Farewell to 2023!

As the year 2023 draws to a close, I wanted to thank all of you – viewers, subscribers, supporters, and those who have bought anything from me, had anything repaired/serviced by me, or donated/sold anything to me – for making 2023 a great year.

I’m sending all of my best wishes to you and your loved ones for 2024 – I hope you get everything that you want out of it and more, and are able to spend it in safety and comfort.

Now that the sobby stuff is out of the way, let’s get down to business.

I didn’t post as much as I wanted to on my website as I’d have liked this year (though I’ve been quite active on Twitter as usual), however it was still a good year for activity – clocking in at nearly 70,000 views and 39,000 viewers across 148 countries.

I’ve still been plenty hard at work in my workshop, so there’s lots of website content for me to catch up on, including some of my favourite repairs/restorations from the past year or so.

1985 Nintendo Entertainment System (NES)

For my birthday this year, my wife bought me a NES, which I’ve since fixed up – new cartridge slot, switch-mode voltage regulator, full recap, and region free modification.

1993 Apple Macintosh LC-III

I bought a boxed Macintosh LC-III and matching CRT display a couple of years ago, and this year I got around to refurbishing and upgrading it – mainboard and power supply recap, upgraded system RAM, upgraded VRAM, a 68k co-processor, and an internal BlueSCSI.

1989 Apple Macintosh SE/30s

I’ve worked on three Macintosh SE/30s this year:

• #1: A good tune-up for a friend’s machine – a digital board and analogue board recap, FDD service, and internal BlueSCSI.
• #2: A good tune-up for a friend’s machine, the same as the LC475 and 12″ CRT – a digital board and analogue board recap, FDD service, and internal BlueSCSI.
• #3: My own machine, damaged badly by battery leakage. I’ve transferred all of the original components over to a replica SE/30 PCB kindly provided by CayMac – it’s not completely working yet, but I’ll get there in time!

1997 Apple Macintosh Performa 6400/180

I bought a boxed Macintosh Performa 6400/180 with matching CRT display early this year, which needed some work – a power supply repair (the TOP200YAI PWM controller had failed short-circuit and the 1R 0.5W fusible resistor at R4 had blown – I fitted a TOP204YAI part as a replacement as this is rated for higher power, apparently a TOP224YN part also works – I also replaced all of the original electrolytic capacitors while I was at it), a new SCSI CD-ROM drive as the original wasn’t reading discs, a new IDE-CF internal drive as the original HDD had failed, and a new CR2032 PRAM battery as the original was missing.

1981 Texas Instruments TI-99/4A

This year I repaired a Texas Instruments TI-99/4A vintage home computer, which apparently worked OK but the screen starts to deteriorate to garbage after running for a few minutes. I figured that this could be a thermal issue with the power supply, RAM, or video chip starting to fail as they get hot over time – this was backed up by the fact that with the case removed and running in a cool room, the bare electronics alone didn’t exhibit a failure even when running for a couple of hours.

I disassembled the unit, reseated all of the socketed ICs and cleaned their sockets with contact cleaner, and cleaned an excessive amount of thermal paste off the video chip. I noticed that the PSU board was running hot – this is a hybrid supply with a switching system for the +5Vdc and -5Vdc rails and a linear regulator for the +12Vdc rail, and the 78M12 linear regulator in particular was running very hot.

I decided to remove the linear 78M12 and its heatsink and replace it with a Traco TSR 1-24120, a drop-in switch-mode replacement which is up to 96% efficient and runs cool so doesn’t require a heatsink, to try and reduce the amount of heat build-up in the case. I tested the PSU under load and adjusted the voltages and it seemed to work OK, with the +5Vdc rail at +5.05Vdc, +12Vdc rail at +11.93Vdc, and the -5Vdc rail at -5.14Vdc.

I refitted the case (with new thermal paste on the video chip) and ran my testing again, and this time the issue started to occur at around the hour mark – a significant improvement due to the PSU running cooler now, but it was apparent that the root cause remained.

