Handling CMOS chips

[owhite]

Hey,

Long time ago, when I was getting chips from the local EE department to build circuits, handling CMOS chips was a real pain because of static.

Has the situation improved at all?

I'm building a timer, and was thinking of using this device: CD4536B

but I dont want to bother if they're still as sensitive as they used to be. Any comments?

Owen

[Al_The_Man]

Owen, Personally I have not had a problem, I believe that alot now have some kind of input protection built in, but it is wise to be carefull when handling any solid state device, especially if you do not have the recommended grounded work station, but usually once they are in place in a properly designed circuit the chances of damage are minimised. If you have some of the pink anti-static styrofoam the IC's sometimes come in, keep the chip plugged into it until you insert it into the board.
Al

[DR-Motion]

Hi Owen,

Even with built-in diodes protecting the inputs on modern CMOS... they are STILL STATIC SENSITIVE.

I live in Ontario, Canada and work on my electronics bench during the dry winter. Winter is the worst time for static electricity to build up. Just try rubbing your feet accross the carpet then touching a door knob... you'll be shocked by the spark.

But if you discharge your body's static charge each and every time before touching your CMOS ( or any other, for that matter) IC's and store them on a conductive sheet of sheet metal while working , you will not cause any damage.

The best investment I ever made was to buy an anti-static wrist-strap and use it every time I work on a circuit, but even before I got the strap carefully grounding myself to a large mass of metal ( eg my milling machine ) or the earth ground on my power supply ( not the negative terminal ) prevented any damage.

Basically it all in the mindset... be aware that static exists and remember to get rid of it.

Good luck on your project.

regards Gary

[Mariss Freimanis]

CMOS devices are almost completely static immune. We have gone thru over 500,000 devices and I have yet to see one damaged by static electricity.

Mariss

[DR-Motion]

Mariss, I understand that you are involved in electronics manufacture. In a previous life as a Test Engineer for a high volume automotive electronics manufacturer, I was impressed by the amount of capital spent on preventing static damage to not only CMOS but even plain old carbon film resistors (all of which can be damaged by relatively low electrostatic potentials )

The tiles on the floor where the SMD auto-insertion machines had a conductive carbon embedded in them, a special conductive adhesive held them to a layer of copper foil underneath. The electrician took special pains to sink a dedicated ground rod through the building's foundation.

All of the factory's personel were educated about the dangers of static damage to all components, static dissipating wrist straps were tested every shift and the operators signed a log. Footstraps were worn in the auto-insertion area and the carts used to transport sub-assemblies in their conductive plastic trays (always black in colour) had chains to disipate any static produced by the cart's wheels.

In short, if industry has seen fit to invest so much to prevent static damage, chances are that it is probably a valid concern.

Static damage is not apparent imediately as it involves the piecing of the metalization layers(in the case of IC's) and the parts remain functional through in-house testing. The problem usually occurs part way through the lifespan of the product which produces the worst possible result... customer dissatisfaction.

A quick search of Google gave this site:

http://www.static-sol.com/61340-5-1/...ensitivity.htm

Please excuse the long rant but your post suggests that static is a none issue and because of your strong reputation in this community, this might result in a misunderstanding.

Sincerly Gary

[ESjaavik]

"Please excuse the long rant but your post suggests that static is a none issue and because of your strong reputation in this community, this might result in a misunderstanding.
"

And in a strong misbelief in the products made by your company.

I have worked with industrial electronics for quite some years, and seen the number of repairs coming back declining. Those customers (now quite few) who does not take this seriously still return products. Often after having run for weeks even months before they failed.

I agree with DR-Motion on all points.

[ToyMaker]

The US Navy did a study of electro-static discharge (ESD) that found that 1/3 of ALL electronic failures were the direct result of ESD or could be traced to ESD.

robotic regards,

Tom
= = = = =
There can be no such thing as a necessary evil. For if a thing is really necessary, it cannot be an evil. And if it is an evil, it is not necessary.
- - Tiorio

[owhite]

well hey, I appreciate the comments people. Recall the original question -- I was wondering if the situation with cmos chips has improved. In the 90s I lost many chips that were already tied to ground in the prototype board to static. I'm just hoping that its better'n it used to be.

