DYNO QUESTION

[kilowatts]

I want to know wether you calculate the loss through the wheels as a pecentage or is it set figure .For example if you dyno,ed your 200 kw at the flywheel car and got 150 kw at the wheels, thus a 50 kw loss and then modify the car and increase the hp will it still be 50 kw lost through the drive train or is it percentage. Hope you understand , it was hard to put to words.

[edwagon]

Percentage. normally around 20-30ish%

[fivebob]

Percentage. normally around 20-30ish%

**cough**bullsh!t**cough**

See http://www.pumaracing.co.uk/power3.htm for a more reasonable treatise on the subject.

[[V8ROLA]]

blah blah

[kilowatts]

Thanks fivebob for that link ,it was most informative .

[UltraSonic2]

yip 20 - 30 % that's what the " car acceleration test " program comes to

[GT4 20]

% loss depends on transmission. FWD tends to have the least losses (would also be the same for a car like an MR2) followed by RWD and then 4WD.
From what I understand..

FWD - 17%
RWD - 20%
4WD - 25%

Now bear in mind that the above is a very rough guide.
Gary

[fivebob]

yip 20 - 30 % that's what the " car acceleration test " program comes to

Which is a complete load of crap

% loss depends on transmission. FWD tends to have the least losses (would also be the same for a car like an MR2) followed by RWD and then 4WD.
From what I understand..

FWD - 17%
RWD - 20%
4WD - 25%

Now bear in mind that the above is a very rough guide.

Very rough and IMHO still way too high, might be close to correct for a 100hp car, though the figures I've seen quoted are 15%-FWD, 17%-RWD & 22%- 4WD

For a more powerful car the % calculation is way too much, e.g. how does a 500rwhp MR2 get rid of 100hp (~75kw) of energy, it would boil the oil in the gearbox and self destruct very quickly, not to mention toasting the engine bay along the way.

[Akane]

Come on, more powerloss = free horsepower!

[strx7]

its not a HP loss, its a READING DISCREPENCY for the type of dyno it is.
take car X, dyno it on all the different types of dyno's there are, and you'll get a different result each time. there is no "HP LOSS" in it.

Yes you loose HP between the flywheel and the tarmac, when someone works out a way of being able to TRUELY record what that loss is, let me know. Short of taking, an engine dyno, and bolting it to your axle, its all a guessing game.

[fivebob]

Come on, more powerloss = free horsepower!

Yes, but it's amazing how much the tuner charges you for that "free" horsepower

[fivebob]

Yes you loose HP between the flywheel and the tarmac, when someone works out a way of being able to TRUELY record what that loss is, let me know. Short of taking, an engine dyno, and bolting it to your axle, its all a guessing game.

Strain gauge on the input shaft and one on each driven axle should give the required data, but those sort of things cost big $$$

[jondee86]

Short of taking, an engine dyno, and bolting it to your axle, its all a guessing game.

This is exactly what the DYNAPACK chassis dynamometer does. Eliminates
tyre pressure, alignment and roller slippage losses.

it would boil the oil in the gearbox and self destruct very quickly

Which is why race cars have have transmission and differential oil coolers
as well as engine oil coolers. There is a limit to how much heat that can
be dissipated from the surface of the gearbox/diff. More power means
more heat and more surface area is needed to stay within the temperature
range of the oil.

Cheers...... jondee86

[RedMist]

it would boil the oil in the gearbox and self destruct very quickly

Which is why race cars have have transmission and differential oil coolers
as well as engine oil coolers. There is a limit to how much heat that can
be dissipated from the surface of the gearbox/diff. More power means
more heat and more surface area is needed to stay within the temperature
range of the oil.

Cheers...... jondee86

Seriously, get yourself a 75kw heater the size of a gearbox. Then add the surface area of a tranny cooler... then rest your nads against any part of the rapidly melting surface...

[fivebob]

Short of taking, an engine dyno, and bolting it to your axle, its all a guessing game.

This is exactly what the DYNAPACK chassis dynamometer does. Eliminates
tyre pressure, alignment and roller slippage losses.

Yes, but you can't attach it direct to the engine and compare readings because the dynapack can't handle the revs. This means that you can't compare output at the shaft with output at the axles as you would have to connect a different measurement device to the flywheel and then the discrepancies in the dynos will make the drive train losses inaccurate.

it would boil the oil in the gearbox and self destruct very quickly

Which is why race cars have have transmission and differential oil coolers
as well as engine oil coolers. There is a limit to how much heat that can
be dissipated from the surface of the gearbox/diff. More power means
more heat and more surface area is needed to stay within the temperature range of the oil.

We're not saying that there isn't a %loss component, but rather that the loss is made up of a fixed component and % component. e.g. on a 100hp engine the loss might be 10hp + 10%, give a 20hp loss, if all that was a % figure that would mean that a 200hp engine would lose 40hp, a 500hp engine would lose 100hp, at those rates most transmission coolers wouldn't keep up.

If however you used the formula of 10hp fixed + 10% loss then that would mean a 200hp engine would lose 30hp and a 500hp engine would lose 60hp, much more realistic figures, though they're still just a guess. BTW 10% +10hp example used is just to illustrate the point and in no way should be taken as the actual loss figure. Also it is likely that the %loss componenet may increase as hp rises, though not by much more that 1-2%

As stated before, the only way to measure transmission loss is to use the same instrumentation to measure output at the flywheel and at the driven axles anything else is just a guesstimate and in a lot of cases pure fantasy

[RomanV]

And at the end of the day it means sweet $&#$% all.....

What good is power at the engine, if it cant get to the wheels?

Having 200hp at the wheels from an '800hp' engine is no more impressive than 200hp at the wheels from a 220hp car.

But yes, this topic gets bought up quite often, and there is a lot of BS floating around about it. We should use the swear filter in TS to replace 'transmission loss' with 'pumaracing.co.uk'.

Would save a lot of time.

[ROBODISCO_20v]

yip 20 - 30 % that's what the " car acceleration test " program comes to

Which is a complete load of crap

% loss depends on transmission. FWD tends to have the least losses (would also be the same for a car like an MR2) followed by RWD and then 4WD.
From what I understand..

FWD - 17%
RWD - 20%
4WD - 25%

Now bear in mind that the above is a very rough guide.

Very rough and IMHO still way too high, might be close to correct for a 100hp car, though the figures I've seen quoted are 15%-FWD, 17%-RWD & 22%- 4WD

Dude you're splitting hairs here, you're saying GT4 20 is wrong then you're saying the almost same thing

[jondee86]

Here ya go.... this is a pretty good explanation of how to figure
transmission losses.

"The efficiency of a manual transmission drive train can be closely
estimated by multiplying together the efficiencies of each loaded gear pair
or other working mechanism of the complete drive train. For helical gear
gears used in transmissions in an EV at torques averaging 20% to 30% of
max and with a low loss lubricant, the efficiency should average around
0.97 per gear pair. An allowance is needed also for the other operating
but unloaded gear pairs, since they are resisting by some friction and
continually churn the lubricant. An 0.99 factor is considered adequate for
these unloaded gears. The Fiesta 4 speed gearbox then, which uses
but one gear pair at a time, under these conditions has an efficiency of
0.97 x 0.99 or 0.96. Counting the differential drive, another helical gear
pair and 0.97 factor, the motor-t-drive axle efficiency becomes 0.93.

Manual transmissions typically used in front engine rear axle drive cars
(those with 1:1 ration for "high" gear) are connected "straight through"
and load none of their gears when in "high". Efficiency then averages
about 0.98, with the moderate 2% loss going into the unloaded but
lube-churning gears. In the lower gears, however, these "straight
through" transmission must use two helical gear pairs, so the gearbox
efficiency alone becomes about 0.97 x 0.97 = 0.94.

There are still more drive train losses before power reaches the driving
wheels. While constant velocity U-joints are low-loss devices as are ball
or roller bearings, considerable and frequent angularity changes (such as
in front-drive axles) cannot be achieved without considerable sliding and
rolling friction losses. 2% loss, or 0.98 efficiency, is reasonable for
modern front drive axles, (including the bearing and seal losses). For
front engine, rear drive cars, a 1% allowance for drive shaft u-joints is
adequate (if joint angularity is small) and another 1% for typical axle
bearings and seals."


