80-Bus News


November–December 1983, Volume 2, Issue 6

Page 29 of 67


By Richard Beal

I recently added an eight inch drive to my Gemini system. This article describes the various problems I encountered and how to overcome them. My system has a Gemini GM829 disk controller card, with Teac 5.25 inch drives which have the same interface as the 48 tpi Pertec drives that Gemini used to supply. The eight inch drive is a double density double sided Pertec FD650, put experiences with any other eight inch drives would probably be similar.

I powered up the drive with a separate power supply and obtained a cable to connect it to the disk card from Henry’s Radio. I generated a version of SYS to support the drive (as a standard single density single sided drive) and booted up the system. To my satisfaction it worked first time, and I started to use it. [Ed. – certain Gemini CP/Ms also have 8″ support already incorporated. If the CP/M sign-on message has an ‘E’ in its version number (e.g. BIOS Vers 2.4 2-SME) then it means that the drive following the 2 Single sided Micropolis drives is an Eight inch one – SSSD. If you want 8″ support and have not got the support contact your dealer for BIOS upgrade details.

I started doing some text processing work and as the drive and fan were a bit noisy I turned it off when I wasn’t using it. When I started trying to read the files I had written on the Teac drives, I found that whole tracks had CRC errors that were permanent, although other parts of the disk were all right. Since this happened on both Teacs, I realised that the problem must lie not with the disks or the controller, or the software, but must be something to do with the 8″ drive, which wasn’t even switched on.

The problem is caused by noise of some sort coming from the cable leading to the 8″ drive. This happens only when there is no power to the PCB in the 8″ drive, and is also intermittent. This explained why the errors were CRC errors not RNF, as the data written was corrupted and could never be read back, but the sector headers were unaffected. The solution was to rewire the power supply connections and the switch in the 8″ drive case so that the switch controlled the drive motor and fan, while the transformer and drive PCB are always live when the entire unit is plugged in.

While doing the rewiring I was uncertain about one of the connections and tried to pull off the insulating cover with a pair of pliers. The next moment there was a terrible sensation through my whole body and then I found myself at the other side of the room, with bits of furniture hurled around. The metal connector had been bent through 90 degrees, which I would not normally have had the strength to do. When I had recovered I found that the 3 amp mains fuse had blown, which probably saved me. Remember that 250V is rather more than the usual 5V we get used to handling, and that it is sensible to unplug everything from the wall before starting work!

The system then worked reliably, but still had one annoying problem. If no disk was in the drive, or if the door wasn’t closed, the system would simply hang up indefinitely. This was disastrous when there was no disk in the drive but the door was shut, as the door locks when the drive is selected. Since I couldn’t open the door to insert a disk, the only answer was to reset the system. The software is designed to detect the status of the Ready line, but this was not connected. On the 5.25 inch drives, the software times out when the motor stops, but on 8″ drives the motor runs continuously.

Page 29 of 67