from the chassis dyno results. "Flywheel" means this is not what is recorded on the actual roller, which would be similar for the HP, but very different for the torque because of possible gear multiplication and roller vs tire diameter "gearing" effects.

Here's the screen for picking these new data types. Note the "Corr. flywheel HP (after losses)" is also picked but farther up the list.

Here's what the results look like, using HP as the data type.

Notes:

As shown in the RPM graph above, it is best if the accel portion and decel portion of the test extend through the same RPM range.

If you are doing this type of test, the Coastdown data in the Dyno Specs to estimate just the dyno losses are not used. That is because these losses are also determined during the coastdown to determine the vehicle losses.

Doing a coastdown test of the vehicle you just tested to estimate the losses in the driveline, to then come up with flywheel HP can be misleading. The reason is because the losses with power ON are much higher than when coasting, especially for an automatic transmission. For example, if the losses from coastdown at 4000 RPM show 32 HP, the losses with full power will be much higher (perhaps 100 HP for a 500 HP engine) when accelerating. Tire slippage will be 3-10%, which equates to a direct loss of that percentage in power. (During coastdown, tire slippage is 0%.) The difference between full power and coasting gear box losses are the difference between power-On efficiency and simple "spin losses". Therefore, the Total Correct Torque and Total Correct HP are still not as high as what is happening at the flywheel. But because several users have requested these results (because other dyno systems perform them), we have also provided them.

When trying to achieve accurate inertia dyno results, what are the main things to consider? By accurate, I mean as close to true horsepower as possible.

Correct for weather conditions to some "standard weather day". Engine power changes significantly with weather conditions. You should record the weather conditions PROPERLY and correct for them, typically done automatically with software. Also, only compare results that are corrected to that same "standard weather day".
Accurately account for all the inertia in the system. This includes the inertia of the engine (crank, flywheel, etc). Again, this is best done in software.
Accurately account for all friction in the system. Friction includes bearing friction and windage of the inertia wheel spinning in the air. This can be measured by doing a coastdown test, spinning the flywheel system up and then letting it coast down with the engine disengaged. Again, this is best done in software. Note: A coastdown test does not account for all friction. The friction losses when power is ON is higher than when just coasting down. However, it is the best estimate we have of these losses. Also see FAQ below, "...repeatable inertia dyno results?"
Use a clutch that locks up well below the RPM where you want accurate data. If you want good torque and HP data at 3000 RPM, use a clutch that locks up at 2000 RPM.
Try to rev as high as the engine will safely live. If you want accurate data at 10,000 RPM, you will need to rev to 10,500 or higher.
Use moderate acceleration rates, like 300-500 engine RPM per second. Very quick acceleration rates, like 1000 RPM per second are less accurate.

What are the main considerations when trying to achieve repeatable inertia dyno results? By repeatable, we mean test-to-test and day-to-day consistency

Correct for weather conditions, as described above.
Always start your run with the same engine temperatures. To most kart builders, this means head temp. However, block temperatures should also be included.
Always do the same acceleration rate. With an inertia dyno this will happen automatically if you use the same gear ratio between the engine and dyno, use the same clutch (lockup RPM), and same starting RPM.
Because weather correction factors are not perfect, always try to run critical comparison tests on the same day, same engine, same operator, etc. Don't try to say Cam A is better than Cam B because Cam B got better results on one day on a different engine than Cam A.
Try to keep frictional losses as low and constant as possible.
- Flywheel bearings should be the same approximate temperature. Cold bearings in the winter (before your shop warms up) have higher losses than summer bearings where the shop is always above 80 degrees.
- For critical measurements, do not change the gear, chain, or belt ratio. In a perfect world, this should not matter. However, it appears that changing ratios can effect frictional losses.
- Keep the chain lubed and in good condition if chain drive.
- Keep the belt tension constant and the belt in good condition if belt drive. V belts are typically less repeatable than flat, toothed belts (like timing belts).
Do repeat tests. Try Cam A, then Cam B, then go back to Cam A to see if it repeats, then go back to Cam B. This is a lot more work, but it is the only want to get a repeatable measure of what Cam B gets over Cam A.

for more tips for testing peaky engines.

What are the main considerations when trying to achieve accurate and repeatable brake dyno results?

for tips to get accurate results for an inertia dyno. Most all these tips also apply to water brake dynos.

for tips to get repeatable results for an inertia dyno. Most all these tips also apply to water brake dynos.

for tips about checking the zero reading on your dyno before testing.

Check if there are things putting torque on the dyno other than the engine. For example, in the picture below, the water hoses are going out to the side of the dyno. As the pressure in or temperature of these hoses change, the hose will stiffen or go slack, changing the torque preload on the dyno from the hoses. The same applies to cables. If someone moves the cables or hoses, it can change the preload torque on the dyno.

click image to enlarge

A better system for the hoses and cables would be for them go go straight back and be supported at a point behind the dyno. Then changes in the hose or cable forces would only put thrust loads on the dyno. Note: The hose or cable support should allow for some movement in the hoses and cables, so they do not inhibit any rotation of the dyno itself. See pictures below.

Click here to increase image of Land&Sea dyno with hoses to the side, changing torque with changes in water flow and pressure. This introduces errors in the torque reading, especially when doing small motors.

Click here to increase image of Land&Sea dyno with hoses rearranged, nearly eliminating effect of hoses on dyno torque and improving dyno accuracy.

With an inertia dyno, the acceleration rate is typically quite smooth and repeatable because of the high inertia of the dyno. However, with the low inertia of a brake dyno system, you need a repeatable control method to do smooth and repeatable sweeps through the RPM range. Good control can be done manually if you have good experience. Otherwise you will need some type of controller, like Performance Trends' Dyno Controller.

To help compensate for different acceleration rates from test to test, you can correct for engine inertia effects. for an FAQ on using engine inertia correction. for the Dyno DataMite user's manual where you can check for "engine inertia" in the index or doing a word search for that phrase.

for more tips for testing peaky engines.

When I dyno an engine using an Inertia Dyno and your DataMite, the HP seems to peak too early, say 9,500 RPM, when I actually rev the engine to 10,500 RPM on the track. Why?

There are several reasons:

For best performance you do want to rev the engine past the HP peak. For best performance, you want to keep the engine in the highest HP range available. This usually means revving past the peak, so the lowest RPM on the track is still near the HP peak.

The dyno can only show performance for the RPM range you have tested. In simplest terms, it can't show you torque and HP at 10,000 RPM if you don't rev it to at least 10,000 RPM.

Most all dyno data looses accuracy near the point where you open or close the throttle. This is usually the start and the end of the run. One reason can be due to not knowing exactly where the throttle closes. Another is that in the process of filtering (smoothing) data, the data happening a small time before and after some event gets averaged in with the event. This means that data at 9800 and 10,200 RPM gets averaged in with data at 10,000 RPM. Since 10,200 RPM may be very low power (closed throttle), it will bring down the average data at 10,000 RPM. This filtering effect is even more pronounced with inertia dynos, which must also diferentiate the data. This means that data at 9,500 RPM and 10,500 RPM gets averaged in to produce the result at 10,000 RPM. (The RPMs above are not exact, but are just examples to illustrate the point.) The point is, to accurately see what is happening at, say, 10,000 RPM, you should rev the engine well past 10,000 RPM during your dyno test. Note, however, you must consider the safe RPM limit for both your engine and dyno !

The Torque Peak and HP peak reported on the Main Screen is slightly different than what I see on the graph which is slightly different than what I see on the report. Why?

This is still good information, but Version 3.2 (released around October 2002) provides a Preference to ensure the torque and HP filtering and RPM step sizes for torque and HP are always the same, thus ensuring that you always get the same numbers on the Main Screen, in the Graphs or in Reports. To set this, click on Preferences, then "Graphing" tab at the top, then set "Specs for RPM Graphs and Reports" to "Increments/Filtering match Main Screen".

The torque and HP peaks are the highest values the program finds in the data. However, the user must understand that the data has been arrived at by doing a "bunch of math". For example, from an inertia dyno, the data is dyno wheel RPM and engine RPM measured at various periods of time, not torque and HP measured at various RPMs. For absorber dynos, the data is volts measured at various periods of time. In addition, all measured data has a certain amount of error associated with it.

In the process of doing the math, the program "smooths" or filters the data to help reduce the effects of measurement errors and so the data looks more like what the user expects. The Dyno DataMite program gives the user a lot of freedom to choose the amount of filtering, and also the RPM increments for reporting data. Both of these settings will affect the numbers and therefore the highest number (peak number) you will see. The Figure below shows a torque curve from an absorber dyno. You will see that the top curve with the least filtering is also the "jumpiest" curve. It has more highs and lows and therefore produces the highest "Peak Torque" of 856 ft lbs at 5600 RPM. Applying some "Light" filtering reduces the peaks and fills in the dips. Now the Peak Tq is only 849 at 5650 RPM. Applying "Heavy" filtering greatly reduces the peaks and fills in the dips, to produce a Peak Tq of only 844 at 5450 RPM. This is all with the exact same recorded data, just doing the math on the data differently.

You will now ask "Which number is right?" and the answer is "They all are, if you include the math which produced the results." If we keep getting this question, Performance Trends may need to take away the user's freedom to set various amounts of filtering and the RPM increment size for reporting data. This would be to avoid confusion on the part of the user. Currently, these are your options:

Main Screen: Data is reported at the RPM increment and filtering level specified in the Preferences menu.
Graphs: Data is always graphed at 50 RPM increments, but 4 levels of filtering are available in the Graph Settings screen.
Reports: Data is reported at the RPM increments and 4 levels of filtering are available in the Report Settings screen.

If you make the RPM increment size and filtering the same for the Main Screen and the reports, you will obtain the same peak numbers. Graphs are done at 50 RPM increments (smaller than available with reports or at the Main Screen, and therefore will be slightly different).

When you make comparisons from run to run, you must use the same filtering and RPM increments to do so. That is why the best comparisons are done with graphs, putting both or several runs on the same graph. Then you know the RPM increment and filtering for all runs are exactly the same. Comparing Peak numbers between runs is not a good way to compare runs.

Can I use just engine RPM or Dyno Wheel RPM (not both) to calculate torque and HP?

Yes, there are several options. These options are set in the Dyno and DataMite screens. The Dyno screen is available by clicking on Dyno at the top of the Main Screen. The DataMite screen is available by clicking on DataMite at the top of the Main Screen.

To use just Engine RPM for an Engine Dyno, select Dyno Type of "Engine, No Clutch" and the Gear Ratio to the correct ratio (at lower left in the Dyno specs screen). Then in DataMite specs screen, be sure to change the Sensor & Calibration of the dyno wheel RPM to something other than Dyno Wheel RPM, or set it to Not Being Used. Not finding Dyno RPM, the program will use engine RPM and the gear ratio set in the Dyno specs screen to calculate the Dyno Wheel RPM, torque and HP, assuming NO CLUTCH SLIP. If you graph or report the Dyno RPM channel, it will show the actual recorded Dyno RPM, not the calculated RPM.

To use just Dyno RPM for an Engine Dyno, select Dyno Type of "Engine, No Clutch" and the Gear Ratio to the correct ratio (at lower left in the Dyno specs screen). Then in DataMite specs screen, be sure to set the Sensor & Calibration of the dyno wheel RPM to Dyno Wheel RPM. If necessary, you can set Channel 1 (typically Engine RPM) to Dyno Shaft RPM in the Pro version of the software. Finding Dyno RPM, the program will use it and the gear ratio set in the Dyno specs screen to calculate the Engine RPM, torque and HP, assuming NO CLUTCH SLIP. If you graph or report the Engine RPM channel, it will show the actual recorded Engine RPM, not the calculated RPM.

To use just Dyno RPM for a Chassis Dyno, select Dyno Type of "Chassis, No Engine RPM" (at lower left in the Dyno specs screen). Gear Ratio in the Dyno Specs screen now becomes disabled (grayed out). Then in DataMite specs screen, be sure to set the Sensor & Calibration of the dyno wheel RPM to Dyno Wheel RPM. In Test Conds, you must set the correct gear ratios and tire size so the program can correctly calculate the Engine RPM, torque and HP, assuming NO CLUTCH SLIP. If you graph or report the Engine RPM channel, it will show the actual recorded Engine RPM, not the calculated RPM.

If you have the Pro version of the software, you can get Calculated Engine RPM to show up on the Current Readings screen on the gauges in Preferences. If you have chosen one of the options above and are using Dyno RPM, you can do the following: Click on Preferences at top of main screen, then the Calculations Tab. Set "Engine RPM is Calculated RPM" to Yes, then click on OK to keep this change. Now you can see this Calculated Engine RPM on the Gauges if you select to display Engine RPM. BE CAREFUL, because if you have entered too low a gear ratio in the program, the Calculated Engine RPM will be too low and you will over-rev the engine. Also, be sure to turn this off when you start to MEASURE Engine RPM again.

