Image Map Image Map
Results 1 to 7 of 7

Thread: Possibly dead Tekram DC-680T cached VLB IDE controller

  1. #1
    Join Date
    Apr 2020
    Location
    Belfast, Northern Ireland
    Posts
    38

    Default Possibly dead Tekram DC-680T cached VLB IDE controller

    I decided tonight to have a go at installed the Tekram DC-680T controller that I picked up from ebay a while back.

    Its possibly dead though as on boot it gives a looping single flash on the LED and an error message "Controller Failure! ERROR CODE = 1"

    Looking at the manual

    http://www.minuszerodegrees.net/manu...0%20DC-690.pdf

    Page 46 would suggest this means 1 of 4 things

    Controller is not functional
    No Cache DRAM is installed
    TAG SRAM test failure
    Parity error on cache DRAM

    Ram is installed (it came with 4x 256kb simms but I have also tried others and tried bank 0 only). What does it means by TAG SRAM test failure or Parity error on cache DRAM? Assuming this means a fault with the memory modules I know at least that the simms I tried from my own collection do work fine so its not that.

    Sounds awful like its controller is not functional. The fact that its reporting an error code makes me think the bios on the card is fine. There is a lot of 74 logic sitting between the ram and the main IC so I suppose something there might have gone but its hard to test. The card is spotless, no evidence of previous work or bad solder joints etc.

    Interestingly if I connect a floppy drive it does the floppy seek at startup but the light on the floppy drive stays on, it sits like this for maybe 5 seconds then the error message appears on screen. The single flashing light is constant from when the machine is powered up.

    Any ideas?

  2. #2
    Join Date
    Apr 2015
    Location
    Austin, Texas
    Posts
    2,268

    Default

    Quote Originally Posted by Glen M View Post
    What does it means by TAG SRAM test failure or Parity error on cache DRAM?
    TAG RAM is a special bit of memory used to hold a cache lookup table. It holds entries for data in the cache and has a bit of circuitry to generate a hit or miss when the CPU is requesting said data. It can be implemented with a discrete memory chip and some 74 series logic, or it can be a completely integrated single package with both the RAM and the logic in one chip. A TAG RAM failure can either be the memory itself is bad, or the circuitry to drive the hit or miss logic for the CPU is bad.

    Parity errors are an early form of memory error detection. An extra bit of data will be added to a memory module, in the case of a 30 pin SIMM, one bit will be added, making the module 9 bits instead of 8. This extra bit is used to calculate odd or even (1 or 0) parity for each byte, and if the value read back has a different parity value than originally written, an error is generated. In early computers, this usually results in an NMI being sent to the CPU and halting the system with a message like "Parity error detected, system halted." Later systems made this an optional BIOS feature.

    In either case, you have some significant memory issues. I would test the 256k memory modules in a motherboard using an early version of Memtest86+.


    The TAG RAM is going to be a lot harder to diagnose, it'd be nice to have a high resolution picture of the card to see where the TAG is at, and if it's removable.

  3. #3
    Join Date
    Apr 2020
    Location
    Belfast, Northern Ireland
    Posts
    38

    Default

    Here is the board

    https://imgur.com/IlX7WRw

    I don't see any memory chips other than those in the simms slots, just a lot of 74 logic so perhaps its implemented in there.

  4. #4
    Join Date
    Dec 2008
    Location
    libtard capital, California
    Posts
    1,123

    Default

    the hyundai hy62256 above eproms are sram.

  5. #5
    Join Date
    Apr 2020
    Location
    Belfast, Northern Ireland
    Posts
    38

    Default

    Quote Originally Posted by dorkbert View Post
    the hyundai hy62256 above eproms are sram.
    Yeah that's sort of obvious now that you mention it. I suppose I could probe it to look for inactive data lines but it'll be hard to do given the location. Might have to just remove the motherboard from the case...

  6. #6
    Join Date
    Apr 2015
    Location
    Austin, Texas
    Posts
    2,268

    Default

    Wow, that must have been an expensive card back when it was new, it has a full blown 80186 on it.

    I think that 64k of SRAM may be for the 80186 from its location near the ROMs. That leaves the question where the TAG RAM is. All of those DIP chips in the center are buffers and flip flops.

    I do see some DIP chips hiding between the RAM slots, I'm wondering if those are it. The SRAMs could be doing double duty though, it'd be interesting to reverse engineer the ROMs to see what's happening.

  7. #7
    Join Date
    Apr 2020
    Location
    Belfast, Northern Ireland
    Posts
    38

    Default

    It is a really nice card and I'd like to get it up and running if possible. Would be nice to have a raid setup in my 486!

    I've also been chatting with someone over at vogons who has a fairly good understanding of how the card works. He says that error code 1 along with the duty cycle of the flashing LED suggests that the card is booting but not seeing the simms and is convinced the fault is either physical damage to the card or a failed chip. I've gone over it with a fine toothcomb and no sign of any damage. Next thing will be to check the data and address lines to see if anything is stuck hi/low. I'm hoping its just a bad logic chip which I can swap out easily.

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
  •