[RTTY] K3 In NAQP RTTY

[J. Edward (Ed) Muns]

I used a field-test K3 and an Icom ProIII in NAQP RTTY yesterday.  The K3
worked well, and flawlessly, for its first major RTTY contest.  I've been
using the same pair of radios this past month in the weekly NCCC Sprint (CW)
practice contests as well as a bit during IARU.  Since folks are eager to
hear about the K3, I'll describe how it worked in RTTY mode.

Interconnection

The K3 has excellent interconnection capability that has been well-thought
out for how we actually use our radios.  For example, in addition to the
microphone and headphone connectors on the front panel, both are
additionally available on the rear panel where the cables can be out of the
way of the operating area.  In my case, I use a Top Ten DX Doubler for the
SO2R switching between two radios and their standard pig-tail radio cable
was easy to install four connectors on to connect to the K3 rear panel
(microphone, headphones, key and PTT).

A standard RS232 cable connected the K3 to a COM port on the PC for radio
control with the logging software.  Of course, the station ground buss and
antenna coax connections were made along with the APP 12VDC supply cable.

For RTTY, interconnection is much simpler than any of the Kenwood, Yaesu or
Icom radios I've used.  For AFSK, all that is needed is two off-the-shelf
1/8" phone plug audio patch cables running from the K3 LINE IN and OUT jacks
to the PC sound card LINE OUT and IN jacks respectively.    The K3 has audio
isolation transformers built-in for each line, so the typical add-on
transformer dongle is eliminated.  The levels in each of these lines will be
settable in the K3 configuration menu and will stay constant as the AF Gain
control on the front panel is varied for operator headphone level.  (After
initial setting, the sound card levels can be readily adjusted by assigning
a front panel button to bring up the menu item or the Windows sound card
level applet can be used.)  These connections are independent of the
microphone, headphones and speaker connections, so nothing needs be
disturbed to add RTTY connections.  The only additional thing I did was
insert a Y-adapter in the K3 LINE OUT jack so I could run the receive audio
to a Hal DXP38 in parallel with the MMTTY software decoder.  For assurance
against ground loops, I did use an isolation transformer, but I don't think
it was really necessary.

For FSK, the typical level-shifted keying interface used for CW or FSK is
connected from a PC COM port to the DIG0 input on the Accessory DE15 jack on
the K3 rear panel.  PTT is also on this connector so both signals can run
from the same COM port.  Of course, for FSK the audio cable connecting the
K3 LINE IN with the PC LINE OUT for AFSK is not needed.

What I also eliminated from my normal setup was the JPS NIR-12 which gave me
independent headphone level control as well as additional AF DSP filtering.
With the K3 IF and AF DSP, plus its internal independent LINE and headphone
level controls, the NIR-12 is no longer needed.  And if it weren't for my
parallel hardware modem, no audio isolation transformers would be needed.

Note that with the K3 I will never have to move the headphone cable(s) when
switching from CW/phone to RTTY in order to have headphone level adjustment
independent of decoder input level.

KRC2

I have been using the KRC2 band decoders since I field-tested it several
years ago.  I've used them on Kenwood radios, K2s, Icoms and now the K3.
One on each radio controls my band pass filters and SixPak antenna switch.
I made up a cable to connect the AuxBus between the K3 and KRC2, downloaded
newly-released firmware with AuxBus control code and had automatic station
control from the K3.  There is also an Array Solutions SixPak control box
with LED indicators for the band selected for each radio and manual switches
that are wire-OR'd with the KRC2.

K3 DATA Mode

The user interface for the data modes is more powerful, yet simpler than any
I've used.  As always, AFSK can be run with the K3 in LSB mode.  But, AFSK
can also be run in DATA mode and enjoy all the features of FSK, including a
dial frequency of the Mark signal.  Now, when someone says, "meet me on
14083.7", both the AFSK and FSK operators of K3s can dial in the same
frequency and be zero-beat.

So, first the MODE button is pressed to move the radio into DATA mode (other
choices being CW, LSB, USB, AM, FM).  Then, the DATA MD button is held for a
second to bring up the DATA MODE menu.  Here the operator selects the data
mode with the VFO-B knob: AFSK, FSK and PSK are currently implemented.  The
speed is selected with the VFO-A knob, although for now there is only one
choice for each of these three modes (45, 45 and 31 baud respectively).
Third, the PITCH button is held for one second to bring up the TONE and
SHIFT menu.  VFO-A selects the Mark tone frequency, currently 915, 1275,
1415 and 2125 for AFSK and FSK.  VFO-B selects the shift frequency,
currently the only choice is 170 Hz for AFSK and FSK.  If you prefer
listening to lower tones, select 915 for example and simultaneously select
915 in your decoder.  But whether you are on FSK or AFSK, the dial frequency
is the same, regardless of what Mark tone or shift you select.  (Remember
this is running either AFSK or FSK in the K3 DATA mode.  If you run AFSK in
LSB on the K3, then the dial frequency is the LSB suppressed carrier
frequency and therefore skewed from the actual Mark frequency that is on the
dial in DATA mode.)  So, this is directly analogous to CW and varying the
pitch.

Holding the ALT button on the K3 reverses the Mark and Space frequencies on
both transmit and receive.  In addition, there will be a menu item (that can
be assigned to a user-definable button) that reverses the tone frequencies
only on receive.  There is also a FSK Polarity menu setting to change the
logic on the FSK keying line in case you need that.

Soon to be implemented is a dual-tone DSP filter for data modes like AFSK
and FSK that have two tones.  The user interface being discussed is very
slick, but let's wait and see what emerges.  Functionally it does a similar
thing as the Icom Twin Peak Filter, but I expect the K3 implementation to
much better.

Decoder/Encoder (generally useful outside a contest)

Currently implemented are decoders for CW, RTTY-45 and PSK31.  When the
DECODE button is held for a second, the Decode menu comes up and VFO-B
selects ON or OFF while VFO-A selects the detection threshold level.  A
level of 5-10 out of 0-30 seems to work best in most cases.  Exiting this
menu and tuning in a signal will result in the decoded text scrolling across
the VFO-B display area.  There is also a selectable CWT tuning indicator
that converts half the S-meter display area into a tuning indicator so you
can zero-beat signals in each of these modes without external hardware or
software tuning indicators.  And, the text can be output from the K3 to a
terminal program.  This is all pretty cool but not something you would
normally use in a contest.  However, it is extremely useful for a reading
RTTY or PSK without a PC connected or any other external gear or software.

Future enhancements include improvement of the decoding algorithms,
improving the tuning indicator, adding more speeds and shifts and data
modes, and an encoder for all modes that uses CW sent to the K3 as input.
Thus, one could use have a RTTY or PSK QSO using only the K3 with a paddle
connected.  Obviously, these capabilities could lead to making the K3
decoder output available to the RTTY portion of the logging or RTTY control
software, as well as having that same software send ASCII text to the K3 for
encoding there.  There are higher priorities for the firmware team right
now, so availability of some of these enhancements is not yet set.

NAQP RTTY Operation

Subjectively, the K3 performed as well or better than the ProIII for me in
NAQP.  I used a 250Hz roofing filter followed by a 200Hz DSP width setting.
I never felt that was too narrow.  It helped sort out the few pileups I had
by printing only those stations that were zero-beat.  For stations that were
within 200Hz or so, one of their tones was noticeable and I used the RIT to
quickly bring them in.  Thankfully, the K3 rig control is patterned after
the Kenwood command set and 'Clear RIT' is implemented.  My CQ and TU
messages all had the WriteLog '%O' parameter to clear RIT and that is a
welcome feature after having to manually press 'Clear RIT' on the ProIII.
Icom is the only manufacturer that doesn't have a clear RIT command in their
radio interface.  As always, I used FSK as I prefer it over AFSK, but that
is transparent in the K3 DATA mode.

I wasn't aware of any severe IMD situations (close-by strong signals) that
should have highlighted the K3's 250Hz roofing filter vs. the ProIII's 15KHz
roofing filter.  There was one instance of a S9 signal 1KHz away that put
some light popping in my passband, but I'm assuming that was transmission
impurities, not a K3 IMD issue.  Other than that, the passband was clean and
clear with only the tones that should be there.  Widening up the roofing and
DSP filters demonstrated a lot of signals nearby that were completely
missing at the narrow settings.  This can be a disadvantage though, if I am
able to receive clearly while close to other strong signals, because I may
not notice them.  Other stations trying to work me may not have as good
filtering and reception.  So, it is prudent to open up the filtering to
check around you, when you suspect a nearby "neighbor" that may be
interfering with your operation.

The K3 simply "sounds better" on all modes, including RTTY.  The band noise
is less severe to listen to, and the difference is striking just having the
ProIII in one ear and the K3 in the other with no signals in the passbands.
Yet, the K3 dynamic range accepts a much larger range of signal levels and
the K3 pre-amp can be left on even on the low bands where the attenuator
does NOT need to be turned on except in severe situations.  I ran with the
pre-amp on, maximum sensitivity, and the attenuator off on all bands 10-80
for the entire NAQP.  The K3 was less fatiguing to listen to for long
periods of time with the audio sounding very natural, whereas the ProIII
does sound digitized or artificial.  (Maybe if you're not old enough to have
spent many hours in front of a Drake or Collins receiver, you can't
appreciate this point!)  There were many times when I had perfect copy on
stations that could not hear me, even with no one else calling them.  This
occurred on all bands 15-80, so it wasn't just local noise level at the
other end on the low bands.  There were a few times, though, when I had
trouble printing a station that seemed to be copying me fine.  I don't know
what the cause of that was and I didn't take the time to swap radios to see
how the ProIII would handle it.

The K3/KRC2 combo also worked perfectly, never skipping a beat.  After
having automated band control for several years, it is hard to go without
it.

I do have to say that I really miss the band-scope.  Over the past two
years, I have grown accustomed to using it as a quick general indication of
band activity and openings.  While transmitting on one radio, I can quickly
switch bands on another radio and see if 10 meters is open, for example.
Sometimes I use it to then quickly QSY to indicated signals, but mostly I'm
using it in a macro sense.  Another use is seeing close signals that may be
masking my transmitted signal for stations I am trying to work (see above).
Elecraft has talked about doing a band-scope accessory in the future.
Meanwhile, I can run the IF output from the KXV3 to the ProIII and use its
band scope!

Summary

Based on my K2 experience, being in the K3 focus group for a couple years
and using the K3 for several weeks now, I am extremely optimistic that a
pair of K3s will be my mainstay SO2R station for all modes, especially RTTY,
for the foreseeable future.

73,
Ed - W0YK