ROTATIONAL TORQUE TRANSDUCER

[0PETER0]

Hi,

I am thinking of using an rotational torque transducer from some brand to measure the outgoing torque at the end of an rotating axe. There will be no external load at the end of the axe. Will this work? Or do I have to put the torque sensor between the axe and an external load (flywheel) or something similar?

What do you think?

 

[MikeHalloran]

If there's no load on the axle, there's no torque to transmit, and the transducer's output should be zero, unless it's broken.

If you want to measure the reaction torque as the axle accelerates, then you have to insert the torque transducer between whatever moves the axle and 'ground'. It's probably cheaper and easier to just instrument whatever the prime mover is, measure the reaction forces, and compute the torque.



Mike Halloran
Pembroke Pines, FL, USA

 

[0PETER0]

If I understand you well I have to do it this way :
(engine)axle -> torque transducer -> ground
Is this right?

Could the ground be a kind of flywheel with a certain inertia? What is the influence of the inertia of the flywheel regarding the measured torque?
Is the flywheel just necessary to be able to have a load on the axle through the torque sensor.

 

[MikeHalloran]

Rotary torque transducers are very expensive, mostly because of the 'rotary' part.

It's not clear what you are actually trying to do, but I'll assume it has something to do with an automobile and wing it:

You could use a rotary torque transducer in between the engine and the transmission, or one with a larger capacity spliced into the driveshaft, or one with a still larger capacity spliced into one axle. You can buy the latter as a wheel spacer that can measure torque and lateral forces. Bring your wallet.

You could use a _static_ torque transducer as the only rotational restraint for the engine, softening up the engine mounts in the rotational direction. This would measure driveshaft torque.

Or, you could put a displacement transducer on each motor mount, measure the spring rate of each motor mount, and use a computer to infer the forces being applied to each mount, and the torque necessary to do it. Something like this was offered as an aftermarket add-on at least ten years ago, giving the driver a dynamic display of power developed. The installation was very expensive, so the market was pretty much stolen by dashboard mounted accelerometers.

Similarly, you could fit strain gages to all of the fore/aft locating links for the axle, and compute the net torque on the axle.





Mike Halloran
Pembroke Pines, FL, USA

 

[0PETER0]

I supposed that those torque sensors are expensive but that they are SO expensive I am not aware of.

Like you guessed it, I would like to build a neat small machine that I can attach instead of the wheel.
I need two of them on a 2 wheel driven car.

In that way the car is attached to the two machines.
Here you have the idea

http://www.rototest.com

I am for the moment only interested in measuring the produced torque. But the way of attaching has to be similar.

That is the reason why I thought that the torque sensor are ideal.

I can also try to measure the angular acceleration of a flywheel that I connect to the car it's axle. T = I . alpha

The thing is that the torque sensors are direct and all the rest is in fact torque based on calculation. Which of course is an estimate.

What is your idea on the subject? You seem to have some ideas.

 

[MikeHalloran]

For rotary torque transducers of all kinds, these are the go-to guys:

http://www.himmelstein.com/WheelTorque.asp

they will tell you the torque while you drive the car.

The difference between that and the rototest is that you have to drive the car, or provide an external simulation of road load, i.e. a big flywheel or a big brake. It appears that the Rototest provides brakes, and optionally motors for nonpowered wheels.




Mike Halloran
Pembroke Pines, FL, USA

 

[0PETER0]

I have seen their system. very nice but I can not use it for the moment. My meaning is to have a portable dyno like the rototest. A brake is to expensive (or they are asking crazy money for a simple thing), the problem is I can not build it myself (so it cost money again). What I well can do is to place a reasonable flywheel at the other end of the sensor. In this way the engine of the car has to overcome this inertia.

Assume this works, what will be the influence of the inertia on the quality of the torque measuring results?
If I take a small inertia, the flywheel will spin up very fast. If I take a big inertia flywheel the accelerationn will take longer. I suppose al this has no influence on the torque readings of the device (not looking at sample frequency of the system of course)?

Do you agree on this? I am not sure.

 

[MikeHalloran]

You don't, strictly speaking, need to bolt on a flywheel, or even a torque transducer. You already have a large inertance in the mass of the car itself. An accelerometer can provide enough data to let you calculate the net torque at the wheels, provided you have measured the car's weight.

The accelerometer- based instrumentation that's sold to drag racers today is surprisingly sophisticated, and probably within your budget.




Mike Halloran
Pembroke Pines, FL, USA