Forum Replies Created
-
Posts
-
Hi
Great work. Music theory is definitely not an easy subject. It was a while ago, but if I’m not mistaken, I used a formula for natural scales and major/minor chords, (not harmonic, diminished or dominant, but I wouldn’t be surprised if there was also an error in my implementation somewhere ;)
I think that CEGBD is the correct order for a major ninth:
https://en.wikipedia.org/wiki/Chord_letters
What scale and chord do you prefer?
It is great if you have improved the code. I propose that you add a pull request in github so anyone can use your solution and compare the changes!
Then I would focus on the lower LED-strip.
- Check that it is oriented correctly (signal in / DI to the right and GND on the bottom.
- Measure connection between DI on the LED-strip pad and pin2 on the arduino. (using continuity tester on a voltmeter)
- Measure connection between +5v on the LED-strip and VCCon the arduino
- Measure connection between GND on the LED-strip and GND on the arduino
You can also connect it to a computer with a DAW software running (handling the midi events), then the tx led should flash for every key press. (Verify this behaviour first using the working midi keyboard)
They are less common and more difficult to find. It is a matter of choice, but I prefer the lower activation force. Here is one example (I have not tried that particular model thought)
https://www.digikey.com/en/products/detail/e-switch/LL1105AF065Q/3777946
(Digikey allows you to sort on activation force)
Strange. I just downloaded the code and it works as expected.
Here are some questions that helps narrow down the source of error:
- Were you able to upload the program (double check that you use the right version)
- Do you see the LED animation at start?
- Does the right red led flash when you press any of the eight pad keys?
The leonardo boot-loader should not cause any problem.
Hi
I just posted LEET Linux – the fruit of all great ideas from this post.It is powerful, cute and sounds great (and you can build your own ;)
I just posted LEET Linux documentation with all files needed to build your own:
https://vonkonow.com/wordpress/leet-linux/
There are lots of sources for the trimpots, try ebay, amazon, wish or alibaba and you should find them. (I bought a kit with 100pcs)
The magnets can also be found thru the same resellers:
Hope it helps!
The ESP32 is certainly an interesting (and powerful) platform, and I considered using ESP32 in the beginning, especially for the possible MIDI over Bluetooth. The reason I went for the pro micro was mainly cost, size and that it has enough I/Os and performance.
However, the S2 has a lot of interesting features for future devices:
· USB host capabilities
· Lots of performance (ram, flash and cpu)
· I2S & 2x DACs
· WiFi (configuration over html)
I see no disadvantage to rebase the code, but for a simple keyboard, I see no real benefit either ;)
Hi
I considered using 3.5mm connectors for serial midi in the beginning, but decided to use USB since it didn’t require external connectors and components, used I/Os or caused (minor) delays if using midi thru. It is possible to use one hardware serial port and one software serial port.
If you use usb to provide power, it’s a lot of extra wires. In theory you can also skip optical isolation and provide power over the fourth 3.5mm connector, but then you have to avoid connecting it to other TRS midi devices without a buffer.
Personally, I think it is better to use a MIDI bridge to connect it to other devices. (It is rather easy to build one using a microcontroller that supports USB host mode.)
Hi
Great timing! I’m just about to document LEET Linux – an improved version of the unit above. You can see it in action and hear how it sounds here: https://youtu.be/RMyfaAVU64k
I have uploaded the 3D models in development: (will crate a dedicated repo once the documentation is done).Let me know what you think!
It’s not ideal but the best solution I found. Options around it:
· Remove the top of the thin wall (will expose the potentiometer)
· Move the trimpot up (the knob will not be centred)
· Rotate the trimpot (makes routing more complicated)
I can probably reduce the distance to the trimpot by 0.1-0.2mm, it will be slightly better, but I’m not sure how standardised the trimpots are (have tested with two different versions). If anyone have any proposals, let me know!
It worked ok-ish for me when I printed with PLA (0.4 nozzle):
I use to fold the protective liner and score the adhesive underneath gently with a knife (without cutting of the copper traces). Then remove the adhesive over the pads, before soldering.
Another way is to remove all adhesive and use a stronger one to keep the FPC in place.Awesome (a bit jealous about the pitch and double sided routing that comes with a real PCB ;)
Nice bonus to have the option to chose between separate LEDs and LED-strips! (but the LED strip seems to be missing data in & out – in)
I guess you are using a ground-plane for switches and decoupling caps (not visible in the picture).
Looking forward to see the build!
I have redesigned the control with a version using trimpots instead (they work great):
Ouch, good to know (I have quite a few versions, but all with pads on both sides).
Thanks for informing. Will try to avoid pads in the middle in the future.
Is it possible to drill a small hole? (guessing they are single sided to save money…)
The LEDs are offset 0.5mm in the versions I have. The later designs compensate for this, but i suspect that the distance could vary on other models…
