Elecraft

I built most of my Elecraft stuff from kits, partly to save a few $ but mostly because I enjoy making things. Like Heathkit years before, Elecraft kits are excellent, providing detailed step-by-step instructions, all hardware and electronic bits (solder-it-yourself components for the K2, pre-made modules for the K3 and KPA500), and if needed responsive online support from Elecraft’s chief engineers plus the community of builders and fans. Everything worked first time and is a joy to use. The thrill of hearing the first callsigns on a receiver you’ve just finished is hard to beat - actually working DX with a homebrew transceiver is superb. These are not beginners’ projects though. The K2 took me a month of evenings to build and the K3 an intensive day. While the K3 is a ‘no-solder’ kit (basically assembling the case and plugging in pre-built modules and circuit boards), learn to solder properly before you try building the K2!

Elecraft K3

The K3 is more sophisticated than the K2 with a fully digital IF and AF using DSP for demodulation, noise reduction, AGC etc., on receive, and digitally generating the modulated waveform on transmit. Mine both have the second receiver option for SO2V. I can monitor the beacons and CW end of 10m simultaneously. It is handy to be able to pick out callers in a pileup and figure out where the DX is listening (in one ear) while simultaneously listening to the DX (in both ears).

The K3 has a true diversity mode with two independent receivers phase-locked by dint of sharing a common reference oscillator. On, say, 30m using a 30m vertical as my transmit antenna, listening to it in my left ear and the beam, vertical or dipoles on the sub-receiver feeding my right ear, I find diversity mode is worth about 3dB due to the vertical partially filling-in QSB dips on the other antennas. Signals sway from side to side in my head, floating across the gap between my ears (no comments please).

Whereas the K2 covers the main 9 shortwave bands 160-10m, the K3 also covers 60m and 6m, and can cover 2m with an optional internal transverter.

There are a few K3 mods and add-ons:

• Solve a nasty transmit spur on K3s released in mid-2009 by removing a single capacitor.

• Rather than constantly reaching over to the rig or adding some sort of external switch box for the DVR/CW memory buttons (M1-M4 plus the REC button used to interrupt sending), Logger32’s Radio Control Panel lets us define up to 36 macros. N1MM+ contest logger can also trigger the memories. The commands to trigger and abort the memories are easy enough, while more complex ones to set up split etc. just take a bit of patient experimenting to get right ... or simply copy/adapt the macros further down this page.

• The Elecraft KPA500 amp gives 500 watts out.

• The Elecraft KPA1500 amp gives - yes, you guessed it - 1500 watts out.

• The Elecraft KAT500 500W auto-ATU is cool if you use 500W or so.
  

Connecting the K3 to your PC

The K3 comes as standard with an RS232 serial port through which it can communicate with logging software, passing information such as its frequency and mode in either direction. RS232 is a well-defined standard. The K3 has a 9 pin RS232 connector.

If your PC has an unused real RS232 serial port, you simply need a suitable cable to connect it to the K3. However most of us need a USB/RS232 serial converter gizmo since modern PCs have USB ports but no RS232 ports. Some hams will insist that you need a particular chipset in the converter to avoid problems, but I guess I have been lucky: all mine have worked OK. USB/RS232 converters are cheap as chips so don’t worry about chipsets, just take a gamble. Buy one and try it. If you are highly risk-averse and have the $$, simply buy the 'official' one from Elecraft.

Here’s how to connect your PC to your K3 (I will assume you are using a USB/RS232 adapter on a Windows machine - if not, your mileage may vary):

1. Plug the USB/RS232 adapter into the PC’s USB port (any one should work). Wait a moment for the PC to recognize it and find the driver. A pop-up message should appear bottom right of the desktop saying something like “New USB device found – installing driver” and then shortly after “USB serial adapter installed”.

2. Now figure out which serial port the adapter is using:

   • Open Device Manager using WIN+X (hold the Windows logo key and tap X) then either click the Device Manager line or press M.

   • One of device types listed should be “Ports (COM & LPT )”: click the little arrow to the right of that line to open up a list of all the serial ports on your machine.

   • One of them will be the USB/RS232 adapter you have just installed, usually called something simple like “USB Serial Port (COM2)”, in which case COM2 is the number you need to use in your logging software (keep a note of it).

   • If you have several ports listed and don’t know which one it is, you can simply try them all by trial-and-error, or unplug the USB/RS232 adapter and wait a second to see which one disappears, then plug it back in to make it reappear.

3. Plug the RS232 end of the USB/RS232 adapter into the RS232 connector on the K3’s rear panel.

4. Configure the K3’s serial port:

   • Press and hold the K3 menu button to open the settings menu.

