Elecraft

Technical question on keying of an SDR radio

Classic

List

Threaded

8 messages Options

Elecraft mailing list

Technical question on keying of an SDR radio

If you look at various old circuit diagrams of ancient radios made from electronic circuits you find that the telegraph keying for the radio involves switching on and off the driving oscillator frequency or something similar.

However, with an SDR, you have other options and I have no idea how it is typically done. For example, a digital signal can be produced that represents the Morse coding and this signal need merely be converted to analog and amplified. Or, is there still an analog circuit being switched on and off for Morse code (telegraphy) keying with a modern SDR.

Can someone describe how this is done in a radio like the KX2. And, is it done differently in the different Elecraft radios — I presume that the K2 would have the more traditional sort of circuit but that is really just a wild guess on my part.

73, phil, K7PEH

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

wayne burdick

Re: Technical question on keying of an SDR radio

Administrator

Hi Phil,

In the K2 the sidetone starts off as a squarewave created by the microcontroller. This is then shaped and injected into the AF amplifier. The sidetone is turned on/off at the same time as the carrier, which is generated by the LO (PLL synth) mixing with a gated TX BFO signal.

In our DSP-based radios (K3/K3S and KX2/KX3), both the audio sidetone and the 15 kHz 2nd IF carrier are created by the DSP. Their rising edges are shaped mathematically using what's called a "raised cosine" or sigmodal waveform. We experimented to find the ideal waveform equation, the result being the cleaned signal possible, with virtually no key clicks. The DSP can of course do other things like apply amplitude or frequency modulation, generate voice and data mode signals, apply ALC and audio EQ, etc. CW is just the simplest case of what can be done.

From the DSP, the digitized audio signal codes to a DAC (part of the audio CODEC IC), which then converts it to analog for injection into the AF amp.

The 15 kHz 2nd IF carrier in the K3/K3S goes to a transmit mixer on the KREF4 module where it's up-converted to about 8215 kHz. It is then routed to the first IF mixer. The 8215 kHz signal passes through two crystal filters enroute mixer. In the KX2/KX3, the 15 kHz IF signal is converted to a pair of IQ signals (90 degrees out of phase with each other) to directly modulate a quadrature up-converter. The other input to the up-converter is a pair of balanced LO signals, again separated by 90 degrees. A result of IQ modulation is that one sideband is cancelled out, resulting in a single RF carrier at the desired frequency.

In all cases, the rest of the T/R sequencing involves the usual amplifiers, PIN diodes, filter switching, etc.

73,
Wayne
N6KR

> On Apr 7, 2019, at 9:58 PM, Phil Hystad via Elecraft <[hidden email]> wrote:
>
> If you look at various old circuit diagrams of ancient radios made from electronic circuits you find that the telegraph keying for the radio involves switching on and off the driving oscillator frequency or something similar.
>
> However, with an SDR, you have other options and I have no idea how it is typically done. For example, a digital signal can be produced that represents the Morse coding and this signal need merely be converted to analog and amplified. Or, is there still an analog circuit being switched on and off for Morse code (telegraphy) keying with a modern SDR.
>
> Can someone describe how this is done in a radio like the KX2. And, is it done differently in the different Elecraft radios — I presume that the K2 would have the more traditional sort of circuit but that is really just a wild guess on my part.
>
> 73, phil, K7PEH
>
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:[hidden email]
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

wayne burdick

Re: Technical question on keying of an SDR radio

Administrator

In reply to this post by Elecraft mailing list

By the way, a similar explanation can be found in the Theory of Operation section of any of our transceiver manuals.

Wayne
N6KR

> On Apr 7, 2019, at 9:58 PM, Phil Hystad via Elecraft <[hidden email]> wrote:
>
> If you look at various old circuit diagrams of ancient radios made from electronic circuits you find that the telegraph keying for the radio involves switching on and off the driving oscillator frequency or something similar.
>
> However, with an SDR, you have other options and I have no idea how it is typically done. For example, a digital signal can be produced that represents the Morse coding and this signal need merely be converted to analog and amplified. Or, is there still an analog circuit being switched on and off for Morse code (telegraphy) keying with a modern SDR.
>
> Can someone describe how this is done in a radio like the KX2. And, is it done differently in the different Elecraft radios — I presume that the K2 would have the more traditional sort of circuit but that is really just a wild guess on my part.
>
> 73, phil, K7PEH

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

Elecraft mailing list

Re: Technical question on keying of an SDR radio

Wayne,

Thanks for the explanation. I should have realized that it might have been documented.

My question was prompted by the appearance of a question posted to the "ham radio" formum of Stack Exhange on-line. The OP (original post) of the question mentioned looking at a 1976 ARRL Handbook and seeing a circuit for the keying and wanted an explanation of how that circuit worked. The answers (at the time I read the post) were somewhat simplistic and directed to the circuit that was shown in the question which of course was technology going back to the 1940s or before.

Thus, I was curious of the improvements possible with DSP.

