Depending on where you're located of course, there still are numerous AM radio stations available. In my area, one of our favorites shares call letters with an FM version. The AM station specializes in "old style country" while the FM is "modern-country-rockabilly-progressive-newage junk".

Assuming that a tester can be found, the tubes are the logical first suspect and they should be tested and any "bad" ones replaced.

A good voltmeter is sufficient to verify that the supply voltages are present and correct. In a 1947 radio, filaments probably will be 6V and the "plate voltage" will be 75V, 90V, or 150V if memory is correct. The proper voltages can be determined by looking up the ratings for the tubes that are in the radio.

The plate voltage (the higher DC) should be checked for "AC content" which, if present in excess would indicate a bad capacitor in the power supply section.

A precision RF signal generator is needed to adjust the frequency of the IF amplifier stage.

Once the IF frequency is set, the same precision RF signal generator would be used to inject signals at "broadcast frequencies" at a few frequencies over the broadcast range (from about 600 KHz to 1800 KHz). The two-stage variable "condensor" is then adjusted so that the difference between the frequency to which the input stage of the condenser is tuned, and the frequency to which the second stage of the condensor tunes the internal RF oscillator is constant over the tuning range and matches the IF frequency. This was usually called the "tracking adjustment" and was generally done by "bending the plates" on the second stage variable condensor. Some skill is required.

A "noise everywhere but no signal" could indicate absence of the internally generated RF, most likely due to a bad tube; or it could indicate that the tracking is off by a very large amount, possibly due to cobwebs or grease and grime on the condensor plates. A third possibility would be a bad "rectifier" tube in the detection (audio) stage.

There is of course always the possibility of corrosion, open joints, missing insulation, short circuits, and other symptoms of simple old age. Radios of this era generally had only 5 or 7 tubes, 3 tank coils, and the tuning condensor, with a maximum of about 3 resistors and possibly a "noise" capacitor at about half the the major components, so there aren't a lot of things to go wrong. Repair and adjustment is labor intensive (i.e. time consuming) if you can find someone with the right equipment who's willing to work on one.

John