[Guest]

One for Droid and others. How do we calculate power loss from flywheel to the driven wheels? This must be difficult to work out on an EVO. Here are the things I believe affect it, but not sure if it makes them better or worse.

1) Flywheel - it has mass, so takes energy to move it. Lighter flywheel, less loss?
2) Clutch - this must create a loss which must depend on how well it engages (slipping clutch|EQU|power loss). What about multi-plate clutches? Better or worse?
3) Gearbox - helical gears feature drag, so losses here. Does a dog box (which has less drag) mean less power loss?
4) Front diff - if you have one, like I do, would that mean more loss from an LSD? Do open diffs (standard) mean less loss?
5) Centre diff /transfer unit - proportions power front/rear, but probably loses some of the power in doing so. The ACD in the 7 is a multiplate clutch, so that must feature slightly bigger losses.
6) Rear diff - as front diff. AYC presumably would have a higher loss than a mechanical diff in the RS/RSX?
7) Driveshafts.
8) Wheel and tyre weight - These have mass and are rotated, so are a load - cause of aloss?
9) Tyre friction - power has to overcome tyre friction, so that must have an effect on loss.

Idea behind this thread is to understand the science of drivetrain power loss, why 4WD is higher than 2WD and what effect drivetrain changes have on power losses.

As a side issue, although RR measurements can only be taken seriously by looking at the power-at-the-wheels figure, it can be very difficult to get accurate power-at-the-wheels figures as it:-

1) Depends on grip and how well cut the rollers are. Cars with big power can slip, hence why they are strapped and have bodies inside them often. My car on it's last RR session had 3 people in it and 6 straps tieing it down and it was still slipping!

2) Ambient temperature and lack of real airflow will stifle power outputs, and give lower readings. It will affect highly tuned engines more than standard ones, as induction kits, cooling kits etc. only really work properly on the move.

3) Gearing - depending on what gear it is done in, need to be close to 1.0 - or recalculate according to the gear ratio

Anything I missed?

 

[Guest]

One for Droid and others. How do we calculate power loss from flywheel to the driven wheels?

Blade,

Good thread starter!

Only just looked at this after replying to you on the M3 thread.

This must be difficult to work out on an EVO

The only way to work it out on any particular car is to bung it on a rolling road and get the at-the-wheels number. Then take the engine out, put it on the bench and work out the flywheel power. That's it. Rolling road operators will argue with me but there's simply no way to measure transmission losses at the wheels during coastdown because coastdown means engine not producing any power .

Here are the things I believe affect it, but not sure if it makes them better or worse.

1) Flywheel - it has mass, so takes energy to move it. Lighter flywheel, less loss?

It's more about the power required to keep it moving rather than the power required to spin it up in the first place. This then depends on the frictional losses due to keeping it moving. These are generally tiny. Also important is how well balanced it is. Usually that should be OK.

2) Clutch - this must create a loss which must depend on how well it engages (slipping clutch|EQU|power loss). What about multi-plate clutches? Better or worse?

Yes, the better the clutch, the less it will slip and the less power will be lost to frictional losses.

3) Gearbox - helical gears feature drag, so losses here. Does a dog box (which has less drag) mean less power loss?

In theory yes. A lot of heat is generated in a gearbox and that means big energy/power losses.

4) Front diff - if you have one, like I do, would that mean more loss from an LSD?

I'm not sure on the LSD/Open thing but the very existence of the front differential is the reason why 4WD loses so much through the transmission compared to FWD/RWD. RWD loses more to FWD for obvious reasons (more components required to transmit the power to the rear wheels rather than the front).

5) Centre diff /transfer unit - proportions power front/rear, but probably loses some of the power in doing so. The ACD in the 7 is a multiplate clutch, so that must feature slightly bigger losses.

Yes. Any component that transmits drive, changes its direction or changes the associated gearing will cause power loss.

6) Rear diff - as front diff. AYC presumably would have a higher loss than a mechanical diff in the RS/RSX?

Yes, I would have thought so.

7) Driveshafts.

Absolutely.

8) Wheel and tyre weight - These have mass and are rotated, so are a load - cause of aloss?

I would only count wheel bearings/CV joints etc. in this. Rolling resistance (loss due to tyres) is usually measured separately from transmission losses. Again, it's not the mass of the component that's important, it's the amount of energy required to keep it rotating at a constant speed that gives you the power/torque loss.

