Searching \ for '[PIC]: Mysterious accidental config word changes' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: techref.massmind.org/techref/microchip/devices.htm?key=pic
Search entire site for: 'Mysterious accidental config word changes'.

Exact match. Not showing close matches.
PICList Thread
'[PIC]: Mysterious accidental config word changes'
2006\07\05@153329 by Sean Breheny

face picon face
Hi all,

I have a small production run of 60 boards using 16LF88s. Most are
working fine but after several firmware updates, I now have three or
four which can no longer be programmed, they always read all zeros for
the program memory. Examining the config word, I find the following:

Normal (programmable) PIC config word: 3F 35 3FFC
Abnormal PIC config word (on the few that will not program): 1E 35 3FFC

That is, there are two bits altered: flash AND data memory code
protect has been turned on.

There are two programmable devices on this board which are programmed.
The procedure is to erase the PIC, leave the board powered up
(including the PIC), program the other device (it is a Maxstream radio
module and IS connected to the PIC's pins but uses no high voltages in
programming), then program the PIC using MCLR (high voltage
programming - LVP is turned off in the config word).

Other than some weird bug with the PIC or a static problem, the only
thing I can think of is that somehow programming the Maxstream affects
the PIC sometimes, even though it doesn't place high voltage on MCLR.

Are there any known bugs which could cause this? Any ideas?

Thanks very much as always,

Sean

2006\07\05@172919 by Gerhard Fiedler

picon face
Sean Breheny wrote:

> I have a small production run of 60 boards using 16LF88s. Most are
> working fine but after several firmware updates, I now have three or
> four which can no longer be programmed, they always read all zeros for
> the program memory. Examining the config word, I find the following:
>
> Normal (programmable) PIC config word: 3F 35 3FFC
> Abnormal PIC config word (on the few that will not program): 1E 35 3FFC
>
> That is, there are two bits altered: flash AND data memory code
> protect has been turned on.
>
> There are two programmable devices on this board which are programmed.
> The procedure is to erase the PIC, leave the board powered up
> (including the PIC), program the other device (it is a Maxstream radio
> module and IS connected to the PIC's pins but uses no high voltages in
> programming), then program the PIC using MCLR (high voltage
> programming - LVP is turned off in the config word).

After erasing the LVP bit is turned on. If you haven't pulled down the PGM
pin and are wiggling the programming pins, strange things could happen.

> Are there any known bugs which could cause this? Any ideas?

This is not a bug, it is by design :)

Gerhard

2006\07\05@174348 by Bob Axtell

face picon face
Sean Breheny wrote:
{Quote hidden}

I do a lot of work with the F88; I have 6 active production designs running.

The F88 and LF88 are very sensitive to the level of the LVP pin,
regardless of the state of the
configuration bits. I'd try that first. A 10K resistor to GND seems to
solve that, in addition to
whatever else is connected to that pin. This is probably your problem. I
place a 10K to GND
on the LVP pin of ALL products I design. And I never have programming
problems (any
more). I realize that Microchip says this isn't needed, but in this case
they are wrong ("Pay
no attention to the man behind the curtain").

Another thing is that the F88 (like some other nanowatt PICs) require
running timers during
powerup, and if the devices are not bypassed while being programmed,
sometimes there
are problems because the oscillators are having trouble starting. 0.01uF
is plenty (10nF).

Try that and let us know.

--Bob


2006\07\12@234942 by Sean Breheny

face picon face
Thanks very much to Bob and Gerhard. That should have been obvious to
me but wasn't :) The fix I implemented was simply to pull the MCLR
line low while I'm programming the radio. I can't pull the LVP pin low
because it is used to program the radio. So far there haven't been any
failures since I made this change. We're about to make 90 more boards,
so we'll see in that run whether it is a permanent fix.

Sean


On 7/5/06, Bob Axtell <spam_OUTengineerTakeThisOuTspamcotse.net> wrote:
{Quote hidden}

> -

More... (looser matching)
- Last day of these posts
- In 2006 , 2007 only
- Today
- New search...