I noticed that when the video problem occurred, the computer didn’t actually crash – the keyboard still worked and you could see the cursor running in the background, it was just that the video became corrupt. If the RAM was the issue the computer would most likely lock up or become unusable, and if the power supply was the issue it would also cause problems with the RAM (as 4116 DRAM requires +5Vdc, -5Vdc, and +12Vdc to function) – the video chip (TMS9929ANL) was running hot, so I figured this was the most likely issue.

I tried a couple of apparently-working replacement parts, and the symptoms seemed to change – I therefore replaced the video chip with a working one, and for good measure I also replaced the video chip socket, which was quite poor quality. I also replaced all of the electrolytic capacitors on the mainboard and power supply, for preventative maintenance.

With the computer reassembled (again, with new thermal paste on the video chip) it was still working happily after soak testing for over four hours, so it seems that the cause of the problem had been fixed.

1991 Acorn RISC PC 600

This year I got an Acorn RISC PC 600, with a 200MHz StrongARM CPU card and 486 PC card – this had minor battery leakage and needed repairs, and also got mainboard and daughterboard recaps, a power supply rebuild, an internal IDE CF card with RISC OS 3.70 and Windows 95, and a 16-bit audio upgrade.

1984 Amstrad CPC464s

This year I repaired a pile of Amstrad CPC464s, with various issues – some completely dead, some with video issues, and some with tape loading problems.

CPC464 #1 didn’t output any video: I tested the socketed CPU, ULA, ROM, and RAM ICs, and all tested OK; I checked the board over using an oscilloscope – the CRTC was running warm, and the CRTC data lines seemed to be floating; the reset line was OK, but there was no activity on the CPU data and address lines; I checked the CPU clock, and it wasn’t present.

I tested this back to RT3 (ZTX312) which buffers the CPU clock signal, and the signal was present on the base, but not the emitter of the transistor; I removed RT3 and tested it out-of-circuit, and it seemed to have failed; I replaced with an equivalent ZTX313 part, and the computer seemed to work OK.

CPC464 #2 had an intermittent power switch, so I dismantled it and cleaned the contacts using contact cleaner and emery paper, then it started to work reliably.

This computer also wouldn’t read tapes on the internal deck – I cleaned the read/write heads & idle pulley and replaced the drive belts for the tape drive & tape counter, which are all common problems – I also noticed a broken rubber tyre on the take-up spindle, which I removed and replaced using a suitably-sized O-ring, which worked great.

CPC464 #3 was completely dead with no video output – on disassembly it seemed to be a later board with an SMD ASIC and only two 41464 DRAM ICs (compared to the eight 4164 DRAM ICs on older boards).

Both RAM ICs were burning hot, so I socketed and replaced them and tested again, but still no video output, and the CPU now ran extremely hot; the CPU was already socketed so I removed it and tested with a known-good CPU, the unit now gave a flicking black screen, and the ROM now ran extremely hot; I socketed and replaced the ROM, and the computer started to boot OK.

This unit seemed to have been damaged by someone plugging in an incorrect PSU, which is a common problem with the CPC464 – people see a DC jack and just shove any power supply into it, regardless of its voltage or polarity or even current rating. The ICs inside the CPC464 are powered directly from the 5Vdc regulated power input, so any incorrect voltage goes directly onto them, and can cause major damage. Thankfully the SMD ASIC on this board was OK, as these are difficult to find replacement parts for, and are very sensitive.

CPC464 outputted video but only displayed a garbage screen. I tested with a known-good CPU and ULA, but no change in symptoms; I removed and tested all eight 4164 DRAM ICs, one had failed; I installed sockets and refitted the original 4164 DRAM ICs except for the failed on which I replaced with a known-good part, and the computer started to boot OK.

CPC464 #5 seemed to boot OK after cleaning out the power switch, however the colours seem to be off on the video output, and the video signal was intermittent.

I dismantled the unit for inspection, and it appeared that the solder was cracked on the video connector; I removed the original solder from the video connector and re-soldered it with fresh solder, and the video output now seemed to work OK. I repeated this process for the rest of the connectors on the mainboard as the soldering seemed to be poor quality.