How do I measure Engine RPM on a chassis dyno?

Because there are so many different engine ignition systems, we have developed several methods to measure engine RPM.

DTM-IPU Inductive Pickup. This method involves tie wrapping a wire from our harness to a plug wire or secondary wire from the ignition coil. We have also had good luck placing this wire near the coil on a "coil on plug" type engine.

DTM-IPUC Inductive Pickup Clamp. This method is the same as the DTM-IPU above, but provides a "timing light style" clamp to clip to a plug or coil wire. This can provide a cleaner signal than just the DTM-IPU with a wire, but can not be used for "coil on plug" type engines.

BB2-IPULV Low Voltage Inductive Pickup This sensor clips to a low voltage wire, like the coil primary wire or fuel injector wire.

DT2-RPMR Reflective Optical RPM Sensor This sensor "watches" a piece of reflective tape you will place on something spinning at engine RPM, like the crank pulley or dampener. This method is VERY reliable and universal, but requires with you coming up with a bracket. The sensor has a red "pointer" beam which greatly aids the setup process.

Calculate Engine RPM from Dyno RPM The software allows you to CALCULATE engine RPM from dyno wheel RPM from simple vehicle measurements or from a simple calibration test (Pro version of software only). This method assumes the clutch and tire slippage is 0, so it does not work well with automatic transmissions. The v4.1 software has new features to make this method even easier than previous versions. Check the Dyno DataMite v4.1 Supplement (in particular pages 248-250) for more details.

Back to Top

Can I use just your software with an existing dyno?

Our software only works with our electronics. You will need to install our electronics and sensors and then you can use our software. We do NO control of the dyno at this time, which is very important with absorber (water brake, hydraulic pump, eddy current, etc) dynos, either a chassis or engine only dyno. By control, I mean the system which holds RPM constant as you make changes to the engine or throttle, or the system which lets the engine accelerate at a slow, continuous rate as you are making a power run. Being inconsistent at how you let the engine accelerate during a run can lead to not-repeatable tests.

If you already have control with your dyno, you should keep that existing dyno system to do the control. Then add our sensors and electronics to do the data gathering, graphs, etc.

Back to Top

How does your software and electronics control the dyno?

How you control the dyno is very important with absorber (water brake, hydraulic pump, eddy current, etc) dynos, either a chassis or engine only dyno. By control, we mean the system which holds RPM constant as you make changes to the engine or throttle, or the system which lets the engine accelerate at a slow, continuous rate as you are making a power run. Being inconsistent at how you let the engine accelerate during a run can lead to not-repeatable tests.

for info on our Dyno Controllers.

If you already have control with your dyno, you could keep that existing dyno system to do the control. Then add our sensors and electronics to do the data gathering, graphs, etc.

You could also continue to control your dyno manually with the throttle and water control valve. You will continue opening the throttle and applying load with the water control valve until the engine is at WOT (wide open throttle) and at the starting RPM you desire. This requires a little experience and practice to get good at it, especially with high powered, turbo'd or "peaky" engines. You will adjust the control valve to get the engine to either accelerate or decelerate at a slow (100 to 300 RPM change per second), continuous rate through the entire RPM range.

A big advantage of the inertia dyno is the "control" is near perfect. You start at a low RPM, hit the gas (go to WOT or wide open throttle) and the engine accelerates smoothly and consistently because of the large amount of inertia in the system.

How will I run a test with a water brake dyno?

All absorber or brake dynos (water brake, hydraulic pump, eddy current, either chassis or engine only) tests are done as a "sweep" test, where the engine goes through its full RPM range and the DataMite or Black Box system records the data at a very high rate. When the test is done, then the software will report or graph the data at standard RPM increments, like every 100 or 250 RPM, whatever you select. This is much more accurate and efficient than trying to stabilize at various RPM steps and then record data, then go to the next RPM step.

Check the other FAQs here about controlling the dyno, and section 4.3 in the user manual at this link: PDFs/DTMMANL7.pdf

Is my load cell working in tension or compression?

The load cell is used to measure the torque being absorbed by the dyno by preventing the dyno from rotating. "S" beam load cells (shaped like the letter S) can work in either tension or compression. Performance Trends needs to know this when we wire up your load cell and amplifier. Typically tension is easier to design for if you are making your own brackets. Old fashioned load cells were hydraulic cylinders and worked only in compression. So if you are replacing an hydraulic load cell with an electronic "S" beam load cell in the same position, it will also be working in compression. Check out the picture below to know if your load cell is working in tension or compression. If your engine or dyno turns in the opposite direction, the working of the load cell will reverse.

click image to enlarge

The load cell signal is typically from 0.3 volts to 4.3 volts. It is possible to wire up a load cell to work in both tension or compression, where 0 ft lbs will be about 2.3 volts. Tension may have the voltage go down and compression may have the voltage go up. However, if we wire it this way you lose half the voltage scale for recording torque, which limits accuracy. For this reason, we wire it for either tension or compression, not both.

for info and parts for reversing the operation of your load cell if you want to switch from tension to compression.

Can most of the USB DataMite III or 4 features be added later if I don't buy them now?

There are 2 options to consider, which require returning the logger to add at a later time:

Internal Weather Station Sensors, part number DT3-IWS: The weather station allows the DataMite software to automatically do weather corrections to the data exactly the same for each test. This helps produce more repeatable data. These sensors and fan must be installed and checked by Performance Trends, which means you must ship the unit back and we then ship it back to you. And the cost for doing this after the box is returned is $25 more that doing this originally. When you include shipping costs which can be expensive if shipping outside the USA, you may want to get this option right away and save all the shipping and extra expense.

Internal 4 or 10 Thermocouple Channels, part numbers DT3-TC4 or DT4-TC10: These thermocouple channels let you plug in inexpensive (typically less than $60) type K thermocouples to record most any kind of temperatures. (You can add external thermocouple circuitry, but it costs more than $215 to add just 2 channels.) This circuitry must be installed and checked by Performance Trends, which means you must ship the unit back and we then ship it back to you. And the cost is $25 more for doing this after the box is returned. When you include shipping costs which can be expensive if shipping outside the USA, you may want to get this option right away and save all the shipping and extra expense.

We are happy to sell to the system without thermocouples, but it is VERY nice to always have the temperatures recorded the same way with each test. If you ever have a question, you can go back and look at the data and say "Yes, the head temp was the same" as some other test, not hope you manually recorded the temp for that test, but it could have just been carried over from the previous test. In just a couple of days, you will not remember. Also, adding a $35 inlet air temp thermocouple can make for better weather corrections, because the temp at the box can be different than temp going into the engine. Barometer and humidity are same at engine as at box.

How can I troubleshoot bad or "corrupt" data I'm getting from the Black Box II Dyno System?

Take the top off the box, but leave the wires attached to the lid plugged into the board. Make sure the ribbon cable connector from the computer connector is firmly plugged into the board connector.

If that was OK, then:

What happens if you just run a test without the engine running, just record data for 20 seconds and download it. Does this show up as corrupted?

If Yes, if you have an optical isolator, what happens if you repeat the step above without the optical isolator in the system?

If not, then unplug all sensors from the box and run a test with engine running and download. Does this show up as corrupted?

Plug in optical isolator for tests with sensors hooked up.

If not, then start plugging in sensors 1 at a time (starting with dyno RPM) and run a test and download. Does this show up as corrupted?

What sensor being plugged in shows up as corrupted?

Let us know what happens when you try these tests.

Can you do a "coastdown" test on our chassis dyno to measure the dyno and vehicle losses, and then correct the chassis torque and HP to get flywheel torque and HP?

Doing a coastdown test of the vehicle you just tested to estimate the losses in the driveline, to then come up with flywheel HP is VERY MISLEADING. Reason is because the losses with power ON are much higher than when coasting, especially for an automatic trans.

For example, if the losses from coastdown at 4000 RPM show 32 HP, the losses with full power will be much higher (perhaps 100 HP for a 500 HP engine) when accelerating. Tire slippage alone will be 3-10%, which equates to a direct loss of that percentage in power. (During coastdown, tire slippage is 0%.) The difference between full power and coasting gear box losses are the difference between power-On efficiency and simple "spin losses". Torque converter slip during power-On can be 30% or more, again a huge loss that is 0 when doing a coastdown.

Back to Top

How can I email Performance Trends a test from my DataMite software, for Performance Trends to check out?

This is done by putting a floppy disk, memory stick, or writeable CD into the computer. Then click on File at upper left of main screen, then select Save to CD/Floppy Disk. (If the drive letter does not match what drive letter you want to use, you can change this in Preferences at the top of the main screen. Click on the Filing System tab, then change the Default Floppy Disk drive letter. Then click on OK button to keep this change.)

This will copy the 3 files which make up the test to the floppy or CD. Take this disk to an email computer and send all 3 files to a Performance Trends email address to evaluate. The Performance Trends tech will give you and email address to send it to.

Starting with Version 3.7, there is an "Emailing" tab in Preferences. Here you can set up your email properties so that emailing can be more automatic. Then, at the main screen you can click on File, then Email These Results and the program will email all 3 files automatically. Once you have these properties set, you can also email graphs and reports.

My Black Box II (or DataMite II) will not communicate to my computer. What should I do?

Remove the optical isolator from the system if you have one. If this fixes the problem, be sure to set the Preference to Enable Optical Isolator power. If this is set correctly, then you may need a powered optical isolator from Performance Trends.

Be sure you are using the cable supplied with the system, called a Null Modem cable. Not all 9 pin cables are the same. You need one where pins 2 and 3 swap through the cable, pin 2 is connected to pin 3 and vice versa.

With the Black Box II, remove the lid (keeping all wires connected) and be sure the ribbon cable from the computer connector is firmly plugged into the connector on the board.

Click on DataMite at top of main screen, then Troubleshoot, then Check Com Ports. Does your COM port pass the "paper clip" test? Can you find a COM port which does pass this test. Note that COM 3 many times may pass this test, but will still pass the test with the paper clip removed, and is not a valid port. If no ports pass this test, you need to take your computer to a shop and have the Com Port activated correctly.

If you find a Com Port that does work, then note which Com Port Number. Click on Preferences at top of main screen, then General Operation tab, then set Automatically Find Com Ports to No. Then click on OK at top right of this Preferences screen. Click on DataMite at top of Main Screen, then set the Com Port to the correct number you've just found. Also be sure to have set the correct Type of DataMite to the correct type for you system. Click on Back at the upper left corner, and say Yes to Saving these changes and Yes to Saving these changes as the Master DataMite Specs.

Click on DataMite at top of main screen, then Troubleshoot, then Check Boot Message and follow the instructions. Do you get a message from the Black Box II or DataMite II? If yes, it means that you can at least receive info from the box, but may not be able to send commands to the box. Contact Performance Trends tech.

If you are using a USB adapter, check its status:
-Right click on My Computer, then click on Properties. The "System Properties" window appears.
-Click on the "Hardware" tab then "Device Manager". This gives you a list of devices on your machine.
-Look down the list for "Ports (COM and LPT) and click the "+" next to it. You will now see a list of ports on the machine. You should see a "Communications Port (COMx)" (probably with "USB" mentioned also) where the "x" is the port #. This is the port # you put in the DataMite specs screen.

Back to Top

What are the advantages and disadvantages of an inertia dyno compared to a water brake?

An inertia dyno uses the inertia of a large flywheel or drum to absorb the power from the engine and slowly accelerate through the RPM range. By knowing the inertia and measuring the acceleration rate, torque and HP can be calculated. An absorber dyno (water brake, hydraulic pump, eddy current, etc) absorbs the power of the engine in some type of "pump" which raises the pressure of the fluid, eventually turning the power into heat. The amount of torque required to keep the housing of the "pump" from rotating is also the amount of torque the engine is producing. Both methods produce accurate results, but each has advantages and disadvantages.

Advantages of Inertia Dyno

Very simple to build. All that is required is a shaft and a sufficiently large flywheel to provide a slow enough acceleration rate for accurate measurements. The power run should be 6-10 seconds ideally.
You only need to measure dyno RPM and do not need a torque sensor.
"Perfect" speed control. You start at a low RPM, hit the gas (go to WOT or wide open throttle) and the engine accelerates smoothly and consistently because of the large amount of inertia in the system.
Because of the simplicity of the system and the "perfect" speed control, the inertia dyno system can be extremely repeatable.
The "bang for the buck" of the inertia dyno can NOT be matched by an absorber dyno, especially for small engines.

Disadvantages of Inertia Dyno

Can only measure performance while the engine is accelerating. You can NOT hold speed at a particular RPM.
Very large amounts of inertia are required for large engines. Flywheels or drums weighing 1000 lbs or more may be required.
Can be unsafe. Due to the large amounts of inertia and speeds involved (especially for large engines), that is a lot of energy waiting to be released should something break, or you slow down (brake) the system too quickly.
Can NOT measure the torque and HP at the exact beginning and end of the run. In order to accurately measure the acceleration rate, we need to know the engine RPM both some time before the current RPM and AFTER the current RPM. At the end of the run, we don't know the RPM some time AFTER the current RPM, because we never get to that RPM. What this means is if you need to accurately know the torque and HP at 10,000 RPM, you have to run your test out to 10,500 RPM or possibly 11,000 RPM.

Advantages of Absorber Dyno

This system CAN measure performance while the engine is setting at a steady RPM. This can be very handy when making tuning changes while the engine is running.
The system can be quite small and compact. All power is converted to heat in the fluid, which is typically quite safe.
You can bring the engine RPM down very quickly because there is so little inertia in the system (very safe).
It CAN measure the torque and HP at the exact beginning and end of the run.

Disadvantages of Absorber Dyno

More complicated and expensive to build. Large dynos require a lot of water to absorb all the power from the engine.
Needs a good torque (force) measurement system.
Need good speed control to produce repeatable results.
May require a lot of cool down time between runs once the dyno water system gets warm.
These systems can be very repeatable, but generally only if you spend a lot of money, especially the control system.

Are there any general tips for building an inertia dyno?

The following is a sample warning report which can be generated by the Dyno DataMite program. It shows some specific RPMs and MPHs based on certain sizes of components describing a particular inertia dyno system. These limits will change for various dyno systems as you change critical dimensions.

------------------------------------------------------------------
The following are general tips to explain that if you are not careful, your dyno could turn into a 'bomb', injuring or killing yourself and bystanders. As with any relatively high RPM rotating machinery, you should have the design checked by a qualified engineer.
------------------------------------------------------------------

The largest diameter in your dyno system is 30. inches.

It is recommended that you do not exceed 1333 RPM

There is no way the program knows exactly how your system is designed. Poorly designed systems should not be used at all, but should definitely NOT exceed 1000 RPM, while well designed systems can operate safely to RPMs of 2000 RPM or higher.

Because Chassis dynos typically have relatively thin wall rollers (drums) and because of the speed limitations of the vehicle and tires, do NOT exceed 92 MPH on this 23.0 inch diameter roller. That means do not test a vehicle to more than 92 MPH on its speedometer.

More Chassis Dyno Tips:

- Chassis dyno rollers are more complicated to build and to analyze for safety than simple flywheels.
- Always note the tire's speed ratings and keep vehicle speeds well below the tires limits.
- Keep tires properly inflated for safety and consistent results.
- The smaller the diameter of the dyno roller, the more tire deformation, the more tire heat buildup and the more HP losses.
- Change transmission gear for testing to a lower one to keep the MPH low. However, if the run is over too quickly (accelerates too quickly), you can move to a higher gear ratio (but keep the top speed below all speed limits discussed above).

The important thing to note is that you want to keep the RPM low to keep your dyno system safe. A flywheel or chassis dyno roller can literally burst if you over-rev it, or cause the entire system to start tumbling violently if something jams or brings the flywheel to a stop (brake) too quickly.

More tips include:

- Karting inertia dynos should NOT use hollow 1.25'' shafts. (Solid 1.25'' shafts are commonly used with 24'' diameter x 1.25'' thick flywheels, but even this is only marginally safe.
- Do not weld flywheels to shafts, unless the shaft is a large diameter.
- Use collars, hubs or flanges to avoid creating a stress riser right where the edge of the flywheel meets the shaft.
- Avoid very large, but thin flywheels. Do not use flywheels with a diameter more than 20 times the thickness.
- Put some type of 'cage' or containment around the flywheel to 'catch' a magnet which could fly off or to allow the flywheel to 'coast' down slowly (rubbing on the inside of the cage) should a shaft or bearing fail.
- Many inertia engine dyno operators like to use a one-way clutch, so the engine can come to a stop while the dyno wheel coasts down more slowly.

There are many more things to consider to make a dynamometer which is safe to operate. Be sure to consult other references like the Machinists Handbook.

Could you elaborate on the advantages (carb tuning related) of the A/F sensor over BSFC?

BSFC stands for Brake Specific Fuel Consumption and is reported in pounds of fuel consumed per hour, per HP. Very efficient engines at full power at their torque and HP peak typically run about .40 to .45. This means the engine requires .40 lbs/hr of fuel flow for every HP it makes. If it was making 1000 HP, the fuel flow required would be 400 lb/hr (or approximately 1 gallon of fuel/minute, GPM).

BSFC is a measure of how well the engine is converting fuel into HP, NOT how rich the engine is running. For example, if you retard the spark 20 deg, the BSFC goes higher (say from .40 to .50) because the same fuel is going in, but the engine is making less power. To many people this is indicating the engine is richer. The engine is NOT richer, just less efficient at converting fuel into HP.

A major problem with fuel flow measurements (and therefore BSFC measurements) is fuel flow meters have a limited operating range. At the low end they get inaccurate or don't read at all. At the high end, they get restrictive, changing the fuel pressure significantly in the fuel line.

For example, for our popular 3/8" NPT fuel flow sensor would produce about 3-4 psi pressure drop in your fuel line at about 1500 HP on gasoline (about 700 HP on alcohol). However, below 350 HP on gasoline (170 HP on alcohol), it would not read accurately or at all.

The 1/4" NPT fuel flow sensor will read down to 15 HP on gas, but would produce 3-5 psi pressure drop at just 180 HP on Gas. At 360 HP the pressure drop is up to 10-20 psi. This smaller sensor has proven to be quite restrictive.

Oxygen sensors for A/F have their own set of problems. The sensors degrade and get inaccurate with age and leaded fuel, any air leak or back flow from reversion make for inaccurate readings, etc. However, when they are working correctly, they are the best measure of how rich the engine is running.

Both A/F and BSFC are useful as long as you know the difference and limitations of each.

I have a 2 roller chassis dyno. Do I put the RPM sensor on the front roller or rear roller?

There is a good deal of tire slippage and tire deformation which takes place on a twin roller dynamometer. The RPM between the 2 rollers can easily be 10% different or more.

You must put the RPM sensor on the same roller as the torque absorption unit. If this is a twin roller inertia dyno, put the RPM sensor on the roller attached to the largest amount of inertia. This is typically the front roller, because the tire will better adhere to it as the torque output increases. The tire will tend to pull away from the rear roller.

What is meant by Master DataMite Specs and Master Dyno Specs?

The DataMite Specs and Dyno Specs are critical for accurate dyno test results. For each test you run, a copy of these specs are saved with each test, and that provides for the following advantages:

If you have found you have made a mistake (for example had the wrong calibration for a sensor or wrong number of magnets for the dyno RPM sensor), you can simply correct the mistake and the test is now correct. You don't need to rerun the test with the correct calibrations. This is VERY handy if you find a mistake once the engine is off the dyno and down the road.
If you make a change, like say you make your inertia dyno wheel (flywheel) bigger, you want all new tests to use the new bigger flywheel. However you want all your old tests to use the flywheel specs for that test. By having separate DataMite and Dyno Specs for each test, it is possible to do this.
You may want to analyze (graph, report, etc) results from 2 very different dynos. These may be your dynos, or may be a friends Dyno DataMite results. By having separate DataMite and Dyno Specs for each test, it is possible to do this.

Let's use example 2 above to explain the Master Dyno Specs. So, lets say you are looking at an old test which was run with the smaller dyno flywheel. If you go into the dyno specs, the program tells you these specs were used for the old test (the current test you are looking at). However, these specs do NOT match your Master Dyno specs, the specs for the larger flywheel.

Now, if you start a NEW test, and this old test happens to be on the main screen, you don't want the new test to use the old small flywheel. You want it to use the Master Dyno specs which are for the larger flywheel. The Master Dyno specs are the specs for your dyno settings right now, today. The program is keeping track of the both Dyno Specs for this old test and what dyno specs should be used for each NEW test (the Master Dyno specs).

If you want to see the Master Dyno specs, just go into the Dyno specs and you should get the message saying the current test's settings either DO or DO NOT match the Master. If they match, then what you see on the screen are the Master Dyno specs. If they DO NOT match, then click on File at the top of the Dyno Specs screen, then Open Master Dyno specs. They will be loaded for you to view. When you click on Back to close this screen, you are asked if you want to Save these changes. If you just wanted to see the specs, be sure to answer NO. If you DO want to use the Master Specs for this test, like you are correcting a mistake, then answer Yes and the results will be recalculated based on these new settings.

The Master DataMite Specs work the same. For the Road Race/Circle Track and Drag Race DataMite software, the concept is exactly the same except the Master Dyno Specs are replaced by Master Vehicle specs.

A very similar concept is used in our Port Flow Analyzer, Cam Analyzer, Shock Dyno, and Spring Tester.

Should the density altitude go up with the humidity? I've tried to read up on it but can't make perfect sense of it. Can you please explain?

Density Altitude is really a very old, nearly useless term which can have various definitions. It came about when drag racers used aircraft altimeters to judge air quality. We give you 2 readings in the DataMite software, and other Drag Racing software.

Density Altitude which only takes barometric pressure and air temperature into consideration, which is all old altimeters could do. This should correlate with an old altimeter.
Dry Density Altitude which also takes humidity into consideration, and is a much more useful and accurate term. You will see that as humidity level goes up, the Dry Density Altitude also increases, indicating poorer quality of air for producing power.

I'm getting erratic Engine RPM readings. What can I do?

There are 2 types of "erratic" Engine RPM. One is where the engine RPM can be perfect at some times or at some RPMs, and then do something "crazy", dropping down to 0 RPM, or jumping up 10,000 RPM for a split second, which we will call "erratic". The other is where the engine RPM is constantly switching about 5 or 10 % between 2 different RPM readings, which we will call "dithering".

Erratic RPM:

Engine RPM is critical for accurate dyno results, and can be measured most conveniently by an inductive pickup sensor like a timing light. A very common question we get is "We were getting very clean engine RPM and now we're not. What changed?" Unfortunately, it could be most anything, different engine, ignition or ignition components, plug gap, A/F, spark advance, compression ratio, etc. Anything which affects how much voltage it requires for the spark to jump the spark plug gap will affect the "electrical noise" in the plug wire which is what we use to measure engine RPM.

We have 2 different "clips", the standard inexpensive one DTM-IPUCW and the optional, more expensive "timing light" style DTM-IPUC. Prior to 2013 we used a different, smaller timing light style clip which was not always better than the standard clip. Then we switch to a bulkier style which did seem to be an improvement over the standard clip.

DTM-IPUCW (standard) DTM-IPUC (new, better) DTM-IPU (inductive pickup signal conditioning)

Both clips require the signal to be "cleaned up" by our Inductive Pickup signal conditioning, DTM-IPU. It has a sensitivity adjustment knob, and its operation is described in this Inductive Pickup Notes PDF document (click on blue link).

Other methods of measuring engine RPM include:

Magnetic RPM Sensor
Low voltage inductive pickup (to be picked up of 12 v primary coil wire or EFI injector wire) BB2-IPULV
Tach pulse signal (however this requires the logger to be grounded to the engine and can cause other issues)
Optical RPM Sensor
Calculate engine RPM from dyno RPM, for both engine or chassis dynos
OBD2 engine RPM, excellent for chassis dyno on modern street vehicles

to view a list of various styles of RPM sensors, with prices and pictures.

Suggestions: Here are some things to try to clean up erratic engine RPM readings recorded via the inductive pickup.

Make sensitivity adjustments to the inductive pickup, as described in this Inductive Pickup Notes PDF document (click on blue link).
Sometimes the sensitivity knob does not provide enough adjustment. To get an even stronger signal, wrap the purple wire around the plug wire 2 to 6 times. Some users report success with taping the purple wire along side the plug wire (parallel) for 2 inches or more. Then you still may need to fine tune the sensitivity with the knob.
To check if the signal is too strong, remove the purple wire from the plug wire. With the engine running and the inductive pickup turned to highest sensitivity, hold the purple wire 12 inches away from the plug wire and watch the RPM readings on the computer. Start moving the purple wire close to the plug wire and note where it starts reading RPM. If you have to get within 1/2" or less to the plug wire, it is unlikely the signal is too strong. Then you still may need to fine tune the sensitivity with the knob.
For engines with 2 cylinders or more, move the inductive pickup and other wires away from other plug wires which could "cross talk" to the plug wire you are measuring. See pics below.
Make adjustments in the DataMite III and DataMite 4 Preferences settings, as described in DTM USB RPM Preferences PDF document (click on blue link).

click on image to enlarge and show graph of Inductive Pickup Engine RPM with and without noise, and OBD2 RPM for comparison.

The "fix" for the Inductive Pickup noise in the graph above was to space the wire with the standard clip away from the other 2 plug wires on this bank of a V-6 engine. There was also an adjustment to the Inductive Pickup box sensitivity screw, making it less sensitive. The graph also shows that the OBD2 Engine RPM was clean in both tests, but do lag the more accurate Inductive Pickup RPM by about 0.1 seconds or so. This lag is not critical unless you are doing very fast accelerations, which are not good for accurate dyno testing anyway.

Standard Inductive Pickup clip, plug wires close together Moving this plug wire away from the other 2 with a shop rag fixed the problem

Dithering engine RPM:

This problem is much like having unevenly spaced magnets on a shaft or dyno inertia wheel. to read up on this effect.

Graph of dithering engine RPM sent by a customer

There are many reasons the engine RPM pulses could be unevenly spaced. One could be the engine is unevenly firing, like a Harley Davidson or some early Buick V-6 engines. Another could be due to a single cylinder engine running "wasted spark" ignition. Here the engine fires every rev even though a 4 stroke needs spark only every 2 revs, which is the case in the graph above. What you are seeing is that sometimes the DataMite catches the power stroke (faster) ignition pulses and sometimes the exhaust stroke (slower) pulses. This is random, but why you see the RPM jump between 2 different RPMs, a fast and slow one.

If you have a DataMite III or DataMite 4, there is a "fix" for this "dithering" RPM. Go into the DataMite specs, click on the Sensor and Calibration for the Engine RPM, and select "Every 2nd pulse" for "Use Ign. Pulses", then click on Keep Calib. button to save this change. Now the DataMite will ignore every 2nd pulse, and determine RPM based on pulses that are evenly spaced.

I am trying to get Engine RPM from a coil-on-plug COP engine. What can I do?

If you have the Inductive pickup box, as described in the FAQ above, you need to remove the clip from the purple wire. Then you can try to tie-wrap the purple wire to the coil, to the spark plug, or just push the purple wire down the spark plug hole getting it as close as possible to the spark plug. NOTE: Be sure there is shrink tubing over the end of the purple wire or the spark plug could "zap" the inductive pickup box. The newer inductive pickup boxes (2014) are designed to prevent this, but nothing is perfect.

You can also try wrapping the purple wire around a fuel injector, which is similar to a small ignition coil. If you do not get a signal, try adjusting the sensitivity adjustment on the Inductive Pickup box. If you go to max sensitivity and still do not get a signal, try wrapping the wire more tightly around the injector, or wrapping it 2 or 3 times around the injector. You may have to adjust the "x stroke, x cylinder" setting for RPM to match the number of times the injector fires per revolution. If once per rev, assign this as "1 cylinder, 2 stroke" or "2 cylinder, 4 stroke" which also fire once per revolution.

My DataMite will hang up (stop updating the screen) during a dyno test. What can I check?

I'm going to suggest things for troubleshooting, things that are relatively easy to do to understand the problem. Do not assume these are permanent changes. After each suggested change, resume dyno testing and see if the situation improves.

1) Reboot the DataMite and computer. Shut down and restart your computer and then restart the DataMite program. Before going into the Current Readings screen, unplug both the USB cable and the power to the DataMite
box, wait about 10 sec and plug them in again, typically best to plug in power, wait 5 sec and then the USB cable. Wait about 10 sec, or about 5 sec after you hear the USB connect tone before going into the current readings
screen.

2) The USB communications is very fast for our USB loggers, and the standard USB cable is 10 ft long to keep the communications accurate. If you are using a longer cable, or especially if you have added a USB extension cable, this could be the problem. Go to the standard 10 ft USB cable and eliminate any extensions to see if the problem improves.

3) Try setting sampling rate to 25/sec in the DataMite specs screen. The screen below shows it set to the default of 50/sec.

4) Try reducing the size of the current readings screen. The larger the screen, the more computer time it takes to refresh it. Click on the 'Restore' button in the upper right corner as shown below if you have the Current Readings screen maximized. If it is still large on the screen, try shrinking the borders so it takes up less computer screen.

5) Try reducing the update rate for the Current Readings screen. The default is 10 times per second. The picture below shows it reduced to 5 times per second. Try going even lower to understand what the problem could be.

6) Try moving the DataMite and the computer farther away from the engine, especially the ignition system.

7) Try moving the sensor cabling farther away from the engine, especially the ignition system.

8) Try going to resistor plugs and plug wires.

9) When you are having a day when this is happening, close out of the Current Readings screen. Do a Ctrl-Alt-Del (hold down the Ctrl and Alt key and then press the Delete key, then release them) and see what else is running. Click on the Processes tab at top, then click on the "CPU" label at the top of the "CPU" column 2 times. This will put the processes using the most CPU time at the top. Is something else running which is taking up a lot of CPU time, or is "System Idle Process" 90% or more. It is possible something is running in the background taking up lots of resources the DataMite needs to properly record data.

for info on a related FAQ.

How can I use the weather readings in my Dyno DataMite software to tune carb jetting at the track?

I've tuned my jetting on the dyno at some particular weather conditions. Now I'm at the track and the weather is significantly different. How can I adjust the jetting to bring the A/F back to where it was on the dyno?

There are some issues to consider when rejetting a carb:

A carburetor tries to hold A/F constant, but it does in imperfectly. If you are not sure of what you are doing, it is probably best to leave the jets alone.
How your carb is designed has a lot to do with how you must change jets to hold A/F constant for weather changes. This primarily has to do with the power valve, it's size relative to the jet you will adjust and if there is even a power valve present in the carb.
Weather reading can vary significantly between weather stations, TV or radio report weather data, etc. This is especially true of the barometer reading, either observed or corrected to sea level. For this reason, the best way to determine the rejet required it is best to use the exact same software and weather station for all purposes. This would mean taking your DataMite to the track and letting the Dyno DataMite read the weather just like done on the dyno.

In v4.1 of the Dyno DataMite software we've added a feature to copy the weather readings from a particular dyno test to the Windows Clipboard. Then you can go to Performance Trends Weather Wiz program and paste these readings into either the left side (baseline) conditions or the right side (new) conditions.

If you are rejetting at the track, you would open a dyno test for the engine when it was well tuned at the dyno, then go into Test Conds and click on the 'Copy' button in the Weather Readings.

Then open the Weather Wiz and click on the "<" button under Copy from DataMite Logger label to paste these conditions as the Baseline weather. You must also set up the carb parameters so the Weather Wiz knows what type of carb you are trying to rejet. There are other Jet options under Options at the top of the screen.

Now you can read the weather readings from your Dyno DataMite software and logger at the track, record a short test (even if the engine is not running) and paste the track weather readings into the right side of the Weather Wiz. The Weather Wiz will give its analysis of how much the A/F will change with no rejet or suggest how to rejet to bring back the same A/F you had on the dyno.

You can also use the Weather Wiz's own "stand alone" Weather Station to read in the track weather automatically, or a different style weather station and type in the track weather data (if you have a thorough knowledge of the data it is reporting) to do the same rejetting calculations.

I have a dyno run with noise spikes in the RPM signal which the program can not edit out. What can I do?

The Enterprise Edition has a special "Advanced Edit of RPM Noise Spikes" feature to let you manually cut out a section of data with noise spikes and replace it will an "estimated RPM". Click on Edit (upper left of main screen), then Advanced Edit of RPM Noise Spikes and choose the RPM channel with the noise. A graph is presented of that RPM channel vs time.

click image to enlarge

With the mouse, click and drag a box around the section of noise. Note: Do not go much beyond the section of noise as the bigger the box, the less accurate will be the new estimated RPM will be which replaces the spikes. If necessary, do this 2 or more times if there are more than 1 section of spikes.

click image to enlarge

That section of RPM data will be replaced by a straight line from the first and last RPM data outside the box. Over a small amount of time, this estimated RPM is sufficiently accurate for good dyno data.

click image to enlarge

My RPM measurements seem very jumpy. Why?

Jumpy RPM readings can be due to different things, like vibration, electrical interference, etc. However, when there seems to be a repeatable pattern in the "jumpiness" and you are using more than 1 magnet (or other target), the problem is most likely the magnets are not evenly spaced. Say you have 2 magnets on a wheel, which should be spaced 180 deg apart. If you have them spaced 179 and 181 degrees and the wheel is spinning at a constant 2000 RPM, the RPM measured between 179 deg magnets is actually 2011 RPM. The RPM measured between 181 deg magnets is actually 1989 RPM. Instead of a steady 2000 RPM reading, you see an RPM reading jumping between 1989 and 2011 RPM.

(click image to enlarge)

In the picture you will see a definite, repeatable pattern in the green wheel RPM trace. The wheel has 4 magnets on it, making the job of getting all magnets evenly spaced difficult. However, because these wheels are very large, 4 magnets made the RPM measurement better because the RPM was quite low. The driveshaft had only 1 magnet, so it was perfectly evenly spaced, as shown by the blue trace.

Note on Inertia Dynos: Inertia dynos use the acceleration rate of the inertia wheel to determine the power output. Acceleration is determined by measuring the rate of change in RPM. If part of that change in RPM is caused by the magnets not being evenly spaced, the acceleration rate can not be determined as accurately. For that reason, we typically recommend using just 1 magnet to eliminate this spacing problem, unless the RPM is very low. If the RPM is going to be less than about 500 RPM, we would recommend 2 magnets, and if the RPM is going to be less than 300 we would recommend 3 or 4 magnets on the wheel or shaft.

My Inductive Pickup does not seem to read RPM. Why?

If could be that the ignition signal is too strong. Be sure to reduce the sensitivity first to see if you start getting an RPM reading.

If that does not produce any type of RPM reading, it could be the sensitivity adjustment of the Inductive Pickup does not go sensitive enough. We put a 1K (1000 ohm) resistor in series with the sensitivity adjustment pot (knob), so 1000 ohms of resistance is as low as the sensitivity will go. If you are good with a soldering iron and electronics, you can solder in another 1K resistor in parallel with the 1K resistor there now, called R1. Now the sensitivity will be double of what it was before. See picture below.

click image to enlarge it

Notes:

A very common mistake is calling a motor with "wasted spark" a 4 stroke motor. Wasted spark means the 4 stroke motor's ignition fires at TDC compression (like it should) and also at TDC overlap. You should set this as a 2 stroke motor because if fires on every TDC just like a 2 stroke motor. Trying to make a wasted spark engine produce good readings by setting it up as a 4 stroke motor could waste a lot of time. for more info on dealing with "wasted spark" ignitions.

Another common mistake is you have not set up the software correctly to read engine RPM. Run the dyno and see if you get dyno RPM. If you start getting engine RPM once you get dyno RPM, you have the software set up to calculated engine RPM from dyno RPM in Preferences. Also be sure there is a Yes in the Used? column in DataMite specs and you have the engine RPM plugged into the correct connector on your logger.

R1 was reduce in value to about 750 ohms in 2019 for better sensitivity.

R1 can be in different locations on different generation boards. Do not go lower than adding a 1K resistor without contacting Performance Trends. When you are done soldering in the additional 1K resistor, check with a volt meter that the resistance across R1 is now somewhere between 400-600 ohms.

Wrapping the purple wire several times will produce a stronger signal, making this resistor modification unnecessary. You can also tape the purple wire along the length of the plug wire for 2-4 inches. This also produces a stronger signal.

I need my brake dyno to measure torque in both directions of rotation. Can you set up a load cell to do that?

This is a common question we get, especially from chassis dyno owners who want to test RWD and FWD vehicles on the same rollers. The FWD cars would be backed onto the rollers, and thus spin the rollers in the opposite direction. And, Yes, we can set up a load cell amplifier to measure torque in both directions, under compression and tension. However, there is a price to pay for this.

For a normal load cell, say, always in compression, we have a full 0-5 volts to work with. For example 0-5 volts could be 0 to 1000 lbs. If you want this load cell to do both compression and tension, then we only have 2.5 volts to work with. For example, 2.5-5 volts could be 0 to 1000 lbs and 2.5-0 volts would then be 0 to -1000 lbs. You would loose half of the resolution of your torque measurement. You would also have to change the calibration number(s) in the DataMite program for the torque channel. Typically this would involve just changing the sign of one of the torque calibration numbers.

Another option that is easier for most mechanically inclined people to do, would be to reverse the load cell mounting. When you go to FWD from RWD, you could just change where the load cell is mounted. This would keep it always measuring in, say, compression even though the dyno would be spinning and absorbing torque in the opposite direction. With this option, you do not need to change the calibration numbers in the program (unless the torque arm length changes with the second mounting location). You also do not loose any resolution in your torque measurement.

Recent Development: Also, we have developed an adapter cable which can reverse some wiring at the load cell amplifier. This way you do not have to change the load cell mounting. Check with us for some options.

I'm using the Engine Inertia correction for my water brake dyno. How can I see how much effect it is having on the results?

When you are doing an accelerating test, some of the engine's power is going into accelerating the engine's and dyno's rotating inertia, so results will be lower than during a steady state dyno run, where you set for, say, 2 seconds at each RPM. The faster the acceleration rate, and/or the more inertia (larger engine flywheel, heavier the crankshaft, etc), the more the power loss. During a decelerating test, power is being released from the spinning inertia, so results will be higher. This is one reason people like decelerating tests more. It produces higher numbers.

