Drivetrain losses

Author
Discussion

andygtt

Original Poster:

8,344 posts

263 months

Monday 29th September 2014
quotequote all
Ive been on 3 different UK dynos on my noble mostly slightly differing states of tune so not directly comparable.

Last one was a hub dyno and they wouldn't even entertain a flywheel estimation... its this much at the hub your guess of flywheel figure is as good as ours.... the other claimed 60bhp and 100bhp... all are claimed to be conservative and more accurate than everyone else's.

Anyhow speaking to a respectable american firm who specialise in tuning extreme nobles they have a car with a comparable engine to mine.... they told me they were running massive flywheel power, but that they had 200bhp drivetrain losses?.... at reasonably close wheel power levels to what i run.

Other than engine dyno back to back with a hub dyno is there a way to test these losses and have conclusive figures?

stevesingo

4,848 posts

221 months

Monday 29th September 2014
quotequote all
Towards the bottom of this page gives a breakdown of difference between measured hub power and manufacturers claims.

http://rototest-research.eu/index.php?DN=39&Me...

It would appear that there is a 9% loss on average between manufacturer claims and measured power at the hubs on the Rototest dynos.

The losses on rolling road dynos are massively variable. The largest variable is the tyre to roller interface which is hugely dependant on wheel speed. The faster the wheel speed the higher the losses, so dependent on gear you will get a different result.

Pumaracing

2,089 posts

206 months

Monday 29th September 2014
quotequote all
The most comprehensive discussion of transmission losses on the web is on my old website archived here.

https://web.archive.org/web/20110902010921/http://...

Look at Power and Torque 3 - https://web.archive.org/web/20110909141356/http://...

Coastdown Losses - https://web.archive.org/web/20110903073319/http://...

How not to set up a car on the rollers - https://web.archive.org/web/20110918114959/http://...

Plus any of the other articles that take your fancy.

Hub dynos lose about 7% of the flywheel bhp on fwd cars and 9% on rwd ones with inline engines where the drive has to be turned through 90 degrees.

Rolling roads driving through the tyres are discussed in the articles.

Normally I take wheel bhp + 10 bhp and then divide by 0.9 for fwd cars and wheel bhp + 10 bhp and divide by 0.88 for rwd cars to get to the flywheel bhp. This produces a decreasing total loss percentage as power goes up.

andygtt

Original Poster:

8,344 posts

263 months

Monday 29th September 2014
quotequote all
ok this sounds similar to what i was thinking.

so basically for a noble that runs a mondeo ST220 V6 engine and gearbox (a fwd car)... even though its making 700hp at wheels its not going to be loosing 200bhp through the drivetrain and thus have 900bhp at fly.... more like 70-100bhp max?

what would you expect the difference in figures between a hub dyno and a set of decent rollers?... its just the wheels and tyres that are removed?

PeterBurgess

775 posts

145 months

Monday 29th September 2014
quotequote all
Forget 'rules of thumb' that is all they are, rules of thumb and thumbs can and do vary. The Rototest figures are hub measured then compared to what manufacturers claim, depends how the manufacturers test engine compares to the retail ones don't it!

Tyre speed seems to be the biggest loss and tyre width. UJs lose more bhp than fixed axles, thick oils lose more bhp than thin oils. To give an idea. Minis and Midgets about 14.5 bhp losses (low speed and not very wide tyres), MGBs 22 ish for road speeds (100 mph) and tyres, 26 ish for race tyres and speeds(120 mph). VW Kampers 20ish bhp. TR6 loads of UJs which one can hear creaking 30 ish bhp. Renault Alpine PRV V6 horrible UJs, bif f**k off tyres and 140+mph 36 bhp loses. I can find no correlation as a percentage of bhp at engine, just a fixed amount depending on transmission type and oils and tyres and speed. As an example as to why you cannot do rules of thumb, MGB had a leak on back axle, owner put gloop in to cure it, losses went up from 22 to 27 even though all else same! Tyre pressures, going down from 30 psi to 18 sapped 7 more bhp on a VW!

At the end of the day you want as much power as you can get at the wheels on a repeatable value rolling road, the flywheel power is not really relevant, coast down losses are useful to compare same make cars to make sure they are 'about right' for losses, if not, find out why, added to wp they are an approximation only of flywheel power, but, put you in the ball park area rather than a rule of thumb guesstimate.

