[arm-allstar] Distorted audio (sampling?)

Alessio Sacchi iz4efn at gmail.com
Sun Sep 15 18:24:48 EDT 2019


David,  will replace the power supply on my next visit with a traditional
regulator getting 5V out of the 12V where everything else is connected to,
that's for sure.

Regarding ground loops and RF noise, I am quite lucky the location has no
high power TV or FM broadcast but just a number of other PMR repeaters on
160-175 Mhz and 460 MHz, fire station system links at 70Mhz (this is very
Italian) and then us. However you bring up a good point I always thought
about: the URI have no opto coupling for PTT and COS, and audio is also
directly coupled. I did put two radioshack audio transoformers on the audio
lines for the 1200 MHz port before taking it up to the hill. I was lazy for
the oters, my bad. I guess every one of you do that on top of the URI
instead of plugging it as it is to a radio? BTW, when the distortion
occurred, it was affecting also 1200 users who were isolated at least on
the audio line. Could still have been a ground loop from the common ground
of PTT and COS though.

dmesg - thanks I didn't think about it. I looked at dmesg since the last
reboot and found a few things I don't like:

[    6.204436] FAT-fs (mmcblk0p1): Volume was not properly unmounted. Some
data may be corrupt. Please run fsck.
[...]
[13436.127396] Transfer to device 5 endpoint 0x2 frame 980 failed - FIQ
reported NYET. Data may have been lost.
[13436.127698] Transfer to device 4 endpoint 0x2 frame 980 failed - FIQ
reported NYET. Data may have been lost.
[14513.031406] Transfer to device 7 endpoint 0x2 frame 636 failed - FIQ
reported NYET. Data may have been lost.
[14513.031717] Transfer to device 5 endpoint 0x2 frame 636 failed - FIQ
reported NYET. Data may have been lost.
[82883.543637] Transfer to device 7 endpoint 0x2 frame 716 failed - FIQ
reported NYET. Data may have been lost.
[82883.543947] Transfer to device 5 endpoint 0x2 frame 716 failed - FIQ
reported NYET. Data may have been lost.
[95631.951656] Transfer to device 4 endpoint 0x2 frame 324 failed - FIQ
reported NYET. Data may have been lost.
[95631.951949] Transfer to device 7 endpoint 0x2 frame 324 failed - FIQ
reported NYET. Data may have been lost.
[107403.807689] Transfer to device 7 endpoint 0x2 frame 276 failed - FIQ
reported NYET. Data may have been lost.
[107403.807996] Transfer to device 4 endpoint 0x2 frame 276 failed - FIQ
reported NYET. Data may have been lost.
[132821.807783] Transfer to device 4 endpoint 0x2 frame 548 failed - FIQ
reported NYET. Data may have been lost.
[132821.808094] Transfer to device 7 endpoint 0x2 frame 548 failed - FIQ
reported NYET. Data may have been lost.
[134794.135780] Transfer to device 5 endpoint 0x2 frame 652 failed - FIQ
reported NYET. Data may have been lost.
[134794.148734] Transfer to device 4 endpoint 0x2 frame 665 failed - FIQ
reported NYET. Data may have been lost.
[154885.673839] Transfer to device 5 endpoint 0x2 frame 1310 failed - FIQ
reported NYET. Data may have been lost.
[154885.686353] Transfer to device 4 endpoint 0x2 frame 1323 failed - FIQ
reported NYET. Data may have been lost.

Alessio IZ4EFN

On Mon, Sep 16, 2019 at 12:00 AM David McGough <kb4fxc at inttek.net> wrote:

