I’ve been collecting vintage computers and other interesting technology on and off for years now, and in recent times I’ve become more and more interested in early portable computers. Back in the 80s and 90s I was fascinated by this area, and used to daydream about all sorts of ways of computing on the move. Unfortunately I couldn’t afford to actually indulge most of these ideas but I still remember a lot of the technology and the real sense of wonder and opportunity.

I did however get some hands on with one particular legendary computer back in the 90s which was a Tandy 102. A lot has been written elsewhere about this computer and its place in history so I won’t talk about that any further here, except to say that it must be one of the candidates for being the first genuinely useful laptop computer ever sold. It had built in ROM applications including a programming language, a serial and parallel port, a built-in modem and an excellent keyboard. To top it off it had long battery life and was light and easily transportable.

I was lent it by a friend, and the idea was that I was going to buy it to use it for my GCSEs but for some reason that never happened and it disappeared from my life again. I did have a lot of fun with it for a spell, and it was the first taste I’d had of actual mobile computing. It even came with a disk drive and dot matrix printer. I’ve always wanted to get my hands on one again, and recently an opportunity arose and I jumped at the chance.

I bought it as fully working, and when it arrived it was in excellent cosmetic condition and with a fresh set of batteries immediately booted up to the familiar applications screen.

However, things immediately went wrong because I could not navigate around the screen using the four arrow keys in the top right. This is a major problem! I could launch BASIC by pressing ‘Enter’, but I found quite a few of the main keys didn’t work and also I was spuriously getting apparently random keypresses appearing on the screen as well as the occasional ‘Menu’ (ie F8) command. With a bit more experimenting I found that many of the F keys along the top row did not work either.

So whilst I had hoped for an easy ride it was obviously going to need some work. Fortunately there is a lot of useful information out there, including detailed technical reference manuals published by Tandy. The first step was to have a close look at the keyswitches for the top row.

Dismantling the computer is very straightforward and documented in the manual – a few screws underneath and then the top casing lifts off. The screen and keyboard sit on top of the PCB and are connected by flexible cables which can be gently disconnected. Having a close look at the top row of keys was illuminating:

There was clear evidence of corrosion to the pins on almost all of these switches and it seemed highly likely that at best the solder joints were compromised or at worst the whole switch units had rusted. Desoldering one and soaking it in isopropyl alcohol overnight did not help at all so it was clear that I would have to replace them.

Finding spare parts for things of this age is always a hit and miss affair – some things are readily available whereas others vanished years ago. Keyswitches in particular are often very specific to manufacturers and can be impossible to find. However – the top row of keys are not really keys but more ‘buttons’ and I was very pleased to find that you can still buy identical replacements brand new:

So I was able to desolder the faulty ones and replace them with the new ones, which fitted perfectly and were compatible with the switch tops. On reassembly I was pleased to find that everything worked properly and the cursor could move around freely.

The next thing to do was look at the main keyboard. After some testing it appeared that most keys worked but a few didn’t. I tried hammering on them a few times to see if I could free them up but nothing happened. So after a bit of reading around, I decided to remove the keytops and see if I could do anything. I was able to pull the keytops off and I could see the contacts were corroded:

So after a good scrubbing with some isopropyl alcohol the green rusty substance came off and the contacts were clean and shiny. Once cleaned up and put back together again, they worked fine – so now the keyboard was fully functional again.

The final issue to tackle was the random spurious keypresses. I was getting ‘Enter’ pressed repeatedly, as well some other groups of letters appearing. The excellent ‘Technical Reference Manual’ (why don’t they write these any more) showed that the letters were arising from the same part of the keyboard membrane. There wasn’t anything really obviously wrong but there were some crusty looking joints here and there. I also noticed the problem was worse when the PCB was under pressure (no problem if the casing was off, for example). So I blindly reflowed the solder on a few of these and around the problematic area. This solved the problem, and so I now had a fully working keyboard.

So a good start… but more is to come!

It’s now just over a year ago since I wrote a post about repairing a Lakeland Dry:Soon clothes airer as part of the Repair Cafe which I’ve started in my church.

The response to this post has been nothing short of amazing – through an initial post on Mumsnet and then writing about it on here, I’ve seen a massive response with lots of people describing successful repairs which they have done. I’ve also been able to help people who for various reasons haven’t been able to tackle it themselves, and I now offer a postal repair service (please do comment below if you are interested and I’ll get in touch).

