A real 803 logic fault, Tic TAC Toe and searching the ICL archives
Post this page to popular social media
In a bumper edition this week we have...
A real 803 hardware logic fault is discovered and fixed
When Peter 'facial fuzz' O switched on the 803 on Saturday things did not go as expected which was put down to the cold weather overnight. However, after waiting a while for it to warm up it still did not want to load and run programs from tape. Initial suspicion turned to the tape reader but loading a short program from the word generator (console) also failed so something more fundamental was causing the problem. Several test instructions were performed which narrowed the problem down to the accumulator being loaded with -1 instead of +1. Time to look at the logic diagrams to see what could cause this. After a few hours of signal tracing with a scope and meter, and help from John S and Adam, the problem was tracked down to a faulty diode. A quick search of John S's 'secret component store' found a suitable replacement which was fitted and the system was operational again. Nice work Peter.
However, that was not the end of the investigations... yes, the top 4K of store was failing again... so, another round of scoping and metering began. After much head scratching and finding nothing obvious, it was discovered that pushing one of the core store control cards in fully resulted in the fault going away. The board was removed and edge connector pins cleaned but this did not fix the problem. So this looks more like a mechanical fault rather than logic one this time or something wrong on the backplane. As it was the end of the day, further investigations will have to wait until the next time Peter O or John S are in.
The TAC computer is powered up after years of silence
There was an unusual gathering of people around the TAC (Transistorised Automatic Computer - a mid 1960s system used to monitor a nuclear power station) on Saturday. Someone had switched it on and it was showing green lights on the console. That someone turned out to be prospective volunteer Colin H, who some of us later found out was part of the team that installed the system in Wylfa in 1965, and was thinking about joining us to work on the system again. Colin had only intended to do a visual inspection of the system but having seen the interest fellow volunteers and public had in what he was doing, he decided, having completed his checks, to see if it would switch on. We had tried in the past but for some reason it did not work. It turned out that one of the power supply rails had tripped a switch but the trip switches used in the system had no obvious indications they had tripped. With suitable inside knowledge, Colin identified the cause, pushed the relevant trip reset button and the system powered up for the first time in many years. After checking some of the console activities he decide to try and load a test tape, however this failed either because he had forgotten how to load it or more likely due to a fault. So a significant amount of progress was made, even if it was not intended. We hope to see Colin return in the not too distant future to continue working on the TAC.
An important ICL circuit diagram is found for the 2966 operators console
Pete H continues to work on repairing the operator consoles but is still struggling without any circuit diagrams. On Saturday, Pete had help from two of our newest volunteers; Nick and Delwyn. The decision was made to do some detective work with the ICL archive that the museum has. This consists of thousands of punch cards, each containing a 35mm negative of circuit diagrams, manuals, software and associated information for much of the equipment that ICL produced. They were somewhat hampered by not having a full index system so they spent all day searching what index they had. Nick was the first one to find something, that being a parts list for the console. Further searching then identified a reference to the circuit diagrams for the console, and with 5 mins before closing, the punch cards containing the elusive diagrams were finally located. The next hurdle is how to get a printed copy, but that will have to wait until next week. So Pete H has, at last got some good news and will hopefully be able to make progress on fixing the consoles. Well done Pete, Nick and Delwyn.
Work starts on restoring the Harwell computer
Under the watchful eye of working group lead Tony F, work has now started in earnest on restoring the Harwell Dekatron Computer (HDC). A new volunteer, Eddie W, has joined the museum and the HDC working group and his expertise is in Post office Relays, which the HDC has in abundance. Eddie has done an initial inspection on some of the relays and concluded that they all need cleaning and adjustment (no surprise there given their age!). This is going to be a very time consuming and fiddly task for Eddie, but he is certainly up for the challenge which has started already.
Tony F has now an initial inventory of all the valves and tubes which, fortunately or perhaps not, does not clash with the valves used on Colossus. He is working with John Pether on searching the museums large stock of valves to see if we have any that are useful. He has also been in contact with valve enthusiast groups in the UK and US who are now on the lookout for suitable replacements and spares - the major worry are the Dekatrons, which are very rare and the system uses 900 of them! - if you have any Dekatrons or valves you are willing to donate please get it touch with us. Tony has also made an initial inspection of the power supply units and made a start on cleaning and checking the rectifier unit but he has already discovered the PSU does not match our circuit diagrams so a new one will need to be drawn up.
Peter 'double agent' O has been making steady progress in reading and decoding some of the 5 hole paper tapes used by the HDC on his home brew tape reading 'Gizmo'. Some of the tapes had become very translucent and had initially been unreadable, but with a few hardware mods, Peter O was able to improve the reader enough to cope with the problem. There are still many boxes of tapes to copy so Peter still has a lot to do before we will know what we have - this is hindered a bit by some of the boxes being mis-labelled.
Happy Birthday to you... happy birthday to you...
We had an impromptu mini birthday bash at our weekly 'pizza' meeting when we surprised one of the trustees and fellow volunteer with a birthday cake and a noise loosely related to the happy birthday song... however we hit a slight snag when lighting the candles.... no-one had any matches and our 'smoker' in the camp had decided to give up a few weeks earlier so no longer carried their lighter... Anyway, Kevin B saved the day by finding some matches in his bag, so with a huff and several puffs, the birthday boy blew out all 12 candles - however it must go on record that he took two goes to blow them all out! The cake was nice though.
We won't add to Kevin M's embarrassment by mentioning who's birthday it was...Doh!