>
> This is very strange behavior. When the audio issues are occurring, do you
> see any kernel error messages?  From a bash command prompt, use the
> "dmesg" command to see the kernel/console messages.
>
> A marginal or faulty 5V power supply would be my #1 suspect. A reboot
> radically changes the power supply load, which might temporarily snap it
> back to proper operation (I have experienced this with a wall wart
> supply)...OR, perhaps some type of ground loop or RF/noise getting into
> the low-level analog audio.....I have experienced ground loop noise that
> comes and goes.
>
>
> 73, David KB4FXC
>
>
>
> On Sun, 15 Sep 2019, Alessio Sacchi wrote:
>
> > Thanks David. Got some samples converted.
> >
> > Restarting asterisk did not solve the proble. Rebooting the PI fixed it
> > twice - after the first reboot I described above, it started degrading
> > again and I had to restart the RPi again after about 24 hrs. We are now
> > running since 3 days from that restart and audio is still good. But most
> > importantly no - stopping and starting asterisk was not helping at all.
> >
> > Alessio
> >
> > On Sat, Sep 14, 2019 at 2:05 AM David McGough <kb4fxc at inttek.net> wrote:
> >
> > >
> > > Alessio,
> > >
> > > Interesting that rebooting the RPi3B fixes the issue. Does stopping and
> > > starting Asterisk also fix the issue?
> > >
> > > To convert uLAW (pcm) files to wav files, one way is with the sox
> > > software, like:
> > >
> > > sox -t ul -c 1 -r 8000 allstar_audio_file.pcm  output_file.wav
> > >
> > >
> > > 73, David KB4FXC
> > >
> > >
> > > On Fri, 13 Sep 2019, "Alessio Sacchi via ARM-allstar" wrote:
> > >
> > > > Danny, David
> > > >
> > > > I could certainly take a generator up to the site on my next trip.
> > > Speaking
> > > > of adding a tone, I remember I used a "generate continuous tone" in
> ASL
> > > > usb-tune tool. That helped me at quantifying the distortion back in
> my
> > > ASL
> > > > testing days.YOu kept it running for seconds and the tone broke up
> > > > intermittently eventually.
> > > >
> > > > There's a quite important update on the degraded audio issue. I
> continued
> > > > my controlled manner testing by disabling things one by one.
> > > Specifically I
> > > > had a script checking gpio to sense loss of mains power - killing it
> > > > doesn't change anything. I eventually tried a reboot which fixed the
> > > > problem immediately. But the issue come back in the same form less
> than
> > > 24
> > > > hours later. Users reported this thing gets worse over time. Another
> > > reboot
> > > > fixed it again. I am back doubtful of my power supply, unless you
> have
> > > > other ideas. Am I wrong thinking 3 ports could be too many for a
> single
> > > PI
> > > > 3B?
> > > >
> > > > I'll procure some uLAW recordings as soon as I can listen to them.
> What
> > > do
> > > > you reckon to play the uLAW files with? I tried  importing them as
> raw
> > > > files with audacity with no luck.Maybe I am configuring it wrongly
> during
> > > > the import.
> > > >
> > > > Thanks,
> > > > Alessio
> > > >
> > > > On Thu, Sep 12, 2019 at 5:46 PM "Danny K5CG via ARM-allstar" <
> > > > arm-allstar at hamvoip.org> wrote:
> > > >
> > > > > Alessio,
> > > > >
> > > > > It sounds like you need to do some testing in a controlled manner
> to
> > > > > eliminate as many unknowns as possible.
> > > > >
> > > > > Do you have a service monitor? Generate input RF, add a tone,
> record
> > > it.
> > > > > Is the distortion still there?
> > > > >
> > > > > Danny
> > > > > K5CG
> > > > >
> > > > > ----- Original Message -----
> > > > > From: "ARM Allstar" <arm-allstar at hamvoip.org>
> > > > > To: "ARM Allstar" <arm-allstar at hamvoip.org>
> > > > > Cc: "Alessio Sacchi" <iz4efn at gmail.com>
> > > > > Sent: Thursday, September 12, 2019 9:55:51 AM
> > > > > Subject: Re: [arm-allstar] Distorted audio (sampling?)
> > > > >
> > > > > Hi David, thanks. I will immediately begin recording with uLAW and
> > > find a
> > > > > few samples to share.
> > > > >
> > > > > >>is the same noise heard on all your AllStar connected receivers
> or
> > > just
> > > > > one?
> > > > > All connected receivers. To clarify:
> > > > > - A distorsion over a signal coming from the 50MHz port is heard on
> > > > > echolink, on 70cm and on 23cm
> > > > > - Distorsion can happen regardless of the input ports - except for
> > > audio
> > > > > coming from echolink. Meaning that signals coming from each of the
> > > radio
> > > > > ports can be affected by distorsion. Audio from echolink seems to
> be
> > > pushed
> > > > > out clean at all times.
> > > > >
> > > > > Alessio
> > > > >
> > > > > On Thu, Sep 12, 2019 at 4:01 PM "David McGough via ARM-allstar" <
> > > > > arm-allstar at hamvoip.org> wrote:
> > > > >
> > > > > >
> > > > > > BTW, I noticed that the dropbox .wav files are in the GSM CODEC
> > > format.
> > > > > > For diagnosing audio issues, I recommed saving the files using
> the
> > > much
> > > > > > higher audio quality uLAW CODEC, rather than GSM. The negative
> > > drawback
> > > > > is
> > > > > > that the files will be bigger, of course.
> > > > > >
> > > > > > To enable saving uLAW formatted files, in your
> /etc/asterisk/rpt.conf
> > > > > file
> > > > > > node stanzas, add:
> > > > > >
> > > > > > archivetype=pcm
> > > > > >
> > > > > >
> > > > > > 73, David KB4FXC
> > > > > >
> > > > > >
> > > > > > On Thu, 12 Sep 2019, "David McGough via ARM-allstar" wrote:
> > > > > >
> > > > > > >
> > > > > > > Hi Alessio,
> > > > > > >
> > > > > > > To me, the distortion heard in your recordings doesn't sound
> it's
> > > > > coming
> > > > > > > from the HamVoIP/AllStar software. My first suspicion is noise
> on
> > > the
> > > > > > > receiver audio, which leads me to a question: is the same noise
> > > heard
> > > > > > > on all your AllStar connected receivers or just one?
> > > > > > >
> > > > > > >
> > > > > > > 73, David KB4FXC
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > >
> > > > > > _______________________________________________
> > > > > >
> > > > > > ARM-allstar mailing list
> > > > > > ARM-allstar at hamvoip.org
> > > > > > http://lists.hamvoip.org/cgi-bin/mailman/listinfo/arm-allstar
> > > > > >
> > > > > > Visit the BBB and RPi2/3 web page - http://hamvoip.org
> > > > > >
> > > > > _______________________________________________
> > > > >
> > > > > ARM-allstar mailing list
> > > > > ARM-allstar at hamvoip.org
> > > > > http://lists.hamvoip.org/cgi-bin/mailman/listinfo/arm-allstar
> > > > >
> > > > > Visit the BBB and RPi2/3 web page - http://hamvoip.org
> > > > > _______________________________________________
> > > > >
> > > > > ARM-allstar mailing list
> > > > > ARM-allstar at hamvoip.org
> > > > > http://lists.hamvoip.org/cgi-bin/mailman/listinfo/arm-allstar
> > > > >
> > > > > Visit the BBB and RPi2/3 web page - http://hamvoip.org
> > > > >
> > > > _______________________________________________
> > > >
> > > > ARM-allstar mailing list
> > > > ARM-allstar at hamvoip.org
> > > > http://lists.hamvoip.org/cgi-bin/mailman/listinfo/arm-allstar
> > > >
> > > > Visit the BBB and RPi2/3 web page - http://hamvoip.org
> > > >
> > >
> > >
> >
>
>


More information about the ARM-allstar mailing list