On Tue, Aug 14, 2001 at 10:25:42AM -0400, Jeff DeMaagd wrote:
>
> > BTW. Is the CUMP project still alive? Has anybody heard about the
> > another open source/open hardware programmer able to programm different
> > families uf uC (8051-like 8952-like, PICs, AVRs, 68HCs... ), and different
> > EPROMS & EEPROMS?
>
> > It probably could be done with single 16F877 and a little
> > of analog stuff, but has anybody tried it?
>
> There was a huge flame wa^h^h^h^h^h^h discussion about it a few months ago.
> Unless the people that want one contribute work on it, it probably isn't
> going to happen.  There are a few technical limitations involved, on top of
> the fact that it is a_huge_ variety of chips involved, often each with their
> own quirks.  That spells headaches.
>
Well, I can see a few difficult details:
1) For chips with LVP - everything can be done in software, just connect
   the pins of 16F877 (or 8952) to the ZIF socket, provide the power supply
   and "only" ;-) write the appropriate software
2) For standard PIC's there is need for "high voltage driver" which should
   provide the "high voltage" (usually 13.5V or so).
   It can be done quite easily, eg. with circuit like this:

  "high volt."--+------+          HV OC TTL = "high voltage open collector TTL"
                |      |
                |      [] R2         N1  N2    Pin voltage
                 \E B  |             L   L     vorbidden !!!
               Q1 |----+             L   H     ca. GND
      D1   R1    /C    |             H   L     ca. HV
  +5V-|>|-+===+-+      [] R3         H   H     ca. +5V
                |      []
           N1   |      |             Q1 - bipolar PNP
 HV OC TTL -----+-pin  |
           N2          |
 HV OC TTL ------------+

3) For verification with raised/lowered Vdd, there is a need for special
   drivers interfacing the programmer's uC (powered from +5V) with
   programmed uC (powered from other voltage)
   It can be done eg. with circuit like this:

               R1          pin
   uC's Vdd --+==+---+-----+---     R2
                    C|     |     +-+==+----+5V
                  B /   R3 []    |_______ data out   Q1,Q2-bipolar NPN
    data in -------|Q1     |  B / C
                    \      +---| Q2
                    E|     |    \ E
                     |  R4 []    |
                     |     |     |
           ----------+-----+-----+

    It is not a problem with devices supporting serial programming, but when
    we have to deal with parallel data bus =:-( ...

    I would be nice if we could change the Vdd of programmer's uC together
    with verified uC, but I really don't know if it is safe to change Vdd
    in working PIC. Has anybody any experiences with that? Doesn't it cause
    the spurious reset or other temporary malfunction?

4) Probably the programmer should have one circuit from point (2) and a few
   circuits from point (3). Now the problem is with connecting the drivers
   to the ZIF-socket pins. The only idea I have is the "cross-wiring plug"
   connecting the ZIF to the drivers. There should be different plugs
   for different chips (or at least chip families).

   Does it make any sense? Any remarks are appreciated.
--
                                TIA & Regards,
                              Wojciech M. Zabolotny
        http://www.ise.pw.edu.pl/~wzab  <--> wzab@ise.pw.edu.pl

http://www.freedos.org  Free DOS for free people!

--
http://www.piclist.com hint: To leave the PICList
mailto:piclist-unsubscribe-request@mitvma.mit.edu