Motor with torque sensor/control?

Thread Starter

Alice24

Joined Apr 22, 2011
46
Hi guys, do any of you know where I could find a motor with an accurate torque sensor? I don't mind if it's expensive; as long as it gives me torque reading that I can use/program/control.
 

MaxHeadRoom

Joined Jul 18, 2013
28,617
Torque is usually directly related to current, you could use a ACS712 ebay for $1.50 on a PCCT.
What is the nature of the motor?
Max.
 

Thread Starter

Alice24

Joined Apr 22, 2011
46
You're telling me I can use this "ACS712" to accurately measure torque along with a PLC? Really?

The motor is connected to a plastic device which I measure for various applications. Sorry I can't be specific, but let's say it's a lot of plastic measurements for the plastic industry and I want to have better control of my torque tests.

Frankly the more built-in the solution is, the better it is for me; even if it requires I pay a few hundred bucks more.
 
Last edited:

MaxHeadRoom

Joined Jul 18, 2013
28,617
Did you already decide on a PLC? If so you would need a analogue input card, if using a μP then it may be easier.
It may also depend on the mechanical design of the operation as to whether this method is suitable.
Is there an existing method in place that needs improving? Or?
Max.
 

Thread Starter

Alice24

Joined Apr 22, 2011
46
There is an existing method. It's a simple motor on one side, torque measuring device on the other side. I put a plastic part between them, turn the motor on and write the results the torque measuring device gives me (sometimes I bring the plastic till breaking point, sometimes I measure till specific amount, etc)

I'd like the motor to be more involved and controllable. Say, it reaches a certain torque, I want it to stop automatically without me turning the switch off.

Currently the best solution I see is buy a "Rotary transformer" and put it instead of the motor. If I understand correctly a rotary transformer is kind of like a DC motor with a built-in sensor that outputs electrical output based on the torque it senses. Correct?
 

MaxHeadRoom

Joined Jul 18, 2013
28,617
Do you know the nature or model etc of the existing torque measuring device? What media is it actually measuring?
Is the motor operating some kind of Geared winch method?
Max.
 

Thread Starter

Alice24

Joined Apr 22, 2011
46
Yes the current torque measuring device is a Mark-10 MTT03-12 , quite a fancy measuring device. It even comes with a propriety software that records/shows results live on the screen via USB, I've contacted them but they won't give me the code so I can't program the data it outputs in real time order to control the motor.
The motor is a simple DC motor -> http://www.micromotorssrl.com/gear_motor_b138f.html
 

MaxHeadRoom

Joined Jul 18, 2013
28,617
Too bad as it should be simple for the OEM to make a change themselves to the software to output a signal at a operator pre-set level.
Max.
 
Seems to me you are making things more complex than they need to be...
I agree that you can measure torque by measuring current but with basic DC motor, actually any motor, there will be a huge variation with speed...

The point it isnt linear or even likely to be all that repeatable, this I know because it is something I use regularly when I need a rough guide to tell me a rotor is locked or even nearly locked.
My application uses a standard o to 10V plc input card and the chip suggested above... I am driving DC gear motors at 24V that open spear valves or guide veins. Of course there are limit switches but they dont help if the valve closes on debris... bring on an overcurrent trip that can be set programmically.

However of you actually want to measure torque I would suggest the following...
Mount your drive motor in a gymbol so that it can rotate about its output axis.
Constreigne it with an arm, of known length.
Measure force at the end of the arm.
N m is the very definition or torque and PLC compatable load cells are not hard to find.
You will probably end up working in g/mm and converting but its just maths.

If you want to measure torque in a drive chain then you will need some sort of sprung coupling...
There are comertial units available but they are far from cheap.

Considder a dog clutch or jaw coupling.
http://www.abssac.co.uk/p/Shaft+Couplings/Jaw+Couplings/39/#&panel1-8

If the compressible eliment is not loaded then there is little or no displacement, output relative to input.
Add torque and the compresible element deforms to a degree proportional to the load.
If both input and output parts have some form of encoder attached the displacement can be measured, You could do this with simple optical pickups.
Of course you would need to calibrate the thing unlesss you are buying a coupling with known properties, designed for the job and the compressible eliment would almost certainly have to be a spring to minimise temperature effects but it would work.

I suppose you could even have some sort of self powered circuit with a load cell in the coupling and wireless data, a freequency perahps, transmitting the torque in real time back to a reciever that then talked to the PLC but in all honesty a calibrated load cell on a well built torque are is almost cirtainly the easiest and likely most accurate option.

https://en.m.wikipedia.org/wiki/Torque_sensor ... A bit more detail although I still think an arm is the way tom go in your case.

Hope that helps,
Al
 

cmartinez

Joined Jan 17, 2007
8,218
There is an existing method. It's a simple motor on one side, torque measuring device on the other side. I put a plastic part between them, turn the motor on and write the results the torque measuring device gives me (sometimes I bring the plastic till breaking point, sometimes I measure till specific amount, etc)

I'd like the motor to be more involved and controllable. Say, it reaches a certain torque, I want it to stop automatically without me turning the switch off.

