Using new decoder settings DSD+. Harris systems

[jets1961]

Anyone have tips for using the new decoder setting on a P25-II Harris system also IMBE/AMBE settings?

I don't get much traffic on the system so it's hard to make a raw audio recording. Anyone manage to tune your Harris system to get improved audio?

thanks in advance.

 

[slicerwizard]

Using the proper RF gain and channel bandwidth in FMPx and a direct link between FMPx and DSD+ is all the tuning you've got.

 

[jets1961]

OK thanks slicer,

I am using FMP24 link, but I might have gain high or what I think is high in DSD+ it's 49.9. In the source audio window I see clipping of the waveform.

So the AMBE setting have no effect on voice audio quality?

 

[slicerwizard]

Why do you have it so high?? Is the signal super-weak??

Post a screen grab of the FMP24 spectrum and the source audio.

You can try dicking with the AMBE unvoiced audio level, but I wouldn't expect miracles.

 

[jets1961]

No strong is well if i get on my roof I can see the tower LOL:

 

[jets1961]

better images:

 

[jets1961]

tuned to CC:

 

[merlin]

I would reduce the gain for starters. try -g14.5 in your FMP24 batch file string.
My tower is in eyeshot, 4 miles and my 700 Mhz batch file I have -g8.7

 

[AM909]

That's not clipping – it's exactly what the regenerated 4-level waveform should look like I think, though I guess it's seems to see a stream of symbols without a +1 in them. Here's a local example:

Encrypted of course, so no audio to call "good" – just showing a similar waveform from a known working system.

[Add] Yes, the gain is rather high, though it doesn't seem to be causing any problems in the pic shown – the noise floor is at a consistent and reasonable level (though we don't have a waterfall to show us whether it bounces around over time as potentially overloading sigs come and go.

 

[Adenosine]

Another quick way to check how good your doing is look at your DSD file scrolling. To much gain or to little you will have Errors in the link.

 

[slicerwizard]

RF gain level is ok, but it could be toned down about 10 dB.

PPM error adjustment isn't correct. FMP24 is warping the tuning by 2.1 kHz to get centered on the signal. Errors like that will have a negative effect on voice decoding. Adjust PPM until FMP24 is fine tuning by no more than 100-200 Hz. If you're running separate CC and VC dongles, fix the PPM correction on both of them.

Audio waveform is exactly what one should expect from Phase II PSK modulation after FMP24 processes it.

 

[jets1961]

OK so I've done some more observations of my setup and think I am onto something.

First of all I am using the one dongle setup: 1R/FMP24. I have dropped my gain to 30 and adjusted my PPM to get close to zero tuning correction on the CC. Notice the bold that's on the CC.

I notice when the radio tunes around to the various VC/slots there is lots of correction going on. The FMP24 screen shot is from an older DVB DAB dongle. My best dongle is a Thumbsat but but had a broken antenna, I have since fixed that and switched to the Thumbsat.

I get much better result now but when tuning the CC but when the various VC/slots tune the tuning deviation is still quit large on some channels. Values go from +20Hz on the CC to +300 to +400Hz on voice channels. Earlier I thought I was seeing larger deviation around +700Hz on VC?

This system is mostly encrypted and this is one of the nights the few clear TG's are off air, go figure.

I have a few days off for Remembrance Day so I will continue to play. I'm going to switch back (been a while) to a two dongle setup. The DVB TAB will be used for CC tuning then the Thumbsat for VC tuning. I have a V3 (1 PPM) on way to replace the DVB DAB. I should be able to observe the CC tuner and keep it close to zero until I get better v3 dongle.

I just kind of wonder if the Harris systems are not a tight as the Moto systems. BTW the bandwidth of the system is 2.25MHz between the channels on the system so not out of range for a 1R setup.

Thanks, will update in a few days.

 

[slicerwizard]

I get much better result now but when tuning the CC but when the various VC/slots tune the tuning deviation is still quit large on some channels. Values go from +20Hz on the CC to +300 to +400Hz on voice channels.

That's not much deviation and it's all in FMP24's auto tuning logic and the PLL's tuning granularity.

I just kind of wonder if the Harris systems are not a tight as the Moto systems.

Faulting Harris hardware instead of a hobby setup?

BTW the bandwidth of the system is 2.25MHz between the channels on the system so not out of range for a 1R setup.

1R doesn't care how much spread the channels have. I've monitored a site here that had 410, 450 and 460 MHz channels.

 

[LD723]

FMP24 doesn't like my local harris system. All I did to fix my issue is I disabled auto tuning just for the local harris simulcast P25 system because if the signal slightly dipped, the signal would look like it was drifting a little causing fmp24 to drift along with it causing the deviation levels to drift all over the place and error decodes. Turning off the auto drift solved majority of the errors. Fyi I have a nooelec dongle with the typical 0.5 ppm tcxo and this sort of behavior does not occur on any other local P25 systems simulcast or not but are all Motorola.

 

[boatbod]

If you're using RTL dongles it's been my experience that you can expect to see +/-300Hz frequency variation as you tune around. For this reason it's not unusual to have the CC tuned accurately and have the voice frequencies off a bit.

 

[jets1961]

LD723, how are you turning "auto tuning" off, with a command line switch in FMP24 bat file, or in the GUI Decoder settings?

I will look at documentation in meantime to see if i can find the switch. But if anyone knows please chime in.

FMP24 doesn't like my local harris system. All I did to fix my issue is I disabled auto tuning just for the local harris simulcast P25 system because if the signal slightly dipped, the signal would look like it was drifting a little causing fmp24 to drift along with it causing the deviation levels to drift all over the place and error decodes. Turning off the auto drift solved majority of the errors. Fyi I have a nooelec dongle with the typical 0.5 ppm tcxo and this sort of behavior does not occur on any other local P25 systems simulcast or not but are all Motorola.

 

[jets1961]

Actually I found it, I didn't even realize you can hit ? in the FMP24 window to display options, learn something new today.

I knew about ? in the DSD window.

I have turn Auto tune off with A command.

 

[LD723]

Actually I found it, I didn't even realize you can hit ? in the FMP24 window to display options, learn something new today.

I knew about ? in the DSD window.

I have turn Auto tune off with A command.

Yea all I do is hit A on my keyboard on the fmp24 window but I have no idea if there's command line for it but I switch through different systems that all don't have issues with the auto tune besides the one Harris system I monitor