SDRTrunk Beta-3 brief or no audio

[John Jones]

My system is a WIN 10 with 32 gb. Have been running Version 0.5.0 Alpha 6 for quite a while, no issues. Just started testing with 0.5.0-beta3 Gradle 7.4.2. The application starts just fine and early on the audio has no issues. After a few minutes I get little to no audio. An occasional blip of audio but now mostly quiet. I see plenty of calls going in the events list. And listening to the same P25 system on OP25_Boatboat running on a Pi-4 audio is just fine. I jump back to 0.5.0 Alpha 6 and all is well.

Any suggestions? Thanks, John.

My 0.5.0 beta3 log

WARNING: Using incubator modules: jdk.incubator.vector
16:41:25.315 INFO  i.g.d.icon.IconModel - loading icons file [C:\Users\John\SDRTrunk\settings\icons.xml]  [11MB/512MB 2%]
16:41:25.320 INFO  i.g.d.icon.IconModel - Icons file not found at [C:\Users\John\SDRTrunk\settings\icons.xml]  [11MB/512MB 2%]
16:41:25.691 INFO  i.g.d.log.ApplicationLog - Application Log File: C:\Users\John\SDRTrunk\logs\sdrtrunk_app.log  [18MB/512MB 3%]
16:41:25.735 INFO  i.g.d.log.ApplicationLog - SDRTrunk Version  : 0.5.0-beta3  [20MB/512MB 4%]
16:41:25.735 INFO  i.g.d.log.ApplicationLog - Gradle Version    : Gradle 7.4.2  [20MB/512MB 4%]
16:41:25.735 INFO  i.g.d.log.ApplicationLog - Build Timestamp   : 2022-05-04T14:55:06.134-0400  [20MB/512MB 4%]
16:41:25.735 INFO  i.g.d.log.ApplicationLog - Build-JDK         : 18.0.1 (BellSoft 18.0.1+12  [20MB/512MB 4%]
16:41:25.735 INFO  i.g.d.log.ApplicationLog - Build OS          : Linux (amd64 5.13.0-40-generic  [20MB/512MB 4%]
16:41:25.735 INFO  i.g.d.log.ApplicationLog -   [20MB/512MB 4%]
16:41:25.740 INFO  i.g.d.log.ApplicationLog - *******************************************************************  [20MB/512MB 4%]
16:41:25.740 INFO  i.g.d.log.ApplicationLog - **** sdrtrunk: a trunked radio and digital decoding application ***  [20MB/512MB 4%]
16:41:25.740 INFO  i.g.d.log.ApplicationLog - ****  website: https://github.com/dsheirer/sdrtrunk             ***  [20MB/512MB 4%]
16:41:25.740 INFO  i.g.d.log.ApplicationLog - *******************************************************************  [20MB/512MB 4%]
16:41:25.740 INFO  i.g.d.log.ApplicationLog - Memory Logging Format: [Used/Allocated PercentUsed%]  [20MB/512MB 4%]
16:41:25.740 INFO  i.g.d.log.ApplicationLog - Host OS Name:          Windows 10  [20MB/512MB 4%]
16:41:25.740 INFO  i.g.d.log.ApplicationLog - Host OS Arch:          amd64  [20MB/512MB 4%]
16:41:25.740 INFO  i.g.d.log.ApplicationLog - Host OS Version:       10.0  [20MB/512MB 4%]
16:41:25.740 INFO  i.g.d.log.ApplicationLog - Host CPU Cores:        6  [20MB/512MB 4%]
16:41:25.740 INFO  i.g.d.log.ApplicationLog - Host Max Java Memory:  7 GB  [20MB/512MB 4%]
16:41:25.740 INFO  i.g.d.log.ApplicationLog - Storage Directories:  [20MB/512MB 4%]
16:41:25.740 INFO  i.g.d.log.ApplicationLog -  Application Root: C:\Users\John\SDRTrunk  [20MB/512MB 4%]
16:41:25.745 INFO  i.g.d.log.ApplicationLog -  Application Log:  C:\Users\John\SDRTrunk\logs  [20MB/512MB 4%]
16:41:25.745 INFO  i.g.d.log.ApplicationLog -  Event Log:        C:\Users\John\SDRTrunk\event_logs  [20MB/512MB 4%]
16:41:25.745 INFO  i.g.d.log.ApplicationLog -  Playlist:         C:\Users\John\SDRTrunk\playlist  [20MB/512MB 4%]
16:41:25.745 INFO  i.g.d.log.ApplicationLog -  Recordings:       C:\Users\John\SDRTrunk\recordings  [20MB/512MB 4%]
