I had the same thing happen to me several times with seagoing radars. Sometimes you had to just drill a small hole in the bottom of the radar antenna and see if water came out. Then I would reseal it with silicone seal. I would go through a lot of trouble before I got to that point, however, until I had experience with water leaks in the antenna. It never looked like there could be a leak. But looks (leaks) are often deceiving...
Had the same thing happen with an aircraft radar waveguide. It worked fine on the bench but when we pressurized it and leak tested we found a small pinhole caused by RF arcing. Once airborne the leak had allowed more moisture in and pressurizing nitrogen out.
All broadcast transmission lines that have an "air dielectric" which is virtually all broadcast transmitters, are pressurized to eliminate moisture in the lines. It can be plain air that has been run through a dehydrator, or nitrogen from a bottle. Dehydrators can either use a dessicant, or a condensation coil to remove moisture from the air. Pressures are recommended by the transmission line manufacturers, and usually are in the range of 2 psi up to possibly 15 psi maximum. High pressure is not necessary, only enough to create a positive pressure differential. You do not want too high pressure, or you risk "bursting" the pressre window that keeps the pressure from entering either the antenna structure or the transmitter tuning cavity. The pressure window is a thin piece of mylar type of material (think of a 35mm photographic slide...similar material and thickness). We generally always have the lines pressurized. This also gives us the ability to tell when transmission line damage occurs, as the flow meter will start to increase. Moisture in transmission lines is not good, not only because of corrosion, but when dealing with 50Kw and higher, arc overs become a real problem.
The pressure in the cable does serve to keep water out, for most situations. BUT the use of dry nitrogen also has other bemefits, in that it provides a dry atmosphere, which reduces the tendancy for arcs to develop, and it avoids the effect of oxygen on both the insulation material and the conductor surface. In addition, if all of the air is displaced, the formation of ozone is inhibited, which ozone is detrimental to all of the materials used in RF circuitry.
They don't necessarily need dry gas pressurization. Many broadcast transmission systems use solid dielectric insulation and thus are not presurized. In fact it is atypical for an AM facility to use pressurized lines. In the past, the primary reason for using pressurized lines was slightly lower loss. Second, FM broadcast antennas often require dry gas pressurization to insure no moisture ingress into the antenna mounted high up on the tower.
I recently had a client whose new solid state FM transmitter was unhappy with its antenna. When the weather got cold the miss-match got worse over time. This was supposed to be pressurized line but the client was lax about keeping the line filled with dry air. I had them drill a small hole in the line at the lowest point outdoors on a day above freezing and lots of water came pouring out. The temporary fix was to replace the entire air dielectic line with sold dielectric line that the tower rigger had in stock from a cellular tower job.
The client's old tube transmitter didn't have a problem with some water in the line as it could be tuned to match into almost any load. The nature of the design of transistor transmitters, many combined smaller amplifiers, requires a precise 50 ohm match on their outputs. While one can specify output matching networks for some brands of solid state AM transmitters, that is not the case with FM or TV. It has to be 50 ohms or bust.
Solid line is heavier than air dielectric line. That could raise a mechanical issue with tower loading. Many broadcasters opt to use rigid line sections that look like copper sewer pipes. The benefits are even lower line loss and the ability to repair a damaged section without having to replace the entire coaxial line. But the installation labor is higher for rigid line sections.
As for troubleshooting an AM antenna system, one would normally first test the line in two conditions with a simple ohmmeter, open at the far end and then shorted or terminated in a 50 ohm load. This will immediately detect or rule out shorts or opens in the line itself. The old GR bridge has long been superceded by far less expensive and more informative test equipment such as vector impedance analyzers. Many AM directional phasors come with built in operating impedance bridges similar to the old GR bridge but designed to handle full power from the actual transmitter to help maintain correct impedance matching. This is especially important when using transmitters with tuneable output networks or matching networks as its possible to accidently adjust both the phasor input and transmitter output far from the correct 50 ohms of the interconnecting line.
I once got an urgent call from a broadcaster who had just enough knowledge to be dangerous. He had a brand new 10 kW digitally modulated AM transmitter running at 5 kW. He decided to swap in a different audio processor to take advantage of the superior sound capability of his new transmitter but he neglected to slowly turn it up. The transmitter dutifully produced over 200% modulation and the excessive RF voltage was not received well by the antenna system. The result was a bunch of alarm lights and an unhappy, but unscathed, transmitter. Upon arrival I disconnected the transmitter from the antenna line and put an ohmmeter on it. It might have been my old Simpson 260P VOM or my newer Simpson 488 digital multi-meter. Dead short! So out to the tower went I. I disconnected the line and the short went away. Whew! That ruled out an arc-over problem in the 500 feet of non pressurized buried line. However, the short was evident on the output of the antenna matching network. Indeed, after separating everything, the tower itself was still shorted to ground. This insulated hot tower had a cell site tenant onboard with two isocouplers to permit the cellular equipment's coaxial lines to pass RF to and from their antennas without grounding out the tower. One of those isocouplers had internally shorted.
Actually, some coaxial line manufacturers caution against using dry nitrogen in their air dielectric lines. Apparently under certain conditions the nitrogen will react with the Teflon spacers/insulators while a normal air mixture will not.
At this year's MD&M West show, lots of material suppliers are talking about new formulations for wearables and things that stick to the skin, whether it's adhesives, wound dressings, skin patches and other drug delivery devices, or medical electronics.
Focus on Fundamentals consists of 45-minute on-line classes that cover a host of technologies. You learn without leaving the comfort of your desk. All classes are taught by subject-matter experts and all are archived. So if you can't attend live, attend at your convenience.