Robert Jackson / Notes

2026-06-04 Sachuest Point

Thu, 04 Jun 2026 00:00:00 GMT

Sachuest Point National Wildlife Refuge, US-0516, was a good fit for a compressed morning. I had to be on Aquidneck Island anyway to pick up baked goods for my son's high school graduation, and I had about an hour and a half while the order was being finished. I hiked a little over half a mile each way into the refuge, out to Sachuest Point itself, which also made the activation qualify for Pack Mule. It was a beautiful morning: mid-60s when I arrived, mid-70s by the time I left, bright morning light, and a steady breeze coming in off the water.

At a glance

Where: Sachuest Point National Wildlife Refuge, US-0516, Aquidneck Island
When: June 4, 2026, 14:16 to 15:23 UTC
Activation: Pack Mule; 38 CW QSOs on 20 meters
Radio: Elecraft KX2 with the internal battery
Antenna: KJ6ER Challenger 25-foot whip on the REZ Antenna Systems mini tripod
Power: 5 watts from the KX2 internal battery, about 0.4 Ah used
CW gear: KXPD2 paddle/keyer, headphones

Field notes

Sachuest Point National Wildlife Refuge, US-0516, has stricter setup rules than a typical park. No stakes in the ground, and no wires in trees. The Challenger is a great antenna and I use it often, but this site changed the base rather than the antenna: instead of staking it into the ground, I screwed the KJ6ER Challenger into the small REZ tripod and kept the whole station no-impact.

I kept walking until I was out near the point, then stepped onto the big rocks along the shoreline and found one large, flat slab that worked like a natural operating platform. The chair, radio, tripod, and antenna all fit on the rock, and the open water made the site feel much more exposed than a typical inland park.

The only real mechanical concern was the wind. The operating spot was right on the water, and the breeze was steady enough that I did not want to trust the tripod on its own. Once the bag was unpacked, I set it over the tripod leg that was pointed into the wind. The bag probably weighs around 30 pounds fully loaded, and even after unpacking it still had at least 20 pounds of gear in it. Being a habitual over-packer helped for once. That worked perfectly for this site because the ocean breeze had one clear direction. I would not count on the same trick in swirling wind, but for a coastal morning like this it was simple and effective.

I used the KX2 on its internal battery with the KXPD2 paddle and headphones. The radio was balanced on my leg for the whole run, which worked, but it was the finicky part of the operating position. Every time I shifted I was aware that the KX2 could slide or tip if I was not careful. The headphones were not optional at that spot. The wind and surf were part of what made the morning so good, but they would have made weak CW much harder to copy through the KX2 speaker alone.

Once I found a place on 20 meters, the contacts came fast. I had only planned for about an hour, and the log filled almost continuously: 38 QSOs from 14:16:49 to 15:23:33 UTC. It felt like the fastest sustained CW run I have had so far. There were a few CQs in the middle, but nothing like the long CQ-repeat stretches from the previous day. This was mostly one caller after another until I had to stop and pack up.

The battery use was interesting too. The previous day's two-hour activation used about 1.4 Ah. This one used only about 0.4 Ah in roughly an hour. Some of that is just the shorter operating time, but the pace also meant I was listening more and repeatedly calling CQ much less.

CW copy felt better than it has on some recent activations. I still needed repeats, especially when a few stations came back at once, but I was able to pull one or two letters out of the pileups and build from there. I also copied several callsigns cleanly on the first pass.

It was great to get a few friends in the log this morning also. Today I got some friends from Discord N8JMS and KF8CNK, along with N1BS (a local Rhode Island ham who has been helping me get into RI POTA and CW generally). It is always fun to get people I know in the log. Conor, KF8CNK, has been camping in another POTA park in Michigan with limited cell service, it was great to get him for a park-to-park.

One small CW detail still makes me grin: when folks send me the proper roger with that elongated dah in the R, it sounds fantastic. I cannot really send it from paddles without switching into a bug-style mode, but hearing it makes me want to spend more time with a straight key, bug, or cootie key. Thanks for the proper roger N8JMS!

What worked

The Challenger on the mini tripod gave me a no-stakes, no-tree setup that fit the refuge rules.
The half-mile-plus hike each way and 38 QSOs made this a Pack Mule activation.
The flat shoreline rock made a stable natural platform for the chair, radio, and antenna.
Anchoring the windward tripod leg with the gear bag was enough for the steady coastal breeze.
Headphones were the right choice with wind and surf noise at the operating spot.
20 meters produced a nearly continuous CW run for the full hour available.

