On Sat, 14 Jan 2006 00:42:20 GMT, you wrote: >In message > Mike Harrison wrote: > > >> Contact bounce is not an issue as it only happens when switching in or out of programming mode > >Right. Four-pole relay it is then. Small 3V coil relays were hard to find when I was looking (can't remember if they were not listed, or just out of stock at the time), 4pco in 3V is probably unobtainium so a pair of 3v ones may be needed.. My setup powered the isolator from the host power, on the basis that it would always be there, and allowed the isolator to be a simple in-line addition to the ICD cable. There would be some benefits to tapping into the 5V supply within ICD2, one of which would be the ability to use more easily obtainable 5v relays. Maybe just mod ICD2 to put 5V down the unused pin on the RJ connector. >> Don't really see the need for protection - the pin voltages would be clamped to the PIC pins the >> ICD2 is connected to. Failing that a couple of 5V6 zeners would be more than adequate. > >I was thinking of "catastrophic failure protection" - a 5V-to-12V supply >bridge. I've seen it happen when motor controller ICs go pop... > >> Why would you want 25V isolation ? > >The most you're likely to see on any PIC board is 12V. Theoretically, the 12V >could get bridged to the 5V line. The protection circuit should ideally be >able to block twice the rated voltage - 24V. I just rounded it up. So you just want protection, not isolation. 3W 5.6V zeners on the data and power lines, and 14V on MCLR should be more than adequate. If you have something that blows these your target board will be serious toast anyway... -- http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist