Wires in the Rigging: Junior and the SunCat, a Perfect Match?

As shown in the precedent article, the deployment of a wire vertical antenna from a boat, using a freestanding S9 14-foot telescopic mast, is only feasible in the absence of wind and waves. The S9 mast is heavy and friction alone will not hold its sections in position when it is continuously rocked. Rain is also a problem, as water sips inside the joints between the different sections facilitating their collapse. Hence, if the W3EDP Jr. is to be used “Portable Afloat”, an alternative method needs to be found for its deployment onboard.

The Com-Pac SunCat 17 is a unique design with some very special characteristics. It is a “gaff-rig” sailboat with a very short and hinged mast (as required for easy hoisting and also for towing on a trailer). From deck to top the mast measures 18.5 Ft.. The LOA (length over all of the boat) is about the same: 17.3 Ft., and the mast stub sits on deck 15 Ft. from the stern and 2.3 Ft. from the bow.


Now, this is amazing: after all the geometry is done and the arithmetic is cleared, the two lines joining the top of the mast to the bow and the stern added together make for 41 Ft., i.e. , one foot less than the length of the W3EDP Jr.! It is as if Clark Mills had designed the SunCat 17 for the exact purpose of deploying the W3EDP Jr. as an inverted Vee…

The angle of the antenna at the top of the mast would be close to 46º. However, it is difficult to predict the effect that the proximity of the metal in the rigging (i.e., the aluminum in the spars and the steel in the standing rigging) might have on the efficiency of the antenna. One way to provide some evidence in this regard is to give it a try.

The tip of the W3EDP was attached with a nylon cord to the bow of the vessel. A small pulley was attached to the top fitting at the mast and a thin double line with a small loop was rigged along the side of the mast. The antenna wire was threaded through the loop and its mid-portion hoisted to the top of the mast. The twin line portion was left to rest on the wood of the “gallows” at the stern (i.e., the support for both the boom and the mast when trailering) and then redirected underneath it to the 4:1 unun on the aft cushion of the cockpit. (As before, the unun was connected to the Elecraft KX3 via a toroidal choke). With this arrangement some interesting QRP contacts were made:

LZ1MS (CW, 15m, LZ1)
HI8CSS (RTTY, 20m, HI8)
AA2FM/4 (RTTY, 20m, FL)
W5ZR (CW, 20m, LA)

However, the overall impression is that although receiving may be mostly unimpaired, the W3EDP Jr. as a V on board the boat may have a reduced transmitting efficiency. This may be due to the proximity of the metal rigging or the higher take-off angle, typical of inverted “V” antennas.

The following curves show the impedance and the SWR as function of the frequency for most of the HF portion of the spectrum, as measured with the mini-VNA analyzer:


And here are the SWR values as determined for each band (at the center of each CW portion) read directly from the Elecraft KX3 with the ATU bypassed or after tuning:

Band (m)

w/Unun, no ATU w/Unun, w/ATU


4.6:1 1.4:1











5.8:1 1.0:1


7.7:1 1.0:1




40 8.9:1



5.5:1 1.0:1




A Single Gybe to Horaceville for a Night at Anchor in the Moonlight

On August 19 the forecast seemed appropriate for another sailing “expedition” to Horaceville. The plan was to spend the night at anchor in Pinhey’s cove under the quasi-full Moon (the “Sturgeon” full Moon had been August 18).

I parked the car at Dick Bell Park early in the morning, but was careful to leave on the dashboard a special permit from the Sailing Club, as the entire parking area fills-up later in the day with the vehicles of avid Pokemon Go hunters.

The wind was steady at 5-10 Kts. from the SE, which allowed for a single-tack gybe all the way to Pinhey Point. With the Pinhey mansion on sight, a decision had to be made as  whether to anchor at the Point or proceed further up-river to Baskin Beach, likely the anchorage best protected from SE swells. I decided in favour of Pinhey’s cove.

As many times before I put my trust on the Belgian 5Kg Bruce anchor and its chain rode. I dropped the anchor in 6-7 Ft. with plenty of scope. The swell was strong due to the long fetch all the way from Britannia and to the wind blowing against the current.

The activities of the day included a refreshing swim from the boat and rowing the dinghy around the point, which in spite of the many signs indicating that this is a “No Public Access Ecologically Fragile Area”, is regularly visited by humans on foot, as witnessed by the many inukshuk “sculptures” at the tip of the Point.


