Acorn Electron 39870 restoration
eBay purchase of a 'non working' Electron. I have found previously that without the keyboard connected the Electron does not power up, the strange white crystalline growth around the connector made me think (hope) that it was just a poor connection. I removed and thoroughly cleaned both the connector and the board with white vinegar, reconnected the ribbon and it powered up fine. However, a lot of the keys were unresponsive.
|
Main board and power supply removed and thoroughly cleaned with IPA and a soft brush. Reconnected and retest OK. power supply and voltage rails all within spec. I still wanted to recap (electrolytics and tantalums) of both the PSU and the mainboard so went ahead with that. Retested OK still. |
|
|
Keyboard next.... a little dirty (hazmat gloves time :) ) |
|
Started by desoldering all the keyswitches
Separate the keyswitch assembly from the PCB
Remove the keyswitches from the mounting plate
Separate the keyswitch assembly from the PCB
Remove the keyswitches from the mounting plate
I could have just worked on the unresponsive keyswitches but it would only be a matter of time before the currently working ones failed, so, I decided to put all the keyswitches through the same process to bring them all in line. This involves partially disassembling each switch and cleaning the contact pillars.
I use a pair of pin vices, clamping down on the pin then unscrewing them, leaving them in the pin vice. With both pins removed I flood the switch with DeOxit D5 spray switch cleaner. I also apply cleaner on to a cotton bud and wipe each contact pillar until all tarnish has gone.
Then I reinsert the pillars and retest, if the resistance is greater than about 0.05 ohms then more work is required. Note: out of the 56 switches, only 6 needed more attention.
I use a pair of pin vices, clamping down on the pin then unscrewing them, leaving them in the pin vice. With both pins removed I flood the switch with DeOxit D5 spray switch cleaner. I also apply cleaner on to a cotton bud and wipe each contact pillar until all tarnish has gone.
Then I reinsert the pillars and retest, if the resistance is greater than about 0.05 ohms then more work is required. Note: out of the 56 switches, only 6 needed more attention.
This involves a complete strip of the key, cleaning each part before reassembly. In all cases I got the resistance down below my self imposed 0.05 ohm value.
Slight aside...(I was bored with cleaning keyswitches!) The keytops I cleaned in a solution of UBIK2000 in an ultrasonic cleaner. and The case was washed in a solution of UBIK2000 and dried.
Slight aside...(I was bored with cleaning keyswitches!) The keytops I cleaned in a solution of UBIK2000 in an ultrasonic cleaner. and The case was washed in a solution of UBIK2000 and dried.
Now for the reassembly....
Keyswitches dry fitted into the retaining plate. PCB installed and soldered up. Keycaps fitted
Keyswitches dry fitted into the retaining plate. PCB installed and soldered up. Keycaps fitted
|
Not far to go now, I added a small heatsink to the ULA because why not, then reassembled the case, etc.
I left it running for several hours then checked for temperature issues, the only one was potentially the 6502 running at 56 degrees C, too high in my book so I added a heatsink to it and left it another few hours, bringing it down to a respectable 40 degrees |
|
|
Finally, I coated the keyboard, main and power supply boards with an Ambersil Conformal Coating before refitting. Then just a case of making sure all connectors, speaker, etc. were connected then reassembled the case. |
|
FINAL TESTING
- All keys very responsive, no contact bounce or intermittent contacts
- RGB socket: good signal with my RGB-SCART cable
- Power socket and Original PSU brick tested OK
- Cassette port: tested with my Acorn Data Recorder, save and load OK.
- RF output: worked but my modern TV does not like the dirty signal from these modulators! but at least I got a workable picture
- Video output: worked fine but mono so added LK4 to add the colour back in. Note: the mono output is far crisper than with the colour signal mixed back in.
- Edge connector: Thoroughly cleaned and treated with DeOxit. Then tested with my elkSD128 cartridge.