On Sun, Jun 24, 2001 at 07:29:37PM +0200, =C5ke Neehr wrote:
> Hello!

Hi. I thought I just answered this question less than a week ago.=20

>=20
> In Microchips directions of programming
> for PIC16F87x  devices can You read
> following:
>=20
> "To disable low voltage ICSP mode, the
> LVP config bit must be programmed to'0'.
> This must be done while entered with high
> voltage entry mode (LVP config bit=3D 1).
> RB3 is now a general purpose I/O pin."
>=20
> I don=B4t understand exactly what they say,
> therefore I=B4m asking the autority on the
> PIC-list before I destroy the chip,

It's very unlikely that you'll destroy the chip in any case.

> Could it be like that:
> If the chip is programmed with LVP ON
> from the factory so must the pin RB3
> beconnected to VDD (+5V) under the
> programming before it can change from
> LVP on to LVP OFF?

No. Here are the rules:

1) The chip can always be programmed in HVP mode. Always.
2) The chip can be programmed in LVP mode only if the LVP config bit is s=
et.
3) The LVP config bit can only be changed in HVP mode. LVP mode cannot ch=
ange=20
   the LVP config bit. This prevents an LVP programmer from making the ch=
ip=20
   not programmable with said LVP programmer.
4) When LVP config bit is set, RB3 must be held low except during LVP=20
   programming. So that means that RB3 must be held low during program=20
   execution and that RB3 is unavailable during execution. It's the only=20
   design mistake I see in the whole process. LVP should have been assign=
ed to=20
   a end pin of a port instead of a middle pin.
5) There is an errata sheet that stipulates that RB3 should be held low
   during HVP programming even if the LVP config bit is off. It implies t=
hat=20
   the chip could be put into a weird mode if RB3 is high (or floating) w=
hen=20
   trying to program in HVP mode. Essentially the chip tries to go into L=
VP
   programming mode before it makes it to HVP mode.

   You can find that errata sheet here:

   http://www.microchip.com/Download/lit/suppdoc/errata/80076b.pdf

   As a precaution I always tie RB3 to ground through a resistor even if =
I'm
   building a HVP programmer for 16F87X parts.


That's it. That's all there the rules for HVP/LVP programming on 16F87X p=
arts.

So here are the issues in answer to your specific question.

1) If it's an LVP programmer (setting RB3 to +5 as you propose) you canno=
t=20
change the LVP config bit according to rule #2 above.

2) If it's an HVP programmer, you should ground RB3 through a resistor=20
(10k recommended in the errata sheet) so you would not set it to +5. An H=
VP
programmer would be able to change the LVP config bit.

So there is no way to change the LVP config bit if RB3 is set to +5.

Hope this helps,

BAJ

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email listserv@mitvma.mit.edu with SET PICList DIGEST in the body