Why you can't trust dynos

[TURBO2GO]

I have so far ran my CTSC NSX on 4 different dynos. Same car, all SAE "corrected", two on Dyno Dynamics dynos, one on a dynojet, and 3 runs on a dynomite.

My results have varied from a low of 312 to a high of 396. That is an 85 HP variance. There is NO WAY my car is acting that different on different days. Every shop claimed that the other dynos are off, and that theirs is the "true" number.

So all these comparisons on this forum... of how a guy does a mod and posts a graph, and someone else thinks "wow that is different than his graph" and it must be the mod... or how some NSX's are R motors and some are not, all these results are highly suspect. Just on two different Dyno Dynamics dynos I posted a 311 on one and 373 on the other. Now... this is the same kind of dyno at two different shops.

The ONLY relavent number is before and after on the same dyno. On all the dynos my A/F readings were flat and rock solid all the way across the band. Boost levels were all even. On an average of 3 runs on each dyno, all runs were within 1-2 HP of each other which showed the dynos were accurate within their own parameters.

My overall gain with the CTSC and GT-One header and exhaust was +113 HP. That number is perhaps relavent. I really don't know what to think of the rest.

 

[s14_tat]

i dont know, but all the dyno shops in new york uses dynojets. they all seem to read pretty damn close to each other.

 

[Alan C.]

I've had the same experience in the past. If you want to see big numbers put a hot cup of coffee under the dyno's temp sensor ;-)

 

[john@johnhudakracing]

I have so far ran my CTSC NSX on 4 different dynos. Same car, all SAE "corrected", two on Dyno Dynamics dynos, one on a dynojet, and 3 runs on a dynomite.

My results have varied from a low of 312 to a high of 396. That is an 85 HP variance. There is NO WAY my car is acting that different on different days. Every shop claimed that the other dynos are off, and that theirs is the "true" number.

So all these comparisons on this forum... of how a guy does a mod and posts a graph, and someone else thinks "wow that is different than his graph" and it must be the mod... or how some NSX's are R motors and some are not, all these results are highly suspect. Just on two different Dyno Dynamics dynos I posted a 311 on one and 373 on the other. Now... this is the same kind of dyno at two different shops.

The ONLY relavent number is before and after on the same dyno. On all the dynos my A/F readings were flat and rock solid all the way across the band. Boost levels were all even. On an average of 3 runs on each dyno, all runs were within 1-2 HP of each other which showed the dynos were accurate within their own parameters.

My overall gain with the CTSC and GT-One header and exhaust was +113 HP. That number is perhaps relavent. I really don't know what to think of the rest.

As I understand it this is Normal. I've had several operators tell me that they can honestly make a dyno show anything you want within a certain envelope. The other issue is people comparing differing dynamometers such as Mustang A to Dynojet B, along with the host of other variables from altitude/ambient/gas/charge temp/tires/pressures/maps, to usually the more significant operator and dynamometer software calibration/firmware revisions. That's just the tip of it apparently too.

It is for this reason that for best results, you need a baseline and to stick with one facility for the duration of your experience (tuning with intent) so as to minimize variables as much as possible between runs. Otherwise, take them with a grain of salt... comparing HP dyno charts- especially peak numbers on internet forums are often about as meaningful as the off topic section.

Even when the manufacturers test a new powerplant on an engine dyno they often pay to have an SAE certified employee standing right there next to pre-tagged hardware to ensure the test was carried out properly.

 

[KSXNSX]

uh oh. Someone is going to revisit the whole AIS thing again...

 

[nsxotic911]

Butt dyno is fairly accurate, no? I think I have 500 RWHP after adding better gas and cleaning my air filter :biggrin:

 

[Malibu Rapper]

You're not gonna get all proud with us over your high dyno are you? :biggrin: I got a shop close to here that might make my NA put up some bigger numbers. :wink:

 

[nsxtacey]

who cares about hp...its all about sound dave :wink:

 

[Da Hapa]

The ONLY relavent number is before and after on the same dyno.

Absolutely correct. Nothing else needs to be said.

