Cool. Based on both Russell's and your recommendations, I redrew the circuit incorporating this. It's at: http://www.geocities.com/roinesreenig/simple_pwr_protection2.JPG The 5V 2A power supplies that I am using are unregulated. I'm in the process of trying to make decisions about the following set of issues: 1. I've put two caps in the circuit, C1 and C2 for cleaning up the line noise and handling intermittent undercurrent (is that the right term to describe extremely brief lapses in power?). I plan to pick a small value like 330nF for the noise and a big value like 470uF for the undercurrent. I noticed that Philip's voltage regulator zener diodes like the BZD27 series incorporate transient suppression so maybe these caps are unnecessary if I pick an appropiate Zener with TVS? 2. Zener choice. Here's the Philips zener datasheet:http://www.semiconductors.philips.com/acrobat/datasheets/BZD27_2.pdf This one looks good. I'm thinking of picking the BZD26-C5V6. This selection would tie in with the thyristor triggering issues. 3. Triggering of the thyristor. I'm worried that a transient voltage might cause the SCR to trigger. Should I put a cap in parallel with the resistor, R1 to filter out stuff that might get past the Zener. If I think about it, I get confused because the cap may charge up using the leakage current and then discharge to trigger the thyristor in which case I shoudn't put it there. I need help thinking about that. I looked at the BT148 series thyristor from phillips. http://www.semiconductors.philips.com/acrobat/datasheets/BT148_SERIES_1.pdf Looks like a gate trigger current of 15uA and a gate trigger voltage of 0.4V. I didn't find an IV curve for the BZD26-C5V6 but I suspect a sustained input voltage of about 6Volts would definitely trigger the thyristor. 5.6V - 0.7V (drop across Zener) = 4.9V. 4.9V/4.7kOhm =~ 1mA so sufficient to trigger thyristor. Is that calculation right? 4. Diode D1, I'll use a 1n5400 as you described. Did I understand you correctly about this together with the thyristor forming the inverse parallel diodes? Thanks, Roines Steve Smith wrote: Add the thyristor to the first circuit and let the wall wart do the regulation if its 5 v then don't regulate it 5v its there already use a 5v1 for the zener it will make the thyristor work at about 5.7v allowing for gate sensitivity. Don't waste the fet put an inverse parallel diode in parallel with the thyristor it will blow the fuse if backwardz rate at about 3A (1n5400) alternatively use circuit 2 with a higher input voltage (10v) Regardz Steve -----Original Message----- From: pic microcontroller discussion list [mailto:PICLIST@MITVMA.MIT.EDU] On Behalf Of roines reenig Sent: 12 May 2004 06:22 To: PICLIST@MITVMA.MIT.EDU Subject: Re: [EE:] polarity reversal and overvoltage protection Resending with url's to the images instead of attachments: http://www.geocities.com/roinesreenig/simple_pwr_protection.JPG http://www.geocities.com/roinesreenig/pwr_reg.JPG roines reenig wrote: Hello, I read the threads about polarity reversal and overvoltage protection with interest . In my situation, I've got cheap access to 2Amp unregulated 5 Volt DC power supplies at about USD $3 a piece. I have a bunch of unprotected equipment that take 5 Volt input needing currents from around 500mA to 1A. I'd like to provide protection for these circuits. Here's what I've come up with based on what I've read so far. I'll put in a FET to protect against voltage reversal and a Zener to crowbar the supply if an overvoltage condition is reached. A 2Amp hand replacable fuse would protect against overcurrent and also protect the power supply if we crowbared the supply. I've attached a jpeg schematic of what I'm thinking of. What do you all think about this? Should I add some caps to provide minimal spike and line noise filtering? If so, what kind of cap sizes should I put in? Would a small cap to take care of the high frequency noise and a large cap to take care of any transient line drops be a good idea? I realize that the above doesn't provide regulated 5 Volt output. If I wanted to provide regulated 5 Volt, then I would add a thyristor, a 5 volt regulator and use a Zener on the other side to trigger the thyristor on thus crowbaring the supply again. I've attached a jpeg of this. I stuck an LM7805 as the regulator but I guess this is not a good choice since I've only got a 5Volt input. Any suggestions? I looked at an LM309 and some others but didn't spot anything that looked good as a 5volt-to-5volt regulator. Thanks, Roines --------------------------------- Do you Yahoo!? Yahoo! Movies - Buy advance tickets for 'Shrek 2' -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.