Any ESD concerns w/ Acrylic boxes/boards?

[jim_s]

Just curious if there are (or should be?) any ESD concerns w/ the acrylic enclosures and dev board/plate/sled types of things? Will these build up a static charge that could endanger the mounted device? (BBB, RPi, etc) Any recommended preventative measures, aside from the regular ESD precautions? (grounding before touching, wrist strap, etc).

Thanks!

[adafruit_support_mike]

There shouldn't be any problems with a case, for a couple of reasons:

First, acrylic is mostly a 'triboelectric' material: you can generate a static charge on it by rubbing it against something else, especially when the humidity is low. It doesn't spontaneously generate charge though. There has to be some kind of mechanical action moving the electrons around. There isn't likely to be much rubbing inside a case.

Second, the nature of electrostatic repulsion is such that any charge on a hollow object will appear at the outer surface of the object. Everyone knows that like charges repel, but most people don't understand how strongly. If you took all the free electrons in a chunk of copper 0.3mm to a side (about 1 Coulomb) and removed all the protons, the resulting force would be strong enough to make a Nimitz-class aircraft carrier (about 100,000 tons) go from zero to 250 MPH in one second.

Stray electrons want to be as far from each other as they possibly can. They can get farther away on the outer surface of an object than on the inner surface, so that's where they go.

Gauss's Law says that the electrical field inside a hollow object with a charge on its surface is zero. Basically, if there was an imbalance capable of creating a field inside the object, the electrons on the outer surface wouldn't be as far away from each other as they could. Left to themselves, they'll redistribute to get rid of the imbalance.

It's always a good idea to touch a grounded connection before playing with anything that has high impedance during the low-humidity part of the year, but the interaction between your shoes and the carpet presents more ESD risk than a case.

[jim_s]

Thanks for the reply and the info, Mike. How about the plates/sleds? Just curious if I should ground a plate/sled w/ something akin to a wrist strap, or if having the device bolted (via metal bolts that contact the mounting points of the board, which at least in the case of the latest BBB, I think are part of the board's ground plane, and then bolt into the plate/sled) is enough or what. (Perhaps this is as much a general thoughts discussion as much as a real-world-advice discussion - I appreciate the input!)

[adafruit_support_mike]

There still isn't much to worry about.

The average static charge has a density of a few microcoulombs per square meter, and a field strength around 15kV. The mounting plates we sell have a total surface are of about 0.03 square meters, so the accumulated charge would be in the neighborhood of 50 nanocoulombs.

Most antistatic films and mats have a resistance in the high megohm to gigohm range under normal circumstances, but it drops to the mid-megohms when you apply a few kilovolts to it. That still sounds pretty close to "nonconductive", but the numbers still crunch properly. Applying 15kV across a 1G resistor produces 15 microamps of current. That's enough to dissipate our estimated charge in about about 30 milliseconds. If we assume the resistance drops to around 50M, the charge will dissipate in about a millisecond.

That reasoning holds for dissipating an existing charge, but preventing charge formation takes even less current. Triboelectric charge generation is phenomenally slow by electrical standards. Deliberately scuffing your feet on the carpet to acquire a charge takes about 10 seconds, but if you were touching a gigohm resistor connected to GND, the charge would dissipate faster than you could build it. Ambient static charges accumulate even more slowly, so it takes even less current to keep them from accumulating a dangerous charge.

It takes special techniques to maintain a resistance more than about 20M. At that point, the fingerprints you leave on the surface of a resistor when touching it are enough to 'short' it. The surface film accumulated through normal handling is enough to keep dangerous charges from accumulating on the surface of a mounting plate, and yes, mounting screws that connect to the PCB's ground plane will shunt any accumulated charge to ground.

Again, the biggest risk is from a previously isolated body coming close enough to the device for a spark to jump (usually 3-5mm). Antistatic mats give you a large surface that can absorb charge when you first approach the bench, and will dissipate any charge you develop while you're working at the bench. Grounded wrist straps and rings have much lower resistance and are mostly used to swamp out the 60Hz noise the human body picks up as an antenna near mains wiring or flourescent lights.. I've been working on some >10M circuitry recently, and can see the signals change on the scope if I wave an ungrounded hand above the PCB.

If you're concerned about ESD, touch a metal component connected to GND when you first approach a device. If you want to be fancy about it, solder one lead of a resistor to a fender washer.. any resistor value will do, though ESD specs say it should be 1M to prevent injury if the device's GND plane gets connected to a dangerous voltage. Then connect the resistor's other lead to some grounded point on the device.. wrap it around one of the mounting screws, for instance. Touch the washer first any time you approach the device, and you'll drain any charge you may have been carrying.

[jim_s]

Awesome info, Mike - a true mini education on ESD - THANKS! (I hope that others find this and read it, and that they get as much from it as I have!!