To adjust next time

Do not assume the bag-on-tripod-leg trick will work unless the wind has one stable direction.
Keep the tripod base option handy for other no-stakes or no-tree sites.
Find a kneeboard or leg strap so the KX2 is not just balanced on my leg.
Keep practicing pileup copy; pulling partial calls out of the noise is starting to feel more repeatable.

JL Curran State Park Pack Mule Activation

Wed, 03 Jun 2026 00:00:00 GMT

JL Curran State Park, US-6992, is one of my closest parks, and I have activated it a lot. I am slowly working toward a Kilo award here; the local ledger had me at 740 QSOs from JL Curran before this activation, so this log should move that to about 763 once it is merged. This morning I turned the familiar park into a small Pack Mule outing, hiking a little over half a mile in to operate from a quiet spot in the woods.

At a glance

Where: JL Curran State Park, US-6992, wooded operating spot off the trail
When: June 3, 2026, about 9:25 to 11:35 AM EDT
Activation: 23 CW QSOs; 3 on 40 meters and 20 on 20 meters
Radio: Elecraft KX2, freshly reassembled after the USB-C charger and side rail install
Antenna: 51-foot end-fed random wire, thrown roughly 35 to 40 feet high and running in an upside-down U shape through the trees
Power: Mostly 5 watts; the last two QSOs were logged at 10 watts
CW gear: Zippy Key paddle by K8CES; headphones and the KX2 speaker

Field notes

The activation started later than I wanted because the radio was still apart from the USB-C charger installation. I finished putting the KX2 back together that morning, packed it, and did not get to the park until around 8:45 AM. That made the morning feel a little compressed before I even got on the trail.

The weather was gorgeous, and I found another excellent operating spot in the woods where I could stand, move around, and not feel pinned to one position. There were boulders, a low stone wall, and enough tall trees to make the antenna worth the effort. The throw line went perfectly on the first throw, right where I wanted it. That has been a running skill to improve on recent outings, and this was one of the mornings where the practice showed up.

The 51-foot random wire went up high, maybe 35 to 40 feet, and the available supports made it run in an upside-down U shape through the trees. It was not the shape I would draw on paper, but it loaded and it got out.

The on-air part was slower than the final log makes it look. I made the basic POTA activation fairly quickly, then spent a long stretch trying to get over the Pack Mule line. There was a lot of QSB, and the pace kept stalling enough that I moved between calling CQ, hunting around for stations that could hear me, and checking what else was active.

I used headphones when I needed to copy carefully, but for the long CQ stretches I switched to the KX2 speaker and paced around the operating spot while listening. That was one of the nicer parts of this setup: the site gave me room to stand, move, and keep the activation from feeling like sitting still and waiting. Final count was 23 CW QSOs, with 22 unique callsigns. For a familiar local park, this one took more patience than expected.

What worked

The throw line landed exactly where I wanted it on the first try.
The high random wire was worth the effort, even with the odd upside-down U-shaped run through the trees.
JL Curran has enough quiet woods close to home to make a real hike-in operating spot practical.
Using the speaker during long CQ stretches made it easy to pace around and stay comfortable.
Staying with 20 meters eventually got the Pack Mule count, even though the pace was slow.

To adjust next time

Finish radio work the night before a planned activation, not the morning of the activation.
When QSB is that heavy, expect the second half of the log to take longer than the first ten contacts.
Keep looking for a cleaner wire path at this spot; the site is good enough to revisit.

Installing USB-C Charging and Side Rails on My KX2

Wed, 03 Jun 2026 00:00:00 GMT

I finished two small upgrades to the KX2: the KXUSBC2 internal USB-C charger from G7UFO and the Side KX rails from Gems Products. The KXUSBC2 replaces the KXIBC2 that came factory-installed in this radio, using the same side-plate location to add USB-C charging. The rails replace the stock side plates with protective handles, which should make the radio easier to pack for portable operating without worrying as much about the knobs and front panel.

This is a field note, not a replacement for the manuals. I am not a serious tinkerer, and opening up a brand-new radio still makes me pause. The install went fine, but it was not a mindless follow-the-pictures job either. My KX2 was manufactured in mid-May 2026, and one bit of the right-side hardware did not look like the manual photos. The useful part of this note is probably that moment: slowing down, comparing the parts in front of me to the instructions, and figuring out what Elecraft changed.

At a glance

Radio: Elecraft KX2, manufactured mid-May 2026
USB-C charger: G7UFO KXUSBC2
Rails: Gems Products / Gem's Products Side KX for the KX2
Extra parts: non-rail KXUSBC2 side plate, plus an uninstalled Pro Audio Engineering heatsink
References: the KXUSBC2 project notes and the Side KX KX2 manual
Why: USB-C charging in the radio and better side protection for portable operating
Result: USB-C charging worked; the charge indicator glowed green after plugging in a USB-C charger

The parts

The G7UFO kit includes the charger board and a replacement side plate. I also bought the protective rail version for that same side, plus the matching rail for the other side of the radio. Before taking anything apart, I laid out the parts and took a few photos so I could tell which screws, washers, and plates came from which kit.

That was worth doing. There are enough small screws, washers, thermal pads, and connectors involved that I did not want to rely on memory once the radio was open. It also means I have the normal non-rail KXUSBC2 side plate left over if I decide later that the handles are more than I want for a particular pack or trip.

I also picked up the Pro Audio Engineering heatsink, but I left it out for now. It adds a few ounces, and most of my portable activations are five-watt CW/SSB outings rather than long digital-mode sessions. For now I would rather keep the KX2 light. If I start pushing it harder, especially on digital modes, I already have the heatsink and can add it later.

Opening the KX2

I worked on a towel and treated the first part as slow disassembly: battery out, side hardware off, screws kept together, and no tugging on small internal connectors.

The KXUSBC2 uses the same two RF-board connection points as the factory charger board. In my radio the visible labels were B and E, and the replacement board had matching red and white wiring. This was simple enough, but I was glad I had photographed the original routing before disconnecting anything. A phone photo is cheap insurance when the radio is open on the bench.

The stock side plate

The factory KXIBC2 side plate came out as a small assembly with the connector wiring, thermal material, and heat-transfer surfaces still attached.

This was the first place where "remove the side plate" covered more detail than I expected. The parts are small, the clearances are tight, and the plate has enough bits attached to it that I wanted to understand what was structural, what was thermal, and what was simply coming along with the old assembly.

The newer KX2 screw difference

The trickiest part on my radio was one right-side screw/fastener arrangement that did not look like the pictures I was using. My KX2 was built in mid-May 2026, and the area around the PA/heatsink/side plate matches the newer mechanical revision that Gems Products notes for KX2 radios after serial number 5441, roughly since May 2025.

The manual described a small screw with only a few threads. On my radio, the part behaved more like a longer screw threading into a small barrel-shaped standoff. I would call it a threaded standoff or internally threaded spacer: not a normal nut, but a little sleeve with threads inside it. As far as I can tell, it is only receiving that screw. The screw does not pass all the way through the standoff and attach to something on the other side.

That was the moment where I had to stop treating the manual photo as literal and look at what the radio was actually doing. Once I realized that the longer screw and standoff were the newer version of the same attachment point, the rest of the install made sense again. The important thing was not to force it.

Moving the thermal interface over

The Side KX plate still needs to preserve the thermal path that the original side plate provided. I reused and moved the thermal material as instructed, then checked the coverage and alignment before reassembling the side.

This was another place where going slowly mattered. The rail is a mechanical upgrade, but it still has to behave like the original side plate electrically and thermally. I did a dry fit before tightening everything down so I could make sure the wires were not pinched and the plate was seating correctly.

Fitting the rails

Once the side-plate details were sorted out, the rail install was mostly the kind of mechanical work I expected: line up the plates, confirm the connector openings, and tighten the hardware evenly.

I checked the fit from a few angles before closing the radio. Nothing should bow, the connector openings should line up, and the internal wiring should still have a natural path. That is all obvious once the part is seated, but it is much easier to catch before the case is back together.

Closing the radio

After the internal checks, I reinstalled the battery cover and looked over the radio from the front, back, and sides. The rails make the KX2 look a little more field-ready, and they give the radio a much better handhold without adding a lot of bulk.

The last real test was simple: plug in USB-C power and see whether the charger behaved.

It worked. I plugged in a USB-C cable, got the green charging light, and the radio looked happy. The first real field use came later that morning during a Pack Mule activation at JL Curran State Park. That does not make the install fully proven yet, but it did get the radio through a normal portable CW outing.

