to vividracing:

on your top graph which line represents afr and which is boost?

 

It shouldn't do anything. The only thing I am considering to do is to raise the ramp up rate slightly to better represent the EVO's shorter gearing. For example, our dyno results for the EVO use a ramp-up rate of 100. Dyno Comp used a ramp up rate of 200. That means our run lasted nearly twice as long as his (say, 5 seconds instead of 10 seconds). Our method will place more thermal loads on the car and make it more knock prone. I use 100 for moderately stock WRXs. I'll probably end up bumping it up to 120 for EVOs since they have much shorter gearing and tend to accelerate considerably faster as far as engine RPM goes. This shouldn't raise the results more than 1-2%, all things equal. And it will only apply to EVOs, not WRXs.

The biggest cause for the hp differential between our dyno and that of Dyno Comps had to do with barometric compensation. They were using standard corrections for their slightly lower barometric pressures. This introduced a rather large positive correction factor. We don't use any such barometric compensations for turbo cars from Audis, Mitsus, VW, Subaru, etc,. Their engine control systems naturally compensate for slight barometric differences by running slighly more or less boost (above ambient). In other words, they all tend to regulate their boost pressure to the same absolute pressure or mass airflow. Such standard barometric compensations really only apply accurately to naturally aspirated or belt-driven SC'd cars. So, we will won't use them.

We also used different air temp sensor placement and inertial factors which also influenced final raw numbers.

Needless to say, Richard and I are taking advantage of having the same dyno and will be doing our best to make sure they see eye to eye.

Cheers,
Shiv
www.vishnutuning.com

 

Damn that's short!

-Nathan
4-second settle, 15-second ramp time, 4th gear. Sometimes when I am feeling real nasty, I’ll torture test a map at 30-second ramp time.

 

Wont changing these variables destroy the ability to compare the resulting numbers to other cars, namely the wrx? This seems like a tilting of the numbers in favor of the evo, or maybe I'm misunderstanding.

 

nmyeti-- Those durations were just examples for comparisions sake. Depending on the car and what we're trying to test, a dyno pull could last as little as 10 seconds or go as long as 20. Anything more than that is best left for applications where I'm trying to test the towing capability of a car

Nzo-- The numbers should still be comparible. Sometimes its easy to forget that a good dyno test replicates real world situations. In the case of the EVO, the shorter gears will make it ramp up faster on the road. Trying to slow it down on the dyno to behave like a relatively long-legged WRX isn't very realistic or fair. But, in the end, I wouldn't expect more than a 1-2% change so there's not much to be concerned about. I think whatever small change there is will be for the better.

Also, if you notice on the graph on page one, the ramp up rate was set so high that full boost didn't occur until ~5000RPM. As a result, you'll also notice that the car didn't even have enough time during the dyno run to develop max torque at the rated ~3800rpm. This is why it doesn't look anything like a standard EVO torque curve.

Cheers,
shiv
www.vishnutuning.com

 

Shiv,
Just for clarification sake here, are you saying that you tune different WRX's to different load and ramp times?

 

shiv, do you use different ramp up #'s for different cars? or in the past have you used the 100 on all cars.. s4,wrx,dsm,etc?
--bobby

 

any idea when vivid will have updated #'s from the equalizing of dyno settings?

 

The ramp-up rate induces a load. There is no discrete "load" input. Load is generated when the tires apply torque against the rollers. Is the roller ramp-up rate which is controlled by the operator to replicate the desired situation.

That said, we do tune cars with ramp-up rates best representing their real-world acceleration rates. In the case of a stock WRX, the ramp up rate will be reasonably slow. For a 500hp WRX, it will be considerably faster. Of course, at the end of the session, it's not unusual to test both ends of the spectrum to ensure that no det is incurred and that boost level is controlled. This is very much like running the car in a higher and lower gear when running on a Dynojet.

BobbyD-- We typically use ramp up rates between 100-120, depending on the car and its real-world acceleration capabilities. It really doesn't make much of a difference as far as peak numbers are concerned. But it can affect the spool up characteristics (and, as a result, the shape of the torque curve) of a turbo car. Especially one with a big turbo that needs as much time as possible to bring it up to speed.

Cheers,
shiv

 

It has been reported to me that the "ramp rate" setting has a great deal of impact on the final dyno graph. From my understanding, it affects both the peak HP and the shape of the curve. For instance, when Dyno Comp first opened up they dynoed a mostly stock WRX (hacked air box IIRC) at around 188hp or so when they were in pure inertia mode and around 160 when they were using middle of the road ramp rates. I find it interesting that this same brand of dyno is often used in Australia to dyno stock WRXs at around 130whp. IIRC one of the dyno comp guys mentioned that when run at very high load settings the stock WRXs make about that power over here.

I fully understand the idea of matching time actually spent in the dyno pull to the time a car would spend in the gear on the street, but clearly it appears to sacrifice consistency on this type of dyno.

So the question becomes why not use the same settings all the time?

 

I'd agree. That would make it consistent right? So why wouldn't you set it and then throw away that knob so it couldn't be messed with?

It almost sounds like you could "dial" hp in or out with just that. That could be misleading at times I think.

 

I can't speak for Dyno Comp, but the range of ramp-up rates we use don't have much of an effect on wheel output. I don't think I've ever even used a ramp up rate as big as 200 so I can't comment on its effects.

I usually keep it between 100-120, depending on what conditions I'm trying to reproduce. FWIW, that's like the difference between a 10 second pull and a 12 second pull. If those extra 2 seconds of load time is yielding significant output changes in either torque or hp, I'd say that there is a problem with the car or testing proceedures.

I honestly can't think of a more repeatible dyno based on all the others I've used. But like everything else, a dyno is a tool. It can be mis-used just as easily as it can be used. One just has to know what he's doing. Playing the numbers game is fun. But it's how the car behaves on the road/track that is important. The dyno is simply a tool used to replicate those conditions as closely as possible while holding other other variables as constant as possible.

Cheers,
shiv

 

Found some differences in why our results were higher then Shivs...

Look for new dyno numbers and our retest with our Cat Back on it!

Stay Tuned...

 

cant wait to see..

 

yeah i am looking forward to seeing the revised results

also it looks like i may end up with a 99 eclipse to mod so an Evo is on hold