Lot's of mis-information which I think I correct below I'm pretty sure the first shipments of the 23FD was in January 1971 the with 2835 Storage Control Unit for the 2305 Fixed Head DASD and the S/370 M155; it might have been simultaneous but the 2835 did attach to existing S/360 M85 2880 Block Multiplexer channel and the M85 shipped in December 1969. It was followed later that year with the shipment in the S/370 M165 and then the rest of the S/370. FWIW, the 2835 SCU was very similar to the 3380 SCU, mainly differing due to the parallel data transfer from the drive. The 23FD was a read only device Dave Noble started on what became the 23FD in February 1968 and it passed IBM's A-Test a year later. Al Shugart moved to Memorex in April 1969. The first read/write FD's were shipped in 1972 by Memorex and others, it not clear who was first; they were incompatible which each other, but the Memorex could read and write the 23FD disk. The IBM Diskette 1 was announced with the 3740 in January 1973 and shipped in May 1973 - and the industry began Tom -----Original Message----- From: Fred Cisin <cisin(a)xenosoft.com> Sent: Monday, June 8, 2026 5:18 PM To: Murray McCullough via cctalk <cctalk(a)classiccmp.org> Cc: Murray McCullough <c.murray.mccullough(a)gmail.com> Subject: [cctalk] Re: Floppy disk On Mon, 8 Jun 2026, Murray McCullough via cctalk wrote: I think that Alan Shugart (and staff) started work on it in 1967, and it made it to market in 1971. Then, the patent for the disk was issued on June 6, 1972. July 18, 1972 was probably the patent fot the 8" floppy DRIVE. Both of which were already shipping in 1971. They were originally intended as a read-only device for loading "microcode" into the IBM 3330 Shugart moved to Memorex in 1972, but in 1973, founded Shugart Associates. Xerox bought Shugart Associates in 1977. Shugart then started a new company in 1978, and tried to name it "Shugart Technology". But, when Xerox had bought the company in 1977, that purchase included the name "Shugart Associates". So, to avoid trademark overlap legal issues, in 1979, he renamed it "Seagate" NEVER name your commpany after yourself! If you do, then if/when you sell the company, you're selling off all the rights to your name also! OB_irrelevant: [Unless you are POTUS, in which case, you can go around tagging everything in sight, like a teenage spray can gang. By taking immediate action starting after announcement, the opposition, and those who wouldst object to doing things that you aren't empowered to do, can't get organized to block it until it is too late.] -- Grumpy Ol' Fred cisin(a)xenosoft.com

The 360/85 had the microcode for its base instruction set in ROS (ROM) so it would not have needed a device like the 23FD to get started.  It was also shipped starting in December 1969 which is before the date the 23FD is reputed to have been ready.   A paper I found at https://dl.acm.org/doi/epdf/10.1145/1476936.1476969 describes instructions in the the 360/85 to load words into the writeable control store from main memory.   The 370 systems on the other hand had no control store in ROS so needed a device to do the Initial Microcode Load (IML) to get the machine of the ground and I suspect the 23FD and later IML microcode sources where just part of the IML  system as something would have to stuff the microcode into the control store.  I see some 370 documents mention a "Service Processor" that loads the microcode, one candidate for this service processor would be one of the family of Universal Controllers that IBM used in a wide variety of machines dating back to the early to mid 1970s. Paul. On 2026-06-09 20:21, Jon Elson via cctalk wrote:

I see by historical records that IBM patented the 8" floppy disk in 1972. It produced and released the disk in 1971. Was there a reason for this? Happy computing. Murray 🙂 On Wed, Jun 10, 2026 at 2:25 PM Tom Gardner via cctalk < cctalk(a)classiccmp.org&gt; wrote:

What was the application date / priority date for that patent application? They may have applied and shipped product at roughly the same time. The wheels of the Patent Office turn slowly. De

That's a bit puzzling as the 23FD I saw was not under the console, it was definitely on the main panel at the top left. So I did a little searching and the 155 was not what I saw either. Finally I found this: https://drivearchive.co.uk/drivelarchive/blog/_archives/2007/3/19/2817813.h… The 370/135 was what I saw in the warehouse, not the 145.

