VOGONS


First post, by kinetix

User metadata
Rank Member
Rank
Member

Greetings to the Vogonists
I need some help with a difficult task.
I have "saved" this motherboard from a recycling center (when I go there I have to check a lot and carefully to find something older, since almost everything is more modern, post-2000 hardware). I couldn't find it on TheRetroWeb, so I contributed it, and the guys identified it as an "Ultima Electronics Corp. 3802". It's a 386. I don't have 386 CPUs now (except an SX without legs,), but I do have a TX486DLC/E-40GA, and I understand that it is a drop in upgrade. Even without being sure it is supported, I want to restore the motherboard (the other 386 I have is a small Foxcon M396F, but I don't have CPUs to improve the poor 386SX it has).
But the condition of this motherboard is BAD.
On the one hand, it received a hit that destroyed an array of resistors, but that is easy to repair.
The bad thing is that the battery had leaked and severely damaged the circuit in its vicinity.
I cleaned it in a general way but on the damaged part I did it in the following steps: vinegar+baking soda, water+soap+toothbrush, soft acid+toothbrush (some crust remained), water+soap+toothbrush, vinegar+baking soda (neutralizing any remnant of the last acid), water +soap+toothbrush , dry well . now you can taste the surface with your own tongue, taste nothing, the Ph is 7, totally neutral,
In the images, how it turned out. The images have been digitally treated to highlight details, contrasts and colors, making it easier to assess damage (those here are scaled down for vogons attach).
You can see that some lines have disappeared, although at the bottom they are not so bad, maybe just tinning is enough.
As you see, this is a work for "necroware" (I'll have to find and raise my inner equivalent )
Now I need to do some unsoldering and maybe more localized cleaning. The layout of the missing veins can be inferred, but to be sure I´ll need references, a "ground truth".

For this reason, if anyone has a sample of this motherboard, I would appreciate it if you could share images of that part, the best possible images, from the 4 sides at the top and one from the bottom, clearly showing that part of the mb.

Ah!! and some clarification on whether the TX486DLC/E-40GA would actually work on it is appreciated.

Attachments

  • 386 front.jpg
    Filename
    386 front.jpg
    File size
    845.82 KiB
    Views
    754 views
    File license
    Public domain
  • 2a.jpg
    Filename
    2a.jpg
    File size
    855.47 KiB
    Views
    754 views
    File license
    Public domain
  • 1a.jpg
    Filename
    1a.jpg
    File size
    625.58 KiB
    Views
    754 views
    File license
    Public domain
  • UPPER.jpg
    Filename
    UPPER.jpg
    File size
    860.52 KiB
    Views
    754 views
    File license
    Public domain
  • BACK.jpg
    Filename
    BACK.jpg
    File size
    535.06 KiB
    Views
    754 views
    File license
    Public domain
Last edited by kinetix on 2024-03-09, 00:52. Edited 3 times in total.

Reply 1 of 11, by Horun

User metadata
Rank l33t++
Rank
l33t++

Can't help on the soldering, other than you need a good solder station, good digi volt meter to check ohms <not continuity> as traces have to be about 1-3 ohms max to work proper iirc.
I am curious why it has a 30Mhz Xtal. Boards using the same/similar chipset and layout (Hot-307, Jetway386WB, TAM33-P1, etc) have one that is 2x the cpu clock so for 33Mhz have 66Mhz xtal.
if all is similar yours with a 30Mhz xtal would run the cpu at 15Mhz as 386 use a divide by 2 on the clock.... just an observation...

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 11, by kinetix

User metadata
Rank Member
Rank
Member

will go tender and slowly. maybe one vein a day. for now will use those MB and other similar as references, as I see they are very similar around that place, some seem the same. thanks
just look this one: https://theretroweb.com/motherboards/s/jetway … -other-revision

Reply 3 of 11, by kinetix

User metadata
Rank Member
Rank
Member

finally completed the repair of the destroyed lines. I am reinserting some components that I removed to facilitate the fix (the most difficult to remove was the external battery connector)
In addition to a couple of lines elsewhere broken as a result of a blow. and that same blow destroyed half of an array of resistors. Since I don't have one with the same number of resistors, I will adapt a smaller one to replace the destroyed ones. I already did that in a floppy drive with one damaged resistor array

Last edited by kinetix on 2024-03-09, 00:53. Edited 1 time in total.

Reply 4 of 11, by kinetix

User metadata
Rank Member
Rank
Member

Today I just "completed" the repair of this motherboard. It is the first one I have done with such a level of damage and it took me a while. some lines had just disintegrated. thank goodness I took enough pics beforehand, plus some found around in the web.
BUT...
When I was measuring everything for the last time before doing the first test, I realized that I had put back the power connector on the motherboard backwards!!!!! ouch!!!
It was thanks to verifying the pins, and the 5V pin position. When I went to connect the PSU I saw that they were on the opposite side, hahaha.
Thank goodness I took that last measurement and my short-term memory jumped right away.
Tomorrow I will unsolder them again. It is something that is a little difficult for me without a desoldering station.
If everything is OK I´ll put some pics

Attachments

  • repair3.jpg
    Filename
    repair3.jpg
    File size
    424.34 KiB
    Views
    425 views
    File license
    Public domain
  • repair2.jpg
    Filename
    repair2.jpg
    File size
    264.33 KiB
    Views
    425 views
    File license
    Public domain
  • repair1.jpg
    Filename
    repair1.jpg
    File size
    330.17 KiB
    Views
    425 views
    File license
    Public domain
Last edited by kinetix on 2024-03-10, 22:18. Edited 2 times in total.

Reply 5 of 11, by kinetix

User metadata
Rank Member
Rank
Member

Well. I finally tested the motherboard today.
Yesterday I corrected the big mistake of having soldered the power connection the other way around, which, thank goodness, I realized. Everything was OK, everything measured correctly, all fixed lines ok, and no shorts.
But it seems that Lady Luck is not on my side.
the motherboard is this one I contributed to theretroweb: https://theretroweb.com/motherboards/18207 . there are several similar ones, with just small variations.
Since I went to a friend's house to use his scope, I was a little lost. As a stand on a glass table, I used the cover of a magazine as support.
The PSU I used is from an Olivetti. These Italians used yellow for the -12V cable and blue for the +12, somewhat confusing as it was the opposite of the standard.
well, connect to the motherbar, without CPU or memory, just to start testing signals, and...
...the PSU would not boot. or it starts and stops.
...???
I thought it would be a low load problem and added an HDD. But then I saw more clearly that the PSU started and stopped cyclically.
I disconnected everything and I decided to check the voltages with just the disk, and the power supply started!
When I measured the connector pins on the motherboard, I got a short GND to -12V!!
I dedicated myself to checking all the pins and contacts, in case any had been bent or a particle of tin had been stuck, or anything else. straighten those that had been bent and in danger or possible contact and try again.
nothing...
but when I look at the cover of the magazine I saw that it still had one of the metal clips, and raised!!! an enormous probability of having made contact with the circuit.

I don't know how -12 volts are used on a motherboard , beyond being in the slots to be used by expansion cards that need it. so I did an experiment. I disconnected the -12 volts, and the power supply started.
I measured the clock signals on the ISA bus and on the CPU socket.
In the ISA pin 20 (CLK) there was between 13 and 14 MHz, and in the socket 80MHz, I forgot to measure on pin 30 (OSC) in the ISA slot. The crystal on the motherboard is 30MHz. The worst thing is that after some tests, on and off, the signal in the socket disappeared, leaving a small 15 KHz signal, or something like that, at 2V and 1V peak to peak. I include images.

several things may have happened
-I am only seeing an original problem with the motherboard, when I picked it up I don't know if it worked before the big damage caused by the battery. but before there was no short.
-The clip touched the circuit and damaged it. Although it was far from where -12V is, at least where I know -12V exists, in the ISA slot...
-The short could be a component that failed when it received power after many years, perhaps a capacitor, although in this case it would have exploded, right?

Any idea what should I try?

