rf - Wireless water level sensor, in sump. Installation best-practices for longivity

My wireless water-level sensor is working fine, and when deployed in the sump-well, it is giving me periodic readings quite well. At this stage, I have the circuit still on a breadboard, and now plan to put it on a veroboard, and finally install this thing in the sump-well for long-term reading purposes.

There are few challenges I am facing and few that I am perhaps imagining. Thus looking for answers on how to fix / work-around.

1. 
   

   The sump has an iron cover. The ISM band ASK/OOK transmission cannot be detected, even 5 feet away, when the iron cover is closed. With the cover open, I am getting a decent range of about 20 feet, which is what I need.

   
   


2. 
   

   I had left the circuit installed for around 36 hours, and next day when I took the circuit out, I was surprised that it was still working and had not shorted out. It was full with in water-drops that had condensed overnight. The iron-sump cover is a dark shade of brown, and gets extremely hot during the day, in the sun. In the night, the temp drops significantly, and thus the result.

   
   
   


3. 
   

   Given how how it gets inside the sump in the day (I've not measured, but I'm guessing that it'd get like 50-60 deg, with relative humidity above 95%, in the sump), what can I expect about the life of my 4S AA batteries ?

   
   

Due to these challenges/observations, I am think of splitting the circuit into 2 parts, i.e. the ultrasonic sensor, mounted on a tiny PCB, with the soldered/exposed pins all potted with hot-glue. This part stays inside the sump. The rest of the circuit is housed outside the sump, in an IP66 type of enclosure. The enclosure kept outside, would let the RF TX work well, keep the batteries in saner temp, avoid excessive condensation. Of course, here the challenge is, how to wire the 2 parts together, as the wire coming out of the sump, will have to avoid being damaged by the sump cover crushing it, once closed tight. One problem with this is, I cannot keep the IP66 enclosure within 5 feet of the sump, so the wirelength might become an issue (?)

Should I alternatively, just do something to keep the RF TX antenna outside the sump ? But won't the wire of the antenna extension, alter the antenna behaviour (altered gain for the necessary frequencies) ? If I do this, I still don't know, how to address the water condensation issue. Or how well the batteries would do.

Edit: Finally, I am not sure, if the water condensation will (and how badly) impact the ultra-sonic range sensor itself, i.e. the transducers themself, as those need to remain exposed. I can imagine a drop of water (from condensation) hanging on the transducer and easily throw-off the readings completely. No ?