VOGONS


First post, by Mezmaron

User metadata
Rank Newbie
Rank
Newbie

So I got the itch recently to throw together an old system. After digging out multiple motherboards and testing over the past 2 weeks with no success, I thought I came across one that was finally going to work.

It's an old Gateway system that I actually had in use at least 12 years ago. It's Gateway designation is "ATXSTF FED Peformance 1000", with what I guess is a Gateway "Fedora" motherboard, which is more or less an Intel 815 motherboard with a Pentium III 1ghz CPU in it. It worked fine years ago before it was put away for storage. It was a little finicky at first, but I eventually got it to fire up and POST, so after throwing a hard drive in it, I went about installing Windows 98SE.

The install went without a hitch, though a few times here and there when I would shut it down and power it back up, I'd have to hit the power button several times. I got the basic drivers installed, including the Intel motherboard drivers and on-board video, and when I got home from work today I planned on finishing it up, installing the unofficial SP2 for Win98SE, along with trying to get an AGP card working in it.

But it was not to be. It sat all night plugged in but powered off, and upon pressing the power button today, I got absolutely nothing. After pulling the plug, holding the power button in for a while, and then plugging it back in, I finally at least got the power button light to come up orange, and the CPU fan slightly jerks, but that's it. (The power button is green in normal operation.)

I've tried everything I know of; Held the power button down longer powered off and on, pulled the CMOS battery and did the same thing (brand new 2032 battery, by the way), different RAM sticks, pulled all of the PCI cards, hard drive and CD-ROM, pulled the CPU and put it back in, booted with the setup jumper on, etc., nothing. I have 2 other power supplies that I tried, one of which did nothing, and the other actually made the CPU fan come on, but still with the orange light and no POST/video.

All of the caps on the board and inside the power supply look fine. I realize that doesn't always mean anything. It also wasn't kept in the best of conditions over those past 12 or so years, sitting in a non-heated room and sitting through many winters. The board doesn't have any rust on it, and I cleaned any old dust off of the board thoroughly.

It's very frustrating to think that out of 5 or 6 boards I tried, not a single one of them work. I realize that no matter what it looks like, going through temperature fluctuations and such over that period of time likely did it in. The main thing I'm curious about is, what exactly goes bad in these things, when there are no visible signs of physical damage/corrosion/etc.? Anyone know of any other tricks to potentially zap it back to life?

Reply 1 of 5, by treeman

User metadata
Rank Oldbie
Rank
Oldbie

I would be looking at the easiest thing first which is psu, try a different one or at least measure the output on a meter and test the rails on the motherboard once you confirm psu gives good power

Reply 2 of 5, by rasz_pl

User metadata
Rank l33t
Rank
l33t

classic bad caps symptoms, big ones on the motherboard near CPU socket and probably also psu

https://github.com/raszpl/FIC-486-GAC-2-Cache-Module for AT&T Globalyst
https://github.com/raszpl/386RC-16 memory board
https://github.com/raszpl/440BX Reference Design adapted to Kicad
https://github.com/raszpl/Zenith_ZBIOS MFM-300 Monitor

Reply 3 of 5, by treeman

User metadata
Rank Oldbie
Rank
Oldbie

everytime I had bad caps they looked perfectly fine to the eye but they were the problem.

Once I inspected a motherboard and looked perfect, as soon as it powered on white smoke came from like 3 caps around the cpu area.

Reply 4 of 5, by Mezmaron

User metadata
Rank Newbie
Rank
Newbie

I suppose I kind of knew that it was likely the caps, but didn't want to accept it. In my past experiences with bad caps, they were always visibly bulged. However, I realize that the electrolyte drying up on the inside happens, especially with the conditions this motherboard was exposed to.

After making my post, I realized there was another system I didn't try. I had an old HP Vectra VL400 PIII/733 desktop that I didn't try firing up. Thankfully it didn't give me any issues, so that's what I'm going forward with.

I may or may not try replacing the caps on the other board. The boards I've replaced caps on in the past were total nightmares with the pasty lead-free solder.

Reply 5 of 5, by treeman

User metadata
Rank Oldbie
Rank
Oldbie

It all depends on the tools and right technique.
You need at least a hand solder pump, I removed capacitors and even through hole chips using a iron and a hand pump. with lead free solder you really need flux and mix in new lead solder with the old solder and be really quick to jump on it with the pump while its hot.

Ofcourse a vacuum desolder gun with a station is a mercedes after learning this with a hand pump.

It is important to use a little or a little more flux so all the solder mixes in properly, it may look mixed from the top of the joint but you can't see in the middle, so if there is still stuck solder add more flux and new solder on top and suck again