Thanks,
phil, K7PEH

> On Apr 7, 2019, at 10:25 PM, Wayne Burdick <[hidden email]> wrote:
>
> By the way, a similar explanation can be found in the Theory of Operation section of any of our transceiver manuals.
>
> Wayne
> N6KR
>
>
>> On Apr 7, 2019, at 9:58 PM, Phil Hystad via Elecraft <[hidden email]> wrote:
>>
>> If you look at various old circuit diagrams of ancient radios made from electronic circuits you find that the telegraph keying for the radio involves switching on and off the driving oscillator frequency or something similar.
>>
>> However, with an SDR, you have other options and I have no idea how it is typically done. For example, a digital signal can be produced that represents the Morse coding and this signal need merely be converted to analog and amplified. Or, is there still an analog circuit being switched on and off for Morse code (telegraphy) keying with a modern SDR.
>>
>> Can someone describe how this is done in a radio like the KX2. And, is it done differently in the different Elecraft radios — I presume that the K2 would have the more traditional sort of circuit but that is really just a wild guess on my part.
>>
>> 73, phil, K7PEH
>

Vic Rosenthal

Re: Technical question on keying of an SDR radio

In reply to this post by wayne burdick

Oh no! NOW you tell me that all these years that I have believed that I
have been operating CW, it's really SSB!

(Disclaimer to avoid the otherwise inevitable explanations that a pure
audio tone transmitted via SSB is identical with CW: I know that).

73,
Victor, 4X6GP
Rehovot, Israel
Formerly K2VCO
CWops no. 5
http://www.qsl.net/k2vco/

On 8 Apr 2019 08:24, Wayne Burdick wrote:

> Hi Phil,
>
> In the K2 the sidetone starts off as a squarewave created by the
> microcontroller. This is then shaped and injected into the AF
> amplifier. The sidetone is turned on/off at the same time as the
> carrier, which is generated by the LO (PLL synth) mixing with a gated
> TX BFO signal.
>
> In our DSP-based radios (K3/K3S and KX2/KX3), both the audio sidetone
> and the 15 kHz 2nd IF carrier are created by the DSP. Their rising
> edges are shaped mathematically using what's called a "raised cosine"
> or sigmodal waveform. We experimented to find the ideal waveform
> equation, the result being the cleaned signal possible, with
> virtually no key clicks. The DSP can of course do other things like
> apply amplitude or frequency modulation, generate voice and data mode
> signals, apply ALC and audio EQ, etc. CW is just the simplest case of
> what can be done.
>
> From the DSP, the digitized audio signal codes to a DAC (part of the
> audio CODEC IC), which then converts it to analog for injection into
> the AF amp.
>
> The 15 kHz 2nd IF carrier in the K3/K3S goes to a transmit mixer on
> the KREF4 module where it's up-converted to about 8215 kHz. It is
> then routed to the first IF mixer. The 8215 kHz signal passes through
> two crystal filters enroute mixer. In the KX2/KX3, the 15 kHz IF
> signal is converted to a pair of IQ signals (90 degrees out of phase
> with each other) to directly modulate a quadrature up-converter. The
> other input to the up-converter is a pair of balanced LO signals,
> again separated by 90 degrees. A result of IQ modulation is that one
> sideband is cancelled out, resulting in a single RF carrier at the
> desired frequency.
>
> In all cases, the rest of the T/R sequencing involves the usual
> amplifiers, PIN diodes, filter switching, etc.
>
> 73, Wayne N6KR

EricJ

Re: Technical question on keying of an SDR radio

Not to worry, Victor. The really important part of generating and
demodulating CW happens in your head.

Eric KE6US

On 4/9/2019 3:08 AM, Victor Rosenthal 4X6GP wrote:

> Oh no! NOW you tell me that all these years that I have believed that
> I have been operating CW, it's really SSB!
>
> (Disclaimer to avoid the otherwise inevitable explanations that a pure
> audio tone transmitted via SSB is identical with CW: I know that).
>
> 73,
> Victor, 4X6GP
> Rehovot, Israel
> Formerly K2VCO
> CWops no. 5
> http://www.qsl.net/k2vco/
>

k6dgw

Re: Technical question on keying of an SDR radio

In reply to this post by Vic Rosenthal

A technique pioneered by Art Collins many years ago in the KWM2 and S-line.

73,
Fred ["Skip"] K6DGW
Sparks NV DM09dn
Washoe County

On 4/9/2019 3:08 AM, Victor Rosenthal 4X6GP wrote:
> Oh no! NOW you tell me that all these years that I have believed that
> I have been operating CW, it's really SSB!
>
> (Disclaimer to avoid the otherwise inevitable explanations that a pure
> audio tone transmitted via SSB is identical with CW: I know that).
>
> 73,
> Victor, 4X6GP
> Rehovot, Israel

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

Bob McGraw - K4TAX

Re: Technical question on keying of an SDR radio

And some of today's software applications {HRD} have provision for
MCW. One must be careful how the interface between the computer and
radio is constructed so as not to have anything in the audio path which
would contribute to less than a pure sinewave signal. Actually this
is exactly how AFSK and other data modes work. A tone frequency is sent
to the radio to produce the MARK signal and a different frequency tone
is sent to produce a SPACE signal.

In all cases, a clean signal path between the computer and the radio is
found to be mandatory. Hum, buzz, clicks, pops and a mike left open
are all a NO NO!

73

Bob, K4TAX

On 4/9/2019 1:16 PM, Fred Jensen wrote:

> A technique pioneered by Art Collins many years ago in the KWM2 and
> S-line.
>
> 73,
> Fred ["Skip"] K6DGW
> Sparks NV DM09dn
> Washoe County
>
> On 4/9/2019 3:08 AM, Victor Rosenthal 4X6GP wrote:
>> Oh no! NOW you tell me that all these years that I have believed that
>> I have been operating CW, it's really SSB!
>>
>> (Disclaimer to avoid the otherwise inevitable explanations that a
>> pure audio tone transmitted via SSB is identical with CW: I know that).
>>
>> 73,
>> Victor, 4X6GP
>> Rehovot, Israel
>
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:[hidden email]
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
>