VOGONS


First post, by MaxWar

User metadata
Rank Oldbie
Rank
Oldbie

Alright, i set myself on an epic 486 revival journey! I went at my dad's place and retrieved two 486dx33 computers. My dad never gets rid of his old computers 😀 I want to set up a cool machine for myself, and make sure the other one is in good storage.

Both computers boot, i took them apart to clean everything from the pounds of dust and to inspect all the parts. I am not very familiar with AT hardware, so i am posting this here as some of you are very knowledgeable.

One of the computer is the one i used to play on as a kid, so its a special moment for me to take it back from its sleep 😀
It also appears to be the best motherboard + case of the two so i think its the one i will work on.

Heres the case after dusting.
caseview.jpg

Partial motherboard view :
gigaboard.jpg
Here's a link to the mobo's details (GA-486US):
http://stason.org/TULARC/pc/motherboards/G/GI … 6-GA-486US.html

There are a couple things unusual with this machine:

-It has 32mg of 30 pin memory, which is pretty overkill for a
machine of the era, i think i will scavenge some of it for a GUS pnp card, i just need to identify the big sticks :
ram.jpg

-Then the best cool thing with this machine is that its no longer a 486dx33!!!
The beast has been outfitted with an Evergreen overdrive upgrade chip! It can be seen from the motherboard view but heres a closer shot :evergreen.jpg
I really need to find some more infos about this chip, i dont even know how rare it is, and make sure its properly optimized. Any insights on the subject are welcome!

Now to the sad part 🙁
There was a battery leak issue in the past. The leaking battery has been replace with a remote one, but the acid corrosion is still there, see on the picture:
corrode.jpg
My dilemma is what to do about it? : should i try to carefully clean it off, and possibly jump some dodgy traces with wire just to make sure it does not fail in the future? Or is it on the other hand to risky to do so? It looks pretty ugly to me, im actually glad the computer worked well when i test ran the thing. plz advice.

Of course, i still have some other stuff to check and consider about this board, any recommendation or commentary on other aspects of the subject are welcome!

Reply 1 of 71, by luckybob

User metadata
Rank l33t
Rank
l33t

imho that evergreen upgrade is worth more than anything else in that case. If it still works, post the autoexec.bat and config.sys files from the root of the hard drive. If there are any special utilities, that will tell us where/what they are.

as for the corrosion, I'd personally take the board under some water and wash it with soap and water. use and old toothbrush to try to remove all the crud you could. and let it dry before turning it back on!

What does the other board look like?

edit: if you have 32mb of ram in this 486, then all 8 sticks are 4mb sticks. which is the max for 30 pin simms.

It is a mistake to think you can solve any major problems just with potatoes.

Reply 2 of 71, by MaxWar

User metadata
Rank Oldbie
Rank
Oldbie

Thanks for reply! I could try to check on the hard drive that was in there for a special util for the evergreen. Knowing my dad, i doubt it's on the hard drive but i can test as i know it can boot. I will check again but best might be to simply ask him if he has a floppy for it somewhere. He has the insane ability to remember and retrieve stuff like that.

I found a bit of info on that evergreen btw, not exactly like mine, but very close:
http://www.ukcpu.net/Collection/Processors/Ev … een/486/486.asp
I could try some sort of benchmark utility for dos and see how it performs!

So you would go for the thootbrush and wash that green grime?
Sounds good to me, but i wonder if the brushing action will not completely remove trace remnants.
I can always test traces and fix them afterward i guess.

As for the second board, i completely forgot it as i was writting the original post, 🤣 . And well, its a pretty funny board.
opti.jpg

No clear model number on it, although i found this, very close.
http://stason.org/TULARC/pc/motherboards/S/SI … 6-OPTI-486.html
However mine has 2 EDO ram slots that is not on the opti-486 schematic.
I simply searched for OPTI 486 and found this. Schematic map seems to fit pretty well.

Reply 3 of 71, by luckybob

User metadata
Rank l33t
Rank
l33t

neither board really light the world on fire for me. But if I was in your position I would use board #2. if board #1 has 32mb of ram you can put that into board #2 and have that and whatever is in the simm slots. From the picture, it looks like board #2 has a 33mhz processor, so its likely the upgrade will work in this board.

also, salt and vinegar works to remove corrosion. I use old ketchup! And if the traces break with a toothbrush, they probably weren't connected or working.

but check the hard drive before you go washing that board.

It is a mistake to think you can solve any major problems just with potatoes.

Reply 4 of 71, by MaxWar

User metadata
Rank Oldbie
Rank
Oldbie

You may be right about board#2, at first i thought ill of board 2 because it was somewhat noname and the schematics i found rated it as a max 40 mhz, against 50 mhz for the other one. But i guess the system bus speed is what really matters since i will be using the evergreen. I do not know the bus speed for sure for either. Is there a solid standard for DX boards ? Otherwise board specs seem similar, minus corrosion.

Edit: I do have old ketchup!

