Just a quick post so I don’t lose this command… (!)
$ parec --latency-msec=10 -d alsa_input.pci-0000_00_1b.0.analog-stereo --stream-name=Sidetone-source | pacat --latency-msec=10 -d alsa_output.pci-0000_00_1b.0.analog-stereo --stream-name=Headset-sidetone &
I’m setting up for iCW (Internet CW) on Linux and this is the pulseaudio pipe command to get a low latency monitor of my laptop’s line in (FT-817 audio is plugged in as a keyer) to my headphones. By default, if I load the pulseaudio loopback module, the latency is far too high to work the keyer… but all those months trying to get pulse working for the network radio setup have paid off! Instead of using pulse’s loopback module, the command above pipes the specified input source to the specified output sink and allows me to specify a target latency. 10ms is just a first guess here but it seems to be OK.
More to follow, probably a new page when I’ve got it all working…
Finally got around to hooking my new Red Pitaya up to the PA0KLT VFO and having a play around with the oscilloscope and spectrum analyser applications. Very easy to hook up on the dining table, connected to the laptop via WiFi and seems fairly responsive given that it’s driven through a web browser so a minimum of cables out on the table:
Basic operations on the scope seem easy enough:
The frequency measurement on the scope appears fairly accurate (the VFO isn’t calibrated yet) but the precision seems limited to 4 significant figures, not so useful when you move above 10 MHz:
The “Spectrum Analyzer Pro” application seems pretty basic and I couldn’t figure out how to set the span to a smaller range, need to do some reading I think. The trace shows a lot of noise in the lower part of the spectrum, could be local interference (I’m staring at the Red Pitaya’s own tiny switch-mode power supply…):
Winding the VFO up in frequency moves it out of the noise:
The previous version of the spectrum analyser is still available in the Application Marketplace and that seems to be more usable despite the smaller display area. It seemed quite obvious that you could zoom in by selecting the zoom control and drawing a box around the area of interest:
To finish off I’ve had a little look at the digital capture of the oscilloscope to visualise the switch bounce problem I have on the PA0KLT VFO (100nF capacitors across each switch are listed as “optional” and not included in the kit but I’m thinking they’re probably “essential” and will be fitting some this week!):
Panning left and right on a scope capture seems a little clunky (you can drag the bar/trigger point at the top of the display but the trace only updates when you release the drag), hopefully someone will improve this.
A couple of quick snaps of a PA0KLT VFO coming together on a 3D-printed test panel:
Firstly a pass from NOAA 19 around 1330h, only Channel A shows a picture:
…and then later on, around 1515h, only Channel B shows an image (and very little movement in the weather):