Connect with us

16f628 - wont program!

Discussion in 'General Electronics' started by Quack, Aug 28, 2004.

  1. Quack

    Quack Guest

    Hi,

    I recently ordered a new batch of PIC chips, only this time i got the
    16F628A instead of the 16F628 i was using previously.

    I am using a p16pro paralel port programmer, and picall for windows
    (latest version), it apparently supports the 16F628A.

    It can not write even the first byte, its asif the chip is not
    entering programming mode ... i checked the voltages, they are okay,
    and it works with the 628 no problems.
    Does the 628a need more programming voltage ?
    Has anyone got experience with these chips and this progammer ?

    Thanks!,

    Alex.
     
  2. Robert Baer

    Robert Baer Guest

    Manufacturers have a habit of making an "A" version that has rather
    different programming requirements than the older non-A versions.
    Perhaps your software for the programmer does not support the "A"
    version; if so, it will be a completely different entry, and if you
    cannot find it, then you are using an incorrect programming scheme -
    which might (or might not) damage the "A" parts.
    I have seen "A" versions that used lower programming voltages,
    different programming pulse schemes, and different repeat-then-verify
    cycles to ensure reliable but fast programming.
    The first part of using lower voltages could lead to part damage if
    the older, higher voltage is used (depends if manufacturer process of
    new part can tolerate the higher voltage).
    The rest could lead to bit "dropout" - or not; when there is no
    disclosure of methodology, there is no way to tell.
    Back in the dim dark 80s, there was a lot of discussion of methodology
    and the whys; many speedups and increases of reliablity of programmed
    bits resulted from the open community efforts.
    Then companies made slight changes (some cosmetic, some not) and
    copyrighted or otherwise held actual details secret; now it is so bad
    that the companies themselves may not know (lost in an old dusty file or
    may even have been tossed).
    Naturally, when a manufacturing process changes, the internal
    characteristics change, and the programming scheme should be changed for
    optimum results.
    Due to the secrecy it is unknown if any changes are made and if so,
    what they may be.
    Design and circuitry changes, along with possible process changes can
    make for a decidedly improved ("A"?) product, with everything changed.
     
  3. Quack

    Quack Guest

    Thanks for that, but it doesnt help me :(.

    I was under the impression the 16f628a is very recent, and being pin
    compatible to both the 628 and older f84 chips, i figure there would
    be plenty of info to program the bastard :(.

    There is one saviour though, i can build a small icsp programmer and
    do them that way, theres lots of info for doing that around the place.

    I noticed some earlier posts of people asking pretty much the same
    question, there were links to some docs describing the difference in
    detail between the 628 and the 628a on microchip.com - but the links
    were broken and the docs dont come up in a search.

    In the 628a or 628 Docs, i find no relevant info regarding programming
    changes.

    My software (www.picall.com) definantly has the programming scheme for
    the 628a specifically, so i can only guess that my programmer
    (hardware) wont work with it.. but why ? very strange.

    Oh well, any recommendations of a simple ICSP programming circuit that
    definantly works with these 628a's ? It doesnt have to support
    anything else.

    Alex.
     

  4. Microchip has an errata sheet for early silicon of the "a" parts. Maybe
    this explains the failure?

    http://ww1.microchip.com/downloads/en/DeviceDoc/80151J.pdf
    ..
     
  5. Quack

    Quack Guest

    Microchip has an errata sheet for early silicon of the "a" parts. Maybe
    Thanks for that, but doubt it :(.

    I have since realised that i am using PICBASIC not PICBASIC PRO, and
    that doesnt necesarily support the '628A.

    But Picbasic is the compiler, not the programmer - the programmer
    should still program the chip, even if the code will operate
    unexpectadly.

    Or am i wrong about this ? something compiled for a '628 would not
    even get programmed onto a '628a ? is there some kind of header or
    identifier that the chip would reject when in programming mode ? i
    doubt that too .. :(

    It fails on the very first byte in the programming process, so it
    seems to be a hardware incompatibility. I cant find anything that says
    specifically the 628a programs different to the 628.

    My programmer sux, looking for a new one..

    Anyone use 628a's ? - what programmer do you use ?

    Thanks!

    Alex.
     
  6. The 628a is supposed to be upwardly compatible with the 628, and
    it ought to program the same. If you would actually read the reference
    I gave you, you'll notice that early silicon for the 628a had problems
    that changed the programming requirements.
     
  7. Quack

    Quack Guest

    I did read that in there, but i dont understand the implications.

    I am completely not sure what my hardware supports, i got it without
    documentation, its worked with everything i have tried (16f877's,
    628's, f84's and a few others) and i use it with PicAllW.indows.
    (kit96 settings)

    Surely picallw should work ? it says it supports 628a. So must be
    hardware ?

    If its hardware, then do you know of a programmer that will take the
    altered programming requirements into consideration and work with
    these buggers ? :)

    Thanks again!

    Alex.
     
Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day

-