VOGONS

Search

Common searches

Zeos 386SX-16 board ROM-BIOS Checksum Error Repair

Topic actions

First post, by GeGnome

Posted on 
User metadata
Rank Newbie
Rank
Newbie

Hi everyone,

New member here. Just a biology guy, but have learned tons by reading the Vogons threads over the years. Two weeks ago, I purchased about 15 or more 286, 386, and 486 boards, most with CPUs, and I've been testing, repairing traces, changing caps, etc, and I have resurrected several so far. Learning a lot and having a lot of fun. This one is a bit different, and was hoping you could help.

It's a Zeos 386SX-16 board with CPU. Both banks are full of 256 chips for a total of 1MB. I'm testing with post card and/or 16-bit VGA card only.

https://www.win3x.org/uh19/motherboards/5796

First, this did have minor battery damage including one bad trace below the battery and a badly corroded IC socket just above it. I replaced the socket, cleaned up the IC legs and jumped the trace on the back side, confirming continuity. I also had a couple of exploding tantalums and shorted power rails, which I traced and addressed with new tantalums (only three have been replaced on this board at this time). So at this time, it powers up but halts.

****

Currently upon boot, I get a good temporary VGA BIOS with all post card LEDs lit shortly after bootup, with AMI BIOS post codes 19 18, but then the "ready" led goes out and I get post code 20 1A, and a halt due to back checksums on the BIOS. At this point, I can go no further (although I have yet to add a floppy controller to try to boot, but I am assuming that this is truly a halt).

I don't really understand how best to interpret the post codes, but can I assume that the steps in the codes I have seen have passed? So are 1B and 1C , or perhaps 21 next?

http://www.bioscentral.com/postcodes/amibios.htm

+ 18 CGA color mode set OK
+ 19 Attempting to pass control to video ROM at C0000h
+ 1A Returned from video ROM
1B Display memory read/write test OK
1C Display memory read/write alternative test OK
1D Video retrace test OK
1E Global equipment byte set for proper video operation
1F Ready to initialize video system
+ 20 Video test OK
21 Video display OK

Anyways, I have reseated and Deoxit treated the BIOS chips to no avail. I just got a GQ-4X EPROM programmer in the mail and some 27C256 EPROMS should be delivered in the afternoon/evening, so once I figure out how to use it, I plan to pull these EPROMs, check for corruption, and if so, write new ones using the correct BIOS images I found here:

Re: 80386 BIOS image collection

But this is assuming that I just have a bad BIOS chip. or two Is there anything else that I should be looking at (besides more bad traces)? I am still new at this, so I could just be looking under the lamp post at the minute.

Oh, I am also preparing a 3.6V NiMH phone battery for the external batt. header, since it does generate 5v+ and I do not see any jumpers on the board or documentation to switch this off.

https://www.win3x.org/uh19/motherboards/5796

Could you guys take a look and let me know your thoughts? Does this look like bad BIOS ROMs or something else?

I will work on this today and will provide updates as I go.

Thanks!

Attachments

  • 20211113_081415.jpg
    Filename
    20211113_081415.jpg
    File size
    1.46 MiB
    Views
    459 views
    File comment
    Relatively minor damage - after cleanup and replacing the IC socket
    File license
    CC-BY-4.0
  • 20211113_081428.jpg
    Filename
    20211113_081428.jpg
    File size
    1.17 MiB
    Views
    459 views
    File comment
    The issue...
    File license
    CC-BY-4.0
  • 20211113_081400.jpg
    Filename
    20211113_081400.jpg
    File size
    1.9 MiB
    Views
    459 views
    File comment
    The board as tested - 386SX-16
    File license
    CC-BY-4.0
  • 20211113_081350.jpg
    Filename
    20211113_081350.jpg
    File size
    1.31 MiB
    Views
    459 views
    File comment
    Final post code at ROM-BIOS checksum error screen
    File license
    CC-BY-4.0
  • 20211113_083216.jpg
    Filename
    20211113_083216.jpg
    File size
    1.49 MiB
    Views
    459 views
    File comment
    First temporary post code during VGA BIOS screen
    File license
    CC-BY-4.0

Reply 1 of 4, by Horun

Posted on 
User metadata
Rank l33t++
Rank
l33t++

Old BIOS chips can go bad so if you can read them with something like a TL866 that may show if one or both are bad. Halting after reading the Video ROM with that checksum error does suggest bad BIOS.
Your board appears to have 1Mb ram, it can have up to 4Mb using 36 1Mbx1 chips (instead of 36 256Kbx1) iirc. My old board is in storage somewhere (I lost it somehow) so hopefully those BIOS images are good and will will help.
Good luck with your board and let us know what you figure out.

Hate posting a reply and then have to edit it because it made no sense 😁 First computer was an IBM 3270 workstation with CGA monitor. Stuff: https://archive.org/details/@horun

Reply 2 of 4, by GeGnome

Posted on 
User metadata
Rank Newbie
Rank
Newbie

Hi Horun, thank you for the tips! Just an update:

I received my EPROMs, learned how to read and write them, and and prepared new odd/even BIOS chips based on the upload above. It went well, and I even learned how to write a 128kb image to a 256 kb EPROM.

Turns out both new and old BIOS chips work...JP21 was missing, and needed to be set at 256Kb (vs 128 Kb). I really think I had already tried that, but perhaps it was the combination with some other missing jumpers, but it now starts to POST and I see the Zeos AMI Bios header.

It now halts before the BIOS string is displayed and goes to a black screen or blinking cursor with post codes 23 and 24:

23 - BIOS vector table initialization
24 - BIOS vector table initialization

Something else is amiss, but this progress. I may check all the RAM chips one at a time in case one is partially malfunctioning. Or perhaps my kbd controller is malfunctioning or has a bad trace, hence stopping before post code 25 (Turbo check of 8042 keyboard controller)?

If anyone has any tips, please let me know! I will try another AMI 8042 chip from another board and will post again if I make progress, if not only for posterity.

Reply 3 of 4, by Horun

Posted on 
User metadata
Rank l33t++
Rank
l33t++

Well you made some progress. Have you tried all jumpers as default per the UH19 no at https://www.ultimateretro.net/motherboard/manual/32409.pdf
make sure the Turbo select jumpers on J32 on 1-2 and then jumper the Turbo Switch jumper on the side of the board.
Also my board did not have J20 jumpered and the docs say to leave alone, maybe yours needs jumpered...

Hate posting a reply and then have to edit it because it made no sense 😁 First computer was an IBM 3270 workstation with CGA monitor. Stuff: https://archive.org/details/@horun

Reply 4 of 4, by GeGnome

Posted on 
User metadata
Rank Newbie
Rank
Newbie

Thanks for the tips! I had this jumper document, and I had previously attempted to get all these in the default settings, but I did not have the turbo jumpers selected correctly.

Unfortunately, those along with the J20 in either configuration, did not change anything. I also tried a known working AMI 386 SX keyboard controller, with no change. I may need to look for another bad trace or two.

However, following your lead, I did get a second "unknown" 386 SX-16 up and running by adding at least 4 missing jumpers. It now fully posts and boots to DOS from a floppy!

I will have to continue digging into this ZEOS board, but I am glad to got another one moved from the "broken" to the "working" box.

Go to top of page Go to top of page
Back to General Old Hardware
Close window
Close window