It turns out, that remote use is another reason that it is good to build the K2 in one chassis and the 100 watt amp in another. It is perfectly safe to disconnect the power to the base (QRP) K2 with a timer. This avoids the need to interrupt the high-current connection to the 100 watt amp. I was assured that this is safe by Elecraft guru Don, W3FPR. If you ever ask a techical question (especially troubleshooting), Don is likely to give you the answer. Here is what he had to say about this (posted with permission – thanks Don):

Yes, it is sufficient to kill the power to the base K2 (when the KPA100 is mounted remotely).  Unlike the K3, there is little to be lost (bits mixed up) if you turn the power source off for the K2 during normal operation.

——— And he went on to say in another message ——-

If the base K2 and the KPA100 are mounted in the same box, then the situation is different – you would have to interrupt the higher current connection (to the KPA100) if everything were connected normally.
Now, there “is a way” — if you do not connect the AUX 12V connector from the KPA100 to the base K2,  the base K2 will not be powered by the KPA100 power supply connection – you will have to power the base K2 from the coaxial connector.  That makes the power off conditions the same as with the remotely mounted KPA100.

The antennas:
1 – A “long” 80m Dipole, ladder line, 4:1 balun, then a long coax run
to the rig
2 – A fan dipole with 80/40/20 meter elements (these are close-spaced
elements, with 80m and 40m using opposing legs of a single piece of
300-ohm twinlead), 1:1 balun, then a long coax run to the rig

This was a brief test. I tried only 10/15/20/40/80 meters, and tested
only a few frequencies on each band, including something close to the
bottom and top of each band.

The IC-7600 internal tuner was able to tune both antennas over most of
the 10/15/20/40/80 meter bands with the following exceptions:
1 – Could not match the 80m Long Dipole at the high end of 80m or
anywhere on 40 meters
2 – Could not match the Fan Dipole anywhere on 20 meters

The External LDG-100 PROII was able to match all the test frequencies.
The IC-7600 tuner is very good, and tuned most of the frequencies,
including some that appeared to be worse than 3:1 SWR (the stated
spec) on my antenna analyzer.

Please note – I am NOT knocking the internal tuner, and with a little
antenna work (pruning antennas and feedlines – not fun) I could
probably get matches everywhere I need to operate. So, while the
internal tuner is very good, the LDG (which has a nice ICOM interface)
covers a wider range. The LDG was sometimes very slow (10s of
seconds), but once a match is found it is stored and recalled quickly
next time you need it.

I have had the IC-7600 for a few weeks and have had a chance to do a bit of testing. I worked a short stint on a RTTY contest, a 10-hour CW contest, and made a few SSB and PSK-31 contacts. Operating the rig has been very enjoyable so far. It just seems to be very well thought-out and the IF DSP filtering is outstanding. The built-in RTTY and PSK decoders are great. Just plug-in a USB keyboard and go.

The internal antenna tuner works great but only handles up to about 3:1 SWR. Since I have some “ugly” wire antennas, I have acquired an LDG external tuner – the AT-100 PROII. It has an Icom interface that integrates it nicely with the rig (as well as the internal tuner, in my opinion). It definitely allows me to match a broader set of antenna situations than the internal tuner (as the specs would lead you to believe).

My approach is the use an Arduino board to monitor the PTT line from the K2. If that is active longer than the predetermined timeout period (FCC mandates not more than 3 minutes), it disconnects power to the rig.  I have the power disconnected for a minute, then restored. The K2 will not power-up in a transmit state, so the problem should be resolved.

Skype seems to work fine for the audio interface in both directions. I got good audio reports and had a solid 20-30 minute SSB QSO. Note that I was actually in the same room as the rig, but I was using the Internet connection. I was using a laptop with a headset and remotely operating the rig.

I had a small issue when I connected the Arduino board (using it for the dead-man timer) to the station control PC. The Arduino software install seems to have failed and it locked-up the PC. I had no time to work out the bug, so it has to wait until my next visit.

Merry Christmas.

