# Thread: Dyno talk - let's pull all the facts, and maybe even rumors about dyno differences to one place

1. OK - So we always hear that dyno X reads higher than dyno Y, and that dyno A is more consistent than dyno B. We also know that dyno operator John vs. dyno operator Mike can influence more of a difference in the results than anything else.

These things are bantered about in every engine technical forum, but rarely with any supporting data.

Are there any 'case studies' of dyno data out there we can share with each other to build an approximation of what the differences are between inertia and load style dynos, dyna-pax vs. dyno dynamics vs. mustang vs. dynojet, etc.

It would be cool to collect this stuff in one place.

2. We currently own 2 dynos. (Superflow Chassis and Engine dynos).

First re: the Chassis or wheel dyno. There are a minimum of 2 types of dynos. Inertia and brake loaded. The inertia dyno simply has a large roller (or set of pinch rollers) that are driven by the wheels. This roller usually weighs several tons and can represent as much as 3000 lbs of mass. During the run, the computer measures the acceleration rate of the roller and computes HP each few milliseconds. It then uses the formula HP=RPM x Torque / 5252 to back into the Torque at the those RPM points.

The Brake looaded dyno does it a little different. It puts a breaking force on the roll via a water brake or Eddy current brake. During the run, a strain gauge is used to measure the Torque at many RPM points. After the run the same formula to compute the HP.

Our dyno can run in either brake loaded or inertia mode. (Some can only do one or the other.)

An inertia dyno has the best repeatabilty and accuracy when used correctly.

A brake loaded dyno is generally not quite as accurate, but has more capability for tuning methods.

Another thing muddying up the water is correction factors. To compute the HP a correction factor is used. This is SAE, STP, STD or one of several others. It is based upon Baro pressure, Altitude, Temp, humidity. It is used to insure that the HP of the car is comparable from one day (or even hour) to the next. Some dyno operators use different correction factors. (and or don't have true weather stations that input the data to the computer real time.)

More coming....

3. Since the dyno is really a tuning tool, accuracy isn't terribly important. Repeatabilty is. What you are really wanting to do is tune the car for maximum output. Whether it is 506 HP or 495 HP doesn't really make much difference. What is important is the shape of the HP/TORQUE curves and is it better or worse than the last run.

The different chassis dyno makers also have some internal calculations the attempt to make up for things like aerodynmic and frictional losses in the rollers. Small pinch rollers require lots of adjustments to compensate for the frictional losses. In our experience, most all of the major dynos in inertia mode are pretty close. It seems DynoJet reads in the 5%+ per cent high range. This could be the above calcs or simply a marketing effort. They told me at a PRI show, that "the higher your dyno reads, the more customers you will have."

Assuming a chassis dyno in inertia mode, there isn't much the operator can do to throw the numbers off. (Unless he blatently cheats.)

More... Engine Dynos..

4. MOST engine dynos are of the brake loaded type. SOME of them have the operator control the load manually. This does allow an operator to fudge a little, either consciously or uncomsciously. Better dynos automatically control the load and acceleration rate of the engine. Typically 100, 200 or 300 RPM per second rates are used. Measured HP will vary at different rates and the more the engine rotating mass the more the variance.

Again, what you are really interested in is is it getting better. Not whether it is 506HP or 495HP.

BTW, We built our first dyno in about 1967 using a Hydromatic transmission with a scale and a locked ourput shaft. My HP/TORQUE readings were probably 20% off, but I knew when it was getting better or worse.

Jim

5. And finally......

When a chassis dyno is used the car should USUALLY be run in a gear that is closest to 1:1. It will be close in most any gear above 1st, but there are usually some additional thrust loads in low gears and the accelleration rate might high enough so that engine inertia losses become prevalent. That being said, running a 90HP car on a 2500 lb inertia dyno may take all day to get to 6500 rpm (or whatever), so 3rd gear may be preferred. Again.... The exact NUMBER isn't important. Only repeatabilty is.

Hope this helps....

Jim

6. It's pretty futile to use a chassis dyno to do engine development work unless you are talking major power increases and that is why all car companies use engine dynos. As you know with an engine dyno you can control the majority of variables and get excellent accuracy and repeatability if the dyno is well maintained and the tests conducted to SAE or similar standards. All of the B.S. factors of chassis dynos is eliminated and you get honest, accurate results with a properly conducted engine dyno test.

7. An engine dyno is not the "best way" to do it. It's just another way to do it Regardless, if a motor makes 500chp (crank), but only makes 375whp it's better to tune for the drivetrain losses. Of course, this can only be done on a good loading dyno where steady-state tuning can be done.

This brings us to another point. An inertia dyno for anything other than making WOT runs is just an expensive toy. It doesn't have much use other than producing a graph saying you made 'xxx.x' amount of power to the wheels. Even then, the numbers are ordinarily higher than reality.

Inertia only dyno manufacturers eventually got tired of hearing these complaints from their customers and began either incorporating retarders (electromagnetic brakes / eddy currents) into their dynos or simply offering it as an option. There are several problems with this design.

1.) While attempting to hold the vehicle under load, the retarders have a tough job. They're trying to control the entire mass of the roller(s). These rollers can be quite large. You can probably see how this is not a great design? The retarder(s) will do their best to hold at say, 3000rpms, but when modulating through the different throttle and tuning the different load points, the RPMs or wheel speed will have trouble holding and will actually wander around. This makes tuning difficult resulting in longer tuning sessions.

2.) Continuing from above, the wandering gets even worse the longer tuning goes on. The coils in the retarders begin to overheat and you get to a point where they must be cooled down before any more tuning can be completed.

3.) These loading-inertia dynos (some, not all) also can't provide necessary tuning data while holding steady-state like torque or hp. You need this type of feedback so you can see if your increases or decreases of ignition, fuel, boost, etc are making positive or negative gains.

There are several dynos like this on the market. I don't want to get too much into brands, but you'll probably ask which ones they are. Some of the more well known ones offered are; Dyno Jet (retarder offered as an option), Super Flow (retarders standard on inertia units), Mustang (retarders standard on inertia units), Dynocom (I won't even go there LOL), and maybe a couple that I'm forgetting.

So let's talk about low-inertia, loading dynos. These are built around a steady-state design out of the box. Some of these dynos are made by; Mustang, Land & Sea, Dyno Dynamics, Super Flow, Mainline (not well known in the U.S.). The nice thing about these dynos is their ability to hold load in steady-state and do it for long periods of time. They also have a small footprint generally speaking. Some of them won’t give as much data details as others though. For example, Dyno Dynamics will tell you wheel torque, but more importantly will tell you ‘tractive effort’. This is the work that’s being done at the contact patch of the tire. It’s considered far more important than wheel torque. It will also tell you what your wheel horse power is as well all while holding the RPMs steady under load.

Unfortunately, some of these loading dynos have an issue with ‘wandering’ while trying to hold load in lower RPMs. Not quite as important for some of the 4 cylinders (unless running a small turbo), but the big 6s, V8s, V10s and V12s want to tune these areas out easily without the hassle of the load wandering around. The issue goes away the when the RPMs are allowed to be brought up.

Yet, there are more differences. Traction comes to mind first. These low inertia rollers generally incorporate two smaller rollers (rather than one big one). When designed properly, they are able to hold traction better than the large inertia drums and even do it in a lower gear. The reality is that most of these have a belt that connects the rollers. This is why you’ll see some of them state that it is traction limited. It’s not the size of the rollers that’s the problem, it’s the fact that they’re connected together, the knurling design of the rollers, and the strapping method.

Strapping rollers together also creates inconsistent readings. Especially as wheel speed increases, the rollers are forcing the tire to spin at the same speed in between them. This is not desirable since the tire is wanting to turn into (move forward) the forward rollers. The tires then begin to get pinched down between the rollers etc etc.

