My apologies for jumping in late in the middle of the discussion. I must admit I have not yet read the rest of responses to this to see if any of my points have already been made.
Pete Base, as some one who has both a physics degree and a Masters degree in Electrical Engineering, I beg to disagree. Because of my physics background, I understand the the 'why' of ESD protection, and while it is the best policy, is not as blanket as you describe.
First, let use consider what happens when humans generate and discharge static electricity. As has been mentioned elsewhere, the amount of humidity in the air changes its conductivity, and depletes a static charge faster or slower in humid versus dry air. This phenomenon is well known to almost anyone who lives in an area where the indoor relative humidity changes throughout the year. Sliding or separating surfaces (pulling scotch tape of the dispenser generates a HUGE static potential!) build up charge. Whether that charge is deposited on the semi conductive skin of a person or a non-conductive plastic item, the electrons that accumulate charge the item like a capacitor, and generate a potentially very high voltage. This charge is then rapidly depleted when you touch and 'zap' something.
The next point to keep in mind is that there are two important measures of how damaging this zap can be, and they are the voltage built up, and the peak current. A piece of charged plastic can have a very high voltage, but because it is so insulating, little current flows when one part of it is discharged. On the other, human bodies are conductive release most of the charge very quickly, hence the painful zap. The true strength of the zap is the product of the current times the voltage, sometimes given in joules for sure protectors, and that is what makes some zaps more painful than others. And just as small capacitors take little charge to build to high voltages, big capacitors can soak a lot of charge before reaching the limiting voltage. And unfortunately, human bodies make fairly good capacitors as far as ESD is concerned.
Now we look at what happens to the device. The damage is mostly thermal, due to localized overheating of a specific spot or junction, and not the total voltage. However, the amount of power needed to burn a hole through a junction varies wildly with junction type. Modern CMOS devices have gates on insulators and so take very little charge to reach a high enough voltage to suddenly break down and pass all the current, causing a burn through. But power devices are designed to dissipate lots of heat and pass huge currents, and may be completely unaffected. Older devices like discrete diodes and transistors were so huge they took lots of charge to make a small voltage, and did not have the small gap dimensions that allowed reverse breakdown at small voltages, so they could take a walloping big zap and keep on working.
To put this in perspective, an unprotected, high density CMOS device, like a CPU or memory chip, may have one or more junctions destroyed by a zap so small you can't feel it, and from as much as 12 inches between your finger and the device. ANd you can also create latent damage that does not cause the chip to fail until some time later. It is for this reason that all electronics companies have massive ESD training programs (I have been through many annual recerts myself). But when it comes to microcontroller chips, with +/- 20 milliamp drivers on almost every pin, and reverse voltage and clamping diodes on EVERY pin, that are safe for handling from any but the most powerful, and painful ESD discharges.
So statement "There is nothing to be debated here, you must
use ESD protection with any semiconductors or other ESD sensitive
components - end of discussion." is the official position of all electronic companies for their assembly lines, but is NOT universally true in ALL situations. Even modern chips can be designed to be ESD resistant to the point that no special handling is required.
However, you will never know if it truly WAS required until after you destroy something, so stick with things you are confident are self-protecting and cheap to replace (;!
Mike Bushroe
--- In Homebrew_PCBs@yahoogroups.com, Benjamin Blumer <b3nzilla@...> wrote:
>
> It is compelling to have someone with credentials stand on one side of the
> fence. But I think it would be much more compelling, and more satisfying,
> to have said person explain their reasoning.
>
> Sent from my iPad
>
> On 2011-12-25, at 7:01 AM, Pete B <pete.basel@...> wrote:
>
>
>
> I am a degreed electrical engineer who has done both analog and digitial
> design for over 30 years. There is nothing to be debated here, you must
> use ESD protection with any semiconductors or other ESD sensitive
> components - end of discussion.
>
> http://www.linkedin.com/in/petebasel
>