Re: It was "fine", then... strange behavior #draws #setup #betaimage

Jack Spitznagel

Hi Basil,

I was able to play with settings again today after I finished putting
together the DRAWS/SweetPi console. Spoiler alert - I got back to
function... eventually.) I tried messing with the audio level settings again
- going back and forth between rigs and software. So... please get a beer
(or whatever your favorite libation is) and try to be amused with my
continuing fumbles and the eventual solution.

1. I shut down direwolf, tried your new scripts (in message #2327), and it
did nothing for the lack transmit sound levels for fldigi and wsjt-x, but
also put me back to where the direwolf levels were way too high for the
IC-7000. I reset them to:

PCM L:[-31.00dB], R:[-31.00dB]
ADC Level L:[-2.50dB], R:[0.00dB]
LO Driver Gain L:[0.00dB], R:[11.00dB]
IN1 L:[Off], R:[Off]
IN1 L:[10kOhm], R:[10kOhm]

... and direwolf now works again with good transmit audio. So in desperation
I hooked up the FT-817 to repeat the test, but of course had to switch back
to keying with GPIO because I don’t have a working CAT control setup just
yet. I had been trying to key the IC-7000 with flrig/C-IV/CAT alone! With
GPIO 12 enabled, the 817 transmitted the very noisy audio! Hmmm. I went back
to the IC-7000 left it keyed by GPIO and it transmitted dirty audio.

So here is a "key" learning: *Keying with flrig C-IV/CAT alone does not
work... the card apparently does not send audio unless a GPIO port is
enabled.* <headsmack>

I don't remember seeing that anywhere in the UDRC/DRAWS discussions. The
fldigi docs were not helpful... implied keying could be done with flrig.

So now I am where I can try the deviation/level settings for fldigi/wsjt-x
again... without munging up direwolf.

2. Using fldigi, I adjust PCM levels up to:
PCM L:[-21.50dB], R:[-21.50dB]< the last increment to here
clears the audio!*
ADC Level L:[-2.50dB], R:[0.00dB]
LO Driver Gain L:[2.00dB], R:[11.00dB]
IN1 L:[Off], R:[Off]
IN1 L:[10kOhm], R:[10kOhm]
* this is a familiar #! John put it out with some of his early configs.
Had a nice QSO with KD8MDJ who heard me trying an on air test... looks like
my IMD and S/N ratio are acceptable even though the ALC reading is very
high... I could not see splatter on the pan scope.

3. Now the acid test... did these setting mess up direwolf? Re-enable
ax25/direwolf and...
NO - Problem from Wednesday solved. Put tape on the settings.

So here is the ask:
*I still want to know if there is a way to reduce audio drive so the ALC
reading on the radio drops to near 0 as recommended for digital operations
without the audio getting noisy.*

Is there a way to leave the sound where it is clean? Reducing the LO Drive
Gain or the PCM level immediately puts me back in the high noise plus signal
state. I don’t understand the other ADC settings well enough (yet) to make
an educated guess if dialing in some resistance network in the audio output
is possible? Enlightenment please.

Thanks for your help and your patience.

Jack - KD4IZ

-----Original Message-----
From: <> On
Behalf Of Jack Spitznagel
Sent: Wednesday, December 19, 2018 14:37
Subject: Re: [udrc] It was "fine", then... strange behavior #betaimage
#setup #draws


Thanks for the quick response.

I understand setting of PCM and LO Driver Gain. If I started with PCM at
zero, I could not turn LO Drive down far enough to prevent the IC-7000 doing
the "protect -self" routine and shut down transmit. Had to turn it way down
from there to have the stupid little beast transmit anything particularly a
noisy but decodable PSK31 sig!

Quick follow up with more info. At those drive levels below, direwolf is
still behaving fine - the audio is loud and clean enough that my D74A and a
creaky old AEA PK232 running as a KISS TNC can both handle it.

It is with fldigi direct to sound that is having the noise and modulation
problems? Confusing for me, but might tell you something. I am not aware of
"transmit power" adjust in fldigi like there is in wsjt-x. It might be
something the program is doing? I have used fldigi a bit, but usually in
"winders 10" prefer other digital programs.


