url: http://escribe.com/health/thesilverlist/m60257.html Re: CS>$$$ perpectives From: Mike Monett Date: Mon, 16 Jun 2003 06:48:22
I wrote: CMOS is very sensitive to esd. TTL draws more power, but is pretty much immune. The headaches really scramble my thinking. This statement is true, but useless in this context. CMOS can be damaged or destroed by ESD events. There is protection at the inputs, but it is only good enough to pass the industry test so the product can be shipped. It will not survive the strong esd events that can occur in ordinary use. Another problem is an esd event can cause the device to go into latchup and be destroyed if the power supply can deliver enough current. Usually several hundred mA will do it. This is what I was thinking of when I said TTL is immune. TTL cannot go into latchup, but it is most certainly affected by esd and rfi. However, the comparator inputs are not TTL. I don't know which device you guys are using - very few that can withstand 30 volts differential across the inputs. If the device is CMOS, I'd make sure there was very good protection at the inputs, especially on systems that have long leads. The same filter techique to protect against rfi works for esd. But you really have to check it carefully. ESD risetimes can be sub-nanosecond with currents in the hundreds or thousands of amps. The problem this causes is the voltage can arc across the series resistor at the input. If the bypass cap inductance is high, or the traces are too long, this can allow very high voltage to appear at the input to the device. This can damage the thin oxide or create latent defects that fail months later. You will never know why. It won't occur often enough to seem to be a serious problem. But each time it happens, someone will lose their unit until it is repaired. However, rfi or esd protection can ultimately be defeated. No matter how much you install, there is always some place where it won't work. Walking across the carpet in Colorado during the winter can draw 1 inch arcs. This is over 30kV. That really takes good filtering to minimize the effects. You will never eliminate them completely. If you are in an industrial area, being next door to an arc welding outfit is not a good idea. If there is a high powered transmitter nearby, it can also cause havoc with sensitive comparators. I once found a huge problem when a laser company moved in next door. They used 10 KW heaters to melt the glass. The temperature was controlled with a simple bimetallic thermostat. When the contacts opened, they produced a small arc. This resonated with nearby wiring and pruduced a huge spike at 40 MHz. I meaured the frequency by triggering a scope and seeing the rf pulse in my lab. This severely disrupted the product I was working on. The cure was simple in this case. A snubber across each contact provided enough damping to kill the arc. The snubber was a 47 ohm resistor in series with 0.1uF right at the contacts. But it took a while to figure out where the noise was coming from. And there will always be the next one... Best Regards, Mike Monett -- The silver-list is a moderated forum for discussion of colloidal silver. Instructions for unsubscribing may be found at: http://silverlist.org To post, address your message to: [email protected] Silver-list archive: http://escribe.com/health/thesilverlist/index.html List maintainer: Mike Devour <[email protected]>
