# Is there a cheap and accurate electromechanical method to provide a configurable counter-torque?

Given a continuously rotating motor shaft, is there a cheap and accurate electromechanical method that would provide a configurable counter-torque?

Rotation is in one direction only, RPM range is 0 - 12000, torque range from 0.001 to 1 Nm.

edit: Before the edit I was asking for a constant counter-torque regardless of the motor's RPM. I have removed this as it just adds complexity, increases cost and is not strictly necessary IF the magnitude of the counter-torque is precisely known at whatever RPM the motor's shaft is currently spinning at.

• Define 'cheap' and 'accurate'. What is the expected rpm and torque range, and why do you need constant counter-torque? – Bruce Abbott Nov 24 '19 at 2:39
• define RPM range and Torque level including allowed negative RPM and negative rotor position range to pickup slack. Also define skill level to implement servo controls. There will need to be sensors for max position velocity +/- acceleration during slack and tolerance for overshoot. – Tony Stewart EE75 Nov 24 '19 at 6:57
• I have seen competent Mechatronics Engineers get the right equipment but fail to understand how to get the software to work and resulted in expensive material damage being overlooked. – Tony Stewart EE75 Nov 24 '19 at 7:03
• From the desired torque range of 0.001 to 1Nm I presume you want an accuracy of 0.1%? Are you trying to make a dynamometer? – Bruce Abbott Nov 25 '19 at 6:00
• @BruceAbbott Yeah, that would be one of the main applications. – xnor Nov 25 '19 at 21:28

Do you understand the concept behind using the Prony Brake for measuring torque?

You can use electromechanical feedback (a simple servomechanism) to maintain constant torque at arbitrary speeds.

• How precise is a prony brake though? Aren't there significant heat losses? – xnor Nov 24 '19 at 18:11
• As precise as you want to make it. And any form of countertorque is going to involve dissipating the power involved one way or another. – Dave Tweed Nov 24 '19 at 18:30
• Of course but if I understood it right then the heat loss is not part of the measured torque of a prony brake. It only measures the actually "transferred" force. How would I get it arbitrarily precise if it involves this unknown? – xnor Nov 24 '19 at 19:34
• @xnor Why are you worried about getting arbitrarily precise when you specified cheap? It's not like you're willing to spend an arbitrarily high amount of dollars in the first place. And where do you think the heat is coming from? That output power = torque x RPM has to be dissipated somehow. Spinning motor shafts don't spontaneously radiate heat from nowhere. They can only transmit rotational motion. I think you have your cost vs performance priorities and mental model power of power all mixed up. – DKNguyen Nov 25 '19 at 3:08
• @xnor BTW, a smooth running DIY prony brake is not as straightforward as it looks. Get earplugs. They can be ear splittingly loud in my experience, even small ones. All that stiction and high frequency motion. – DKNguyen Nov 25 '19 at 3:17

The cost will depend on the range of torque and speed you need and particularly the maximum horsepower, torque multiplied by speed. The required steady-state accuracy and dynamic performance will also be factors. You also need to consider how much of the system you are going to build yourself and how much you are going to purchase.

The prony brake may or may not be the least expensive option depending on the complete requirements. Look at this Dynamometer information

If you use a trunion-mounted absorber, the torque can be measured directly as the force exerted on the lever-arm that prevents the stator of the absorber form rotating. That method of measuring torque should be the least expensive. A prony brake can be arranged to serve as an electrically-controlled, trunion mounted absorber. At the power level in question, wear and power dissipation should not be problems. At the low end of the torque range bearing friction could become significant. That would not interfere with measurement, but it is a residual torque that will set the lower limit of applied torque.

• I thought about using a brushless motor as a generator with a resistive load but this just adds more (unknown) variables and would require calibration of torque ... which is kinda what I'm asking for in the first place. – xnor Nov 24 '19 at 18:19
• @xnor whatever system you eventually design and install, calibration will be needed. – Solar Mike Nov 24 '19 at 18:52
• Yeah but if it requires a \$2000 measurement device then that defeats the whole purpose of the question. – xnor Nov 24 '19 at 19:25

There are "Torque Motors" that provide approximately constant torque independent of RPM. An example application would be a driving a take-up spool on a web process.