Failure of Pics.

we have some control electronics that employ pics in an industrial environment. The other day we found out something interesting, we installed a mini board with a pic, on a rotating item of a machine. made sure everything was secure etc. this unit does spin kind of fast. mostly made from Aluminum and employs slip rings to supply the hub with its power and signal return.

for what ever reason, after a few hours of spinning, the PIC gets corrupted software and has to be reloaded.

We are still looking at the reason why. We almost think the movement is causing some electrical field generation that is effecting the pic.

--
"I\'m never wrong, once i thought i was, but was mistaken"
Real Programmers Do things like this.
 Click to see the full signature
Reply to
Jamie
Loading thread data ...

Electrons have mass. You're spinning it fast enough that they centrifuge out of the flash ROM.

:^)

Tim

-- Deep Fryer: A very philosophical monk. Website @

formatting link

Reply to
Tim Williams

I believe it maybe gravity waves that's causing the problem.

Reply to
fjtenthy

what G force is it subjected to ?

Colin =^.^=

Reply to
colin

Probably a design flaw. Find a consultant (local, preferably) to efficiently solve your problem. There are probably other latent issues which perhaps can be found in time.

Best regards, Spehro Pefhany

--
"it\'s the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
 Click to see the full signature
Reply to
Spehro Pefhany

Flash-based PIC? Low-voltage programming still enabled?

Have you used flash-based memory microprocessors in similar slip-ring- supplied devices?

I suspect that a ground opened up and the noise and hash on the other lines got interpreted as programming commands.

Tim.

Reply to
Tim Shoppa

We're using a DC-DC (isolated) . Think we may have it down to a static build up due to the nature of the operation.

--
"I\'m never wrong, once i thought i was, but was mistaken"
Real Programmers Do things like this.
 Click to see the full signature
Reply to
Jamie

No it wasn't particularly. It just sounds like something that could be found in a relatively short time by someone with a lot of relevant hardware experience, and your guesses lead me to think that it would be more efficient to get help, and we're not going to be able to help much from a ng, most likely.

They are used in zillions of such applications. The problem is NOT in the microntroller chip... it's SOMEWHERE in YOUR application. Maybe mechanical, maybe a floating input, maybe passive component behavior, maybe EMI, maybe triboelectric effects, who knows? Maybe just the unit you are testing now is defective. There are even more possibilities that come to mind.

I would suggest getting to the bottom of whatever is causing it. What you don't understand generally comes back later to bite you on the a** at a most inconvenient time.

Best regards, Spehro Pefhany

--
"it\'s the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
 Click to see the full signature
Reply to
Spehro Pefhany

This unit works fine if the machine is operating at low speeds. High speeds seem to cause failure after a few hours of operation.

I don't know if your reply was a derogatory one in nature, if so, it would be nice to keep them to your self. This was just a general request incase some one had some experience using these devices in high vibration or G-Force environments. I may try out a small AVR Uc, I've used those also for a numerous of things.

--
"I\'m never wrong, once i thought i was, but was mistaken"
Real Programmers Do things like this.
 Click to see the full signature
Reply to
Jamie

You think too highly of yourself. You need help, or you wouldn't have asked. He gave you good avdice, then you shit on him. Typical.

--
Service to my country? Been there, Done that, and I've got my DD214 to
prove it.
 Click to see the full signature
Reply to
Michael A. Terrell

Two possibilities: rotating wiring in a magnetic field gets induced currents, have you tried soft-tinned-steel shielding? And, is there a quartz crystal in the circuit? Stresses on a piezoelectric crystal can change the oscillator characteristic, or vibrations can cause spurious output. An LC oscillator should be easy to substitute in.

There are also piezoresistive effects in stressed silicon, but that can only be fixed by redesign or mechanical isolation (potting?) of the whole thing.

Reply to
whit3rd

Well if it is an industrial environment and there are strong EM fields in the immediate vicinity, spinning the part at large velocity through both types of fields induces voltages and currents throughout the entire IC and every other node on your circuit board. Depending on the type of memory you have, this could certainly translate into accumulated time causing memory cell corruption. The answer would then be strict high attenuation EMI shielding, both magnetic and electric, a steel box, filters on all I/O.

Reply to
Fred Bloggs

Just curious: how many RPM and at what radius?

-- Joe

Reply to
J.A. Legris

^^^^^^^^^^^^^^^ Very Possible ^^^^^^^^^^ We have a fabricator on the machine as we speak repositioning the deck that houses the electronics. There is a DC drag clutch very close by. And after some testing, we found it to get a heavy drive of current at higher RPM's to maintain tension.

No quartz, using an RC osc based PIC.

I hope we don't have go that route. We'll see. Thanks.

--
"I\'m never wrong, once i thought i was, but was mistaken"
Real Programmers Do things like this.
 Click to see the full signature
Reply to
Jamie

could it be something in the software that overflows or similar after a number of revolutions or somthing else related to speed, that ends up trashing the flash? So it wont have the problem at low speeds until after a lot longer time or maybe never?

-Lasse

Reply to
langwadt

You should heed your own advice. You also shouldn't throw glass in a stone house.

--
Service to my country? Been there, Done that, and I\'ve got my DD214 to
prove it.
 Click to see the full signature
Reply to
Michael A. Terrell

10" radius@3500 RPM when it's screaming. and if you know manufacturing? its never fast enough!
--
"I\'m never wrong, once i thought i was, but was mistaken"
Real Programmers Do things like this.
 Click to see the full signature
Reply to
Jamie

Maybe you need a bleeder resistor across the galvanic isolation barrier.

Best regards, Spehro Pefhany

--
"it\'s the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
 Click to see the full signature
Reply to
Spehro Pefhany

centripetal acceleration = w^2 * r = (2pi * 3500/60)^2 sec^-2 * 0.25m = 33584 m-sec^-2 = 3427 g

Even if you solve the electrical problems, the mechanical ones will get you sooner or later.

-- Joe

Reply to
J.A. Legris

I meant the mechanical forces acting on the electronic parts. Have you successfully spun other circuits under these conditions?

-- Joe

Reply to
J.A. Legris

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.