16:41:25.745 INFO  i.g.d.util.ThreadPool - Application thread pool created SCHEDULED and CACHED executors threads  [20MB/512MB 4%]
16:41:25.750 INFO  i.g.dsheirer.gui.SDRTrunk - Home path: C:\Users\John\SDRTrunk  [20MB/512MB 4%]
16:41:25.755 INFO  i.g.d.p.SystemProperties - SystemProperties - loaded [C:\Users\John\SDRTrunk\SDRTrunk.properties]  [20MB/512MB 4%]
16:41:25.755 INFO  i.g.d.p.SystemProperties - SystemProperties - application properties loaded [C:\Users\John\SDRTrunk\SDRTrunk.properties]  [20MB/512MB 4%]
16:41:26.090 INFO  i.g.d.s.t.m.TunerManager - Discovering tuners ...  [7MB/56MB 12%]
16:41:26.200 INFO  i.g.d.s.t.m.TunerManager - LibUsb API Version: 1.0.262  [8MB/56MB 14%]
16:41:26.200 INFO  i.g.d.s.t.m.TunerManager - LibUsb Version: 1.0.22.11312  [8MB/56MB 14%]
16:41:26.310 INFO  i.g.d.s.t.m.TunerManager - LibUsb - discovered [18] potential usb devices  [8MB/56MB 14%]
16:41:26.320 INFO  i.g.d.s.t.m.TunerManager - Discovered tuner at USB Bus [4] Port [8] Tuner Class [RTL-2832]  [8MB/56MB 15%]
16:41:26.320 INFO  i.g.d.s.t.m.TunerManager - Discovered tuner at USB Bus [3] Port [3] Tuner Class [RTL-2832]  [8MB/56MB 15%]
16:41:26.320 INFO  i.g.d.s.t.m.TunerManager - Discovered tuner at USB Bus [4] Port [6] Tuner Class [Airspy]  [8MB/56MB 15%]
16:41:26.320 INFO  i.g.d.s.t.m.TunerManager - Tuner: USB Tuner - RTL-2832 USB Bus:4 Port:8 - Added / Disabled  [8MB/56MB 15%]
16:41:26.320 INFO  i.g.d.s.t.m.TunerManager - Tuner: USB Tuner - RTL-2832 USB Bus:3 Port:3 - Added / Disabled  [8MB/56MB 15%]
16:41:26.320 INFO  i.g.d.s.t.m.TunerManager - Tuner: USB Tuner - Airspy USB Bus:4 Port:6 - Added / Starting ...  [8MB/56MB 15%]
16:41:26.380 WARN  i.g.d.v.VectorUtilities - CPU supports maximum SIMD instructions of Species[float, 4, S_128_BIT]  [11MB/56MB 20%]
16:41:26.480 INFO  i.g.d.d.f.c.ComplexPolyphaseChannelizerM2 - Sample Rate [6000000.0] providing [240] channels at [25000.0] Hz each  [13MB/56MB 23%]
16:41:26.480 INFO  i.g.d.s.t.m.TunerManager - Tuner: USB Tuner - Airspy USB Bus:4 Port:6 - Applying Tuner Configuration  [13MB/56MB 23%]
16:41:26.490 INFO  i.g.d.s.t.m.TunerManager - LibUsb Hotplug event notification Is Not Supported on this platform.  [13MB/56MB 23%]
16:41:26.490 INFO  i.g.d.s.SettingsManager - SettingsManager - loading settings file [C:\Users\John\SDRTrunk\settings\settings.xml]  [13MB/56MB 24%]
16:41:27.701 INFO  i.g.d.p.PlaylistManager - Loading playlist [C:\Users\John\SDRTrunk\playlist\saved playlist ver 3.xml]  [18MB/56MB 33%]
16:41:28.753 INFO  i.g.dsheirer.gui.SDRTrunk - starting main application gui  [57MB/284MB 20%]
16:41:31.722 INFO  i.g.d.c.c.ChannelAutoStartFrame - Channel auto-start canceled by user  [154MB/284MB 54%]
16:41:58.401 INFO  i.g.d.a.c.m.JmbeAudioModule - Loading JMBE library from [C:\Users\John\SDRTrunk\jmbe\jmbe-1.0.9.jar]  [159MB/540MB 29%]
16:41:58.412 INFO  i.g.d.a.c.m.JmbeAudioModule - JMBE audio conversion library loaded: JMBE Audio Conversion Library v1.0.9  [160MB/540MB 29%]
16:41:58.413 INFO  i.g.d.a.c.m.ImbeAudioModule - JMBE audio conversion library IMBE CODEC successfully loaded - P25-1 audio will be available  [160MB/540MB 29%]
16:41:58.447 INFO  i.g.d.d.f.c.ComplexPolyphaseChannelizerM2 - Sample Rate [6000000.0] providing [240] channels at [25000.0] Hz each  [161MB/540MB 29%]

