UniTrunker P25 decoding tips...

[Forts]

Anyone have any tips for P25 decoding with UT? I'm running release 19 from a tapped Maxtrac and I can only seem to get about 30% decode quality (EDACS decoding is 100% usually). I've tried messing around with my tap a little bit, as well as using the Mic in vs Line in (no difference). If I boost the volume in the mixer at all the decode rate seems to drop. My current tap is setup with a 10uf cap and 27k resistor, as well as a 4.7k between the cap an 27k to ground.

Even with this low decode rate UT seems to still work pretty well, however I'm fussy and would like to see a higher decode rate

Any tips appreciated!

 

[SCPD]

Any tips appreciated!

What does the waveform look like?

 

[Forts]

Very flat... as in basically a small ripple. As soon as I increase the volume slider in the mixer my decode rate drops to almost nothing.

 

[slicerwizard]

My current tap is setup with a 10uf cap and 27k resistor, as well as a 4.7k between the cap an 27k to ground.

So you've wired it like so?

tap---10uF---+---27k---mic in
             |
            4.7k
             |
gnd----------+---------gnd

That's pretty much the same as

tap---10uF-------27k---mic in

gnd--------------------gnd

Neither provide the required voltage divider. Without it, 4 level decoding is really going to suffer.

The 4.7k goes after the 27k, not before.

 

[Comint]

Double post - deleted.

 

[Comint]

The correct voltage divider diagram.

tap---10uF---27k--+-----mic in
                  |
                 4.7k
                  |
gnd---------------+-----gnd

An alternative layout, which is electrically the same as above.

tap---10uF---+
             |
            27K
             |
             +------mic in
             |
            4.7K
             |
gnd----------+------gnd

--
Comint

 

[Forts]

Thanks for the suggestions... I'll crack the radio and see how I did it, but I believe I have the 4.7k before the 27k.

 

[Forts]

Well, moving the 4.7k actually made it worse. I was using the Line In on my sound card and the decode dropped from ~30% to 0. Switching to Mic In gives me about 15%.

 

[SCPD]

Forts; can you post a picture of the waveform?

 

[bowser22]

i put a 10uf cap and a 10k resistor inline of the tap like this

Disc---+10uf-----10k resistor--------Line In+


Gnd------------------------------------------Line in-

It dropped my audio levels so low dsd could barely discern the audio.
After i removed it and went straight through i get a clean decode of p25 with 40% inlvl.

 

[gary123]

on the maxtrac take your signal from P601 pin B. this is a buffered and high passed filter point. you should see an exellent waveform here.

I am using this point for DSD with exellent results. I do use a 10u cap for DC isolation.

 

[Forts]

Forts; can you post a picture of the waveform?

Here is the waveform I'm currently getting..

 

[Forts]

on the maxtrac take your signal from P601 pin B. this is a buffered and high passed filter point. you should see an exellent waveform here.

I am using this point for DSD with exellent results. I do use a 10u cap for DC isolation.

Thanks, I will look into that!

.... ok, where in the world is P601? I'm not seeing it here at all

 

[SCPD]

Here is the waveform I'm currently getting..

The waveform is C4FM which explains the low decode rate. The program performs much better with CQPSK. This is a deficiency I'm working to improve.

 

[Forts]

Ah ok, thanks for clearing that up! This is a new demo system that Harris is tinkering with, and the only P25 system within earshot for me.

 

[SCPD]

The waveform is C4FM which explains the low decode rate. The program performs much better with CQPSK. This is a deficiency I'm working to improve.

C4FM decoding has greatly improved for the next release.

 

[Forts]

Great! Looking forward to it!

 

[Mike_G_D]

Odd - I always thought that C4FM was the "easy one" for everyone versus CQPSK being the tougher to deal with?

C4FM = Compatible 4-level Frequency Modulation - pretty easy to demodulate with basic FM hardware.

CQPSK = Compatible Quadrature Phase Shift Keying - not so much, especially with LSM and other multipath constructive and destructive phasing issues.

-Mike

 

[Forts]

Hate to dredge this up again, but I'm still not having much success with Release 22. I moved the tap in my Maxtrac to Pin B for JP601 and same as before.... EDACS decode is excellent, P25 sucks (around 15% at best). Tap is setup as Comint described in post #6 above, and the waveform still looks the same as what I posted earlier.

Anyone else running a tapped radio on a C4FM P25 system and getting good decode? If so, would you mind describing your tap setup for me?

Thanks!

Update: I was able to pull in the MPSCS P25 system from Michigan tonight (which is CQPSK) at around 85-90% quality, so it seems my issues are only with C4FM systems.

 

[GroundLoop]

on the maxtrac take your signal from P601 pin B. this is a buffered and high passed filter point. you should see an exellent waveform here.

From what I can tell, pin B picks up the signal after the 3kHz low-pass filter, but BEFORE the 300hz high-pass filter. If you want the high-pass filter included, that should be "A", no?