Then it was time to do some radio “afloat” and try the W3EDP Jr. antenna from the boat. The S9 14-meter long telescopic pole was deployed and secured at the stern. With the W3EDP Jr. thus hoisted quasi-vertically, its twinlead end was connected to the KX3 via the 4:1 unun using a toroidal choke. Not surprisingly given current propagation conditions, signals were heard only in 20m. Contacts were made with Karoly HA8IB (RX RST 559)  and Michel F5MXQ (also Rx RST 559). Station ZV2016RIO (in celebration of the 2016 Olympics) was heard CQ-ing and making contacts, but it never acknowledged my call. More signals were heard, but suddenly the movement of the boat caused the telescopic pole to come collapsing down. Next I will explore deploying the W3EDP Jr. as an inverted “V” using only the spars in the rigging.


Dinner was clam chowder warmed up while witnessing the sun going down behind the trees. The moon was already up and almost full and remained on sight the entire night.

During the night the boat pitched and rocked while swinging at the anchor, with the waves occasionally finding the resonance of the hull and causing the bow to leap higher. The wind never seemed to subside and the rocking brought memories of a momentous dash from Alert Bay to Heck Seamount, just to experience the Pacific and use Celestial Navigation offshore with no land on sight. That offshore trip had taken place 20 years before during a circumnavigation of Vancouver Island in a Gin-Fizz Jeanneau 37 and the offshore leg had ended with a landing at Hot Springs and a visit to Tofino’s Harbour in the west coast of Vancouver Island.

At dawn, towards the West, the Moon was still above the trees, while the top clouds were being tinted in pink by the Sun, still below the horizon on the other side of the World and the river:

Version 2

In the morning, I rowed the dinghy to land and discovered an old apple tree that provided a primeval albeit frugal breakfast.


Back on the boat it was soon time to weigh anchor (with a big chunk of mud from the bottom on account of the waves that relentlessly pounded the bow the entire night). The now head-wind had not changed. Sassy tacked back for a couple of hours. Then her “iron spinnaker” was called to duty, which it did flawlessly for the rest of the way to the marina.

At the tip of the peer a large group of Go Pokemonners had already gathered for the day (and the night), and I could only imagine what would become of the “Ecologically Fragile Area” if Pokemons started to appear in Pinhey Point.







A 42′ Portable Endfed Multiband HF Antenna with no Wire on the Ground: the “W3EDP Jr.”

The “senior” (full-length) W3EDP, as originally described (QST March 1936, pp. 32-33), is likely the best option for a portable, HF multiband (6m – 160m), tuned antenna not requiring wires on the ground. However, it is 85 Ft. long, often too long for vertical deployment with telescopic poles and not too practical in highly constrained environments (i.e., a small balcony or a small boat). In these instances a half-sized “junior” version might prove more practical. However, the W3EDP Jr. does not seem to have received as much attention or to be as popular as its “senior” relative, which is rather surprising considering its potential characteristics:

  • can be hoisted vertically with a 43 ft. telescopic mast
  • can be tuned to all HF bands down to 80m
  • does not require any wire on the ground

Similar to its bigger relative, it does, however, require a tuner and a 4:1 unun.

It can be viewed as an amputated version of the better known G5RV Jr. and is the same length as the popular S9v43′, itself designed for use from 80m to 6m. In fact, when hoisted vertical the W3EDP Jr. could be consider as an S9v43′ antenna with a single juxtaposed counterpoise cable 1/5th its length (actually, it can be easily converted into a portable version of the S9v43′ by disconnecting the coax shield from its shorter branch  and connecting it instead to one or more counterpoise ground cables).

Since the full-length W3EDP can readily tune 6m – 160m, its half-size version would likely tune up to 80m. Dropping the ability to tune in 160m seems an acceptable trade-off for the advantage of having a shorter multiband end-fed antenna with no counterpoise cables on the ground.

The measures used were exactly one half of those used for the W3EDP: 8.5 Ft. of twin-lead wire with one of its wires soldered to 33.5 Ft. single wire for a total length of 42 Ft.. The materials were the same as those used in the construction of the “Flimsy W3EDP” (https://thewakesileave.wordpress.com/2016/05/22/a-flimsy-w3edp-portable-antenna-la-manquita/).

