Transmission losses on a dyno - appropriate estimate
10-01-2002, 03:30 AM
#1
Instructor
Thread Starter
Transmission losses on a dyno - appropriate estimate
As I am sure almost everyone reading this knows, chassis dynos measure rear wheel horsepower (RWHP) rather than flywheel horsepower (FHP). RWHP is lower due to the frictional losses getting from the flywheel to the wheels. FHP can be estimated from RWHP by assuming what this transmission loss is.
As FHP is a bigger number, it is the one everyone likes to quote. Unfortunately, from what I have read this causes certain tuners throughout the world to use "optimistic" assumptions of transmission losses.
For example, if a car makes 200 RWHP you could assume anywhere from 5% transmission loss to 30%. This would give a FHP estimate range of 211-286 - a 75hp difference!
So, I would like to initiate a bit of a discussion on what people have seen as an appropriate transmission loss (or correction factor) for 911s (or 930s).
My thoughts are:
- it might be different between dynos (my dyno tuning was done on a Dynapack which attaches directly to the hubs (wheels removed)) and this may cause a difference
- it might (should) be different between transmissions (ie a 901 transmission is lower than a 915 which might be lower than a G50 or a 930 fits)
I would also expect that a transaxle transmission would have much lower losses than a front engine - prop shaft - rear diff drivetrain.
Finally (long post, huh), the dyno operator I used sticks to 1.1x (or 9.1% losses) for a 911. Bearing in mind this is on a hub dyno (which may measure lower losses than one which has wheels and tyres in the measuring path), this is still way down at the conservative end of the calculations I have seen. Gearbox is a 915 from an SC.
I have also seen 11-12% losses on the same car on the same shops other Dyno (a Bosch rolling road). It estimated the losses using coast-down. This was with the old 901 gearbox.
Useful (but not working today) links are at:
<a href="http://www.pumaracing.co.uk/power3.htm" target="_blank">http://www.pumaracing.co.uk/power3.htm</A>
<a href="http://www.pumaracing.co.uk/trans.htm" target="_blank">http://www.pumaracing.co.uk/trans.htm</A>
<a href="http://www.pumaracing.co.uk/coastdwn.htm" target="_blank">http://www.pumaracing.co.uk/coastdwn.htm</A>
As FHP is a bigger number, it is the one everyone likes to quote. Unfortunately, from what I have read this causes certain tuners throughout the world to use "optimistic" assumptions of transmission losses.
For example, if a car makes 200 RWHP you could assume anywhere from 5% transmission loss to 30%. This would give a FHP estimate range of 211-286 - a 75hp difference!
So, I would like to initiate a bit of a discussion on what people have seen as an appropriate transmission loss (or correction factor) for 911s (or 930s).
My thoughts are:
- it might be different between dynos (my dyno tuning was done on a Dynapack which attaches directly to the hubs (wheels removed)) and this may cause a difference
- it might (should) be different between transmissions (ie a 901 transmission is lower than a 915 which might be lower than a G50 or a 930 fits)
I would also expect that a transaxle transmission would have much lower losses than a front engine - prop shaft - rear diff drivetrain.
Finally (long post, huh), the dyno operator I used sticks to 1.1x (or 9.1% losses) for a 911. Bearing in mind this is on a hub dyno (which may measure lower losses than one which has wheels and tyres in the measuring path), this is still way down at the conservative end of the calculations I have seen. Gearbox is a 915 from an SC.
I have also seen 11-12% losses on the same car on the same shops other Dyno (a Bosch rolling road). It estimated the losses using coast-down. This was with the old 901 gearbox.
Useful (but not working today) links are at:
<a href="http://www.pumaracing.co.uk/power3.htm" target="_blank">http://www.pumaracing.co.uk/power3.htm</A>
<a href="http://www.pumaracing.co.uk/trans.htm" target="_blank">http://www.pumaracing.co.uk/trans.htm</A>
<a href="http://www.pumaracing.co.uk/coastdwn.htm" target="_blank">http://www.pumaracing.co.uk/coastdwn.htm</A>
10-01-2002, 07:09 AM
#2
Race Car
You are absolutely correct in assuming that a Porsche transaxle will have less loss when compared to a car with a driveshaft/differential/transmission...there's a lot of torque lost on accel and so forth.
A local racing shop uses 20% for a normal setup (engine/trans./driveshaft/diff.) and 15% for 911s.
I figure this is a good figure to deal with, may even be a very conservative figure as losses may be less than 15%, quite possibly even ~10%??
Either way, when considering my friend's RSA pulled 218HP on a wheel dyno, if you add 15% to that figure, it brings you pretty close to factory HP figures.
A local racing shop uses 20% for a normal setup (engine/trans./driveshaft/diff.) and 15% for 911s.
I figure this is a good figure to deal with, may even be a very conservative figure as losses may be less than 15%, quite possibly even ~10%??
Either way, when considering my friend's RSA pulled 218HP on a wheel dyno, if you add 15% to that figure, it brings you pretty close to factory HP figures.
