I’ll also experiment with stripboard to link together the pins on the busbar. If the copper strips will carry enough current this will be a less tedious job of soldering than using copper wires.
2 Likes
yes this method is a far better idea in the outcome totally. you are on the right path and experimentation is the key to success
1 Like
I posted to the wrong thread accidentally, but in an effort to keep things together here’s a link to the AVIATION CONNECTORS!!!1! discussion.
2 Likes
I think the circuit board headers would work, but they don’t really address my unease about flimsiness in power connectors. The GX16-5 connectors I’m investigating are about as heavy duty as I could wish for. I’m attracted by the reasonable price and the solidity of the connection system.
This 5 pin system could provide earth plus +/-12V plus +5V and +3.3V, which are the power levels I’m most likely to want to work with.
Now all I need is a sturdy base. I hate to admit that serious construction work may be in my future, but I’ll start by using my preferred traditional crafting materials to see how far I get before having to mess with drills and saws and the like. Glue and greyboard could, in theory at least, do a better job.
2 Likes
The Noun Project also has useful collections of synth related symbols. You will need to consider the licensing models and appropriate attribution.
“Sine Wave” search
3 Likes
Just when things seem to be settling down I discover the collet I last used in my Dremel has disintegrated and its remnants are stuck inside the drive shaft. Well that’s it from me and collets, chucks only from now on.
Meanwhile I have a useless Dremel until I either replace the motor assembly or work out how to get rid of the ruins of the collet from the inside of the drive shaft.
So my plan for a safe Mean Well enclosure is in abeyance. I’ll go ahead and finish that Kristian Blåsol kit I bought last month. I’m really keen to get back into the rhythm of construction again.
3 Likes
Some sprung doorstops arrived. I took a couple and mounted them on one of my A5 plywood prototyping panels. You can adjust the position along the spring at which the panel clamp holds the doorstop, which introduces a primitive tuning capability. The doorstops are mounted simply by clamping the edge of the panel between the panel clamp of the doorstop and the first loop of the spring. There is also a mounting hole which could be used to mount it more firmly using a bolt or a screw.
I’m using a desk vice to hold the bottom part of the panel down but the top just rattles free, so you’ll see that I held the top using my fingers, which to my ears produced a purer sound.
This capability makes the doorstop more versatile than I expected.
5 Likes
I bought what appears to be a quite good Tack Life kit for less than one-third of what I paid for the Dremel. RTD35ACL is the model.
This brand seems to have a decent reputation so I’m pleased to get it for about £26 on Amazon Prime. Somehow the cheapest price on eBay is £10 more.
5 Likes
Good kit. Motor power is low so match speed to task and it should last.
2 Likes
It appears what I have to do with the broken Dremel is find a close match to the collet size among my drill bits and then get my big scary electric drill out and drill away the remains of the dead collet. Somehow I must contrive to do this while not damaging the steel shaft in which the dead collet is embedded.
Alternatively I can buy a new armature for the Dremel 4000 at a cost of £40. I wish I’d known about the integrated armature when I bought this model. Many other models have separate replaceable motors and drive shafts, which make for relatively inexpensive repairs.
Edit: the technique of drilling out the collet is demonstrated in this video about repairing a different brand of rotary tool with exactly the same problem.
Ironically I chose to get a Dremel precisely because I don’t want to work with big scary electric drills on delicate projects.
2 Likes
I just ordered a couple of these new Bela Trill hubs. They’re £10 a pop. If you are planning a hybrid synth that will be configurable through I2C for external sensors this is a bargain.
Edit: on September 12 they arrived. I2C is a very simple digital protocol so electrically there isn’t much there. The convenience of the Grove connectors, and being able to switch sensors in and out, is the main reason for these hubs.
2 Likes
I finally got up enough time and energy to find the big scary drill and the step bits, and drill ridiculously large holes to fit the glands (yes, that’s really what they call cord grips nowadays) to my power supply enclosure. I will be able to complete this job without special equipment so I’m putting it aside for now while I go to care for my daughter. Any further drilling will be small bore so I’ll use my Dremel (if I ever get it mended) or my Tack Life.
