Hi Olin Mucho thanks. I've been out for a few days. My future wife and I just closed on a new house and I spent the last few days with the movers schlepping her and my junk to our new place. The only thing worse than moving is moving twice! I'm back to work but my aching back is still out. I tried the ribbon cable but it made no difference. The thing would program if the device was blank with either the ribbon cable or the RJ-12 cable so I was fairly confident there was a problem with erasing the device not the cable. My soldering skills are legendary (a very dark and disturbing legend) but the cable I sent worked fine with my trusty crusty ICD2. I did add the 47nF caps to the PGC and PGC lines. I added them to the adapter board portion of my programming cable not to the target board itself. The point of using the SM connecters is that they are a stock item used else where in other designs and that the connector can be replaced every few days for only pennies so they aren't expected to last very long. We just shipped 250 units of a 16F676 design and the same connector is still going strong. My technician is a woman (my sister with the same steady hands genes as my brother the watch maker) with tiny hands. If it were me with my ten thumbs I seriously doubt I could get any where near that level of usage. Even if I have to replace it every couple of hundred boards then it is acceptable. I copied the interface of my circuit from the schematic of a PICDEM 2 PLUS development board from Microchip so that is how I ended up with the cap on VPP. I'm more than happy to not load that part...fewer parts is better in my mind. Do you (or anyone else out there) have any idea why Microchip puts a 10nF on VPP?? I'm wondering if I should remove it from the 16F676 design as well?? Phillip Things should be as simple as possible but no simpler Phillip Coiner CTO, GPS Source, Inc. Your source for quality GNSS Networking Solutions and Design Services, Now! -----Original Message----- From: piclist-bounces@mit.edu [mailto:piclist-bounces@mit.edu] On Behalf Of Olin Lathrop Sent: Friday, April 20, 2007 5:13 PM To: Microcontroller discussion list - Public. Subject: [PIC] Pro programer/head in dark place troubles > I'm going to send it FedEx today. I got it yesterday (Thursday) and didn't have time to really look at it until today. > My programming cable has a handedness the pictures on the CD will make > this clear. The he target board is mounted on an RF board. Remember to > disconnect the target from the RF board (just pull them apart) as it is > not 5V tolerant. I have it disconnected and intend to leave it that way. I don't care about your device working, only that the ProProg can program it. Anyway, here is what I found so far: The first thing I did was plug in a ProProg and see if it could ID the unit. It could. It programmed and verified at 4.2V and 5.5V but then got a verification error on a config bit. Then I tried it with a USBProg and it worked fine. It programmed and did two verification passes in 59 seconds using your HEX file. But every time I tried with a ProProg I got different results, so I looked at the board more closely. You never did some of the things I asked you to, and I expected you had when trying to understand your symptoms. There were no caps on PGC and PGC and you were still using the ICD2 cable instead of a ribbon cable using the 6 pin header on the proprog. I added two 47pF caps to your board and in the process noticed that one pin on your adapter board that plugs onto the 6 pin header on the targer board, and possibly one pin on that 6 pin header looked like maybe there was a solder problem. These kind of SMD headers and sockets are not meant for much stress. I wouldn't use them for more than mating two boards once that are then held together with screws. Thru hole is definitely more robust for stuff like this. I fixed the suspect solder joints (at least they looked better with a jewler's loupe), and then operation was reliable. Reliably not working that is. To take the flaky connectors, kludge board, and the ICD2 cable out of the equation, I made up a ribbon cable that connects to the 6 pin header on the programmer and soldered the other end directly to the pads of the programming header on the target board. Same result. I'll spare you the various trouble shooting that followed, but I think I know what the problem is. The big difference in signals between the USBProg and the ProProg is Vpp. When not in programming mode, the USBProg holds this low via 22ohms to ground. When enabling Vpp, the signal rises from 1V to 13V in about 200nS. However when the ProProg is driving Vpp high it jumps to 5V, stays there for 10mS, then does a exponential decay to 13V over a few mS. This is definitely not right, and I think explains the problems. The slow rise time from 5V to 13V is due to your cap C21, which I hadn't noticed before. You don't want to put 10nF on Vpp! I don't have a layout of your board, and couldn't find C21 to try removing it and see what happens. If you give me a layout or tell me where it is I'll try it. While 10nF on Vpp is definitely a problem, the 10mS plateau at Vdd is a firmware issue in the ProProg that I'll try to address this weekend. I think I understand why it is happening in this case. However, I think the rise time will still be too slow with 10nF on the line after I fix the firmware. ******************************************************************** Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products (978) 742-9014. Gold level PIC consultants since 2000. -- http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist -- http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist