Hello,

We have trouble with our microprocessor board, we are using a 5VCC as
power supply in our application and we are dealing with AC motors.
The microprocessor reset itself everytime the motor turn on, so we used
a filter into the motor, the problem seems solved but sometimes the
micro still reset itself.
How can i filter my 5VCC ? we saw some disturb on 5 VCC when the motor
turn on/off....anyone got some suggestion ?


Thank you in advance

 

Unfortunately these sorts of problem are not always simply fixed by adding a
few capacitors.

You don't say anything about speed control etc so I may assume you are simply switching the motor on and off, probably using a contactor or relay?

Each time these contacts close or open a burst of radio frequency interference will be generated. Unless proper shielding and filtering are used this burst of RF will be picked up on your processor board, not just on the power supply but on all the signal traces and even within each device on the board.

Clearly if you start getting substantial amounts of noise on the address, data or control lines of your processor it will behave erratically and most likely the watchdog on your board will kick in to reset the processor.

The issues involved in rectifying problems such as this are many and varied
and I'm not going to cover them all here.

You need to keep the processor electronics electrically shielded from any noise sources. This may mean putting the in a seperate enclosure or mean putting them in an enclosure within the main enclosure.

You need to make sure that any connections into the processor enclosure are
not simply going to act as aerials, picking up the noise and conducting into
the processor enclosure where it can re-radiate and cause as many problems
as if the processor were left outside the shield.

You can achieve this by using filtered feedthrough connectors for cabling coming into the processor enclosure, by segmenting the design into "noisy"
and "quiet" sections, by using screened cable for certain signals that go outside the processor enclosure and then being very careful about how you connect the screens to these cables.

You may need to think about having seperate "clean" side and "dirty" side power supplies and you might even need to think about galvanically isolating the control lines to the outside world so that you aren't "exposing" your delicate circuitry to dirty return currents from the outside world.

You need to bear in mind that EVERY conductor has the potential to be an
aerial which could pick up AND radiate energy.

This is a really big subject.

I'd suggest that if you are in the situation that you are in then you need
to employ some expert help to review your design for electromagnetic
compatibility. DON'T just stick a few capacitors around the circuit. You may
believe you've fixed the problem then have your system fail in the field
because the cable layout is different or something equally probable!!

Good luck

Geoff Moore
====================================
Straight Forward Solutions Ltd
Maynooth Road, Prosperous,
Naas, Co.Kildare, Ireland
Phone : +353 (0)45 892739
Fax : +353 (0)45 893880
Mobile : +353 (0)86 8179683
email : [email protected]
====================================

 

Sprinkle bypass caps around on the board and put a ferrite bead on the imput 5VDC lead. make sure the grounding is correct.

Regards

cww

 

Does it have a proper ground plane?
Or even better: have one layer for ground and oen layer for 5V
if possible, the capacitor you create with this have quite good
performance then it comes to RF generated disturbances.

This does not mean the suggestions made by others should not be
looked at first, a couple of godo capacitors and inductors at
the proper places can remove most of the disturbances, but only
those entering thru the external wires, not the ones entering
the board as radio waves. A proper ground plane can however take
care of quite a lot of that (but, of course, not all)

A properly shielded box does of course help too.


/Johan Bengtsson

----------------------------------------
P&L, Innovation in training
Box 252, S-281 23 H{ssleholm SWEDEN
Tel: +46 451 49 460, Fax: +46 451 89 833
E-mail: [email protected]
Internet: http://www.pol.se/
----------------------------------------