There have been countless threads now that have hit upon drivetrain loss percentages. the two schools of thought here are that:

1. it will take a set amount of power to turn the same drivetrain so therefore the actual drivetrain loss is a set number, say 80hp rather than a percentage.

2. the loss is a percentage, so a 500 horsepower motor will lose more power through the drivetrain than a 300 hp stock ls1, accounting for the same exact drivetrain setup.


so if a dyno measures the power of a vehicle by acceleration, of sorts, by using torque and a formula, wouldnt our drivetrain losses be from the acceleration of the components? The faster you accelerate the drivetrain components, the more losses you will have as it takes more effort to accelerate the drivetrain at at increase of 50 rpm per second than it does to accelerate the drivetrain at an increase of 25 rpm per second. meaning, The faster you want to run, the more power it will take to just get all that force to the ground?

Nine-Eight

 

As far as the dyno "measuring" power by acceleration.... Maybe clearer to say that it "calculates" power by acceleration but then that is only true when you are using a INERTIA DYNO. Any dyno that actually measures torque (as none actually measure power) can test two different tq setups at the same acceleration rate if you wish to do that. Even then the force against the drivetrain is greater with the higher hp car so you will have a higher drivetrain loss due to frictional losses etc.

Back to the different acceleration rates... that is true even with the same hp car. You may not be changing anything concerning the drivetrain but anything that increases the acceleration rate will usually lower your measured tq.

Just like when we run a engine on the dyno if we run a steady state or step test, or 100 rpm/sec instead of 300 rpm/sec, we should see a higher number. The engine is not making anymore tq it is just not losing as much, or using as much to accelerate.

 

I agree with you. I had not added anything that specific about the relation a dyno has with torque because we are then hitting on the topic of defining torque, horsepower, and exactly how it applies to acceleration of the car and I didnt want to go that far in depth.

When I say that a dyno measures acceleration I was not being as specific and punctual as you have taken it. I dont mean the dyno actually looks for and measures acceleration as a figure of some kind. two seperate cars, I.E. our 300 and 600 horsepower ls1 cars, will have different drivetrain losses even though they are set up exactly the same, only because the cars are accelerating on the dyno at a different rate. I am trying to be clear that that acceleration is what causes your different drivetrain losses.

For anyone who is curious, a dyno will NEVER, EVER measure horsepower. Ever. Horsepower is a calculation, and only that. the simplest way, although not 100 % correct, to explain torque and horsepower is that horsepower is a figure describing torque over time. torque is a figure of work, meaning 1 ft-lb of torque is a set amount of work, defined just as clearly as a simple measurement, such as 12 inches or one foot. its a figure, and horsepower describes what rate that work can be done. the mathmatical formual for this is:
horsepower= torque x rpm / 5252

there is a crossover that is constant, ie, can someone maybe point out what the interesting thing about a dyno readout is that will occur at 5252 rpm?

A dyno may be able to calculate horsepower, but it can only measure torque. torque is a twisting motion, and horsepower is something entirely different, a rating of that twisting motion.

Your last paragraph is exactly what I was illustrating. if you load the car more, by providing more resistance on the dyno rollers, and therefore accelerating the car at a slower rate but still at max throttle, you will incur less drivetrain loss and this in itself shows that your losses are occuring due to the acceleration of the mass in the drivetrain.

Nine-Eight

 

Just to be clear and accurate.... A dyno CAN measure tq at any throttle opening. It doesnt know if you are full throttle or not. That is how we set up different load points at part throttle when tuning.

When testing no load/inertia only or on a inertia only dyno the dyno actually does not measure torque but calculates horsepower and BACKCALCULATES (the best word i can think of .. LOL ) torque from that.

I think the formula is clear about why hp and tq cross at 5252. For the people who are not clear on this, it is simply the movement of 1 ft lb of work one revolution.

(One ft lb one revolution is in a 2 foot diameter circle, 2ft x 3.14159 (pi) = 6.28318 ft.)

That 6.28318 feet means that we did 6.28318 ft lbs of work.
Because 33,000 ft lbs tq per minute equals one horsepower, when we divide 6.28318 into 33,000 we get our 5252 constant.

Same as above, Inertia dyno testing DOES NOT measure torque but measures the acceleration rate of a known mass and calculates hp from that.
( F = M x A ) Again, Then the dyno software back calculates torque from that hp figure using the rpm reference. I guess there are different ways to express that but I think it is clearer to say it like that for the people that do not understand this.

Of course with a load cell dyno we are measuring torque directly and then we calculate hp from that with our rpm reference.

Dieknow

 

most of the time when one speaks of drivetrain losses it is a hypethetical (?)figure of 10-15% but how much is it exactely ?

you know what would be really nice to see ? that is a turbocharged engine that has been on an engine dyno with various boostlevels is then build into a car and that car is then put on a car-dyno at the same boostlevels as was on the engine dyno, now you can see exactely howmuch drivetrainlosses occur with increase in horsepower.

i would love to see that

 

if you want an equation for drivetrain loss it will look something like this:
(.02 x horsepower) + (constant). this is just an example, but their are certain parasitic losses that will always be there. however, like you originally stated, as the horsepower increases so will the loss due to the moment of inertia of the spinning objects: (flywheel, engine components, clutch, tranny gears, driveshaft, diff gears, rear axle(s), and wheels) if everything was weighed and their centroids were found, then an equation could be produced that would be fairly accurate... seems like a waste of time though. just go to the track! btw, lighter wheels and tires will produce higher dyno numbers... the same concept applies for anything i listed above that spins... the lighter the object the less energy required to change its inertia and therefore the more horsepower you see on your dyno sheet. however, your overall power hasnt changed at all but your dragstrip times will increase by an almost negligable amount considering the fact that the power required to spin the rotating objects in your vehicle is an entire level of magnitude lower then the amount of power it takes to accelerate your car.

 

Inertia Dyno's calculate horsepower by measuring the time it takes to accellerate a known mass (work/time). They then take the intergral to calulate the torque.

Load cell dyno's measure applied force (torque) then derive horsepower from that.

Gotta love Newton


In our tests between engine dyno's and and our Mustang MD-1100SE dyno, we've seen as much as a 19% (100lbtq) drivetrain loss through a manual transmission. There's some food for thought.

 

If it means anything, and probably doesn't. My combo should be together in about 2 months. My builder talked me into dyno tuning on an engine dyno. However, he won't be doing the tuning as he is an engine builder but is getting a dyno cell installed at his shop. The guy doing my tuning has a dynojet at his shop. Neither of them have seem back to back dynos being done each way on the same tune. They want to try it on my car. I will post the results when I have them.