There is one in the 3830 control unit as well. This large box controlled the DASDs (disks) on early S/370s. I have one, and it still works. -- Will On Mon, Jun 8, 2026 at 8:41 PM steven(a)malikoff.com steven--- via cctalk &lt;cctalk(a)classiccmp.org&gt; wrote:

The original 23FD drive as used in the 3830 control unit for 3330 DASD was a read only device and I remember one of the large system CEs showing me the diskettes for them, the sleeve was much stiffer than later diskettes. The read/write 33FD was first shipped in 1973 and when I started at IBM in 1979 several of the terminal controllers started up from one of them.  The head carriage was moved by a crude stepper that turned 90 degrees per step and was coupled to the bottom of a lead screw by a Geneva drive.  The drives did not have a track zero sensor so when initialize they would take about 80 down steps. There where collars fixed on the lead screw that limited the travel and the one at the bottom would hit a plastic tip on the head carriage to stop the head at track zero, and when being initialized would hammer on the down stop several times especially if the head was already close to track zero.  In my early days as a CE I was thrown into fixing a S/34 that I knew nothing about but it turned out to be a problem with the 33FD diskette drive that I was familiar with from working on terminal controllers, the down stop had broken off the head carriage.  I was at a remote location and had no parts, but was able to fix the S/34 by swapping its drive with one from a 3742 the customer was returning to IBM. Before I left field service the first time in 1983 double sided drives where already starting to appear, these drives had a proper stepper motor and a taught band to move the head carriage, but still did not have a track zero sensor. Several years later I was a hardware support person in the Toronto Lab and one of the rooms I looked after had a 3880 controller with a couple 3380 DASD units and the first time I heard the 3880 start up I thought I know that sound, so I looked inside and sure enough it IMLed from a 33FD. Paul. On 2026-06-08 22:46, William Donzelli via cctalk wrote:

Yes the disk rotates continuously, the spindle is driven by an AC motor that runs as long as power is supplied.  Since this is a single sided drive the head is against one side of the diskette and there is a pressure pad on the other side that presses the diskette against the head.  This pressure pad is normally held away from the diskette by the head load solenoid and only dropped when the diskette is being accessed.  I can think of a couple things that might cause what you observed.  The first would be maladjustment or failure of the head load solenoid that would result in the pressure pad always being in contact with the diskette, the other is media degradation, the binder in the media coating breaks down and the coating comes off in a gummy mess. When double sided diskette drives came out IBM used a head carriage design that lifted both heads away from the media when idle. In the family of the 3660 supermarket system there was a controller that was designed for small stores that did not support scanners and operated off of a diskette instead of the usual 3651 store controller that ran off a hard disk.  When it was initially released it would store a cashiers totals in the same location on the diskette, which would mean that every time the cashier hit the total key on here register that spot would be accessed.  When first installed the customer started to complain about diskettes life in the controller, so an investigation was started and after doing some quick calculations it was determined that the number of times the spot on the diskette where the totals where stored far exceeded the specified maximum number of accesses in the time between a new diskette being installed and it wearing out,  The solution was to alter the microcode of the controller to move the spot where totals where stored around so that the same spots on the diskette where not continuously pounded. Paul. On 2026-06-09 06:08, Peter Coghlan via cctalk wrote:

On Tue, Jun 9, 2026 at 3:03 PM Carey Schug via cctalk &lt;cctalk(a)classiccmp.org&gt; wrote: Would a Greaseweazle : https://github.com/keirf/greaseweazle do what you want? It links to a standard 5.25", 3.5" or 3" drive with a plain cable and with a simple wiring adapter to an 8" drive. Plugs into a USB port which powers the Greasweazle microicontroller but not the drive (of course) It works at the flux transition level and can produce 'raw' files of the timings of said transitions. The standard software will also encode/decode FM, MFM, some GCR, etc disks. The bad news is that IMHO the documentation is very poor. You have to deduce some things from the source code (it is 'open', you can download the source to the host tools, to the microcontroller firmware and the schematics). Even then there are things it is capable of that I have not got it to do and that annoys me,. The good news is that it certainly works. And it's not expensive. Even if you 'only' use it to read/write 'normal' disks for PCs, CP/M machines, BBC micros etc, it's worth it IMHO. There are host tools for Windows (runs in the 'Power Shell' command line thing), certainly. And it appears also for Linux and MacOS. -tony