What I would do the same way again

Take photos before disconnecting anything.
Keep the original screws and small parts organized by side and step.
Treat the instructions as the source of truth, but compare them against the actual radio before forcing any part.
Dry-fit the rail before tightening the hardware.
Check USB-C charging before putting the tools away.

What I still want to verify

Do a longer charge cycle and a few more portable operating sessions before calling the install boringly complete.
Decide after a few activations whether the rail handles stay on full time or whether I prefer the normal side plate for some trips.

CQ WPX CW as a Learning Weekend

Sun, 31 May 2026 00:00:00 GMT

I worked a little of the CQ WPX CW contest this weekend (2026-05-30 & 2026-05-31), but not in any serious score-chasing sense. I was on the air for maybe eight hours total and finished with 37 QSOs. The real goal was to use the contest as a learning opportunity and as a very dense source of DXCC entities that I would not normally find so easily.

The contest speed was often far past what I can comfortably copy by ear. A lot of stations were running around 35 to 40 WPM, and I had to work for many of the contacts. I copied most of the station callsigns only after several rounds of repeated CQ calls. After the first few QSOs I started recording the audio on my phone and playing it back slower so I could confirm the serial number for my log. It was awkward, but it totally worked.

At a glance

When: May 30 to May 31, 2026
Operating time: Roughly eight hours total
Result: 37 CW QSOs, with all but one on 20m
Radio: Elecraft K4D at 100 watts
Antenna: DX Commander Expedition (native 1/4 wave for 20m)
Headphones: Heil Pro 7

I used the K4D at 100 watts into the DX Commander Expedition which is basically just a quarter wave vertical for 20m, and it did amazingly well. Most of the log was on 20 meters, with one 40-meter contact to the Isle of Man (!!). I used headphones, but I also kept the K4D speaker active so I could record difficult exchanges on my phone when I needed to.

The K4D's decoder helped sometimes, but only up to a point. When conditions were clean it could give me enough confirmation to stay in the contact. When there was QSB, crowding, or nearby interference, it was not reliable enough to use on it's own as confirmation.

The contest was part of the backdrop for my East Bay three-park POTA rove, but that outing stayed focused on checking off more Rhode Island parks rather than trying to operate inside the contest itself (though the contest did make it VERY crowded).

What the contest actually felt like

The first barrier was nerves. Sending in a contest feels more exposed than a normal POTA contact because the pace is so unforgiving. Once I got past that, though, it went better than I expected. A couple of operators were really kind and matched my speed for the serial-number part of the exchange (but most didn't).

I made plenty of mistakes. I got enough NR? and NR NR requests for repeats that I was hearing it my sleep on Friday night. That was probably the most honest part of the whole weekend: I was fast enough to get into the exchange, but not consistent enough to get through all of them cleanly on the first try.

The one part of the exchange I still do not buy is the RST. Everyone sends 5NN every time. I understand that this is simply how the contest works, but it really does feel like dead weight in the exchange.

Why it was worth doing anyway

The contest was useful because it compressed a lot of rare-for-me DX opportunities into a short window. I was watching the DX cluster, looking for countries I do not already have, especially on CW, and then trying to work those stations even if it took several repeats to get through the exchange.

That strategy paid off. Based on my LoTW baseline before the contest, the weekend added 13 new DXCC entities in any mode:

Crete, Estonia, Cayman Islands, Georgia, Trinidad & Tobago, Hawaii, Isle of Man, Bulgaria, Cyprus, Greece, Poland, Serbia, and Bonaire.

On CW specifically, it added 18 new entities:

Crete, Estonia, Cayman Islands, Georgia, Trinidad & Tobago, Hawaii, Isle of Man, Belgium, Bulgaria, Cyprus, Denmark, Greece, Poland, Serbia, Morocco, Bosnia & Herzegovina, Czech Republic, and Bonaire.

I do not have many confirmed entities yet in any mode, and even fewer on CW, so contest weekends like this are a little like shooting fish in a barrel. That is not a complaint. It was awesome!

What I want to keep

Keep using big contests as structured CW practice, not just as a score contest.
Keep leaning on the DX cluster for targets when the full-rate contest flow is too fast to follow casually.
Trust the 20-meter DX Commander Expedition setup more.

What I want to improve

Get better at copying serial numbers in real time at contest speeds instead of relying on phone recordings.
Keep working on sending accuracy so I do not turn every exchange into a repeat request.
Spend more time intentionally practicing at contest-adjacent speeds instead of only comfortable speeds.