Dyno Dynamics is designed around a non-belted roller design, meaning that the front and rear rollers are not connected. In fact, the rear rollers are smooth, light, and are independent of each other. While the drive rollers are connected all the way across to the retarder, are light, and have a square cut knurling. The wheels are actually pulled down into the rollers, forcing them to dig. When making enough power, the tires will actually climb up to the 11 o’clock position. If properly strapped down, Dyno Dynamics dynos have held traction to 2000whp and maybe more but no car has ever been able to make more than that (actually 1952whp and the transmission popped).

So why would you want low inertia rollers? For one, some are capable of picking up a single misfire or small detonations. The lighter the roller the easier this is. I can tell you that on Dyno Dynamics, this is the case. Most people are blown away that it can do that. Not to mention, if there's ever a drivetrain malfunction or you pop the motor you can actually slam on the brakes. On an inertia dyno, if you slammed on the brakes to prevent further damage from happening, it would not be a pretty site

Strapping vehicles down on either an inertia style dyno or a loading one is very important for traction. Pretty much, most of the dynos on the market strap the vehicle down by holding it from the front and the rear and the weight of the vehicle is what helps keep traction. Most all of the time, they say you’ve got to be in a 1:1 gear. While true for inertia dynos, not so for most loading ones. On a 5spd, 4th gear would normally be used. Traction is better in 4th than 3rd. Dyno Dynamics dynos are actually designed to be operated in lower speeds, which is impressive since all that power can be held without spinning in 3rd gear.

The next kind of chassis dyno are hub dynos. The most well known is DynaPack. I actually like these a lot as wheel slip is never an issue and is probably the most compact. This is one of the only things I like since getting cars on and off is time consuming. They’re generally more expensive than any of the other chassis dynos and require water hook up. Also, they don’t have the ability to get a good ignition pickup for automatic transmission vehicles. Additionally, they’re reading what’s happening at the hub and not the contact patch of the tire.

Accuracy between all dynos. Some dynos have shown to be very inaccurate. See this article http://www.factorypro.com/maga....html I think most of them can be fairly accurate when following the same strapping procedure each and every time, making sure that tire pressure is up, and testing on the same unit. It’s only when you go from dyno to dyno, same brand everything, but the way one is set up versus another is where the changes begin to happen. On Dyno Dynamics, there’s something called a “Shootout” mode. It fixes all of the variables so that nothing can be changed. This is really nice when you want to compare numbers from one dyno to another. Or even when you have a dyno day and you want the conditions to be the same for each and every vehicle throughout the day.

Numbers. It’s widely known that some of the numbers produced from various dynos are worthless but they do help our egos. I had someone tell me recently that Dyno Dynamics numbers sucked and that the “math” was wrong. He then told me that his Dyno Jet numbers were dead on and how could he sell his customers software with the low numbers like that. I told him that the numbers were real and he said that they couldn’t be because he knows his stuff At the end of the day, what really matters is what the car made, before-and-after results. Those are real-world gains. Besides, take it to the track and see what your trap speed is to back up your dyno numbers. That will tell the tale If someone is supposedly making 400whp (400 real HP to the wheels) then the trap speed should reflect that (providing the clutch is good shape etc).

Graphing. This is huge. To the best of my knowledge, no other dynos out there can produce a live graph as the run is taking place other than Dyno Dynamics. As a matter of fact, I believe it's also the only one that doesn't require 'graph smoothing'. Why you ask? If you need to smooth out a graph, then you've got a problem with the tune. Think about it, if you go back in and 'smooth' out the graph to make it look pretty, you have just hid the important details. Dyno Dynamics doesn't give you this choice, either it looks good or you need to go back and work on it. All the other make you smooth their graph. Generally speaking, the resolution in the graphing is lower on most of the other brands forcing the smoothing issue.

