locked Re: #AudioIssues #raspberryPi #linux #PiSDR #ALSA #pulseaudio All Audio Streams Dead to WSJTX - need help please #pulseaudio #AudioIssues #raspberryPi #linux

William Smith

I'm still wrapping my head around all of the variants of Linux sound, and this doesn't really help with your problem, but I've got a:

https://tapr.org/product/wspr/ ($32 plus the Pi, power supply, antenna)

Which puts out about 200 mW using direct synthesis (bit-banging a GPIO pin as an NTP-disciplined-oscillator?) and is heard all over the hemisphere from the Turks & Caicos as VP5WS on 40M WSPR, though usually only at night.

I know you are looking for better understanding of Linux sound, but if what you really wanted was a WSPR beacon, the above might be a simpler solution.

73, Willie N1JBJ

On Jul 30, 2021, at 5:41 PM, WTWisniewski@... wrote:

I am trying to operate WSJTX with SDRAngel and a LimeSDR as the transciever. At this time I intended to run unattended as a WSPR beacon. The host system is a Pi4b with PU2SPY's PiSDR Build 4.2 complemented with my re-installation of pulseaudio and addition of ARM 32 bit WSJTX 2.4.0.

- Audio stream out of WSJTX is silent and no audio or silent audio is processed as a blank waterfall in receive.
- All sensible stream devices (there are 17 to chose from including pulse, default, and on-board sound cards) behave the same - the stream gets registered but audio is silent. Additional pulseaudio streams created as virtual also appear listed in all apps, but still they don't convey sound in or out of WSJTX
- In all cases invoking the selected streams, the following error appears in the parent terminal:

ALSA lib pcm_usb_stream.c:508:(_snd_pcm_usb_stream_open) Unknown field hint
ALSA lib pcm_dmix.c:1108:(snd_pcm_dmix_open) unable to open slave
ALSA lib pcm_dsnoop.c:575:(snd_pcm_dsnoop_open) The dsnoop plugin supports only capture stream

- When pulse is selected, the pavucontrol gui shows the WSJTX streams in both input and output, but the level indicator bars show nothing generated by WSJTX transmit and streams with sound don't reach the waterfall.
- The initial tests were in WSJTX 2.3.1 and upgrading to 2.4.0 changed nothing.
- I have injected audio into pulseaudio using paplay, to then routed it successfully to SDRAngel, the analog sound cards, and it appears in pavucontrol normally as a level, but failed to get it into WSJTX. This suggests that pulseaudio is piping streams to other connections.

I am a newbie in OS audio resources, however I have successfully connected WSJTX and SDRAngel with pulseaudio on several Ubuntu distros successfully - so not a total newbie. This means that my expectations seem reasonable. I succeeded by putting outputs to default devices or virtual sinks, and setting inputs to stream "monitors", but have not made it through the WSJTX sound I/O to the rest of the system on a Pi.

Please help. A direct solution to my problem sure would help get the beacon on the air, but resources to learn how to troubleshoot these sound servers could help lots of computerized hams now and in the future. Software generated messages piped directly to SDRs are the future of most communications. How does one un-blackbox ALSA and pulseaudio and proceed logically? 

Thanks in advance,
73 Wit Wisniewski N7PHD

Join main@WSJTX.groups.io to automatically receive all group messages.