Along with these repairs, I usually perform preventative maintenance on all my CPC464s, including: power socket replacement, and reflowing all other sockets; replacing the thermal paste on the ULA heatsink; cleaning all ports and edge connectors, power switch, and volume control with contact cleaner; stripping and cleaning the unit, including the keyboard; servicing the tape deck, including cleaning the read/write heads & idle pulley. and replacing the drive belts for the tape drive & tape counter.

1989 Amstrad PCW8256

This year I repaired and upgraded an Amstrad PCW8256, which needed a 3″ FDD service (head clean and drive belt replacement, as these have nearly always deteriorated and snapped) and a general tune-up, and also got an upgrade from 256k to 512k system RAM – this involved populating the eight empty RAM sockets with eight working 41256 DRAM ICs and setting SW01 to configure 512KB RAM (A=off, B=on, C=off, D=on).

1983 Commodore 64s

I haven’t worked on as many Commodore 64s as usual this year, only two – both for friends of mine, both requiring repairs – one didn’t output any video (bad VIC-II) and then only displayed a black screen (bad 6510 and bad 6526), and the other was a garbage screen (bad MOS-brand glue logic ICs).

I’ve already written about repairing/restoring my own Commodore 64, and repairing most generations of Commodore 64 mainboards (ASSY 250407, 250425, 250466, and 250469).

1994 Amiga 1200

I’ve only worked on one Amiga 1200 this year, for a friend of mine, which needed a mainboard recap and general clean-up.

I’ve already written about restoring my own Amiga 1200.

1994 Pioneer CLD-D515 Laserdisc Player

I volunteered at the repair café at the Billingham Festival of Thrift 2023, where I successfully repaired a lovely Pioneer CLD-D515 laserdisc player.

2012 Nintendo Wii U

I volunteered at the Climate Action Stokesley & Villages (CASaV) repair café at Swainby village hall in November 2023, where I was able to repair a Nintendo Wii U console for a very grateful mother and twin daughters – the unit wouldn’t connect to controllers, and needed a new Bluetooth module fitting which the owner kindly provided.

1975 Sinclair Cambridge Calculator

This year I fixed a Sinclair Cambridge calculator for a friend of mine, which was not powering on – it had suffered from battery leakage in the past, and needed a good clean up.

1954 CT52 CRT Oscilloscope

I bought a beautiful military CRT miniature oscilloscope untested from Kiwi Trading in Hartlepool – it needed replacement of all unreliable vintage paper/wax, oil, and electrolytic capacitors (over 100 in total, this thing was crammed full!), replacement of three unreliable vintage selenium rectifiers, and fitting of an IEC mains input (the original was a non-standard screw-in DIN connector).

1973 Armstrong 626 Amplifier

I fixed a gorgeous Armstrong 626 stereo amplifier for a friend of mine, which was partially working but needed a service and repair – a mains filter needed to be removed, three bulk capacitors needed to be replaced, two fuse holders needed to be replaced, all of the controls needed to be cleaned, the dial lamps needed to be replaced, and the right audio channel needed to be repaired (two shorted transistors).

1985 Fidelity IS-650 Turntable

My father-in-law gave us a Fidelity IS-650 hi-fi cabinet, his from back in the day, which had been in storage for decades and wasn’t working – it needed a good clean and general tune-up, including a new turntable drive belt and stylus, speed switch repair, speed adjustment, tape deck service, partial recap, and cleaning all of the controls.

1961 Ever Ready “Sky Master” Radio

I bought an Ever Ready “Sky Master” AM transistor radio from Tynemouth Market – it was working but needed servicing, including replacement of all wax/paper and electrolytic capacitors, and cleaning all of the controls

1954 Philips “Music Maid” Radio

I bought a Philips “Music Maid” AM tube radio from Kiwi Trading in Hartlepool – it was working but needed servicing, including replacement of all wax/paper and electrolytic capacitors, and cleaning all of the controls

1962 Bush TR82C Radio

I bought a cheap Bush TR82C AM transistor radio on eBay, as I really like the iconic look of these – it was working but needed servicing, including replacement of all wax/paper and electrolytic capacitors, and cleaning all of the controls