A sewing machine motor speed control - The sound of a functional snubber
The answer to a non-Zen non-koan.
A sewing machine motor speed control - Table of Contents
In Zen Buddhism, there is a tradition of using paradoxical questions (koan) to bring insight to practitioners. A famous koan that even non-Zen Buddhists have heard of is the question “What is the sound of one hand clapping?”
I’m not philosophical enough to attempt to answer that question.
I am practical enough, however, to answer a non-Zen, non-koan question along similar lines.
- Question: What is the sound of a functioning snubber?
- Answer: The sound of a running motor.
A functional snubber in a motor control makes no sound in and of itself. It merely prevents the “kablooey” that occurs if it is not there or not functional.
Bigfoot is operational again, with an upgraded power supply (higher rated parts for the rectifiers) and a snubber to prevent short peaks from killing the flyback diode across the motor.
| Snubber |
|---|
 |
 |
I installed that snubber across the flyback diode in Bigfoot. It should prevent fast transients from the motor commutator and the PWM from killing the flyback diode (and then the PWM MOSFET and the bridge diode.)
Now that things are running again, I finally got to compare the electromagnetic interference (EMI) generated by Bigfoot with the EMI generated by a regular sewing machine motor speed control.
Lacking a radio frequency spectrum analyser and a lab to use it in (as well as the training to use it,) I simply tuned an AM radio between stations to hear what kind of RF “crap” Bigfoot is spewing.
| Bigfoot EMI |
|---|
| Bigfoot EMI |
That sounds lousy. It is loud and certainly nothing that you really want to hear on your radio.
For comparison, I used a standard electronic motor control on a Kayser vibrating shuttle sewing machine that I’m working on.
| Standard electronic control EMI |
|---|
| Standard electronic control EMI |
That’s not any better. If anything, it is worse than Bigfoot.
Bigfoot may not be an improvement (EMI wise) over the standard controller, but at least it isn’t really any worse.
If you listen closely, there’s another difference between the two besides the volume.
While making both recordings, I put my thumb on the handwheel to vary the load on the motor.
In the Bigfoot recording, you can hear the noise getting a little louder when the load goes up. The frequency itself doesn’t change - the Bigfoot PID control holds the motor speed constant despite the changing load.
The standard motor control doesn’t do that. Besides getting louder when the load goes up, the motor also slows down under load. You can hear the motor speeding up and slowing down as I vary the load. That’s the basic difference between Bigfoot and a standard control - Bigfoot maintains the speed set by the foot pedal while a standard control maintains the power to the motor while the speed wanders up and down.
I’m going to make a new circuit board for Bigfoot (someday.) The new board will allow me to properly mount the snubber as well as an NTC resistor I added to reduce the inrush current to the 47µF capacitor in the DC power supply. It’ll also have larger through holes for the UF5408 diode I’ve switched to for the flyback diode.
I’ve also got some small software improvements to make, but that’ll be another day. Right now I want to use Bigfoot. This snubber mess set me back more than a month on things I wanted to sew.