[hpsdr] CW operation

[Luke Steele]

Ok, thanks for all the input.
I've studied again the Flex-5000 Series Owner's Manual v 1.14.0. This has a
detailed description of PowerSDR.
Pages 90, and 101 refer to the Audio buffer and sampling rate, and DSP
settings.
In tests last night, I changed the Audio buffer size to 512, from default
2048, and left the sample rate at 96000. I find the panadapter gives me the
most useful display at 96000. 192000 I find too wide, and the display too
cramped.
The Audio buffer setting change gave me improvement on latency, and tests
on CW and phone gave quite good results. The echo effect on phone was much
reduced, to the point where I can use the Monitor audio as sidetone when I
have the headset boom mic in use in phone modes.
I only had one Q on CW last night, but found it easier.
In the DSP tab, I ended up resetting the buffers to default (Rx: 4096, Tx:
2048) and this gave me the best filter response. With the Rx buffer size
reduced, the filter performance suffered. As a spinoff benefit, when
changing mode from phone to CW and back, there wasn't the pause I was
seeing when the buffers were different sizes for different modes.

To summarise, the best effect was realised by reducing the buffer size in
then Audio setup tab.

I must dig out the oscilloscope to do some objective tests, as the
adjustments are quite subtle and not very easy to determine positively by ear.

When operating, I normally work with headphones, so need the sidetone on
the headphone jack. Any extra wires, even inside the enclosure, would add
risk of RF effects. I have a 100W PA inside the Pandora enclosure, and
sometimes run an external PA, so the EMR in the shack is significant.

73,
Luke

 1294443357.0