[Elecraft] DATA-A tx freq offset with digital modes

[Guy Olinger K2AV]

[In the following text, "both the K3 and the K3S" is represented by "K3/S".
Text applicable to only one or the other will be "K3" or "K3S"]

Hi Graham.

With a warm smile, a soft voice, and friendly demeanor. I have had to go
through this post-analog era "de-analoging" my own radio perceptions, and
have had a fairly painful time of it :>)

The big lesson from my de-analoging was just how much my traditional analog
thinking, ground in by 50 years of analog hamming experience on analog
equipment, was keeping me from moving on. I needed to understand, cope with
and take advantage of very superior digital methods in a new digital era. I
was really pretty clunky. That said, not reporting any genius-level
learning rate on my part to claim superiority...

There is no suppressed carrier in software derived radio (SDR). In the
 K3/S the single sideband envelope is generated directly from firmware. It
begins with data from analog to digital converter chips (ADC) on line in or
mic leads. Firmware converts this to SSB representative data, sent to a
digital to analog chip (DAC) directly outputting SSB at the 15 kHz TX IF.
There is no carrier to null or filter out. There is no opposite sideband to
phase out or filter out. There are DAC and frequency conversion artifacts
which need to be removed by one means or another (as in going through a 2.7
kHz wide 8 MHz crystal filter also used for receive).

CW is not a keyed carrier any more. It is directly generated by the DAC
chip to the 15 kHz TX IF from digital data representing an ideal waveform.
The normal keying "sidebands" of the CW signal are also from that data. The
frequency conversion, and linear amplifier stages following, will add some
additional sidebands based on their degree of linearity, hopefully very
well down.

For frequency display, in CW mode the K3/S will display the single
frequency of the central signal.

For SSB and any mode generated through DATA A, the K3/S will display the
traditional amateur radio SSB frequency, which is what you would get if you
could somehow create SSB output from a DC voltage applied to the K3/S
transmitter line in or mic inputs. Note that some other radio services use
center of bandwidth as the frequency for SSB operation.

Use of audio input for keying should be avoided where possible. Note that
you can generate CW, FSK, and PSK without use of audio input, producing
ideal keying and bandwidth. Signals generated this way are without any
audio input artifacts like 60, 120, 180 Hz hum, audio harmonics from low
level distortion, audio noise or hiss, or miscellaneous unwanted audio from
grounding issues, or feedback from mild RF in the shack, or miscellaneous
audio from other apps on the PC bleeding through PC "soundboards".

The state of digital direct signal generation for CW is so clean that it is
simply poor amateur practice to generate CW from audio tones. There are so
many ways to screw up CW in the audio. The reputation of very clean CW from
K3/S depends on the direct data generation of CW in the K3/S

The key jack input on the back of the K3/S is itself converted to data on
one lead of an ADC chip, and sent to the CPU as data, only one data
"channel" on a single multiplexed data line, informing the CPU that the key
is *reported* high or low. That key jack voltage is not used directly for
anything except to be sampled, and advise the CPU whether the jack is keyed
or not. The CPU directs other stuff based upon the *report* of keyed or
not. It can do exactly the same thing from a string of characters converted
in firmware to the *effect* of *reports* of keyed or not. Ditto FSK D and
PSK D.

In a contest, I would not want to be the next station up or down frequency
from you doing audio generation of CW. I urge you to permanently
discontinue that practice.

I would also urge that you use FSK D and PSK D where possible to transmit
instead of audio tones. These are all successfully employed by some remote
users, though I can't speak to the methods or their complexity.

FSK D and PSK D will occupy minimum bandwidth by virtue of ideal waveforms
and be independent of all the audio train issues of gain, noise,
level-based distortion and AC power artifacts between user PC generation,
audio train, and the K3/S.

Use of the K3/S USB "soundcard" in some setups does remove some audio
transmission issues, but you are exporting the "quality control" of
waveforms to the third-party program running on the PC. Third-party
programmers are not going to take responsibility for K3/S output waveshape.

Sorry to rag on, but not excessively sorry.  :>)

73 and good luck with all the remote stuff.

Guy K2AV

On Thu, Sep 22, 2016 at 1:37 AM, Graham Alston <graham at alston.com.au> wrote:

> Hi,
>
> I’m very new to Elecraft and K3S so hope this is not too obvious….
>
> I have a K3S connected to my PC with audio tones coming in via MIC IN (in
> reality the K3S is the remote end of a K3-Twin remote setup).
>
> When I use DATA-A sub mode for CW, the transmission is 600Hz higher than
> the VFO frequency. I assume this is because DATA-A transmits on the USB and
> the VFO shows the suppressed carrier frequency?
>
> I want my CW signal to match the VFO, how can I do that?
>
> 73, Graham VK3GA
>
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:Elecraft at mailman.qth.net
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
> Message delivered to k2av.guy at gmail.com