QRQ CW over 100 wpm - tips, techniques and specialized QRQcw GEAR

[Chuck aa0hw]

This video is demo’N a setup to create a custom QRQ CW sine wave, raised cosine edged, ‘keyable’ oscillator

There are already 2 excellent programs for LINUX for a software QRQ CW KEYBOARD… FLdigi and TWQRQ...however, neither one of them are ‘keyable’ - this first video in a new video series on “QRQ CW OVER 100 WPM” uses Fldigi as a tool to build a new software QRQ CW pure sine wave, raised cosine edged, ‘keyable’ oscillator...that you can adjust weight both by TIME and by PERCENTAGE…you can also choose the raised cosine ‘window’, adjust pitch, volume, Q, bandwidth etc… NOTE: at QRQ CW over 100 wpm, many of the cw element parameters become critical for good copy… for example, the edge noise, in spite of the cw element having an 8 ms rise and fall time, increases with cw speed and must be reduced...the edge noise can get so loud as to be 50 percent of what you hear… the VSTs filter it out very well...

NOTE: headphones might be needed in order to hear the subtle cw edge changes
NOTE: only one speaker output is used for the video, to prevent 2 speaker phase issues while listening....you can use a STEREO to MONO adapter if you need to hear this video on both speakers..

first a 1000 hertz tone from Fldigi’s RT QSK CONTROL RT CHANNEL audio output goes to the input of the VST HOST’s first VST PLUGIN, a bandpass filter...which is centered at 1000 hertz...the volume of its output is CRITICAL to get the weight by TIME to work in the following plugins… [ - 1.74 db in this setup ]

the bandpass filter connects to the first REAGATE VST Plugin, where again the threshold volume(-.2 db) is critical to get the weight by TIME to work on the next plugin...another REAGATE…

the 2nd REAGATE PLUGIN is used for adjusting the WEIGHT BY TIME on the CW NOTE...the threshold slider in the middle is set to clone Fldigi’s weight and if you raise the threshold slider higher...you get lighter keying...if you lower the threshold slider you get heavier keying…

the next VST PLUGIN is called “sineCW” and is the heart of the system...where the 1000 hertz tone is used to create a new “cloned” cw element..sineCW lets you adjust the CW PITCH, VOLUME, bandpass filter bandwidth and Q, and rise/fall time

the final vst plugin is called the ENGINEERs FILTER, where we remove any remaining edge noise and select the filter ‘window’ to your preference...each window sounds a bit different...and allows more or less edge noise to get through to the audio output…

a text file from TOM’s, W4BQF, article about copying CW OVER 70 wpm is used...and demoN some of plugins while sending...
https://sites.google.com/site/tomw4bqf/copyingcwover70wpm

http://qrqcwnet.ning.com/forum/topics/home-brewed-software-code-pra...
https://www.image-line.com/support/FLHelp/html/plugins/Minihost%20M...
https://www.meldaproduction.com/MFreeFXBundle
https://www.reaper.fm/reaplugs/
http://www.rs-met.com/freebies.html

[Chuck aa0hw]

Here is another example of QSO'n  QRQ CW Over 100 using the latest STUDIO LINK STANDALONE APP

STUDIO LINK

https://doku.studio-link.de/standalone/installation-standalone.html

playlist here:

https://www.youtube.com/playlist?list=PLfFan4sDonOfYgekqoqcLClSBF1q...

[Chuck aa0hw]

another use case example for QRQ CW over 100 wpm -  CW DATA over IP demo

USES a CW KEYBOARD/KEYER APP from Recri Keyer: https://github.com/recri/keyer

that can send CW at "ZERO HERTZ" (0, 1) ON/OFF(keydown/keyup) data stream -

uses TRIGGER MIDI MONO https://lsp-plug.in/?page=manuals&section=trigger_midi_mono

to convert the CW DATA into MIDI NOTE ON/off messages

which "keys" a sine wave CW Oscillator & CW BANDPASS FILTER modules from Recri Keyer: https://github.com/recri/keyer

CW data OVER ip - Transmitter & Receiver - by Gstreamer (EAC3 codec) https://gstreamer.freedesktop.org/