East Bay Three-Park POTA Rove

Sat, 30 May 2026 00:00:00 GMT

This was a three-park evening rove through the East Bay side of Rhode Island: US-6986, US-10546, and US-8293. The first stop was crowded by the CQ WPX CW contest, the second turned into an SWR debugging session, and the last made up for both of them with a fast 20-meter run right on the water at sunset. I finished the rove with 43 CW QSOs: 14 at Simmons Mill, 14 at Eight Rod Farm, and 15 at Sapowet Marsh. See my separate writeup about the CQ WPX CW contest.

At a glance

Where: Simmons Mill Wildlife Management Area, Eight Rod Farm Wildlife Management Area, and Sapowet Marsh State Wildlife Area in Rhode Island
When: May 30, 2026; rove start logged at 19:51 UTC, move to Eight Rod Farm at 21:14 UTC, move to Sapowet Marsh at 23:36 UTC
Activation: 43 CW QSOs total at 5 watts with the KX2; 14 at Simmons Mill, 14 at Eight Rod Farm, and 15 at Sapowet Marsh
Radio: Elecraft KX2
CW gear: N3ZN ZN-Lite II paddle on a magnetic base

Simmons Mill Wildlife Management Area, US-6986

I started at Simmons Mill with the KX2, 5 watts, and the Challenger 20-meter off-center-fed dipole. This was still early enough in the evening that the CQ WPX CW contest was filling a lot of the band, so the first problem was simply finding a place to operate POTA at a speed I could manage. I did not want to mix a POTA activation with contest-speed copying.

Once I found a spot, it came together, just not quickly. I got a burst of contacts early and then had to work through longer gaps before the activation was finished. I ended up with 14 CW QSOs between 19:51 and 21:06 UTC. I was using the new paddle on its magnetic base, but I had forgotten to grab the leg strap out of the bag, so I was basically hand-holding the paddle while keying with my right hand.

This one feels like the baseline stop for the whole rove: perfectly workable, but with enough friction from the contest and the improvised keying setup that I was already thinking about what to clean up at the next park.

Eight Rod Farm Wildlife Management Area, US-10546

Eight Rod Farm was the frustrating one. I reached that stop a little after 21:14 UTC and stayed there until the move to Sapowet Marsh at 23:36 UTC, which already tells the story: a lot of that window went into setup and troubleshooting rather than operating.

I started with a vertical setup built around the REZ 17-foot whip and a loading coil for 40 meters, but the actual antenna was the POTA Performer by KJ6ER with the 40-meter extension radials. That is why I needed the 40-meter coil in the first place. I made one or two contacts while trying 40 and maybe 30 meters, but the SWR was wildly wrong for a setup that should have been resonant. After chasing that for a while, I found the actual problem: a loose coupler at the connection between the choke and the whip assembly. Once I tightened that up, the SWR dropped to essentially 1:1 and the antenna behaved normally.

At the same time, band conditions were getting worse. The K-index had climbed from 1 earlier in the outing to 5 by then, with a more negative Bz and higher solar wind. So this stop turned into a combination of self-inflicted hardware trouble and genuinely degrading propagation.

The good change here was ergonomic rather than electrical. When I reworked the antenna and shifted to 20 meters, I also took the time to get the leg strap out for the paddle's magnetic base. I had been keying left-handed at Eight Rod, hand- holding the paddle for the first part of the stop. Once I put the magnetic base on the leg strap, it was much better. I finished with 14 CW QSOs from 21:59 to 23:15 UTC, with contacts on 40, 30, and 20 meters.

This was also the first time the KX2's internal battery made me pay attention. I expected it to be full enough for the outing, but I do not think I had actually charged it long enough before taking it out the first time. It threw a low-battery warning, so I plugged in the 3 Ah pack and kept going. It never died, and it was not really a problem in the field, but it is a reminder that "probably charged" is not the same thing as charged.

Sapowet Marsh State Wildlife Area, US-8293

Sapowet Marsh was the payoff. I switched there at 23:36 UTC and set up right at the shoreline on the inlet from the ocean into Mount Hope Bay. This one used the REZ Rybakov-style setup: 25-foot whip, 4:1 transformer, and four 33-foot radials. The radials were basically touching the water.

