1
\$\begingroup\$

I have a task where I want to measure a 24Vdc motor's rotational speed using an AC tacho attached to the same shaft of the motor. When running the motor at full speed, the tacho generates around 9Vac at a max frequency of around 1kHz. So I was going to feed this through a precision rectifier and ADC to sample the digital value with MCU but this got me thinking:

If I am only interested in measuring the speed of the motor, do I only need to measure the frequency of the tacho signal? I don't think I need a rectifier and could use a zero crossing detector and counter to calculate RPM?

Or is it more complex and does the voltage magnitude have to be considered too as this changes with frequency?

How is this achieved in current technology?

EDIT

I have measured the AC tacho signal. At full speed it generates around 8.6Vac at a max frequency of 900Hz. When turned slowly by hand it produces approx 150mV at around 25Hz although I won't be running the motor this slowly so I suspect the low end voltage to be at least 0.5V to 1V.

EDIT 2

I am thinking of using an op amp comparator to convert the positive half of the AC tacho waveform to a square wave signal. So the square wave is high during positive cycle, and low during negative cycle.

I am thinking of measuring the time between the pulses to calculate the motor's rpm. I want to use a 16 bit counter that starts counting on the rising edge of the square wave pulse and capture the counter value during the next rising edge. How can I convert this counter value from 0 to 32768 to an rpm range from 0 to 4000?

Thanks for your help everyone!

\$\endgroup\$

1 Answer 1

Reset to default
2
\$\begingroup\$

If I am only interested in measuring the speed of the motor, do I only need to measure the frequency of the tacho signal?

Yes, this will provide a more secure method for inferring speed but, at low speeds the voltage level p-p may be a tad low to adequately detect frequency unless you have provided a signal stage that uses (say) a comparator with a little bit of hysteresis. (or maybe some form of automated gain method).

I don't think I need a rectifier and could use a zero crossing detector and counter to calculate RPM?

I would recommend that you look at the output on an oscilloscope before making this judgement. You may need to jump through a couple of small hoops at low frequencies and only the visual display will tell you that.

\$\endgroup\$
8
  • 1
    \$\begingroup\$ I will examine the output at a range of frequencies and report back with findings. I'll try to get some screenshots too. \$\endgroup\$
    – David777
    Nov 17, 2020 at 13:14
  • 1
    \$\begingroup\$ Of similar importance is the waveform shape. It probably looks a lot more lumper at low speeds. The amplitude range is good so that’s one tick to one box but realistically you want the signal to be monotonic else some filtering is going to be needed. Been there etc.. \$\endgroup\$
    – Andy aka
    Nov 19, 2020 at 21:02
  • 1
    \$\begingroup\$ @David777 what did the tacho output waveform look like at extremes of speeds? \$\endgroup\$
    – Andy aka
    Nov 25, 2020 at 11:19
  • 1
    \$\begingroup\$ No, waveforms are needed, not guesswork. Been there before. \$\endgroup\$
    – Andy aka
    Nov 30, 2020 at 0:30
  • 1
    \$\begingroup\$ Did you get round to producing waveforms yet @David777 \$\endgroup\$
    – Andy aka
    Dec 6, 2020 at 14:47

Your Answer

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

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