Remote/virtualized CW: running CW by connecting a key to your PC and listening to audio on your PC via VAC.
It has long been believed that remote/virtualized CW is not possible due to high signal latencies in a virtualized environment. However, this is not strictly true any longer. Remote CW with PowerSDR and Thetis ranges from "merely doable" to "provides perfect QSK performance", depending upon your operating environment.
Hardware requirements:
- Key or paddle
- 9 pin serial port (USB, built-in or PCIE card)
- Cable between key or paddle and the serial port
- Speakers connected to your PC
- Code oscillator (optional and/or not required--see below)
- ASIO-capable sound interface (highly desirable, required for QSK with Protocol 2)
Connecting the key or paddle:
Method 1, paddle/iambic key,
- In Setup > DSP > CW set Primary to COMx and Secondary to None.
- Attach the DIT connection to the DSR input on pin 6
- Attach the DAH connection to the CTS input on pin 8
- You can use usually use the RTS output on pin 7 as the common connection, the DTR output on pin 4 can also be used.
Method 2, straight key or external keyer output,
- In Setup > DSP > CW set Secondary to COMx and Primary to Radio or None
- In Setup > DSP > CW set PTT Line to None and Key Line to RTS
- Attach the key across the RTS output on pin 7 and the CTS input on pin 8
- (Alternate arrangement: set Key Line to DTR, attach key across DTR output on pin 4 and DSR input on pin 6)
Connecting speakers:
See my VAC tutorial here: https://apache-labs.com/community/viewtopic.php?f=13&t=2655
For optimum performance you want to use an audio interface that supports the high-speed ASIO audio driver. A very good choice is the Behringer UMC202HD, which seems to have the most stable drivers and is at a compelling price point, but there are many other choices. You must spend quality time tweaking your buffer and buffer latency settings to the minimum possible on your system. With Protocol 2 and ASIO always set the PortAudio settings to zero. You also must use either the low latency filter setting (any size) or a linear phase filter of no larger than 1024.
For QSK under Protocol 2 an ASIO interface is mandatory. Normal Windows sound drivers are just too slow.
Configurations vs. available perfomance (PowerSDR implies Protocol 1 firmware, Thetis implies Protocol 2 firmware):
PowerSDR with normal Windows sound drivers--you can send and receive, but latency is pretty bad. No QSK. There will be no sidetone unless you attach a code oscillator right to the key or paddle.
PowerSDR with ASIO sound drivers and interface--with properly adjusted driver and VAC settings, perfect performance. No QSK. Again, if you want sidetone you have to generate it locally yourself the old fashioned way, with a separate code oscillator.
Thetis with the new QSK feature turned off and normal Windows sound drivers--same as for PowerSDR above.
The following configurations require ANAN-200D, 7000 or 8000 series radio hardware.
Thetis with the new QSK feature turned on and Windows sound drivers--too frustrating to use due to latency, not recommended.
Thetis with the new QSK feature turned on with ASIO sound drivers and interface--with a properly adjusted driver and VAC settings, perfect QSK performance. And perfect sidetone is obtained from your own signal coming back into the radio full duplex.
Thanks to recent work by W1PA (software) and N1GP (firmware), this last configuration is really a stellar performer. Set up properly you will enjoy using it quite a bit!
...
Remote (virtualized) CW
Re: Remote (virtualized) CW
Thetis with the new QSK feature turned on with ASIO sound drivers and interface--with a properly adjusted driver and VAC settings, perfect QSK performance. And perfect sidetone is obtained from your own signal coming back into the radio full duplex.
This is very useful information for somebody new to the Anan/Thetis and who mainly operates in CW.
Is it really true that you hear the CW sidetone with a capable sound card (cmASIO setup)?
So far I played with my computer's internal sound card (Realtek) and the ASIO4ALL driver. I get a green indication from cmASIO but beside a terrible sound (changing block numbers did not help) there is no CW sidetone. The sidetone is only available at the audio outlet of the G2 and l assume it is not being streamed. Maybe it is just related to the limited performance of the Realtek soundcard and I should invest in Behringer external device but I guess the sidetone is only available when using DUP and MON together with QSK which would be a No-Go for me as it would heavily stress the relais of my amp.
