RE: Custom NRF52 design. HFXO starts generation, but software does not detect it

rundb — Mon, 10 Dec 2018 10:22:44 GMT

Thank you for fast answer. I understand that my measurement influences the capacitance. Still I see that every time I check the signal, the generation is present, and it is completely similar to generated signal on the development board.

I will try adding more capacitance, thank you. Also I will try to assemble another sample to make sure it is not a soldering issue.

RE: Custom NRF52 design. HFXO starts generation, but software does not detect it

ketiljo — Mon, 10 Dec 2018 10:09:20 GMT

If it's this crystal you're using https://www.digikey.com/product-detail/en/epson/FA-20H-32.0000MF20X-K3/SER3801CT-ND/3198123, then the CL value is 10 pF. This is not the same as the load cap value. The formula is:

Ccap = 2 * CL - Cpin, where Cpin is 4 pF.

So for a 10 pF crystal, you need 15 pF load caps. If you use the wrong value, the crystal may have problems starting.

You will add quite a lot of capacitance when you probe it with a oscilloscope.