NXDN (lots of) missed transmissions

[troymail]

Paul -

I mentioned this in NC discussions but since it still isn't getting any better, I opted to toss this out here to ensure it is being addressed.

I'm monitoring a single frequency NXDN system along with a conventional NXDN channel. Yesterday, I also added 4-5 conventional frequencies that might be analog but are planned/recently licensed for NXDN (had to set them to "Digital Only" to overcome the analog squelch issue).

Anyway - watching the radio, I see the most solid 5 bars you can imagine when the single frequency NXDN system is being scanned - I assume because the frequency is both control and voice. Point being, however, is that the signal is SOLID.

And, yet, I've yet to solid reception of voice calls on this system. I have a TRX-1 sitting here also that is regularly hearing voice calls on this system (as well as the known active NXDN conventional channel) - some fairly lengthy - and the 436 misses well over 90% of that voice activity when I clearly see it scanning through the system at 5 bars. This situation if exhibiting the kinds of things I've seen in the past on simulcast systems but I'm fairly confident that this isn't any type of simulcast issue.

I have tried quite a number of things to see if I could pinpoint situations that are better than others -- HOLD on the system/site, switch threshold settings to MAN, HOLD on the single TGID, program a range of hundreds of TGIDs (most are unknown), switch from ID Search to ID Scan, etc. and I've not been able to make this much better if at all.

Obviously, this started with the initial NXDN firmware (1.14) and 1.15 isn't any better.

If you want a capture of something to look at, let me know (how).

Edit (0926): just installed 1.16 but given it's only a fix for the squlech bug, I'm not expecting much change....

 

[troymail]

More observations -- I was watching the display and am seeing various different things being shown for a system type -- mostly the regular pattern of NX4 but sometimes jumping around from NX4, to NX0 (and/or NXD), IDS, etc.

On a voice call it received, I noticed the signal bars drop from a solid 5 to a solid 4... so still pretty strong...

The biggest observation (which sort of makes sense) is that when the TRX is talking and the 436 is not, the 436 stops displaying ANY system type indicator (NX4, etc.) -- there are still a solid 4-5 bars but nothing else - no voice, no NXDN (or any other) indicator... It sort of makes sense in that the control channel signal on the frequency has been replaced by voice and I suspect that the radio simply missed the channel grant and now has nothing to tell it that the channel is now voice (other than the voice?)....

Following that - it just missed another voice call on the conventional channel also.... note: this conventional channel exhibits regular intervals of data bursts (probably location data) when there is no voice.

 

[troymail]

So, now I am holding on the SYSTEM... signal bars appear to be somewhat less solid/stable but still no less than 3 (mostly 4; not sure why solid 5 while scanning but only 3-4 when holding). I do see the regular interval of the momentary/blinking "housekeeping (?)" dropout. In ID Scan mode, the radio seems to pick up more voice transmissions (which again makes sense since the scanner is now only effectively scanning that one system/site frequency).

Are there no options to receive these type systems if the initial channel grant is missed (which clearly will happen most of the time if you are scanning more than just this one system/frequency)?

 

[UPMan]

Might be worth posting a debug log. It is Monday morning, and I'm prepping for a flight to Arkansas and back tomorrow, so not much time for me to study, today. But, engineering could look at the logs tonight.

 

[troymail]

Might be worth posting a debug log. It is Monday morning, and I'm prepping for a flight to Arkansas and back tomorrow, so not much time for me to study, today. But, engineering could look at the logs tonight.

Refresh me no the steps for the log... and do you want logs of HOLDing, scanning, both, other? Let's make this worth the effort...

 

[KE4ZNR]

Refresh me no the steps for the log... and do you want logs of HOLDing, scanning, both, other? Let's make this worth the effort...

When powering up, hold the AVOID key.

When powered-up, scan only what you want to log. Disable all others with Quick Keys, or de-selecting in the scan list, or whatever.

When scanning only what you want to log, go to MENU, then SETTINGS, then SET DEBUG LOG MODE, then SD CARD.

When you're done, go to MENU, then SETTINGS, then SET DEBUG LOG MODE, then OFF.

Connect the scanner to a PC and get the logs in the DEBUG folder, ZIP them, and attach to your post.

 

[LeSueurC]

I too noticed I was missing lots of transmissions on a NXDN 9600 system I was receiving Friday and Saturday. Have messed with the threshold, no improvement.

 

[troymail]

Might be worth posting a debug log. It is Monday morning, and I'm prepping for a flight to Arkansas and back tomorrow, so not much time for me to study, today. But, engineering could look at the logs tonight.

I have a log but due to size - even compressed - I cannot post it here -- it's not a really busy system but I heard multiple voice calls on the TRX-1 in the last few minutes of this log timeframe - no voice at all from the 436 - still solid 5 bars.

 

[W4ELL]

I have a log but due to size - even compressed - I cannot post it here -- it's not a really busy system but I heard multiple voice calls on the TRX-1 in the last few minutes of this log timeframe - no voice at all from the 436 - still solid 5 bars.

You could open the log and cut/paste half of the text into another text file. Then, compress both of the logs separately and upload them as separate, smaller zip files. (If that doesn't work, do it with 3 or 4 files).

You could also purge some of the log itself... as long as you have an example of your issue somewhere in the log, it should work.

 

[jonwienke]

If the most "interesting" part of the log is the final few minutes, then trim the beginning of the log file until it is small enough to meet the size limit, then post.

 

[troymail]

Let's try this:

https://www.dropbox.com/sh/esjxl306y0fcecf/AAAjW_rDUVq436CR8lSfh8vDa?dl=0

 

[troymail]

So I have definitely concluded that the x36 will almost never receive voice calls on a trunked NXDN single frequency ("composite", VOC) site unless the radio is allowed to monitor that control channel nearly 100% of the time (i.e. don't scan with your scanner). In my case, the 436 is scanning this and only 1 other conventional NXDN frequency and misses more than it receives on the trunk system. It seems pretty obvious that if I allowed to to scan even more, I'd probably almost never hear activity on this system.

I don't know if this can or will be improved upon (but I would think that it could be).

 

[wx5uif]

I wonder if it is because being programmed as trunked it misses the channel grant for the voice, even though it is the same channel.

 

[Voyager]

Program it as conventional?

 

[troymail]

I wonder if it is because being programmed as trunked it misses the channel grant for the voice, even though it is the same channel.

I'm fairly certain that is the case - the voice call replaces the control channel.

Program it as conventional?

It's a multi-site trunked system. Programmed as conventional just results in the radio hanging on the control channel.

 

[Voyager]

Hmmm.... OK. I thought it was a single frequency (that uses VOC).

Even if several sites have that, just program several conventional channels.

Not ideal, of course, but it might allow more effective monitoring.

 

[troymail]

Hmmm.... OK. I thought it was a single frequency (that uses VOC).

Even if several sites have that, just program several conventional channels.

Not ideal, of course, but it might allow more effective monitoring.

Ultimately, when Sentinel is NXDN smart, it will import the system as trunked with multiple sites so none of these suggestions would make any sense at that point (nor do they now)....

For giggles, I've programmed it as NXDN One Frequency to see what happens -- it's better than conventional - at least it isn't hanging on the control channel....

 

[WX4JCW]

Ultimately, when Sentinel is NXDN smart, it will import the system as trunked with multiple sites so none of these suggestions would make any sense at that point (nor do they now)....

For giggles, I've programmed it as NXDN One Frequency to see what happens -- it's better than conventional - at least it isn't hanging on the control channel....

Did you send the logs to UPMan?


Sent from my iPhone 8 using Tapatalk Pro
Jason WX4JCW
Unication G4, BCD536HP

 

[troymail]

Ultimately, when Sentinel is NXDN smart, it will import the system as trunked with multiple sites so none of these suggestions would make any sense at that point (nor do they now)....

For giggles, I've programmed it as NXDN One Frequency to see what happens -- it's better than conventional - at least it isn't hanging on the control channel....

Follow up - Nothing is received when programmed as NXDN One Frequency.... I even programmed a specific talkgroup and did a HOLD on that talkgroup - nothing other than signal bars and a blinking NX4 (when there was no voice).

I reverted back to HOLD on the system programmed as a NXDN trunk system and it receives fine (as long as I don't SCAN.... ).

 

[buddrousa]

Troy I have not tested 16 on the system 80 miles from me but it worked good on 15