• Please review our updated Terms and Rules here

Compaq Portable II: low-level format of HDD with MFM-to-IDE adapter

mikerofone

mikerofone

Experienced Member
Joined
Aug 9, 2020
Messages
69
Location
Switzerland
Hi all,

I got a Compaq Portable II with a Miniscribe 8425 HDD. It's an MFM drive that has an MFM to IDE adapter board bolted to it's underside, labelled WD1001-IWH. The system HDD controller is IDE.
The drive seems to be fine mechanically, but seems to be unable to find tracks. Any read to the drive results in the drive going to track zero and then slowly stepping to the requested track three times (see first video in this album), before returning an error. Seeks work fine though. I tested all this using SpeedStor 6.0.3.

My feeling is that the drive needs a low-level format so it'll find its sectors again. However I don't know how to do that with a drive connected via IDE. The corresponding option in SpeedStor errors out immediately after failing the first attempted write, which makes sense as it's going through the IDE controller. Other LLF tools like sgatfmt4 or the one in the IBM Advanced Diagnostics disk fail in the same way described above.
I didn't find too much documentation on the WD1001-IWH, but it doesn't seem to come with its own LLF routine [1]. I've tried a couple of the common DEBUG addresses and as one would expect, none work.

Any ideas on how, or if even at all, I could trigger a low-level format through the WD1001-IWH adapter? I feel like LLF-ing the drive by directly connecting it (without the IDE bridge) to a real MFM controller might help, but I currently don't have the right components available to do so.

Thanks for any suggestions! :)

Cheers
mikerofone

[1] https://www.os2museum.com/wp/learn-something-old-every-day-part-ii/ says it's similar to a WD1003-WAH which doesn't have an LLF tool either, as per https://minuszerodegrees.net/manual...d WD1003-WAH and WD1003-WA2 - Users Guide.pdf
 
Is it similar to mine? I got this out of my Compaq portable II I assume as well. Disregard the drive attached, A few years back I tried a bunch of drives on it but never got it working. I dont think its faulty I think I just dont understand its proper usage. I am sure its not standard IDE but XT 8 bit ide also. Until I know what to do with it, it lives on my shelf. IMG_20230313_153749.jpg
 
Yup, exactly the same board. It seems that it supports exactly two drive geometries, or at least only two modes are documented in the portable 2 service manual: 20MB (Seagate ST225 compatible I believe) or 10MB, switched by a jumper on the board. This person seems to have been successful with connecting a matching drive to it, but it was already low-level formatted: https://www.os2museum.com/wp/learn-something-old-every-day-part-ii/
 
It does. The others aren't populated or documented either. Makes me wonder whether there might be a jumper setting for "initialize drive". As the drive is nonfunctional and I have a second, working one I put into the Compaq, i might as a last ditch effort just try to see what happens when I connect those...
 
Didn't the utilities that came with the machine provide the format tools? Been forever but I recall there being a format tool.
 
Sorry for the lack of reply, twolazy. I checked the diagnostic disk and couldn't find a specific formatting tool. I found that the IBM Advanced Diagnostics Disk had a LLF tool (via https://minuszerodegrees.net/ibm_xebec/ibm_xebec_llf_add.htm) but that one failed the same way - the disk kept seeking endlessly until finally erroring out.

I managed to use the LLF formatting tools I mentioned in the OP (like SGATFMT and SpeedStor) on a second identical drive that I have, and they worked. The second drive was working to begin with, but showed a lot of bad sectors on head #2 (nearly 50% of sectors under that head have errors) in both tools. I wonder whether LLFing that one using an MFM controller would also revive head #2 - it doesn't look like it has crashed, as it didn't get worse, and the drive is working well. Interestingly, marking sectors as bad permanently on the disk didn't seem to work either, they all showed up again when formatting (which I had to do in SpeedStor as DOS' format.exe would stop with errors).

I still don't have the cables to connect the drive to an MFM controller directly, and I don't know whether the controller I got from somewhere would work either. For now, I hope that the one mostly-working drive that I have stays working. :)

I will report back if I ever get around to LLFing that other drive.

Cheers
mikerofone
 
You could use those very short cables if you can find a way to suspend the drive up in the air while testing...
 
I tried, but they are keyed (one hole plugged), and the headers on my controller aren't. Though, maybe I should just get out them side cutters... if the cables can be keyed, surely the controller end doesn't need those pins either. I just need to triple check that I got the cables the right way around before getting to cutting. ;)

I'll ponder the idea for a bit.
 
Stick a fine tip soldering iron tip in it to open a hole or use a cigarette lighter on a paperchip until its red hot to do the same thing. I have had to do that many times. Early vga cables and cards that had a blocked pin as well
 
I have an awl (? this thing) that I pushed against the hole, and then heated up by holding my soldering iron against it. Worked surprisingly well, after just a few seconds it pushed right into the plastic. A bit of wiggling to ensure the hole is wide enough and now the cables fit well.

I plugged the controller into a Pentium class machine I have and disabled the onboard IDE. The BIOS didn't detect the drive, but after configuring no drives at all, the controller actually presented a "Couldn't get drive parameters from disk, want to launch pre-format tool?" message instead of "no boot device found". The disk started formatting but it's going veeeeeeeeeeery slow. From the sound of it, it processes maybe one track every 10 seconds (audible stepper motor click). Formatting all 612 tracks is going to take a while - I'll have to restart it tomorrow when I have the time to let it run through.

Thanks for the encouragement, this is already farther than I expected to get with this random controller and the very short cables I have to work with. It's all a bit precarious (the cables are under tension, and the controller card isn't fully seated in the ISA slot) but it seems to be working, albeit slowly. :) Here are some not-so-exciting pictures of the process: https://photos.app.goo.gl/tfZLz2cAXYzA2xxs9

Cheers
mikerofone
 
Remember sometimes leaving the drive doing a scan or just running for many hours sometimes "wakes" them back up again. Progress.. Good job!(y)
 
You must log in or register to reply here.
Back
Top