Review request on XTAL for nrf52810

[Gabriel Womelsdorf]

Good morning/afternoon all, I have been working on a BLE air quality monitor using the NRF52810. The first version functioned as expected with the exception of the external oscillators, where neither the HF or LF clock would start up. On the new design, I tightened the layout a bit more, and I was wondering if there were any pitfalls a more experienced designer could see. My plan is to use a 8pF crystal for the LF and HF clock, with 12pF load caps.

My current stackup is a 4 layer board with the top layer being a signal layer, second being a full ground plane, third being a power plane (3.3V) and fourth being signal.

      

[Szabolcs]

Hi Gabriel,

The issue I see with both designs is that there should be more ground vias, close to the ground pins of components. Please see in picture:

On the first one, the HFXO trace should also run closer to the other.

For a CL=8pF crystal, 12pF caps are good. If your substrate is thin (100-150um between top and first inner layer), then the PCB capacitance could be higher, and the caps could probably be decreased to 10pF, but it's not much difference.

Make sure you have more stitching vias all over the board, especially near high-frequency circuits and along the board edge.

If your oscillators fail to start up, that could also be due to configuration error.

[Gabriel Womelsdorf]

Thank you! I’ll make those changes 

[RonaldHise]

Hey Gabriel, thanks for sharing your design! For the nRF52810 and oscillator issues, your tighter layout is a good step. Using 8pF crystals with 12pF load caps seems reasonable on the surface. Perhaps double-check your ground connections near the XTAL; a solid ground is crucial. While you're troubleshooting, have you considered a mental break with a quick game of Pips? Fresh eyes can sometimes spot those elusive problems!