Dividing a 32768 Hz crystal frequency

Anything with power gain can be made to oscillate, but not always with a simple RC or crystal circuit.

Any sub-dollar MCU would run from 2^15Hz input and provide you with a timer to divide by 20. Why bother using these ancient CD4000 parts?

Best regards, Piotr

and it'll have a dac or pwm to do a decent sinewave, but you have to know how to program it ...

All the program needs to do here is to setup a timer for /20 and connect it to a GPIO somehow. With all due respect, if that's considered difficult, I'd suggest changing the hobby.

Best regards, Piotr

An off-the-shelf oscillator would work far better. RTC oscillators share many traits with memcpy(): only selected few can make both *right*. :-)

Best regards, Piotr

1. There's no 64-bit counter that can be clocked from a 32768Hz clock.
2. The interpolators don't work with DMA.
3. The DMA can't auto-restart (without chaining).
4. Many IRQ sources share an IRQ line, so handler dispatching is annoying.

I use the RP2040 to decompress an FPGA bitstream, decode audio files for playback and handle the WS2812B status LED chains, as the chip starts first. The Artix does the rest.

Best regards, Piotr

I've seen higher frequency xo circuits that weren't reliable in production. The circuit and the caps have to match up with the crystal motional equivalents, and thay can vary a lot. Some purchasing agent buys a different crystal and it doesn't work.

The project has a tutorial element. People have to understand exactly what's going on, and walking them through the source code doesn't seem to be an acceptable way of providing it.