I'm done testing the DCO that I got from JLCPCB. It worked flawlessly - after fixing a few firmware bugs that is.
I had to change the function of Data Ready to SPI Chip Select, that fixed some startup problems when the DCO had started before the main MCU powered up. Amplitude tuning works incredibly well and sync hard-syncs like it should.
I did however realise a few things:
The current setup with a 56k charge-voltage-to-current converter resistor and a 1nF integrator cap means we cannot reach 20kHz with an amplitude 0-10V. Also, the MCU draws a lot of current, around 150mA it seems. The current firmware allows 8Hz to 8kHz approx.
In addition, the circuit has both an MCU and a DAC, so if we could tap the DAC directly we could make a wavetable oscillator of sorts.
For version 1.4 I did a lot of improvements:
- Exposing the DAC output and changing the buffer resistors from 47k to 22k. The DAC can only drive 60k load, so by reducing the load for the internal buffer we can hopefully connect a second, external buffer that can be used as a digital audio output.
- I've also exposed the buffered dac output. The cap is already exposed in both ends. This makes it possible to connect a resistor in parallell with the charge resistor, to increase the current to the cap.
- All DAC SPI pins have been exposed
- Three utility pins on the MCU has been exposed, to make debugging easier (connecting leds etc)
- To reduce current consumption, it should also be possible to run the device at 3.3v, and reducing the crystal frequency from 32MHz to 16MHz. This requires a rewrite of the DCO firmware, but nothing too complicated.
I've ordered 40 of these from JLCPCB, fingers crossed :-D
No comments:
Post a Comment