[sdrtrunk]

Check the PPM on all of your tuners.  Changes in version 0.5.0 Beta 3 may cause your previous tuner PPM settings to be lost and reset to 0.0.  

[Chris Simons]

I also experienced this issue, another reason for staying on Alpha 6.  I use Trunking Recorder and would have empty audio files play, I thought it was just me until I tested, there is missing audio in the recorded files.

[John Jones]

So, I have continued to test various combinations to see what may be causing the loss of audio shortly after starting SDRTRUNK. Going to the messages panel when the audio would stop (or hearing just very short bursts) continuous sync-loss messages are display. This is happening on both the Air Spy mini and the RTL-SDR v3 dongles. The RTL-SDR v3 has an average signal strength of -13 while the air Spy mini runs around -37 (depending on the frequency received)

When starting the program I can get PPM as good as .3 PPM. The loss of audio (sync loss) occurs whether "enable decoder(s) to auto-adjust PPM" is checked or not. 

The program starts with the correct control channel freq in the Tuners display. After a little bit that frequency changes as switching occurs. What I am not sure of is why would there ever be a frequency displayed that is not a valid one on the system that I am tuned to. However in the Events display the correct voice frequency is displayed. 

Once we have lost sync to only way get it back is to restart the program.

 

[wemana...@gmail.com]

On Monday, May 30, 2022 at 9:46:18 AM UTC-4 john.f.j...@gmail.com wrote:

So, I have continued to test various combinations to see what may be causing the loss of audio shortly after starting SDRTRUNK. Going to the messages panel when the audio would stop (or hearing just very short bursts) continuous sync-loss messages are display. This is happening on both the Air Spy mini and the RTL-SDR v3 dongles. The RTL-SDR v3 has an average signal strength of -13 while the air Spy mini runs around -37 (depending on the frequency received)

You list in another message you have a Pi with OP25 boatbod on it.... can you SSH X  ( ssh -XCl user pibox ) errrr... wait... you likely don't have OP25 setup the way I do... goto the OP25 terminal, and hit 2 WHILE OP25 is IDLING ON THE CC!! Not a VOICE CHANNEL... ON A CC... HIT 2 this brings up the constellation, and on it you want to have 4 DOTS with as few as possible pixels running around outside those dots. 4 NICE TIGHT DOTS... A few outside of there is fine, like this:

https://user-images.githubusercontent.com/18183123/51884351-e94eee80-2354-11e9-92af-a964b0216699.png

Is this a SHARED ANTENNA setup for OP25 and SDRT????

The goal/aim/purpose is to ensure you have a good signal to start.. the CONSTELLATION In P25 is the key to this, as well as some other, but its easier to understand that one than the others..

Confirming that, means we look hard at SDRT and not RF components ie: antenna, coax, location etc...

So it looks like from your post you are good, but the constellation will tell the tale.... : ) ;) ;) 

When starting the program I can get PPM as good as .3 PPM. The loss of audio (sync loss) occurs whether "enable decoder(s) to auto-adjust PPM" is checked or not. 

What do you have the INITIAL PPM set to??? My Airspy Minis I set at 1.0 for the INITIAL PPM and let it auto adjust from there... they seem to run at 1.4PPM so I see a 0.4 PPM adjustment mostly... Same with the V3's I set at 1.5 and my Hydra I set at 2.6.. this had caused me all kinds of issues on CC rotation (earlier versions, and still, if the CC rotates from Primary to first alternate, we had some base failures which forced this, I keep first alt CC on for issues, just in case..I don't use the auto rotate option....personal choice for me.. even that few seconds of down time is a failure for me, **me**.)
 