I got the station ready about ten minutes before the UTC day rollover and then waited to start calling CQ until after 00:00 UTC on May 31. I did not want to scatter a few pre-midnight contacts into a separate park-day log and then have to start over. Once the clock rolled over and I called CQ, though, the whole thing lit up immediately.

This was the most satisfying operating of the day. The logging app estimated that I was running at roughly 44 QSOs per hour over the first stretch, which matches how it felt: nearly a contact a minute, small pileups, and much better control than I remember having in similar moments before. I was keying with my right hand again here, with the paddle base moved to the right side, and I was noticeably more comfortable and more accurate.

One of the fun moments was hearing a call that sounded like it ended in X and thinking it might be Rory, W8KNX. It was, and he was loud from Michigan. The broader pattern was even better: contacts from California, Nevada, Michigan, Illinois, and Florida, all at 5 watts. Being right on the saltwater almost certainly helped.

The full log ended with 15 CW QSOs from 00:06 to 00:29 UTC on May 31.

What worked

The KX2 kept the whole rove simple. One radio, 5 watts, CW only, and no separate power-system drama.
The new N3ZN ZN-Lite II paddle worked well once I actually used the leg strap with the magnetic base.
Waiting for the UTC rollover at Sapowet Marsh was the right call. The log stayed clean and the run came together immediately afterward.
The shoreline Rybakov setup at Sapowet Marsh was exactly the right antenna for that location.

To adjust next time

Pack the paddle leg strap where I cannot forget it. That was avoidable friction at the first stop.
Double-check every threaded antenna connection before I start chasing propagation or tuning problems.
When a rove depends on a late third stop, I should assume less operating time than the clock suggests. The second-stop troubleshooting ate a lot of margin.

Durfee Hill Wildlife Management Area POTA

Fri, 29 May 2026 00:00:00 GMT

At a glance

Where: Durfee Hill Wildlife Management Area, US-7715, hilltop above the main valley, Gloucester, RI
When: May 29, 2026, 13:12 to 14:16 UTC
Activation: 19 CW QSOs; 4 on 40 meters, 5 on 30 meters, 10 on 20 meters
Radio: Elecraft KX2 (first field outing)
Antenna: Reliance Antennas wire Rybakov; 25 ft wire supported ~30 ft up in a tree via throw line, four ~17 ft radials, 4:1 unun
Power: 5 watts, KX2 internal batteries
CW gear: KX2 built-in paddle

Field notes

Durfee Hill had been on the list for a while, and this morning had everything going for it: 65 degrees, clear sky, no wind to speak of. I also had a new reason to get out — the KX2 arrived yesterday and I wanted to get it in the field as soon as possible.

The park is mostly open wetland and meadow in a shallow valley between low wooded hills. I hiked in roughly a third of a mile — short of the half-mile required for Pack Mule — but the operating position was the main reason I kept walking. The natural choice would have been somewhere down in the valley, but every spot I looked at felt like it was sitting in a bowl. I kept going up the hill instead and found a good edge-of-treeline position at the top, where the field opens up and the trees have real height.

The throw line went to the first branch I aimed at, roughly 30 feet up — the best throw I have made. The problem came right after. The weight cleared the limb cleanly, but the line had tangled inside the bag on the way up, so the whole bag was hanging elevated off the limb instead of paying out. I had to untangle it on the fly while hauling everything into position. Still, the branch was exactly right, and once the wire was up the rest of the setup took only a few minutes more.

I ran 25 feet of RG-316 back from the antenna to the camping chair, plugged in, and let the KX2's internal tuner match the Rybakov. It locked on fast — noticeably quicker than the KX3. I started on 40 meters and made four contacts, then moved to 30 and picked up five more. The best run was 20 meters — ten contacts from 14:00 to 14:16 UTC, including VO1AW in Newfoundland. Brian hunts me constantly — I hear "VO" in a pileup and just autocomplete at this point. Picked him right out.

The RBN spots confirmed I was getting out well across all three bands. I did try to hunt N1BS, who was activating another RI park at the same time, but he could not hear me. Still — for the smallest state in the country, Rhode Island has a surprisingly active POTA scene.

The KX2 was excellent. The internal speaker is noticeably better than the KX3's — though my KX3's speaker may be damaged, so I'll reserve final judgment. The KXPD2 paddle is also a step up from the KXPD3 on the KX3: better feel, cleaner action. The ATU locks on fast. Running on the internal battery was a revelation — just the radio, the coax, and the key, with no separate battery pack or power wires to manage. The whole station felt genuinely simpler. The KX2 also has a built-in Ah usage meter, which is a nice touch — this activation came in around 0.475 Ah.