[Chuck aa0hw]

SEND cw data over ip instead of CW AUDIO over ip - use a soundcard SDR mixer module to convert back to CW AUDIO

this is a follow up to the previous video on this topic starting here: https://youtu.be/hyGMRDvNLsU

in this video the DECODING of the CW DATA into CW AUDIO NOTEs is done with a soundcard SDR program from RECRI KEYER https://github.com/recri/keyer ...using its LO-MIXER to mix/convert the CW DATA received over ip from 0's &1's to CW AUDIO NOTEs with a frequency that you choose on the Recri Keyer LO-MIXER module

2 post Recri Keyer CW audio bandpass filter modules are used to clean up/clean out... the CW ELEMENT audio edge noise coming from the rapid start/stop/make/break/keydown/keyup etc....

THE PINK CW NOTE on the SCOPE VIEW shows the final wave shape of the decoded/demodulated CW from the CW DATA Over IP transmitter source the Gstreamer audio codec = eac3 is used to provide the pathway over ip from the transmitter location to the receiver location

here is an example of the TRANSMITTER GSTREAMER SCRIPT used: gst-launch-1.0 -v jackaudiosrc ! audioconvert ! queue ! avenc_eac3 ! rtpgstpay ! udpsink host=(ip address of the cw receiver) port=7777

here is an example of the RECEIVER GSTREAMER SCRIPT used: gst-launch-1.0 -v udpsrc port=7777 ! "application/x-rtp, media=(string)application, clock-rate=(int)90000, encoding-name=(string)X-GST, caps=(string)\"YXVkaW8veC1lYWMzLCBjaGFubmVscz0oaW50KTEsIHJhdGU9KGludCk0ODAwMA\=\=\", capsversion=(string)0, payload=(int)96" ! rtpjitterbuffer ! rtpgstdepay ! avdec_eac3 ! audioconvert ! jackaudiosink

[Chuck aa0hw]

send QRQ CW DATA OVER IP and convert back to CW AUDIO using a soundcard SDR MIXER MODULE

Gstreamer Wavpack Codec is tested to see how well it can send QRQ CW DATA over IP at speeds over 100 wpm...and how well a sound card SDR MIXER module can convert the CW DATA back to CW AUDIO tones

SDR mixer & filter modules are from:       https://github.com/recri/keyer

WAVPACK CODEC udp send / udp receive scripts are from

Gstreamer: https://gstreamer.freedesktop.org/download/

TRANSMIT WAVPACK CW DATA OVER IP script:

gst-launch-1.0 -v autoaudiosrc ! queue ! "audio/x-raw, channels=1" ! audioconvert ! wavpackenc perfect-timestamp=true bitrate=64000 ! wavpackparse ! rtpgstpay ! queue ! udpsink host=(ip address of wavpack cw data over ip receiver LOCATION) port=2345

RECEIVE WAVPACK CW DATA OVER IP script:

gst-launch-1.0 -v udpsrc port=2345 ! "application/x-rtp, media=(string)application, clock-rate=(int)90000, encoding-name=(string)X-GST, caps=(string)\"YXVkaW8veC13YXZwYWNrLCBjaGFubmVscz0oaW50KTEsIHJhdGU9KGludCk0ODAwMCwgZGVwdGg9KGludCkzMiwgZnJhbWVkPShib29sZWFuKXRydWU\=\", capsversion=(string)0, payload=(int)96" ! rtpjitterbuffer ! rtpgstdepay ! wavpackparse ! wavpackdec ! audioconvert ! autoaudiosink

[Chuck aa0hw]

Here is an example of QRQ CW over 100 wpm that is bandwidth conserving by using a modified A2 Mode of QRQ CW Keying

the modified A2 mode of sending QRQ CW, that is demo'D in this video, that has such tiny spectrum bandwidth when compared to A1 mode QRQ CW keying, is using micro-modulation of a 2Khz VLF RF CARRIER to encode CW KEYING data onto a 2Khz VLF RF carrier, done by FLdigi & Zamgate, and then decoded at the RECEIVE site by another ZAMGATE & sine wave oscillator

see previous videos on the setup info for encoding & decoding micro-modulation of CW KEYING DATA onto a VLF CARRIER

https://youtu.be/FtCCLDYwRTQ

https://youtu.be/wVHZ6H81okI

https://youtu.be/yh7Wx5WA2M8

[Chuck aa0hw]

