Research Machines RM380z

Known Issues
The power supply was blowing fuses and was outputting low voltage.

The bridge rectifiers were also getting very hot after a few seconds.

This unit also exhibited no signal on the video output and no life on the bus, this was due to all the corrosion this specimen was subjected to. This specimen is also missing a floppy drive as that one was very corroded and the bearings were seized that is why the photo has a large hole in the front left.

Causes

 * The large filter capacitors were bad.
 * The tantalums on the cards were bad.
 * Sockets and ICs were corroded

Solutions

 * Tried reforming the large filters however this was unsuccessful so these were replaced.
 * Replace the tantalums which lowered the current draw to what was expected cooling the rectifies down to normal operating conditions..
 * Replaced the sockets and replaced the ICs which cured the inactivity and no video output.

Disassembly Notes
Simple disassembly is easy with two screws on the back of the lid to gain access to the cards however the power supply requires disordering cables and removing the whole case off the base.

Removal of the floppy drives in the vertical configuration is also complicated as they require the removal of the front panel and all the cards. This can be quite difficult especially if the nuts and bolts are corroded like mine, the tight space doesn't help as it can be difficult to get pliers and/ or spanners into the right place. When the lid is removed the cards and backplane are easily seen and the power supply is on the right underneath the CPU serial monitor board. the floppy drive mounts are on the left for vertical mount BASF 6106s. The serial number tag should be visible on the case above the PSU which this one is hand written SN 691

Disk drives
The early RM 380s use BASF 6106 floppy drives which are 2/3 height SSSD 5 1/4. The drives are Shugart interface compatible. Later on IBM PC full Hight drives were chosen and the mounts were changed from a vertical to horizontal. These are easy to replace as they are more common. The early BASF drives can badly corrode depending on storage conditions so be wary of this as 2/3 height floppy drives are not as common.

Keyboard
The keyboard is just standard parallel ASCII foam and foil unit external to the unit via a 15 pin cable. The foam pads in mine are still in good shape but are the same size as most replacement pads making replacement easy. My unit was made by alphametric in the UK.

Keylock
The keylock is a old off the shelf part from RS however the keys are unique to each machine so check to see if you have the keys. This is because the switch controls the power input and also lock function of the front. This one was also seized so was soaked in WD40 to unstick it and allow it to rotate freely.

Cards
The cards are pretty easy to service however the early RM380s use Texas Instruments chips which are know for getting corroded legs. The early ones also use cheap sockets which can corrode. For mine to be a reliable machine large scale replacements of chips and sockets was the only way as the IC legs were breaking in the sockets. Below are the first party Research machines cards in this case as originally configured via the original label inside.

Power supply
''Early supplies use the types below. Later revisions use different types.''

Related Links

 * https://www.computinghistory.org.uk/det/2226/RM-380Z-(Black-Case)/
 * https://www.theregister.com/2013/11/13/archaeologic_the_research_machines_380z_story/
 * https://vt100.net/rm/380z_review
 * https://archive.org/details/rm380zservicemanualpn13821a (This covers the later 380z but most of the schematics and information is still relevant.)