15340TDD@MSU.BITNET (Tom Davis) (07/11/89)
I'm trying to help a friend put a new 20M hard disk in his original IBM PC, and ran into some problems. After getting the disk installed we were unable to run FDISK to setup the DOS partition. It failed with "Error reading fixed disk." We ran the hardware diagnostics and it failed trying to access track 0 on the hard disk. My guess is that the BIOS in this old beast doesn't know what to think about the 20M drive (they were all 10M back then I believe) and improperly identifies it to DOS at which point it goes south. Can anyone confirm my hunch? I think I have two options available: 1) Find some newer BIOS ROMs and installed them in the PC or 2) Install the disk in a newer machine, run FDISK (and FORMAT?), then put it back in the old PC. I hate to go to such trouble if I'm on the wrong track... FWIW, the BIOS is version C1.10 dated 25-Apr-1981, the new disk is a Kalok KL320, the controller is Western Digital.
psb@eecea.eece.ksu.edu (Phil Buckland) (07/12/89)
In article <10315340TDD@MSU> 15340TDD@MSU.BITNET (Tom Davis) writes: >I'm trying to help a friend put a new 20M hard disk in his >original IBM PC, and ran into some problems. After getting >the disk installed we were unable to run FDISK to setup the >DOS partition. It failed with "Error reading fixed disk." >We ran the hardware diagnostics and it failed trying to >access track 0 on the hard disk. > Have you made sure that the two hard drive cables are properly installed (pin 1 of connector on controller card to pin 1 on drive connector via the red striped wire in the cable)? Also, is the drive select jumper on the hard drive in the correct position (If cable is twisted, jumper for DS1 if the first drive select number on the drive is DS0, or for DS2 if the first jumper is DS1. If the cable is *not* twisted, jumper the drive for DS0 or DS1 according to the aforementioned rule.)? Is the terminating resistor pack installed in the hard drive? If there is more than one hard drive, it *must* be installed in the drive attached to the last connector of the 34-pin control cable. Finally, have you run the skeleton formatter on the drive before running FDISK? There should be instructions on how to do this in the documentation sent with the controller. Judging by the error messages you are reporting, it would appear that the drive doesn't have the skeleton format on it (sector addresses only). >My guess is that the BIOS in this old beast doesn't know >what to think about the 20M drive (they were all 10M back >then I believe) and improperly identifies it to DOS at which >point it goes south. Can anyone confirm my hunch? The BIOS has *no* idea what size the drive is - this is a function of the drive controller only. >I think >I have two options available: 1) Find some newer BIOS ROMs >and installed them in the PC or 2) Install the disk in a >newer machine, run FDISK (and FORMAT?), then put it back in >the old PC. I hate to go to such trouble if I'm on the >wrong track... >FWIW, the BIOS is version C1.10 dated 25-Apr-1981, the new >disk is a Kalok KL320, the controller is Western Digital. If the potential solutions outlined above don't get you a working hard drive, E-mail what happened (or repost) and I'll provide some troubleshooting guidelines. Before ruling that the BIOS is bad, it is necessary to prove that the hard drive and controller combination do work together (by installing them in another system, perhaps). It is also possible for problems to occur by having a stuck IRQ line in the I/O channel (the connector the card plugs in to) which affects *only* hard drive operation - you wouldn't see this problem on a floppy only system. Phil Buckland Computer Engineer psb@eecea.UUCP psb@eecea.eece.ksu.edu