It is a shame, after so much time I devote to this motherboard. This is the only 386 motherboard of any quality that I have. the other is a small Foxcon M396F with a 386sx soldered and socket for the coprocessor. I'll have to dig a little deeper into the recycling center to see if I can find another one.
Maybe I'll leave it for a while and go on to restore an Olivetti M24 motherboard that I have in the queue, including making it an 8-bit ISA bus adaptation (I couldn't find its video card or the backplane). or perhaps an XT clone with some historical relevance in my country.

Attachments

  • CPUclk.jpg
    Filename
    CPUclk.jpg
    File size
    130.29 KiB
    Views
    454 views
    File comment
    CPU socket clock signal
    File license
    Public domain
  • ISAclk.jpg
    Filename
    ISAclk.jpg
    File size
    126.46 KiB
    Views
    454 views
    File comment
    ISA pin 20 signal
    File license
    Public domain

Reply 7 of 11, by majestyk

User metadata
Rank Oldbie
Rank
Oldbie

The -12V tantalum capacitor right at the PSU connector can be shorted or one of the 3 tantalums on the left side between the ISA slots.
Shorting when powered on after a long storage time is a typical behavior of tantalums made in that period.

You can remove all 4 of them or one at a time and test when the short disappears.

Reply 9 of 11, by kinetix

User metadata
Rank Member
Rank
Member
majestyk wrote on 2024-03-13, 05:53:

The -12V tantalum capacitor right at the PSU connector can be shorted or one of the 3 tantalums on the left side between the ISA slots.
Shorting when powered on after a long storage time is a typical behavior of tantalums made in that period.

You can remove all 4 of them or one at a time and test when the short disappears.

thank you both of you so much for your answers.
I´ve thought about the one closer to the power connector , but not about the ones between the ISA slots. I will check all caps related to the -12V line. will do it next days.
also the strange behavior of the clck signals

Reply 10 of 11, by CoffeeOne

User metadata
Rank Oldbie
Rank
Oldbie
kinetix wrote on 2023-11-13, 03:44:
Greetings to the Vogonists I need some help with a difficult task. I have "saved" this motherboard from a recycling center (when […]
Show full quote

Greetings to the Vogonists
I need some help with a difficult task.
I have "saved" this motherboard from a recycling center (when I go there I have to check a lot and carefully to find something older, since almost everything is more modern, post-2000 hardware). I couldn't find it on TheRetroWeb, so I contributed it, and the guys identified it as an "Ultima Electronics Corp. 3802". It's a 386. I don't have 386 CPUs now (except an SX without legs,), but I do have a TX486DLC/E-40GA, and I understand that it is a drop in upgrade. Even without being sure it is supported, I want to restore the motherboard (the other 386 I have is a small Foxcon M396F, but I don't have CPUs to improve the poor 386SX it has).
But the condition of this motherboard is BAD.
On the one hand, it received a hit that destroyed an array of resistors, but that is easy to repair.
The bad thing is that the battery had leaked and severely damaged the circuit in its vicinity.
I cleaned it in a general way but on the damaged part I did it in the following steps: vinegar+baking soda, water+soap+toothbrush, soft acid+toothbrush (some crust remained), water+soap+toothbrush, vinegar+baking soda (neutralizing any remnant of the last acid), water +soap+toothbrush , dry well . now you can taste the surface with your own tongue, taste nothing, the Ph is 7, totally neutral,
In the images, how it turned out. The images have been digitally treated to highlight details, contrasts and colors, making it easier to assess damage (those here are scaled down for vogons attach).
You can see that some lines have disappeared, although at the bottom they are not so bad, maybe just tinning is enough.
As you see, this is a work for "necroware" (I'll have to find and raise my inner equivalent )
Now I need to do some unsoldering and maybe more localized cleaning. The layout of the missing veins can be inferred, but to be sure I´ll need references, a "ground truth".

For this reason, if anyone has a sample of this motherboard, I would appreciate it if you could share images of that part, the best possible images, from the 4 sides at the top and one from the bottom, clearly showing that part of the mb.

Ah!! and some clarification on whether the TX486DLC/E-40GA would actually work on it is appreciated.

This board is not in a bad condition, I would say it is in a horrible condition.
As a minimum, you should also remove the ISA connector nearest to the keyboard connector. I believe there is severe corrosion inside.