   • Select the RS232 setting by turning the VFO B knob.

     • Turn the VFO A knob to select the maximum speed (38400 baud).

     • In the unlikely event that doesn't work, select a lower speed.

   • In the same way, configure the PTT/CW setting to be rts-dtr.

     • You will be using the dtr setting to send CW now.

     • The rts PTT setting can be used to put the rig in transmit for audio data modes if you use RTTY, FT8 etc. (I prefer the PTT CAT command)

5. Tap the K3 menu button to exit the settings menu.

6. Optionally put the K3 into TEST mode by holding the mode up button, so that you can test the CW sending in a moment without actually transmitting. TX will flash on the display while in TEST mode. Hold TEST again later to put the rig back into normal transmit mode.

7. Start your logging software and get it to connect to the K3 - instructions for Logger32 and N1MM+ follow below.

8. Test it.

“Instant CW split” macro

When I come across a juicy morsel of DX operating split, I use the following key macro to set up the typical “Up 1” CW split instantly, making use of the K3’s sub-receiver. It is configured on one of the K3’s front panel keys so a single tap on the button, plus a bit of time to tune around and find a good frequency to transmit in the pileup if he’s not listening exactly 1 kHz up, is all it takes. No messing around with the split button, turning on the sub-receiver, resetting the VFOs etc. - it all happens automatically, quickly, correctly and reliably every time:

RT0;XT0;FR0;DV0;BW0100;SWT13;SWT13;UP4;SB1;BW0280;MN111;MP002;MN255;
1 2 3 4 5 6 7 8 9 10 --------------11---------------

1. RT0 = Turn RIT off just in case it was set previously.

2. XT0 = Turn XIT off just in case it was set previously.

3. FR0 = Turn normal split mode off, just in case it was set previously.

4. DV0 = Turn diversity receive mode off, just in case it was set previously.

5. BW0100 = Set the filter bandwidth to 1 kHz to focus-in on the DX, cutting off most callers in the pile [you may prefer even narrower settings, such as BW0060, but I tend to reserve narrow filtering for the few times when I really need it. If you normally operate SSB, use a wider setting such as BW0180 or BW0200 or more, otherwise you won’t be able to copy the DX!].

6. SWT13 = A>B [first time] i.e. copy the DX frequency from VFO A to VFO B.

7. SWT13 = A>B [again] i.e. copy the mode, filter settings etc. from VFO A to VFO B.

8. UP4 = Moves VFO A up by 1 kHz [note: 4 is a parameter not a frequency - see the programmers’ guide for other possible split values. “Up 1 kHz” is just a starting point that works for most CW DXers. SSBers usually split ”Up 5 kHz”, so use UP7 instead. Either way, be prepared to tune around for the best frequency on which to transmit].

9. SB1 = Turn on the sub-receiver to listen to the DX, now he is on VFO B.

10. BW0280 = Open the filter on VFO A to 2.8kHz so I can listen to more of the pileup in the hope of finding someone working the DX.

11. MN111;MP002;MN255 = Listen to audio from VFO B in both ears with VFO A in my left ear only [MN111 is the code for the audio mix function; MP002 is a parameter; MN255 ends the code. The parameter options are MP000 = A B (great if you are able to concentrate on either ear) MP001 = A AB (giving VFO A priority over B by being in both ears) MP002 = AB B (giving B priority) and MP003 = AB AB (both VFOs in both ears). Naturally, this command only works if the sub-receiver is installed, enabled and turned on!].

12. Optional extra LKB$ = Lock VFO B to avoid me accidentally tuning away from the DX station by knocking the VFO B knob [I prefer not to lock the VFO so I can follow the DX around if he should QSY, without having to release the lock, but this means I must be careful not to knock my little knob].

Notice that I do not use the K3’s built-in split mode to operate split in the conventional manner. Instead, I listen to the DX on VFO B and transmit (as usual) on VFO A. Advantages of this ‘reverse split’ configuration are:

• QRQ QSK still works fine, whereas split operation disables QRQ mode.

• I’m tuning around the pileup for a transmit frequency with my big knob, leaving my little knob well alone.

• I can use the K3's CWT tuning indicator to zero-beat my transmitter more accurately on other callers if I want to tail-end them, and if the DX station isn't tuning around his pile.

I'm tempted also to turn on the attenuator to reduce the audio QRM from the pileup, using RA01; but meanwhile I simply use the AF/SUB knob (with the K3 configured for b) to reduce the volume of VFO A relative to B. [Some ops use external audio mixers to combine different rigs/receivers.]