 

[BobMc]

My overall gain with the CTSC and GT-One header and exhaust was +113 HP. That number is perhaps relavent. I really don't know what to think of the rest.

Dave, was this +113 hp on the same dyno at the same facility? Your car did feel very strong!

 

[goldNSX]

uh oh. Someone is going to revisit the whole AIS thing again...

Yes maybe there's no effect at all as I would suspect. :wink: It's just a effect by chance within the variance of the dyno itself.

I'm not astonished by the results at all. There's a big incentive for the dyno-shop to measure BIG numbers because people want to see these.
Here in Switzerland we've two different dyno-shops, legally proven dynos on the one hand, not proven dynos on the other hand. The speciality of the first ones is that the police use them for cars that seems to be highly modified and prove that they are. These dynos are monthly calibrated and more trustful as the others because noone wants to get a fine for his stock s2000 with 1'000 hp.

 

[areout]

Huh horsepower can vary significantly because of temperature, air density, elevation etc.

 

[TURBO2GO]

Huh horsepower can vary significantly because of temperature, air density, elevation etc.

We know that. But I am not changing elevation that much... all runs were done inside enclosed shops close to a 75 degree ambient temp. Plus that is what SAE correction does, any modern dyno takes all these into account. You just cannot blame an 85 HP difference on these criteria. Dynos in general are not very accurate machines when compared to one another. Their accuracy is only good in comparison to themselves.

I spent a lot of money on dyno runs... but what it shows is that there is no reason to get overly excited because some guy in some other state did a mod and posted a graph....

I could have posted my 311 and everyone would say wow there is something wrong with my car. Its so weak... I could have posted the 396 and everyone would want to know what I have done... "must be that exhaust you have" everyone might be saying. It's all the same car of course.

I plan on using one dyno and one dyno only from now on for all my measurements.

 

[TURBO2GO]

Dave, was this +113 hp on the same dyno at the same facility? Your car did feel very strong!

Yes, I went from a baseline of 249 to a high of 368 at Mechanics Direct, Mitchell's shop in Lowell. +119 on the best run. The average was +113. Same dyno. The only changes were the CTSC, and the GT-One header and exhaust. That's the only number that has any meaning really.

 

[BD]

who cares about hp...its all about sound dave :wink:

I'll take a step further!!!

How cares about hp and sound, it's all about the car you drive!!!:biggrin:

And you drive a car that is highly desirable WITH CTSC!!! Faster than my car will ever be!!!

 

[BioBanker]

Nah, I still say that it is that exhaust you've got.

Clearly it is tempermental. Works great sometimes, works terribly the rest of the time.

I will take it off of your hands. You dont need that hassel.

 

[sahtt]

.. you should already know this but different types of dyno's measure different things entirely or they use different methods to reach the power figure.

Each dyno has positive and negative attributes, even if it's just up front cost. The programming on some dyno's software also allows you to basically increase the "real" figure by a given %.

So all of these dyno's are actually "right", they are just using different inputs [whether it's inherent to the dyno or software adjustments] to produce different outputs.

Even if you use the same dyno, it's important to realize the % difference is what really* matters. 245 before and 290 after on one type will be proportional to 280 and 335 etc. on another that reads higher.

 

[clarkm]

I am relatively new to this forum and perhaps this subject has been covered in the past, but what sort of experience has anyone had with the Escort G-Timer? It claims to be able to calculate hp, and has some pretty good credentials. clarkm

 

[JeffShoots]

Butt dyno is fairly accurate, no? I think I have 500 RWHP after adding better gas and cleaning my air filter :biggrin:

FOR THE LAST TIME IT"S THE AIR FILTER NOT THE SCOOP :biggrin:

 

[Vega$ NSX]

A good way to test the horsepower is to take it to the 1/4 mile. Mikey @ Factor X can nail down a whp from a 1/4 mile run plus or minus 10 hp. It's all in the trap speed mph. You can use it to confirm which dyno was the closest. The trap speed is less dependant on the launch and should get you a good idea of what kind of whp you are putting down.

 

[Coz]

Dyno's mean nothing unless you're in front.
They are only a starting point.