-----Original Message-----
From: <> On
Behalf Of Jack Spitznagel
Sent: Wednesday, December 19, 2018 13:52
Subject: Re: [udrc] It was "fine", then... strange behavior #betaimage
#setup #draws


That test got me part of the way there. DRAWS was apparently overdriving the
audio front end of the IC-7000, hence the flashing TX led. I did not know it
did that! The ** output is at the end of the email.
Initially ** said:
PCM L:[0.00dB], R:[0.00dB]
ADC Level L:[-2.00dB], R:[0.50dB]
LO Driver Gain L:[0.00dB], R:[11.00dB]

Using the ** script, I was able to turn the levels back to
a point that the tone was "fairly" clean. The AGC indication was still very
high with ** now saying:
PCM L:[-31.00dB], R:[-31.00dB]
ADC Level L:[-2.00dB], R:[0.50dB]
LO Driver Gain L:[6.00dB], R:[6.00dB]

When I turned PCM back any more than that, the background noise (frying
bacon) being transmitted increased and quickly overwhelmed the tone level.
That was the ALC pumping I was seeing.

Observation: As it is set now, the circuit noise level is extremely high.
Apparently the IC-7000 input is pretty sensitive and/or badly impedance

Looks like I will need to build some kind of attenuation if the DRAWS board
does not have the ability to dial it in? (I have not gotten that far in the
documentation yet)

Unfortunately, I will have to put off any additional tweaking until Friday.

Jack - KD4IZ

I did not see anything grossly out-of-line in the ** output
pi@kd4iz_draws:/bin $
==== Sound Card ====
udrc card number line: card 0: udrc [udrc], device 0: Universal Digital
Radio Controller tlv320aic32x4-hifi-0 [] udrc is sound card #0 ==== Pi Ver
==== Pi 3 Model B+ Mfg by Sony UK Has WiFi ==== udrc Ver ==== Found a

Name: hat
Product: Digital Radio Amateur Work Station
Product ID: 0x0004
Product ver: 0x0204
UUID: 78feb572-30c1-420b-8596-f37186c9cb16
Vendor: NW Digital Radio

==== sys Ver ====
----- /proc/version
Linux version 4.14.79-v7+ (dc4@dc4-XPS13-9333) (gcc version 4.9.3
(crosstool-NG crosstool-ng-1.22.0-88-g8460611)) #1159 SMP Sun Nov 4 17:50:20
GMT 2018

----- /etc/*version: 9.6

----- /etc/*release
PRETTY_NAME="Raspbian GNU/Linux 9 (stretch)"
NAME="Raspbian GNU/Linux"
VERSION="9 (stretch)"

----- lsb_release
No LSB modules are available.
Distributor ID: Raspbian
Description: Raspbian GNU/Linux 9.6 (stretch)
Release: 9.6
Codename: stretch

---- systemd
Static hostname: kd4iz_draws
Icon name: computer
Machine ID: 2f51e140e908474eaa86ae805f03a6d8
Boot ID: b9e8d2eefb9f401a8884c0d1f9d7c2d3
Operating System: Raspbian GNU/Linux 9 (stretch)
Kernel: Linux 4.14.79-v7+
Architecture: arm

---- modules
tlv320aic32x4_i2c 16384 1
tlv320aic32x4 32768 1 tlv320aic32x4_i2c
udrc 16384 3
snd_soc_core 188416 3 tlv320aic32x4,snd_soc_bcm2835_i2s,udrc
snd_pcm 98304 6

-rw-r--r-- 1 root 6388 Dec 8 19:59
-rw-r--r-- 1 root 38156 Dec 8 19:59
-rw-r--r-- 1 root 6408 Dec 8 19:59
-rw-r--r-- 1 root 8332 Dec 8 19:59

---- kernel
||/ Name Version Architecture Description
ii raspberrypi-kernel 1.20181112-1 armhf Raspberry Pi bootloader
ii udrc-dkms 1.0.4 armhf DKMS source for the UDRC