The program starts with the correct control channel freq in the Tuners display. After a little bit that frequency changes as switching occurs. What I am not sure of is why would there ever be a frequency displayed that is not a valid one on the system that I am tuned to

Can you be specific on this??? I've seen OP25 display WRONG FREQUENCIES when it decodes things wrong, which happens... I've NEVER SEEN SDRT do that if what you are describing is the same thing... ie: a voice grant to the WRONG TRAFFIC CHANNEL/FREQUENCY.... The CRC detection in SDRT I don't think would permit it if it decodes bad data...... OP25 doesn't do that as much.. so it happily goes to what ever it gets told to do ...

SDTR will start with the CC you give it.. and then it will "adjust/retune" the CENTER FREQUENCY of the SDR to try to accommodate channels within the bandwidth it has....My Hydras set at 853.xxx which is the HIGHEST channel we have, and then works down from there... and 95% of the channels fit in that 2.4Mhz range, theres a couple that a 2nd tuner of the Hydra is brought up to handle, this is rare except at high volume periods..

So you will ON INITIAL SETUP see things sort of jump around a bit on the waterfall/spectrum as the SDR settles into a groove to handle the channels for your system... So the FREQUENCY DISPLAYED ON THE SPECTRUM AND TUNER TAB will HAVE NO CORRELATION to ANY CHANNEL IN YOUR SYSTEM, my one hydra is locked at 852.700 that ain't a channel I am assigned. Another setup is at 770.33125, that is not assigned to that system either (this from the tuners tab!)  It would be great if the spectrum showed only valid channels/frequencies, in the way the band plan assigns them, ie: 12.5khz, 6.25khz etc...

 

. However in the Events display the correct voice frequency is displayed. 

Once we have lost sync to only way get it back is to restart the program.

I seen similar issues when I was playing with something, but I chalked this up overloading the USB with 6-10--12Mhz of bandwidth those boxes didn't like... hmmmm... test production box is running 050b2, main production is A6, as not a chance to update it yet....which when I ran these on other boxes on earlier versions I didn't think I had issues.. in re USB saturation... so I will have to test those out again ....hmmmm...

Just curious... if you STOP, RESTART the channel it still won't sync? A a HARD RESET is required???? I'd have to rerun the tests again, but I though I could restart things with just a stop/start on the channel...

Just for testing sake... if you are using the POLYPHASE CHANNELIZER ****SWITCH**** to the HETRODYNE one and see does this change any result???

 

[John Jones]

Dean, thanks for the message and details.

You list in another message you have a Pi with OP25 boatbod on it.... can you SSH X  ( ssh -XCl user pibox ) errrr... wait... you likely don't have OP25 setup the way I do... goto the OP25 terminal, and hit 2 WHILE OP25 is IDLING ON THE CC!! Not a VOICE CHANNEL... ON A CC... HIT 2 this brings up the constellation, and on it you want to have 4 DOTS with as few as possible pixels running around outside those dots. 4 NICE TIGHT DOTS... A few outside of there is fine, like this:

My constellation display is good.

Is this a SHARED ANTENNA setup for OP25 and SDRT???? Dedicated antenna for each dongle - In fact the AirSpy mini is on the 800 mhz beam antenna. The main site is only 9 miles from my house.

The goal/aim/purpose is to ensure you have a good signal to start.. the CONSTELLATION In P25 is the key to this, as well as some other, but its easier to understand that one than the others..

Confirming that, means we look hard at SDRT and not RF components ie: antenna, coax, location etc...

So it looks like from your post you are good, but the constellation will tell the tale.... : ) ;) ;) 

When starting the program I can get PPM as good as .3 PPM. The loss of audio (sync loss) occurs whether "enable decoder(s) to auto-adjust PPM" is checked or not. 

What do you have the INITIAL PPM set to??? My Airspy Minis I set at 1.0 for the INITIAL PPM and let it auto adjust from there... they seem to run at 1.4PPM so I see a 0.4 PPM adjustment mostly... Same with the V3's I set at 1.5 and my Hydra I set at 2.6.. this had caused me all kinds of issues on CC rotation (earlier versions, and still, if the CC rotates from Primary to first alternate, we had some base failures which forced this, I keep first alt CC on for issues, just in case..I don't use the auto rotate option....personal choice for me.. even that few seconds of down time is a failure for me, **me**.)

My initial PPM is set to -1.0 and like you I see about the same adjustment. I have done both the single control channel and the rotate option. Same results for both. With my offset of -1.0 I have a measured error of 68Hz.
 

The program starts with the correct control channel freq in the Tuners display. After a little bit that frequency changes as switching occurs. What I am not sure of is why would there ever be a frequency displayed that is not a valid one on the system that I am tuned to

Can you be specific on this??? I've seen OP25 display WRONG FREQUENCIES when it decodes things wrong, which happens... I've NEVER SEEN SDRT do that if what you are describing is the same thing... ie: a voice grant to the WRONG TRAFFIC CHANNEL/FREQUENCY.... The CRC detection in SDRT I don't think would permit it if it decodes bad data...... OP25 doesn't do that as much.. so it happily goes to what ever it gets told to do .

Yes, this is the part that confuses me as shortly after startup the frequency displayed may be one that is not active on my local site. Once this has occurred a few times we start seeing the Sync Loss messages. My control channels arr 853.4375 and 853.9375. Voice channels are from 851.825 to 853.6875.  

The AirSpy typically has a signal strength of -48db. While the RTL-SDR runs around -20db. Each, while on the same 850 mhz beam antenna. 

SDTR will start with the CC you give it.. and then it will "adjust/retune" the CENTER FREQUENCY of the SDR to try to accommodate channels within the bandwidth it has....My Hydras set at 853.xxx which is the HIGHEST channel we have, and then works down from there... and 95% of the channels fit in that 2.4Mhz range, theres a couple that a 2nd tuner of the Hydra is brought up to handle, this is rare except at high volume periods..

So you will ON INITIAL SETUP see things sort of jump around a bit on the waterfall/spectrum as the SDR settles into a groove to handle the channels for your system... So the FREQUENCY DISPLAYED ON THE SPECTRUM AND TUNER TAB will HAVE NO CORRELATION to ANY CHANNEL IN YOUR SYSTEM, my one hydra is locked at 852.700 that ain't a channel I am assigned. Another setup is at 770.33125, that is not assigned to that system either (this from the tuners tab!)  It would be great if the spectrum showed only valid channels/frequencies, in the way the band plan assigns them, ie: 12.5khz, 6.25khz etc...

Thanks for the operational details. That explains the frequencies I am seeing.

 

. However in the Events display the correct voice frequency is displayed. 

Once we have lost sync to only way get it back is to restart the program.

I seen similar issues when I was playing with something, but I chalked this up overloading the USB with 6-10--12Mhz of bandwidth those boxes didn't like... hmmmm... test production box is running 050b2, main production is A6, as not a chance to update it yet....which when I ran these on other boxes on earlier versions I didn't think I had issues.. in re USB saturation... so I will have to test those out again ....hmmmm...

Just curious... if you STOP, RESTART the channel it still won't sync? A a HARD RESET is required???? I'd have to rerun the tests again, but I though I could restart things with just a stop/start on the channel...

Yes, once the system has gone south a stop/restart will not bring things back to life. A full shutdown and restart of the APP is necessary to get things going again. 

Just for testing sake... if you are using the POLYPHASE CHANNELIZER ****SWITCH**** to the HETRODYNE one and see does this change any result???

Been testing in Heterodyne mode since yesterday. The receiver has stayed running  over 24 hours on RTL-SDR V3 dongle. I just re-started SDRTrunk on the Air-Spy mini. I want to see how long it runs on that dongle.

In retrospect I am beginning to think that the issue might be twofold. 1. the Air Spy has a lower signal sensitivity than the RTL-SDR. 2. possibly the amount of PPM offset, I suspect maybe the signal is dropping too low and SDRTrunk cannot recover.   

Next I will try going back to PolyPhase and see how it behaves.

Thanks guys for the help. 

--
You received this message because you are subscribed to a topic in the Google Groups "sdrtrunk" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/sdrtrunk/uJaxqcux52A/unsubscribe.
To unsubscribe from this group and all its topics, send an email to sdrtrunk+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/sdrtrunk/ed83393b-6c70-4f40-b81c-69122eb0e8f4n%40googlegroups.com.

[John Jones]

Running on the AirSpy again tried both PolyPhase and Hetrodyne. Give or take an hour or so I started to get Sync Loss messages on either channelizer mode. Looked at the Channel display and the signal had dropped from -35 db to -68 db. I then went to the Tuners tab and the Gain Control was on Linearity was set at around a 2 setting (had not adjusted this control before) I bumped it up to 18 and my signal level raised to --33db and the audio returned immediately. I am going to let it run on Hetrodyne for a while and see what happens. If it runs for a good long time I'll then try Polyphase.

The RTL-SDR V3 dongle seems to be a little more forgiving when it comes to gain performance in BETA 3. I'd prefer to run the Air Spy for the wider band coverage as our local system is recently expanding to almost 4 mhz.

Thanks, John.