RN-52 Powering Electret Microphone

I’ve got an RN-52 (on the breakout board and 3 more without it) and I’m trying to get an electret working decently well with it. The RN-52 is being powered by a small LiPo battery (3.7V nominal) and everything is working fine (speakers stream audio, mic even works…kinda).

The issue is that the audio from the mic has a ton of noise (terrible SNR) and is almost unusable. While monitoring the audio on my PC, I’ve found that powering the electret circuit from the same battery that the RN-52 is connected to sounds great even while the RN-52 is powered on and connected to my phone. However, as soon as I start one of the Bluetooth profiles like by streaming audio or making a call through the RN-52, the noise gets crazy high (so I presume the RN-52 is doing something wonky to the voltage there when it is really doing stuff).

I’m not sure if my diagram got uploaded, but I’m basically running the RN-52 3.3V pin (also connected straight to LiPo) to the electret ‘+’ side through a 10kOhm resistor and the RN-52 GND directly to the electret ‘-’ side. Then I run GND to Mic L- and Mic L+ gets connected to the electret ‘+’ side through a 100 nF capacitor. Again, sounds terrible once the RN-52 actually does something useful.

So then I tried powering the electret mic through a separate 3V supply (two CR2032s) and when connected in the same way to the RN-52 (only Mic L- now connected to the GND on the CR2032s), I can get audio to come through the RN-52 clearly (as in during a phone call). I’ve had little success using M_Bias to power the mic and I’m wondering if that is because the audio ground pin is not exposed in the breakout.

Is there something I should be doing with the bias or something to “clean up” the battery voltage when the RN-52 is working. I don’t have an oscilloscope or anything to test, so I feel like I’m kind of in the dark right now. I really would like to get the mic on the same battery supply as the RN-52. I also have the MEMS breakout and the 3-pin electret breakout if those would be better to use, but I’d rather get this working with a regular electret mic.

It sounds like radio frequency interference and power supply noise is getting into the microphone. The microphone and leads should be in a small shielded and ground metal case. Also, you will need a decoupling filter on the microphone power supply (say, 1K Ohm, 10 nF AND 10 uF in parallel), outside the metal case. Similar to below but C1 would be two caps in parallel.

https://i.stack.imgur.com/6EXwj.png