I’ve lost count now of how many I’ve done myself, but it must be about 20 (as I started off with 20 spare capacitors!), and I have been told of at least as many more success stories from people who have followed the instructions themselves and I’m sure there are many more which I’ve not heard about.

It is such a common fault that it’s either a design issue, or perhaps a bad batch of manufacture, but either way it seems to regularly happen just outside of the manufacturers warranty and Lakeland show no interest in helping and just advise people to throw them away, which would be at a real environmental (not to mention financial) cost.

It’s been so exciting for me to have been involved in this, and I certainly had no idea when I started tinkering with the first one what it would lead on. So a big ‘thank you’ from me to everyone who has got in contact with me about this. Please keep the reports coming, and do let me know if you’d like help with a repair.

Meanwhile we have had lots of interesting items come through the cafe and I’ve learned loads about how to repair all sorts of things. So I will continue posting whenever I can with details of what I’ve learned and hopefully this will also be of interest.

Through the Repair Cafe project I’ve been running I’ve been asked about all sorts of things which has led to some interesting projects. Just recently, a friend brought me a Honeywell CM927 central heating controller which had been causing trouble.

This appears to be a very commonly installed device, with a battery-powered wireless control unit linking back to a powered relay at the boiler. There do seem to be various issues with them, one of which is a problem with the connection with the screen resulting in a corrupted display. However, this was different – the screen was fine, but some of the keys were poorly responsive and the screen would regularly blank out for a few seconds when trying to set the time or date. This often happened when trying to get the ‘OK’ button to respond which needed several presses.

I found a useful Youtube video which describes a fix for the screen issue (which didn’t apply) but which does give useful instructions for opening the casing and getting the PCB out:

My first thought was that this was something to do with the power supply, so perhaps a problem with the contacts on the batteries. The faulty buttons were probably due to a poor contact between the button.

The internals are very simple – just a screen, a few chips and a relay and the buttons.

The square pads with the interlocking pattern are the contacts for the buttons. The rubber button in the top casing pushes down, and the back of the button has a conductive surface which bridges the two sides of the contact. Several of these – particularly where the ‘OK’ button connected – were dirty and so I cleaned them up with isopropyl alcohol to give the best chance of a good contact.

On the back of the rubber buttons there was a rough surface which is conductive, and on testing them the ‘OK’ button showed almost no conductivity at all. This is a common fault with TV remotes are other device which use this design, and the solution is to renew the surface using conductive paint. I found something suitable on Amazon and painted the back of the OK button with it.

The two large square contacts on the bottom left and right of the board are where spring connectors from the battery make contact. Looking closely at them, I could see two small holes on each pad where it appears that the springs have worn through the pad. I was pretty confident that this was responsible for the cutting out, due to the power momentarily being interrupted. I flowed some solder on to the pads to restore the connection, and also bent out the springs slightly.

On reassembling everything I was pleased to find that the buttons now worked correctly, needing only a gentle push and there was no more cutting out. So overall a great result for a relatively small effort!

At the most recent Repair Cafe a friend asked me about a wireless subwoofer for a soundbar which was not working. It was completely dead with no signs of life.

On closer inspection it proved to be a Sony HT-CT290 soundbar (which was working fine) and a separate SA-WCT290 subwoofer.

It’s common especially when things are completely dead that the fault lies in the power supply, which often relatively easy to repair (bad capacitors or similar issues). So I was hoping to find something similar here. A bit of quick searching around revealed this to be a relatively common fault with a voltage regulator which steps the 18V down to 3.3V for the control circuits.

Getting access to the circuit board is fairly easy, just a question of removing all the screws on the back of the unit. The PCB is fixed to the rear panel, and there are two cable connectors to unplug from the main board (one for the LEDs and one for the speaker). Then it’s quite easy to see what’ going on:

The large brown PCB is the main power supply board, and the green one has the rest of the supporting circuits. This board steps down the 240V AC to 18V DC. With the board powered up it was possible to check the voltages at various points as outlined in the service manual. The 18V seemed to be getting through OK which ruled out any problem with the brown board.

