piHPSDR with G2 - summary of my issues

[dl4zbg]

Hello all,

here a short summary of my current issues with piHPSDR/G2 harware

1) Encoder menue no storing changes - already reported
2) After the latest update I cannot find the sample rate setting (radio menue)
3) knacks/glitches in rx audio -already reported EDIT at 96kHz sampling rate RX audio seems to be ok.
4) Pure signal is very effective bringing IM3 down to app. -55dBc. BUT tx audio is interuppted (you can hear it as a glitch in monitor audio and see it on my Siglent spectrum analyzer). Enabling/disabling PS is not working properly. Without PS two tone is clean.
5) I can define transverters, but switching to these bands does not work.

Listening to the Rx audio is exhausting. Mostly using Thetis at the moment.
73

Volker

DL4ZBG

[JohnFT61]

1. I enter in the info manually for now. Strange things happen when one uses the "Encoders" menu to do it. Left-Bottom and Top-Right are reserved when using that way. Just enter in manually for now.

I have "COMP" for Top-Right Switch, but it only works if you press that button, and then press the Switch button for Bottom-Right (but also toggles the switch button I have for Bottom-Right (ANL) at the same time.

2. The sample rate is adjusted in the "Rx" menu.


John
N2KBE

[dl4zbg]

2. The sample rate is adjusted in the "Rx" menu.


John
N2KBE

TNX. I overlooked it. Now at 96kHz sampling rate the RX audio sounds better.

Do you edit any config file for fixing the encoder assignments?

73

Volker

[JohnFT61]

I manually edited the props file in the same folder as the pihpsdr program, to setup my Encoder Switches, inner and outer dials:

The file is called "D8-3A-DD-07-E6-A7.props" on my G2. I have the files required for pihpsdr in "/home/pi".

I replaced the following lines in the file so that:

Top-Left Switch = PS On/Off
Inner Dial = Mic Gain
Outer Dial = TX Drive

Bottom-Left Switch = AF On/Off
Inner Dial = AF
Outer Dial = AGC

Top-Right Switch = COMPR On/Off (only works if you press the Switch first, and then press the Bottom-Right Switch, I hope this will be fixed!)
Inner Dial = ZOOM
Outer Dial = PAN

Bottom-Right Switch = ANF
Inner Dial = RIT
Outer Dial = XIT

*********** Replaced the following lines *************

encoders[4].switch_function=NONE
encoders[4].top_encoder_function=NONE
encoders[4].bottom_encoder_function=VFO
encoders[3].switch_function=ANF
encoders[3].top_encoder_function=RIT
encoders[3].bottom_encoder_function=XIT
encoders[2].switch_function=COMP
encoders[2].top_encoder_function=ZOOM
encoders[2].bottom_encoder_function=PAN
encoders[1].switch_function=MUTE
encoders[1].top_encoder_function=AFGAIN
encoders[1].bottom_encoder_function=AGCGain
encoders[0].switch_function=PST
encoders[0].top_encoder_function=MICGAIN
encoders[0].bottom_encoder_function=TXDRV

*************************************************

In other words:

Encoder 0 is Top-Left
Encoder 1 is Bottom-Left
Encoder 2 is Top-Right
Encoder 3 is Bottom-Right



John
N2KBE

[ct1iqi]

@dl4zbg
There is a with and without control front-panel version ('wo').
Can confirm the audio problems also for the 'wo', as elaborated but not resolved in another thread, when one tries to use the embedded pihpsdr.
( viewtopic.php?f=7&t=4792 )
An additional issue, specific to the 'wo' is that, when starting embedded pihpsdr, one first has to kill the p2app process that is running.
When both p2app and pihpsdr run on the CM4, in tx mode there is a clash between them in comms to the FPGA, that in tx mode makes the G2 do tx<->rx at roughly a 1 s. rate.
When using the 'wo' from a Linux PC via LAN, using p2app in the G2, and exactly the same pihpsdr/wdsp as present in the G2, but compiled for an amd64 PC with Debian 11, the audio is fine.
So basically the same conclusion as yours with Thetis.
p2app uses xdma for communication with the FPGA. And so does G2 embedded pihpsdr.
It is puzzling why the G2 local pihpsdr produces these little gaps and hick-ups in the audio.
It does not have to send anything time critical via ethernet, so its job supposedly would be easier.

[dl4zbg]

Thank you John,

Encoders are working now the way they should. Switches of the encoders do not work. Seems that my encoder bottom left has no switch.
Will investigate further.

Now using the latest build 2 08 from 9/16

73

Volker
DL4ZBG

[JohnFT61]

Make sure you feel a depression on the Switches of the Encoders when you press them. If not remove them, position them correctly (there are slots that line up), and then push them back on... then try again. I had to do that with my G2.



John
N2KBE

[JohnFT61]

Build 2 08 was from a week or 2 ago, didn't help me with the Encoder issue. You see piHPSDR saturn (Protocol 2 v8) on XDMA (on the top of the pihpsdr program, right?

[w-u-2-o]

It is puzzling why the G2 local pihpsdr produces these little gaps and hick-ups in the audio.

Better to continue this discussion in the original topic (viewtopic.php?f=7&t=4792), but it's a good bet that piHPSDR running on a CM4 just can't keep up with the real-time data stream and frequently finds its audio buffers empty.

Since people report CM4 CPU utilization is not excessive, this could be just a matter of tweaking the piHPSDR code.

[dl4zbg]

It is puzzling why the G2 local pihpsdr produces these little gaps and hick-ups in the audio.

Better to continue this discussion in the original topic (viewtopic.php?f=7&t=4792), but it's a good bet that piHPSDR running on a CM4 just can't keep up with the real-time data stream and frequently finds its audio buffers empty.

Since people report CM4 CPU utilization is not excessive, this could be just a matter of tweaking the piHPSDR code.

I hope so

[ct1iqi]

on the topic of 'which thread' it would perhaps be an idea to separate issues that are specific to the G2 having a controller, and issues that all G2s have in common.

[dl4zbg]

Hello all,

here a short summary of my current issues with piHPSDR/G2 harware

1) Encoder menue no storing changes - already reported
2) After the latest update I cannot find the sample rate setting (radio menue)
3) knacks/glitches in rx audio -already reported EDIT at 96kHz sampling rate RX audio seems to be ok.
4) Pure signal is very effective bringing IM3 down to app. -55dBc. BUT tx audio is interuppted (you can hear it as a glitch in monitor audio and see it on my Siglent spectrum analyzer). Enabling/disabling PS is not working properly. Without PS two tone is clean.
5) I can define transverters, but switching to these bands does not work.

Listening to the Rx audio is exhausting. Mostly using Thetis at the moment.
73

Volker

DL4ZBG

After a couple of updates of firmware/FPGA, p2app und piHPSDR all 5 issues are history and I enjoy working with this radio, 99% using piHPSDR. Thanks to DL1YCF and N1GP for their work and support.

73

Volker