> The chip internal programming must work because it programs properly.

That hasn't gone unthought about ;-)

Although /mclr is pulled up to ~13V during programming

> Have you tried writing a new and simple program that just writes 8
> bytes to 0x1840 and then read the chip back in a programmer?

Ignoring the outer '8-byte' loop ? Yes

> Any floating input pins? - they can cause weird things to happen -
> "for some reason it came right twice" :-)

All pins except /mclr are o/p, which has 10k/100n. There's about
30mV noise below Vcc on the pin during the self-write

Joe


-----
No virus found in this message.
Checked by AVG - www.avg.com
Version: 2016.0.7797 / Virus Database: 4656/13185 - Release Date: 10/10/16

--=20
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
.