The chip circled in red on the green board is a TPS54334 step down voltage converter (sometimes called a DC:DC or buck converter) which is a general purpose device but in this case is set up to reduce the 18V input to 3.3V output for the low voltage circuits. The service manual suggests measuring the voltage at one of the connectors, but for some reason neglects to mention the two large ‘test pads’ just next to it (on the left of the red circle in the picture above). The service manual states that if the voltage is anything other than 3.3V to replace the board.

On the one I had, the voltage was 2.90V – so not much lower, but definitely not 3.3V and actually more than 10% out of tolerance. Other people have reported the output as being 0V, but given that the reading was clearly wrong it seemed logical to start there. This chip is a surface mount device and unfortunately I don’t (yet) have the kit or the skills to rework this. There’s a great video where someone does just this:

So I was contemplating doing this, but then I came across an interesting thread on iFixit which describes a similar problem and solution. However, further down the thread comes the genius idea to ignore the SMD work completely and replace the part with a separate buck converter board (which are widely available very cheaply) with the input 18V taken from the power connector (first three pins), and the output 3.3V fed into the test points described above. Rummaging around in my junk box I was delighted to find just such a module left over from an old project, in this particular case based on an MP1584 chip:

These converters all work in the same way, in that you feed an input voltage in one side and then adjust the small screw to set the output voltage. This particular one has a tiny and very sensitive screw, and after connecting it up to an 18V source it took quite a lot of very fine adjustment to get it to output exactly 3.30V.

It was then simply a matter of wiring the input to the 18V supply pin on the power connector (I used pin 1) and the output to the test pads (the positive one helpfully labelled ‘+’). The ground is all commoned together so I only wired in one negative lead. I stuck the module to a spare piece of board with a ‘sticky fixer‘ and secured the leads with tape. I also took the precaution of snipping off all the leads of the old IC although I didn’t actually remove it. This is hard to do as it has a big blob of solder on the back holding it to the board for heat dissipation.

With the board powered up, all the voltages looked good so it was a matter of putting everything back together and looking for signs of life:

I was very pleased to see a green light appear – and everything appeared to be working properly. Not having the actual bar I could not test it myself, but the owner later reported that it was back to full working order which was very pleasing.

Once again, a relatively complex piece of electronics was defeated by a simple power supply problem and the generous provision of test points made it very easy to replace. SMD devices are a pain for a casual hobbyist like myself, although it is now getting cheaper and easier to obtain the kit to work them it’s still a whole lot more difficult. Overall though, a great result and this does appear to be a very common problem with this model. There are also reports of the same thing happening in the soundbar itself.

Let me know in the comments how you get on if you attempt something similar

Update (Nov ’25) – the response to this post has been amazing, and as as you will see there are loads of comments reporting successful repairs of these airers. This article should tell you everything you need to repair one yourself, but if this isn’t something you’d like to tackle I’m offering a postal service to repair it for you. If you are interested please add a comment and I’ll get in touch with you. I’ve not got exact numbers, but I have personally repaired 30+ of these, and at least as many more have commented below that they have had success. A big thanks from me to all who have commented and helped this be such a success. So read on, and let’s save more of these from landfill together!

I’ve recently started a ‘Repair Cafe‘ in my local church. As hopefully is clear from other articles on here I love finding out about how things work, and fixing them when they break. This is good both for my own personal satisfaction, but also it’s good for society at large in terms of reducing waste and saving resources. I’ve also really enjoyed meeting people through it and being able to help them.

At the most recent event, someone brought in a Lakeland Dry:Soon clothes airer. These are quite expensive items, and it arrived looking brand new and still in its box:

However, in spite of the immaculate appearance, it didn’t work. When you plugged it in, the display flashed on and off and there was a constant beeping sound.

Having a quick look around the Internet it appears that this is a common problem – on various forums people were complaining about it, and whilst some were lucky and were able to get their items replaced by the supplier, many others were told that nothing could be done and they should buy a new one which feels like a bad deal for all concerned, and a real waste.

On closer examination these rails are very simple devices, with one small control panel which switches a heater in the rail itself. The behaviour of the panel was strongly suggestive of a problem in the power supply circuit – it is obviously starting up OK (showing ‘on’ on the display and beeping) but then something is going wrong and it’s resetting. I’ve seen this sort of thing before where the power supply is inadequate causing the circuit to reset over and over again.