By the way, Logger32’s Radio Control Panel lets us send arbitrary CAT commands to the K3. Logger32 also has the brains to recognize split-frequency DXcluster spots (such as when DXers spot some juicy DX with "UP" in the comments, or more naively “UP 2.4” or “QSX 14027.4”). It is possible to have Logger32 send the split-frequency macro to the K3 when we click a split-spot. But, as always, don’t forget to LISTEN FIRST and find the best place to transmit. Don’t be a cluster crab.

“Instant CW pileup” macro

The following macro is handy for me to start ‘listening up 1’ with a single button press, when I generate a CW pileup of my own:

SWT13;SWT13;UP4;FT1;DV0;SB1;RT0;XT0;LK$1;BW0270;BW$0060;MN111;MP001;MN255;
1 2 3 4 5 6 7 8 9 10 11 ---------------12---------------

1. SWT13 = A>B [first time] copy the frequency from VFO A to VFO B

2. SWT13 = A>B [again] copy the mode etc. from VFO A to VFO B

3. UP4 = Move VFO A up by 1 kHz

4. FT1 = Turn split on (SWH13 would do this too)

5. DV0 = Turn diversity off

6. SB1 = Turn on the sub-receiver to keep an ear on my TX frequency, while also listening to the pileup

7. RT0 = Turn RIT off

8. XT0 = Turn XIT off

9. LK$1 = Lock VFO B, allowing me to tune through the pileup on VFO A without the risk of accidentally knocking my TX frequency on VFO B

10. BW0270 = Open the filter on VFO A (the pileup) to 2.7kHz to catch callers who are not right on my announced split freq (I can always narrow the bandwidth on A using the width control if the pileup gets too big, but mostly I prefer to pick out individual callers ‘by ear’)

11. BW$0050 = Narrow the filter on VFO B (my TX frequency) to 500Hz to avoid hearing split callers too close to me, while still listening for those who fail to split at all plus any kops who QRM my TX freq

12. MN111;MP002;MN255 = Listen to the pile in both ears, plus my TX frequency in the right ear only

“Unsplit” macro

Here’s a key macro to clear any split, locks etc., resetting the K3 instantly to my default unsplit setup:

SWT11;BW0270;SWT13;SWT13;FR0;SB0;RT0;XT0;LK0;LK$0;MN111;MP001;MN255;
1 2 3 4 5 6 7 8 9 10 --------------- 11 --------------

1. SWT11 = Swap VFO A with VFO B [since I have usually been listening to DX on VFO B with the reverse split setup, and want to continue listening to the DX when unsplit].

2. BW0270 = Set VFO A to 2.7kHz bandwidth, my preferred bandwidth for all modes.

3. SWT13 = A>B [first time] i.e. copy the DX frequency from VFO A to VFO B.

4. SWT13 = A>B [again] i.e. copy the mode, filter settings etc. from VFO A to VFO B.

5. FR0 = Turn off split.

6. SB0 = Turn off sub-receiver.

7. RT0 = Turn off RIT.

8. XT0 = Turn off XIT.

9. LK0 = Unlock VFO A.

10. LK$0 = Unlock VFO B.

11. MN111;MP001;MN255 = Set the audio mix to A AB [so if I later turn on the sub-receiver, I will hear it only in my right ear].

12. Optional extra DV1 = Turn on diversity mode - useful when I have a diversity antenna anyway!

Audio mix macros

These simple key macros set the audio mix directly, without having to navigate to, adjust and save the CONFIG menu options each time:

• SB1;MN111;MP001;MN255; turns on the sub-receiver and sets the audio to A AB feeding the main receiver to both ears and sub-receiver to the right ear.

• SB1;MN111;MP002;MN255; turns on the sub-receiver and sets the audio to AB B feeding the sub-receiver to both ears and main receiver to the left ear.

I have these macros assigned to the PF1 and PF2 keys on my K3s for convenience. You may prefer some other way of triggering them e.g. two of the FN keys on the P3, or sending them from your logging software via CAT.

Turning off the sub-receiver returns the rig to the normal A A audio mix with ‘stereocode’ effects if AFX is on. AFX makes quite a difference over plain monophonic audio on CW. It’s cool. Try it! In the K3’s menu (tap the Menu button), set AFX MD to dELAY 5 for the most pronounced effect.

K3 tips

If, like me, you use Logger32 with the K3 and if, also like me, you prefer to use the K3 in CW-REV mode rather than CW, there’s a simple workaround to get the K3 to stay in CW-REV when you click on a DXcluster spot - see my Logger32 page for details, also how to get Logger32’s DVK function to trigger the K3’s DVR memories.

