IMPORTANT: DO NOT USE THESE FOR ANYTHING YET, THEY ARE VERY MUCH PROTOTYPES AND HAVE SOME PROBLEMS!!!!
Files for PCB modules for use by usb_midi_clocker project.
- IMPROVE: reduce size of jack holes to make a snugger fit with less wobble?
- IMPROVE: Some nice panel graphics!
- DONE: FIX: Replace the voltage regulator with one that will actually be able to deliver enough amps to power the screen and Teensy facepalm
Can work around this for the moment by connecting directly to 5v Eurorack power with eg Dupont cables- FIX: Fix/bridge/remove power diode that prevents power from USB going to screen
- IMPROVE: Make the i2c port into a 2x5 connector
- IMPROVE: Add 100ohm resistors on the SPI connectors
- DONE: IMPROVE: Cutout for display pins in order to allow the panel to sit flat on top of the display without bending like it currently does?
- DONE/UNTESTED: FIX: Make the encoder hole bigger, so that the encoder will fit through
- FIX: ?? fix whatever the problem is that is making SPI unreliable af and lights randomly seeming to not go off
- DONE/UNTESTED: one of the protection diodes isn't connected to ground
- FIX: eratic behaviour where LEDs stay lit, don't light up properly, etc
- seems to be fixed/worked around by dropping SPI rate and putting a 22pf cap between MOSI+GND, one between GND+CS1, on between GND+SCK, one between CS1+SCK, and a 100ohm resistor between CS1+SCK too
- FIX: when put into 'inputs' mode, all seem to read HIGH when there isn't anything connected?
- FIX: when put into 'inputs' mode, two of the SEQUENCER channels randomly flicker on/off...?
- FIX: the switch pins currently connect tip to ground; should this go through a large resistor to act as a pull-down? or not happen at all?
- FIX: layout of the 'sequencer' outputs isn't right, for unknown reasons? (worked around in software)
- FIX/UNTESTED: outputs 13+14 had their resistor values accidentally swapped - 1k <-> 47k!
- IMPROVE: make reversible in / out, or label as CLOCK / SEQUENCER
- IMPROVE: make header 2x5 instead of 2x4?
- Seems to be working on first attempt!
- IMPROVE: pitch spacing of the resistors/capacitors to allow common components to fit a little better (slightly wider resistor spacing, slightly narrower capacitor spacing)
- IMPROVE: maybe add protection diodes to protect the Teensy's GPIO pins in the event of wrong thing being connected?
- IMPROVE: Make panel double-sided (in+outs 1-4 on one side, 5-6 on the other)
- IMPROVE: probably possible to do this in 3hp instead of 4hp
- TODO: Add a way to directly power USB Host from case's 5V
- TODO: IMPORTANT! Label the USB Host pins so that can tell which end is ground!
- DONE: correct the ADC inputs, as they are reversed on the first boards! (A was connected to C and C was connected to A) facepalm
- FIX: find correct footprint/component to use for the +12/-12v protection diodes (workaround is to just bridge the connections -- remember to do this BEFORE spending 6 hours debugging everything else!! Without this, it will look like the DAC chip is working except you won't get any output at all!!!)
- IMPROVE: move the power connector to the inside edge of the board and use right-angle pin headers
- IMPROVE: use right-angle pin headers/socket for the expander port; maybe get rid of one of the duplicate expander port to give more room for the power connector
- IMPROVE: label the stripe on the expander port(s)
- IMRPOVE: scale the LED output so that brightness doesn't max out halfway through range (LEDs are https://www.ebay.co.uk/itm/254275891818 - R/G 575nm Diffused Lens)
- NOTE: short leg of LEDs goes in square hole to ensure that green is positive and red negative
| Address | A1 | A0 | Note |
|---|---|---|---|
| 0x4C | LOW | LOW | correct |
| 0x4D | LOW | HIGH | incorrectly seems to stick at 0x4C as if A0 is stuck low? |
| 0x4E | HIGH | LOW | correct |
| 0x4F | HIGH | HIGH | untested - presumed non-working as A0 seems to be stuck low in this hardware revision? |
| Bank | A3 | A2 | Note |
|---|---|---|---|
| 0 | LOW | LOW | Correct |
| 1 | LOW | HIGH | Correct |
| 2 | HIGH | LOW | untested but presumed working |
| 3 | HIGH | HIGH | untested but presumed working |
- Daughterboards to replace the CV input (currently reliant on Pimoroni 24v ADC module):
- Own replacement for the Pimoroni board (believe this to be beyond my current ability)
- Daughterboard to add 3xGate outputs, instead of inputs, based around a GPIO i2c
- Expander-for-the-expander to add switches to select uni-polar/bipolar modes -- with GPIO i2c on there, too, so that we know in software to switch modes and use different calibration data