Getting the control panel out was relatively easy. The first thing to do of course is disconnect everything from the mains. The panel is held in by two screws at the back of the panel, so remove these and you can pull the module out. Turning it over you’ll see three small screws which hold a plastic cover, so remove these too and take the cover off. You’ll then see the underside of the circuit board and you can pull the board out of the plastic housing.

The next thing to do is disconnect the three wires. The outer two are the mains power supply (live and neutral) and the middle one is the output to the heating elements. It’s important to label the wires before you disconnect them. They are ‘FASTON‘ or blade connectors and they have a metal retaining tab which makes them hard to pull off. The best way is to use a very small flat screwdriver to push into the connector and this will release it and then it should pull off easily.

You can now get a good look at the circuit board:

It’s a pretty simple device – the main components are displayed above. On the other side of the board there is a microprocessor which runs the timer and controls, and some additional components to bring the voltage down from 240V and convert from AC to DC. The relay is marked ’24V DC’ and is clearly used to switch the heating elements on and off. I measured the voltage at the output of the rectifier, and it was showing only about 5V which is clearly not enough to switch the relay and confirmed the suspicion that something was wrong.

I did not know a huge amount about how these circuits worked as traditionally you’d have a transformer to reduce the voltage down (in this case from 240V to 24V). However after some reading I identified this as a ‘transformerless power supply’:

https://circuitdigest.com/electronic-circuits/transformerless-power-supply

The link gives lots of detail, but in essence the key component is the ‘X2 capacitor’ which is the yellow rectangular component above. If this fails, then nothing will work properly. They often do fail, and when repairing vintage computer or audio equipment I have frequently found old ones to have exploded and generated a lot of smoke and a horrible smell.

I also found an article about a similar timer / control board where the X2 capacitor proved to be responsible:

https://forum.allaboutcircuits.com/threads/fault-with-ac-to-dc-using-db107-bridge-rectifier.138563/

I decided to try replacing it. Fortunately these are cheap and readily available. The original one is marked 0.39uF at 275V. It’s also important to note the physical dimensions when looking for a replacement to make sure it fits properly as there is not much additional space on the board. The pitch (ie distance between leads) is 15mm. I looked everywhere and I could not find a 0.39uF part which would fit, but plenty at other ratings. The exact rating of these components is not critical (especially voltage, which must not be any lower but can be higher) so I ended up buying a 0.47uf at 315V version instead on eBay, for the princely sum of about 31p each (when buying 10).

Replacing the part is a simple soldering job – desoldering and removing the old and replacing and resoldering the new one. I was pleased to find that the new one was absolutely identical in size and appearance to the original and it was easy to solder in to place.

So then it was simply a question of reassembling and testing. I connected the wires, switched on and…

Success! I could hear the relay engaging, and the display is clear and steady with no beeping to be heard. There a few odd things which on reading the manual are by design – the main ‘on / off’ switch lights up, but only when the heating element is off – apparently to remind you that you to press it. On fully reassembling the unit I was pleased to find that it now all works perfectly.

I shudder to think how many of these have been consigned to the tip because of such a simple and easily remedied fault. It’s likely either a design fault, or perhaps a bad batch of capacitors but either way I feel that the supplier could do better than telling people it’s not repairable and that they need to buy another one.

So if you have one of these with this problem – I’d definitely encourage you to consider attempting the repair (or finding someone to help you with it). If you have a local repair cafe or similar then the information in this post should allow them to help you with it too.

Please do let me know in the comments if you have experienced this or if you want any help.

So the next thing to do was the strip the unit down and look for the culprit. As stated above service manuals are readily available these days (I found mine here) and I was able to find the right one on archive.org. Service manuals are strange things – having seen quite a few now they are very variable in how much detail they give, and whilst obviously they are aimed at people who are supposed to know what they are doing the variation is quite puzzling. So this particular one is very helpful although does miss a few specific points which I’ll describe here.

The disassembly diagram shows a couple of ways to remove the mechanism – in view of the complexity of the device there are two separate circuit boards as well as the mechanism itself. You can either remove the main board, and then the mechanism, or else the whole ‘mechanism deck and main board block’ together. The latter of these is what you need to change the belts… but I didn’t realise this until too late and did the former. It doesn’t really matter but does mean removing more screws and connectors than you actually need.