---- compass
Compass preference file not found: /etc/apt/preferences.d/compass
---- compass apt sources list file
deb [arch=armhf,amd64] cedar main
---- compass package files
-rw-r--r-- 1 root 2201 Nov 25 23:31
-rw-r--r-- 1 root 17567 Jul 2 22:52
-rw-r--r-- 1 root 43315 Nov 25 23:31

----- Dire Wolf DEVELOPMENT version 1.6 A (Dec 9 2018)

==== boot config ====

# Uncomment this to enable the lirc-rpi module #dtoverlay=lirc-rpi

# Additional overlays and parameters are documented /boot/overlays/README

# Enable audio (loads snd_bcm2835)
# dtparam=audio=on


---- gpsd
● gpsd.service - GPS (Global Positioning System) Daemon
Loaded: loaded (/lib/systemd/system/gpsd.service; indirect; vendor
preset: en
Active: active (running) since Fri 2018-12-14 20:49:52 EST; 4 days ago
Main PID: 4274 (gpsd)
CGroup: /system.slice/gpsd.service
└─4274 /usr/sbin/gpsd -N -n /dev/ttySC0 /dev/pps0

Dec 14 20:49:52 kd4iz_draws systemd[1]: Started GPS (Global Positioning
lines 1-8/8 (END)

-----Original Message-----
From: <> On
Behalf Of Basil Gunn
Sent: Wednesday, December 19, 2018 12:13
Subject: Re: [udrc] It was "fine", then... strange behavior #betaimage
#setup #draws

In general for debugging problems:

Send output of:

Use the following instructions for to verify PTT &
transmit paths.

Using another radio listen to the tone produced by the

The most common problem, if the tones do not sound right, is the output is
being over driven causing it to clip.

Receive path:
Tune a radio to the APRS 1200 baud freq. 144.390 MHz run the following in a

tail -f /var/log/direwolf/direwolf.log

You should see received packets scrolling by.

Let us know what you find & thanks for testing the DRAWS hat.


Jack Spitznagel <@flyingfrawg> writes:

Basil, John, All,

Initially, Beta 5 seemed to work out of the box but I ran into some
DRAWS audio and keying strangeness that is baffling me after 3 tries
at installing from newly flashed microSD cards and following Basil's
latest procedure on GitHub. Hope you can provide insight. Rig is a CAT
controlled IC-7000.

Tested Direwolf with Xastir and YAAC with both RX and TX successfully,
noticed the audio was a bit hot - signal here looked normal on the SpecLab
so I left it alone.

Then disabled AX.25/direwolf (script) and configured fldigi and wsjt-x.
Both programs worked in receive but the TX audio to the IC-7000 was very
"hot". Put fldigi in tune then to bring the ALC reading as close to zero as
possible by reducing audio levels with the PCM drive control. As audio level
is reduced - the ALC starts to "pump", as did the power out indication -
unexpectedly, there was a increase of noise on audio and harmonic spikes got
higher as level of signal was reduced. Returned levels to where they had
been and meter indications returned to where they were before. Was able to
have a PSK31 QSO and got no comment about the signal quality.

I re-enabled AX.25/Direwolf using the script. Now both Xastir and YAAC
longer would transmit audio. When the rig is keyed, the TX LED on the rig
flashes 3-4 times quickly and there is no modulation of the FM carrier.
Reboot did not change behavior. Going back to non-AX25 mode, now fldigi
would key the radio, but no modulation occurred.

Re-installation of the DRAWS beta 5 from the ground up (x2) did not
help -
none of the programs transmit audio and symptoms are the same.

To be sure rig was OK, I hooked up an old SIgnaLink to a laptop, ran
and it drove and keyed the rig just fine. So I have to assume that the
problem is software or DRAWS hardware related.

Where should I be looking to fix the audio? The keying?
Settings should not so brittle. Thanks in advance,

J Spitznagel
Science River LLC

J Spitznagel
Science River LLC

J Spitznagel
Science River LLC

Join to automatically receive all group messages.