I’m sure there’s a lot more but I don’t feel like typing anymore

Modified by DD Todd at 9:38 AM 2-9-2008

Modified by DD Todd at 9:39 AM 2-9-2008

Modified by DD Todd at 9:45 AM 2-9-2008

Modified by DD Todd at 3:51 PM 2-9-2008

8.  Quote, originally posted by raceware » It's pretty futile to use a chassis dyno to do engine development work unless you are talking major power increases and that is why all car companies use engine dynos. As you know with an engine dyno you can control the majority of variables and get excellent accuracy and repeatability if the dyno is well maintained and the tests conducted to SAE or similar standards. All of the B.S. factors of chassis dynos is eliminated and you get honest, accurate results with a properly conducted engine dyno test.

I would argue this point. If you are careful to keep the conditions consistent from run to run, including the additional variables that the driveline/running gear present, it is a benefit to tune the system as a whole in the car, as it will operate on the road/track.

I do like the dynapac system as it removes what I find to be the most inconsistent variable - wheels/tires and their condition/pressure/etc. I also like it because GRD has 'sponsored' my car by letting me use theirs for a great rate, so maybe I am biased. The dyno-dynamics information above is really great to read - that is the other dyno that I am most familiar with, as we use it to tune another car that I crew with that lives in St. Louis area. That said, we are trying to put together the accessories needed this year to run the motors on an engine dyno over next winter - just to make lots of tuning easier (no need to drag the car to the dyno and strap it down every time - just build the 'B' motor to match the 'A' motor and leave it there to lots of iterations of testing).

9. You guys seem to be the right people to answer this question.

I had my 08 gti run on the dyno recently and was pretty pissed off with the results. The operator swore he was doing it right but After 5 passes on the dyno with readings all over the place i was sick of it and just told them I would go back to my tuner to do it on there dyno.

Here is what took place: The operator did not run the car with a load (the drum). Now I did five runs on his dyno with crazy readings. the first few were reading around 135hp and like 145tq. The best reading i got was a 188hp 207tq. on the last run. During the last 2 runs I watched how he revved the car up and it didn't seem right. He said he was hitting max boost at 56 mph in 6th gear. Now I know the guy was Bs'ing me because when you floor a tiptronic in 6th @ 2k rpms it will down shift to 3rd and then put out max boost. As I stated I watched the boost gauge and it never went past 15lbs of boost.

My car has these current mods
GIAC Reflash
Boost increase from 14lbs too 21lbs
Intake
Lightened Pulley
deleted noise pipe

Now my question is as follows:
The stock gti has been hitting around 175-180 whp. With such a large boost increase shouldn't I be hitting over 200hp and how much effect does no load on the drum have on actual whp.
It just seems to me that in order to have a car put out maximum power it should have a load on it to produce the most whp?

10.  Quote, originally posted by SR Racing » Another thing muddying up the water is correction factors. To compute the HP a correction factor is used. This is SAE, STP, STD or one of several others. It is based upon Baro pressure, Altitude, Temp, humidity. It is used to insure that the HP of the car is comparable from one day (or even hour) to the next. Some dyno operators use different correction factors. (and or don't have true weather stations that input the data to the computer real time.)More coming....

Anymore on how these correction factors work? Do they take the ambient conditions and calculate a density altitude or to ISA where the dyno pull was made then correct for errors at the altitude?

