Forum » General Tuning Discussion » Dyno to Road Inconsistency

Dyno to Road Inconsistency

General Tuning Discussion

Discuss all things tuning in this section. News, products, problems and results. 

Page 1

Hello Tuners!

I have a problem which has come up twice now with two different cars on two dynos with two tuners (one, me!).

The cars are turbo cars. Both cars on the dyno struggle to make boost. Mine was running with max solenoid setting maxing out at 1.4 bar. On a track day, I immediately hit boost cut with very little throttle at 1.8 bar. Subsequent road tuning has shown we can get 2 bar (ish).

The other car was suffering overboost on the road, but on dyno again couldn't get boost up.

Presumably this is to do with the load the dynos are offering the car on ramp runs - mine was a 6 second ramp run. The other car, however, was longer.

Is my assumption correct, or are there more subtleties to this?


what kind of dyno?

maybe u need more load on the dyno?

This was on both a rolling and a hub dyno. I couldn't tell you the manufacturer right now, unless that is needed?

On a ramp run, you increase the load by extending the duration, right?

The dyno operator wasn't keen to do this for some reason.

We need to understand the dyno type - was it just inertia based (in which the dyno operator has no control over the run length), or load-controlled? If it's a load controlled model (like a Dynapack hub dyno), then you can set the start RPM, and the delay before beginning the ramp run. If so what were these parameters? What gear ratio were you using? How about some details on the engines -- displacement, turbo size, etc.

I think RPM starting point needs to be sufficient RPM to build at least some exhaust energy, or the entire run will be an example of turbo lag, especially if the turbo is large. With an inertia dyno, I would probably try to replicate this by using the brakes as I apply throttle to full while keeping the speed down, then release the brakes to make the full run.

Both sets of runs done on load bearing dynos.

The engines are 2ltr running turbos of size similar to Garret GT35.

Typically on the road they "get going" at 3500 -> 4000 RPM with full boost between 4500 -> 5000.

We've never had trouble on the road, but first time on the dyno.

I should expect the same results on dyno as road though, right, so step 1 should be to be able to replicate road results, and go from there? I.e the question is I shouldn't see different results on road and on dyno - which suggests we've not configured the dyno right...not an oddity of the car / mapping process?

Thanks for help.

Yes you should be able to duplicate your road experience on the dyno. Let's explore what is different..

-- What gear/RPM to you start the dyno run? When you compare road results -- are you the same gear/ starting RPM? So if you start mashing the throttle at 2000 RPM in 3rd gear, what happens when you try that on the road?

-- On the road, how long does it take to reach redline from the given starting gear/RPM? Is the dyno set to this same time? If not, why not?

-- It would be helpful if you are logging data in both cases so you can determine exactly what starting RPMs were used and/or time was required.

It sounds like your testing isn't the same, for example you stated that you get boost cut with very little throttle, but in the dyno case you weren't testing at very little throttle, but at wide-open-throttle, correct? Were those in the same gear? What about throttle traces in the data logs?

Hi David,

Thanks for the thoughts. As I read them, it makes sense.

To me, boost is boost - I can't create less boost if I have mashed the throttle wide open, so when I hit boost cut on the track day at mid throttle I was surprised.

If we're considering flat out power / flat out tuning, I want the dyno to replicate the "edge" of my car's capabilities. As we were going in cold, we didn't know that the car would make 2 bar easily without WOT (nor did we want it doing that!).

We do have trace logs of the track day, but they're irrelevant as we don't have the same for the dyno session.

My thoughts are that next steps need to be:

- repeat current "on road" behaviour - understand gearing and timing to hit observed boost

- replicate the same on the dyno... if the logs say it took X to hit Y boost in Z gear then we want to duplicate that on the dyno

- assuming we can get there, then we should be able to enhance the tune

What bugs me is that we really "stressed" the car on the dyno, or that's what it felt like. Hearing the car on the dyno vs. hearing it on the road, it seemed like a lot more effort - but perhaps that is just perception. I'm convinced it's something we're not doing right dyno side, because the mechanicals are the same... just annoying that we've seen it on two separate occasions with different dyno operators and we've not been able to make ends meet.

Anyway, thanks for your tips!

Were you solid state tuning on the dyno or was it all sweep tests?

I have to say that a 6 second run, especially on a hub dyno is a short run for a full sweep e.g. 1500-6500RPM. 6 second run give an acceleration rate of over 800rpm per second, if you watch some of the dyno operation webinars Andre comments that he usually starts using a rate of approx 500rpm per second. On my dyno (Dynapack) I usually vary tests starting about 500rpm per second dropping to about 400rpm per second, the only time I shorten the runs if there is excess heat or ignition break down or general problem solving.

On a rolling is there any chance you had traction issues? It sounds like you've not managed to load it on the dyno properly, where abouts in the UK are you?

Hi Chris,

If we focus on one car as it's a simpler setup, it was 6 second sweeps on a rolling road (not hub).

The car is based in the North West (sometimes London depending on where we're racing). I think you're up in Scotland aren't you (sure I remember that from a while back?!).

General consensus seems to be too short on the sweeps.

I'm doing some testing this weekend, so will try to get a better feel for the time it takes in real world to do a pull (as per David's comment) and see how that compares.



Yeah I'm just south of Glasgow so if your in the area feel free to pop by. Let us know how your testing goes