VOGONS


386SX-40 Motherboard Not POSTing

Topic actions

Reply 40 of 101, by appiah4

User metadata
Rank l33t++
Rank
l33t++
quicknick wrote:

On my board the keyboard controller was defective (among other problems). On yours, I see what looks like corrosion on two vias near the KBC (above R43 / D2). Check to see if you have continuity between KBC pins 1 and 39 and the data/clock pins on the keyboard connector.

I will do this tonight, thanks for the heads up!

Any ideas why it might be booting SoooOOoOoOoOOoooo SlooOOoOooOOooowww.w.....?

Retronautics: A digital gallery of my retro computers, hardware and projects.

Reply 41 of 101, by jesolo

User metadata
Rank l33t
Rank
l33t
appiah4 wrote:

Any ideas why it might be booting SoooOOoOoOoOOoooo SlooOOoOooOOooowww.w.....?

You need to get into the CMOS setup. Chances are that your system boot up speed is set to "normal" and not "fast".
At the same time, check your turbo header on your motherboard, make sure it's connected and that your turbo switch is on.

Reply 42 of 101, by appiah4

User metadata
Rank l33t++
Rank
l33t++
jesolo wrote:
appiah4 wrote:

Any ideas why it might be booting SoooOOoOoOoOOoooo SlooOOoOooOOooowww.w.....?

You need to get into the CMOS setup. Chances are that your system boot up speed is set to "normal" and not "fast".
At the same time, check your turbo header on your motherboard, make sure it's connected and that your turbo switch is on.

Umm.. The case I am currently using for test has no turbo button so the turbo switch is open.. *Facepalm* I will short it with a shunt tonight. That should probably fix the SLoowoOWooWow issue but then I need to check the keyboard controller traces.

Progress is so much better than frustration 😁

Retronautics: A digital gallery of my retro computers, hardware and projects.

Reply 43 of 101, by Murugan

User metadata
Rank Oldbie
Rank
Oldbie
appiah4 wrote:
quicknick wrote:

On my board the keyboard controller was defective (among other problems). On yours, I see what looks like corrosion on two vias near the KBC (above R43 / D2). Check to see if you have continuity between KBC pins 1 and 39 and the data/clock pins on the keyboard connector.

I will do this tonight, thanks for the heads up!

Any ideas why it might be booting SoooOOoOoOoOOoooo SlooOOoOooOOooowww.w.....?

I kinda have the same problem. Keyboard lights briefly flash when I power on but after that 'no keyboard' error. The board sustained a lot of damage from the leaking battery when I got it.
Will test continuity ASAP. Thx for the tip 😀

My retro collection: too much...

Reply 44 of 101, by appiah4

User metadata
Rank l33t++
Rank
l33t++

With Turbo header closed it takes roughly 1 minute to complete memory test and report CMOS and keyboard errors so that is twice as fast as before but still terribly slow.. Hopefully it will be better once I can get into BIOS and save some CMOS settings..

Retronautics: A digital gallery of my retro computers, hardware and projects.

Reply 45 of 101, by appiah4

User metadata
Rank l33t++
Rank
l33t++
appiah4 wrote:
quicknick wrote:

On my board the keyboard controller was defective (among other problems). On yours, I see what looks like corrosion on two vias near the KBC (above R43 / D2). Check to see if you have continuity between KBC pins 1 and 39 and the data/clock pins on the keyboard connector.

I will do this tonight, thanks for the heads up!

Any ideas why it might be booting SoooOOoOoOoOOoooo SlooOOoOooOOooowww.w.....?

I have checked pins 1 and 2 on the keyboard connector and they are indeed connected to pins 1 and 39 on the controller chip.

Out of curiosity I checked the others, and I found no continuity between pin 5 of the keyboard port and any of the 5V pins on the power connector, so I decided the issue had to be there. See here, once again:

Unknown-OPTi-82-C291-Battery-Damage.jpg

It is connected to the Component to the left of it, which is barely visible in this photo, but it is marked as F1 - so I assume it is a fuse.

There is no continuity between the two legs of this fuse so I suppose it is busted. I would replace it but a) it's a hassle b) I hav eno idea about the rating..

So Vogons, either tell me what the rating of a fuse on a keyboard port 5v rail should be and advise me on why it is wise to replace it, or just tell me to add a jumper wire between the two legs under the board and bypass it 😀 I'm thinking it will work afterwards!

Last edited by appiah4 on 2019-06-28, 06:45. Edited 2 times in total.

Retronautics: A digital gallery of my retro computers, hardware and projects.

Reply 46 of 101, by TheMobRules

User metadata
Rank Oldbie
Rank
Oldbie
appiah4 wrote:

So Vogons, either tell me what the rating of a fuse on a keyboard port 5v rail should be and advise me on why it is wise to replace it, or just tell me to add a jumper wire between the two legs under the board and bypass it I'm thinking it will work afterwards!

On my ASUS VL/I-486SV2GX4 the fuse is 125V 3A:

kbfuse.jpg
Filename
kbfuse.jpg
File size
107.93 KiB
Views
1039 views
File comment
Keyboard fuse
File license
Fair use/fair dealing exception

I believe the fuse is there to protect the motherboard from surges that may occur when hot-plugging the keyboard. I suppose you can use a jumper wire to test if that fixes your problem, but be sure to replace it if possible.

Reply 47 of 101, by root42

User metadata
Rank l33t
Rank
l33t

So close to having this board running! 😀 great job!

YouTube and Bonus
80486DX@33 MHz, 16 MiB RAM, Tseng ET4000 1 MiB, SnarkBarker & GUSar Lite, PC MIDI Card+X2+SC55+MT32, OSSC

Reply 48 of 101, by Cyrix200+

User metadata
Rank Oldbie
Rank
Oldbie

Seeing this (almost) success story really want me to fix my own 386 motherboard. Put it away frustrated a while back after making some progress, but will have to get it out again.

1982 to 2001

Reply 49 of 101, by HanJammer

User metadata
Rank Oldbie
Rank
Oldbie

What you did to make this board almost running? I have 386DX40 motherboard that also stops on POST code 13...

New items (October/November 2022) -> My Items for Sale
I8v8PGb.jpg

Reply 50 of 101, by arteq

User metadata
Rank Member
Rank
Member

I have two mobos with stop POST with code 1312.

I have checked memory ( 8 modules and 4 modules ). But with same results. Even without memory modules post code remian 1312.

I have noticed that without keyboard bios POST process takes very long time, but also stop on 1312. I have replaced keyboard bios but no difference.

Any idea what could be the issue ?

Some items for sale 😀
https://docs.google.com/spreadsheets/d/1YNG8R … dit?usp=sharing

Reply 51 of 101, by HanJammer

User metadata
Rank Oldbie
Rank
Oldbie

Same as above + I replaced KBC (8042) with another - working one and it didn't changed anything too...

New items (October/November 2022) -> My Items for Sale
I8v8PGb.jpg

Reply 52 of 101, by appiah4

User metadata
Rank l33t++
Rank
l33t++

Post code 13/12 is not keyboard related it is a memory initialization error; the system can not find usable RAM in Bank 0. Check your socket connectors, traces under the Bank 0 memory sockets (possibly requires desoldering them), traces between the Bank 0 memory socket and cpu/ chipset for oxidation and battery damage, or as in my case make sure you did not populate the wrong bank (Bank 0 was on the board age in my case, I had never seen this before) or the correct bank with an unusable configuration of RAM (some boards with 8 slots only accept 4x256K 30-pin FPM RAM for 1MB in Bank 0 from what I gather).

Retronautics: A digital gallery of my retro computers, hardware and projects.

Reply 54 of 101, by appiah4

User metadata
Rank l33t++
Rank
l33t++

Progress report:

I bridged the 5V leg of the keyboard connector socket to the diode right after the Fuse, and now the socket has connectivity with the 5V Rail. Yay.

OPTI-291-WB-Motherboard-Repair-04.jpg

Time to boot.. So I set it all up and push the power button..

And there is immediately a change, the Num/Caps/Scroll Lock lights on the keyboard light up as soon as the system turns on..

The memory count is done, and the Debug card continues past the point where it hanged int he past! My heart races.. and then sinks, as I again get two beeps and CMOS checksum, CMOS display type and Keyboard errors.

There is a difference though - even though I can't proceed to the BIOS, the system responds to any key press with two beeps! This is new. Also, we have a new debug card code:

OPTI-291-WB-Motherboard-Repair-05.jpg

So.. Any further ideas? All connections to the keyboard socket are fine, so that means the keyboard controller is busted somehow?

Retronautics: A digital gallery of my retro computers, hardware and projects.

