[comp.sys.ibm.pc.hardware] Seagate157A

jos@udonix.hobby.nl (Jos Jansen) (05/07/91)

The following problem has, so far, defeated my efforts.

-    machine: unknown AT-clone sold as 'Humbird', obviously MIT;
     backplane design with CPU board bent like a banana to make it fit

-    BIOS:

-    disk:     Seagate ST157A

-    controller:    due to the nature of the disk, this should be an
     IDE controller. The sole IC is marked ACER M1505 9042.

-    OS: Dos 3.3 (OEM version)

Symptoms:

The usual FDISK - FORMAT routine creates the usual partitions.

As soon as disk writes are made, the disk gets screwed up; CHKDSK shows
many 'allocation errors' and useful files such as COMMAND.COM get
clobbered. This is NOT an intermittent problem; just copying one
directory tree back and forth terminates the file system. 

We used NDD to check the disk (lots of errors, of course). 
We did NOT use Disk Manager-type software to repartition the disk. 

History

The disk had been supplied and installed as a 40Mb hard disk, as
replacement for a broken 20Mb Seagate. According to the setup, this
disk is supposed to have 7 heads, which sounds unlikely to me.

Also according to my information, the ST157A has a capacity of 45Mb. 

No documentation was supplied with the system. No useful support can be
obtained from the vendor (who employs N salespersons and one
delinquent who's really good at bending CPU boards). 

The vendor claims that the disk worked when he installed it. My
colleague - who is now reduced to using pencil, paper and calculator -
says it didn't after he copied DOS to a subdir. I believe him (see
below); I saw the resulting mess and could reproduce it exactly.

Query: 

-    does anyone have specs of this disk? 

-    does anyone know how to give this disk a factory-level (i.e. low-
     level) format? 

-    am I right in suspecting drive-controller mismatch?

-    is there a way (now I'm just skirting libel) to recognize drives
     which have been re-conditioned? In the past six months the vendor
     supplied four disks as add-ons or replacements; all but three were
     DOA (i.e., total failure within 2 months of 9-5 use). 

-    what's true about allegations that one should not apply low-level
     format software to IDE disks?

Any reply will be appreciated

-- 
Jos Jansen              =     UUCP:     {...}!mcsun!hp4nl!hgatenl!udonix!jos
tel: (+31)1729 8288     =     Domain:   jos@udonix.hobby.nl