how do you measure the torque of something that is rotating?

Thread Starter


Joined Oct 3, 2010
I had an idea to make a homemade bench dynomometer for testing the mechanical power output of motors (both electric and possibly small gas engines) at home. I realised that I have no idea where to start because I don't understand the principle behind it. I understand that hp= torque X RPM. I understand that torque = distance to center X force.

I think of a torque wrench when I think of torque. you put the wrench on a nut and turn the wrench until the needle points to the desired number and then you stall. I could simply measure the torque of motors by stalling them out with a torque wrench and read the number (if I wanted to lose fingers and be bashed repeatedly about the face and head) but that wouldn't give me much useful data because the motor isn't spinning. So how on earth do you measure torque why spinning? I guess I could put a 1' diameter brake disk on the shaft and apply a constant braking force and multiply that constant by the RPM but i'm not sure if I've got the principle right. Is that the right concept (ignoring the fact that the brake would probably lose braking force as it heats up and all other real world variables)?


Joined Sep 30, 2009
Your on the right track. There are two basic types of dynamometers in use to day, the brake style and the electric eddy current style.

In the brake type instead of a disk brake the use a band around a drum and an arm attached to the band that applies force to a force gage.

The eddy current uses two disks and a electromagnet to act like a clutch. The disk thats not connected to the motor has an arm and load cell like the brake style.

Theres also the hydraulic coupling style that is like a torque converter.

The measurements of all of them are only approximations, no two read exactly the same even two of the same style made by the same company.

An electric motors torque is highest at stall,like a steam engine, Thats why there used as traction engines, like a train. An internal combustion engine makes maximum torque and horsepower at a certain RPM.

Thread Starter


Joined Oct 3, 2010
good to hear, thanks for the confirmation.

Here's another idea I ran across using inertia:

We came up with Horsepower= (Inertia*Alpha*Faita)/746. In other words, (Inertia* the change in Angular Acceleration* average time) / 742. The inertia of my drum, we came up with, is 5.8, plain and simple. Angular Acceleration (A) equals the change in Radians/Second, over the change in Time. And Faita (W) equals the average time taken to spin the drum. And then we divide by 746, because there are 746 watts in 1 HP. Maybe this chart will help.
5.8 Kg/m^2
change in Radians/Second
(R2-R1) / (T2-T1)
Average time taken
(T1+T2) / 2
If I could simply mount a fixed weight (heavy) flywheel on a shaft and use this math then I think I could get away with not using too many parts and save time & money. Has anyone seen this math before? is it correct? sound theory?


Joined Feb 11, 2008
You can use any convenient load like a water fan or air fan too.

The shaft torque generated by the motor is also reflected in an equal torque applied to the motor body, so you can use any convenient variable load and then measure the torque applied to the body of the motor. That gives you a lot more options for loads.


Joined Nov 30, 2010
RB beat me to it. I was going to say that I've seen horsepower/torque measuring machines that dissipate the energy into water, and measure the reactive force that is pulling against the mounting platform. Seems simpler than trying to grab something that is rotating, but only just a certain amount.