Peter

stevieturbo

17,229 posts

246 months

Monday 29th September 2014
quotequote all
Whether it's your fancy or not...

1/4 mile traps, 1/2 mile traps etc are far better and more reliable indicators of actual power to the ground.

Dyno's just vary too much to try and make comparisons elsewhere.

And trying to compare with US dyno's....is even more difficult. Everything is bigger over there !

Or if standing starts arent great. Some of the 30-130 or 60-130 type pulls are also a good indicator of power and can be far more reliable measured across continents ( unless you're running downhill with a massive tailwind )

anonymous-user

53 months

Monday 29th September 2014
quotequote all
Tune your engine on an engine dyno, so you can get repeatable data during development, allowing you to eak out every last 100 watts.

Test you car on a race circuit / drag strip / rally stage or whatever. The car is as fast as it is fast, no more, no less.


Anything else is just pub ammo, unless you happen to have access to a full inertia simulator chassis dyno, and even then, you'll find repeatability and measurement issues make a mockery of any "absolute" numbers you might care to suggest.......


(generally, roto-test and similar hydraulic hub chassis dynos can't work out coastdown losses very well because they don't have enough rotating inertia!)





PeterBurgess

775 posts

145 months

Monday 29th September 2014
quotequote all
NA up to 600 bhp at the wheels seems very consistent on our rollers if you take your time and don't rush things. Turbos are much harder, the 1% down for every 3 degrees C hotter shows up beautifully on the power lines.
The rolling road is a tuning aid, no more no less and in my humble opinion is better than guessing. As you say, when the flag drops the bulls**t stops smile

I reckon, those that knock rolling roads would use them if they owned one to tune their cars even enough to get em down the strip for final tuning if required, I dont think many folk who own rolling roads would not use them in some way to preset their cars.

Our Dynocom rollers are 4 and a half years old now and we have a lot of runs under our belt to see consistency and correlation.

Peter

Pumaracing

2,089 posts

206 months

Tuesday 30th September 2014
quotequote all
andygtt said:
ok this sounds similar to what i was thinking.

so basically for a noble that runs a mondeo ST220 V6 engine and gearbox (a fwd car)... even though its making 700hp at wheels its not going to be loosing 200bhp through the drivetrain and thus have 900bhp at fly.... more like 70-100bhp max?

what would you expect the difference in figures between a hub dyno and a set of decent rollers?... its just the wheels and tyres that are removed?
Estimated actual flywheel bhp for fwd. (700 + 10)/0.9 = 789 bhp.

Hub dyno estimated reading would therefore be 789 x 0.93 = 734 bhp

So of the 89 bhp total losses, 55 bhp is in the transmission and 34 bhp in the tyres.

If someone with a hub dyno couldn't give you an estimated flywheel figure then they are idiots basically and don't understand how their own equipment works, but hey, there's nothing new in that as I see above.

Modern helical cut transmissions are all basically identical in design and in respect to their power losses which are always about the same percentage of engine bhp regardless of how powerful the engine is. So hub dynos are very predictable tools and the 7% and 9% rules I have given can be applied to any 2wd car.

Please ignore all talk of RR coastdown losses above which have nothing whatsoever to do with actual transmission losses and are utterly meaningless in estimating a flywheel bhp.

PeterBurgess

775 posts

145 months

Tuesday 30th September 2014
quotequote all
I cannot help but think you are trying to pigeon-hole everything neatly into nice concise patterns instead of accepting there are many shades of grey Dave.

At the end of the day it is the wheel power, what is left of the flywheel power, which propels the car, being able to compare cars for wp and some form of transmission effect reductions in that resultant wp helps to see if cars are about the same or if they have a problem with engine or transmission.

As an example from the real world, we had a race B which gave different power lines in a sort of random way. After about ten runs I realised I was using the brakes to drop the revs down after warm up runs to equalise the engine temps and sometimes the servo was holding on a tad keeping the power down. we disconnected the servo for testing and the figures settled down. Also realised most race bs dont have and don't need servos so advised removal of the servo for his application.

Peter

Pumaracing

2,089 posts

206 months

Tuesday 30th September 2014
quotequote all
Ok I'm now thinking you might be feeding us bum information. Looking at your previous posts and other stuff on the net your rolling road figures will already have been converted to estimated flywheel numbers and won't be wheel bhp. So I'm now thinking you have an estimated 700 bhp flywheel bhp not wheel bhp which will probably be in the low 600s.