The K3’s DVR memories work well on SSB and CW but I find the rig's user interface confusing ... cue the helpful DVR operating instructions by KE7X.

Here's a tip for AFSK RTTY users on the K3. The tones transmitted by the PC should match the tones expected by the K3, particularly if you have the TX audio filters turned on (it’s a menu option). If they don't (for example if you try to respond to a signal too high or too low in the audio frequency range), the PC audio will trip the VOX but little if any RF will emerge and the K3 sidetone/monitor will be quiet or silent. I discovered this when changing to a new sound card. I'm now using a second card to separate PC bleeps, MP3s, YouTube audio etc. from my pure RTTY tones. While playing with the settings, I discovered the K3’s PITCH setting in AFSK modes. The low tones (915Hz) sound nicer to my CW-tuned ear than the default high tones (2125Hz), so I set my K3 to 915. I neglected to change the MMTTY setting - largely because I tend to use NET in MMTTY to set the transmitted tones to match the received tones, which lets me 'tune' on RTTY just by clicking in the middle of RTTY sigs on the MMTTY waterfall (I hope my convoluted description makes sense!). With NET turned off, MMTTY defaults to high tones which the K3's line input evidently filters out (although the VOX presumably triggers on the raw unfiltered audio).

Having subsequently set the 915Hz tones in MMTTY, I saved the MMTTY profile to stop it reverting to the high tones every time MMTTY is re-started. The same applies when using MMTTY within Logger32: set the tones then save the profile.

According to someone on the Elecraft reflector, the K3’s numerous birdies can be significantly reduced if not eliminated by carefully dressing the cables within the case, knocking out the remainder using the clever LO displacement function in firmware. With a quiet receiving location, birdies on the high bands are the most annoying so perhaps I should give that a try, next time I have the case open.

K3 REF CAL re-calibration

Re-calibrating the K3’s VFO frequency is easy if you can have a high quality frequency reference such as a GPS-trained Rubidium source, or can hear one of the standard frequency transmissions such as WWV or WWVH:

1. Turn on your K3 and wait a good while (e.g. a couple of hours) for the rig’s internal temperatures to stabilize in your normal shack operating conditions. Don’t even bother trying this if your shack temperature fluctuates wildly while you are operating.

2. Tune to your frequency reference in CW mode. You should hear a carrier at roughly the same pitch as your CW sidetone (if you are uncertain of that, hold the SPOT button to hear your sidetone and check its pitch against the receiver audio). Pick the highest frequency reference you can receive a strong enough signal to discern its pitch clearly (e.g. 20MHz) as that accentuates any calibration error compared to using lower frequencies (e.g. 10MHz).

3. Hold the K3 mode-down button to switch to CW-REV mode on the other sideband. If the pitch changes, your VFO calibration is not quite right. Switch back and forth between CW and CW-REV a few times, listening carefully for any difference in the pitch.

4. Hold the MENU button to open the settings menu.

5. Turn the small VFO B knob to find the REF CAL option and make yourself a little note of the current value - well, the final 3 or 4 digits at least.

6. Turn the main VFO knob to adjust the REF CAL value up or down by a few Hz or tens of Hz - you should hear the audio pitch change a little as you do so.

7. Flick back and forth between CW and CW-REV to hear whether the pitch is the same. If the difference is greater than before step 5, you may have moved the VFO CAL the wrong way, so go back to the original value and shift it a bit the other way. You may also have gone the right way but too far: don’t worry, you’ll get there in the end!

8. Repeat steps 5 and 6 repeatedly, making smaller adjustments of the VFO CAL setting as you approach the point at which you can no longer hear any difference in pitch between CW and CW-REV.

9. You’re done! Tap MENU to exit the settings menu.

If you are not good at discerning small differences in pitch by ear, or want to be even more accurate, use audio spectrum analysis software such as SpectrumLab (ideally using a fine resolution setting such as FFT input size of 16384) to compare the pitch more accurately between CW and CW-REV. WSJT-X also offers a frequency calibration mode. You should be able to get to within 2 Hz of the true value fairly easily, and with care within 1 Hz.

If you screw up completely, reset the REF CAL to the value you noted in step 5 and either give it another go or just “live with it”. Minor frequency errors really don’t matter at all in normal ham usage. I re-check my K3 every few months, ending up with much the same REF CAL value each time. I only bother with this rigmarole at all because I want to record and report the actual frequencies of HF beacons as accurately as I can, partly so I can identify them later from the same information having tuned them in carefully using the K3’s CWT function.

