Hi all,

after several days of experimenting with my homemade ICD (modified Vpp
supply circuit) I came to a painful conclusion - it will not work this way.
It will program the chip, but will fail in debug mode. The reason is realy
curious - althought there is the Q3 to switch Vpp on and off, MPLAB
firmware fiddles with the charge pump on/off _and_ also applies a special
condition when the charge pump is switched off and both Q2 and Q3 are
opened. There are three or four such pulses 100msec long when the target is
being reset. The purpose could be to remove remaining charge from C10-C12.
With modified Vpp circuit it of course cannot establish propper signal levels.

My applogy to all, who already built the modified ICD. Lacking any
experience with the original ICD I didn't spot the problems right away and
published the circuits. As time permits, I will switch back to the original
charge pump solution and rework the PCB. Notification will be sent the the
list.

Josef

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]: PIC only [EE]: engineering [OT]: off topic [AD]: advertisements