The Pro and Enterprise Editions of the Dyno DataMite let you correct for these inertia effects. The goal is to produce the same power whether accelerating or decelerating, whether fast or slow, whether lots of rotating inertia or not. However, because you must be somewhat careful to keep track of things to use this feature accurately, we somewhat hide it. This feature must be turned on in Preferences under the "Calculations (cont.)" tab. Now you will see a new setting in the Test Options screen as shown below.

click image to enlarge

When this is turned On, the settings which effect how much inertia the engine has is set in the Engine Specs screen, as shown below.

click image to enlarge

A user who was using the Engine Inertia correction wanted to know how much effect this was having on his results. So, like with many things you my want to investigate, you should first make a copy of a test under a new name. I saved is "10" test under the name "10 no inertia". Then all I did was set the "Correct for Engine Inertia Effects" in Test Options screen to No and compare the results, as shown below. (Note: I had the Preference "Show Comparison Run Also" set to Yes under the Main Screen tab, so my changed file appeared on the main screen graph, compared to the original "10" graph. You would have gotten a similar graph by clicking on the "Graph" dropdown at the upper left of the main screen and choosing "10" and "10 no inertia" for graphing.)

click image to enlarge

You can see from the graph above and columns of numbers down the left side, this accelerating test gained about 20 HP (402 peak HP down to 382 HP) with Engine Inertia Correction turned On, or about 5%. This engine accelerated at about a rate of 280 RPM/sec. If it had accelerated more quickly, like 600 RPM/sec, the effect would have been more than double. The idea would be that both the 280 and 600 RPM per second test would produce 402 HP with the Engine Inertia correction turned ON. However, with the Engine Inertia correction turned OFF, 280 RPM/sec accel would produce 382 HP and a 600 RPM/sec would produce about 360 HP. Obviously you want results which are repeatable, independent of the acceleration rate.

to see some more data where the Engine Inertia correction and our Dyno Controller produces very repeatable data

The torque and HP will peak at different RPMs depending on if I measure or calculate the engine RPM on my chassis dyno. Why?

The graph below shows that when the engine RPM was measured with an inductive pickup on a spark plug wire, the torque and HP peak at a lower RPM than when it was calculated from the dyno roller RPM on this chassis dyno. When you calculate engine RPM from dyno roller RPM, the assumption is the ratio between the engine and dyno RPM remains constant throughout the RPM range, and is the same as when you calibrate the ratio (called "factor" in the DataMite program). If this is not constant, then the calculated engine RPM will not be as accurate as measuring it.

click image of power curve comparison to enlarge

However, the good news is the torque and HP levels between the 2 runs are about the same, just that the RPM scale is somewhat changed.

The graph below shows graphing out the engine and dyno RPM vs time for the test with measured engine RPM, and include "Calculated Gear Ratio" (the ratio between the 2 RPMs).

click graph of RPMs and Calculated Gear Ratio vs time to enlarge

You will see the Calculated Gear Ratio varies about 15% during the run depending on the power level being put out by the engine, the higher the power the higher the gear ratio. This is likely indicating that there is slip occurring somewhere in the system during the test, perhaps the clutch or the tire on the roller, or perhaps tire deformation on the roller. (Obviously if you disengage the clutch, this ratio will change greatly and be meaningless.)

So, to summarize, it is best to measure the Engine RPM on a chassis dyno if the Engine RPM signal is clean. If you will calculate Engine RPM from Dyno Roller RPM, remember that the amount of clutch, converter or tire slip you get will change, and can change where the RPM where the torque and HP occur.

When running my dyno, the computer will sometimes lock up and I have to restart it. How can I troubleshoot?

Here are some important things to know when you are trying to troubleshoot a problem:

1) When troubleshooting, you want to try things which are very easy and quick to do. You are not necessarily trying to come up with a permanent fix, just trying to understand what may be creating the problem.

2) To test a change, you do not need to run full power tests to save fuel and wear on the engine. You can do "dummy" tests accelerating the engine at half throttle to only 5000 RPM when a normal test would go to 7500 RPM. You do not have to run the engine at all if that will test the change you have made. The program will record data with the engine not running. It just will not show power curves and the software will want to call the test a "Custom" type of test because it does not look like a power run.

3) Get used to clicking on the "Troubleshoot" button, touching the top of the graph on the main screen. It shows the raw data from the 1 or 2 channels which are most critical to creating power curves, graphed versus time. Time graphs are much more useful for troubleshooting problems.

click image to show Troubleshoot button and graph enlarged

Here's a check list of things to try to understand what the problem could be. If you try a step and the problem is gone, now we have information for solving the problem.

1) Check if the problem is present with the engine not running. Watch the Current Readings screen, perhaps watching the weather or power volts channels to ensure you have communications. You should see reasonable readings on those channels and the numbers changing slightly.

2) Repeat item 1, but this time with the engine running, perhaps idling or a "dummy" test.

3) If the problem is not there when trying steps 1 or 2, it could be electrical noise from the ignition system or engine vibration producing intermittent connections in one of the connectors. Full power running produces more electrical noise and vibration. Try the following:

    - Move the computer farther from the engine if within 10 ft of the engine.
    - Move sensor wires farther from the ignition system and possibly disconnect the engine RPM lead from the DataMite.
    - Try disconnecting some or all sensors from the DataMite. If problem is fixed, the problem is that particular sensor or cable, or noise being picked up by that sensor or cable. If problem is not fixed with nothing connected to DataMite, problem is with DataMite box (which is very unlikely) or power to DataMite or USB cable. USB cables longer than 10 ft can produce problems.

4) If problem is still there when trying items 1 or 2, then problem could be software setting or PC.

    - Do you get a "blue screen message" or an error message which does NOT have "DataMite" in the message? If it is a Windows error message, problem is most likely with your computer.
    - Is it possible some program starts up in the background when problem occurs, like a virus scan or backup program?
    - Try reducing the size of the Current Readings screen from being maximized. Maximized screen requires a lot more computer resources.
    - Click on Options in the Current Readings screen and reduce the Update Rate to, say, 5 or 2 per second. 10/sec requires a lot more computer resources.
    - In DataMite specs, reduce the sampling rate to perhaps 25 or 10 samples per second. 50/sec requires a lot more computer resources.
    - Like stated above, these are not permanent fixes, but just trying to understand what the problem is.

Once you have tried some of the items above, and you can still not figure out what the problem is, contact Performance Trends.

Why should I measure weather conditions when engine testing? Does this produce what they call Corrected Torque and Corrected HP?

We've got a very thorough Motorsport Blog which describe this issue in great detail.

for the Motor Sport blog describing weather affects on engine performance and how to compensate (correct) for them.

My RPM Reading sometimes reads double of what it should, and sometimes reads correct. Why?

The first thing to realize is we live in an imperfect world. Every cylinder firing in the engine is slightly different than the last, and every ignition pulse is also different. The Inductive Pickup is trying to create a signal from the electrical noise emitted from the spark plug or coil wire. It has a sensitivity knob to help adjust it to produce a clean, accurate reading. However, this electrical noise changes greatly depending on engine conditions like, load (throttle opening), RPM, A/F, etc. So you could adjust for a clean signal at idle, but then at full power (WOT, or wide open throttle) the reading completely change.

When an RPM signal jumps between reading accurately to double or half the correct RPM, the ignition system typically has a "wasted spark" system. Wasted spark is when the ignition fires double what is needed. 4 stroke Briggs type motors typically have wasted spark, meaning they fire every TDC, like a 2 stroke motor. A 4 cycle motor needs to only fire every 2nd revolution. The wasted spark is occurring during TDC overlap. Well the ignition signal produced during overlap is typically much lower than during the compression stroke. See the figure below.

The circles show when an RPM pulse is read by the system. The top picture shows how if you adjust the sensitivity knob, you can get the system to read correctly. The 2nd picture shows what happens with "wasted spark", where the overlap signal is lower than the compression signal. Depending on how you set the Engine RPM calibration in DataMite specs, and how you have the sensitivity set, you can get correct or incorrect readings. Now, say you have it set to see just the compression pulses at some particular condition, say lower RPM or lower load. Then at higher RPM or higher load, both signals increase. All of a sudden you are seeing both the compression pulses and the overlap pulses. The RPM signal will immediately double.

Typically we say with a 1 cylinder, "wasted spark" ignition to set up the Engine RPM calibration in DataMite specs as a 1 cylinder 2 stroke (not 4 stroke). This assumes you will see both compression and overlap pulses. Then adjust the sensitivity low enough to pick up both compression and overlap pulses.

Note: Circles shown on each ignition signal pulse show the pulse that will be passed through the inductive pickup box to the DataMite to be recorded.

The picture above also goes into some details about what the inductive pickup does to "clean up" the ignition pulses, by not allowing some of the "ringing out" of the signal to be measured. If it was measured, you would get possibly 2 up to 5 pulses for each single firing of the plug, and very erratic RPM readings.

So you can see there can be lots of reasons for the inductive pickup to read half or double the correct RPM. But with some understanding of what is happening, you will be able to adjust the sensitivity knob for the best accuracy throughout the entire RPM range.

for a movie demonstrating this.

How should I size a fan for my dyno room?

You need a certain amount of air flow through your dyno room to be sure you get rid of all the exhaust. This is both for accurate engine testing (air with exhaust produces less power) and for your safety. You also need to get rid of the heat generated by the engine, especially the exhaust.

One way to get rid of the exhaust is to run a complete exhaust system and route it out of the room. However, for most race motors running low restriction exhaust or open headers, this is not acceptable. A common method of routing exhaust out of the room is pictured below.

Diagram of typical dyno room exhaust Picture of kart motor dyno room exhaust pipe

You can provide suction pump(s) to suck the exhaust and extra room air out of the room. Or you can force feed the room extra air to force the exhaust and extra room air out the room exhaust pipe(s).

The engine's air flow requirements are approximately 1.5 CFM for every 1 HP, so assume 2 CFM per HP. So a 400 HP engine will require fans or blowers which will produce at least 800 CFM. If you are sizing suction pumps for the exhaust system shown above, you want to size larger than that so there is always extra room air being sucked out also. Also, exhaust is hotter than air so it is more expanded than room air. So a safer assumption would be 3 CFM per HP. And because the engine is generating heat you will want blow out of the room, you may also want to size blowers a little bigger than 3 CFM per HP.

Some people worry that a blower may put too much pressure or suction in the room. If there was much more than 0.1" mercury pressure (1.4" water), you probably could not either open or close the door. A standard 6'8" x 36" door is 2880 square inches. A pressure of 0.1" mercury is about .05 psi, which sounds like nothing. But that .05 psi pressure over 2880 sq inches is 144 lbs on the door. Also, if you have one of our data loggers with internal weather sensors, and have that mounted in the room with the engine, any small variation in room pressure is measured and corrected for by the barometer weather sensor.

So to summarize:

Provide some method of routing the exhaust out of the building per the diagram above
Provide at least 3 CFM air flow for every 1 HP of the engines you will test.
Mount your DataMite data logger with internal weather sensors in the room with the dyno to correct for any very slight change in air pressure from the ventilation fans.

I have a fuel flow sensor from a different data logger company, but am reading it with your DataMite. Why doesn't it read until the flow is quite high?

Most fuel flow sensors have some type of turbine wheel inside and produce an RPM type of signal. Ours uses a very low drag type of sensor, but others use a magnetic sensor. The magnet tends to make the turbine wheel "stick" until the fuel flow is quite high.

To show the difference, we did a test where we put a pulse of air through our meter and let it coast down. It took over 30 seconds to coast down to a completely 0 reading as shown by the blue line. We did the same with a magnetic "Brand X" sensor and you can see it coasted down to near 0 in about 1 second, the green line. The much higher drag of the magnetic sensor means it requires a lot more fuel flow to read. It can not read low fuel flows. Where ours with the much lower drag turbine wheel can read very small fuel flows. And, ours is about half the cost. Just 2 more reasons to pick a Performance Trends' DataMite logger.

How does a brake dyno (water brake or eddy current brake) dyno work?

The word "brake dyno" comes from the fact that in the early 1900s, the absorber was actually a friction brake, like brakes in your car. They were very difficult to control and got hot very quickly. A modern brake dyno puts a load on the engine with something that can absorb power for long periods of time. These include something like a hydraulic pump, a water impeller like a water brake, or an electrical brake like an eddy current brake.

The image below explains the basics and how you need some type of load cell measuring force (lbs) at some distance out from the center of rotation (feet) to measure the torque (foot lbs) being absorbed by the dyno.

click image to enlarge it

My torque and HP are low and both peak at the same RPM. Why?

This typically occurs because you are not at WOT (wide open throttle, full power) throughout the run. For example, if you are slowly "easing into the throttle" during the run to avoid tire slippage on a chassis dyno, your tq and HP will be low until you get to WOT. In the picture below, it looks like the engine finally got to full throttle right before the end of the run, when the throttle was closed. Both the torque and HP peak at nearly the same RPM, which is very unusual. The torque should almost always peak 500-1000 RPM or more before (at a lower RPM than) the HP peak.

Here's an even more extreme example.

To fix the problem, be sure you are WOT before you start your run. If you can't do this because of some slippage or other limitation, you must realize you are not getting full power performance curves.

The air for my motor is different than the air for the DataMite logger. Will the Weather Corrections still be accurate?

Three measurements go into Weather Corrections, Barometric Pressure, Humidity, and Air Temperature. The barometric pressure at the motor and at the DataMite are exactly the same (unless you have a special "altitude test cell". The humidity at the motor and at the DataMite are the nearly the same, unless you have special A/C to change the humidity of the air going to the engine. Also, the fan in the lid helps ensure that the sensors inside the box can quickly track any changes in humidity and temperature in the room. (For accurate measurement of absolute humidity, the air temperature for where the relative humidity is measured is also required. The DataMite system is set up to use the air temperature inside the logger for this, as this is also where the relative humidity is measured.)

However, the temperature can change very quickly and can be significantly different between the DataMite and the air going into the engine, especially if that air is coming from a separate duct. For this reason, the DataMite software lets you assign a thermocouple channel as "Std Thermocouple, Eng Intake Air". If this channel is assigned, the software uses this temperature for the correction factor. In this way, the DataMite system will accurately know all 3 conditions for doing accurate weather corrections.

for the Motor Sport blog describing weather affects on engine performance and how to compensate (correct) for them.

Is there an easy way to adjust the inertia of my inertia dyno to get the torque or HP number I want?

First, figure out how much you want to factor the reported torque or HP up or down. Say you want 100 HP and dyno is showing 80 HP. The factor will be 100 / 80 = 1.25 or you want 1.25 times more inertia.

Say the Total Inertia in the lower right corner of the dyno specs for the 80 HP run is 121. To increase the HP by 1.25, you will need 121 x 1.25 or 151 inertia. Adjust the inertia wheel width and watch the system's total inertia number until you get 151. Save this and back out of the dyno specs screen and the power will be refigured based on this new total inertia, showing close to 100 HP.

Notes:

For a Chassis Dyno, do not adjust the roller diameter. This diameter is used to measure the vehicle's speed (MPH or KPH).

Chassis dynos will always report less power than an engine dyno. Expect chassis dyno numbers to be less. The current Pro or Enterprise software has a Preference setting to allow you to enter a "Drivetrain Losses %" so you can factor up chassis dyno numbers for these losses. This loss is also reported in the printouts.

If this is an engine dyno, where 1 flywheel makes up most of the inertia, just multiply the flywheel thickness (not diameter) by the factor, 1.25 in this case.

For good repeatability, DO NOT keep adjusting the dyno specs. Once you set them, do not change them unless you are truly making a change to the dyno.

You say your water brake controller does not control the RPM sweep rate, but the valve opening rate. Why?

It is our goal that our Dyno Controller control the RPM sweep rate, like 300 RPM/second or 600 RPM/second. However, if this is not done perfectly the controller can go into oscillation. That could easily damage your $50,000 motor and would be a much bigger problem than not having a perfect sweep rate.

The picture below shows a dyno controller going into oscillation. The customer had our Dyno DataMite logger system for recording data but was using another company's controller. It was set to produce a 350 RPM/second sweep rate, but went into oscillation at the end of the sweep. You can also see some "blips" at other points during the acceleration. Luckily the overspeed in the system shut it down and prevented major damage to the motor.

click image to enlarge

It can be quite difficult to control an engine dyno, because there is so little mechanical inertia in the system and the engine can have so much power. For this reason, chassis dynos are easier to control because the rollers, wheels and tires, etc add lots of inertia. Inertia makes it so the RPM can not change so quickly and the controller can keep up. Other things that make an engine difficult to control and more likely to go into oscillation are things that produce quick changes in power, like a shot of nitrous, or turbocharged motors.

For this reason, as of 2019 or water brake engine Dyno Controllers are just controlling the sweep rate of the water valve's opening and are much less likely to allow for this oscillation.

How can I mount Performance Trends sensors on my Land and Sea Dyno?

Performance Trends has retrofitted many different dynos, including Mustang ^tm, Clayton ^tm, SuperFlow ^tm, DynoJet ^tm, Stuska ^tm, Land and Sea ^tm, and others. A couple of customers who fitted our Dyno DataMite to their Land and Sea dynos sent in some pictures to share with other dyno customers.

Jimmy Brandon from Bland, MO first used our Land and Sea RPM Conversion cable to keep using the original L&S RPM sensor. However, it produces some noise spikes and he converted it over to our standard, inexpensive RPM sensor. Jimmy was quite clever in that he used the existing hole for the L&S RPM sensor for mounting is fabricated bracket, and plugged the hole. Jimmy says "I got the new RPM sensor installed today. The RPM reading on the dyno is rock steady now, before it jumped around a little. It even reads the cranking RPM when I am starting the engine, which it bounced all over the place before. I used the existing threaded hole in the water brake where the old sensor installed to attached the bracket, that was the only hard part as it needed a 9/16 fine thread Allen head bolt. If anybody else has a Dynomite 13 inch brake and needs a sensor feel free to show them how I did it."

Comment from Performance Trends: We have slightly better luck with the magnet mounted on top of the hub with epoxy, not buried flush into the aluminum.

click images to enlarge

Ryan Freeman of Raptor Racing Engines, Randleman NC devised this method for mounting both the load cell and RPM sensor on the smaller L&S dynos. The L&S dynos have special arms designed by L&S with the "load cell" built into the arm. However, this "load cell" is only a quarter bridge load cell, which can work. However, the load cells Performance Trends sells are full bridge load cells, which are typically much more stable. They don't drift with time, temperature, minor resistance changes in the connectors, etc. Ryan fitted this L&S arm with our full bridge load cell as shown in the picture. Its got a ball joint (heim joint) top and bottom which is good because it avoids any side loading on the load cell. It also has a rubber damper which reduces torque spikes, and protecting the load cell from damage. Ryan also mounted the RPM sensor in his arm. Check out the pictures below.

Comment from Performance Trends: If there is not much torque going through the arm, things should be fine. However, this is the thinnest part of the arm and may fatigue in time. You could also mount a bracket to the bottom of the arm with a couple of small screws, as pointed out in the picture.

click load cell images to enlarge

click RPM sensor images to enlarge

Here's how Joe Harkness, Leslie MI josephis34@aol.com mounted his RPM sensor and load cell. Joe tests chainsaw motors, and his pics can give you chainsaw guys some good ideas.

click images to enlarge

Can I keep using my SuperFlow (tm) controller if I use your DataMite for data logging and analysis?

Both the Performance Trends' DataMite and the SuperFlow controller need a dyno RPM signal to operate. We are working on sharing the SuperFlow RPM sensor signal for both systems (June 2021) and may have something in the future to allow this. However, if you can mount a new DataMite RPM sensor on anything that spins with the dyno (shaft, coupling, engine, etc), that works best. Then the SuperFlow controller can keep using the SuperFlow RPM sensor. The picture below shows how Dave Bisschop in Canada mounted the DataMite RPM sensor in his bellhousings for DataMite RPM.

click image to enlarge it

My peak HP (but could also be peak torque) seems to change a lot and is not repeatable. Why?

This is a common question when people just like to look at peak numbers and not the full power curve. For example, if you have VERY flat torque or HP peaks, a very minor change could move the HP peak from, say, 443.4 HP at 5250 RPM to 443.6 HP at 6000 RPM. 0.2 HP could move it 750 RPM. You may think the tune of the engine has completely changed, but it's just a very flat HP curve. Flat curves are very typical of restricted motors (engines with restrictor plates or small carbs to meet race track restrictions).

Looking at just peaks can be even more of a problem with very peak engines, like 2 stroke engines. Look at the graph below where there are two HP peaks, one at about 8000 RPM at 24.5 HP and another at about 11,000 RPM at about 25 HP . For most all RPMs the 4 runs agree very closely. However, one test (the black HP graph) it jumped up some at the 8400 RPM peak so that now 8400 RPM was the peak HP RPM, not 11,000. If you did not look at the entire curve, you may think that run was totally different (not repeatable).

click image to enlarge it

Here are some things you can do to avoid thinking tests are not repeatable when they actual are:

Always compare runs on graphs. Do not just look at peak numbers.

For peaky engines, increase the filtering setting. The graph above was done with Light filtering, allowing more detail in the graph. However, for very peak motors, you may want to set the Filtering higher, maybe to Heavy. You will see now the HP peak for all 4 runs is at 11,000 RPM. However, the peak HP number is lower due to filtering (smoothing off the high points and filling in the low points). Because it is slightly lower, most engine builders do not go for Heavy filtering.

click image to enlarge it

You can set up your Dyno DataMite program to use a certain amount of filtering (smoothing) and RPM Increment (100, 200, 250, etc) for doing all reports and graphs. Choosing a larger RPM increment is also like filtering in that the data will be more repeatable, but you will not see possible details. Shown below are the Preference settings you should make.

click images to enlarge them

Look at average numbers over an RPM range. Another option is to use Dual Cursors as pointed out in the first picture. If you turn these on, you can bracket the area you want analyzed as shown below. Then you can click either the Maximum, Minimum or Average buttons to have that calculated for the area between the Dual Cursors. (Look up Dual Cursors and all graphing features in Section 3.3 in the Users Manual for details.) You can see the Average HP between the cursors for these 4 runs varies from a low of 23.45 to a high of 23.67, or less than a 1% difference from lowest to highest. Not bad for an engine this peaky. Note: Reports can also be made with averages over a certain RPM range.

click image to enlarge it

How can I check out my DataMite System without a dyno or engine?

You can make some "dry runs" with your Dyno DataMite system without an engine, or even a dyno. You will not get any torque or HP curves, but you can get used to how the system works and record data from some sensors. With nothing plugged in, you can at least measure and graph and report power supply voltage, and weather sensors if you have internal weather sensors.

for a movie describing how to make a "dry run" with no engine or dyno.

How do I get my DataMite program going on a new computer (transferred over)?

The process outline here is the same for the Road Race and Drag Race DataMite programs, and very similar for most other Performance Trends programs.

There are several ways to do this. The most universal way is to copy the complete folder from the old computer to a memory stick (zip drive). This is done without the Dyno DataMite program running. Right click on Start in the lower left corner, then select Explore (or Windows Explorer, or File Explorer). On the left side of the new screen, look for the C: drive (sometimes called Local Disk C:) and double click it. On the left side, now look for Program Files folder and double click it. (For Win 7 and Win 10 this may be Program Files (x86).)

For Dyno DataMite, look for Performance Trends folder and double click on it.

Look for Dyno DataMite Analyzer v3.7 folder and right click on it and select Copy.

Now find your memory stick or CD drive on the left side of the screen, right click on it and select Paste. You have now copied everything from your old computer to your memory stick or CD. This is an excellent way of making a backup of all you data also.

Now, on the new computer, install the Dyno DataMite program to your new computer to the default location, the location where it automatically wants to go. You can do this from your CD, or to download from our website, and follow all of the default prompts. (Note: Install the version you own, like 4.1 or 4.2, unless you want to purchase an updated version, typically $100 to $250). When asked, select to do a Complete installation.

Now go to your new computer and insert your memory stick or CD. Again, this is done without the Dyno DataMite program running. A message may come up automatically and you want to "Browse" the memory stick or CD. If not, Right click on Start in the lower left corner, then select Explore. On the left side of the new screen, look for the yellow folder Dyno DataMite Analyzer v3.7. Right click on it and select Copy.

Now look for the C: drive through the same screen from where you Copied the folder. You may have to go up a few levels. When you find the C: drive (sometimes called Local Disk C:), double click it. On the left side, now look for Program Files folder and double click it. (For Win 7 and Win 10 this may be the Program Files (x86) folder.)

Look for Performance Trends folder and double click on it.

Right click on Performance Trends folder and select Paste. If you have done this correctly, it should ask you if you want to overwrite some files and say Yes. (Do not make the common mistake of copying the Dyno DataMite folder from the zip drive into the Dyno DataMite folder on the C drive.)

Now install the Dyno DataMite program to your new computer to the default location, the location where it automatically wants to go. You can do this from your CD, or to download from our website, and follow all of the default prompts. (Note: Install the version you own, unless you want to purchase an updated version, typically $100 to $250). This time, when asked, select to do a Refresh installation instead of a Complete installation.

Now you should be able to start your new Dyno DataMite installation on the new computer by double clicking on the desktop icon and all your old test files should be there. Click on File, then Open from All Saved Tests to view them.

Resetting your Master Dyno Specs:

To get your dyno settings and calibrations back, you need to reload the Master Dyno and Master DataMite Specs. In the DataMite program, open one of your old tests which you know represents how your dyno is configured today, a recent good test. Click on DataMite at the top of the main screen. Once inside the DataMite Specs screen, click on File (at top) then Save as Master DataMite Specs. Click on Back to return to the main screen. Click on Dyno at the top of the main screen. Once inside the Dyno Specs screen, click on File (at top) then Save as Master Dyno Specs. Click on Back to return to the main screen. Now you should be ready to start testing again.

Notes:

With newer versions of the software, you can right click on Open when opening a good, recent file, then select Open as Master. The program will prompt you to click on OK or Yes about 6 times to walk you through the process outlined in the paragraph above.

Your program will run for 10 days on this new computer. You will need to get a new Unlock Code from Performance Trends to run past this 10 day demo period. At the top, left corner of the Main Screen, click on File, then Unlock Program Options. Send the Computer Hardware number, Registered Name and Registered Code #. I need all three to send the correct unlock number. If you have a CD, send the 4 digit serial number from the CD also.

For our USB DataMite loggers, you will also have to get the USB Drivers installed to get the logger to communicate on this new computer. Check the Quick Start paperwork that came with your DataMite for how to install the drivers from the CD. for running this USB installer from our website.

for our main Downloads page to install Road Race or Drag Race DataMite software, or most any of our other programs.

How long are the cables for the typical Dyno DataMite system?

click image for typical cable lengths for Dyno DataMite system

Are there instructions that come with the Dyno DataMite system?

There are extensive instructions that will come with your Dyno DataMite system. The larger User Manual mostly covers the software and it's many options. Then we include a "Quick Start" which mostly covers the exact hardware options you got. Below are links to some examples:

for the Users Manual for the software.

for a typical Quick Start our DataMite III for an inertia engine dyno. (Depending on your exact options and sensors, your Quick Start could be different.)

for a typical Quick Start our DataMite 4 for a water brake dyno. (Depending on your exact options and sensors, your Quick Start could be different.)

My updated Dyno DataMite v4.2 A.062 gives an error message about "data missing". What does this mean?

In v4.2 A.062 we added an additional check for data integrity for each data set recorded. If it finds a problem, you will get an error message like that below.

to read the January 2023 newsletter which discusses what to do about this message.

How can I make my flow meter read Gallons per Minute for water?

Most of the time flow meters with the DataMite loggers are used for measuring fuel flow. These sensors put out a frequency signal like RPM because they have little turbines inside. Then this flow can be assigned as fuel to or from the engine, used to calculate total fuel consumed and BSFC. For these calculations, it is important to know the specific gravity of the fuel so volume flow like gallons per minute can be converted to lbs per hour, a mass flow of fuel.

However, sometimes you just want to use a flow meter to measure flow, like water flow, oil flow, etc. In these cases it is easiest to assign these to an RPM channel and use the "Other RPM" calibration. This calibration is not used for any other calculations in the software like clutch slip, engine RPM, fuel flow, etc. Shown below is the calibration for a very large flow sensor, with a K factor of 806 pulses/gallon. Smaller flow meters will have a K factor more in the range of 9,000 to 40,000. Like the picture shows, the # Magnets should be assigned as 1, and the Multiplier is 1 / K factor, or .00124 in this case. The Data Name can be whatever you want. Now this Other RPM channel will read directly in gallons/minute with the label "gal/min" on graphs and reports. If you want gallons/hour, multiply the .00124 x 60, or .0744. If the fluid is water, which weighs 8.33 lbs per gallon, multiply the .00124 x 8.33 for lbs per minute.

NOTE: If the multiplier gets too small for the program to accept accurately, change the number of magnets to 10, and multiply the multiplier number by 10.

My updated Dyno DataMite software is saying "x.xx seconds of data missing" from my dyno test. What does this mean?

This message appears when the software senses there is a problem with the current test you ran, most likely caused by electrical noise. We've always known that electrical noise from the ignition, certain light fixtures, CNC equipment, parts washers, battery chargers, VFD motors, etc. can cause noise in the DataMite data. This shows up as some extreme jumps in data, say for head temp that reads about 200 degrees to read 0 or maybe 1200 degrees for a couple of data points. Here's an example of noise on an RPM channel, where the real RPM is between 800 and 1600 RPM. But there are noise spikes jumping to over 10,000 RPM.

The DataMite software can typically find these noise spikes after the test and edit them out, as shown below.

Well, we've just leaned that this "noise" can also cause the DataMite to not sample at the correct rate. Say the DataMite is set to sample at 50 times a second, that is every 0.02 seconds, or 20 milliseconds. If there is a lot of electrical noise, the DataMite may sample at every 20 mSec for most data, but perhaps 21 or 22 mSec at other data points. This is typically not a big deal, unless you are measuring acceleration rate, like with an inertia dyno. If the time is not what we expect it to be, acceleration will be wrong and torque and HP from an inertia dyno will be wrong.

Here's a graph of back to back inertia dyno runs, one with no noise and the other with noise which shows faster acceleration and therefore more torque and HP. You will notice the noise effect started at about 5800 RPM.

In Dyno DataMite v4.2 A.062 we have now added a check in the software to see if the sampling rate is consistent and accurate. If we detect a problem, we will give you an error message like that below when you first upload a test to your PC.

This message is very helpful for troubleshooting the source of electrical noise. For example for troubleshooting the dyno producing the graphs above, we would get this message uploading a test with the engine not even running. They were also having trouble with their DataMite 4 not connecting to their PC. It seemed to connect only half of the time.

We would record about 10 seconds of data, and see if the message above would pop up, and it did. Then we disconnected most all sensors and recorded 10 seconds of data again, and the message did not appear. Then we plugged sensors back in 3 or 5 at a time to narrow down the source. It turned out to be 2 of the EGT probes which were bringing in electrical noise from an electric pump motor. Their thermocouple extension cable was running close to an electric conduit.

Before we had this software check, all we knew was that some tests produced different power when they should be the same, and we did not know which data was correct. If you own v4.2 software, this is a free update.

The new v4.2B software released July 2023 also checks for this problem, and can correct the data so it is like the electrical noise problem was never there. There is a Preference setting that tells the software how you want this problem to be handled. However, it is still best to try to eliminate the noise problem at the source. See below.

for the text file displayed if you click on the "Click for Info" button shown in the Preferences screen shown above.

My A/F readings seem too lean, especially at idle. What could be wrong?

A/F sensors (wide band sensors, O2 sensors, lambda sensors) read the Air to Fuel ratio based on how much oxygen or O2 they sense in the exhaust. The less oxygen, the richer the mixture. The amount of oxygen in the exhaust is VERY low, like less than 0.5% or 1/2 of 1 percent. BUT, if any room air gets into the exhaust, which is about 20% oxygen, it really makes the exhaust mixture look lean.

Therefore you must be very careful to avoid any leaks in your exhaust between the engine and the sensor. Exhaust systems have strong pulses, both positive and negative and negative pulses can suck small amounts of room air into the exhaust making it read lean. Also, if you put the sensor too close to the end of the exhaust pipe, these negative pulses can suck room air to the sensor from the end of the pipe. This will especially happen at lower RPM and lower flows, like idle.

Click the image below for more details on mounting your A/F sensor for good readings.

Click image for larger PDF version

Drag Racing and Circle Track/Road Racing DataMite Data Logger Questions

My MSD Grid (tm) ignition (or most any other high energy ignition) is producing electrical noise and interfering with the RPM signals. What can I do?

Our DataMite USB IIIs and 4s are very robust for dealing with electrical noise. However, noise can come in on the RPM sensor wires and produce false signals, producing incorrect RPM readings.

"Sportsman Plus" Drag Race System

Burgess Coleman, a good friend of Performance Trends and racer from Ford with lots of "hands on" experience with drag racing ignitions, had these suggestions:

Here are some things to try in order of most likely to fix the problem with the least amount of hassle or affect on engine performance.

1) Click on the link below for what MSD suggests to use for suppressing electrical noise from their ignition systems:

https://www.holley.com/products/electrical/parts/8830MSD

2) Install a smaller 1 uF cap in the same way as the large MSD noise suppressing cap described above. Use a high voltage rating (100 volts or so) cap. Ceramic caps do not have polarity so it is not critical how you install it (which side goes to ground).

3) Run a ground strap from the engine block to one of the MSD box mounting screws.

4) Do not run solid core wires. Use MSD's spiral wound 8 mm wires.

5) Run a shielded 16 AWG wire from the 7AL (or equivalent) to the coil (typically black wire and orange wire). NOTE: This must have insulation which can withstand 600 volts or more. Attach the drain (shield) wire to a good chassis ground on the vehicle or engine block at just 1 end (not both ends).

6) If still problems, consider running resistor or suppressor spark plugs.

My car seems down on power. How can I check it?

Our Road Race/Circle Track DataMite and Drag Race DataMite loggers have options to produce dynamometer quality power curves from doing acceleration tests in your car. Recently I had a need to see how my 68 Mustang was running, and I did some of these tests with great results.

click image to enlarge it

for the Motor Sport blog describing this type of testing and the results.

Can I measure spring and/or shock loads on my car on the track?

We often get asked by circle track racers if our DataMite loggers can measure the shock and spring loads when they are out on the track. They are trying to better understand their car's handling. This can be done with fairly expensive load cell bolts or load cells, which can be difficult to mount and maintain. Or you can more easily estimate these loads using shock travel sensors and spring and shock data.

For example, if you know how much the spring is compressed and know the spring rate you can calculate the spring force. If you know the shock velocity and have shock dyno data, you can calculate the shock force. Add the two together and you have the same data you would get from a load cell mounted on the spring and shock, or a coil over. Our Shock Dyno program works with our Road Race/Circle Track DataMite data logger program to make this process very easy.

click to enlarge screen for exporting data from Shock Dyno program

click to enlarge screen for importing Shock Dyno data to Road Race/Circle Track DataMite data logger program

for a detailed explanation of how this is done. Note, as of Sept 2019 this is still a limited release option. Contact Performance Trends for more info.

I click on the Run Number down the left side of the screen and it does not open the correct run. What am I doing wrong?

When you get a new run from the memory card, the program asks you for a file name. If you just let the program default to the name it suggests, you could overwrite and old test. The programs warn you that you are doing this, and v4.2 is better at suggesting new file names. However, if you just click OK or Yes on the messages without reading them, you may be missing some important info.

click image to enlarge it

for a detailed answer to this question.

Road Racing Software Questions

I want to optimize my car to produce the quickest time to accelerate between 2 different MPHs. How can I do this?

Our Drag Racing Analyzer (std, Pro and Pro-Team Engineer) all can do it. All take the inputs you want to change. Concerning, not starting at 0 speed: the lower level of the program, the more creative you have to be to get your answer.

For example, with the standard $79.95 DRA, you can specify a shift RPM which occurs at your desired starting MPH, and another shift RPM which occurs at your desired ending MPH. (This will take some trial and error and some hand calculations.) These shift ocurances are reported in the results with time, feet, MPH, etc. You can subtract and get time between MPHs. These shift RPMs would change with tire size, gear ratios, etc and would require a lot of effort on your part.

In the $229 DRA Pro, you can specify a starting MPH and MPH increment for the results to be reported at. This way you can automatically get your conditions reported in the output table, but you would still have to do the manual subtraction to get elapsed time between 2 MPHs.

In the $329 DRA Pro Team Engineer, you can specify conditions for a "Special Timer", like a starting MPH and ending MPH. Numerous other "Special Timer" options are possible, like starting MPH and to 400 feet past this point, start of 1-2 shift to start of 2-3 shift, etc. In Team Engineer, you can also automatically optimize inputs (program tries numerous different inputs) for minimizing this "Special Timer" time. This program was designed for exactly what you are doing, but is the most expensive.

Can I use my Data Logger data (from your DataMite II or my own data logger) in your Suspension Analyzer to see what my suspension is doing on the track?

Yes, absolutely! The suspension analyzer can automatically accept data from our Road Race/Circle Track DataMite software. In addition, it can read shock travel, steering travel, acceleration, MPH and other channel data from formatted text files. It can also try to determine the format of data files itself in an input file format we call "free form", to make it easy on the user. This has been used with files from several popular data loggers.

This feature is best explained with a demo movie file available at this link: DTM-RR Susp Anzr Demo.wmv

Can the Circle Track/Road Race DataMite analyze my shock's performance with graphs and histograms?

Yes, absolutely! The Circle Track/Road Race DataMite software has several graph and report options to analyze shock travel and velocity. Check out these graphs, which compare lap 4 and 5 of a particular session. You can also compare (overlay) graphs from different laps from different sessions.

(click image to enlarge)

In Suspension Analyzer, why is the Ackerman Error not the same for both tires?

The reason the errors are not exactly the same angle is because there is a "non-linearity" between how much the inside tire must turn vs the outside tire.

To find Ackerman Error, w first find the turn radius you would get from the existing tire angles. Then we find the ideal tire angles to produce this same turn radius. To do this, the inside tire will have to turn more than the outside tire, so it has a higher error than the outside tire.

In Suspension Analyzer, what is meant by "binding error"?

