Regarding the GRE NFM vs. FM settings - I have run my own experiments and have noted many times in the past on these forums that the GRE approach does not change the audio level but it DOES select a different IF filter bandwidth. The Uniden approach (at least on my BCT15) is just the opposite - they do alter audio levels (or adjust the discriminator differently) to compensate but do nothing with the IF filter bandwidth (which makes it problematic when dealing with closely spaced signals as the adjacent channel rejection really suffers).
Ideally, they should do BOTH an IF filter bandwidth change as well as audio compensation but there you go. To me, it looks like GRE added the NFM setting almost as an afterthought as it is weirdly hampered in its function:
1) No audio compensation (as noted; the "Audio Boost" setting is a "band aid" solution, in my opinion, and not a real solution).
2) Only works in programmed memory channels and is not capable of being activated in any search mode (the "Auto" function only selects between P25 and analog FM [old standard +/-5KHz maximum deviation]).
3) Although you can select NFM mode while also selecting P25 digital voice mode and the display does reflect the setting it actually does NOT change the IF filter bandwidth as it does in analog FM - in effect it does nothing at all. I ran my tests to discover this on my own but it was also confirmed here by a moderator who had a contact within GRE - the NFM mode has no effect on P25 voice even though you can set it and it looks like it has been selected on the screen and in software. Obviously this is a disconnect between the embedded firmware and the hardware that was left in place and never corrected. I can speculate that the reason you cannot use the narrower IF bandwidth is that it is not "friendly" to P25 voice as it is designed in the GRE scanners. Perhaps the group delay is not adequate when the filter is in narrow mode. I do not have a schematic so I don't know how they change the IF bandwidth, whether they use two different hardware IF filters and select between the two or otherwise alter the bandwidth of the filter using external components (klunky unless the filter is really designed for that) or is added post filter using DSP software filtering (as some have speculated to me). In professional gear there are different P25 bandwidths, at least in some, so I know that that different IF bandwidths for P25 Phase 1 voice are used in practice but I don't know the details and how many Phase 1 P25 voice bandwidths there are actually in use.
4) There are some strange artifacts one can detect when using the NFM setting in GRE scanners. Occasionally, when scanning and some channels have analog NFM programmed frequencies the scanner may stop on an inactive frequency (no apparent signal present) and you will hear a soft "pop" sound and the scanner will remain on the channel until the delay period is over (assuming no signal present). As far as I can determine, this happens mostly when the frequency has an interfering signal at some frequency off center from the desired. What I think is happening is that there is a small delay in activating the narrow bandwidth setting during scan mode. So I think it actually has the filter in the wider mode initially during scanning and then switches (if NFM is selected) once a signal is detected; so if an adjacent interferer is present that is within the bandwidth of the normal wider filter but outside of the bandwidth of the narrower filter the scanner will stop on that channel but then kick in the narrow filter so all will be quiet after the quick burst of audio from the interferer (the soft "pop" sound I mentioned). Another strange effect of using the NFM mode in GRE scanners is that the squelch tail will increase in some cases. Not always, insofar as I can tell, but in some cases a channel programmed with NFM mode will have an annoyingly long squelch tail when a signal ceases. If you reprogram the channel in regular FM mode this effect will go away. I can't say why this is and why only some frequencies get affected this way but I am guessing that it may occur when there are interfering signals present that are broadband enough to fall outside the discriminator bandwidth so don't get demodulated but are still there and increase the noise floor in the IF passband causing the squelch noise gate to stay open longer. I would intuitively think that a narrower IF bandwidth would decrease that noise floor, however, so it seems very odd that the squelch tail is lengthened when the narrow bandwidth is used - I can only assume that it has to do with the inadequacy of the design reflecting its "afterthought" speculated (on my part) status.
Also, I am unclear whether the NFM function works in any or all trunking modes. Given that it does not work in P25 conventional mode I would guess that it will not work in P25 trunking mode or on a mixed mode system with P25 voice. My lack of adequate test equipment to test this leaves this an open question for me. I have read reports of some folks who swear that some systems function better in FM mode or vice versa so ???...I don't know.
Anyway, good info and good pictures so thank you!
Last edited by Mike_G_D; 01-02-2014 at 7:39 PM..