On Thursday 6/1 I received an email from a few hams near Midland that reported packet interference to their 147.000 repeater. A snapshot of some packet data identified it as coming from the system here somehow. Midlands repeater being 120 miles away made this report rather astounding to both them and me.
Thursday afternoon I took the spectrum analyzer to the site and did some checking. At first I noted a questionable blip on the meter near the 147.6 input to the Midland repeater. This blip was very weak, and only visible from an antenna pickup up the tower (there was nothing visible when the packet TX was put into the meter directly). Even though the blip was weak, I moved the packet node from 145.07 to 145.09 as a precautionary measure, in the event it was a temperature related item and to give me more time to assemble a better measurement technique and ponder the possible modes that the Midland hams could be receiving the interference. All of my other transmitters appeared clean on the Motorola spectrum analyzer.
Interestingly, when moving the packet node to 145.09, that 'weak noisy blip' (had what looked like FM broadcast modulation in it too) had moved up the band and was now squarely on the 147.64 input for the Benzie repeater-- just 12 miles away. I monitored the Benzie repeater through the evening and there was no packet interference to it. Please note that up until early May, this same node was on 145.09 for the previous year and there was no interference to Benzie, through both cold and hot temperatures.
On Friday I returned to the site with more gear (attenuators, etc) and and measured that 'noisy blip' to be approximately 75db below the 39w carrier level of my 145.07 transmitter. (measured by a 2m dipole antenna about 6' horizontal from the packet transmitters antenna). Surely this weak signal could not have made the 120 mile trip to Midland.
On mulling over the captured packet headers, I now conclude that the reception was NOT of the 145.07 transmitter, but of the 433.100 mhz transmitter. The captured packets do have some irregularities that are rather questionable, so the cause may be something other than technical in nature. The 433.1 transmitter is the 10w UHF packet repeater on site. 120 miles is quite a hike for 433mhz, but, if the Midland repeater happened to have a 433.1mhz link receiver on their tower (and it was carrier squelch and omnidirectional) there is a possiblity it could be heard if conditions were 'just so'. [It would be an interesting test to transmit on 433.1 in the Midland area and see if the 147.00 rptr responded].
As a precaution, I moved the 433.100 packet repeater up 25khz to 433.125 (listening on 438.125) in expectation that this would eliminate the problem if there was indeed a 433.1mhz receiver atop the Midland repeater tower, or if there was a mix of 433.1 occurring at a site external to the Empire site (perhaps in Cadillac or closer to Midland) that was then getting into 147.60.
Time will tell if this is successful in eliminating the interference, as I would expect it depends greatly on propagation effects. As I left the site, there was no combination of transmitters that I could amass that would cause any intermod products (greater than that -75dbm noted earlier) that fell within 100khz of 147.600. I continue to monitor 147.600 at my workshop (TS2000 and a dualband ant at 50') and have yet to hear any packet energy on that frequency. Certainly if it was to make the 120 mile trip from Empire to Midland, it would be audible at 2 miles from my transmitter.
In other intermod related news, The work on the WGFN antenna system performed this past January appears to have eliminated the extremely strong intermod that rendered the 52.42 receiver useless last summer whenever the temperature exceeded 72 degrees. After a week of near 90 degree weather without any intermod (and I looked for it), I am all smiles that it is now absent.
73 John KF8KK