• Please review our updated Terms and Rules here

Bios search for Gateway 2000 P5-60, and other issues

N

NHVintage

Member
Joined
Oct 1, 2023
Messages
24
Hi All! Been doing a fair amount of vintage computer stuff of late but by no means am I an expert (even if in the mid-90s I was building my own PC's). I was given a Gateway 2000 P5-60 tower. At first things were relatively good, except the Dallas battery/clock had a dead CMOS battery, no surprise. I got a new one and after a bit of adjusting got it booting up, set the date and time. I decided to get the cards out and clean them up as the thing was dusty as heck. I replaced the cards in the same slots, doing each card one at a time. Since then, I've not gotten it to boot to the bios screen.

Sometimes I get a single beep. I put a PCI diagnostic card in, and the boot freezes at a variety of places. I've done the usual work of cleaning out the slots to be sure nothing is bad there. the latest stopping point is at the 13 08 post code... and a noise I've never heard from a booting pc's little speaker, best described as a mid frequency buzz, almost a video-game style noise (if you have an apple II you might know what I mean). I've tried moving the cards around, swapping memory, the only thing I haven't tried swapping out yet is the power supply. I'm guessing the BIOS is in fact on the PC still, since I get these codes (via the 2nd PCI slot), right? if the flash was somehow vaped by me putting the wrong jumpers in the wrong place clearing the CMOS (argh) then I'd get no reaction at all, right?

It's quite frustrating to start the day with a working machine (save I couldn't set the date and BIOS settings), get the bios settings set, and then find that just pulling the cards out to clean them off kills the machine, thus ending the day with a dead machine... But hope springs eternal with your help I can figure this out. I appreciate any suggestions you have. I was thinking of getting the BIOS for the machine and reflashing it, just to be sure its good. Any help is appreciated. Let me know what information I can provide. Thanks!
 
Welcome to the forums. You need to check the power supply to make sure the voltages are good and verify the ripple is good too. Do you have a scope?
 
Thanks for the welcome! I don't have a scope but I do have a multimeter so I'll start with that. The diagnostic card also has lights to indicate the different voltages are present on the slot, which for the most part they are, but I noticed not always on the isa slots one had only +12v, another had that and 5v, etc. I'll dig deeper in the morning. Thanks for the suggestions, will let you know how it goes.
 
I would recommend getting a can of either Deoxit Gold G5 or CRC 2-26. The former works better, but the latter is cheaper and more available at Home Despot/Lowes. You don't need to wipe it dry, you can leave both wet on the cards/RAM. It'll also clean the pins inside the socket when you do.

Spray a bit on the contacts of the RAM sticks and either use your fingers to rub the entire edge connector a few times, or use a microfiber cloth. Do the same with any expansion cards as well (AGP, PCI, ISA, etc.)

The problem you have sounds like it could be a few things, hence all of the cleaning. Marginal RAM, oxidized RAM or addon cards can cause the symptoms you're having. A bad power supply can as well, but I'm thinking its more a connection issue from oxidation.
 
Thanks for the welcome! I don't have a scope but I do have a multimeter so I'll start with that. The diagnostic card also has lights to indicate the different voltages are present on the slot, which for the most part they are, but I noticed not always on the isa slots one had only +12v, another had that and 5v, etc. I'll dig deeper in the morning. Thanks for the suggestions, will let you know how it goes
 
Sorry, been a busy day here. Only looked at this a bit so far, still need to multitest, but now I get a beep and usually the BIOS code (AMIBIOS) stops at 4E. given this is a 1994 machine this means either "Start Second protected mode test" or 'Address Line Test OK". Also, I am getting signal out the video card that the monitor can see, but nothing shows text-wise - just black.

This is promising. I'll get back in the next day or so with multimeter results.
 
GiGaBiTe said:
I would recommend getting a can of either Deoxit Gold G5 or CRC 2-26. The former works better, but the latter is cheaper and more available at Home Despot/Lowes. You don't need to wipe it dry, you can leave both wet on the cards/RAM. It'll also clean the pins inside the socket when you do.

Spray a bit on the contacts of the RAM sticks and either use your fingers to rub the entire edge connector a few times, or use a microfiber cloth. Do the same with any expansion cards as well (AGP, PCI, ISA, etc.)

The problem you have sounds like it could be a few things, hence all of the cleaning. Marginal RAM, oxidized RAM or addon cards can cause the symptoms you're having. A bad power supply can as well, but I'm thinking its more a connection issue from oxidation.
Click to expand...
That sounds reasonable, I have some of the liquid DeOxit courtesy my vintage radio work, but not the spray (I have a can of other stuff and I find 99% IPA works pretty well too). I'll let you know how it goes.
 
Since my last update, I've gotten a new Dallas RTC (actually 2, just in case). While waiting for that, I took the motherboard out of the case and confirmed there were no issues on the bottom, and cleaned out the slots and card interfaces thoroughly. I actually decided to leave it out of the case, on a non-conductive surface, and tried it on another good power supply as well as the original. At this point I'm not getting any responses from the board, no post codes on the diagnostic card; I tested with the keyboard, mouse, and every board and stick of RAM removed and various combinations of them in. At this point, I am weighing if it's worth the work recapping the board. I don't have any evidence that's the issue but it seems to be the thing to do. Any other suggestions?

I've determined that this is the "Batman" model Intel board. https://stason.org/TULARC/pc/motherboards/I/INTEL-CORPORATION-Pentium-PREMIERE-PCI-P5-PCI-BATM.html

I happen to have a Pentium II system looking for a new case, it's based on a Chaintech board. the gateway case doesn't have the universal window on the back where the interfaces are on modern boards so it'd require modifying the solid steel back of the case.

So I guess I have some decisions to make. Thank goodness I got my Epson Equity I+ finished this week and it's working great so I have some positive hobby news there at least. :)
 
Last edited:
If removing it from the case stopped it from doing anything, that sounds more like there are broken solder joints on ICs on the board. You're quickly working into the territory of board level repair. I would suggest getting a desk magnifier with a light, and a pair of fine tweezers and start checking the chipset ASIC first's legs to see if any have become disconnected from the board. Then start moving on to the smaller chips if everything seems OK.

If you find any disconnected legs, you'll need to reflow the solder to reconnect the legs.
 
One thing I was careful to avoid is this blue wire.. I've seen such as this on other boards, is this a repair or just as designed?

One other interesting thing I spotted on this Gateway 2000. Every component - the case and it's serial # info, the cards, etc - has a sticker with the same date of 1/29/94 and the same serial # or order #, except the motherboard. Its got a date of 2/22/95 and a different number. Maybe this isn't the original mobo.
 

Attachments

  • 20231007_080015.jpg
    20231007_080015.jpg
    3 MB · Views: 12
Last edited:
The blue wire is a bodge rework, probably done in the factory. Be sure to not disturb or remove it. You may want to take closeup pictures of what pins on what ICs it connects to, in case it breaks in the future.Actually, you may want to check continuity on the wire to make sure it's not broken internally.

As for the dates being off, while it could be a replaced board, I suspect more that the case was probably sitting in a warehouse waiting for a board. It's not uncommon for different components to have a different date, depending on how supply chain logistics worked at the time.
 
That could be true, but this was when G2000 custom built their machines to individual orders, it was kind of their shtick at the time. So I'm pretty sure the serial number on each component and the case denotes someone's original custom order, can't see all those exact cards and case sitting for a year waiting for the mobo. The cards include a token ring card and serial/scsi, and the machine has multiple hdds and before it went south I spotted dB and other software on it suggesting it was built as a server.

Speaking of this machine, I got out my giant illuminated magnifying glass and went Sherlock Holmes all over both sides, and in the slots, and haven't had any luck finding anything, not bad or rusted components or bad soldering, or cracks in the board or paths. That took awhile! Of course those kind of electrolytics don't usually explode or leak like others, so I'll take out the capacitance tester and see what I find as well as check continuity on that wire. always good to give the board a thorough look anyway so it wasn't a waste of time.