The left gland is for live, neutral and earth coming in from 240V 50Hz mains and the right gland is for the five core cable carrying +12V, -12V, 0V and +5V out from the Mean Well switching power supply shown in place in the photo.
The fifth core will go nowhere but outside the enclosure I will have a stage where the +5V is split into +5V and +3.3V, the latter being fed through the fifth core downstream. Together this five core supply will provide a sturdy power base to Eurorack/Kosmo modules, mignons, microcontrollers, Raspberries, Beaglebones, and external devices.
4 Likes
Last thing before bed time: I got my son to be my witness and safety while I tested the output of the power supply. All tickety boo. Now to muck around with those aviation connectors I found so attractive a few weeks ago.
That’s my old test meter from Tandy’s (aka Radio Shack) a long time ago. I keep it in a nice aluminium case from its predecessor, a Soviet meter I succeeded in blowing up in 1979 because it lacked an overload fuse.
4 Likes
I’m not getting anything done. The impending US election is occupying a lot of my time, which isn’t unusual for me during a Presidential election year. That’s an unhealthy hobby because a lot of the news and analysis comes in when I should be sleeping.
Meanwhile I found a reference on banana jack colour schemes. I may borrow from some of these in future.
I have some low temperature solder paste that I hope will make point to point construction very easy with a hot air gun. As usual I’m very short on round tuits.
5 Likes
You and me both brother. My unfinished projects are looking at me with their judging non-existent eyes.
4 Likes
Googly eyes on every knob.
Speaker driver or piezo under that sheet of googly eyes?
1 Like
Hum, these are way better ideas than what I thought!
I gave eyes to my unfinished modules so they can judge me correctly will I’m being lazy:
7 Likes
Doodle on rotary encoder multiplexing, part 1
I’m following @Caustic’s idea for MIDI control but taking a different path. I want to use rotary encoders in preference to potentiometers. I’m thinking about how to use a small microcontroller to read lots of rotaries.
Rotary encoders are much maligned because they’re not that well understood by amateurs. They’re called encoders because the serial output they produce is not just an arbitrary pair of serial pulses, but a gray code with well defined state transitions. Interpreting the transitions correctly is essential to reliably reading the encoder, and what’s more it’s a lot simpler than you’d think. Here’s a good design. It’s also very simple and economical compared to amateur designs.
So for each encoder I need to read a succession of snapshots of the pairs of pins A and B (they’re sometimes marked CLK and DT on breakout boards). One of the most direct ways to read these is to snapshot a bank of four encoders as 8 bits in a parallel-in, serial-out (PISO) shift register and then transfer the contents to the microcontroller. This can be done using the SPI port.
Now that’s all well and good, but there are two problems. First, this ignores the push button built into many encoders. Second, this is a lot of work for just four controllers. And we only have one single SPI port in many microcontrollers. So that’s that, isn’t it?
Well no. The push button state adds an extra bit. 3 bits per encoder instead of 2. First problem solved, albeit 8 pins can only accommodate 2 encoders instead of 4.
At this point we can add a multiplexer. The CD74HC4067 is an analogue/digital 16 bit bidirectional multiplexer. We can connect a PISO shift register to each of the 16 pins of this device and read their contents by stepping through them one at a time. So to perform the stepping action we need a binary counter to present the succession of numbers from 0000 to 1111 (0 to 15 in binary) at the selection pins of the multiplexer. The counter can be stepped forward by a pulse from the microcontroller.
This is great because it enables us to read a total of 128 binary inputs in just 16 SPI reads. If we decide to go with 3 bits per encoder we can put two encoders onto each of the 16 shift registers and we end up with 32 encoders maximum, or we could adopt a more efficient packing scheme yielding a maximum of 42 encoders.
This needs more work. What’s the maximum operational speed of this setup? Will it surpass the Nyquist frequency for the encoders? How much time does it leave for an Arduino Nano to generate and transmit MIDI sequences?
2 Likes