New Power Supply

 I bought a beautiful Farnell L30-5 PSU - sold as faulty, not working. It's a thing of some beauty, with big meters on it. The former owner was correct - out of the box, it produced no output. The fuses were all intact, but when I turned it on, the overcurrent LED lit up, regardless of the setting of the knobs. Now though I was on a conference call at work at the time, I turned the camera off, carried on listening to the conversation, and got my screwdrivers out. 

So, time to go inside. I found a schematic online, and I found my way to the overcurrent LED driver. It was holding the output transistors in high impedance. It was a fairly simple thing to find that it was being driven by an op amp, that compared two voltages - one generated across a low value resistor in series with the output, and one from the current limit pot. Interesting to see that one of the voltages on the op amp was at or near the negative rail. I measured the resistance between the two pins on the op amp - pins 4 and 5, and it measured a few ohms. I concluded that the op amp had failed, I checked on t'Internet to find a close equivalent, and discovered that a good old TL072 would do the trick, and we have oodles of those around for the synth stuff. I even found an 8 pin socket ready to go in. 

Took the board out - and I wish I had taken pictures, but there was a small area, about a centimetre across on the coppers side of the board, which looked like a fuzzy crystalline structure had grown out of the solder connections. Bright, shiny solder except for this one grey, fuzzy mess - like a metallic mould. 

I've read about whiskers affecting older equipment before - it happens with different metals - sometimes zinc, sometimes lead and with a voltage across. Never seen it - a bit apocryphal I have always thought. I wiped the crystals with a cloth, then with a small bit of sanding block used for cleaning PCBs. I then took a soldering iron to it, just to tidy it up. Measured the resistance again, and the near short was gone. Not even a need to replace the op amp? 

Excellent repair. By the time the conference call was over, I had cleaned the insides with isopropyl, hoovered out the dust bunnies and put the screws back in the case. 

An accidental acquisition with an unexpected addition

 So I bought a Selectric III typewriter from eBay. 

I've been fascinated by IBM Selectric typewriters ever since I began working for IBM in 1977. Actually, my interest in IBM typewriters probably started when I was given a Model D during my University degree, which I took apart, lovingly repaired and typed my final year project on (OK, so Chrissie did most of the typing). 

When I began at IBM, all of the secretaries had Selectrics, and later Mag Card machines with the huge card reader/writers beside their desks. I adored the sound they made. Later I graduated on to using an IBM 2741 with an APL golfball, connected from IBM Hursley to the machine in Santa Teresa to run delay simulations on logic circuits. 

Now, many years later I am nearing (OK, shot well past) retirement age, I decided to indulge myself and bought a Selectric III from that auction site, with a view to making it one of my future retirement projects - once I have repaired several motorcycles, the music synthesisers and other bits of professional music gear.  I didn't pay a lot for it; I put it on my watch list, didn't think much more about it, and the auction finished without a bid. The seller reached out to me with a reduced price, so I was hooked. 

When the beast arrived this week - in much, much better condition than I expected, I noticed that just by the power cord, there was a ribbon cable with a D-shell connector on it emerging. So, against my better judgment I decided just to open it up to have a quick look. The ribbon cable with 20 conductors terminates at a third party add-on - a proper commercial product, made in Denmark - with the number GNT3602 on it. The add-on is fixed underneath the keyboard frame and seems to attach to all of the code levers, the space and carriage return and so on. 

Milli ohm meter now working .. !

 Gosh darn it, it seems to function quite well. 

Milli ohm meter getting warmer ...

So the aforementioned PCB arrived back, have assembled. 

I had to play with resistor values and so on,  but now I have a circuit that will pump 500mA through the device under test. There is a x100 amplifier which boosts up the voltage developed across the DUT to a suitable voltage for an Arduino 10 bit ADC to convert. 

So, a 100milli ohm resistor will develop a full scale reading. I'm going to have to add a range switch to lower the current if I want to measure bigger resistors than that. But then we have multimeters that can go down to a few hundred milli ohms in general. 

In general, quite pleased so far. I am going to assemble this into a bench power supply which happens to have one of my old boards that can read switches, write data to a flourescent display (which is only 30 years old). I am also going to float the power supply ground so we don't liberate magic smoke from things when I am using it. 

