The Adventure of the Grassy Curve

DN Staff

August 17, 2009

3 Min Read
The Adventure of the Grassy Curve

By Jerry Lowe, Contributing Writer

It’s a case of mistaken identity in this case involving a noisy motor.

A few years back I was involved in the design and set-up of a new, computerized coil-test machine with two stations. The spec called for testing a sensor coil within about 15 seconds. The first station measured the direct current resistance (DCR) and inductance of the coil. If there were two coils wound on the same bobbin, it measured the capacity between the coils. We built an electronic circuit to produce a sine wave at the different frequencies selected by the data acquisition board in the computer.

The second station had a target wheel powered by a three-phase motor, which was controlled by a variable-speed controller that received data from the data acquisition board. The coil under test was brought to a specific location between the target wheel and the coil. The wheel was then set to the required speed (up to several hundred rpm depending on the coil model number) and the A/D data collected. The computer then analyzed and plotted the data on a curve in a picture box (we wrote the program in VB6).

Everything started going okay, until suddenly we started getting grass growing on the curve. An intermittent condition, it appeared only about half the time on a given coil. It also changed the test results, so we knew the noise was coming in from the coil.

We did all the usual things we could think of to eliminate the interference, including the bypass caps, to no avail. Then it dawned on me that we were practically sitting in the glide path of an international airport. We had an engineer stand in the door and watch for planes coming and going, trying to connect the activity to the occurrence of our noise. But no luck. We also turned off equipment and lighting in the shop, but the noise was still there.

After about two weeks of chasing this problem, I woke up in the middle of the night with a Eureka! moment.  As soon as the machine designer came in I had him on the phone. This machine had been very well designed, right down to the $500.00 coupler between the motor and the target wheel, even though the load on the motor was quite light.

I asked him to replace the coupler with a rubber insert coupling, and I would be there by the time they made the change. When I arrived the original coupling had been replaced with a piece of rubber hose. I thought that looked a bit crude, but I gave it a try and the grass was gone and never came back. The rubber hose in still in place and the machine is still running as far as I know.

And the reason the rubber hose fixed the problem? It was not a noise problem at all, it was the cogging of the target wheel motor at low speeds. This cogging takes place due to the low frequency applied to the motor, causing it to jerk from one pole to the next instead of turning smoothly. (If the motor had been larger in diameter it may have helped to manage the inertia build up.) The rubber coupler just made a soft coupling that absorbed the jerk.

Contributing Writer jerry Lowe started Tri-L Electronics at age 26 in the basement of his home. Throughout his long career (50+ years) he has built several VHF and UHF radios, broadcast audio to mag strip film amplifiers and other analog and digital devices. He has designed circuit, built equipment and wrote software for machine control and converted his 1971 Pinto to an EV in 1975.  In an unexpected career move, he also once worked as a police officer.

Sign up for the Design News Daily newsletter.

You May Also Like