Sunday, November 30, 2008



Zach kindly sent me a sample of his PCB version of the radio interference suppressor that I designed for the GM3 gearmotor.

I have tested it and am pleased to report that it works at least as well as my prototype, possibly better.

Here is the noise spectrum without it fitted.

And here is the result of fitting it: -

More details here:


Just a passing comment... I checked the ferrite core that you are using in this circuit and it's data sheet has "obsolete" stamped across it.
Just wondering why you haven't got a diode strapped across the motor terminals. That motor is a nice, big inductive load, so you have potential for a big voltage spike when it the motor shuts off. If there's a diode across the terminals on the driver board, then the wires that go to the driver are like an antenna. It's better to have a shottky diode right at the motor terminals.
The ferrite was Zach's choice, he reckons there are plenty in stock at Mouser.

My version had a diode but that prevents reversing the motor, which the RepRap motor driver can do.

The filter will reduce the speed of the back emf spike so having the clamping diode at the other end of the cable probably doesn't make much difference from an RFI point of view. It may well do without the filter.
Nophead, et al,

For a reversable motor and electronics, a TVS (transient voltage suppression) diode -- basically back-to-back zeners in a single package, might be a good addition. SA12CA is a popular family, available in leaded package, and appropriate for a +/- 12 volt system. This would crowbar any spikes above 13.3 volts, either polarity.

There are also lower-voltage versions, handy for protecting 5 volt or lower inputs from harm. Cheap insurance....

-- Larry Pfeffer


The TVS is essentially the same as what I was talking about. The point is that if you use a TVS, it makes the whole noise suppressor unnecessary.
>The point is that if you use a TVS, it makes the whole noise suppressor unnecessary.

No really unless you could mount them on the rotor, TVS diodes would just clamp the signal to 13V pk-pk. That is still a lot of RF energy.
Post a Comment

<< Home

This page is powered by Blogger. Isn't yours?

Subscribe to
Posts [Atom]