AVR-Chat

A while ago I put together an 8 channel fan controller, for controlling
the speed (via PWM) of computer fans.

Works fine with fans, but last night I attached a couple of little DC
motors to it (for a little robot project I am working on with my son)
and to my surprise the AVR started resetting.

It would give a short burst, then reset (motor would run for a little
bit - maybe a quarter of a second).

Does anyone have any idea what could be causing it? As I said it works
fine with the fans. Output is via logic level MOSFET's.

Many thanks
Dean Claxton

Switching transients and back EMF noise.

Claxton, Dean J wrote:
> It would give a short burst, then reset (motor would run for a little
> bit - maybe a quarter of a second).
> 
> Does anyone have any idea what could be causing it? As I said it works
> fine with the fans. Output is via logic level MOSFET's.















.

You may need a stiffer power supply. Fans typ pull 100-200 ma, a stalled hobby motor
can be a lot more, easily 1-2 Amps.

Peter Gargano wrote:

> Switching transients and back EMF noise.
> 
> Claxton, Dean J wrote:
> 
>>It would give a short burst, then reset (motor would run for a little
>>bit - maybe a quarter of a second).
>>
>>Does anyone have any idea what could be causing it? As I said it works
>>fine with the fans. Output is via logic level MOSFET's.
>

On Tue, 24 May 2005 11:33:52 +1000 you wrote:

>A while ago I put together an 8 channel fan controller, for controlling
>the speed (via PWM) of computer fans.
>
>Works fine with fans, but last night I attached a couple of little DC
>motors to it (for a little robot project I am working on with my son)
>and to my surprise the AVR started resetting.
>
>It would give a short burst, then reset (motor would run for a little
>bit - maybe a quarter of a second).
>
>Does anyone have any idea what could be causing it? As I said it works
>fine with the fans. Output is via logic level MOSFET's.
>
>Many thanks
>Dean Claxton

I would try a 1000µF capacitor on the supply line. If that doesn't do it then
add a 10µF tantalum + 0.1µF close to the processor on the supply lines.

Such problems are 99% inadequate decoupling.


--
Ralph Hilton
http://www.ralphhilton.org
C-Meter: http://www.cmeter.org
FZAOINT http://www.fzaoint.net