Image Map Image Map
Page 10 of 11 FirstFirst ... 67891011 LastLast
Results 91 to 100 of 106

Thread: How to burn 2708s

  1. #91
    Join Date
    Jan 2007
    Location
    Pacific Northwest, USA
    Posts
    25,658
    Blog Entries
    20

    Default

    Well, a second 2708 would pretty much settle the question, wouldn't it?

  2. #92
    Join Date
    Jan 2011
    Location
    Vancouver, BC
    Posts
    2,580
    Blog Entries
    1

    Default

    I wouldn' be surprised if the programmer, the computer, or both, had issues. They are 40 years old, of course. I have noticed that the programmer doesn't always 'see' the EPROM correctly. Sometimes it'll show weird data when I do the erase test or dump contents... like all 00s or something. Then I reseat in the ZIF socket and it goes back to reading whatever should be there. It did pass that erase test once and even dumped all FFs into system RAM when I ran the transfer command. But as soon as I flipped on the high voltage switch.. it went right back to 0130-013F being all F0. I don't know how EPROM'S fail but I am suspicious of how the affected addresses are in a neat, tidy sequence like that. It makes me wonder if the programmer is reading that area wrong for some reason. When I ran a burn, mistakenly thinking all was good, it did change the first digit of the contents for each address in the affected area to what it was supposed to be. But it left the 0 there. So what should be 86 was showing as 80. I don't know. Just felt like, for EPROM corruption, it was a bit too orderly.

  3. #93
    Join Date
    Jan 2011
    Location
    Vancouver, BC
    Posts
    2,580
    Blog Entries
    1

    Default

    Okay so I got a second 2708 in today and made an attempt with it. I put it in the eraser for '9' (whatever that represents.. I'm assuming minutes). I then put it in the programmer and did an erase test. It all came back FFs. So I loaded in my code and attempted a write. No luck. It seemed like for each address, it was writing the right digit, but not the left. ie: 60 becomes 30, 2D becomes 7D, etc...

    I'm wondering if I didn't let the EPROM erase long enough.. or if we actually have a problem of some kind with the programmer. I've noticed it will sometimes say everything's all clear until you put it in high voltage mode.

  4. #94

    Default

    There is probably something wrong with your programmer. FF is the erased state. 0 is the programmed state. Even an open socket won't program as 0. It is because there is a problem with the buffer on your programmer. It is driving 0's for the 4 LSBs. That seems quite clear to me.
    Dwight

  5. #95
    Join Date
    Jan 2011
    Location
    Vancouver, BC
    Posts
    2,580
    Blog Entries
    1

    Default

    I decided to try it with some known good ICs swapped onto the programmer. Thank goodness the original owner used sockets. After a 30 min erase, the burner board only found one hard error.. a 00ff. I decided to try it again.. same deal. So then I thought I'd see how it programmed, just to see what it produced on the 'good' addresses. Except for 00ff, it wrote everything correctly. I then shut the machine off, turned it on, and read from the chip again. Again, everything perfect but 00ff. Could we still be dealing with a board malfunction here?
    Last edited by falter; Yesterday at 11:18 PM.

  6. #96

    Default

    That sounds more like a software issue. It may not be waiting long enough after turning the programming voltage off before switching the addresses. Each address would tend to get a little bit of programming that it wasn't suppose to but the last address would get a small hit for every change. Is there a schematic some place I could look at?

    After thinking about it some, it is likely that the 6821 chip is not being initialized before you turn on the programming voltage.
    Try this:
    Do a read of the device with an empty socket and the programming voltage off. Then put the EPROM into the socket and program it.
    If that doesn't work, try replacing the transistor.
    I see the schematic and it is possible that the programming voltage isn't bleeding off fast enough. That might be an issue with CR3 being open.
    Dwight
    Last edited by Dwight Elvey; Today at 06:45 AM.

  7. #97
    Join Date
    Jan 2011
    Location
    Vancouver, BC
    Posts
    2,580
    Blog Entries
    1

    Default

    When you say put the EPROM into the socket and program, you mean, with the computer still on?

    I played around a bit more with it and found it's quite random. On one test, I got all FFs when I dumped to memory, but when I went to write I got zillions of soft and hard errors. On another occasion, I read the contents, found some addresses weren't FF (after erasure), then re-read. Then a different address would show the wrong contents. Tried a write anyway but of course it all came out wrong. I powered off, powered on a few times and reread the chip.. and it is consistent in where it's wrong. I've also noticed sometimes if I moved the chip around in the ZIF socket and then lock it again, I might get a different result..

    How long should it take to properly erase a 2708? I wonder if I'm doing something wrong there also.

  8. #98

    Default

    Hi
    You should only insert or remove the EPROM, when the computer is on, the programming program has started, but waiting for you to tell it to program, such that the program has set up the 6821 and the programming voltage switch it off. You should not turn off or on the computer with the EPROM in the programmers socket!
    Erasure time should be 10 to 15 minutes, maybe longer depending on the UV light source you are using.
    You should check the pin contacts with a meter if you are seeing different results by inserting and removing the EPROM. Especially check that the ground pin is making contact. A fault there can damage your EPROM or damage the programmers circuit, if the programming voltage is applied.
    Dwight

  9. #99
    Join Date
    Jan 2011
    Location
    Vancouver, BC
    Posts
    2,580
    Blog Entries
    1

    Default

    Quote Originally Posted by Dwight Elvey View Post
    Hi
    You should only insert or remove the EPROM, when the computer is on, the programming program has started, but waiting for you to tell it to program, such that the program has set up the 6821 and the programming voltage switch it off. You should not turn off or on the computer with the EPROM in the programmers socket!
    Erasure time should be 10 to 15 minutes, maybe longer depending on the UV light source you are using.
    You should check the pin contacts with a meter if you are seeing different results by inserting and removing the EPROM. Especially check that the ground pin is making contact. A fault there can damage your EPROM or damage the programmers circuit, if the programming voltage is applied.
    Dwight
    Ah.. I thought you had to shut it off when you switched the eprom in or out.. thought it might short the chip or something with power on.

  10. #100
    Join Date
    Jun 2012
    Location
    UK - Worcester
    Posts
    1,637

    Default

    The programming software should disconnect all the power supply voltages from the EPROM when it has finished reading from or writing to it.

    Likewise, when the computer is turned on, the programmer design should be such that voltages are not applied to the EPROM socket.

    This should permit you to insert and remove devices from the ZIF socket without powering down your computer every time!

    When I built a 2708 EPROM programmer many years ago, I included red LEDs to indicate the varying voltages. The rule was to only remove or insert an EPROM when all the LEDs were OFF.

    Dave

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •