[Elecraft] K3 RTTY spectrum

[Joe Subich, W4TV]

 > Not as I read the article -

The article was written before the latest update to the K3 FSK
generating DSP code in MCU 4.66 released 2013-03-22.  The article
is simply out of date.

The release notes for 4.66 say:
> FSK-D KEYING BANDWIDTH GREATLY REDUCED:  This change reduces the
> amplitude of the FSK-D keying sidebands by 25-30 dB at 500 Hz
> offset, giving teh K3 one of the cleanest FSK signals available.

The author of the NCJ article reviewed the K3 firmware while it
was still in beta here: http://www.frontiernet.net/~aflowers/k3beta/
The author's own web page contains the information in the NCJ
article here: http://www.frontiernet.net/~aflowers/k3rtty/k3rtty.html
By comparing the before and after spectrogram in the K3beta article
with the "before", 2-Tone, and filtered MMTTY spectrograms on the
"AFSK Bandwidth with waveshaping" page in the original article, you
will see that the new K3 FSK code is every bit as clean as the wave-
shaped AFSK *without* the problems associated with AFSK.

Note that the author's article on the web page shows the benefits of
the K3 AFSK TX filter in terms of being able to clean up all manner
of *intentional* abuse to the AFSK audio - abuse that would render
the AFSK signals from other transceivers an absolute festival of QRM.

Here is what the author wrote of the new FSK code:
> Notice that the unshaped [original - W4TV] keying has clicks that
> will result in an S7 noise floor at 500 Hz spacing that very
> gradually flattens out to an S4 noise floor 1 KHz each side of
> transmitter, which is eventually limited (we hope!) by the IF filter
> in the transmitter. This makes it easy to see how loud signals
> placed every few kilohertz raises the noise floor of the entire band
> for everyone.  The BW-reduced [new code - W4TV] signal is under S2 at
> 500Hz and drops off very rapidly into the noise as one tunes away. It
> is below the S1 noise floor of the receiver when tuned more than 700
> Hz away.

73,

    ... Joe, W4TV


On 2014-06-09 2:07 PM, Kjeld Holm wrote:
> "The AFSK TX Filter cleans up any hum on your sound card from bad
> connections, grounding issues, or distortion from over drive.  However, with
> the K3 one can get the same narrow transmit spectrum by using FSK without
> the issues of AFSK." Written by Joe Subich, W4TV
>
> Not as I read the article - but I may be wrong.
> " Effect of the K3 AFSK Transmit Filter
> Rather than hacking the K3 to transmit
> through a narrow roofing filter, an equal or
> better result can be had by simply enabling
> the AFSK transmit filter in the configuration
> menu. This places a 400 Hz filter before
> transmit audio arrives at the RF modulator
> (this only applies to AFSK-A mode,
> not DATA-A mode). This filter is centered
> around the tones configured under the
> radio's PITCH menu. The effect is similar
> to what MMTTY and other programs do in
> their software, but this is done in the radio's
> DSP firmware.
> The radio's manual says this filter can
> serve to filter any noise that might be on
> the audio input, and it certainly will do that,
> but it has the additional benefit of filtering
> keying sideband energy. Figure 9 shows
> the result of MMTTY's unfiltered phasecontinuous
> AFSK audio into the K3 with
> the K3's AFSK filter enabled. Comparing
> Figure 9 with Figure 3 demonstrates the
> effect of the K3's AFSK filter, as both have
> exactly the same audio input to the K3.
> Attempts to overdrive the radio resulted
> in no significant change in signal bandwidth.
> Even when transmitting wideband
> noise into the K3, the AFSK filter limits the
> bandwidth. While it is possible to transmit
> trash that is difficult or impossible to copy,
> the K3's AFSK filter makes it unlikely that
> it would generate much interference on
> adjacent channels."
>
> "An important point: Even given the K3's
> transmit IMD, the occupied spectrum using
> shaped FSK is much narrower than
> that when using the internally generated
> FSK synthesizer. FSK keying sidebands
> are not unique to the K3; every radio that
> uses phase-coherent for its internal FSK
> generator will generate a wider spectrum
> at essentially any power level. The only
> differences will be in cases where the IF
> filter cuts off the sidebands.
> As an experiment, I decided to see
> just how wide I could make my signal by
> overdriving the K3's line input. I cranked
> the PC's headphone output to 100 percent
> and drove the ALC as hard as I could by
> setting the line input gain to maximum (see
> Figure 7). Some strange spurs show up in
> the spectrum, but even those are below the
> keying sidebands of the FSK transmitter
> (see Figure 2). Even trying to transmit absolute
> trash I wasn't able to make the AFSK
> Figure 4 - AFSK with MMTTY using a 512 tap TX BPF with a
> passband of from 2000 to 2400 Hz, 0 dBm
> Figure 5 - AFSK with MMTTY with 512 tap TX BPF, 100 W
> signal as wide as that of the internal FSK
> generator. The rise in the noise floor over
> the 2.8 kHz bandwidth occurs, because the
> K3 amplifies noise from somewhere; at this
> gain setting it is present whether or not anything
> is plugged into the K3's line in jack.
> This behavior is almost certainly specific
> to the K3, because the K3 scales audio in
> DSP before it ever enters the RF stages.
> Other radios are not going to protect you
> from yourself nearly as well, so you might
> expect harmonic distortion and other bad
> things with no limiting prior to the point of
> RF modulation."
> Oz1ccm Kel
>
>
>