radioguybroadcasting said:
ANd to diplex an 1170 and 1220? Holy batman... that would be one expensive MESS to do. It's usually reccomended to have at least 180khz between stations you're diplexing, i've seen it as close as 80khz (Orlando, FL) and that was an absolute disaster.
I've always heard that the frequency difference in kHz isn't exactly the issue. Rather, it's the RATIO of the frequency difference to the higher carrier frequency that determines the complexity of the filtering required and the probability of success. A time-honored rule of thumb was that the frequency difference should always exceed 10% of the higher carrier frequency. But that rule was broken almost from the beginning of diplexes. I've been told that many (like 70 or more) years ago, 560 and 610 in SF were diplexed. The percentage difference was only ~8.2%. In San Jose 1370 and 1500 are diplexed into a three or four-tower array, and both are reasonably high-powered stations. The difference is only ~8.67%. In Seattle, 820 and 950, both 50 kW by day, are diplexed into a three-tower array. The difference is ~13.7%, which is greater than 10% but it may be the closest frequency spacing for two directional 50 kW stations in the US. I believe the closest-spaced diplex in the US is two low-power ND stations in Santa Barbara on 1290 and 1340--3.73%.
Honolulu must be the AM multiplex capital of the world. I say multiplex rather than diplex because most of the multiplexes are triplexes and there may be a quadriplex or two. Many of these stations run 10 kW. At least, AFAIK, no directionals are involved. A lot of the Honolulu stations that share towers are quite close in frequency, although I don't believe that any are as close as 3.73%.
Here in Boston, three stations, one 50 kW-U DA-2 on 1200, one 25 kW-D/17 kW-N DA-2 on 1330, and one 20 kW-U DA-1 on 1600 are now multiplexed into a five-tower array. IIRC, the third station (the one on 1200) went on the air from the new facility at the beginning of September. None of the stations is yet operating at full-power but the setup appears to be working. The smaller of the frequency differences is a bit less than 10% and considerably less than 180 kHz.