Blacksheep mentioned pin-prick holes in the radome. Interesting follow on to this observation. We had an on-going problem with a B737-200 which regularly flew in the West and Central Pacific area. Lots of moisture. As anyone knows who flies in that area, Cb and lightning abound. We had Bendix RDR-1E radar - if I remember correctly. Problem was that we could only get returns within 80 miles during cruise. Guam, a island around 60 miles long, could easily be picked up at 180 miles on our other 737's - but not this particular 737.
Cruising in cloud, Cb returns would loom up at 80 miles, first as tiny returns, then full blown serious stuff as we got within 30 miles. Made long range assessment of typhoon clouds a bit late. Below around 8000 ft on descent or climb, returns were normal.
Despite snagging the radar numerous times, techs in Hong Kong (our servicing base)could not fault the radar.
In desperation, I took a series of photos of the radar picture of Guam from 180 miles in the 737 with normal radar returns, and again with the 737 with the dodgy radar (ie 80 miles if we were lucky). I sent these to Bendix in USA with a request for ideas.
They said that in all probability, the radome was letting in water through pin-point holes in the radome. The water would eventually freeze inside the honeycomb and form an ice barrier. The radar beam then becomes attenuated and very little energy gets through. Hence poor range at high altitudes. In warmer temperatures, the ice melts and the radar works normally. Hence, ground checked - serviceable at Hong Kong.
Bendix said bake the radome to dry out the trapped moisture, then repair the radome.
No problem after that.