For several rear suspension types, when several links are attached to a solid axle or spindle, the only way for the solid piece to move is if the links attached to it "stretch" a little. This makes the math fairly complex, but we try to hide that from the user to make the program user friendly.

This stretch may be only .001", and in the real world there is enough clearance or compliance in bushings, rod ends, etc that the suspension acts just like there was no stretch. Sometimes the stretch can be much more, and (depending on the compliance in the suspension) make the suspension difficult to move dynamically (suspension bind). Exactly how much stretch will bind on your suspension is difficult to know, and is preset in the program. Obviously production suspensions with rubber bushings can "bind" much more due to this mathematical "stretching" before it is a problem.

We report a suspension as having "binding error" when we feel the user should know binding is possible on some "race" suspensions with very low compliance. Typically this means you are pushing the dynamic motion farther than what this suspension can handle, or you could try different locations for the mounting points to avoid "binding error."

The Spring Motion Ratios calculated in Suspension Analyzer do not match my hand calculations. Why?

Motion ratio has to do with the amount of movement of the wheel and tire compared to the movement of the spring. If the tire moves 1 inch, but the spring is only compressed .75", the Motion Ratio is .75. This motion ratio says how much the spring rate is applied to the tire, which is actually what is important for vehicle handling and tuning.

Wheel Rate = Spring Rate x Motion Ratio ²

So if a spring has a rate of 1000 lbs/in, and the Motion Ratio is 0.5, the Wheel Rate is only 250 lb/in, much less.

Without detailed computer programs like Suspension Analyzer, the Motion Ratio is sometimes estimated as:

Motion Ratio = A / B

from the drawing above. Additional correction can be applied dealing with the angle of the spring, or involve the C and D dimensions from the drawing above. However, all of these are making simplifying assumptions to keep the math easy. Note that both only use the lower arm dimensions, and not the effect of the upper arm.

"Behind the scenes" in the Suspension Analyzer, we are actually moving the vehicle 1" in dive and watching to see how much the spring actually moves. This is done each time you make a data entry or change a setting. This is very complex to accomplish, but is the only way to obtain the true Motion Ratio.

The Suspension Analyzer screen below shows a fairly typical suspension layout. You will note we put the vehicle through 1" of dive and the Left Spring (in red) compresses .65 inches, and shows a Motion Ratio of .65. In this layout, with the upper arm angled down in the normal fashion, the tire will actually tip up as the spring is compressed, rotating about the left (red) instant center. This lets the right ball joint drop down slightly in this condition and contributes to the Motion Ratio determination.

In this screen below, we've moved the upper arm frame mount up about 7". This is not typical, but it really illustrates what the effect of the upper arm angle. Now the tire actually tips DOWN when the vehicle goes through 1" of dive. The ball joint actually moves up instead of down. The spring is now compressed .98" in this condition, which produces a Motion Ratio of .98.

Motion Ratios for the Shocks and Roll Bar are calculated the same way. Accurate Motion Ratios are needed to determine Wheel Rates, which are critical to determining Handling Ratings, Natural Frequencies, and many more tuning parameters calculated in the Suspension Analyzer.

My Suspension is not exactly one of the choices available in the Suspension Analyzer. How can I simulate it?

There are numerous variations in suspension types chassis builders have developed that do not exactly match the types available in the Suspension Analyzer. Many times with some small adjustments, these "special" suspensions can still be accurately simulated by the program. Here are some examples:

Senneker Performance "Bird Cage": This modification to a standard 3 link rear suspension involves using a bearing on the rear axle tube, so it can float like a "bird cage" rear suspension. However, instead of 2 links on each bearing (like a 4 link suspension), there is only 1 which acts like a trailing arm suspension. Here's a link to a picture: https://www.sennekerperformance.com/product/senneker-performance-bird-cage/ This can be simulate as shown in the picture below, by specifying the link mounting point on the axle as the center of the axle shaft inside the axle tube.

Active 3 Link with Center Pull Trailing Arm: This modification to a standard 3 link rear suspension just puts the 3 link mounting hole directly in front (or sometimes behind) of the axle tube. It could be confused with the Senneker which has the bearing. Here's a link to a picture: https://www.portcityracecars.com/CENTER-PULL-TRAILING-ARM-BRKT.html This can be simulate as shown in the picture below, by specifying the link mounting point on the axle as the center of the Heim joint mounted on the axle tube.

Replica GT40 Rear Suspension: As with many independent rear suspensions, this can be simulated with a 5 Link.

Troyer torsion bar dirt modified rear end: The Suspension Analyzer does not have an option for torsion bars on the rear. However, most torsion bars can be approximated by a coil spring (not "coil over" because the shock absorber is separate from the torsion bar). First, use the Spring Rate calculator utility to determine the spring rate of your torsion bars, as shown below. Click on the "Input (clc)" label to be presented with the Spring Rate Calculator. Then choose one of the Torsion Bar options to calculate the spring rate.

Then locate the coil spring on the rear axle where the torsion bar "lever arm" touches or attaches to the axle. Locate the frame attachment point directly above that point on the frame. See pictures below:

Polaris Razor Rear Suspension: The razor uses one control arm which goes forward to attach to the frame, and 2 camber control rods. See picture below.

This can be simulated by a Double A Arm suspension where the camber rod ball joints on the spindle can be the ball joints, the rear frame mounts are the attachments at the center of the frame, and the front frame mounts of the A Arms are both at the control arm attachment point at the frame. The tie rod should attach to the frame at this same point. The tie rod on the spindle will be a point you make up, typically a few inches ahead of the ball joints and a few inches closer to the center of the car than the ball joints. It's exact location on the spindle is not critical.

Chevy Equinox Rear Suspension: This is a pretty common type of rear suspension, somewhat similar to the Razor above except it has a separate tie rod to control toe. See picture below.

click image to enlarge it

(this list will grow with time)

Shock Dyno Questions

Can you explain what the different Shock Dyno Graphs mean?

The Basic Shock Dyno software does only 1 type of graph, force vs (absolute) velocity, which most people are familiar with. The Plus version does many more types.

click image to enlarge

for a PDF explaining these graphs.

My shocks are rated at a 6 compression and 3 rebound. What does that mean?

Different shock absorber manufacturers have different number ratings. A 6 from one manufacturer can be VERY different than a 6 from another. The late Doug Gore wrote a short article on this which is in the attached PDF. Below is a shock dyno graph of 5 different manufacturer's shocks, all rated at 6 compression and 3 rebound.

click image to enlarge

for a PDF of Doug Gore's article.

Do you have any technical info on reading the Shock Dyno results and adjusting the Shocks?

Ohlins has published lots of good, technical info on interpreting the results from your shock dyno, and suggesting how to make shock adjustments.

for one of there user manuals with this excellent info.

I only have the basic version of the software. What's the most accurate way to eliminate the gas force from the data?

When the shock stops at the top and bottom of it's travel, it is at 0 velocity, but there can also be a very large change in force. This is especially true of shocks with high forces, and makes it more difficult for the software to find the exact force at 0 velocity, which is typically the gas force in the shock.

If you have the Plus software, there are methods to manually measure this force. If you only have the Basic version, this is an option:

Zero out the load cell in the recording screen before you install your shock. Install the shock. Do NOT start the dyno, just watch the recording screen and manually move the shock up and down some to reduce the seal drag, or "sticktion". Note the amount of force on the shock, which should be gas force. Then re-zero the load cell to eliminate the gas force offset before you run your test. Now you have eliminated the gas force from the data. You may also want to record the gas force, and any other info you think is important in the Comments section.

Questions on Other Tools

How much blowby should my engine produce and how much is "too much"?

This can vary greatly for different engine types and applications.

for a table to help you make this estimate.

My blowby measurement changes from test to test. Why?

It is critical that all paths for blowby out of the crankcase be blocked off except for the path to the blowby sensor. This includes the PCV valve on street motors, which must be removed and plugged. Any leaks are a source for variability.

Keep the blowby sensor at the exact same orientation for all testing. Tipping the sensor up or down at a different angle can produce variability.

Rezero the sensor with the motor not running and with no possible air flow through the sensor. You may want to plug the outlet hose when rezeroing.

The blowby sensor measures flow out of the crankcase. The assumption is this flow is from gases passing by the rings. However, it could be vapors boiling off from new oil, called the "lighter ends". These are the parts of motor oil (a mixture of various hydrocarbons) with a higher vapor pressure. There could also be water vapor in the oil from condensation. Dyno tests which take only 5-10 seconds make this type of error difficult to protect against, compared to a test taking 10-20 minutes of warm up and stabilizing at every, say, 200 RPM. If you see the blowby coming down from test to test and you don't think it is the rings seating in, it could be from these vapors boiling off and registering as extra blowby flow.

Can the Quick Cam Checker create a cam file to be read by Engine Analyzer Pro or other Performance Trends software?

Version 4.3 of our Cam Analyzer (release at the end of 2018) lets you import files from the Performance Trends Quick Cam Checker. You need v1.1 A.015 or later of the Quick Cam Checker, which is a free download NOTE: The precision of Quick Cam Checker files is not as good as those measured with the Cam Analyzer on our Cam Test Stand.

Without v4.3 of Cam Analyzer you could do the following: Files saved by the Quick Cam Checker are mostly for troubleshooting the data being recorded. They are not saved in a format which can be used by other Performance Trends programs (except Cam Analyzer v4.3). However, you can record data to get Duration at .050", Centerlines and Max Lift. Then you can change the Rated Height to .200" and run the test again to get the Duration at .200". Now you have enough info to calculate a Ramp Rating in one of our Engine Analyzer programs, Plus, Pro and Enterprise. Letting one of these programs calculate a profile from these inputs will produce a cam profile very close to the one measured by the Quick Cam Checker.

(Note: The vA.012 introduced in March 2017 lets you produce a report of a test if you upload the results to the PC before recording more tests. Then you can make a report which will give you duration at both .050" and .200" lift. This lets you avoid having to run 2 tests. for more info on this feature.)

Shown below is the Cam/Valve Train screen from the Engine Analyzer Plus. To create a cam profile from measurements from the Quick Cam Checker, first choose one of the "Spec" choices for "Lifter (profile) Type". This will make visible the Ramp Rating input. Also choose ".050 inches" for the Lift for Rating Events and enter the Centerline, Duration @ .050 and Max Lobe Lift from the Quick Cam Checker. Now click on the "Clc" button by Ramp Rating to open up the "Calc Ramp Rating, %" screen shown on the right.

In this new screen, set "Based on" to "Duration at .050 and .200". The previous data listed is already entered in this screen, as it is carried over from the Cam/Valve Train screen. You can change it here if you want. Now enter the Duration @ .200" and the program shows the Calc Ramp Rating for this combination of results from the Quick Cam Checker. In the screen below, it shows a Calc Ramp Rating of 67.0%. Click on the Use Calc Value at the lower left and 67% is copied back to the Cam/Valve Train screen. You have now created a cam profile which is very close to that measured with the Quick Cam Checker.

Can the Quick Cam Checker check a cam on the stand by barring over by hand?

The Quick Cam Checker is typically used when you turn the motor over with a starter. This gives a fairly consistent rotation speed, which is needed for good accuracy. However, with some different sensors and slightly different procedure, the Quick Cam Checker can be used on an engine stand to measure a cam.

First, you need a special 10 turn rotary sensor, and likely the larger diameter drive wheel. You will also need the head to be off so you can put a 3rd cam lift sensor above the piston to find TDC. A pressure sensor does not work accurately when barring the engine over by hand.

for info on the process, links to a youtube movie, and parts you may want to purchase.

What is meant by "aspect ratio" in the Spring Wiz program?

Valve spring wire typically has a circular or round cross section, which is much cheaper to produce. However, if you want a lower spring bind height, springs can be produced from wire which has an ovate (oval) cross section. Typical round wire springs have an aspect ratio of 1. However, ovate wire has an aspect ratio greater than 1, as shown in the picture below. For example, if the thickness of the spring wire is .140 and the width is .180, the aspect ratio would be .180 / .140 = 1.29.

click image to enlarge

sitemap

Performance Trends, Inc.

Producing Quality Computer Tools for Racers and Engine Builders since 1986

Click on one of the following categories:

Windows Vista, Windows 7, 8, 10 and 11

Click Here for General Questions

Click Here for Engine Software Questions

Click here for EZ Flow and Port Flow Analyzer Software Questions

Click here for Cam Analyzer / Cam Test Stand Questions

Click here for Engine Log Book Questions

Click here for Valve Spring Tester Questions

Click Here for Drag Racing Software Questions

Click Here for Suspension Software Questions

Click Here for Dyno DataMite Data Logger Questions

Click Here for Drag Racing and Circle Track/Road Racing DataMite Data Logger Questions

Click Here for Road Racing Software Questions

Click Here for Shock Dyno Questions

Click Here for Questions on Other Tools

Back to Top

Back to Top

Back to Top

Back to Top

Back to Top