Reply 5 of 71, by luckybob

User metadata
Rank l33t
Rank
l33t

For the 286-486 era board the bus speed is the same as the cpu speed. EXCEPT for the 2nd wave of 486 processors. 25,33,40,50 were the stock speeds. most people had a 33 or 25. 50's were EXPENSIVE as the board had to run at 50mhz and few could do this reliably. starting with the 486/66 they started using multipliers. usually a multiple of 33 or 40. 66,100,120,133 were the more common speeds. the 486/66 was a HUGE seller. normally the board had to support the chip running on a multiplier, but they got around this with the overdrive processors. that evergreen chip is essentially an overdrive processor. And if yours really runs at 133mhz, thats a rare beast. treat that upgrade chip with kid gloves!

It is a mistake to think you can solve any major problems just with potatoes.

Reply 7 of 71, by Tetrium

User metadata
Rank l33t++
Rank
l33t++
MaxWar wrote:

You may be right about board#2, at first i thought ill of board 2 because it was somewhat noname and the schematics i found rated it as a max 40 mhz, against 50 mhz for the other one. But i guess the system bus speed is what really matters since i will be using the evergreen. I do not know the bus speed for sure for either. Is there a solid standard for DX boards ? Otherwise board specs seem similar, minus corrosion.

Edit: I do have old ketchup!

The 40Mhz/50Mhz thing is often just the FSB. Using a clock doubled or trippled chip should/could work.

Btw...ketchup???

Btw, the 1st board looks as if some of the tiny parts are already missing due to the corrosion. Personally I wouldn't even be sure if I wanted to keep it, or strip it for parts

Whats missing in your collections?
My retro rigs (old topic)
Interesting Vogons threads (links to Vogonswiki)
Report spammers here!

Reply 8 of 71, by MaxWar

User metadata
Rank Oldbie
Rank
Oldbie
Tetrium wrote:
The 40Mhz/50Mhz thing is often just the FSB. Using a clock doubled or trippled chip should/could work. […]
Show full quote
MaxWar wrote:

You may be right about board#2, at first i thought ill of board 2 because it was somewhat noname and the schematics i found rated it as a max 40 mhz, against 50 mhz for the other one. But i guess the system bus speed is what really matters since i will be using the evergreen. I do not know the bus speed for sure for either. Is there a solid standard for DX boards ? Otherwise board specs seem similar, minus corrosion.

Edit: I do have old ketchup!

The 40Mhz/50Mhz thing is often just the FSB. Using a clock doubled or trippled chip should/could work.

Btw...ketchup???

Btw, the 1st board looks as if some of the tiny parts are already missing due to the corrosion. Personally I wouldn't even be sure if I wanted to keep it, or strip it for parts

I dont think parts are missing, it just looks like traces are broken. Ill see more clearly when i have cleaned this off. And i am not gonna condemn this board, it is the board i was gaming on when i was 8 years old. It has special nostalgia value, ill get it back on its feet 😀

Reply 9 of 71, by MaxWar

User metadata
Rank Oldbie
Rank
Oldbie

Allright, i checked the content of hard drive of computer1 and the autoexec, config etc. There is no trace of anything relevant to that Evergreen upgrade. I called my dad and he says he might still have documentation for it somewhere 😜 Anyway im making backups of the original drive, just in case.

Other thing is that i tried to boot drdos on motherboard#2 and upon loading himem gave me a PARITY ERROR SYSTEM STOPPED
I booted with a floppy and remed the himem line, now it boots fine from a hd but i will have to look into that issue too. Possibly a bad ram stick ? I checked the parity check in bios and it counts 16mb of ram and posts without issue.

At this point i think i need to do some motherboard cleanup!
Will keep this updated.

Reply 10 of 71, by Amigaz

User metadata
Rank Oldbie
Rank
Oldbie

Running these Evergreen am5x86 thingies in a couple of my rigs..they need no util or similar to work

My retro computer stuff: https://lychee.jjserver.net/#16136303902327

Reply 11 of 71, by MaxWar

User metadata
Rank Oldbie
Rank
Oldbie

Pretty much what i thought about those evergreen, ill have to run a benchmark and see how fast this thing is!

I took the acid damaged board out and now im a bit hesitant as to the best way to clean it. I wonder if i might not be better starting off with water+baking soda or something to neutralise acid in case it gets reactivated by the washing.

Reply 12 of 71, by h-a-l-9000

User metadata
Rank DOSBox Author
Rank
DOSBox Author

I use water, soap and a brush. After that I scratch off the solder resist where traces are miscolored and tin-coat them with the soldering iron. While doing so, some brown crap forms, which is neutralized acid. I scratch that off with a screwdriver and clean off the rest with alcohol. Affected solder joints are also resoldered.

One thing to be aware of is that the acid does not stop at the pins of the parts. It moves inwards and can cause failure, even years later.

What comes out of the battery is actually base. If you want to neutralise, you can try diluted acetic acid or citric acid.

1+1=10

Reply 14 of 71, by MaxWar

User metadata
Rank Oldbie
Rank
Oldbie

Ok i washed the board with old tooth brush :

1. 1/4 ratio vinegar/water solution
2. Soapy water
3. water ( rince)

I also scrapped gently where the corrosion was more stubborn.

