nRF54L15 reference design: are ±0.05 pF tolerances necessary?

[Jonsiann]

I’m referring to the nRF54L15 Product Specification, document 4503_018 v1.0:

Section 10.3.1, “Circuit configuration 1 for QFN48 (QFAA),” starts on page 889.
Figure 180, the circuit schematic, appears on page 890.
Table 87, “Bill of material for circuit configuration 1,” appears on page 891.

Table 87 specifies C6 as 1.5 pF ±0.05 pF and C9 as 2.0 pF ±0.05 pF. These tolerances significantly restrict sourcing compared with readily stocked ±0.1 pF parts.

What measured performance difference drove these requirements? Specifically, what degradation should we expect with ±0.1 pF capacitors?

Are these tolerances essential to meeting the published radio specifications, or do they reflect the particular components Nordic validated? Guidance on acceptable tolerances would help avoid unnecessary sourcing constraints.

[Asbjørn]

Hello Jonathan,

The specification are for the components we have used to characterise and qualify our chip. You can choose to go differently. The outcome would be that the performance between boards will have larger variations. You can choose to do this, but the RF performance might be more uneven between boards you build. If that's something you are ok with, then that is fine. We recommend anyway to do some validation on your own as well with the components you source and use. 

I'd say if you are planning on using the full +8dBm output power, you will probably want to stick with the stricter tolerances as the attenuation of harmonics becomes more predictable. If you are not using the full +8dBm, the lower tolerance +/-0.1pF would be less challenging.

Best regards

Asbjørn