MultiNav Pro+ BLE - any gotchas with the RF front-end near Li-ion charging circuits?

@Tony Campos yeah this is a legitimate concern and honestly one of the more common ways people quietly kill their link budget without realizing it. The TP4056 itself isn't the worst offender since it's a linear charger, but if you've got any switching element upstream of it (like a boost or buck feeding the input) those harmonics absolutely can land in the 2.4GHz band depending on your switching frequency and board geometry.

 

Haven't done formal bench testing with the MultiNav Pro+ specifically at 10-15mm from a TP4056, but the general rule I'd follow is keep your charge circuit ground plane stitching tight and make sure you're not letting that noisy ground current loop anywhere near the RF section's ground reference. The module's sensitivity is pretty exceptional so it's worth protecting — if you trash even a few dB of receiver sensitivity you're leaving real range on the table.

 

Practically speaking, if you can't increase the physical separation, a ground pour slot between the two sections and keeping the charging circuit on the opposite side of the board from the antenna both help a lot. Also worth throwing a small ferrite bead on the battery positive line close to the module's VCC pin if you're sharing supply rails.

 

If anyone else on here has actually scoped the noise floor on this specific layout I'd be curious too — might be worth someone doing a proper before/after with a spectrum analyzer and posting results.

We actually ran into exactly this on an internal prototype a few months back - had a TP4056 about 12mm from the RF section and saw maybe 3-4 dB sensitivity hit at peak charge current, mostly cleaned up by adding a small ferrite bead on the VBAT line feeding the module and stitching the ground plane more aggressively under the charger. If you can squeeze a bit more physical separation or rotate the module so the antenna end points away from the charging circuit that helps too, but the ferrite + solid ground combo honestly did most of the heavy lifting for us.

Tony's experience lines up with what I've seen too - ferrite on the power rail does a lot of work, but I'd also throw a 100nF cap right at the module's VCC pin if you haven't already, sometimes that last bit of HF noise sneaks past the bead and that cap catches it. Also worth scoping the VBAT rail during charge with a decent bandwidth probe before you commit to the final layout, the harmonics can be sneaky and you'd rather know early than respin.

Good callouts from both of you - one thing I'd add is if you're on a two-layer board and can't do proper ground stitching under the charger, a small copper pour tied to GND between the TP4056 and the module does a decent job acting as a cheap shield and I've seen it shave off another dB or so on its own. Also just double-check the TP4056's PROG resistor placement isn't floating near the antenna keepout, seen that cause weird stuff before that took way too long to track down.

All solid advice in this thread - one more thing I'd mention is that the TP4056's thermal pad can act as a little noise antenna if it's not stitched down hard to GND with a few vias, so don't skimp on that even if the rest of your ground plane looks clean. Caught me out once on a tight layout and the ferrite alone wasn't enough until I sorted the thermal pad connection.

All of this is gold and matches my experience - one thing I haven't seen mentioned yet is keeping the TP4056's CHRG and STDBY indicator lines short and away from the antenna keepout too, those little open-drain pins can pick up and re-radiate noise if the traces run anywhere near the RF section. Caught a weird intermittent sensitivity drop once that took me embarrassingly long to trace back to a CHRG indicator trace that was basically acting as a tiny stub antenna right next to the module.