See before/after pictures:

before.jpg
after.jpg
Many traces are damaged. Strangely, it also seems like one of the red LED-like capacitors was ripped off the board at some point ? You can still see the two leads sticking out, and when i was cleaning the board, i found a small part of red plastic hanging on one of the lead. i did not find any traces of the part when i was cleaning the case.

I am thinking that this may have something to do with the onboard battery being replaced by a remote one such as this:
P1010065.jpgThis one is also a bit bulged and there are traces of leaks on the edges of plastic wrapping, will need replacement again, when i booted the machine i had weak battery warning and bios was factory clean reset

Reply 15 of 71, by DonutKing

User metadata
Rank Oldbie
Rank
Oldbie

A battery like that killed my first sound card.
I would still MUCH rather use an external battery then one of those terrible surface mount ones but the one I use have a single cell and a hard plastic shell to help protect against leaks. Even so I am very careful to place them where they can't leak on anything.

Good luck with that motherboard, unfortunately I think it might be a lost cause. That's very serious corrosion all over it. I'd say that the copper in between the PCB layers has been eaten away too so there is damage that you won't even see. Guess you'll just have to boot it up and see.

If you are squeamish, don't prod the beach rubble.

Reply 16 of 71, by MaxWar

User metadata
Rank Oldbie
Rank
Oldbie
DonutKing wrote:

A battery like that killed my first sound card.
I would still MUCH rather use an external battery then one of those terrible surface mount ones but the one I use have a single cell and a hard plastic shell to help protect against leaks. Even so I am very careful to place them where they can't leak on anything.

Good luck with that motherboard, unfortunately I think it might be a lost cause. That's very serious corrosion all over it. I'd say that the copper in between the PCB layers has been eaten away too so there is damage that you won't even see. Guess you'll just have to boot it up and see.

Funny that you would post this. I just took a nap then checked the board again, inspected the battery and suddenly i see it all differently.
Might not be the original battery of the board that spilled electrolyte, but the replacement battery in that last picture!
There are signs of leakage on it, just not very obvious (look where the wires stick out). And it was stuck on the side of the PSU with Velcro right above the corroded area.

On a more positive note, i inspected continuity of the traces and none are completely broken 😳
Still, i would not put alot of trust on that board in its actual state.
I will build my machine around the other motherboard, but will not scrap this one yet.

Another thing i am doubting at this point is the validity of the battery as a replacement. On the plastic cover you can see its a 4.5 volt replacement for a 844 type battery, which is an Alkaline non rechargeable battery! Those era boards are more often seen with the 3.6 NICD rechargeable battery. I found a picture of a board like mine with the typical blue round battery. :
s_p_29533_1__24004_zoom.jpg

Yet there is also a header for external battery, might be that this external connector does not charge. Otherwise that would pretty much explain why the alkaline battery failed... I guess i would have to power the board once again and check for voltage on that header. If no voltage, then it should be fine to put a regular battery there but the replacement was 4.5 v, while i have a feeling the original was 3.6v .

Food for thought!

Reply 17 of 71, by luckybob

User metadata
Rank l33t
Rank
l33t

its 4.5v because non-rechargeable cells are 1.5V each. 3 1.5's = 4.5 RECHARGEABLE cells are typically 1.2V each. and total 3.6v. I wouldn't worry so much about the missing capacitor. it is likely just a filter cap for the voltage input. Having a quality power supply makes the caps on the motherboard rather redundant. New power supplies are 1% tolerance and provide excelent clean power. that was not the case back "in the day" Thats why PC power and cooling made money, because they made the best power supplies. they have made them for the XT and up! Most people dont realize that. Sadly once they wer bought out by OCZ the quality has gone down like a rock in a pond. 🙁

anyway, before I go off on a tangent... it looks like the first isa slot is dead. if you look at your pictures, you see the connectors are badly out of place on the first slot but comparing them to slot #2 makes it painfully obvious.

It also won't hurt to give it another scrub. I still see some green and that stuff tends to spread like cancer if left alone.

It is a mistake to think you can solve any major problems just with potatoes.

Reply 18 of 71, by MaxWar

User metadata
Rank Oldbie
Rank
Oldbie

I scrubbed it a lot and there is still remnants of the green stuff, this thing is tough!

What worries me the most about the battery is that the header would try to charge it, i was reading and found some indication that it might not be the case but its still not very clear. Unless a knowledgeable person can confirm me the port will not charge, i will have to test myself.

If it does not charge then i would just use a 3xaaa battery pack and be done with it, otherwise i can use a regular 3.6 v NICD pack.

Edit : btw i checked that missing cap and it would link the CMOS reset jumper with ground, probably a safety feature ?

Reply 19 of 71, by luckybob

User metadata
Rank l33t
Rank
l33t

most boards will charge it. I'd put 3 rechargable aaa's in a holder and forget about it. There is a way to prevent charging and all it takes is a diode and a bit of soldering, but imho its not worth it.

It is a mistake to think you can solve any major problems just with potatoes.