Al Shuqart Remembers - Technology Information

Computer Technology Review, Jan, 2000 by Mark Ferelli

AL: The three most significant events in mass storage?

MARK: Yup.

AL: OK, well, the first one is the invention of the slider bearing. The original disk drive from IBM had an externally pressurized air bearing where they brought air in to the head and routed it through the head to provide the air burn so the head didn't touch the disk. Then, IBM developed this bearing, which didn't require an external supply of air and they called it a slider bearing and that was the first disk drive that IBM produced that I was responsible for, the IBM 1301. At that point, everybody went to slider bearing. That's what makes the disk drive industry go. I guess nobody used the externalized pressurized bearing. It was too cumbersome. The significance of that was without having to bring air into the machine in each head, you can now put a head on every surface. So that improved access time tremendously.

MARK: I hear that. OK.

AL: Does that all sound logical?

MARK: It sure does.

AL: Actually, I got four things and you can pick out the three that you like. The next one was track following servo. Prior to track following servo, you located the track, usually with a mechanical racket pinion or D10. You put a piece of metal into a rack that defines the track by the teeth in a particular rack. It identifies the location. That was replaced by track following servo, which IBM introduced in their first, I think it was, Merlin disk drive, the 3330 and this permitted you to have higher track densities because the head would read the track and then adjust itself to the precise location. The head would read the information on the track and then feed it back to the voice coil actuary, which was, then, adjusted. It was a fit looped feedback system that permitted you to have the head to follow the track in a very precise manner, which meant that you could have more tracks per inch.

MARK: If they had more track density, does that mean they'd have to make more adjustments in head slides?

AL: No. Not necessarily. There are other reasons for getting the head closer, but anyway, that provided for greater track density, which was the track following servo and then the third one was the low mass lightly loaded head or, as some people call it, the Winchester head. I was still at IBM at the time and this was a very small head. That's the low mass and had a very light load. Pushing it on to the disk started and stopped in contact. That was called the Winchester. So, your head's in contact with the disk and then the disk is sitting still. Then, the disk picks up speed, then the head starts to fly. When you shut it off, the disk slows down eventually and, then, the head comes back in contact with the disk. So, it's a low mass right over the head that starts and stops in contact with the disk and that's what everybody uses now. It was conceived in a company called Data Disk--I remember the president, Armand Miller. IBM negotiated a license to the low mass slightly loaded head and the first IBM product that used that was the Winchester 3340.

AL: OK. The fourth one is the magneto-resistive head. This is significant because no longer are you worried about the speed of the disk going underneath the head because the head is not reading the speed of the flux change. It's reading the actual flux change. That's what magneto-resistive means. And so that was very, very significant. Of course, from that magneto-resistant head it went to GMR, giant GMR, and all that stuff. But the magneto-resistant head was a totally different concept from the inductive head, which required you to do half the disk rotation in order to read it.

MARK: OK.

AL: So those were four. You can pick the one you like.

MARK: Sounds good to me.

Alan Shugart is the founder of Al Shugart International (Soquel, CA), a startup resource company focused on helping entrepreneurs launch new enterprises and on growing small companies into big companies.