[ESjaavik]

If "long time ago" was in 199x, then the situation is not much changed. It has from 1980 or so. At around that time the CMOS components got input circuits with static protection. Which made them much better.

If you have lost many, read up on EMC handling. Just knowing what the problem is, and doing as best you can with simple means should bring your loss due to static down below the 1% failure rate. In industry that is not enough for reasons pointed out by DR-motion. Grounding the chips is not the most important, you are yourself one of the largest dangers to the chips. Wear non-syntetic clothing, avoid the kind of environment that promote generation of static. Discharge yourself against ground as often as possible or even better wear a wrist strap with a 1M or so resistor in series to ground. (The resistor avoids electrocutions). Keep components in antistatic packing until you install them. Don't use office chairs with rollers on a carpet. Use a grounded (=modern) soldering iron. In other words try to emulate a modern production facility as far as you can with the means and knowledge you have. For a hobby project there is no need for an ISO9001 approval.

My main point is go on make something, but don't overlook simple precautions because "components are immune" they are not.

[Mariss Freimanis]

Well, to help ameliorate your "strong misbeliefs" I was addressing the original question, not how one would run a production line.

The original question was regarding breadboarding and whether special precautions have to be taken manually handling 4000B series parts. My experience is no precautions are necessary. I have built more than my share of breadboards yet I cannot recall a single instance where a 4000B series part was static damaged by handling.

Unlike the original, much more static-sensitive 4000A series, the "B" series inputs are resistor-diode clamped to Vdd and Vss, the device zeners on Vdd overvoltage and the outputs are virtually SCR latch-up tolerant.

I do not market breadboarded circuits, they are for my amusement and edification only. If my experience was one failed to work because of electrostatic discharge, I'd say so. They are time-consuming to build after all, precautions would then be sensible.

In production, CMOS parts pour from anti-static tubes directly into linear feeders or are picked directly from 16mm anti-static reels by the pick-and-place machines. No handling, no exposure.

Once on-board, the parts are completely safe. In-circuit impedences, circuit design, zener clamping and distributed Vdd to Vss bypass capacitances insures no applied voltage can exceed safe levels. Circuit boards that can expose onboard devices to damage by static discharge via input/output connections are in my humble opinion, poor designs. Do not microwave the drives though; don't ask me how I know that's a bad thing to do :-)

By the way, what mystifies and amuses me is why 0.05 Ohm, 3W resistors come in anti-static packaging.

Mariss

[arvidb]

Originally posted by owhite *snip* In the 90s I lost many chips that were already tied to ground in the prototype board to static. *snip*

Actually, a chip that's tied to ground would be more succeptible to ESD damage than an unconnected chip (unless *all* of its pins were tied to ground, which seems unlikely).

Remember, it's not the voltage that kills (neither the chips nor you! ), but the current! So if you discharge yourself through a chip partly grounded, the current have a low-resistance path to ground = quick discharge with high peak current = dead or damaged chip.

The 1M ohm resistors in wriststraps not only protects you from electrocution, but also the chips. They make sure the current can be safely (and slowly) dissipated.

This is why ESD equipment is never made out of metal, but instead out of conductive or dissipative materials that's got a higher resistance, while still being conductive.

So don't use a directly grounded metal plate as a workplace and think it will save your devices from ESD - it's more likely to just kill you if you do something wrong.

Arvid

[arvidb]

Originally posted by Mariss Freimani *snip* Do not microwave the drives though; don't ask me how I know that's a bad thing to do :-) *snip*

Hey, how do you know this is a bad thing to do?

By the way, what mystifies and amuses me is why 0.05 Ohm, 3W resistors come in anti-static packaging.

Yes, I have wondered the same thing. Maybe because it's easier for the supplier to just treat all components the same? Or is there really a good reason for this?

Arvid