Busy hands | Disquiet mind

I've had a bunch of watches sitting in a drawer for years now. These are the ones that survived; loose occasion wear, and those with leather protecting the metal. (Strangely my skin rots the backs of watches.) They're not expensive watches, just mass-market fashion watches. But having given up the fitbit lifestyle a long time ago, I feel drawn to some wristwatch action.

( Details )

This project was absolutely an exercise in persistence.

I had previously restored my parents' old turntable, but the plastic hinges on the lid were broken, which made it awkward to use long term. Even the stereo system in that post got shuffled around and packed away in the shed as not-strictly-necessary.

Fast-forward a few years into a more dystopian future and an ongoing pandemic, and we find me, spending a lot of time at home, looking for projects to distract myself. I had recently gotten a 3D printer for the first time, and I came across a kickstarter for the SongBird turntable kit. They would supply the hardware, electronics, acrylic, and 3D models, which you then print yourself at home. This seemed ideal to me, because if anything ever broke I would (or at least should) be able to replace just that part easily. I backed it in May.

The kickstarter campaign fulfilment was a bit delayed (by the creators catching covid), but not for long as all the parts went up on MyMiniFactory in October, followed by a few updates (that resulted in a naming schema I could not follow). I read the instructions, and decided to not make things difficult for myself and to simply use PETG like they do. I'd never used PETG before, so I did a bit of research and ordered some filament (ROSA3D PETG Standard Light Green and ROSA3D PETG Standard Black) and a Spring Steel Sheet applied PEI Flex Plate for my printer. They arrive, and I print out all the parts before the end of the year.

( Parts list )

But we all know a turntable cannot be used on its own, it needs speakers, and a preamp, so I have to dig out, and clean up, my old stereo system and preamp. I found my old Cambridge Audio 640P preamp, and couldn't find the power plug. So I order a replacement. All this takes a while. June rolls around and I, at least, finally get all the parts assembled into a turntable!

All parts assembled
Photo by chebe

I also needed furniture to fit the stereo into my office, which had to be found, ordered, stained, and assembled. Which required rearranging other things, changing power strip layouts, and then I get covid. This takes a while.

In the instructions they say we need a 'Rega Baerwald alignment protractor V2', and they provide a link to a forum, that is shutting down leaving the file inaccessible. I bought an alignment protractor, but I'm glad backers are sharing copies of the file in the comments now (not least because the place I got my alignment protractor no longer seems to be stocking them).

I finally get enough energy to do the rearranging and I realise I do not have two sets of phono-to-phono cables, and that I really need two sets of phono-to-phono cables. I also find out that I have to order them, because nowhere locally physically stocks them?!

Everything arrives, and I'm setting it all up, going through the fiddly calibration guide, and, it's not working. I get the multimeter out and take the turntable apart. But everything seems okay. I try the stereo system AUX with a portable CD player, and that's working. I even test the brand new cables. That leaves the preamp. The power light comes on, but there's no sound coming through. So I order a replacement, the Behringer PP400 Phono Preamp that they mention in the instructions. Which means I can also print the preamp mount part (pa1mk3-1), to tuck the preamp under the metal arm.

Printed in black;
- pa1mk3-1, at fast quality (3 hours), doesn't look great, but does the job. (Slicer also complained model had problems that it repaired.)

A month later the new preamp arrives, which I attach with two M3 x 5mm screws, and I wire everything up. Finally I can listen to some vinyl.

( Pictures )

Close up of needle, just because
Photo by chebe

Clip of turntable in action
Video by chebe

The BYO Keyboard kit I backed on kickstarter arrived. Time for some soldering.

How about 19 minutes of unedited real-time soldering, with a camera that is struggling to maintain continuous auto-focus with a pancake lens? There's some nice, relaxing music. On repeat. It's a chill time. (Okay, I didn't realise the camera was wobbling so much. It's my very first time. Lots to learn.)

( Video )

But after soldering we have to program it. It's Circuit Python, which means updating the version on the device with the newest one for your board (ItsyBitsy M0), by double-pressing the Reset button and dragging the file onto the disk. And then downloading the libraries to match that version. Extracting the zip, and drag-dropping the libraries you need onto the disk. (More on the libraries here.) One day I'll remember all that without looking it up.

I went for the keypresses demo script, so now I'm sending lost 1s-6s into my computer. Success!

BYO Keyboard kit, with Media keycaps, soldered and programmed
Photo by chebe

I find myself wanting to use the same hardware controls in different projects again and again. But in various configurations. Can I make something modular, that I can reuse as needed?

Let's take the Colour Stealing circuit, actually let's take the version with a microphone.

( Details )

Here's the whole system; wrist controller connected to power regulation and signal level shifter board and NeoPixel matrix. For the longer runs of wire I'm using thicker silicone wire that I have loosely plaited.

Overall it has the same problem as watches, in that the heavier section has a habit of shifting around, but if it annoys too much I suppose I could always stick it in place. The nice thing is that it all disassembles cleanly, making everything (not soldered to the protoboard) reusable.

Wrist Controller in the middle of it all
Photo by chebe

The LilyPad Accelerometer (ADXL335). I've been trying to use this for over ten years. Turns out I've misunderstood what it measures for 10 years. I had partial success with just the x-axis using Leah Buechley's code, but it has since disappeared, so I can't even be sure what it did.

I was trying to do something else, and the numbers, very strangely kept coming out to an exact 100.0, which is very wrong. So I went back to make sure the accelerometer was giving me a full value range. Which it didn't seem to be, so I went researching and found this excellent tutorial.

Although the accelerometer gives you raw values between 0 and 1023 on an analog read, it actually measures gravity acting on it as it moves around in 3D space. Converting the raw to Gs is straight-forward;
long xAcceleration = map(xRawPinValue, 0, 1023, -3.0, 3.0);

To then turn this into a colour, simply, you could map the range onto -255 to +255 (the polarity demonstrating direction of change), and then reducing that to just value of change;
int xRed = map(xAcceleration, -3.0, 3.0, -255, 255);
int absoluteXRed = abs(xRed);

Do the same for the y and z. Write to RGB LED/pixel of your choice. (Warning; movement in video is quite sudden.)

( Short video )

Sorry for the rough prototype version, I dismantled the version I made up for parts to make my tiara.

Other than light-up gloves, this could be used for rough gesture control, but doesn't give me the kind of information I needed for the thing I was trying to do as is.