TOM WALKER wrote to ROY WITT:   
      
    RW>> You could have used a Japanese bearing and had the same quality, but   
    RW>> less cost. I used to rebuild IBM hard drives (12" discs) and used a   
    RW>> Japanese bearing in place of the original. They were just as good.   
    RW>> The bearings that IBM used had to be bought in matched pairs and   
    RW>> cost an arm and a leg. I used non-matched bearings and had no   
    RW>> complaints.   
      
    TW> Well in some cases "matched bearings" are manditory if one does not   
    TW> want a early failure.   
      
   These IBM HDs were 512k on 5 or 6, single sided 12" discs...I collected   
   bad discs and drilled a hole in them and used the very large center hole   
   as a picture frame.   
      
   The hub on which the HD discs rode was extra heavy duty and required the   
   invention of a special tool to remove the top bearing (which was glued   
   into place with #609 loctite). I still have that tool, as it was a very   
   innovative design that I designed and made. This tool saved hours of   
   disassembly time, which made the rebuilding procedure take 8 hours, just   
   for one hub. With the tool, we could do 7 to 8 per day.   
      
    TW> I am thinking of my experieces with Gyroscopes used in navigation   
    TW> equipment. They of course spun up much higher speeds than the Old   
    TW> Hard Drives, Up to 24,000 RPM   
      
   I don't recall the speed of the hub, but it rode on 6001 series bearings,   
   which are  very stable at lower speeds. I think that once assembled, the   
   hub was originally balanced with the discs mounted. They did the balancing   
   by drilling a series of holes in it. Not exactly gyroscopic protocol.   
   Anyway, we never re-checked the balance.   
      
      
            R\%/itt   
      
      
   --- GoldED+/W32 1.1.5-31012   
   --- D'Bridge 3.92   
    * Origin: Bow Tie Racers, Been There, Done That! (1:387/22)