i would start by installing a new battery. I usually use a nicd 3.6 v battery pack that I can get for 5$ at local electro store. I install it remotely with long wires. No battery might just be the reason you get this error. I would also check if the old leaking battery had not damaged any traces.

Maybe a faulty ROM chip. It's very easy to replace a ROM with a EPROM, the hardest part will be getting a working copy of the BIOS of your computer

Just now had a single power on where I got something other than the rom checksum error.... saw the VGA BIOS info (copyright cirrus logic 1994 bla bla ... 1024kb) and then it froze.

I can write a couple of new eproms no problem (if I can dig out the software for my old willem writer), but as you say; there's virtually no chance of finding another of these boards.

From my experience 286 BIOS chips are sometimes interchangable between mainboards as back then not so much proprietary crap existed yet.

Also you can try to read the chips and write them into other EPROMS, as possibly the timing of the reader is more relaxed than on the MB.

This is what I got (a single time, maybe 1 out of 20 power-on attempts) yesterday

I guess I can give reading and writing back to new EPROMS a go - I think I have some 27256 chips left over from writing BBC 'B' roms.

Keep in mind that the problem is that your computer can't read reliably the ROM. It may be a problem in your ROMs or a problem with the mainboard, but if you read a faulty ROM in your willem you'll get another invalid BIOS for your computer.

Have you tried searching the model of the board in internet?

Will uniflash work on a 286?
I mean, if you get one of those rare boots maybe you could save the rom to disk.

Zup wrote:

Keep in mind that the problem is that your computer can't read reliably the ROM. It may be a problem in your ROMs or a problem with the mainboard, but if you read a faulty ROM in your willem you'll get another invalid BIOS for your computer.

Have you tried searching the model of the board in internet?

Yep, theres no matches that I can find on stason.org - the nearest are the generic Headland HT12 chipset based AT clones boards, though none match directly (dual BIOS chips, Sims and dips, no of slots etc).

This is it:
download.php?id=8872

> Will uniflash work on a 286?
> I mean, if you get one of those rare boots maybe you could save the rom to disk.

There were no successful boots so far.

> dual BIOS chips
2x 27256 can be converted to 1x 27512 and vice versa

> Sims and dips
might not be that important

> no of slots
irrelevant

http://chukaev.ru54.com/bios.htm .... just found this, it appears to have BIOS dumps for a number of heafland ht12 chipset 286 boards. I think I have enough 27256 chips spare to give a couple a try!

..... and for my next trick I will use a parallel port Willem eprom programmer on any one of my current pc's that doesn't have a parallel port.

🙁

<sigh>

Things are never easy, are they? I guess I need a USB to parallel adapter.

The USB to parallel port things are not really a replacement, unless the Willem designers say so.

h-a-l-9000 wrote:

The USB to parallel port things are not really a replacement, unless the Willem designers say so.

Yes, better get a cheap PCI paralell port card. Those thigs are cheap, and you can select the port address (to match one of the choices in Willem software).

h-a-l-9000 wrote:

The USB to parallel port things are not really a replacement, unless the Willem designers say so.

Apparently in Wine you can map any linux parallel devices to standard dos/win io port numbers which should work. I'll try that first before going the PC I card route.

These things usually only support the printer protocols, not directly manipulating the signals. But feel free to waste your time and money...

Well this one I picked up (for 2.49 gbp!) does appear to present a standard /dev/parport device in Linux, so anything that talks directly to a normal parallel device file in Linux should work ok - however I didn't have much luck getting the WIllem software in Wine to use it.

So.... I ended up pulling out half the devices in my P4 WIn98 machine which has a dodgy PSU, enough bits removed in order for the system to boot and run Windows - this is the only hardware I have left which has a physical parallel port.
I went through my box of 27c256 and 27c128 chips and found enough blank ones to program at least 3 sets of high/low bios pairs.

First of all I ran the chips from the motherboard through the reader and saved the dumps, so I would have something to refer to later. The first BIOS I tried (http://chukaev.ru54.com/bios/2hlm006.zip) didn't do anything - the board just sat there; no bios beeps, nothing.

The second one (http://chukaev.ru54.com/bios/2hlm008.zip) booted and gave me a working system - one interesting aspect of this BIOS is that it adds a 'configure X Kbytes of extra memory as EMS' option, allowing me to add anywhere from 0 to 3072kb of my 4Mbytes as EMS. Quite neat!

I tried the first burned BIOS again, just to see if the working one had spurred something into life. No luck. I then tried my original AMI chips again and wouldn't you know it, but the damn thing burst into life!

Just look at the thing... I bet it's secretly laughing at me - I have no idea why the original BIOS chips are now working... maybe a higher voltage in the programmer than supplied by the motherboard 'cleared' something. I don't know how that would be possible though.... see for yourselves:

I've decided to stick with the original AMI BIOS that came with the board, as the date code is slightly later (1990 compared to 1989 of the other working one that listed 'EMS' options) and also has a 'Fast ISA Bus' option (8Mhz or 12Mhz) in the setup menu that the earlier one doesn't have.

I have one last set of Headland HT12 compatible BIOS roms to try from that Russian site; so I'll see if they have any additional options or features that are not in either of these two.

Better read out these now-working chips again and compare the data.