Microprocessor Stops

J

Thread Starter

John Catch

I have a situation where a number of Micromint Domino 1A (8052 based) CPUs just stop while otherwise operating perfectly normally. They start up correctly after being power cycled. Units that work fine in one area of the plant stop when placed in certain trouble areas of the plant. Conversely, units that keep stopping work fine in the good areas of the plant. We have tried simulating noise with flourescent lights, universal motors, and soleniods in parallel with the input power. We have also looked at grounding issues but find nothing remarkable. The units are not near noise makers such as VFD drives, or high horsepower motors. Any troubleshooting ideas are much appreciated.

Thanks,

John Catch
http://www.inflowinc.com
 
M

Michael Griffin

I would suspect a problem with grounding and isolation. There may be a ground current flowing through the device via the serial, analogue, or digital lines. I have said "ground current", but it may not necessarily be a current that travels to "ground". You may just have a current path going through the device in a way which causes one of the chips to latch up. These sorts of devices are often not isolated and can be subject to these problems.

This sort of problem can be very difficult to troubleshoot. You can sometimes find it by measuring the signal voltage with respect to ground and with respect to each other and comparing it to the specs. Pretty often though you just have to deduce the cause from the symptoms and the design.

Many people will isolate (not just "protect") everything in order to avoid these sorts of problems. Simply grounding everything can sometimes make the problem worse because you are providing more potential ground current paths without actually removing any of the voltage differences which drive the ground currents. Sometimes what you really want to do is to "float" everything. This can be hard to do in practice though.

The biggest sort of problems may arise when you are pulling non-isolated analogue signals from various distant sources into a board and then sending the results out over an RS-485 network. You are pretty much guaranteed to have "ground" differences between the analogue signals, the board power supply, and the RS-485 network causing undesired current flows.
 
C

Curt Wuollet

Try putting some high value low esr capacitors across the power terminals at the board. If that stops it, it was power noise. If not try shielding the wiring, then the board. After that, you would want to investigate on board causes.

Regards,

cww
 
Top