What worked

Hiking past the bowl to find a real elevated operating spot paid off for propagation and for the antenna height.
First-throw throw-line accuracy continues to improve — hit the right branch first try at 30 feet.
The KX2 internal ATU matched the Rybakov quickly and held across all three bands.
20 meters was the right band to finish on; the run came together fast once I moved there.

To adjust next time

Repack the throw line bag after each use so it pays out cleanly. The tangle was avoidable.
A third of a mile short of Pack Mule qualification. Worth planning the route to the half-mile mark next time.

Black Hut Wildlife Management Area with a 40m EFHW

Wed, 27 May 2026 00:00:00 GMT

Black Hut Wildlife Management Area, US-6984, was a new one for me, and I went in planning to make it a hike-in activation. From where I parked it was roughly a mile to a mile and a half each way, which also made this a good candidate for a Pack Mule entry if I could get past 22 contacts.

At a glance

Where: Black Hut Wildlife Management Area, US-6984, hike-in operating spot in the woods
When: May 27, 2026, 13:23 to 15:13 UTC
Activation: 25 CW QSOs in 1 hour 50 minutes; 12 on 40 meters and 13 on 20 meters
Radio: Elecraft KX3
Antenna: Spooltenna Ultra 40m EFHW; far end in a tree around 35 to 40 feet up, feed point elevated on a POTA Explorer 20-foot mast
Power: 5 watts
CW gear: VK3IL paddle

Field notes

The station was simple, but the antenna setup felt like progress. I have been working on my arborist throw-line technique instead of just accepting whatever low branch is easy to reach. This time I found a useful opening, got the line where I wanted it, and pulled the far end of the EFHW roughly 35 to 40 feet up into the tree. The feed point went on my 20-foot POTA Explorer mast with a ground stake, and a short run of coax came back to the KX3.

Near the operating spot there was an old stone structure that looked like a cellar or foundation. I do not know exactly what it was, but it was partly below grade with a few feet still above ground, which made it a convenient standing table. I set the radio up there and operated the whole activation on my feet.

40 meters started fast. I made six contacts almost immediately, all within about five minutes, and then the pace dropped off. After a while of calling CQ without much coming back, I moved to 20 meters and started picking up stations again. That helped, but it was not steady. The longest quiet stretch was about half an hour of calling CQ on 20 meters from 13:49 to around 14:20 UTC.

I wanted the hike and I wanted the log to count toward Pack Mule, so I stayed with it instead of packing up after the first run faded. Once I had stalled out on 20 meters, I moved back to 40 at 15:04 UTC. That did it. Another run of callers came through quickly, and I finished with 25 QSOs total: 12 on 40 meters and 13 on 20.

The hike in and out was straightforward, and the site itself was beautiful. More than anything else, this was a good reminder that a hike-in activation does not need a complicated station. The KX3, the VK3IL paddle, and the 40m EFHW were enough, and the throw-line success made the whole setup feel much more repeatable than it has on some earlier outings.

What worked

The throw line finally felt reliable enough to make a higher wire deployment worth the effort.
Starting on 40 meters and returning there at the end both paid off.
The stone structure made an unexpectedly good standing operating position.

To adjust next time

Keep practicing throw-line accuracy so the first good shot is closer to the one I actually want.
Expect longer dry spells on a hike-in activation and switch bands sooner when the run fades.

Memorial Day Rhode Island POTA Rove

Mon, 25 May 2026 00:00:00 GMT

Memorial Day turned into a five-stop Rhode Island POTA rove: six park references, 64 CW contacts, and a pretty good reminder that a small radio setup can cover a lot of ground if I keep moving.

The radio for the day was the Elecraft KX3. Most of the activations were simple: set up, get the contacts, pack down, and drive to the next spot. A few were not simple at all. That is usually where the useful notes come from.

Misquamicut State Beach, US-7719

I started at Misquamicut State Beach and set up directly on the sand. The beach was cloudy, overcast, and loud, but the weather was doing me a favor. On a clear Memorial Day this would have been a very different place to operate.

I used the KX3 with the REZ Scout, the REZ Compact 17-foot whip, and the K8CES Zippy key. Setup and teardown were each about 10 minutes, and I logged 15 CW contacts between 16:56 and 17:39 UTC.