9) Tyre friction - power has to overcome tyre friction, so that must have an effect on loss.

Counted as part of rolling resistance.

Idea behind this thread is to understand the science of drivetrain power loss, why 4WD is higher than 2WD and what effect drivetrain changes have on power losses.

As a side issue, although RR measurements can only be taken seriously by looking at the power-at-the-wheels figure, it can be very difficult to get accurate power-at-the-wheels figures as it:-

1) Depends on grip and how well cut the rollers are. Cars with big power can slip, hence why they are strapped and have bodies inside them often. My car on it's last RR session had 3 people in it and 6 straps tieing it down and it was still slipping!

2) Ambient temperature and lack of real airflow will stifle power outputs, and give lower readings. It will affect highly tuned engines more than standard ones, as induction kits, cooling kits etc. only really work properly on the move.

3) Gearing - depending on what gear it is done in, need to be close to 1.0 - or recalculate according to the gear ratio

The above things are exactly why I like at-the-wheels numbers from rolling road ... they all tend to lead to pessimistic readings rather than optimistic ones.

Only things missed I can think off (off the top of my head) and I've mentioned some already:

- wheel bearings
- CV joints

I'm sure there are plenty more. This isn't my strongest area of knowledge as you can probably tell

Ian.

 

[Guest]

Blade,

Just started getting into this on another thread here (I think the one on RR results). I re-post here:

Ok, admittedly this is based on very few (24) observations, and on results from two separate dynos. But from RR results found on the two sites mentioned on this thread so far, the AVERAGE loss from wheel HP to crank (or P-Norm ) HP is 30%. There are some outliers-- lowest loss is 22%, greatest loss at 33%. But most common loss figures are between 27% and 30%.

Not sure what exactly this means... I wouldn't extrapolate much from this rather simplistic exercise. But is anyone out there aware of a more comprehensive analysis on this matter? My car (E5) is showing 233 HP at the wheel. The dyno on which this figure was taken does not return estimated flywheel HP figures, but assuming the 30% average loss figure is correct, that would mean I'm at about 330 HP at the crank.

The car is surely faster than stock-- a recent head-to-head on the highway with my father's Evo 7 had me pulling away, though not by much. And definitely NOT as if I had 50 HP more than his car. Plus, his car has just been run-in, so the engine hasn't loosened-up yet.

Ideas anyone?

 

[Guest]

I believe from my experience that its about 25-30% for a 4wd vehicle.
For some data my EVO III stock made 150 Kw at all 4 wheels.
This is a 25% loss from the quoted engine power

 

[Guest]

Can't say much for drivetrain losses as it's already been covered but you are right that the power output will vary with ambient temperature and atmospheric pressure. Manufacturers power figures take this into account as there is a 'Power correction standard' so everyone is on a level playing field. Basically a correction factor is calculated from the measured temp and pressure at testing and then you multiply the measured power by the correction factor to get the corrected power.

Depending on the standard used the Correction factor |EQU| A1.2 x B0.6
Where:
A|EQU| 99/Measured Barometric Pressure (KPa)
B|EQU| Temperature during testing (K)/298

Corrected Power (kW) |EQU| Correction factor x Measured power (kW)

I believe most Chassis Dyno's use this correction factor for working out the final power/torque plot as they have an intake temp sensor, barometric sensor and possibly a relative humidity sensor attached to the vehicle or nearby (humid air contains less oxygen than dry air but it is insignificant). Figures can be fudged though by fiddling with these.

Check out the following info on Dyno's (you need to subscribe to get the full articles well worth it though, this site is full of info):
http://www.autospeed.com/A_0192/page1.html
http://www.autospeed.com/A_0151/page1.html
http://www.autospeed.com/A_0625/page1.html

Also the 'Bosch Automotive Handbook'

I am afraid if you want to know the exact answer of power at the flywheel then you need to run it on an Engine Dyno. Chassis Dynos are inherently inaccurate because there are so many variables that effect the result, even simple things like placing the intake temp sensor in a slightly different position can make a difference to the final results. It is best to just use them to track changes rather than to get absolute results.

Sorry not very helpful as you want to know how to work out the transmission losses but you could spend ages trying to work them out to get the flywheel figure and it will still be wrong just because Chassis Dynos are not that accurate.

(Little know fact No.4678: Engine power drops by roughly 1% for ever 100m increase in altitude.)