K3 + Sennheiser headset

Santa brought me a fabulous Christmas present one year, a Sennheiser PC 230 headset. I have been a fan of Sennheiser headphones for decades. I had been happily using a pair of Sennheiser PX 100 lightweight folding headphones to monitor the HF bands on my K3 for several hours a day, while working at the PC. The audio quality and comfort were excellent, while the open-back earpieces meant I could hear things happening around me, such as someone knocking at the door. The PC 230 basically adds an electret microphone on a little adjustable boom so I no longer need to grab the old Kenwood handheld mic for my rare forays onto SSB. A switch in the boom enables the mic it swivels down into place, and disables it when I turn it back out of the way. The mic has a noise-cancelling pickup on the outward side of the boom, and the headphones have a rotary volume control neatly built into the right earpad. Cool!

To make it work on the K3, I had to enable the microphone bias (press ‘2’ when in the MIC menu option to toggle bias on/off) and adjust the mic gain and compression to suit. The PC 230 mic seems to have plenty of gain: I have it set at just 6 on the low-gain setting (press ‘1’ in MIC to toggle between high and low gain).

The PC 230 lead is comfortably long enough to stretch underneath the operating desk and plug into the K3’s rear panel phone and mic connectors, reducing the front panel clutter and clearing my desk a bit. I leave the front panel mic settings configured for a different noise-cancelling closed-earpiece headset that I prefer for contesting.

K3 wish-list & bugs

The K3’s firmware is amazing in what it can do but inevitably there are a few little bugs and flaws. Here’s my current wish list:

• There is a little firmware bug (acknowledged but unfixed by Elecraft) that stops CW-in-SSB working if I power-cycle the K3 in SSB mode, with QRQ configured on, on any band except 6m (!!). Until the bug is fixed, power-cycling with QRQ configured off restores CW-in-SSB;

• For some reason, CW-in-SSB also does not work if I assert the PTT line or operate split;

• I used to use a footswitch to hold the PTT closed during a CW over, mostly to stop my old valve amplifier dropping out between words. If I released the PTT footswith while sending CW (relying on VOX to continue holding the PTT) or even if I turned on the PTT while the rig was already sending on VOX, the K3 sometimes messed up the character being sent at that instant. PTT handling should really be a lower priority than CW sending ... [The workaround for this bug is to use QSK with a suitable amp such as the KPA500!];

• Fast tuning is a pain: even with two speed-control buttons (fine and rate) and using the clarifier as a fast-tune knob, it is awkward to get between, say, the 10m beacon sub-band and CW or SSB sections. “Ballistic tuning” was once mooted by Elecraft, meaning that the tune rate would increase if we spun the VFO knob/s faster ... but never materialised. Meanwhile I use the RIT/XIT encoder knob to QSY quicker, leaving the VFO on round kHz values;

• The V--> M button should be physically distinctive - I’ve coloured mine red with a marker pen until I learn not to overwrite a memory when I really meant to recall it to the VFO;

• There is no ‘quick memo’ function like on the TS850, FT1000 and others. I found that a really useful way to store the current settings, whizz off to check something out, and return to where I left off. On the TS850, a single button-press was all it took, or I could tune between 5 quick memo stores using a knob. I’m now using M2 as a single temporary memory - not quite as easy or flexible as a true quick memo function but better than nothing.

Other software defined radios are updated from time to time but users appear to have little say in what changes are made by the manufacturers. Elecraft claims to be very responsive to customers but I have to say that so far I have singularly failed to persuade them to fix a few reproducible bugs. C'est la vie.

Elecraft P3 panadapter

The P3 was, in effect, a freebie thrown-in by the seller when I bought my second K3. I didn't think I would find it any use ... but I've grown to love it. For example:

• In pileups, I can often spot the station currently being worked by the DX simply by looking for his signal appearing at the relevant times, I can guage the upper and lower limits of the pile, and I can often find holes in which to call the DX;

• I can monitor the 10m beacon sub-band for active beacons;

• In contests, it's a little easier to find a quiet frequency on a busy band.

All in all, I'm pleased with it.

Inside the case, there is a vertical circuit board adjacent to the front panel for the display and buttons, and a few cables leading to the sockets on the rear panel ... and plenty of space inbetween. I'm sure I could fit, say, a 13.8V 20A power supply in there but it would need to be carefully designed to avoid interfering with the panadapter electronics, perhaps using a toroidal transformer to reduce magnetic QRM.

KPA500 amplifier

See amps.