Reply 55 of 101, by appiah4

User metadata
Rank l33t++
Rank
l33t++

Well.. If nobody has any further ideas, I will probably put this away for the moment and proceed with my 386SX-25 motherboard without cache 😒

Retronautics: A digital gallery of my retro computers, hardware and projects.

Reply 56 of 101, by jesolo

User metadata
Rank l33t
Rank
l33t
appiah4 wrote:
Progress report: […]
Show full quote

Progress report:

I bridged the 5V leg of the keyboard connector socket to the diode right after the Fuse, and now the socket has connectivity with the 5V Rail. Yay.

OPTI-291-WB-Motherboard-Repair-04.jpg

Time to boot.. So I set it all up and push the power button..

And there is immediately a change, the Num/Caps/Scroll Lock lights on the keyboard light up as soon as the system turns on..

The memory count is done, and the Debug card continues past the point where it hanged int he past! My heart races.. and then sinks, as I again get two beeps and CMOS checksum, CMOS display type and Keyboard errors.

There is a difference though - even though I can't proceed to the BIOS, the system responds to any key press with two beeps! This is new. Also, we have a new debug card code:

OPTI-291-WB-Motherboard-Repair-05.jpg

So.. Any further ideas? All connections to the keyboard socket are fine, so that means the keyboard controller is busted somehow?

What happens if you just unplug the Debug card and, let it count down and press "DEL" to enter the BIOS? You sure your keyboard isn't perhaps busted and is properly plugged in?
The CMOS checksum & CMOS display type errors are common where the CMOS has lost all its settings due to no battery backup (i.e., the battery no longer has any charge to hold the settings).
Try connecting an external battery and see if there is any change.
Also, check that the Display type jumper on your motherboard is correctly jumpered for "Color" (if you currently have a VGA card plugged into the system) or "Mono" if you have a monochrome display card plugged in.

Reply 57 of 101, by Murugan

User metadata
Rank Oldbie
Rank
Oldbie

This makes me want to break out my problem board and try to fix it. Will have to do it after the unboxing is done in my house :p
@Appiah: do your LED's stay on when the board is booting?

My retro collection: too much...

Reply 58 of 101, by appiah4

User metadata
Rank l33t++
Rank
l33t++
jesolo wrote:
What happens if you just unplug the Debug card and, let it count down and press "DEL" to enter the BIOS? You sure your keyboard […]
Show full quote

What happens if you just unplug the Debug card and, let it count down and press "DEL" to enter the BIOS? You sure your keyboard isn't perhaps busted and is properly plugged in?
The CMOS checksum & CMOS display type errors are common where the CMOS has lost all its settings due to no battery backup (i.e., the battery no longer has any charge to hold the settings).
Try connecting an external battery and see if there is any change.
Also, check that the Display type jumper on your motherboard is correctly jumpered for "Color" (if you currently have a VGA card plugged into the system) or "Mono" if you have a monochrome display card plugged in.

To address these in order;
1. The keyboard is plugged in fine - all three Lock lights do light up at first boot, and the PC actually responds to keypresses (more below). It also works with all my other AT motherboards.
2. CMOS errors are not my concern but I can not get into BIOS; when I get the Keyboard error, pressing F1 or Del does nothing - but the system DOES register the keypress, it immediately responds with two long Beeps. So the keyboard does get the 5V power, and the signals from the keyboard do go to the controller chip. In other words, the port is wired to KClock, KData and +5V; I can't seem to understand where the issue is - possibly the keyboard controller is busted or maybe some other pin on the KBC is not making contact with the chipset/CPU? Any particular connections I should check?
3. I have tried it with and without an external battery that I know is good because it works in my other 386SX board
4. I do not know which jumper does this on my board, feel free to speculate and tell me what to try 😀

Unknown-OPTi-82-C291.jpg

Murugan wrote:

This makes me want to break out my problem board and try to fix it. Will have to do it after the unboxing is done in my house :p
@Appiah: do your LED's stay on when the board is booting?

The case Power Led is always on, the Lock Leds on the keyboard light up only briefly at first boot, after that pressing Num/Caps/Scroll Lock does not result in the Leds lighting up, only two long beeps from the PC Speaker.

I have also managed to snap off the ends of one of my multimeter terminals, so I will be waiting for a replacement for a few days. In the meantime I removed the botch wire bypassing the fuse and ordered a 3A 250V through hole fuse to replace it. If I am gonna give up on it I may as well replace the components that I know are failed, no?

Retronautics: A digital gallery of my retro computers, hardware and projects.