It might be anyway a good idea to go with an external sound card as I have all sort of issues with the Realtek (VAC 2 not working at all, only acceptable quality with windows drivers on VAC 1, poor quality together with the ASIO4ALL driver,...). However, I assume this will not solve the CW sidetone. Has anybody played and successfully setup a WINKEY emulation app between N1MM+ and Thetis? There is an app called CW Sidetone but I still have not figured out the correct settings and get it working. Is there somebody who has solved this?
Cheers,
Michael
Re: Remote (virtualized) CW
Sidetone via Thetis/PC audio is only available with QSK. If you don't want to run full-break-in QSK because of relays then no sidetone via that path. This is because your "sidetone" is really your signal going out over the air that you are listening to in full duplex TX/RX mode.
ASIO4ALL is not a true ASIO device. It is WDM audio in a fake ASIO wrapper. It generally does not work well and it certainly does not deliver the speed of a real ASIO driver. If you do wish to use cmASIO it is best to get a sound card or sound interface that has a native ASIO driver.
This topic is quite old. FWIW, now that I've started to dabble in CW myself, I've found the QSK sidetone mechanism to have too much latency. My solution is to use an external keyer with built-in sidetone generator. I bought an OpenCWKeyer MK2. It is possibly the perfect keyer for ANAN/Thetis users as, like our radios and software, it is completely open source. The hardware design is on Github, and it runs the impressive K3NG software, also on Git.
I quickly tricked out my keyer by upgrading the Arduino Nano it came with to the more powerful Arduino Nano Every. The Every has enough memory to add the display, USB serial control, and Winkeyer functions. I also added a cute little 8x2 OLED display.
ASIO4ALL is not a true ASIO device. It is WDM audio in a fake ASIO wrapper. It generally does not work well and it certainly does not deliver the speed of a real ASIO driver. If you do wish to use cmASIO it is best to get a sound card or sound interface that has a native ASIO driver.
This topic is quite old. FWIW, now that I've started to dabble in CW myself, I've found the QSK sidetone mechanism to have too much latency. My solution is to use an external keyer with built-in sidetone generator. I bought an OpenCWKeyer MK2. It is possibly the perfect keyer for ANAN/Thetis users as, like our radios and software, it is completely open source. The hardware design is on Github, and it runs the impressive K3NG software, also on Git.
I quickly tricked out my keyer by upgrading the Arduino Nano it came with to the more powerful Arduino Nano Every. The Every has enough memory to add the display, USB serial control, and Winkeyer functions. I also added a cute little 8x2 OLED display.
Re: Remote (virtualized) CW
Thanks, Scott - yes, this or another keyer would do . I assume then there is no need for a native ASIO device anymore as a latency of 100ms or even more is less critical when replying in CW as long as the sidetone is in sync with your paddle movements.
I am still looking for a soft solution like the CW Sidetone app which is in between Thetis and a software supporting winkeyer as e.g. N1MM+. Maybe somebody has a similar setup.
73, Michael
I am still looking for a soft solution like the CW Sidetone app which is in between Thetis and a software supporting winkeyer as e.g. N1MM+. Maybe somebody has a similar setup.
73, Michael
Re: Remote (virtualized) CW
Scott,
as for some reasons the CW sidetone app is causing a lot of issues with the different virtual cables, I am inclined to buy either a OpenCW Keyer, or a Nanokeyer or a Winkeyer. However, I still have a question when operating with a headset. I assume that the sidetone is only available via a internal speaker of the keyer or is there a way to feed the sidetone to the VAC?
73, Michael
as for some reasons the CW sidetone app is causing a lot of issues with the different virtual cables, I am inclined to buy either a OpenCW Keyer, or a Nanokeyer or a Winkeyer. However, I still have a question when operating with a headset. I assume that the sidetone is only available via a internal speaker of the keyer or is there a way to feed the sidetone to the VAC?
73, Michael
Re: Remote (virtualized) CW
I would suggest modifying what ever keyer you decide upon with an audio out jack. Then obtain an inexpensive two-input headphone mixer from Amazon (or wherever). Use the headphone output of the Realtek. Combine the two audio sources in the mixer. Cheap, simple, and no latency.
Re: Remote (virtualized) CW
Thanks - this sounds like a straight forward, simple solution.
Re: Remote (virtualized) CW
I played a bit with a WK USB Keyer provided to me by a friend for testing the device with the G2 but I am not sure if I miss something.
I am using the keyer with N1MM - setting N1MM to the WK port and selecting winkeyer. Thetis is CAT controlling N1MM. When using N1MM (e.g. crtl+k and writing some text), I hear the CW text in the WK and MOX is engaged but no RF is being sent. RF is only transmitted when connecting a cable between the WK output port (e.g key 1) to the input key at the G2. I assumed that the usb connection between WK and the computer is providing the keying via CAT control to the radio but I am apparently wrong, or have I missed a setting?
I also hear a strange sound - like a double relais switching - when using N1MM for keying (eg. free text or the function buttons) whereas when keying with the paddles, I can hear only one relais switching. Is this normal or related to a wrong setting (e.g. engaging the PTT by CAT control and by the keyer input with some time difference due to the latency?)
In case the keying cable is needed then the WK is not really a solution (besides a cleaner CW timing) as I can directly use a virtual com port, assign it in N1MM to DTR=CW and within Thetis in DSP-CW as a secondary connection. The paddles could then be connected directly into the radio.
73, Michael
I am using the keyer with N1MM - setting N1MM to the WK port and selecting winkeyer. Thetis is CAT controlling N1MM. When using N1MM (e.g. crtl+k and writing some text), I hear the CW text in the WK and MOX is engaged but no RF is being sent. RF is only transmitted when connecting a cable between the WK output port (e.g key 1) to the input key at the G2. I assumed that the usb connection between WK and the computer is providing the keying via CAT control to the radio but I am apparently wrong, or have I missed a setting?
I also hear a strange sound - like a double relais switching - when using N1MM for keying (eg. free text or the function buttons) whereas when keying with the paddles, I can hear only one relais switching. Is this normal or related to a wrong setting (e.g. engaging the PTT by CAT control and by the keyer input with some time difference due to the latency?)
In case the keying cable is needed then the WK is not really a solution (besides a cleaner CW timing) as I can directly use a virtual com port, assign it in N1MM to DTR=CW and within Thetis in DSP-CW as a secondary connection. The paddles could then be connected directly into the radio.
73, Michael
Re: Remote (virtualized) CW
Originally your question was about obtaining good, local sidetone with the ANAN (assumed by me to be) distant from the operating location.
Now it's about using a Winkeyer with N1MM and the ANAN is at the operating location. Two different subjects.
IMHO, using Winkeyer with N1MM is redundant. N1MM directly to the Thetis is a lot simpler. Some folks have reported problems with N1MM direct, but I suspect these are folks who want to send at very high speeds.
Now it's about using a Winkeyer with N1MM and the ANAN is at the operating location. Two different subjects.
IMHO, using Winkeyer with N1MM is redundant. N1MM directly to the Thetis is a lot simpler. Some folks have reported problems with N1MM direct, but I suspect these are folks who want to send at very high speeds.
Re: Remote (virtualized) CW
Hi Scott,
no - you are right - this is what I am trying to achieve. Working cw from a remote location within the same network and having a sidetone available but I fail to understand how this could be achieved with the WK as it needs to be connected to the radio via the output port otherwise there is no RF output. Could you please outline how it should be connected and which settings are mandatory?
Best regards,
Michael
no - you are right - this is what I am trying to achieve. Working cw from a remote location within the same network and having a sidetone available but I fail to understand how this could be achieved with the WK as it needs to be connected to the radio via the output port otherwise there is no RF output. Could you please outline how it should be connected and which settings are mandatory?
Best regards,
Michael