You can break down the drive train efficiency like this:

Typical modern front-drive transaxle (Civic, Fiat 128, Fiesta, Rabbit)

Manual transmission
Number of loaded gear pairs 1. (0.97)
Allowance for idling gears: lube churning plus extra bearing and
selector drag (0.99)
0.97 x 0.99 = 0.96
Differential Drive
Helical gears (0.97)
Drive Axle
(0.98 )
Overall Efficiency
0.97 x 0.99 x 0.97 x 0.98 = 0.91


Transaxle with bevel gear axle (VW Beetle, Renault LeCar)

Manual transmission
Number of loaded gear pairs 1. (0.97)
Allowance for idling gears: lube churning plus extra bearing and
selector drag (0.99)
0.97 x 0.99 = 0.96
Differential Drive
Bevel gears (0.96)
Drive Axle
(0.98 )
Overall Efficiency
0.97 x 0.99 x 0.96 x 0.98 = 0.90


Typical rear drive (not 1:1) (Chevette, Datsun 210, Pinto)

Manual transmission
Number of loaded gear pairs 2. (0.97 x 0.97)
0.97 x 0.97 = 0.94
Drive Shaft
(0.99)
Differential Drive
Hypoid gears (0.89)
Drive Axle
(0.99)
Overall Efficiency
0.97 x 0.97 x 0.99 x 0.89 x 0.99 = 0.82


As above but in 1:1 "high" gear

Manual transmission
Number of loaded gear pairs 0.
(0.98 ) Allowance for idling gears: lube churning plus extra
bearing and selector drag
Drive Shaft
(0.99)
Differential Drive
Hypoid gears (0.89)
Drive Axle
(0.99)
Overall Efficiency
0.98 x 0.99 x 0.89 x 0.99 = 0.85

Of course, none of this matters a toss if all you want to do is measure
what improvement your tuning or mods have produced. Just use the
same dyno and overlay the results. The only thing that you need to
measure is power/torque at the driving wheels.

Cheers..... jondee86

[fivebob]

yip 20 - 30 % that's what the " car acceleration test " program comes to

Which is a complete load of crap

% loss depends on transmission. FWD tends to have the least losses (would also be the same for a car like an MR2) followed by RWD and then 4WD.
From what I understand..

FWD - 17%
RWD - 20%
4WD - 25%

Now bear in mind that the above is a very rough guide.

Very rough and IMHO still way too high, might be close to correct for a 100hp car, though the figures I've seen quoted are 15%-FWD, 17%-RWD & 22%- 4WD

Dude you're splitting hairs here, you're saying GT4 20 is wrong then you're saying the almost same thing

Dude you're failing the reading comprehension test I've highlighted the important bit for you

BTW since when was 15%=17%, or 17%=20% or even 22%=25%, they're all different by a large margin, but if that's your idea of accurate then I hope you don't set you torque wrench with the same tolerances, or you will be spending a lot of money on helicoils.

[fivebob]

Here ya go.... this is a pretty good explanation of how to figure
transmission losses.

"The efficiency of a manual transmission drive train can be closely
estimated by multiplying together the efficiencies of each loaded gear pair
or other working mechanism of the complete drive train. For helical gear
gears used in transmissions in an EV at torques averaging 20% to 30% of
max

All very interesting, but you'll see the bit I've highlighted indicates that this only holds true for the input torque figures quoted, says nothing about the losses at other values, given that friction from the lubricant being churned up is unlikely to change with a change in input torque, though it might change with input/output rpm and things like seal and bearing drag do not change with torque, unless these is some increased radial load change then those %loss figures would not apply at higher input torque values, which is what I've been saying all along, you can't have one %loss figure and apply it over all torque and rpm values and come up with anything meaningful

BTW do you really believe that a 500hp engine will lose 100hp through transmission losses? If so then where does all that lost energy go?

Of course, none of this matters a toss if all you want to do is measure what improvement your tuning or mods have produced. Just use the same dyno and overlay the results. The only thing that you need to
measure is power/torque at the driving wheels.

In most cases that would be correct, but say for example I wanted to compare the output of a 3S-GTE in a SW20 with one in a ST215, without some idea of the actual drive losses I wouldn't be able to determine which motor was outputting more power, of course I could just swap the motors and dyno them on the same setup, provided I took all the other variables out of the equation like tyre slip, atmospheric conditions and minor intake/exhaust differences then I might get a rough idea, but the same engine will not always produce the same results on the same dyno. I suppose in that case an engine dyno is the only way to go.