I took a break on it after the exam to work on another project, my retro computer booth. It's getting there.
 

Attachments

  • 20231007_120312.jpg
    20231007_120312.jpg
    1.6 MB · Views: 6
As it turns out the "good" power supply I was trying - an ATX with an AT adapter - must have been a specialty one from one of my newer desktops as it lacks -12V and a couple other voltages. So I'm looking at buying a new PS, but I had a question about them that actually seems to me would be more general interest than just my project so I asked separately. The original PS is showing only 9V on the 12V bus, tho nearly 6V on the 5V bus, so clearly it has issues.

EDIT: By the way, I checked the voltages on the PS previously, early on after getting the machine, and they were fine. Looks like it must have started failing partway through diagnosing. That's my luck I guess.
 
Last edited:
NHVintage said:
As it turns out the "good" power supply I was trying - an ATX with an AT adapter - must have been a specialty one from one of my newer desktops as it lacks -12V and a couple other voltages. So I'm looking at buying a new PS, but I had a question about them that actually seems to me would be more general interest than just my project so I asked separately. The original PS is showing only 9V on the 12V bus, tho nearly 6V on the 5V bus, so clearly it has issues.

EDIT: By the way, I checked the voltages on the PS previously, early on after getting the machine, and they were fine. Looks like it must have started failing partway through diagnosing. That's my luck I guess.
Click to expand...
Was the power supply under load when you made those voltage readings? Many designs will not regulate properly without modest loads on at least +5 and +12.
 
shirsch said:
Was the power supply under load when you made those voltage readings? Many designs will not regulate properly without modest loads on at least +5 and +12.
Click to expand...
No it wasn't, not at either time. But this is news to me that you need a load, as I've measured others without load and they've at least been in the general area (within a volt) of being correct. Thanks for letting me know! I've ordered a new supply, at the least, I need a spare around here anyway.
 
The new PS didn't do anything either. I'm fairly stumped. Think I'll move on to another mobo. There's a recycler nearby where I can get older mobo's at reasonable prices. I should be able to get one that matches the case - and he tests them to POST before selling them.
 
Does your new PSU have a -5v rail? Some motherboards won't boot if the -5v rail is missing, they have circuitry to detect it and keep the board from coming up.

You can fudge the -5v rail by hanging an LM7905 off the -12v rail. The current draw is so low that a linear regulator is fine.
 
I went ahead and replaced the board with an Asus TXP-4. The machine is now operational, with one quirk - the only video card I could use was a PCI NVidia Geforce 6200 board, it wouldn't recognize either a PCI Diamond (ATI) Radeon 9250 or a PCI Riva TNT2 (verified by the Award BIOS one long, three short beep code). Contrast with a Celeron (PIII) variety SBC machine I have with the 440BX chipset, which recognized all three cards. the 6200 is the newer card and I would have thought the 430TX chipset on the ASUS would NOT recognize the 6200 but recognize the older ones, however reverse is true. The TX chipset doesn't have AGP support, but the BX chipset does, so that perhaps is why, though I'm not sure that matters on a PCI card.

All of which is to say: Perhaps its nothing to do with AGP support and the issues with the board and vid cards are related, and I don't have -5 off the rail. Will get my diagnostic card out in the next couple days and check, as it has indicators for all the different voltages on it.

EDIT: Now that the coffee's kicked in, I realize: of course the vid cards have nothing to do with -5V, they're PCI not ISA :D ... I should know better than reply before coffee. Ah well, in any case I have a Voltage Blaster on order as part of my testing suite along with the diagnostic board. And meanwhile I'll live with the Pentium 1 MMX 233 having a better vid board than the Celeron 733 system...
 
Last edited:
Try cleaning the PCI edge connectors on the video cards with Deoxit Gold G5, they tend to get oxide buildup on them and can cause connection issues.
 
You must log in or register to reply here.
Back
Top