Torque Measurement 5kV Induction Machine

Hi,

I've been given the task of measuring the torque produced by an 5kV induction motor using two of these load cells.

They are attached to a bar on the frame of the motor perpenicular to the axis of rotation. The load cells are mounted one foot apart on either side of the machine.

Does anyone have any idea what would be the best way of going about this task? Any advice would be appreciated.

Eoghan

Five thousand volt induction motor? Are you serious?

Paul

I've worked on 3.3kV wound rotor motors and I can assure you no one in their right mind would let an amature within cooee of them. ...... plug in some load cells and monitor...?? not likely.!!!

Hi,
5kV or 5kW?

Tom... :slight_smile:

Sorry I was half asleep when I made the post. Its a 5 kVA 3 phase machine. Even typed it twice, I'm such a dope.

Those load cells don't look very suitable for an Arduino project. Or maybe there's another component you need which attaches to the 6-wire load cell and gives you a 0-5V, 0-10V or 0-20mA output. Or maybe there's a further component which takes that analog output and gives you a digital ouput, maybe as TTL serial if you're lucky.

An HX711 module would work fine with that load cell. I understand you'll look at it and find it incredibly cheap, but there's a huge jump in price or complexity between the HX711 modules/chips and trying to build something yourself.

I see that RS Components sells the "Z-SG" signal conditioner for these cells and has an RS485 Modbus output. If you want to spend that 180 euros then I'm certain someone can help you out with working with a modbus and Arduino -- not familiar with it myself but I know it's doable.

Right forgive my ignorance on this matter but why do I need a conditioner for use with these. I just presumed I might be able to amplify the output of the signal compare what I get to empirical results for torque from speed and power. There's an arduino mega and TFT screen measuring the machine speed in place already so that why I need to use the arduino.

Is this not feasible?

Hi,
You can use the 6wire as a 4wire load cell.

Tom.... :slight_smile:

eoghan1985:
Right forgive my ignorance on this matter but why do I need a conditioner for use with these. I just presumed I might be able to amplify the output of the signal compare what I get to empirical results for torque from speed and power. There's an arduino mega and TFT screen measuring the machine speed in place already so that why I need to use the arduino.

Is this not feasible?

Yes, it's feasible but "just amplify" is not exactly what is going on here. A strain gauge is a resistor. It's stretched by the strain on the metal substrate. This makes it longer and thinner so the resistance increases. The difference in resistance is tiny. Really really tiny. You need some sophisticated electronics to be able to measure this tiny difference. An Arduino is not capable of doing this by itself.

If I was you, I'd spend the 180 Euro on the converter listed under "essential accessories" and I would be using the RS232 output from that converter. But even that's not as simple as it seems - the serial input and output on the Ardunio is TTL, which needs a voltage converter to reach the levels specified by the RS232 standard.

It's still not going to be very good for your application. The device you linked to is good for measuring weight, where the thing being weighed is not moving. Your 5kVA motor is going to be vibrating, possibly strongly vibrating, so there will be some filtering required in the analog circuitry to get the answer you are expecting.

The "signal conditioner" IS an amplifier. Just a funny name. It's the same thing as the HX711 (albeit more expensive and with a modbus output. And probably better quality).

TBH this is a college project that has bee appended to my FYP at the last minute so its unlikely that I could get a €180 module (or two). THe load cells have been on machine for a few years now without being connected so I guess they'd be non-essential.

I was hoping that that the fact that there a pair of load cells on either side of machine that there would be a certain amount of vibration cancellation without too much need for filtering. Is that wishful thinking?

I was thinking of building an instrumentation amp from 3 op-amps for each load cell. I could use the HX711 instead (only a fiver each on ebay!) if there was much advantage in doing this, although I'm sure the more of this I design and build the better.

google prony break

eoghan1985:
I was thinking of building an instrumentation amp from 3 op-amps for each load cell. I could use the HX711 instead (only a fiver each on ebay!) if there was much advantage in doing this, although I'm sure the more of this I design and build the better.

The HX711 is an amplifier and a 24bit ADC. You can build your own Inamp, or buy an Inamp, but you still won't get the precision that the HX711 will provide.

Greensprings:
google prony break

That's kinda what he already has here.

Right I'm convinced. Just ordered a pair of HX711's hopefully they do the job

It might help if the OP posts a photo of his setup that shows how the load cells are mounted on the machine. (the prony brake)

What's the motor's rated voltage ?

raschemmel:
It might help if the OP posts a photo of his setup that shows how the load cells are mounted on the machine. (the prony brake)

What's the motor's rated voltage ?

Its a is 400V machine.

Ive included some pictures here. You can see the pair of load cells a foot perpendicular from the centre axis. The machine is coupled to a DC machine so it can be motor or generator.