Currently the best solution I see is buy a "Rotary transformer" and put it instead of the motor. If I understand correctly a rotary transformer is kind of like a DC motor with a built-in sensor that outputs electrical output based on the torque it senses. Correct?
Hello Alice. How much power does motor whose torque you want to measure have? As Max just said, a good way to monitor it is through how much current and at what speed the motor is running. But if you want a more direct measurement, there are dynamic torsional load cells out there that you could use. Just google images for "torque sensor".
 

strantor

Joined Oct 3, 2010
6,782
+1 on what dyslexicbloke said.

The gimbal/load cell is a good idea.

The amps of the motor correlates to torque at the motor shaft. If you have any kind of gearbox or screw after the shaft, it ruins your repeatability, is greatly dependant on lubrication and wear, and could introduce nonlinearity into to the amps:torque relationship.
 

Thread Starter

Alice24

Joined Apr 22, 2011
46
@Dyslexibloke
I think your suggestion sounds a little over complicated. All I want is to control a motor to stop at a certain torque.

Aren't rotary torque sensors basically DC motors with sensors? Can't I just use that coupled with a PLC?
 

cmartinez

Joined Jan 17, 2007
8,218
@Dyslexibloke
I think your suggestion sounds a little over complicated. All I want is to control a motor to stop at a certain torque.

Aren't rotary torque sensors basically DC motors with sensors? Can't I just use that coupled with a PLC?
No. They're basically torsional load cells rotationally coupled.
 

Sensacell

Joined Jun 19, 2012
3,432
@Dyslexibloke
I think your suggestion sounds a little over complicated. All I want is to control a motor to stop at a certain torque.

Aren't rotary torque sensors basically DC motors with sensors? Can't I just use that coupled with a PLC?
Won't work. - Yes torque is proportional to current, but this idea totally ignores the energy stored in rotational inertia.
You can have current = 0 while the motor is putting out a massive amount of torque, dissipating it's inertial energy.

Imagine a motor running at high speed, cut the power just before it hits a hard stop, you can create gear-wrecking torque and still have an armature current of 0.
 

MaxHeadRoom

Joined Jul 18, 2013
28,617
There is an existing method. It's a simple motor on one side, torque measuring device on the other side. I put a plastic part between them, turn the motor on and write the results the torque measuring device gives me (sometimes I bring the plastic till breaking point, sometimes I measure till specific amount, etc)

I'd like the motor to be more involved and controllable. Say, it reaches a certain torque, I want it to stop automatically without me turning the switch off.
Is the torque sensing procedure of the plastic involve less than one turn of the material, or does it twist it with multiple turns?
What is a typical measurement at breaking point?
Max.
 

Thread Starter

Alice24

Joined Apr 22, 2011
46
@cmartinez
@Sensacell

Yes, you're right, I've had the realization yesterday it's not a motor with a built-in sensor, it's just a sensor.

@Max
It's lots of different plastics, it can be one turn, half turn, 10 turns. Measurements also vary, can be 10 N/cm, 50 N/cm, 150 N/cm...
 
Complex?
The level of complexity is a function of how accurate you want to be....

I would be inclined to suggest that the most standard way to do what you want is to use a shaft mounted torque sensor, which will of course be a significant cost.

If you attempt to build something then getting it accurate is going to be tricky if the build is complicated and or small.
Given the requirement to cope with small rotation as well as multiple turns, going with two displaced pulse encoders coupled with something deformable seems like a poor plan and as I said earlier any malleable material will likely deform differently in different temperatures.

Gambolled motor... Not hard at the scales you are talking about...
If pick a motor with a 90 degree box and a long shaft...


Not checked specs on this , just a pic to illustrate the point.
Mount the motor using bearings on both shafts, one of which will connect to your test sample assembly exactly as you do now.
Obviously if you drive the motor like that the whole thing will try to spin round the shaft, so you constrain it with a single point attachment where you also measure force... Simple

The longer the arm the smaller the force you measure for a given torque. and if your force sensor has a limited range or is fixed, like a switch with a known pressure to push it, then you can still set different torque values by moving the detection point.
In fact now that I think about it... How about simple mechanics.

http://glossary.periodni.com/glossary.php?en=unequal-arm+balance

If the motor shaft is placed at the point where the red triangle is and you add a counterweighted arm to the gearbox, ballanced when m2 is at 0, you could set a precise torque required to lift the arm simply by adding mas at a known distance.
When the arm lifts you are at the set torque and a contact, which the arm touches, stops the motor.

Calibration is inherent... Balance is easy to check, just visually, and if the arm is only moving a very small distance the change in effective length will be so small as to be insignificant so your measurement will be as accurate as your distance and mass measurement.

Two bearings and a ruler... it doesn't get much simpler than that.

Al
 
Top