Yet Another Transistor Matcher - my version

 I found an article or a blog post on how to match transistors with a bit of slick interface. Matching transistors is pretty essential for exponential converters and ladder networks for filters on synths. Now you can do it manually with precision measurements, or you can use this circuit to show a quite groovy method. 

YATMA

I adapted it slightly - I have a whole drawer of old LF13331 analogue switches which I have used with an Arduino to sequence through, and a x1000 amp on the back side. It takes up to 8 NPN or PNP transistors, and plots their Vbe on a scope. Now since I am an old git and I like to be able to see the components I am soldering,  I am using pretty much all PTH components. How quaint, I hear you say. 

So, I have a small KiCad PCB made by Seeed Studios: 

The small Arduino Pico on the top just runs a small loop that switches two of the 16 switches on, connecting a current source (a chosen resistor) to one of the eight transistors in the ZIF socket on the other side. I had to put the ZIF socket on the back, so I can mount the board backside to the front panel with the socket sticking through. 

The jumpers can be moved to allow either NPN, or PNP (or simpler devices) to be selected. Then, the big BNC socket on the board is used to connect to a scope - I found the small old world scope managed to hold a sync much better than the digital one. 

The net effect is in this bit of a scope picture. I have AC coupling on so the tops of each step slope slightly in this picture - but you can easily see how you can select pairs of transistors with similar characteristics. 2 and 4 in this case are really close, 6 and 8 quite close and so on. It's surprising how quickly the levels change with a finger on the tranny - so it takes a bit of time to settle once the transistors are put in the sockets. 

And it works!  Quite pleased to be honest. 

More Kicad stuff ...

 Playing more with KiCAD these days. I designed a milli-ohm meter (it's not original - it's another of Rod Elliots designs, but I have done a schematic and laid it out. Quite pleased with the results. KiCAD is slowly getting easier to find my way around - clearly a lot of hot key work needed to make it slick. 

Archer 2779601 Power Supply

 So I have an ancient (1986) bench power supply from Tandy in the US, which has a spot on my bench. I have other power supplies for specific reasons, but this beast is 20v, 2.5A with a current limit and seems to survive lots of abuse. I have used it for decades. 

It does suffer from some ageing components though, and I dug around in the paperwork from my stay in the US and found the manuals for it. Someone on T'Internet might find them useful ...

Please find PDFs of the manual (with schematics and wiring diagram) and the assembly instructions, bill of materials and so on. 

Archer 2779601 Power Supply

Archer 2779601 PSU Assembly Instructions

Yet more Minibrute

 So last year I bought a minibrute ... made it kind of work. I did some more work on it recently - replaced the big connector between the analogue and digital boards. That helped enormously with the stability, and functions working. There were a couple of bits left over - the pitch bend didn't work, the LFO doesn't seem to do anything and the aftertouch was flaky. 

I just spent a happy couple of hours sorting out the aftertouch and the pitch bend, and I have now worked out that the issue with the LFO is the waveform selection switch - which is completely open circuit. There is a Mouser part (Alpha?) that seems to measure up to fit, so I will be ordering one in the near future. In the meantime I just wired across the switch to give a sine wave LFO waveform - and now, things are a whole lot better. 

I've done a knob job on it - the sliders were looking very tatty, and I paid a silly amount of money for some replacement rotary knobs. So now it looks like this ... 

More Drumbrute ...

All back together and working ... 

Arturia DrumBrute ...

 So, I was attempting to repair a bit of music electronics yesterday evening. Fault in the power supply somewhere - one of those overload ones, where you turn it on, the lights flash briefly and then all quiet. I hooked it up to a power supply with a current meter, found it was sucking a few amps before dying. 

There's a resettable fuse on the input, which goes open circuit when it fails. 

Aha, I thought. I can tell what is shorting out, since whatever it is will get hot quickly, and then cool down. Having worked my way round the power supply, looking at likely culprits (is that cap a bit bulgy, shall we lever it off the board and retry?) .. I thought I would test the first voltage regulator. It's a tiny thing, (a TPS5431) brings the supply down to 5V pre-regulated before going off to a 3.3v reg, to a switcher which feeds +12 and -12 etc. 

Turned on again, with finger applied. Now that got stupidly hot, very very quickly. And behold, it is showing a resistance of about 7 ohms to ground on the output of the regulator. Trouble is, I couldn't tell if it was the voltage reg or the schottky diode (SS16) next door that suddenly got so very hot. So, I am now looking at the blister on my finger to decide if it is 8mm x 6mm or 6mm by 4mm burn.

Well, the diode is easy to lever one end off the board and have another measure. Ah, it's not the diode. So must be the voltage regulator itself which has gone deady-byes. I hope that whatever happened to the 5V during the time of failure hasn't done too much damage (the 3.3v was still regulating just fine as were the 12v ones).  

I should have been more circumspect and noticed that the voltage reg was showing signs of a little patch on its surface where it has become extremely hot. I will need to lever that off the board now, and substitute in a small external regulator to see if the beast comes up or not. If it does, I might just leave it with a replacement voltage reg on a small board. 

Hot weather killed T'Internet

 My home network has had to be simplified a little, since one casualty of the sudden few days of summer-like weather has been a Netgear switch killer. So, I have now reorganised the wiring to use the other switch in the loft. 

New Keyboard ...

 Have invested in a Logitech Bluetooth keyboard. I managed to trash another Mac keyboard by spilling a minute amount of tea on it. Now that is the second time I have destroyed a Mac keyboard with tea, so I have now invested in a different model... 

This time it's a Logitech K380 .. and a very nice keyboard it is. The symbols and key layout are very close to the Mac ones, and it has a selection mode so I can use it with more than one computer with instant switchover.

Well, it would do if Mint would recognise the Bluetooth adapter I have in Liberator ... 

More Minibrute

So, someone has been in this machine before, and rather brutally broken the connectors. This is what the connector looks like once all of the hot glue (!) has been removed. 

Looking at the other end of the 34 pin cable, I think the hot glue is original; it looks like when they manufacture these things, they hot glue the connectors as an added protection against falling apart? 

So as far as I can tell, someone wanted to get into the machine, has hacked away at the hot glue on the other end of the connection, and in the process have destroyed the connector latches. When they were done, they kind of pressed the assembly together again, hoping for the best. So unsurprisingly, the connector and socket have drifted apart, causing the beast to cease to function. This 34 pin cable is the main interface between the digital board (where the keyboard scanning, CPU, Midi etc is) and the analog board - where the sound comes from. 

Now I have implemented a hack of my own, which is not something I am too proud of but it is a bit of a bugger getting 34 pin 2mm surface mount sockets to suit - so herewith my hack ... 

It works a lot better! Not perfect, but we have some sounds. I also found what was the reason for someone having a go at getting inside the beast. The analog board is covered in a heavily corroding mess of dried diet coke. So, I am gradually working around the board with isopropyl alcohol, and a toothbrush. 

And we have some more sounds ... it kind of makes noises, seems to be largely controllable. The LFO doesn't seem to make any difference to sounds at the moment, but the keyboard all works, the filter filters, the main oscillators and vibrato, pitch bend and the amazing arpeggio all seeem to work. There is some cosmetic work as well  a couple of missing knobs, the slider knobs are all pretty worn and shabby. 

But, all in all, good news - it lives! 

Arturia Minibrute

 Struck again on an Internet auction site - this time an Arturia Minibrute. 

"Turns on but no sound" says the seller. Ah well, might be something silly. Worth a shot especially at a third of the average selling price for a second hand one. 

So, it needs the odd replacement knob or two, and its a bit scruffy. 

And there is a missing screw on the bottom plate. And all of the other screws are loose .. oh dearie me, someone else has been in here. At least it is as described, power light comes on, but nothing - like NOTHING comes out of the jack. 

I was trying to work out how to listen to it, put a short jack cable into the sound interface on the huge 27 inches of loveliness. Moving the boards around to put some insulating material between the digital and analogue boards, and I thought I saw something flash. Was that a burst of sound? Then I saw how badly mangled the "main connector" was on the digital board. Clearly someone has been in this beast before, decided to  try to get the connector out and has broken the retaining clips. Then they decided to stick the connector together with a hot glue gun - badly. 

So I have checked to see if the usual places sell a mating pair of 34 pin, 2mm IDC plugs and sockets. The quickest and most reliable thing to do would be to replace the IDC connector on the cable (using the tried and tested method of a vice to compress the connector and cable together) and replace the motherboard one. I am finding it difficult to locate one of the surface mount kind for the board though. 

Should I solder the 34 connections of the cable to the board ? 

Virtual cameras in Zoom and Teams

 Just leaving this here to remind me how to get Teams to work with an OBS virtual camera again after an update: 

*****UPDATE*****

As of version 1.3.00.28778 the process has changed and the above workaround is no longer valid.  In order to get this fix working on the latest version of Teams, additional commands will need to be entered.  Keep in mind, these commands are relative to where you have Teams installed, so if it is in a different location, you'll need to adjust the file path accordingly.  So as it stands currently here are all of the commands (must be entered one by one and not pasted in one big block) that will need to be entered:

xcode-select --install (only if you haven't already)

sudo codesign --remove-signature "/Applications/Microsoft Teams.app"

sudo codesign --remove-signature "/Applications/Microsoft Teams.app/Contents/Frameworks/Microsoft Teams Helper.app"

sudo codesign --remove-signature "/Applications/Microsoft Teams.app/Contents/Frameworks/Microsoft Teams Helper (GPU).app"

sudo codesign --remove-signature "/Applications/Microsoft Teams.app/Contents/Frameworks/Microsoft Teams Helper (Plugin).app"

sudo codesign --remove-signature "/Applications/Microsoft Teams.app/Contents/Frameworks/Microsoft Teams Helper (Renderer).app"

Lockdown Loaf

 So, I decided during this kind of lockdown that I would have a go at breadmaking. I'd ordered up the ingredients from Sainsbury's last week just in case i felt like it. 

For today, I had put a casserole in the slow cooker this morning and I jiust fancied some olive bread to go with it, so I decided to make some. Here's my first effort at a loaf ...

New lab has a new iMac

 Well, new to me anyhow. Working all day on a small Macbook (though I love it for performance and battery life) is bad for the eyes, so I have furnished the lab with a pre-loved iMac 27 inches ... 

Access to the Lab is now improved ...

So instead of walking across a series of stepping stones embedded in mud, we now have some proper decking outside the lab. It is huge. Well done, Kris - it really does look spectacular. After this, the next steps will be some tidying up at the bottom, and the rockery / wild garden / pond .. looking forward to some post lock-down family gatherings here.

Fixed the Cambridge P500

Another one off the list of things to sort out .. turned out to be one of the darlington output transistors. They have integrated emitter resistors, and one had gone from 0.22 ohm to something in the hundreds of ohms, so it got replaced. While I was there I checked all the caps (no problems) and I replaced all of the heat sink compound, as there was pretty much none on the transistor I changed. Actually I changed both on that channel to make sure I had a matching pair. Applied power, and success! Sound in both.

There's quite a lot of discussion on t'Internet about the bias current, but I found in the spec for the trannys themselves that they recommend 40mA. So I put some meters on, looking for 17.6mV across the two 0.22 ohm resistors. They seemed to want to settle at 20.4mV which is close enough.

This is a budget amp, so I wasn't expecting too much, but the heatsink is a bit pissy and it feels cheap. The whole amp is not worth saving, really but it might do for a bench amp at a push. At least it is now off my bench so I can move on down the list of things to be repaired.

Changing HDD to SSD in an iMac

For a friend; I volunteered to change the hard disk to an ssd in a 2012 iMac. Well, doesn't sound too difficult, does it? I mean, how difficult can it be?

First, there is the getting in to the thing in the first place. It involved delicate prying of the screen away from the back, using guitar picks and a scalpel (very, very gently) to cut through the glue. It fought back, so I did what most people do and went to Youtube for guidance. The main thing seemed to be get it warm with a hair dryer. So, being a very hot May Day, I left the thing out in the sun for three hours to get nice and toasty inside and out.

This got me to the point where the scalpel and guitar picks started to work, and a suction cup gently pulling the front eventually got it to give way ..

So the next step is to fit the SSD in the hole where the old HD was. Now, the old HDD looks like this:

.. and the new one is too small. 

So it's off to Amazon to order up a converter. The first one wasn't the right style to fit into an iMac despite all of the promises on the web site, so I had to have another go. I also ordered up the sticky strips designed to work in an iMac to put the thing back together again. 

All done .. restoring from a Time Machine backup, only took 12 hours for 2TB of data. That has to be quite fast, but clearly benefiting from the fusion drive it was coming from as well. 

Look, Heidi - I have 27 Inches Of Loveliness in the Lab! - if only temporarily ...