I haven't used 18 series parts or the ICD2 yet, but the problem description
is similar to a problem one of our technicians had with an ICD1 and
16LF876's running at 3.3V. In that case I found the solution at the very
bottom of the ICD1 readme file.

*****From README.ICD******
-----------------------------------------------------------------
Low Voltage Operation
-----------------------------------------------------------------
When using the MPLAB ICD with a target voltage of less than 4.5v,
be sure to do the following:
 1.Turn off the Erase All Before Program feature under the
   Program Options.
 2.Make sure all code protect features are turned off. Once the
   device has code protect turned ON, you will need a 4.5v target
   voltage to clear it.
 3.It is recommended to set the End Address under the Program
   Options to a little past the end of your code to save
   programming time.

CAUTION: DO NOT USE the ERASE or Erase All Before Programming
Function. If you do, you will not be able to reprogram the device
until you supply a 4.5v target voltage.
*****************************

Maybe this also applies to the ICD2?

Paul

>-----Original Message-----
>[mailto:PICLIST@MITVMA.MIT.EDU]On Behalf Of Michael Rigby-Jones
>Sent: Wednesday, July 09, 2003 11:12 AM
>
>1) Does anyone know a source of a programming adapter suitable for an
>18LF8620 (80pin TQFP 0.5mm pitch) preferably suitable for use with a PS+
>(i.e. 40PIN DIP)?
>
>2) I am having a nightmare trying to program the above devices using ICSP.
>I have been trying a Promate2 (with MPLAB 5.74 and MPLAB 6.30) and an ICD2
>(with MPLAB 6.30) and the results with both are the same.  The PIC has to
>run at 3.3 volts, and the power is shared by many devices that are not 5v
>tolerant.
>
>With a brand new PIC, if I perform a blank check, the programmer (either of
>them) reports that the device is not blank.  Specificaly, the
>program memory
>always has the same pattern in memory starting with 0xC0C0 at address
>0x0000.  If I try to erase the PIC at 3.3volts, it fails to erase the
>program memory, but the PROMATE then reports that the configuration fuses
>are clear.  This is exactly the opposite to what logic dictates as the
>datasheet states that the configuration fuses can only be cleared by a bulk
>erase at >=4.5volts.  In any case, the PIC fails to program.
>
>After pulling my hair out for a day, I stared lifting the Vdd pins on the
>PIC so that I could power it from an external 4.5volt supply and perform a
>bulk erase.  After this, the PIC will program happily at 3.3 volts.
>However, this is really not practical as a) I have a stack of boards that
>all exhibit this problem, and b) the PCB is made from a material suitable
>for high frequency use and the copper delaminates very easily compared to
>FR4.
>
>In the (recent) past I have seen a simmilar issue with a few other 18F
>series PIC's, notably 18LF452 and 18LF6620. However, it was only the
>occaisional device and I was able to apply 4.5 volts without causing damage
>to other components in those cases.
>
>Has anyone else seen this, and does anyone have any suggestions as to what
>to try next? (Microchip has been contacted, but deadlines are looming!)

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email listserv@mitvma.mit.edu with SET PICList DIGEST in the body