I'm aiming to add an 8MHz xtal to a Nucleo-L496 board but I'm struggling a bit. The recommended part isn't available, I've found one that will fit but the load capacitance is 10pF compared to 8pF of the one in the manual.
However, what is confusing me is the recommended load capacitors - they recommend 4.3pF capacitors for the 8pF xtal, this seems extremely low. My understanding is that as a rule of thumb you take around 5pF off the load capacitance value for board parasitics then double what is left to get the load capacitor value.
However, using 4.3pF caps would suggest the parasitic capacitance is nearly 6pF - which seems odd, as elsewhere ST themselves say it's more likely to be closer to 2pF and they only use 5pF as a worst case scenario.
So, are the Nucleo boards really bad for parasitic capacitance, or has someone messed up the calculations?
For my 10pF crystal I'm not sure if I should be using 8.2pF (assuming around 6pF parasitics) or something closer to 15pF (assuming a more normal 2.5pF parasitic).
Any ideas of experience of fitting different crystals to these boards?
> Any ideas of experience of fitting different crystals to these boards?
There is quite a bit of wiggle room. I assume 4 pF parasitic capacitance and never had issues.
Capacitors themselves have quite a bit of tolerance (typically 20%) as well.
You've probably seen this:
Yes, thanks, that's what I used as a reference to make sure I wasn't missing anything specific to the STM32 series. Using the equations there I get a considerably higher value than ST recommend in the docs for this board.
I've also noticed that they use the same xtal elsewhere (on the embedded ST Link) and there they use 10pF capacitors. Just makes me wonder more if there is something specific about this connection, or just a mistake in the product manual...
I'll get some 10pF and 12pF caps so I've got margin to change them if the frequency is too far off. We are going to be using the USB so it needs to be fairly accurate.