As a future net admin i have had the repair classes, and do it all the time in my free time on 1 of my 8 systems. i have always hered about ESD and use a strap or toch the case to prevent it. i can sort of see how it could fry a port but curos how this actaly happens. is it just from the 1KV+ with little amps or is it just because the sudden power to a single part. i can see 1kv frying something but static has so very little amperage i think it is odd it could do something like this. would like to hear any explination, stories about esd.

 

It's a capacitive discharge, inside the integrated circuits. The many wires of the circuit are like little capacitor plates next to each other, with a very wimpy dieletric insulation layer between them that cannot stand up to much voltage differential.

The high voltage low amperage spark arcs across the gap between adjacent circuit traces, and can literally burn a hole right through insulating layers that either cause a short or open. Boom, the chip is dead.

But this spark and insulation damage is not always so strong as to absolutely kill the chip in one shot. It can arc across the gap and partially weaken the insulation. So sometimes the chip works fine, sometimes it doesn't.


You don't necessarily even need to see a visible spark for damage to occur. Insulators can carry charges on them, and doing things as minor as setting a styrofoam coffee cup next to some bare chips on a table can blow them up internally with no obvious sparking.

Separating charged plates tends to cause the stored voltage on the plate to rise, so just picking up a static charged object can cause a rise in voltage that arcs and blows the chips nearby.


ESD protective foam is a conductive material that pinned chips are inserted into. It basically grounds all the chip pins relative to each other so any charge exposure is rapidly balanced across all pins, and damage is prevented.

ESD protection bags work the same way, distributing charges evenly around a protected device so that there is no concentrated voltage in any one spot that can arc and cause damage.

 

Simply walking across the floor in socks can generate in excess of 15kV!

 

The current spikes can be in the order of nanoseconds, and they can be many amps. The duration is short, but the intensity is killer.

ElectroStatic Discharge

This is the article I wrote for the AAC book.

 

A good read, Bill.

 

In all of my long career with electronics manufacturers I have only seen one single case where ESD was the apparent villain. That particular circuit had extremely high first test failures in a product that had but a single CMOS IC, and replacing the IC would fix the problem... but with a similar failure rate.

This was a hybrid module, meaning the IC was in bare chip form and each "pin" was wired individually using gold wires. Looking it over the only idea I had was to add a note for the person doing the actual wire bonding to connect VSS and VDD pins first, then the other pins. This way the protection diodes were in place before any sensitive input pins were subjected to ESD from the wire bonding.

Just this one little change pushed our yields back into the high 90's percentile.

The basic lesson is "loose" parts are very ESD susceptible, but parts already in circuit are fairly well protected.

That said, I always strap myself in when touching any shippable product. Safe is sure.

 

ESD should be a sticky. This one keeps coming up.

If ESD caused outright failure that would be just wonderful.

It is the microscopic damage to the semiconductor structures that is the more serious problem.

The unit may pass all tests in the factory but it is when it fails in the field that the replacement costs escalates.

ESD protection must be taken seriously.

 

I kind of understood why but never actaly knew the pricipals thanks guys. als conformned this my self today. had an old junky dell that was rediculisly slow i was given for parts or what ever. pulled what i could use then zaped the mobo with my mini plasma globe. looks like it fryed the cmos IC acts odd on power pluged in. puled the few caps off it and then recyle the rest of it. i think mr chips may be righ about the sticky. i don't ever recall seeing a ESD thread so i made one. also may be nice to know why it is so important to avoid it. (i like to keep my ICs in a bin in a divided box should i ues an esd bag or stick them in tin foil so the all ground or something or is that ok. i oly have a few mostly what look like 555 timers.) should i be conserned about transistors also since there also simiconductor?

 

Yes. I had some bad experiences with the old reliable 2N2222A's back in the Rockwell Int days. I would leave the boards running as long as possible to check the new components (replacing the failures). Some would die. I didn't know why back then, but it was ESD (and total lack of protections).

Since then (before they closed) they had implimented the standards. It works, never had the problem since. I never had the same problem at home, I suspect the stock guy had shoes that generated lots of ESD.

 

In a professional service station. The ESD bracelet is longe gone. The floor, chair and workbench. Some times even the ceiling are made of weak conductive materials. And tied to the same potential. The workers use shoes with conductive soles

 

In a professional service station. The ESD bracelet is longe gone. The floor, chair and workbench. Some times even the ceiling are made of weak conductive materials. And tied to the same potential. The workers use shoes with conductive soles

I'll have to disagree with that. I work for a modern military contractor, and while conductive shoes, floor, smock, table, chairs all help, nothing replaces the wrist strap. Nothing. Nowdays wrist straps have active testers built in so the operator/tech knows if a strap failure has occurred real time.

All the rest contributes to reduction of static in total, but people wiggle and move constantly, and in so wiggling generate fresh static that has to be bled off as quickly as possible.

Really high speed components can be zapped with as little as 10V. In both shops I worked in since the 90's I had parts that went from DC to way past 100Ghz, and are extremely (emphasis) sensitive.

 

I figured out how to un-static my plastic drawers by spraying them with a graphite spray designed for locks. After it dries, you wash it with detergent to remove the oils and it measures less than a meg per inch.

 

I agree with Bill.

I recently did some work experience at a manufacturer near where I live, and they still you wrist straps with the features Bill mentions.

 

i normaly just use the techniqe of always touch the case of the divice, i have a strap but usaly don't feel like geting it out of the tool box. actaly got it at a yard sale for nothing because they had no idea what it is. nothing fancy but works also has built in resistor to the clip lead.

 

For home use that is what I do when working on a computer. I do make the distinction between what I do for home and what I do at work. It is lazy, but the stuff I generally work on is not that expensive, nor is anyones life on the line if it fails, unlike all the stuff I do at my job.

If you are working on something that is uber expensive I would go the extra effort. If you are implementing something at work I would do the basics, if you are in charge you set the standard. The tech is not hard nor is it expensive. It can be a bit inconvenient, but not as much as you think.

 

I was looking at the NEETS Module 21, Test Methods and Practices and found this static charge info ...

 

i have worked for 2 electronics companies and in both places the same problem persisted, we would wear straps but if we would have a problem the engineer would come to the bench and pick up boards and not even THINK about a strap, one time i called an engineer out on it and he said that in his ESD class he was told that if he handled boards by the edges he was safe lol

 

if he handled boards by the edges he was safe lol

in one of my computer classes the book sort of said something similar. they discused esd and straps, pads, etc. but then they say hold bords by the edge. witch i can see to avoidpoping a chip or breaking something. maybe he hered somthing similar and misunderstod. grabing by the edge i would dout would prevent esd isues, alot of boards have traces right up to the edge and poibly internal ones. 1KV + would posibl;y just arc to it and its toasted the board.

 

Hello,

On the topic of ESD there is a lot written.
Take a look at the links of this page of the EDUCYPEDIA:
http://educypedia.karadimov.info/electronics/esd.htm

Bertus

 

Alot of components these days are very delicate compared to those used in say radio's of the 1950's and are easily damaged by smaller voltages. If picking up a board by the edge it's highly likely that at some point during the inspection of the board that accidental contact with a component will occure, also PCB tracks will often run real close to the edge of boards at times and a charge could easily be released into the board and destroy a delicate gate input somewhere, if not immediately noticeable, it may eventually fail in the field and will cost money for a field engineer to go out and replace it. my point is that for a few seconds that it takes to strap up, isn't it worth it?

Neil.