It has become clear that way too many of the 18F1220 pics used in the SSR4 do not want to program start addresses correctly. I blame the firmware but can not find any reason. A Test program using microchips code cut and pasted into a new firmware that does nothing else but program the eeprom, will work on the chips that will program on the SSR4, but when run on a pic that will not program on the SSR4 they also fail to program on this test program.
I also have found a couple of other examples of people having issues with a few models of pic18 chips and eeprom.
So to remove this issue from the users without requireing you to be a microchip stock holder (having lots of extras). I am in the process of rewriting the firmware to move the start channel from eeprom to flash memory. In testing so far the bad chips will program flash OK.
I am tweaking it now, as I have it mostly finished after a long couple of days. It should be posted in the next couple of days for use. I hope that this will remove the issue as a whole. I am sorry I did not find it earlier but the first few I built for protos had chips that programmed fine so I was not aware until everyone started to have issues. Then it appeared as bad chips since replacing them fixed it in almost all cases.
RJ