I was thinking about this today, so hear me out. A chassis dyno measures torque, nothing more. The horsepower figure is a numerical derivitave based off of torque. This equates to....

torquexRPM/5252= HP

Consequently, at 5252 RPM, torque and horsepower will always be equal.

Now, lets say that my engine makes 300HP and 300TQ. On the dyno the torque does not get multiplied by the fourth gear ratio of the transmission. Let's use a 1.0:1 (assuming 4th gear) ratio. So there is 300 ft lbs of TQ coming out of the transmission(minus frictional losses) and going into the read end. Let's assume that the rear end gears are also 3.0:1. That means that there is 900 ft lbs of TQ available to both axles/wheels. The chassis dyno sorts out all the mathematical formulas and spits out your rear wheel numbers. Now, let's say that all you did was change the rear end gears to a 4.0:1 ratio. That would be...

300TQ(on motor) going into the 1.0:1 4th gear which is still 300, and goes into the 4.0:1 rear end gears and comes out as 1200 ft lbs to both axles/wheels. Now, would the chassis dyno spit out a different(higher) number because the available torque has increased?? Also, seeing as how the horsepower figure is based off of a now higher torque number, could it be safe to assume that the rear HP figure would be higher also??? What do you think? I was just kickingthis idea around and figured I would post it up and get some comments. The reason why I am asking is because it will be time to do some chassis dyno tuning for me pretty soon and I have an open differential with 3.42's in the pig. I plan on running a 9 inch ford rear end with a posi and 3.73's. If I get my car tuned with the 3.42's and then get the other rear end, I don't want to have to get it tuned all over again because the rear end gears threw off the numbers...

 

I think it uses mph vs rpm vs time to figure out how much torque is produced. I dont know the exacts, but I am pretty sure thats how it trys to keep things accurate.

Regardless of gear, your horsepower reading should be the same if not close.

 

ppl will argue that it wont change the numbers but i've seen it happen on a dyno where a guy lost 15lb ft of torque peak with higher numerical gears. based on my theory i think it was software error

some dynos i think calculate the speed at which the roller moves and the time to get it accelerating to its peak acceleration. somehow it takes that to calculate tq, but i'm not sure. deeper gears will make the car run thru the power band quicker and should make the roller move quicker but the time it takes to do so should decrease. thus the software should calculate no change in dyno numbers

 

All chassis dynos use the speed of the wheels to correct the torque reading. Any differences are due to differences in drivetrain loss, test conditions, or normal measurement error.

Without correction, you'd see differences even just from changing the tire size.

 

Forget which one it was, but a couple of years ago a mag had an article about determining horsepower. Engine dyno was the most accurate. In decreasing order, the next most accurate methods were 1/4 mile performance, engine equipment, and chassis dyno.

 

indeed, because a car that runs 12's at 110mph at a weight around 3600lbs should put down more than 308whp. my buddy was abit dissapointed in those numbers but the car runs faster than what ppl would think based on those figures

 

What Apeiron is trying to say in a nice way is that inertia type dynos pretty much suck at determining actual horsepower. They are very accurate in determining changes in horsepower. The weight and inertia of the drum that the wheels turn is a known constant. The dyno knows how much horsepower it takes to accelerate it. The problem is that the inertial weight of the tires, driveline, and ring gear are added to the weight of the drum to be accelerated. These variables have to be inputed so the computer can compensate for them. This is where the inaccuracy comes in.
A brake dyno measures actual torque produced by the wheels and measures horsepower at a constant rpm speed. It is much more accurate at determining actual net horsepower because acceleration and inertia are not a factor. These are used to measure horsepower on semi trucks.
The idea with inertia dynos is that the drivetrain rotating mass that must be accelerated will make a given car faster or slower at the dragstrip also. This is why drag cars use aluminum driveshafts, 9" converters, light flywheels, and low mass tires. The inertia dyno will see changes from this where a brake dyno would not.
Confused yet?

 

Fascinating thread...

 

Good responses and info in this thread.
Generally, steeper gears lose a tiny bit of efficiency and will show just a little less hp on a chassis dyno.
Horsepower (rate of work) is the direct output from an inertia dyno. Torque is calculated from hp and engine rpm, not roller rpm.


Orr89, If he's putting most of his power to the ground right off the line, then that 308 rw number sounds about right to me. My 3950 lb car shows around 310 STD rwhp on a Dynojet and runs 12.3's at 107+ in good air. It 60's about 1.60.

 

When I had my 2000 Tahoe on the dyno they had to run it in second gear (A4). Reason was the tires on my Tahoe did not like the 120+mph of the dyno pull in 3rd gear. Anyways there was an 18% correction factor to be added to the horsepower rating because of the 2nd gear run. So the lower the gear ratio the more horsepower you would be "losing" on the chasis dyno.

 

OK, I'm confused.
My truck stock had 3.73 gears and 265/75/16 tires.
My truck now has 4.56 gears and 315/75/16 tires.

With the same engine, my truck would register more HP/TQ with the 3.73's and 265's? Because of the smaller tire diameter?

 

In theory it won't read any differently.

 

guess I'm worrying about nothing

 

with the larger tires you would have more loss due to both tire friction and extra mass that needed to be accelerated.

 

I was thinking about that...I wonder how much