Look for speedstor...great Tool!

Many BIOSes from the 286/386 era have LLF option.
If your doesn't, then there's HDINIT.EXE from Checkit 3.0 package, Speed Stor, and many others.

OK thanks, I'll look for those.

Speedstor is great, thanks! The database is not very accurate, though... had a Toshiba model's sector count double what it really is...

What's a reasonable "default" interleave for a 42mb 3.5" MFM? I don't want to obsess over this, just get it working well. 3, perhaps?

precaud wrote:

What's a reasonable "default" interleave for a 42mb 3.5" MFM? I don't want to obsess over this, just get it working well. 3, perhaps?

3 is a good point to start.
Then you can run eg. Calibrate from Norton Utilites - it can detect the optimal interleave.
Or simply run any HDD benchmark, if the transfer rate is around 30 KB/s you need to reformat with greater interleave.

The AMI BIOS in my 286 had a low level format util with a built in speed test.

Thanks for you help, guys. Of the seven HDD's tested / LLF'ed (all were MFM made in 1980's), two had excessive bad sectors and will be tossed, two were ok, and three were exceptional. The three exceptional ones were all Seagates. Makes me think that their platter technology was better than the others...

Do a teardown of the bad drives and post pics.

OK, it will probably have to wait until the weekend. Are you expecting to see something in particular?

I simply use AMIBIOS low level format in diagnostic section of the setup. Works fine.
Funny thing is I managed to fix couple of problematic ST157A-s (IDE HDD! and it states on the sticker on the drive to not LLF it at all!) using this util as well. After reformatting the drives suddenly started working. Although I had to use native CHS layout instead of translation one.

Not really. I just want to see a teardown of an MFM drive since I've never had the chance to take one apart.

cyclone3d wrote:

Not really. I just want to see a teardown of an MFM drive since I've never had the chance to take one apart.

Here's the innards. This one is a Toshiba MK134FA 42MB MFM drive with four platters, seven heads. Not sure what if anything they do with the 8th surface. There is definite discoloration of the surface where the heads spend the most time. The errors were concentrated there, as well. I would say, this drive saw heavy use beore I got it.

The heads are moved across the platter surface by the little stepper motor in the bottom left of the 1st pic. The actuating length (motor to arm) seems pretty long to me, a possible source of errors.