I was reading a user's manual for a Dynojet dyno and it shows a way of measuring the actual drive train loss for each vehicle tested, rather than assuming the 14 or 16 or 18% loss. Has anybody gotten their dyno tuner to do this for them, and what were the results?

 

Interesting....How is it measured?

 

Let us know how its done. :thumbs:

 

I imagine it measure coast-down time, but would love to hear more.

 

The dyno I went to in Germany did this.

I will try to post the plot, I am not yet sure how this works:
http://www.corvetteforum.de/attachme...entid=722&sid=

The grey line you see at the bottom is called "Schleppleistung" in German. It shows the drivetrain loss vs rpm. I think you start at max rpm and take your foot off the accelerator. At max rpm, energy is stored in the rotation of the dyno. This rotational energy of the dyno is consumed by the friction in your drivetrain (negative power), resulting in dropping rpms. So you get the power consumed by friction vs rpm.

Of course there is some calibration involved, since the dyno also has internal friction.

I am not sure if you press the clutch during this procedure, and calculate corresponding engine rpm from the rpm of the dyno. I will call the tuner on monday and ask.

Anyway, for my car I got a drivetrainloss of 17% in fourth gear and with a 3.73 diff.

Till




[Modified by till, 9:36 AM 3/8/2003]

 

You do press in the clutch to do it, and its loss per MPH, rather than RPM since RPMS go to idle when you press in the clutch (otherwise you'd get the engine compressing air causing the slowing down of the dyno, looking like a 150+hp loss!).

Anyway, I'm not sure what to make of the graph, as it shows I was only losing 35rwhp at 140mph. Maybe you are supposed to take the number the dyno jet gives you and then add a static number to the graph as well to get total loss, since there's no load on the drivetrain and thus, probably less friction. I could take a pic of the graph and post it, its strange looking, and its definitly on a curve, not a linear thing (increasingly lossy as mph goes up).

 

I did a coast down test on a dynojet and it showed something like 8% drivetrain loss. At this point I dont know what to believe is true, 8%, 15% or just a standard loss of 75HP across the board once you get over 500 rwhp...

vetterdstr :cheers:

 

This enforced my speculation that drivetrain loss is not a percentage within our operating power band, but rather a fixed number closer to 40-45HP. I'll play with our dyno this afternoon and see what I can come up with. :lurk:

 

I would agree that drivetrain loss is probably not a constant % of HP, but is still not a constant number (since friction losses increase with RPM in any common mechanical assembly). It seems logical that a given drivetrain setup would take X amount of horsepower to turn at a given RPM, no matter whether there is a 100hp engine driving it, or whether there is a 1,000hp engine driving it.

Example: take a standard transmission/driveshaft/axle combo hooked up to a 100hp engine. Current "% of HP" calculations would tell you that 15hp of this engine's output is being taken up via "drivetrain loss".

Now swap that 100hp engine for a 1000hp engine. "% of HP" calculations tell you that the exact same drivetrain that used up 15hp in the first test is now using up 150hp. Why are drivetrain losses significantly changed just because the engine has changed? Doesn't make sense, does it?

I would think that drivetrain loss would be on a scale totally independent of, and unaffected by, engine HP. It would be a sliding scale starting at zero HP (at zero drivetrain RPM), and be a fixed number for any given point on the drivetrain RPM scale.

I'm surprised no one has done this already -- seems like a simple test to conduct. Get a C5, pull the engine, and attach an electric motor/measuring device to the tranny input shaft:
* How much HP to drive a C5 manual trans drivetrain at 1k RPM? At 3k RPM? At 6k RPM?
* How much HP to drive a C5 auto trans drivetrain at 1k RPM? At 3k RPM? At 6k RPM?


[Modified by Kent1999, 3:47 PM 3/9/2003]

 

I think you'll have a better time modeling this if you remember that (a) the purpose of gears is to transfer, or translate, rotational energy or torque (not horsepower) and (b) for any given system (your drive-train, for example) there are inherent losses that can be modeled to give you the (average) efficiency as a ratio (or percentage) and (c) it's not negative horsepower that you're discussing but a loss (a+b).

Sir Isaac Newton did a really good job in trying to explain this mess with his amazing insights into the laws of motion and the conservation of energy. Not sure he'd have driven a 'vette though what with him being a Brit ;)

 

Nobody has done it because then you couldn't take your rear wheel numbers, and calculate a 15% loss and tell everyone inflated engine HP numbers. :lol: :lol:

 

I am also curious about the diff in rwhp/trq with a 1 to 1 run (4th gear, 3:42 rearend in our case) and changing to a 3:90, 4:10 etc

 

Sliding dynamic friction is more or less constant, regardless of velocity. With an increase in velocity, the force of drag does not increase; heat increases.

A car also has fluid in the drivetrain. Drag due to motion through a fluid increases with the square of the velocity. The coefficient of friction for these fluids is very low (in the case of bearings, etc.), so the velocity has to get quite high before there is a significant effect. The transmission and rear-end have gears moving through fluids; here the drag is higher, especially in areas where the teeth are perpendicular to flow.

The old 15% rule probably isn't a bad ballpark guess given the constant and dynamic friction in a car's drivetrain.

 

The F1 guys measure these frictional losses by running a 'special' ignition program where cylinders are 'not fired' every once in a while* (but fired often enough that thermodynamic changes are avoided). By measuring the engine at full song and then remeasuring the enigne with (say) one cylinder not firing, the friction stays the same, but the engine is only producing 9/10 of the power it was before (10 cylinders, 9 of them firing).

Knowing the frictional losses allows them to determin the enegry captured by the piston and transfered to the crankshaft. Knowing the geometry of this system, they can compare the expected energy with the measured energy. THis tells them about the efficiency of the ignition and burn rate ot the 'gasoline'.

In principle, one could do the same by removeing one spark plug/wire. The difference IS the power produced by that cylinder.

*They have to mis-fire cylinders in a special way to avoid things like crankshaft failures,...

 

Or we can just not give a crap and do something similar to the move from Gross to Net HP. Just have the manufacturers advertise RWHP and RWTQ.

 

Even more useful would be to also quote HP/TQ as the total area under the curves throughout the RPM range.