dkelvey at hotmail.com
Sun Jul 23 10:22:30 CDT 2017
I have a system that uses hard sectored disk( 8 inch 32 sectors ) but only used two sectors 16 sector holes long.
I understand that the Wang systems used hard sectored with the holes around the outside of the disk.
There is more than size that is strange.
From: cctalk <cctalk-bounces at classiccmp.org> on behalf of Chuck Guzis via cctalk <cctalk at classiccmp.org>
Sent: Friday, July 21, 2017 1:16:54 PM
Subject: Re: Diskette size
On 07/21/2017 12:19 PM, Paul Koning wrote:
> Interesting. Another example, slightly later, is the audio unit of the PLATO IV terminal (1974 or thereabouts). It uses a rather large disk, perhaps 10 inches diameter, brown oxide, no grooves. It's a random access device, with 128 tracks. Each track has 32 sectors; a given audio clip can be up to 127 sectors long (though I'm not sure what happens if it's more than 32 sectors -- does it switch tracks? Seems unlikely).
There was also a 1954 Timex Magnetic Disc recorder--there was also a UK
product of a similar date and time--Pye, but that used grooved discs.
The advantage of the Timex format was that you could fold the disc up
and mail it to someone:
> The track seek is done with a binary encoded pneumatic cylinder
assembly, 7 cylinders -- low order stroke is one track pitch, next is 2
tracks, next is 4 tracks, etc. So the binary track number would select 7
air valves which would feed supplied "shop air" to one or the other side
of each piston, moving the read/write head to the correct track.
I remember the same system used on the Univac Fastrand II drum; a great
big dual sewer-pipe unit, complete with fluorescent-lit observation windows:
At the time, I was fascinated by the positioning system. A later (late
1970s) system was proposed for 8" hard drives using a setup of
concentric cups. The outfit, run by a friend was called "Ontrack" and
never came to fruition. There's probably a patent on file somewhere.
> Although Chuck mentioned Dysan putting on hub rings, Verbatim (who
> were selling more) didn't do so until later. Therefore, some had
> hub-ring; some didn't. Office workers might as well have been told,
> "put the ones with hub rings into the drive with the asterisk",
> since lack of hub-ring and lack of asterisk mean nothing. Note: One
> of the options for The Berkeley Microcomputer Flip-Jig (MY first
> retail product) was a jig on the side of it for aligning and
> installing hub-rings.
Yup--the Dysan guys showed up with a jig and a bunch of self-sticking
hub rings before they were standard on their floppies. They helped.
One of the things that apparently complicated matters is that Shugart
had the patent on the expanding-cone hub clamp. If you look at a
Micropolis drive of the same time, they used a cup-and-rigid tapered
cone arrangement, which wasn't nearly as effective.
Of course, by the time 1.2MB (high density) drives came out, the problem
had been resolved on production drives, so almost all 5.25" HD floppies
are sold without reinforcement rings.
Those hub-reinforcing rings could be a pain in the posterior. I vividly
recall one of our customers yelling about one of our copying products
producing bad copies. Since we verified after writing, this seemed
somewhat unlikely. We had them check the spindle speed--it was right on
After a few suggestions yielding no fruit, we had them send us some samples.
Sure enough--you copied to the diskette and couldn't read the result
successfully. I was going nuts.
So I grabbed the Kyread/Visomag/Magnasee (whatever it was called back
then) and my microscope to see what I could see. It was right about
then that I noticed that the samples had *two* reinforcing rings
installed--one on top of the other and offset slightly.
This interfered with the clamping action of the drive hub to produce an
effect that the same disk could never be clamped in the same position
The problem was solved by the customer trashing his stock of blanks.
> These terminals also had a back-protection setup for the plasma panels (the "slide projector", actually more like a microfiche projector). Same sort of setup, but with X and Y both done by binary weighted sets of 4 air cylinders each.
"The first thing we do, let's kill all the spammers."
More information about the cctalk