AHHHH!!! now the light goes on. So, let's assume 'pic' specs, of ... oh crap... Is this saying what I think it is? Note 2 on page 417: =20 http://ww1.microchip.com/downloads/en/DeviceDoc/39762e.pdf PIC18f97J60 "2: No clamping diodes are present." That presents an interesting set of challenges... But assuming for a minute that had a different reading than I think it=20 does - and is , what I was going to say was that basically If I had a=20 input resistor of say 330 ohms, and a more typical maximum clamp current=20 of 20ma, the voltage at the 'input' to the resistor will be around=20 330x0.020=3D6.60+Vdd? So I really just need to ensure a clamp to no more=20 than roughly 9.6V in this example? Sounds like I could use a separate=20 rail (probably produced with a S-diode from VDD), and 'clamp' to it, and=20 then use a 5V1 zener to ensure that rail never gets over 5ish volts.... And back on the '97j60 issue... looks like I need to find a really low=20 Vf diode for all the inputs if I use this... That PIC32 part is looking=20 better and better, except for the external ethernet phy. -forrest On 12/31/2010 2:41 AM, Mark Rages wrote: > On Fri, Dec 31, 2010 at 3:15 AM, Forrest W Christian = wrote: >> Assuming for a minute there *are* input protection diodes, what exactly >> would a external set of diodes gain me, if they don't have a Vf less >> than the input diodes? > Suppose you have a micro that has a 0.3 V protection diode, and an > external 0.7 V protection diode. Suppose your micro can tolerate 1mA > in its protection diode. > > If you have a resistor of (0.7-0.3) / 1mA =3D 400 ohms between the > external diode and the micro pin, you can guarantee that the input > current is limited to a safe level. > > Regards, > Mark > markrages@gmail --=20 http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .