MSP430FR2633: CapTIvate - SYNC Pin in Dual MCU System

Sorry I can't share results right now.
My hardwaredesign is a little bit tricky. I use neighbor buttons which are placed close to each other as a GND reference for the active buttons (not active buttons are internally switched to GND). Therefore I need to make sure, that MCU2 waits for MCU1 to finish task.

I am using 2 MCUs and self cap mode, because it is way more robust in terms of EMC performance. I need to resist high disturbances (like >=10 V/m).

The Documentation tells, that the SYNC PIN is designed to wait for an external event (edge triggered) to start measurement and that is exactly what I am looking for. MCU2 should sleep till an edge event triggers my MCU2 measurement cycle. But I can't find documentation on how to setup and use this feature.

Isn't there any example project or detailed documentation on how to use the SYNC Pin feature?
This would help me a lot.

Thanks for the insightful information.
Now I realize under which conditions the SYNC pin can be used at all. I did not find this information in the documentation.

I will try the next week to implement the synchronization with a GPIO. For this I would probably disable the timer sync control at MCU2 (bTimerSyncControl = false) and take over the control via a GPIO interrupt.

Thanks also for the other tips. Noise immunity with frecuency hopping is already used by my firmware.
Just to explain my unusual design: there is no GND signal anywhere near the cap buttons. So i need the inactive buttons to be GND for the active buttons for a solid touch function.

The synchronization works perfectly now, thanks again for the support.

I have two more questions:

1) After a call "CAPT_appHandler" or more precisely "CAPT_updateSensorWithEMC" the sensor touch status ("SENSOR.bSensorTouch") does not seem to update directly. In my assigned callback function "SENSOR.bSensorTouch" is still "false" (after the first measurment) with an Finger on the Sensor. Only "SENSOR.bSensorDetect" seems to update itself directly after a single call (but seems for me much more unreliable). After a second call "CAPT_updateSensorWithEMC", "SENSOR.bSensorTouch" was set to "true". How can I query the touch status after just one update call (to save power, I don't want to measure twice as much as needed)? My ugly workarround atm is that I call "CAPT_updateSensorWithEMC" a second time within one CAPT Timer Cycle if bSensorDetect == true.

When exactly is "bSensorDetect" be set? I've seen "bSensorDetect" sometimes change to "true" with no fingers on a sensor.
For testing I used slow CAPT_Timer cycle times of 2s. That means it takes up to 4s before my application get notice of a touch (with only a single measurment call each CAPT Time Cycle).

2) After some captivate measurements, I would like to put the MSP430 temporarily in the LPM4 until it is woken up with a port interrupt. The RAM (or the calibration of the sensor) should be retained.

CAPT_stopTimer ();
CAPT_disableISR (CAPT_TIMER_INTERRUPT);
setup_my_port_interupt (); // setting up falling edge at GPIO of my desire
__bis_SR_register (LPM4_bits | GIE);
__no_operation ();

Works so far, but the power consumption is much too high for me (~ 18μA for one of the two MSP430). For testing I use __CAPT_NO_INTERFACE__ and no external clock. I tried a few things, but I could not manage to reduce the power consumption with help of API functions. Maybe I am missing the trick. What exactly do I have to do (uninitialize) to further reduce the current consumption? Using Proximity is not an option for my application.

*edited*:

Meanwhile I tried a few things and managed to reduce Power Consumption to  ~2,21µA (both MCUs @ LPM4) by adding this line to the above:

MAP_CAPT_selectTimerSource(CAPT_TIMER_SRC_VLOCLK);

Is there a more "correct" way to totally deactivate CAPTIVATE for lowest power possible?