Fix RAK3401 SKY66122-11 FEM control: enable CSD/CPS for proper PA/LNA operation

[weebl2000]

preamble: I was talking to someone about their RAK3401 board on Discord and dived into datasheets and current implementation. I asked them to test and come back with findings, but I strongly believe the FEM isn't being controlled properly and this can cause RX sensitivity to be bad and perhaps even cause power/overheating issues.

The RAK13302 1W module uses a Skyworks SKY66122-11 front-end module with three digital control pins (CSD, CTX, CPS) that must be actively driven by the host MCU. The previous code only managed CTX (GPIO 31) — toggling it for TX/RX — but never initialized CSD (GPIO 24) or CPS (GPIO 21), leaving them floating with no pull-up/pull-down resistors on the PCB.

With floating CSD and CPS, the SKY66122 was in an undefined operating mode:

• The 30 dB TX PA may not have been reliably engaging
• The 16 dB RX LNA was never reliably active, degrading receive sensitivity

I have 4 variants for you to try:

1. This PR, fix SKY66122-11 FEM being driven correctly
2. This PR, fix SKY66122-11 FEM being driven correctly, and SX126X register patch
3. This PR, fix SKY66122-11 FEM being driven correctly + AGC reset SX1262
4. This PR, fix SKY66122-11 FEM being driven correctly + AGC reset SX1262 + SX126X register patch

I expect variant number 4 to have the best improvement. When replying with your findings please mention the builds you have tested!

Following the comment from Heltec engineer I also applied the register patch here. It should help with keeping AGC in check.

I don't have a RAK13302 device so I cannot test this myself.

References:

• RAK13302 docs
• SKY66122-11 FEM datasheet
• SX1262 datasheet

[txkbaldlaw]

I had good results from using variant 4 above. Stock 3401/13302 1W booster from Rak. No other add-on boards.

Firmware appears stable. Turning off the RXBoost via 'set radio.lna off' results in immediately and reproducible 3-3.5 db gain in advert packet reception from distant repeaters.

[robekl]

is NRF52_POWER_MANAGEMENT enabled in the build for rak3401?

[weebl2000]

It's not currently it seems.

[robekl]

could we see this problem?

FEM control sequencing in begin() can briefly enter TX mode before CTX is forced low.

• In RAK3401Board.cpp:31-35, CSD and CPS are driven HIGH first.
• CTX is only configured/forced LOW at RAK3401Board.cpp:37-38.
• SKY66122 mode table says CSD=1, CPS=1, CTX=1 is TX mode. If CTX is floating/high during boot, you can get a transient TX/boost-on condition before RX-safe state is asserted.
• maybe set CTX LOW first (or configure pulldown very early), then enable CSD/CPS.

[weebl2000]

could we see this problem?

FEM control sequencing in begin() can briefly enter TX mode before CTX is forced low.

• In RAK3401Board.cpp:31-35, CSD and CPS are driven HIGH first.
• CTX is only configured/forced LOW at RAK3401Board.cpp:37-38.
• SKY66122 mode table says CSD=1, CPS=1, CTX=1 is TX mode. If CTX is floating/high during boot, you can get a transient TX/boost-on condition before RX-safe state is asserted.
• maybe set CTX LOW first (or configure pulldown very early), then enable CSD/CPS.

Agreed, it's probably cleaner that way. I doubt it would have caused any issues currently. At most it would have amplified a bit of noise for a few microseconds. :) Changing it.