>Well, it was topic on the list often enough to know that setting
>code protection bits on a JW device is not recommended.
>
>
>According to the programming specs, a normal cell receives up to
>25 programming pulses. If it still doesn't read back correctly,
>the chip is considered bad. Another 3N cycles are applied then,
>to be on the safe side for the whole voltage range.
>
>That would make for a max of 100 pulses for a normal cell.
>
>
>Someone on the list pointed out that his PIC12 actually reads back
>correctly after one single programming pulse, and that he skips
>the 3N overprogram pulses to speed up his development cycle (reduced
>erase time).
>
>If that holds true for the C7x, a normal cell receives just 4
>programming pulses or maybe 8, but no near to the max of 100.
>
>
>The config word (code protection) however does _always_ receive
>100 pulses (making up a total of 10ms programming time), according
>to the programming specs.
>
>
>I wonder whether the UV erase problems with code protected devices
>are caused by UV resistant coverings on the chip surface, or
>simply by _overprogrammed_ code protection bits?!
>
>Has anyone noticed if programmed XTAL or WDOG bits expose the same
>resistance to erasure as code protection does? ("No" would speak
>for UV opaque covers)
>
>Can you confirm that these devices really verify OK after just one
>single programming pulse usually?
>
>
>I'm looking forward to your comments on the matter..
>