The whip went straight into the sand near the waterline, with the counterpoise running down toward the surf. Nothing about it was elegant, but it was fast, compact, and good enough to get the first activation done.

Woody Hill Wildlife Management Area, US-6983

The second stop felt like a different day. Misquamicut was surf and sand; Woody Hill was wet leaves, green woods, a narrow trail, and an old stone wall nearby.

I used the KX3 with the KJ6ER Challenger and stayed on 20 meters. I logged 11 contacts from 18:51 to 19:20 UTC. The whole stop was quick: about 40 minutes including setup and teardown.

The Challenger was a good fit for this kind of stop. It went up quickly, did not need much room, and kept the activation from turning into a project.

East State Beach, US-7865

East State Beach brought me back to the water. By then the sun had come out and the day had turned beautiful.

I made 15 contacts on 20 meters from 20:15 to 20:51 UTC. I used the KX3, the REZ Scout, the REZ 25-foot whip, and the Palm Radio Pico paddle. At the time, I thought I had set up the usual REZ Rybakov setup. During teardown I realized I had not.

What I actually used was just the 25-foot whip with four 33-foot ground radials, leaving the KX3 to deal with the mismatch. The KX3 tuner handled it without complaint. I would not have planned it that way, but it worked, and it is worth remembering.

Ninigret National Wildlife Refuge, US-0515

Ninigret was the fast one. I used the KX3 with the KJ6ER Challenger off-center-fed dipole and the Modern Morse Nameless key. All 11 contacts were on 20 meters, logged between 21:37 and 21:58 UTC.

The logistics were as clean as they get on a rove: about five minutes to set up, about five minutes to tear down, and then I was back at the car. At that point I still had enough daylight and momentum to try one more stop.

Kimball Wildlife Refuge, US-7518 / Burlingame State Park, US-2871

The last stop was a twofer, and it was the one that made me work for it. I parked at Kettle Pond Visitor Center and hiked in. The trail was easy and well marked, but it was still a hike-in activation after several earlier stops.

I used the KX3 with the KJ6ER Challenger off-center-fed dipole in 20-meter mode. I started with the Palm Radio Pico paddles, but after the first contact I started getting a stuck dit. I could not send anything with multiple dits reliably, so I swapped keys in the middle of a contact and sent something like, "sorry, I broke the key." I hope that made sense on the other end.

The K8CES Zippy paddle saved that activation. Once I switched keys, the next contacts came pretty easily. Then another station landed directly on my frequency while working the western U.S. For a few minutes the reports were confusing: some callers were answering me, and some were answering him. I moved up a couple of kilohertz and carried on.

The bigger problem came after about eight contacts. I had been running 5 watts from a 3 Ah Bioenno LiFePO4 battery, the same battery I had used on a six-park rove two days earlier. I had not charged it after that trip. The KX3 shut off in the middle of the activation, so I turned the power down to 2 watts and hoped there was enough left to finish.

There was. The last few contacts were at 2 watts, and the reports were still good. Ohio and Tennessee were still reachable from the woods in Rhode Island. I finished with 12 contacts from 22:28 to 23:06 UTC and got both park references in the log.

Afterward

The day ended at 64 contacts across six park references: US-7719, US-6983, US-7865, US-0515, US-7518, and US-2871.

The lessons are simple enough. Charge the battery before a rove. Carry a backup key. Do not assume the matching network is installed just because the antenna looks familiar. And keep moving, because even the messy stops can still make the log.

Reworking this site into a public workshop

Sun, 24 May 2026 00:00:00 GMT

The site is moving away from a pure identity page and toward a small public workshop: notes, field logs, technical experiments, and project context that can accumulate without needing a full essay each time.

The goal is not a content machine. The goal is a place for useful fragments that are easier to find here than in a chat log, issue thread, or local scratch file.

Portable HF logging workflow notes

Mon, 18 May 2026 00:00:00 GMT

Portable radio work has a lot in common with developer tooling: the workflow has to survive imperfect conditions, tired operators, and small mistakes.

This note is a marker for the kind of radio content that belongs here: concrete setup notes, what changed, what failed, and what should be repeated next time.

Rust CLI defaults and configuration boundaries

Sun, 10 May 2026 00:00:00 GMT

Good command-line tools need defaults that work for the common path and escape hatches that do not turn every invocation into configuration archaeology.

This is the kind of software note this site should support: a focused observation with enough context to be useful later.