So a hub dyno figure would be mid 600s bhp.

fatjon

2,143 posts

212 months

Tuesday 30th September 2014
quotequote all
Some interesting posts on this subject here along with a bit of science.

http://www.pistonheads.com/gassing/topic.asp?f=6&a...


PeterBurgess

775 posts

145 months

Tuesday 30th September 2014
quotequote all
Dave, up above you say you reckon it is a constant 10 bhp for tyre losses so you must mean low 600s plus 10 and not low 600s wp to mid 600s hub dyno?

Peter

stevieturbo

17,229 posts

246 months

Tuesday 30th September 2014
quotequote all
PeterBurgess said:
NA up to 600 bhp at the wheels seems very consistent on our rollers if you take your time and don't rush things. Turbos are much harder, the 1% down for every 3 degrees C hotter shows up beautifully on the power lines.
The rolling road is a tuning aid, no more no less and in my humble opinion is better than guessing. As you say, when the flag drops the bulls**t stops smile

I reckon, those that knock rolling roads would use them if they owned one to tune their cars even enough to get em down the strip for final tuning if required, I dont think many folk who own rolling roads would not use them in some way to preset their cars.

Our Dynocom rollers are 4 and a half years old now and we have a lot of runs under our belt to see consistency and correlation.

Peter
There is no argument that using rollers for tuning can be a good thing. Using rollers for comparison is a totally different thing though.

As for the coastdown thing, never really understood that.
Drivetrain losses will get higher as load through them increases ?

yet a coastdown measurement is using a completely unloaded drivetrain ?

Tune it, make it go fast...ignore numbers unless they're real world road/track results.

Pumaracing

2,089 posts

206 months

Tuesday 30th September 2014
quotequote all
PeterBurgess said:
Dave, up above you say you reckon it is a constant 10 bhp for tyre losses... Peter
I don't say anything of the sort and suspect as usual you're being deliberately obtuse.

Pumaracing

2,089 posts

206 months

Tuesday 30th September 2014
quotequote all
stevieturbo said:
yet a coastdown measurement is using a completely unloaded drivetrain ?
Absolutely which is why it bears no relation to actual "under power" transmission losses which will always be a percentage of power being fed into the system and creating friction and heat in the gear teeth and bearings. This is also why the "coastdown loss" can never change for a given vehicle at a given road speed and rpm regardless of how much power the engine makes or is further tuned from stock.


Edited by Pumaracing on Tuesday 30th September 11:04

PeterBurgess

775 posts

145 months

Tuesday 30th September 2014
quotequote all
Normally I take wheel bhp + 10 bhp and then divide by 0.9 for fwd cars and wheel bhp + 10 bhp and divide by 0.88 for rwd cars to get to the flywheel bhp.

Is what you said so I am not being obtuse

Peter

Pumaracing

2,089 posts

206 months

Tuesday 30th September 2014
quotequote all
PeterBurgess said:
Normally I take wheel bhp + 10 bhp and then divide by 0.9 for fwd cars and wheel bhp + 10 bhp and divide by 0.88 for rwd cars to get to the flywheel bhp.

Is what you said so I am not being obtuse

Peter
Albeit it is apparently only in your own mind that somewhere in that sentence I say the 10 bhp is anything to do with tyre losses which I don't. I'll leave others to decide how obtuse you're contriving to be which is "very".

PeterBurgess

775 posts

145 months

Tuesday 30th September 2014
quotequote all
Pray what is the 10 horsepower and what does it represent that it is the same on all cars?

Peter

Pumaracing

2,089 posts

206 months

Tuesday 30th September 2014
quotequote all
PeterBurgess said:
Pray what is the 10 horsepower and what does it represent that it is the same on all cars?

Peter
Jesus wept. It's an equation. It links what I've found to be the best correlation between accurate wheel bhp and accurate flywheel bhp over many years, many vehicles and engines of varying power from 60 bhp Minis to 600 bhp V8s. The 10 is not tyre losses any more than the 10% or 12% are transmission ones. It's just a f*cking equation. Did your parents ever get you tested when you were a kid to see if you were "special" in any way?