robert a. moeser wrote:
>
> hi!
>
> i am still struggling with the last details of my 16C505 burner.
>
> i have reported here some weird behaviour, specifically the chip
> seems to go into code protect when i innocently burn some zeroes
> in non-code protect bits of the CONFIG at 0xFFF.

> 5. burn the chip again, taking CONFIG to 0xFD4.
>
>     all i expect from this is the change of RB4 to an output pin, but in
>     fact the chip goes dark, that is it cannot be verified or otherwise
>     read.
> i've been over the data sheet, but i still cannot see what i am doing
> wrong. i am cheerfully willing to admit that i am an idiot and i would
> love to have someone prove it by telling me what i am missing.

Hi Rob, so you have checked page 27 of the datasheet?
Note 3. says that all code protect bits must be set to
the same value. Check this?

Why don't you use the __CONFIG stuff in the 16C505.inc
file that comes with MPLAB?? This sets all the bits in
a nice sensible "&" fashion. Will your programmer
work with that?? Here is the stuff:

_MCLRE_ON                    EQU     H'0FFF'
_MCLRE_OFF                   EQU     H'0FDF'
_CP_ON                       EQU     H'002F'
_CP_OFF                      EQU     H'0FFF'
_WDT_ON                      EQU     H'0FFF'
_WDT_OFF                     EQU     H'0FF7'
_LP_OSC                      EQU     H'0FF8'
_XT_OSC                      EQU     H'0FF9'
_HS_OSC                      EQU     H'0FFA'
_IntRC_OSC_RB4EN             EQU     H'0FFC'
_IntRC_OSC_CLKOUTEN          EQU     H'0FFD'
_ExtRC_OSC_RB4EN             EQU     H'0FFE'
_ExtRC_OSC_CLKOUTEN          EQU     H'0FFF'

        __CONFIG   _MCLRE_OFF & _CP_OFF & _WDT_OFF & _IntRC_OSC_RB4EN

Maybe you can work it out from that. :o)
-Roman

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.