First of all, thanks for all the time you are putting in.
See suggestion for PCM, LO Drive gain below.
Tried to take a scientific/stepwise approach with setting the DRAWS board
audio out today.
I went through the entire level and deviation check routine again today and
wound up with very similar settings to my previous attempts. I then tested
direwolf VHF-FM first and then after disabling AX25, fldigi and wsjt-x on
HF. Results "ain't purty", but at least everything is being consistent.
Software function status:
VHF - with direwolf/Xastir transmitting and receiving on the local packet
network does work .
Comment: I don't really have a an easy/better way to test than function on
APRS network locally. I have a sense that the audio may be high, but I won't
know until I have time to putter with Spectrum Lab. It will become my
"Deviation Meter" with an IC-7100 functioning as my "sensor".
Check this out by John Ackermann N8UR
Look for: "The No-Test-Equipment Packet Adjustment System"https://www.febo.com/packet/layer-one/transmit.html
Do you have a handy talky?
use the measure_deviate script: n7nix/deviation/measure_deviate.sh
Pick a tone, bring-up alsamixer Set PCM to 0db & drop LO Drive until there is no
clipping ie. no distortion or just follow John Ackermann's suggestion.
UHF - have not tried anything yet. TBD
HF - With the alsamixer settings below (visually they produce the "lowest
splatter" per an IC-7300 panscope and using Spectrum Lab 2 with the IC-7300
internal audio subsystem via USB. Rig coax is grounded, RF gain at 20% with
20dB of attenuation:
1. Fldigi drives the IC-7000 up to full power (if set) but transmits a very
dirty sig at any power above 5%.
Comment: The sig can be copied by others, but the ALC bar is reading 60%.
Not making any friends in the PSK world. These are ALC symptoms. Reducing
either the PCM level or LO Driver Gain results in very much increased noise
and a feedback like whistle at a different freq. Increasing the two settings
results in more splatter and harmonics.
2. WSJT-x receives just fine but DRAWS produces only an extremely low
transmit audio level according to the IC-7000 and signal being received on
other transceivers. Only testing with FT8 no other JT modes tested. TBD:
Comment: In transmit there is no Power Out deflection on the IC-7000 - even
if I adjust the PCM and IO Driver levels to the max, together or separate.
So the PCM/LO Driver Gain level settings that "work" in fldigi, don't work
in wsjt-x. I even went back to check that invoking wsjt-x did not somehow
tweak the alsamixer settings. they remained the same.
Key Question: Why is DRAWS overdriving in fldigi, and then not putting out
enough audio level to drive the IC-7000 to set power level in wjst-x FT8
mode with alsamixer settings remaining the same between apps?
I don't have an answer but your description is a good data point.
Here are my current alsamixer settings:
PCM L:[-21.50dB], R:[-21.50dB
ADC Level L:[-2.00dB], R:[-2.00dB]
LO Driver Gain L:[6.00dB], R:[6.00dB]
IN1 L:[10kOhm], R:[10kOhm]
IN2 L:[Off], R:[Off]
So PCM set that low is very suspicious. It means you are not using many
bits of the CODEC. I would play with returning PCM at least close to
zero & dropping LO Driver Gain down.
I am stuck until I can figure out how to coerce wsjt-x into transmitting at
set power with the same audio input levels to the rig as provided in fldigi.
At this point I don't even care if the overdrive issue isn't solved. I have
decided to put the FT-817 aside because other are working/reporting on it. I
am only testing with the Beta 5 provided software and the IC-7000 until I
get these critical pieces working reliably because that will be my main
radio for this DRAWS kit.
Now need to go pay attention to other stuff (like family and the local zoo
of 4-legged furries we keep here on the farm).
Yep. I understand.
Thanks for thinking this through with me.
Thanks for taking such a methodical approach.