Image Map Image Map
Page 3 of 5 FirstFirst 12345 LastLast
Results 21 to 30 of 42

Thread: 5150 Not Passs Mem Refresh Test

  1. #21
    Join Date
    Jun 2012
    Location
    Livermore, CA
    Posts
    322

    Default

    I plugged in Ruud's Rom again and had some progress.
    See attached. I can now get passed the 2k ram test.
    I am not sure what the ram failure means but it looks about the same as the supersoft ram test results.
    So far I have replaced 3 of the 9 bank zero ram chips. Should I keep going pulling/replacing random chips until I can pass the tests? (checking continuity along the way)
    Thanks
    Attached Images Attached Images
    Atari Falcon030, Atari MSTE, Atari PS3000, STACY 4, 4160STE-CosmosEX, Atari PC1, 400/800/1200XL,800XL,800XE,130XE, XEGS, Apple IIGS, TRS-80 4P, TI 99/4A, Co-Co 3, Amiga 500, C64/128D

  2. #22

    Default

    What does the SS diags say, Can you post a pic, From the pic you posted it appears there are 8 bad ram chips ??

  3. #23
    Join Date
    Jun 2012
    Location
    Livermore, CA
    Posts
    322

    Default

    SS pic below. I think it shows the same info for bad bits?
    Attached Images Attached Images
    Atari Falcon030, Atari MSTE, Atari PS3000, STACY 4, 4160STE-CosmosEX, Atari PC1, 400/800/1200XL,800XL,800XE,130XE, XEGS, Apple IIGS, TRS-80 4P, TI 99/4A, Co-Co 3, Amiga 500, C64/128D

  4. #24

    Default

    The output has changed since post #8, I would try replacing the ram chips in bank 1 and test again using the SS diags.

  5. #25

    Default

    Desoldering thru hole parts from multi layer boards can be difficult. Here is how I do it.
    First, I clip the leads close to the package, so I have a small piece of pin, I can grab.
    I put the board in a small table vice so I have easy access to both sides.
    Using a clean tinned tip on the iron, I apply fresh solder to the bottom of the board.
    I hold small tweezers next to the IC pin, left in the board.
    When I see it wiggle, I grab it with the tweezers and pull it out quickly before the heat is lost to the tweezers.
    When all the pins are removed, I use a hand held solder-pul-it to remove the solder. Again with the iron on one side and the sucker on the other.
    Any hole that doesn't clear I refill with fresh solder before trying again. I don't use one of those useless blue small suckers.
    It is a timing thing. You want to lift the iron at the same time as you trigger the sucker.
    I don't recommend using the combination sucking irons unless you know what your doing. These also tend to be difficult to remain tinned making it easier to damage traces.
    Still, when removing solder from holes, connected to grounds, with the board in a vice it is easy to heat from both sides and used the sucking iron on one side.
    Dwight

  6. #26
    Join Date
    Jun 2012
    Location
    Livermore, CA
    Posts
    322

    Default

    I have been removing and and adding socketed ram chips to the first 16K block.
    Each time I add one. I run SuperSoft. I have used many different ram plastic chips tubes I have. With each, I still get the same results as I show in post #23.
    I check continuity from U37 to U45 for all pins each time. All seems good. There is no continuity for pins 2 and 14 between any chips. I see the same thing in the other banks, so I'll assume that is normal.
    I still have 4 chips to replace.
    Thoughts?
    Also what is the best way to remove flux after soldering. ISO and a toothbrush just seem to move it around. I try to use just a bit on each hole but there is always some leftovers.
    Again thanks for all the help. I'm leaning, and have learned a lot through all these posts, and threads.
    Atari Falcon030, Atari MSTE, Atari PS3000, STACY 4, 4160STE-CosmosEX, Atari PC1, 400/800/1200XL,800XL,800XE,130XE, XEGS, Apple IIGS, TRS-80 4P, TI 99/4A, Co-Co 3, Amiga 500, C64/128D

  7. #27

    Default

    It might not be the RAMs fault. Addressing DRAM is not like static, where the address lines sit static until the data is written or read. DRAM requires the address to be provided in two chucks on the same sets of address lines. When ever you see all the bits failing it is likely an addressing problem. Places to look are things like the address mux.
    Then, it could be a problem in the test program as well.
    Dwight

  8. #28
    Join Date
    Jun 2012
    Location
    Livermore, CA
    Posts
    322

    Default

    Looking at the RUUD test above. MDA + Mem failed. Does this suggest that the 8283 is bad? Or 8237? Or is it another chip?
    Atari Falcon030, Atari MSTE, Atari PS3000, STACY 4, 4160STE-CosmosEX, Atari PC1, 400/800/1200XL,800XL,800XE,130XE, XEGS, Apple IIGS, TRS-80 4P, TI 99/4A, Co-Co 3, Amiga 500, C64/128D

  9. #29
    Join Date
    May 2006
    Location
    Melbourne, Australia
    Posts
    6,290

    Default

    Quote Originally Posted by KLund1 View Post
    Looking at the RUUD test above. MDA + Mem failed. Does this suggest that the 8283 is bad? Or 8237? Or is it another chip?
    You have a CGA card installed. Therefore, as expected, the 'CGA + [its video] memory' check is passing. You do not have an MDA card fitted and that is why the 'MDA + [its video] memory' check fails. Expected behaviour. The Supersoft/Landmark diagnostic behaves the same.

  10. #30
    Join Date
    May 2006
    Location
    Melbourne, Australia
    Posts
    6,290

    Default

    Quote Originally Posted by KLund1 View Post
    There is no continuity for pins 2 and 14 between any chips. I see the same thing in the other banks, so I'll assume that is normal.
    That is expected. Pins 2 and 14 on a 4116 chip are the data in and data out pins respectively.

    Also, on a 16KB-64KB motherboard, for each RAM chip, pins 2 and 14 are connected together.

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
  •