11. So much difference of opinion, go to the SEMA show and every dyno manufacture tell you they are the best, most accurate piece of equipment on the planet! They usually have huge booths and most have some extremely custom or exotic car on there top of the line model that most buyers will never see but a few times, and even then have nothing to compare it to. Want to impress me show me a line of new rental cars all in stock form running on this dyno over and over and then give me all the data to compare to factory specs. Why not show crank horse power and crank torque, as well as wheel, real time air fuel, real time boost, temps, and more. Why not give me something I can compare to the factory #s ? answer> mostly because most dyno manufactures are not huge global companies, not to say that they donâ€™t sell globally but companies like dynodynamics started out as a couple of guys in Australia trying to find a way to measure the power of the boats they repair, dynojet started out small with motorcycles. Its all well and good and a lot of these company have extremely smart talented engineers doing very good work but the fact remains they donâ€™t have the budget to do all the research and development to make a truly accurate piece of equipment. In short they do the best they can to offer an affordable piece of equipment. Ok so by now your thinking what a blow hard, and thenâ€¦. what is the answer, its easyâ€¦ run and tune your car on a dyno that the car was designed and engineered on in the first place. One of the other post stated that they have a 08 GTI, and got horrible unrepeatable results, well the only way to compare is to run it on the same dyno that Volkswagen did in the testing lanes at the factory. The answer to all the arguing is to use a MaHa dyno, ya I know what is a MaHa; few have ever heard of one let alone used or seen one. Well itâ€™s the dyno that the majority of the worlds manufacture use. VW/Audi, BMW, Mercedes, Porsche, Ferrari, Mitsubishi and on and on. MaHa is the dyno that most of the manufactures do all there tuning on and is the only dyno that is factory approved (that means that you wonâ€™t void your warranty by running your VW on it). No more unplugging your Haldex unit to run your TT or R32 ether. Whatâ€™s the catch?! There are only 3 public units in North America. Why am I so happy about that?!! I just got the third. More info go to Goodspeedperformance.com

12. You won't void your warranty by running your Volswagen on any dyno.

13.  Quote, originally posted by chois » Volswagen

This spelling thing is getting out of hand, Chris. Your friend, the dyno queen!

14. Naw. I was just referring to a VW from Tennessee. They're all vols down there.

15. I was thinking about something today. Aside from different dynos reading different base numbers, what if 40 hp/tq was added to the engine, would these dynos all recognize this as being 40? Or would the lower-reading dynos show 20 percent less?

16. I always assumed that you would see reproduceable % changes across platforms, but now that you mention it - I can see where that might not be the case, especially for the systems that rely on accelleration time and mathmatics to calculate torque, rather than measuring it.

17. Maybe we should link outside the Dyno forum?

18.  Quote, originally posted by SR Racing » And finally...... When a chassis dyno is used the car should USUALLY be run in a gear that is closest to 1:1. It will be close in most any gear above 1st, but there are usually some additional thrust loads in low gears and the accelleration rate might high enough so that engine inertia losses become prevalent. That being said, running a 90HP car on a 2500 lb inertia dyno may take all day to get to 6500 rpm (or whatever), so 3rd gear may be preferred. Again.... The exact NUMBER isn't important. Only repeatabilty is.Hope this helps....Jim

Is there a measured difference in HP/TQ from gear to gear? If two cars with same engine but slightly differring tranny gearing both dynoed in same gear, but to due slight gearing difference would the dyno read a different figure (assuming engines output and drive loss were equal)?

19. I think the difference between dynos vs. the difference between diff gear ratios is greater for sure. When we dyno'd a few of the same cars in different gears, they were VERY close. So close that it wasn't an issue. I'm not comparing 2nd to 5th of course. Obviously you would want a gear ratio closest to 1:1. Any answers to the previous question???????

20. I have a 2000 1.8T Jetta and looking to add some mods. Performance exhaust and down pipe. I also have a 2002 Passat and looking to add the same mods. Any suggestions?

21. the purpose of dynoing in your closest gear to 1:1 is to get rid of any gear underdriving or overdriving. gear ratios over 1:1 such as 4:1 provide torque multiplication.........HP is figured by torque, rpm, and a constant(5252)..............therefore if you have 150ft/lbs at 1:1 then at 2:1 you would theoretically have 300ft/lbs at the same rpm.

22. ive seen ALOT of dyno places do pulls in 3rd gear, cars that ive researched the gear ratios and 3rd is underdriving quite a bit. one of the easiest ways to cheat your customers

23. Not totally true. Some dynos allow you to enter the total gear ratio, so that they are able to accurately measure torque.

24. yes but yet again, that requires the operator putting in the ratio.

25. agreed

26. Telling the dyno what gear ratio you happen to be in is not going to change the measured tq. There is not even a correct way to calculate drivetrain loss accurately. We have the ability to enter the gear ratio, but this is only used to generate a tach signal on diesels, or cars that are a pain to get an accurate rpm pickup from the usual methods. I am only intimately familiar with the dyno we have, but I know it is not possible to calculate real-world hp/tq from whp by knowing the gear ratio of the transmission. It would be like saying "FWD drive cars have 10% drivetrain loss" which is not always true especially between different car manufactures. If there is in fact dyno software that will manipulate the measured tq. and correct it based on a ratio input from the operator, it would be at best a very rough estimate.

Modified by Dubmekanik at 3:05 AM 10-19-2009

27. Right. Torque is measured, and the values won't change, they would just be assigned to the wrong engine speeds, and thus the HP values will change with gear ratio errors.

edit - looks like I typed the wrong word a few posts up - meant to say hp not torque...

Modified by chois at 7:29 PM 10-19-2009

28. im running a v9 stg1 s/c on a 2000 vr6

do you have to get the ecu upgraded first before taking it to get tuned?

29. I've been doing some hands on research of this for almost 10 years now
I'm nearly ready to put it all together into one easy to access document
http://forums.vwvortex.com/showthrea...from-2002-2011

30. My car was run on a dyno before I bought it, the dyno showed the car was at 170WHP and about 185TQ after my stg.1 software my car was running 210WHP and 280TQ. I was wondering if this sounded right, it seems like it is since WHP is less than BHP.

31. Can any dyno operators go into detail about how dyno's struggle with Haldex and other torque limiting factors in automatic transmissions, it annoys the hell out of a lot of R32T owners here that they can't get correct readings for their cars.

32. Listen I just had a sh_t ton of work here and it was done damn right

http://enhancedstreetperformance.com

Dave the owner was great to work with also and does "Dyno Days"

Check them out they do it all ~

Listen I just had a sh_t ton of work here and it was done damn right

http://enhancedstreetperformance.com

Dave the owner was great to work with also and does "Dyno Days"

Check them out they do it all ~
Dave's been my dyno man for the past 4 years. Great guy, great dyno days. Unlike other dynos, I don't need to make a big deposit and then collect the payments from the participants (who frequently no show)
I'm able to book the day and he gets me staff and I get him people. I average 20-25 people in a day starting at 830am and ending around 430ish. They provide food and drink. The viewing area is immense, so there's little to no crowding. Lots of area to shoot video and photos

I've also perfected the 'efficient dyno day' with him... I provided some spreadsheets for signup and running that has gotten us into a rhythm of about 1 car every 20-30 minutes. We do all the registration information up front (the day before). All cars are input into the system one day ahead of time so you just give the operator your name and you get on the dyno... no waiting around as they input your info.
On top of that, we basically have a 'batters box' where you wait for the guy ahead of you to finish.

It's all very common sense, but actually DOING it is a different story.

Here's the last video

check out my channel for LOTS of dyno video

Listen I just had a sh_t ton of work here and it was done damn right

http://enhancedstreetperformance.com

Dave the owner was great to work with also and does "Dyno Days"

Check them out they do it all ~
No sure if you have some affiliation with them but please refrain from posting that same stuff in my own Dyno day thread. Im not interested in someone advertising someone else's services in a thread for my shop's event

Thanks

35. Question:

About 3 years ago I dyno'd my old setup on a dynojet. The operator asked us if we had USB sticks. If so he could give us the WinPep file for the run. I happened to have one. I took it home and opened it in WinPep. It allowed me to choose between the "smoothings" we often see and also to turn OFF the SAE correction method for the atmosphere that day.

How did turning this on/off vary a full 25whp on a 370whp run? The temperature outside that day was a perfectly average 70 deg F and normal barometric pressure.

#### Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts
•