Using a simple audio transformer on the PC at the Control position provides audio that sounds great. Only problem is there still seems to be interaction between the send/receive channels on Skype. Is Skype really a half-duplex link? That would actually be OK as the ham rig is only half-duplex. The trick would be controlling the “T/R” switch on Skype so it is in sync with the Transmit/Receive control for the rig.

I worked SO2R this year with an Elecraft K2 as my primary radio. The second radio, an ICOM IC-7000, was used to search for multipliers and for calling CQ on bands where there were still openings, but not enough stations to make it worth sitting there and waiting for calls.

My main goal the last couple years has been to get the Clean Sweep (all 80 US and Canada sections), but I was also set on getting well over 500 QSOs this year. By 10 AM on Sunday I had 78 multipliers. Maybe this would finally be the year I get ‘em all. I spent a fair amount of time looking for the remaining two as I worked the stations on the various Bands. About an hour before the end of the Contest, I got Nebraska (never had trouble getting that one before). So I now had only the ever-elusive NWT Section to work.

As time ran out, I never even heard NWT – not sure I ever worked that one. So I come up short again with 79 out of 80 multipliers. I DID work all US Sections in the 24 hours and I made about 550 QSOs. Not bad, but still a long way from being really competitive.

So I had a great time with CWSS this year, but still couldn’t get the broom
Maybe next year…..

I added a second antenna and planned to add a third. I now have two dipole/doublet antennas oriented in nearly the same direction. This supports my 2-radio operation but does not help much for directivity. When you use a doublet there are several directional lobes, resulting in some dead spots in the far field. Having another antenna, perpendicular to these, would smooth out the dead spots.

Unfortunately, when I arrived at the operating QTH  the day before the contest, a steady wind of 30 MPH or more was blowing. Cancel plans for the third wire antenna. I have a 6BTV Vertical antenna, so I put that on a ground post, just in case something happened to one of my wire antennas during the contest. I also found that the center of one antenna was about 8 feet lower than normal. A broken limb had become entangled in one of the end supports. Nothing I could do to fix that today, so I go with a slightly lower antenna for the contest, and hope it stays in the air.

In the Shack, I was nearly set for the contest and just needed to fire up the station and make sure N1MM and my DX Doubler (DXD) were ready to control everything. I had previously used an old laptop and recently replaced that with a small (mATX form factor) PC that was more “server-like.” About an hour before the start of the contest I realized that the DXD was not being controlled properly. The TX Focus was not switching between the radios from the keyboard. It turns out the PCI-based parallel port I added was not working. No time to play with that – back to the Laptop.

Now everything appeared to be ready to go.

The Arduino will maintain a “watchdog timer” which will be started any time a transmission begins. The watchdog counts down until it either expires (at the three minute duration point), or it is reset because the transmission ends. If it expires, an automotive relay is deactivated, removing the 12 VDC power from the rig. After some timeout period, an attempt will be made to reset the power.

An LED is used to provide a warning to a local operator when it is getting close to time to disconnect the power. This is mainly for test purposes. An optional buzzer output is also provided for the same reason.

Using TeamViewer 6, I have been able to configure the remote station using Ham Radio Deluxe. The Host computer has TeamViewer installed as a Service under Windows XP. When you install it, you must configure it for incoming remote access, and you must check the box in General Options, to allow it to “Interact with the Desktop.” Not sure what exactly that means, but if you do not check it, when the PC starts, TeamViwer will NOT start until a user logs in to the machine. That really defeats the purpose of the remote setup.

For audio, I have settled on Skype. TeamViewer audio was just not reliable and robust. Sound quality varied drastically. Skype is free and has proven to have very clean audio. It does take some fiddling to configure it with your station, but they provide a Test call that will help. You can setup Skype to automatically answer incoming calls. When you place a call to the remote station, you will need to mute your mic, or suffer problems with management of the duplex audio. This will have to be dealt with in more detail later, but for now I am only receiving the audio and have no need to send audio to the remote station

Following these tips, I now have a well configured receive-only station. I will be testing it for reliability over the next couple weeks. Now it is finally time to get serious about transmitting from the remote location. Should be fun.