Matrix user forums

Apologies for what is most probably a dumb question but I'm getting really confused and frustrated.

When I develop projects I build in steps. When a step is successful I baseline it and move onto the next piece of functionality building the project in stages. What I’ve noticed is occasionally when I revert back to a known working baseline the previously successful baseline no longer functions correctly.

It’s almost as if I changed the chips character/fuses but when reverting, the chips not returned to its same state as it was.

To try and prove my point, I substituted a brand new (18F44K22) and burnt a previously successful baseline and it worked as expected. I then burnt the next unsuccessful baseline which of course failed again. I then reverted and the previously working version is now failing.

Does this make any sense and is there a way or returning a chip back to the same state as a new device. I’ve tried but apart from substituting a new chip I can’t seem to find a way to “go backwards” when this happens. It seems to affect the higher end devices more than the simpler varieties.

Many thanks

anzacinexile wrote:Does this make any sense and is there a way or returning a chip back to the same state as a new device. I’ve tried but apart from substituting a new chip I can’t seem to find a way to “go backwards”

My apologies however this makes no sense as a K or F series chips has flash memory which you can program and re program hundreds of times a limit on Flash programming I have seen at Microchip data sheets are 1000 times,even if you have a problem hex file or baseline if you reprogram with a baseline which worked the chip should run as before
What programmer are you using ?

To my mind you have a programmer that destroy your chips or you are not setting up the configuration every time you program
it is hard to judge what is the problem on your side with the info we see here in the forum.

The chips are not being destroyed when I go back, its just that they do not seem to behave in the same way. I'm using a PickIT 3 as a programmer but as I said, the chips are not damaged in any way and I would agree, re-programming them I would have thought would have returned the chip back to the same as before but testing does not seem to confirm this

Perhaps there are some fuse settings that are not being cleared. You could try reading back the whole chip contents and comparing one that works with one that doesn't. There are other options in the PICkit programming software that select which areas of the chip to clear/reprogram and so you could look at these a bit more. The MPLAB IPE tool will allow you to change these options.

Thanks Steve - I'll have a look and report back

Hi anzacinexile,

I just wondered if this link would help, it may be interesting to you:

http://www.microchip.com/forums/m640643.aspx

Thanks
Stuart