At 07:00 AM 2/25/02 +0200, you wrote: >I know this should be OK in theory as it is in the manufacturer's specs, but >in practice I would never design a curcuit for mass production like that, or >do anythink that could take any pin outside the supply voltage range, even >using schottkey diodes to clamp the pins could be dangerous. There is a >parasitic scr in every chip waiting to be fired! Junction-isolated CMOS chips (which is most of them), yes. If you guarantee the current will never get to the trigger current (which is tested) then you'll be safe on that account. That's why you need a resistor that won't arc over for even an instant, with say 3-4,000 volts applied. Remember that similar voltages are created by walking across a carpet, so pins that are brought out need similar protection. P.S. I agree with distrusting application notes. I have been burned in the distant past by using application note circuits (from RCA), bought samples and wasted several days before I realized that the app note circuit could *never* work reliably. Now I always do the calculations from the min/max figures on the data sheet, and don't trust the people who wrote the app notes (who seem to be young and inexperienced engineers trying to be clever and sell chips, and don't put the care in there to make a commercial conservative design that will always work under a wide set of conditions). In a way, it's a bit like suckering you in.. here buy our part and you can make this product with only 17 parts. If they showed you the 34 you really need to do a good job, you'd shrug and flip the page. Unfortunately, some are suckered to the point of actually trying to produce the "design" with the 17 parts and get burned. I've just been struggling with such a PIC "design" that someone else did. 8-( Best regards, Spehro Pefhany --"it's the network..." "The Journey is the reward" speff@interlog.com Info for manufacturers: http://www.trexon.com Embedded software/hardware/analog Info for designers: http://www.speff.com 9/11 United we Stand -- http://www.piclist.com hint: To leave the PICList mailto:piclist-unsubscribe-request@mitvma.mit.edu