High-frequency (inaudible by humans) acoustic noise from stepper motor

[aalnaif]

Hello,

I am using stepper motors for some neuroscience experiments on rats. The problem is that our experiments are being corrupted by high-frequency noise. The acoustic frequency of the noise is high enough that humans can't hear it, but rats can (humans can hear up to 20 kHz, whereas rats can hear up to 80 kHz). Using a microphone, I measured the acoustic frequencies of the noise coming from the motor, and I observed peaks at 22.5 kHz, 43.25 kHz, 88 kHz, 110 kHz (note that the larger frequencies are harmonics of 22.5 kHz, with the exception of 65 kHz which is missing for some reason). The weird thing is that the motor does not need to be moving for this sound to happen; simply supplying idle current to the motor generates this sound. I checked to make sure that this sound was not present when no current was supplied to the motor, and I also ensured that the sound is indeed coming from the motor and not another piece of equipment.

The stepper motor we are using is the 2-phase Vexta PK223PA (http://www.orientalmotor.com/products/pdfs/2009-2010/C/usa_st_2pk_pv.pdf), and the driver is Applied Motion Products Si3540 (although I don't think the driver matters since no movement is necessary for this sound to occur).

My questions are:

1. Why would simply supplying idle current cause such high-frequency noise, regardless of whether the motor is moving?

2. Would a 5-phase motor also produce this sound?

3. Are there any quieter options? How do servo-motors, linear motors, and solenoids compare to stepper motors, in terms of acoustic noise levels?

 

[AlephZero]

I think the driver does matter. Most likely it is not producing a DC output when the motor is stopped. The first thing I would do would be put an oscilloscope across the motor input terminals. My guess is you will see some sort of pulse waveform, with no 3rd harmonic content.

You should be able to hold the motor in a fixed position with a DC current and no noise. Stepper motors are always noisy when they are rotating, because the driving force is a series of pulses, not a continuous torque.

 

[MRFMengineer]

I think the driver does matter. Most likely it is not producing a DC output when the motor is stopped. The first thing I would do would be put an oscilloscope across the motor input terminals. My guess is you will see some sort of pulse waveform, with no 3rd harmonic content.

You should be able to hold the motor in a fixed position with a DC current and no noise. Stepper motors are always noisy when they are rotating, because the driving force is a series of pulses, not a continuous torque.

From the Applied website:
•Dual MOSFET H-bridge, 3 state, pulse width modulated amplifier switching at 20-30 kHz

The PWM circuit is running at 20-30kHz, so the primary drive frequency and harmonics are what you are picking up. Maybe the output could be filtered to reduce this?

Any type of PWM drive will probably exhibit this characteristic regardless of type of motor. I use an Aerotech motion controller with a linear motor, the drive output has an adjustable filter presumably to reduce this effect. I cannot hear anything, but I have not measured with a mic. Maybe you could look into a linear drive for your stepper instead of a PWM drive?

 

[aalnaif]

MRFMengineer, AlephZero,

Thanks for your help. Both of your posts were enlightening.