Maximizing Dyno results on NA car - HOW ?

**goldNSX** · 05-31-2012, 10:34

Originally Posted by Sumeronthechain

While its a valid theory, I don't think the NSX without a power adder is going to be spinning any tires on the dyno.

Over-inflate, IMO. Also, use the lightest wheels you have access to.

I observed it nearly EVERY dyno I did. Maybe it would help to clean the dyno with brake cleaner first to avoid some of it.

**dirtbag** · 05-31-2012, 11:33

Here's some good pointers that have been floating on the net for a couple years if it already hasn't been posted. Obviously it varies between dyno types and things have advanced since then. But still good reading.

http://kennebell.net/KBWebsite/Commo...tVariables.pdf

**MvM** · 05-31-2012, 23:09

Filled up the car for the second time with 98 Octane.
One thing I have already noticed:

My mileage seemed to have improved:
98 Octane: 10 km per litre = 23.52 mpg
95 Octane: 9.1 km per litre = 21.40 mpg

For the record, I have not changed my driving style.
Funny thing so far is that this way, the higher price of the higher Octane fuel seems to be compensated by a better milage. So my average price per mile is about the same

**goldNSX** · 05-31-2012, 23:20

That's a pretty impressive difference in mpg. But there are so many factors beside your driving-style that you only see a difference after a significant numer of fuel tanks fillings.

**MvM** · 05-31-2012, 23:34

Originally Posted by goldNSX

That's a pretty impressive difference in mpg. But there are so many factors beside your driving-style that you only see a difference after a significant numer of fuel tanks fillings.

I agree with you. However, it's my FIRST time I'm getting 10 km/litre EVER in the NSX. Just filled the car up again with 98 and will keep doing that at least until after the dyno day.
This weekend I will be cleaning the UNI filter and take a look at how dirty the intake seems to be. Perhaps cleaning that will also help a bit.

**goldNSX** · 05-31-2012, 23:58

Originally Posted by dirtbag

http://kennebell.net/KBWebsite/Commo...tVariables.pdf

That's a pretty good article even though it contradicts to my observations. But again, dynos in Europe are more sophisticated than in the US. A run in Europe takes up to 60 seconds.

**MvM** · 06-07-2012, 07:29

This morning I had the Dynorun with the people from the Dutch Autoweek car magazine.
The guys were very friendly and also very interested in the NSX.
I told them about the modifications I have made to the car, of which, for the dynorun, the intake scoop, UNI-filter, Taitec headers and GTLW exhaust were the most interesting of course.
And naturally, I was very curious to see the results.
I have driven the car on 98 Octane gas for the last two weeks, cleaned the UNI airfilter, checked the oil, check the tire-pressure, run a bottle of injector cleaner through the car and then it was up to the NSX itself.

The results were VERY surprising.
I had told the guys from Autoweek (and the guy from the tuner shop) that standard output for the '98 NSX is 280 HP (in Europe) and that I expected a bit more than that because of my modifications.
The previous owner had a dyno done when the car was still stock and the results then was 285 HP at the dyno. So, I expected perhaps 20-22 HP more than that.

The result: An amazing 355 HP AT THE ENGINE !!

Of course, I was a bit surprised. This was not a Dynojet type of dyno, but one with an internal brake which is able to measure the engine power. This kind of dyno is fairly typical in Europe but usually, these also lower results than a Dynojet.
Of course, I expressed my surprise about the unexpected results and left a very happy guy. Never said anything about the results, except to a close NSX friend who was there for mental support.

I have asked for and received the raw data from the run and have taken a close look in it. It turns out that the shop uses a conversion factor of 1.36 for kWh to HP and 1.25 to convert raw Wheel HP to Engine HP which is a bit high for the NSX.
So, I have recalculated the results using the often used 12% power loss ratio for the NSX.

Even then, the results are still very healthy with 286 WHP / 323 Engine HP at 7000 rpm.

Here is a picture of the graph:

When I left, some other cars (they were doing 7 cars in a row) had already arrived. I am interested to learn how they will perform on this same dyno.
For your information, Autoweek car magazine has done a number of these dyno runs in the last year and usually, the cars they test are showing lower numbers than the official claimed output. For example, a F430 Spider F1 had 450 HP instead of the offical 490 HP.
So, what do you guys think?

**goldNSX** · 06-07-2012, 08:07

Originally Posted by MvM

For example, a F430 Spider F1 had 450 HP instead of the offical 490 HP.
So, what do you guys think?

In short: you should add the CTSC and beat that cheating F-car.

I'm not used to this kind of dyno. If it was a MAHA I would have trusted it more.

No way it will have 355 engine hp with these mods.

If you fitted your CTSC then yes.

What data are they publishing now?

**MvM** · 06-07-2012, 08:21

Originally Posted by goldNSX

In short: you should add the CTSC and beat that cheating F-car.

I'm not used to this kind of dyno. If it was a MAHA I would have trusted it more.

No way it will have 355 engine hp with these mods.

If you fitted your CTSC then yes.

What data are they publishing now?

I agree with you that 355 Engine HP is not possible with simple bolt on mods. That's why I recalculated the raw data.
Personally, I had expected at most something like 310-315 HP.
My previous 3.0 NA1 NSX was dynoed at 276 WHP (314 Engine HP) with basically the same mods.

The data Autoweek will show in their video and magazine will be the 355 HP because that´s what the guys from the racing tuner shop said it is.
Will be good advertising for the NSX I think

**goldNSX** · 06-07-2012, 08:28

Does it feel like a 355 hp car or a 323 hp car?

Why is the torque shown just every 1000 rpm?

**dablackgoku1234** · 06-07-2012, 08:33

I'm interested to read the whole article to compare the different cars, scan it when it comes out!

**Sumeronthechain** · 06-07-2012, 08:37

Seems a bit too high for me to believe. But hey what do I know?

**MvM** · 06-07-2012, 11:38

The data I got was spaced at intervals of 250 rpm.
But I agree fully that it seems very lineair between between the full 1000 rpm numbers.
My guess is that they shop simply doesn't measure all the data.
I have other data from my previous NSX as well which was spaced at intervals of just 5 rpm. Made for a spreadsheets with 1260 lines.
Perhaps the shop simply didn't want to spent that much time.
Don't forget they had 7 cars to do today, complete with all the photography and videowork for the articles and video so they might have made some shortcuts.

I will post a link to the video and upload the article when it comes out.
Don't know a date yet of course and they will only publish one car at the time in every issue so it might take some time.

As for the feel of the car, the cars certainly doesn't feel like a 355 HP NSX, more like a 323 HP car. And even that might be a bit optimistic.

**N Spec** · 06-13-2012, 08:55

This is an interesting situation. The tuners believe every car loses 25% of power to the wheels, but the NSX has proven on the dyno that 12% is usualy the difference from the claimed flywheel hp. It is also interesting that I have seen many different Ferraris dyno a bit lower than their claimed hp, as in closer to the 25% loss in powertrain vs the usual 18%.

My humble opinion is Honda just underrates their engines very often, because I find it hard to believe that they are achieving only 12% drivetrain loss(even with a more efficient tranny/drivetrain), whereas everyone else is a lot closer to 20%-25%.

Ultimately, what matters is the whp. Impressive numbers either way, ~290 whp is not exactly slow for a ~3000 lb car. I can't wait to see the article.

**MvM** · 01-08-2013, 05:02

FINALLY, the video of the Dutch Autoweek car magazine appeared on the website.

You can find the link here:

http://www.autoweek.nl/video/3188/op...nk-honda-nsx-t

Of course I know most of you guys will not be able to understand the language, but the numbers will be recognisable