Here is another example of CW OVER 100 WPM -  using AM modulated CW audio over IP

SETUP

2 ops have their own LPAM Transmitter's running with the output of their PC SOUND CARD's LINE OUTPUT or HEADPHONE OUTPUT going to the LPAM's modulation audio input port... both OP's also have their own AM RECEIVER's to pick up the LPAM Transmitter's CW SIGNAL on an agreed upon frequency

both OPs send a nice sounding CW sidetone AUDIO OVER IP to the other OP

using one of the many choices for AUDIO OVER IP apps out there

ie SONOBUS

https://sonobus.net/

[Chuck aa0hw]

QRQ CW over 100 WPM - FLdigi keying a TWIN-T oscillator - sending CW over SonBus - GATE & EQ example

https://sonobus.net/

[Chuck aa0hw]

RAISED COSINE GENERATOR on a Pi4: is it FAST enough to send perfectly formed & timed CW over 100 wpm

A continuation of using CARLA as a plugin host for QRQ CW operations this time using JACKTRIP as the APP to send CW audio over iP

https://jacktrip.github.io/jacktrip/

https://github.com/jacktrip/jacktrip/releases

https://help.jacktrip.org/hc/en-us/articles/1500009727561-Build-a-R...

See previous videos on this pi4 setup for more info:

https://youtu.be/tdo36XFHxGM

https://youtu.be/EnN6M9XbySI

[Chuck aa0hw]

Asymmetrical Exponential shaped CW at QRQq(101 wpm) \\Raspberry Pi4// CW Keyer - LIVE demo

Raspberry pi4 CW KEYER apps create the Asymmetrical Exponential shaped CW NOTEs see these 2 video demos for more info on how the PI operates as a CW (RE)KEYER (REgenerator)

https://youtu.be/HRhqweSciu0

https://youtu.be/T51cr5l-_a0

the processing time for the pi4 to RE-Generate the TLC555's audio INPUT into a MIDI KEYING /CW KEYER/ new ASYMMETRICAL EXPONENTIAL shaped CW AUDIO output, is only 4 -5 milliseconds(depends on the bandpass FILTER 'LENGTH' setting in the CW REGEN APPs setup)

[Chuck aa0hw]

UNiXcw & the FASTLOOP alsa loopbacks for a QRQ CW copy by ear practice session from 100wpm to 135wpm

unixCW is sending CW at 100Khz, 105Khz, 110Khz, 115Khz, 120Khz, 125Khz, 130Khz & 135Khz at 100wpm, 105wpm, 110wpm, 115wpm, 120wpm, 125wpm, & 135wpm

respectively ....

in order to be able to do that,  you have to modify the source code files before compiling https://unixcw.sourceforge.net/

- to allow a sample rate over 48Khz

- to allow a cw speed over 60wpm

- to allow a CW PITCH/FREQUENCY over 4Khz

the FASTLOOP alsa loopbacks, which can operate at a sample rate up to 768Khz, was compiled easily by following the instructions here: (note i am using ubuntu 18.04...and the DEV is not sure how it will work on other OSes, distros ) https://www.sdrutah.org/info/high_rate_loopback_websdr.html

to operate a laptop sound card above its native limit, the JACK AUDIO CONNECTION KIT 'dummy' virtual sound card was used... the sample rate was set to 384Khz to allow operations at 100Khz - 135Khz using the 8 sub-channels of the FASTLOOP alsa loopback cable, 8 CW sends were setup on unixCW to send CW out at 100Khz to 135Khz at 100wpm to 135 wpm to capture every QRQ CW signal from those fastloop alsa loopback cables, Gstreamer pipelines were used to capture the signals on each FASTLOOP alsa loopback cable and bring them all into the jack audio connection bay... then the RECRI KEYER's SDR mixer and filter module (& hilbert cell module) were used to isolate each signal, demod /downconvert it to CW AUDIO and send it to the laptop's native sound card, REALTEK HD