I have a device using a PIC18F24K22 micro that has worked well for 7+ years at a few dozen locations. One location has reported glitching problems that scramble and lock up the micro. New units installed at this location exhibit the same behavior, so the problem seems to do with the environment where the device is used.

After asking a number of questions and giving the problem some thought, the problem seems to be static discharge. The problematic devices are in an area with a cold, dry climate. Most of the devices are used where it's more rainy and damp, where static discharge would be much less common.

The problem comes to troubleshooting a fix to the issue – how to repeatedly generate static bursts conveniently. The budget to do this is approximately zero.

I'm wondering if a spark ignitor from a gas BBQ might be an adequate simulation? Has anybody tried this? Other suggestions?

Thanks for your comments.

 

A long time back I needed to make a test fixture to perform an ESD discharge test per some spec. The model was a capacitor with a series resistor. Charge the cap to 1,500 VDC and let it discharge into the unit under test.

I was able to source that voltage by using a transistor tester, the kind with the cathode ray tube and a step generator. It could crank up that high.

Sorry I do not remember the cap nor resistor values, and have no memory of what was used as the switch.

 

Thanks for the info. That's overkill for what I need. My device glitches when a headset plug touches a jack box about a foot from the device. Since the glitch occurs as soon as the plug touches the jack (as opposed to being inserted into the jack), a static discharge near my device seems the likely cause.

 

The problem may be a faulty ground. If there's a spark it could be that the ground line isn't capturing that spark and pulling it to ground. As for generating a spark - I wouldn't recommend a piezo spark generator. Not only do they produce a high voltage they also produce a lot of current. Current that could damage your device.

An ESD (Electro Static Discharge) can be as high as 50,000 volts. And when you get hit with it - you definitely feel it. However, the amount of power associated is so low that it generally causes no harm to the human. Electronics are much more sensitive and a spark voltage as low as 50 volts can harm the most sensitive electronics. The likes of which you would only find on a space satellite or other space borne vehicle. Much of the common electronics world we live in have built in protection against common ESD levels. However, even though generally self protecting - they can still be damaged. Often that damage doesn't result in a full on failure, but rather weakens the circuit so that when under heavy use it fails because of a weakened pathway inside a microchip.

If your micro is scrambling when there's an ESD event - I'd look at grounding of the housing. A Faraday cage is a metal inclosure that prevents static from disturbing the internal electronics and is especially effective when properly grounded. If you're not completely enclosed then ESD can have an affect. Or a wire that is not shielded properly can also bring an ESD event inside the cage.

I wouldn't go about trying to generate ESD to test electronics - ESD is damaging if not deadly to electronics. You could be testing your device into permanent fatal failure. Look into grounding. But grounding is not the be-all, end-all of it either. Spike suppressors and surge suppressors may be needed. Especially in dry climates. I live in a very dry climate. Winter time I HATE filling the gas tank on the car. Get out of the car, touch the door and BAM! 50KVS (Kilo Volts Static) and you jump half way to the back of the vehicle. Just make sure you touch metal before you touch the gas tank or the nozzle. Fuel vapors have been ignited time and again by people who get out of the car, stick the pump in the tank, sit in the car while they fill the tank, get out of the car, touch the nozzle in a vapor rich environment - and wham - - - you have a fireball. It's not cell phones that cause this - it's Nylon Pantyhose or other man made fabrics and car seats with fabric covers - or worse, plastic covers. ESD is not the way to test an electronic circuit.

 

It's not cell phones that cause this - it's Nylon Pantyhose

That's why I always wear silk ones when I pump gas.