Secondly and most importantly the manual does not tell you the whole story about the flexible ribbon cables of which there are two. The manual simply shows that they need to be removed (CN303):

So I just pulled them out, assuming that they were the variety that were held in by friction rather than any kind of locking mechanism. However, when I came to put it back together again I could not work out how to push them back in again. The cables were very floppy and despite a lot of effort I could not get them to engage in the sockets. I was beginning to get worried as I could see there was a high risk of damaging the cables and being left with a completely non-functional unit.

After a lot of messing around and getting nowhere I finally had an insight when I realised that the ends of the ribbons were completely floppy and not rigid at all – so were obviously designed to work with a locking connector. Looking closely at the connector I realised there were two black tabs, and these could be eased forward which released the clamp. Once this was done it was simple to slide the cable back in and close the clamps again. I really did have a lucky escape here, both when removing the cables and fruitlessly trying to push them in again. Other Sony service manuals clearly explain how the connectors work so it’s frustrating that this one did not do so.

So I ended up with the ‘mechanism deck’ and the main board separated out which allowed me to get to the source of the problem.

In all the excitement it seems that I did not get pictures of the crucial part of the mechanism, which lies under the large metal plate in the picture. You can see the floppy ribbon cables and hopefully appreciate how delicate they are.

The diagram in the service manual is quite helpful though:

It was apparent on getting a look at it that the belt ‘FR-A’ was extremely loose, and in fact had a very obvious flat spot on it where it had sat probably for 20 years. This flat spot accounted for the strange ‘pulsing’ effect I was getting. It was a simple matter to replace it with a suitable modern equivalent.

I had a good look at belt FR-B but it appeared springy and in good condition. Replacing it was a bit more involved, with the counter mechanism needing to be removed. I had a good look at this and did in fact remove everything but I didn’t have quite the right belt, and in any case it looked good.

So it was now simply a matter of putting it all back together again – as above this led to some nasty moments, but eventually I had it all together and on testing was pleased to find the fast winding in both directions worked smoothly.

So overall I’m very pleased with this item and with the result. I’d love to get it on a wow and flutter meter but I don’t have one (or the right tape) but there are apparently some ways of doing it on the computer that I’ll need to have a look at. Still I’m very happy to have a rare bit of Sony history to look after.

I was at the ‘Retro Tech UK’ fair recently (which I would highly recommend) and on my third time round the room my eye was caught by what appeared to another variant on the ‘slimline’ design but was a little more expensive than the usual few pounds, and also had some odd looking controls. The seller immediately sprang into action, and told me that it was the “very unusual” Sony TCS-2000. I had never heard of this before, but he explained that the capstan (ie the thin metal rod which is used to drive the roller which feeds the tape through) was driven directly by the motor (rather than connected by the usual rubber belt – an universal point of failure in old devices) and that as a result it had lots of clever tricks.

First of all it mean that the tape transport was extremely accurate and reliable with the minimum of the dreaded ‘wow and flutter’ (being variations in speed which often result from slack or slipping drive belts). Secondly though it made some very special functions possible – such as running the tape at four times its normal speed, reading the high speed audio in and through digital processing playing it back at normal speed in 8s chunks. This means that you can scan through a tape much more quickly than usual but yet listen to snatches of audio played at normal speed helping you to work out where you are in the recording. You can also record index marks as you record and can skip between these again with high accuracy. The seller told me that the main market for these had been ‘the security services’ where they had been used to record interviews, and sure enough it came in a sinister-looking black vinyl case. That clinched it… although as I told the seller “you had me at ‘very unusual’”.

When I got it home, I did look around on the Internet but could find very little about it. There is one article on ‘Walkman Review’ which gives a good (if rather brief) summary of it, but beyond that – nothing. I did find the service manual, and a couple of mentions in catalogues from the time but little else and sadly nothing to corroborate the ‘security services’ story although it does sound plausible. What was interesting was that no other portable Sony cassette – not even the famous WM-D6 range – used this direct drive approach and this must make it one of the most technically advanced and accurate portable cassette players ever made.

