I have noticed that frequencies at the edges of the tuner spectrum have poor performance. What is the best way to keep SDR Trunk from the unstable edges?
If you were using just one device (like an Airspy), it's pretty easy to do by creating two dummy P25 channels, 0 max voice channels on each one. Set those two dummy channels to start before any of the trunked systems you are monitoring. Those two P25 dummy channels would be the max low and max high freq. But I don't know how that will work in an environment with 6 dongles, since you are really wanting to limit the scope of each dongle. Perhaps you'd have to set up 6 pairs of dummy channels, one pair for each dongle.
An example of my setup (Airspy 10 mhz, perhaps 8.5 mhz usable):
Lowest freq on systems I monitor: 851.1625
Highest freq on systems I monitor: 858.4625
Total bandwidth required: 7.3 mhz
P25 dummy channel 1: 850.8125 (no signal ever present on it) - this one starts first
P25 dummy channel 2: 858.8125 (no signal ever present on it) - this one starts second
Two P25 trunked system sites: these start after the dummy channels
This keeps the Airspy pretty much in the center of the "good area".
With you having multiple dongles, I think it's a different story. And in my mind I can't wrap my head around what you may need to do. Hopefully somebody else will chime in with some ideas.