ESD Damage

[Legion6789]

Just watched this amusing video from 1985 discussing ESD and the Apple II: http://www.youtube.com/watch?v=2WNb3TUNf0A

Is ESD still as big of an issue? I mean, this video is showing styrofoam coffee cups inducing enough charge to short an IC. Have ICs become more resilient or is this still just as big of an issue?

[Klaus2m5]

No, manufacturers put electronic components and circuit boards into antistatic bags just to impress their customers.

Seriously, internal ESD protection of integrated circuits has improved, but so has the danger to zap an IC with its higher density despite the protection.

[GARTHWILSON]

Many of the handling measures recommended on websites may seem extreme and even cost-prohibitive for a private individual to do his own construction. It is possible to handle ESD-sensitive parts without going to extreme lengths if you understand what does the damage, and constantly keep it in mind as you handle the parts. Dropping your guard can result in damage in an instant. For equipment, consider a grounded anti-static mat on the workbench to be a minimum, and keep skin (like a bare forearm) in contact with the mat at all times while handling static-sensitive parts. For minimal cost, an anti-static wrist strap (connected to something grounded of course) goes a long way, removing the requirement to always be in contact with the mat to discharge static. I use the anti-static mats but no wrist strap and certainly none of the other special things like the ionizer, special flooring, special clothing, etc., in fact my floor is carpeted; and just being careful, I have never damaged anything with static.

Where I worked in the mid-1980's in applications engineering at a VHF/UHF power-transistor manufacturer, we acquired a new line of MOSFETs from Siliconix. These had no protection diodes like the logic ICs do, so they were extra easy to damage when they were not yet in their circuits. We had a new production manager decide he was going to get really tough on employees, regarding the use of finger cots, anti-static lab coats, and the whole bit; yet the high loss rate did not improve. I pointed out that he was barking up the wrong tree. Employees in die attach, wirebond, DC test, RF test, fine-leak test, capping, etc. followed all the regulations, but then when they would take a set of parts from one station to the next, even though they weren't touching the parts themselves, the parts (which looked like little satellites, with their wide, flat leads sticking out on all sides) had leads touching the sides of the metal trays which got picked up by hand as if they were somehow no longer static-sensitive.

[BigEd]

If you have a 1% chance of killing a component, that would be near unnoticeable to a hobbyist, but could wreck a company's profit in the cost of returns and service.

[Legion6789]

I guess it surprised me a little as I routinely assemble PCs and never give a second thought to handling CPUs, video cards, etc without any kind of static protection. To date, I've never encountered an issue. I guess I should pick up an anti static mat and a wrist strap.

Does the wrist strap need to be grounded to true earth ground via an outlet's ground plug? Or is something like the metal table leg good enough?

[GARTHWILSON]

Legion6789 wrote:

I guess it surprised me a little as I routinely assemble PCs and never give a second thought to handling CPUs, video cards, etc without any kind of static protection. To date, I've never encountered an issue.

Once they're in circuits, they're less vulnerable. For one thing, if there are lots of things on a bus line, and each of them has protection diodes, there's more protection. Next, if you reach into a PC with the cover off, you'll usually touch the case first, even if unintentionally, bringing the PC's ground and your body to the same potential. When handling cards that are not in the PC, I always touch the metal strip at the back first, for the same reason.

Quote:

Does the wrist strap need to be grounded to true earth ground via an outlet's ground plug? Or is something like the metal table leg good enough?

There's usually a place to snap the strap onto the mat. Ideally the mat has a cord to ground (and yes, the ground prong of wall outlets normally qualifies); but even if it does not have such a cord, often it will get an equivalent grounding through equipment that's sitting on it, if the equipment has 3-prong power cords. In my case, I have two oscilloscopes with metal tilt-up bars that are sitting on the mats, and these go to the oscilloscopes' chasis, which get grounded through the power cords. If yours have plastic insulation between, well, beware. I also have a separate cord from the mats to an alligator clip which goes to the ground lug on the front of one of the 'scopes. The electrical in the room I'm in, built in 1949, does not have grounded outlets, so I puta wire to a ground rod pounded into the dirt just outside. We have another ground rod next to the garage (maybe 70 feet away) where the phone line comes in, and I was surprised to find there was well under ten ohms between the two rods, even though it had not rained for months when I measured! (A really low resistance is not necessary just to discharge static though.)

My main soldering iron does not have a ground prong, so I put a separate wire from its heater to an alligator clip which I clip onto the edge of one of the mats.

[Dr Jefyll]

GARTHWILSON wrote:

I use the anti-static mats but no wrist strap and certainly none of the other special things like the ionizer

Here's an issue not mentioned so far. Much more basic than an ionizer is just a simple domestic humidifier. Some homes have a humidifier attached to the furnace, but not all. And in some cases the humidifier has fallen into disrepair due to lack of periodic maintenance.

Dry air is the enemy of static-sensitive parts! As a superior insulator, it allows static charges to accumulate on your body and other objects, whereas static charges tend to harmlessly bleed away in normal humidity. If you notice signs such as a little shock touching the doorknob, or a faint crackling noise when you comb your hair or pet the cat, consider yourself warned!

Due to condensation, outdoor humidity levels drop during the winter months, and therefore indoor levels also plummet (except in the most modern, well-sealed buildings). A portable humidifier is one inexpensive solution. BTW: aside from electronics and ESD, arid air also has a direct bearing on comfort and health (especially respiratory complaints) and property (such as guitars, pianos and so on, which are subject to damage due to wood shrinkage).

-- Jeff

[GARTHWILSON]

Agreed. I tend to forget about that since we live in a mild climate near the ocean and the airflow is usually off the ocean, so we don't tend to get that dry or need much heating in the winter.

[BigDumbDinosaur]

Dr Jefyll wrote:

Much more basic than an ionizer is just a simple domestic humidifier.

I have small humidifier in my shop and another in my office, the former to keep ESD from decimating the inventory and the latter to keep me from drying up like an old fossil and blowing away.

My previous home had a whole-house humidifier that was a royal pain to maintain. It also contributed to heat exchanger corrosion in the furnace. When I bought this place I decided to put small humidifiers in places were they were of benefit. It has worked out well.

Speaking of inadvertent ESD, Commodore 64s and 128s were often victims, due to their joystick ports being exposed on the right side of the computer where the user could easily contact the pins. The flat 128 was especially prone to this because the reset button was right by one of the joystick ports. The ESD would clobber CIA number one, which among other things, was responsible for keyboard communication and in the C-64, generating the jiffy IRQ. If the discharge was sufficiently severe, the CIA would be toast and the computer would go completely belly-up. I fixed more than a few of those back in the day.

[bogax]

Years ago I had an old flourescent tube
mounted on the wall next to my bench
with a contact on one end and the other
end grounded.

Somewhat suprising how often I could make that
thing flash.

[BigDumbDinosaur]

bogax wrote:

Years ago I had an old flourescent tube mounted on the wall next to my bench with a contact on one end and the other end grounded.

Somewhat suprising how often I could make that thing flash.

Depending on the fluorescent lamp type (preheat, "instant" start, etc.) and its temperature, it will take anywhere from 400 to 1000+ volts to get it flash. So you have some idea just how much voltage you're generating as you shuffle across the carpeting in your home. Your average 65C02 isn't going to be amused when it gets hit by enough juice to make a fluorescent lamp flash.

[Simon]

When I was a student and working as an electronics tech (to pay for University) I remember someone telling us one of the issues with ESD is it might not necessarily kill a chip outright. It could damage or weaken a chip in a way that a failure might not show up for some time or it could make a chip intermittent. When I worked in a manufacturing factory they were very strict on ESD precautions.

Simon