I was very impressed by the sheer quality of the build. The controls are largely mechanical (rather than the ‘full logic’ electronic controls more commonly seen on high-end gear) but the keys are brushed metal and feel extremely solid. There are lots of interest looking buttons and switches on it, and very unusually for a unit like this it has two built in microphones providing for a stereo recording. One irritating – and important – thing is that like a lot of modern slimline units (but unlike the ‘classic’ ones from the 80s and before) it does not have a built in power supply and so you need to use either batteries or a 6V supply via a barrel connector. This is a ‘centre negative’ polarity – for some odd reason, as most are ‘centre positive’. I have found some audio gear (including synths, effects units etc) which have the same polarity. So if you don’t have the original Sony power supply (which unfortunately I don’t) then you have to be careful as most power supplies with this connector are wired the other way around.

So with it powered up properly, I was pleased to find everything working. I couldn’t resist hooking it up to my Spectrum and loading a few tapes and as expected the signal was rock solid and the games loaded perfectly. So whilst playback worked perfectly I did notice that on fast forward or rewind the tape wasn’t winding smoothly, and instead was moving in bursts suggestive of a worn or slipping belt. The motor for the winding mechanism is separate from the drive motor, which is again one of the unique features as usually it is the same motor for all functions.

So the next thing to do was strip it down and look for the problem… but this article is long enough so the story continues in Part 2.

Over the last year or so I have really got into cassette players. It started with my interest in 8 bit computers many of which needed a domestic cassette player to get started. The fancier ones (eg the BBC Micro, Atari 400 / 800 etc) had officially branded (and no doubt very expensive) units but the more affordable ones (like my first computer, the Sinclair Spectrum) did not. For this and other reasons there was a huge market for low cost ‘compact cassette’ players, quite different from that for the enthusiasts ‘high fidelity deck’ designs. There is a whole separate history about the rise of the ‘compact cassette’ itself which has been extensively covered elsewhere so I won’t go into that here.

What is evident is that there was also a clear interest in self-contained units with a microphone and loudspeaker built in which were reasonably portable and could be run from batteries (in many ways the fore-runner of the ‘Walkman’). I’ve not really done enough research on this yet, and so I’m not sure quite when it started but the form factor which emerged was a small rectangular unit, similar in width to the cassette with the speaker above and a series of mechanical controls below. These often were referred to as ‘slimline’ designs.

I began picking these up for a few pounds here or there initially because I wanted something to for loading cassette games into my Spectrum, but after a while I became fascinated by the sheer range and variety of these devices, and their longevity. The earliest one I have found dates I think from the late 1960s and is a Philips model. I actually have 2 of these, bought on successive weeks from different sellers for the local car boot. On a recent visit to the official Philips museum in Eindhoven I was very pleased to see an identical (although much cleaner!) model.

They were certainly still popular in the mid to late 1980s fuelled no doubt by the home computer boom and possibly beyond that. I’ve lost count of exactly how many I have – probably 20 or so – and every one is different. Some are clearly bargain basement designs (such as those sold by Dixons in the 80s) whereas others are heavyweight and clearly serious pieces of kit with multiple controls, or else ‘ruggedised’ with bomb-proof thick plastic cases which were probably for educational purposes. I have some others too of different designs including the hifi style decks, and whilst I do like those too I think the slimline units are actually more interesting, more varied but yet much less regarded or documented.

As I’ve written about before I am interested in old technology, primarily computers from the 8-bit era but also other things more generally. Lately I’ve rekindled an interest in old audio gear as well, particularly cassette recorders but also other things too. Cassettes seem to have experienced something of a renaissance lately, not on the scale of vinyl of course but definitely more than previously. I grew up with cassettes, never owning a record player (well not really… I did have a 1960s ‘picnic basket’ one for a while which I bought for 10p at a jumble sale) and relied on them well into the CD era for music on the move. So I do have a real affection for them.

I’ve always enjoyed car boot sales and I’ve been going to these more lately as well, and I’ve managed to pick up a whole range of interesting stuff, most of it in various stages of disrepair. What has been interested is how little documentation there is on this stuff out there – you can often find service manuals, but rarely anything else.

So I’m going to use the blog to document what I’m finding, what I’ve needed to do to bring it back to life and any lessons learned, and also what I think about them. Given the level of interest in retro these days, there is still a huge amount preservation to be done and I’m hoping to play a small part in doing that. I’ve not been very active on the blog lately but I have a lot of stuff to talk about so let’s begin!