VOGONS


First post, by BoraxMan

User metadata
Rank Newbie
Rank
Newbie

About 15 years ago I found a Headland Technology 286 system on the street, which I had stored in my parents garage. The machine has 4M of RAM, a 1.2M drive and a 43M hard drive and VGA graphics card. Recently, I took it home to get it working, and have had little luck with the mainboard.

Attached is a photo of the mainboard. I have tested independently the hard drive and graphics card, and they do work. When I first started, it went through the memory check, but oddly it would either get stuck part way through, or pause, then continue. After getting into the CMOS setup, I noted that the time displayed in the setup would move forward a few seconds, then reset to midnight. After a reboot or two, the machine would then report "CMOS INOPERATIONAL. SYSTEM HALTED" almost immediately on power up. After that, every time I started it up, I would get that message almost immediately. I removed the barrel battery, as it had started leaking, but it made no difference. There is a small amount of corrosion near where the battery sat, which may or may not account for the problem.

I have also removed and reseated the two BIOS chips, as well as the RAM. I don't believe the absence of the battery should result in the machine not starting, just not retaining CMOS settings.

If I left it off and disconnected from power for at least a few days, I could get it working, albeit briefly. Recently, I had it unplugged for about a month, and it was able to run long enough for me to configure the floppy disk drive and boot DOS from it. It ran for about 30 minutes or so, while I tried to get the hard drive set up working, then back to "CMOS INOPERATIONAL. SYSTEM HALTED". It seems the longer it is left off and disconnected, the longer it will run?

A photo of the board is attached. I suspect the problem may be the real time clock, wihch may be the object labelled YCL 9024 between two ISA slots?

I've reached the limit of my troubleshooting abilities here, and looking for some help as to where to go from there.

Thank you,

Attachments

Reply 1 of 3, by jakethompson1

User metadata
Rank Oldbie
Rank
Oldbie

Funny that you mention finding it in the street. When I was middle school aged I found a 486 on the side of the road. Ended up putting Win95 on it and giving it away to a neighborhood kid who had no computer at home.

Maybe this thing has some kind of sanity check where it's noticing the CMOS not retaining the settings? I can't explain why leaving it off helps, unless a dead CMOS battery is treated different than corrupt settings.

Is that not a ready-made external battery header to the right of the keyboard port, though? Mine has one of the four pins missing as a key pin, though. You can get an external 3xAA battery holder that either has a connector already on it, or you can retrofit one. In the meantime maybe you can rig up something to supply 3-5 V to that header? I'm guessing the pin closest to the kbd connector is positive due to the white mark, and the pin farthest away is negative. Careful that it might try to charge the battery though. Do you have a multimeter to test the voltage either of the two contacts where you removed the rechargeable battery, plus that external battery header?

Reply 3 of 3, by BoraxMan

User metadata
Rank Newbie
Rank
Newbie
jakethompson1 wrote on 2020-07-12, 04:33:

Oh, searching for the error I happened upon this: (Solved) Bios POST error message: CMOS INOPERATIONAL. SYSTEM HALTED.

I did see that threat before when I first came across the problem, and I do have a very similar motherboard! He replaced the HD146818 chip with a Dallas RTC code DS12887, but then ran into similar problem that I have, where the POST memory count may pause or freeze, and then back the same CMOS Inoperational error. It seems the poster there didn't resolve the problem, even after replacing the HD146818

I cannot see that chip on mine, so if I was to replace something, I'm not sure what. It seems likely that my issue may be relating to something holding charge.

By the way, a friend also found a Pentium 150 on the side of the road too, complete with 64M ram and a 3.something G hard disk drive and sound card, which I now have. Some people leave out working computers. Unfortunately, they are harder to find now.