At 09:22 PM 2/24/02 -0300, Alexandre Guimar=E3es wrote: > It is almost impossible to keep it under the maximum current allowed= all >the time !!! Remember that you have many high voltage spikes going on any >regular power line. Even more if you are using a dimmer that generates lots >of noise. Imagine that switching a heavy inductive load such as a halogen >lamp transformer ! The internal diodes cannot protect from these spikes.= You >have to use external protection to make sure you do not violate the specs. >That application note has always amased me because it is just pure bad >enginnering. I would never use that to make a real product. At 03:15 PM 2/25/02 +0000, Michael Rigby-Jones wrote: >This topic has been discussed numerous times on the list. IIRC, the= general >feeling was that injecting currents into the protection diodes caused >unpredictable behaviour and was best avoided. The magnitude of current was >well below that which could cause any kind of SCR latchup as well. I can comment on both the above statements: I *do* use a PIC input pin to sample zero-cross in several products. There= =20 are 2 - 1 Meg resistors (0.25W, 400V) in series to sample a 120 Vac=20 input. Some boards have 1n4148 clamps to Vdd & Gnd, some boards have a=20 4.7V zener clamp to Gnd, some boards have no external diode clamps. I=20 don't use a capacitor filter at the pin if the intent is to sample zero=20 crossing (too much phase shift) but do use a capacitor if the requirement=20 is only use the AC mains as a timebase. I have had *NO* failures related to this out of several thousand boards -=20 even those without external diode clamps. But I did a couple of things=20 ahead of time: I qualified the resistors I use by testing several hundred= =20 pieces with a pulsed 1200 Vac supply to ensure that they do not break down= =20 or arc *and* I have a MOV at the front end of the board - where the 120=20 Vac enters the card. I feel that 2 resistors in series, each resistor=20 having a 10x test margin, is adequate. I *always* use clamp diodes if the PIC has a/d input that I am using. I=20 have measured significant a/d errors on the 12c671, 16c71, 16c73 parts if=20 there is *ANY* substrate current resulting from the ESD protect diodes=20 conducting because some other input pin was above Vdd or below Gnd. I=20 don't know about the comparitor inputs on the '62x family of parts - I'll=20 find out when I do a project that actually uses those those features. On the other hand, minor substrate currents (0.1 mA or less) appear to have= =20 no effect on the purely digital inputs on the PICs I have used to date. Finally, this (2 resistors + MOV) seems to work for me with a 120 Vac mains= =20 supply. All bets are off at higher voltages. dwayne Dwayne Reid Trinity Electronics Systems Ltd Edmonton, AB, CANADA (780) 489-3199 voice (780) 487-6397 fax Celebrating 18 years of Engineering Innovation (1984 - 2002) .-. .-. .-. .-. .-. .-. .-. .-. .-. .- `-' `-' `-' `-' `-' `-' `-' `-' `-' Do NOT send unsolicited commercial email to this email address. This message neither grants consent to receive unsolicited commercial email nor is intended to solicit commercial email. -- http://www.piclist.com hint: To leave the PICList mailto:piclist-unsubscribe-request@mitvma.mit.edu