0
\$\begingroup\$

I am looking to do some speed-varying torque tests on a system and drive it with a sensorless vector drive motor. However I'm now unsure of what to expect for accuracy on speed or position at low speed?

The waveform may be sine, triangular or square, and the pole count may be between 3 and 9

  1. Is the requested motor speed respected, at least on average with the drive, or does it slip?
  2. How much angular misalignment might be expected between the achieved position/phase and speed versus the requested position and speed?
  3. What would I expect for the output of a continually varying speed request against a varying load? Is the error on the order of a few degrees or a few revolutions?
  4. What motor layouts, permanent magnet etc. are compatible with vector control?
  5. What influence does waveform shape and pole count have on the positional accuracy?
\$\endgroup\$
5
  • \$\begingroup\$ It depends entirely on the motor. \$\endgroup\$
    – Andy aka
    Commented Nov 29, 2020 at 11:30
  • \$\begingroup\$ Drive waveform and whether induction or permanent magnet rotor matters greatly. Please define requirement more closely. Also number of poles. \$\endgroup\$
    – Russell McMahon
    Commented Nov 29, 2020 at 11:47
  • \$\begingroup\$ Concerns managed in edits. \$\endgroup\$
    – J Collins
    Commented Nov 29, 2020 at 12:07
  • \$\begingroup\$ Vector drive may be used to describe an induction motor, which inherently slips wrt drive waveform, or a permanent magnet rotor drive which does not. || Waveform may be complex to address torque ripple or simple square wave or sinusoidal or some mix. There is no one size fits all answer. But there are online references which cover many variations.search for Benjamin Vedder vesc, as a starting point. \$\endgroup\$
    – Russell McMahon
    Commented Nov 29, 2020 at 12:26
  • \$\begingroup\$ You will find out for your specific motor when you run the tests. Until then there are no simple accurate answers. Books and theses on Field Oriented Control maybe like krex.k-state.edu/dspace/bitstream/handle/2097/1507/… but nothing appropriate here. Induction and PM motors are very different, for a start. \$\endgroup\$
    – user16324
    Commented Nov 29, 2020 at 13:43

1 Answer 1

1
\$\begingroup\$

The waveform may be sine, triangular or square, and the pole count may be between 3 and 9

Pole count in motors generally refers to magnetic poles. Since there is a N pole for every S pole, pole count is always an even number.

  1. Is the requested motor speed respected, at least on average with the drive, or does it slip?

Induction motors have slip synchronous motors don't.

  1. How much angular misalignment might be expected between the achieved position/phase and speed versus the requested position and speed?

There is an angular relationship between the magnetic fields of the stator and rotor. Torque is proportional to the sine of the angle. The average speed error for a synchronous motor will be zero with respect to frequency or as close to matching the request as the resulting frequency.

  1. What would I expect for the output of a continually varying speed request against a varying load? Is the error on the order of a few degrees or a few revolutions?

That depends on the multiple factors that are involved.

  1. What motor layouts, permanent magnet etc. are compatible with vector control?

As far as I know, any valid structural motor design is compatible.

  1. What influence does waveform shape and pole count have on the positional accuracy?

Higher pole count results in fewer degrees of magnetic field displacement "electrical degrees" per mechanical rotation degree "mechanical degrees."

Relative position control is usually performed based on the moving components of driven machines that need to maintain alignment. It usually requires encoder feedback from the alignment points. The machine alignment point will probably need to tolerate an error due to the motor's torque angle variation for the expected load variation.

\$\endgroup\$
1
  • \$\begingroup\$ Excellent answers thankyou! \$\endgroup\$
    – J Collins
    Commented Dec 1, 2020 at 10:01

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.