Trouble with 3390-2 platters.
Nigel Davies
One "Wed, 19 Jul 1995 04:38:58 EDT", Diminic Rooney wrote:
> We were alerted to potential problems with our 3390-2's
> . . . .
> We were informed that the problems were limited to units produced
> at the Mainz factory over a particular period.
I had the good fortune to attend a set of "Executive Briefings" at IBM
Mainz a few years ago. At that time they acknowledged a problem with
"early" 3390-2 platters, whereby after a period of time the oxide
coating would start to "flake off". This would account for the apparent
"instantaneous" failure with little or no warning.
The oxide coating on these "early" units took the form of a ferrous
oxide impregnated paste that was sprayed onto the aluminium platters.
The solution was to employ an entirely different coating technique known
as "sputtering". This involved putting a charged platter at one end of a
sealed chamber and firing oppositely charged oxide particles at it from
the other end of the chamber. This meant that they were able to achieve
an almost molecular level bonding of oxide to platter.
Hope that explains it :-).
Nigel.
------------------------------------------------------------------------------
T Nigel Davies, Senior Systems Engineer VM/VSE Mid Glamorgan C. C.
Tel: +44 (0)1222 820280 Fax: +44 (0)1222 820195 IT Services Division
Internal OV : MIDGLAM2(DAVIETN) County Offices
Mail Exchange : IBMMAIL(GBC3F33F) Greyfriars Road
Internet : gbc3...@ibmmail.com Cardiff, CF1 3AB, UK
-------------- Never confuse moving fast with going somewhere. ---------------