[Elecraft] Re: RTTY using LSB versus AFSK-A

[Richard Ferch]

Mike K2MK wrote:

> Yes I just tried AFSK-A with the DPB filter off and the bandwidth opened
up 
> and it seems to provide the same comfort level that I had on LSB. Aside
from 
> the advantages of having the levels set differently from LSB and the 
> compression turned off is there a received signal advantage to using Data 
> Mode instead of LSB assuming both bandwidths are set the same.

I doubt whether there is any received signal advantage, but there are
several operational differences:

In AFSK A (or FSK D), if you use the tones chosen using the PITCH control
the dial display and most software will both display, log and spot the mark
frequency, in agreement with standard amateur procedure. If you use LSB or
DATA A, the dial displays the suppressed carrier frequency, which is not the
frequency you are transmitting on. You or your software will have to add
(DATA A) or subtract (LSB) the audio mark frequency from the radio's dial
display to get the correct frequency for spotting and logging.

In DATA A and LSB you can use whatever audio frequencies you want to within
the bandpass, whereas in AFSK A you may want to (and in FSK D you must)
realign your tuning so that the signal you are working is on the chosen
audio frequency. However, this is not really all that limiting - if you find
you need to narrow the bandwidth because of QRM, you will probably want to
be on that frequency anyway, because that is where the DSP filters will be
centred. Some software (including MixW) has an Align or Opt feature that
will do this for you automatically, i.e. if you click on a signal in the
waterfall, you can click on the Align button and it will retune the radio so
that signal is placed on the chosen audio frequency.  

The K3's SPOT tuning aid works properly for RTTY in AFSK A and FSK D (again,
assuming you have the PITCH set to the tones you are using) - it doesn't
work at all in LSB, and in DATA A it produces the CW sidetone, which is very
unlikely to be appropriate for RTTY.

The K3's CWT and TEXT DEC features also work for RTTY in AFSK A and FSK D,
but not with the other mode choices.

You mentioned that you are using MixW. MixW is a bit trickier to configure
for the K3 than for some other radios. Here, FWIW, are the settings I use in
MixW for the K3:

Under Configure -> TRCVR PTT/CAT, I have CAT set to KENWOOD and the Model
set to "All Kenwoods" (*not* to "Elecraft K2", which doesn't work very well
with the K3). The only check boxes I have checked are "CW is LSB" and "AFSK
in place of FSK". The CW pitch setting is the same as the K3's sidetone
pitch for CW. I have the Default digi mode set to RTTY (this selects the
K3's DATA mode, but does not distinguish among the four DATA modes - more on
that later).

Under the Details button, I have the port set to 38400, 8, None, 2, RTS set
to PTT and DTR set to Always Off (I use a Winkey for CW). Correspondingly,
the K3's CONFIG:PTT-KEY is set to rtS-OFF.

In order to set the data submode, I make use of MixW's "OnStartMode" macro
in the "Macros for this mode" mode-specific .mc file for each of the modes
that I use.

For RTTY, my OnStartMode macro contains <CATCMD:MD6;DT1;PC100><INVON> . This
sets the K3 to DATA mode, submode AFSK A, power 100 watts. It also sets MixW
to Inverted RTTY (MixW's default is USB, but AFSK A is LSB). The View ->
Spectrum display setting should be either RF,LSB or AUTO. You will need to
set the TX and RX frequencies under Mode settings to your centre frequency
(2210 Hz if the K3's pitch is set to 2125-170, 1530 if it is set to
1445-170, 1360 if it is set to 1275-170, and 1000 if it is set to 915-170).

To log the mark frequency, you should set the CAT correction for Digi to the
negative of your chosen centre frequency. This will ensure that logged
frequencies and cluster spots will be correct, but the waterfall/spectrum
display frequency legend will display the centre frequency, not the actual
frequencies of the traces in the waterfall. This 85 Hz offset occurs with
all radios; it is caused by the fact that for some reason MixW is programmed
to use the centre frequency instead of the mark frequency. You can adjust
the CAT correction so that the waterfall shows the correct frequencies, but
then the log and cluster spots will be off by 85 Hz. I prefer to log the
correct frequencies and live with the error in the waterfall.

I program one of my MixW macro buttons to contain <ALIGN:xxxx> where xxxx is
the centre frequency. This button is what I use to snap a signal back to the
chosen pitch at the centre of the filter bandpass. I have this button
defined in my mode-specific macros for RTTY, because I use a different align
value in other modes.

For PSK31, the first change is to reset MixW's CAT correction for Digi back
to 0. I have not found any way to automate this change, unfortunately. You
will also have to set MixW's View -> Spectrum display setting to RF,USB ,
because MixW's AUTO setting thinks the K3 is in LSB in DATA mode, whereas
DATA A is actually USB. My PSK31 mode-specific OnStartMode macro contains
<CATCMD:MD6;DT0;PC040;> , which puts the K3 in DATA A mode and sets the
power to 40 watts. This ensures that I don't inadvertently start
transmitting at 100 watts in PSK31, which would result in splatter. My Align
macro button contains <ALIGN:1500> for PSK31, which is where the K3's
filters are centred in DATA A.

73,
Rich VE3KI