Interesting article on dynos

[EspenW]

Inaccuracy and variation are keywords:

http://s-cars.org/postnuke/downloads...sdynoAug04.pdf

[jogi730]

Very interesting article. I guess the only way to get true crank bhp is to pull the engine and do it that way. That leaves thew question - what sort of discrepancy is there between this type of engine dyno?
Cheers
Chris

[EspenW]

Very interesting article. I guess the only way to get true crank bhp is to pull the engine and do it that way. That leaves thew question - what sort of discrepancy is there between this type of engine dyno?
Cheers
Chris

Good question. An engine dyno eliminates many of the components and variables of a rolling road, so that will increase accuracy, one must hope.

[quattro84]

Yes, but running the engine outside the engine bay gives other issues if not sorted by the operator.
That's were Point #2 of the article comes to play...
I work at a site with 34 similar dynamometers and we have to be able to swap engines internally without losing accuracy. That asks for calibration of all the measurement devices twice a month. Next to that calibration of the primairy sensors & emission bench takes place multiple times a day!

To be true to measure under similar conditions you should find pressure & temperature drops/ratings coming from the vehicle and setting it up on your dyno. How? Make everything adjustable into a (preferable digitally controlled) environment to have the right pressure drop on the IC, fuel/oil/water temperature, vehicle on-board voltage, ... and check these during & after measurement because they influence the performance a lot. Air conditions should be monitored and HP-corrected (we all know what a misty, dense, cold morning/evening feels like in our car...)

I know an average dynoshop doesn't have these facilities to measure your performance that accurate, and that isn't necessary either. That's why i like the article's conclusion; Find a shop that is capable of running your car and stay there if you want to measure performance gains. Don't stick onto the exact numbers, but onto the gained performance.

[Jamo]

Yes, but running the engine outside the engine bay gives other issues if not sorted by the operator.
That's were Point #2 of the article comes to play...
I work at a site with 34 similar dynamometers and we have to be able to swap engines internally without losing accuracy. That asks for calibration of all the measurement devices twice a month. Next to that calibration of the primairy sensors & emission bench takes place multiple times a day!

To be true to measure under similar conditions you should find pressure & temperature drops/ratings coming from the vehicle and setting it up on your dyno. How? Make everything adjustable into a (preferable digitally controlled) environment to have the right pressure drop on the IC, fuel/oil/water temperature, vehicle on-board voltage, ... and check these during & after measurement because they influence the performance a lot. Air conditions should be monitored and HP-corrected (we all know what a misty, dense, cold morning/evening feels like in our car...)

I know an average dynoshop doesn't have these facilities to measure your performance that accurate, and that isn't necessary either. That's why i like the article's conclusion; Find a shop that is capable of running your car and stay there if you want to measure performance gains. Don't stick onto the exact numbers, but onto the gained performance.

Exactly, seems to be an obessions over dyno's by certain members. They are only barometers to be used to judge changes made to the cars performance from modifications. As long as they are fairly accurate then I'm happy with their results.

The only time I scoff is at the american figures quoting WHP which are in reality Crank power

[EspenW]

Problem is that most customers and many operators aren't aware of the error margin and variations, so folks get a baseline on one dyno, then do changes and dyno again someplace else, expecting the results to match, as well as expecting (and being told) that the actual crank figure will be quoted with perfect accuracy.
Euro magazines do this all the time, and NAF (the RAC of Norway) actually sued Nissan, based on flywheel numbers off a rolling road. Needless to say, they lost in court.
Whp from an American Dynojet will be close to crank hp on a lower powered car (say below 250hp). The higher the power, the more representative for true whp. Whp from an American Mustang dyno will be true whp regardless of power level, as long as the calibration is correct.

[jogi730]

Yes, but running the engine outside the engine bay gives other issues if not sorted by the operator.
That's were Point #2 of the article comes to play...
I work at a site with 34 similar dynamometers and we have to be able to swap engines internally without losing accuracy. That asks for calibration of all the measurement devices twice a month. Next to that calibration of the primairy sensors & emission bench takes place multiple times a day!

To be true to measure under similar conditions you should find pressure & temperature drops/ratings coming from the vehicle and setting it up on your dyno. How? Make everything adjustable into a (preferable digitally controlled) environment to have the right pressure drop on the IC, fuel/oil/water temperature, vehicle on-board voltage, ... and check these during & after measurement because they influence the performance a lot. Air conditions should be monitored and HP-corrected (we all know what a misty, dense, cold morning/evening feels like in our car...)

I know an average dynoshop doesn't have these facilities to measure your performance that accurate, and that isn't necessary either. That's why i like the article's conclusion; Find a shop that is capable of running your car and stay there if you want to measure performance gains. Don't stick onto the exact numbers, but onto the gained performance.

So this leads to another debate that never really has any conclusions - which is the better way to map an engine:

1) Rolling Road
2) Engine dyno

I guess neither truely represent road conditions, so either way fine tuning should take palce on the road.

[Jamo]

So this leads to another debate that never really has any conclusions - which is the better way to map an engine:

1) Rolling Road
2) Engine dyno

I guess neither truely represent road conditions, so either way fine tuning should take palce on the road.

Mute point IMO

All the rolling road does is show the bhp output, if your using the same dyno with the same settings there is no no issue as you'll be able to see the difference in changes.

[quattro84]

Removing the engine is too much of a hassle, and i'd like it to get it tuned in the setting it'll be running at last, in the engine bay!
Why? Eliminating possible issues which you haven't looked for in the first place. IC efficiency can pe poorer, so can be engine cooling, next thing you know your custom mapping sucks because the effects are more clear on perfomance and for instance can lead to engine knock.

I'd go for hub dyno's because they are great:
No deviations in roll resistance
" dynamic tire circumference
No wheelspin
No interaction of the differential or capable of running a LSD/MIG diff
Can cope with sick amounts of caster/camber/...
Mega low dyno inertia
More safety for the operator and your car!
...

As for numbers: I wouldn't use a dyno for maximum screen numbers, but as a tool to check the engine mapping in a safe, controlled environment to monitor engine parameters next to the performance.

[JP]

Only problem with hub dyno's is that there is very few about in this case in England only one I can think of is Thor Racing. And well I wouldn't go there if I was paid. Second problem with hub dyno's is that they are very harsh on the gearbox.

Like the article states find your favourite dyno operator and mapper and stick with them.

[EspenW]

So this leads to another debate that never really has any conclusions - which is the better way to map an engine:

1) Rolling Road
2) Engine dyno

I guess neither truely represent road conditions, so either way fine tuning should take palce on the road.

I see some folks over at places like 6speedonline.com make a point of their car being tuned on an engine dyno, but I agree with quattro84 that the engine should be in its natural environment. A loading chassis dyno will be good for the mapping session, then check it on the road.

I also agree that hub dynos will be the most accurate, in particular the Swedish Rototest dynos.
Interesting to note that a stock 4wd car rated at 480ps and 620nm at the flywheel pulled 464ps and 622nm at the hubs.
We got a tuner here in Norway that uses 25% drivetrain loss from the hub dyno when guesstimating flywheel numbers on a 4wd car. The owner of this stock car would have gotten papers for 618ps and 830nm had he dynoed at this tuner. The car in question is somewhat underrated, but not much. No wonder we see so many ridiculous power numbers being pushed as "dyno proven" gospel at forums and elsewhere.