Interesting drive failure...

I'm attempting to dump an image of the Ridge 32/330's original SMD drive (A Fujitsu MK2322) -- I'd been running it off of a spare, because I didn't want to disturb what was on the original drive until I could back it up. Now that the spare has died I figured it would be an opportune time... The Ridge "DISCUTIL" tool (loaded from QIC) is pretty powerful, and amongst its weapons is the "DS" command which will dump a sector of the disk to screen, in ASCII HEX. The Ridge has been running for the past eight days dumping the entire 168mb disk, in HEX, at 9600 baud. Yes, I'm insane. Along the way I discovered that the integrity of the data on the disk is pretty poor. There are lots of bad sectors. In particular surfaces 2 and 3 are completely unreadable for all 823 cylinders, surfaces 4 and 8 are marginal most of the time. After a lot of retries I've managed to get most of everything (except for surfaces 2 and 3 which yield no data at all). So I figured that heads 2 and 3 are toast -- perhaps bad analog logic, maybe something in the digital logic that selects the head, maybe even bad heads, who knows. DISCUTIL also has a "FORMAT" command that will format individual cylinders, on a specified surface. So after I'd read out everything possible to save, I tried it on a cylinder (801) known to have trouble sectors, that also containing all zeros so I wouldn't wipe out anything vital. After doing a format on cyl 801, head 2 I could read all sectors on that cylinder/head without issue. Same with head 3, 4 and 8. Suspecting a possible addressing fault I tried reading every surface on cylinder 801 after formatting all surfaces on that cylinder, and they all read without issue. (I'd expect that if, for example, when I asked the disk to format surface 2 it actually formatted surface 5, then reading surface 5 later would fail due to an invalid sector header for surface 5...) So, I suppose this is somewhat good news in that I may be able to use the drive, but I wonder what would cause the data on the disk to fail in such a manner (and I'm still hesitant to reformat the drive without recovering stuff off of surfaces 2/3...) Any thoughts? - Josh