[werk_ag]

Watching both waterfalls at the same time, it appears that when it transmits the noise stops. Those breaks you see on spectran appear to happen when it is transmitting.

If the noise appears to stop when there is a transmission, maybe the source of the noise isn't the transmitter, as it is only powered during the time required to send each data packet.

In any case you are having a remarkable value for the RF Efficiency, I never achieved a value such high (my installation is not in an urban environment too), and I never expected to reach it too, mainly because it I'm using three TX units and some other RF devices, all transmitting to the receiver unit, so some collisions are expected which end in the lost of some data packets, which affect the RF Efficiency math.