Here is an initial analysis of the antenna using the mini-VNA analyzer. The plot shows SWR (blue curve) and Z (green curve) values for the 3-30 MHz range of the HF spectrum. The W3EDP Jr. was hoisted to its full length (using the S9 telescopic mast resting on the top branches of a tree) and was being fed via a 4:1 unun.

W3EDP Jr.jpg

Similar to the S9v43′, it resonnates around 5.5 MHz. The SWR for the most part is below 10:1 and the impedance remains within tunable range, being highest for 40m (as expected for an end fed 1/2 HWL wire).

Here are some data taken with the Elecraft KX3 and its internal ATU (all measurements were made in the middle of the CW portion of each band using 1 Watt and the configuration of the antenna and the connection to its twin line portion were the same for each band):



No Unun

No Unun w/ATU



6m 5.8:1 1.0:1 3.1:1


10m 29.9:1 1.2:1 10.9:1 1.0:1
12m 9.9:1 1.0:1 7.7:1 1.0:1
15m 2.2:1 1.0:1 2.4:1 1.0:1
17m 2.0:1 1.0:1 5.5:1 1.0:1
20m 14.0:1 1.1:0 6.9:1 1.0:1
30m 8.9:1 1.0:1 2.7:1 1.0:1
40m 31.9:1 1.0:1 10.3:1 1.2:1
80m 10.4:1 1.0:1 6.9:1 1.0:1
160m 21.0:1 8.5:1 6.9:1


The antenna was also modeled using CocoaNEC-2:

W3EDP Jr. Elevation.jpg

The elevation angle in vertical configuration appears to be low with a single lobe for the low bands and increasing and becoming “hat-shaped” for the higher bands (this and the vagaries of propagation (and sprint participation) may account for the apparent predominance of relatively nearby 20m contacts in the initial trial below).

Field tests in every band will have to wait for better propagation conditions. However, initial results in 40m and 20m seem to confirm the above predictions: here are the initial CW QSOs (QRP) while operating portable during a recent SKCC WES (i.e., with realistic Rx RST reports):




40m 589


K1LEE 40m 559 CT
W3DF 20m 579 MD
KA3OCS 20m 599 VA
NO8C 20m 559 OH
W4AFB 20m 589 FL
KC3DOF 20m 339 PA
K3WW 20m 559 PA
N8KR 20m 559 IN
W9DLN 20m 559 WI
F6HKA 20m 559 F

Although here it has been used in a vertical configuration, similar to the full-length W3EDP it could easily be used in other shapes and configurations,, something not readily evident for the S9v43′.

The W3EDP Jr. is being added to the backpack as another choice for a practical /QRP/P antenna, together with the Alexloop, the PAR EndFedZ’s (mono and tribander) and its “senior” relative: the 85 Ft. W3EDP.


Note added Aug 17 2016: Operating /QRP/P under similar conditions as above, 4 more CW contacts were made, adding two new bands to the list of QRP QSOs credited to the W3EDP Jr.: 15m and 30m:

– OH6NVC Mika, Finland (CW, 20m)
– K6RB Rob, CA (CW, 15m)
– K6DGW Skip,  NV (CW, 15m)
– DL1NKS Stefan,  Germany (CW, 30m)




The NAQP CW with the “Ottawa Valley QRP” Team

The NCJ (National Contest Journal) is a publication of the ARRL. It sponsors two North American (NA) short contests in three different modes (SSB, RTTY or CW): the “NA Sprint” and the “NA QSO Party”. The latter is a 12-hour contest whose CW versions take place in January and August. This are classic events in radio-amateur contesting in North America, with 2016 being their 25th year.

A feature of the NAQP is that it encourages team participation. In the past I have participated just as myself, but this year the Ottawa Valley QRP Society decided to operate as a team. I offered to join and was accepted. I knew I was not going to break any records but being part of a team effort, even in friendly competitions like this one, demands some form of commitment. Hence, I set to try my best.

The rig was the KX3 at 5 Watts. I preferred the KX3 over the ICOM 703 mainly due to the ease for changing CW speed at the turn of a knob (i.e., without the use of a menu). However, the superb receiving and filtering abilities of the KX3 are what really sets it apart from its competition in the shack.

My best choice at hand for an antenna was the long-wire-up-the-tree (end-fed via the LDG RT-100 and with a set of elevated 1/4 WL wires as counterpoise). This “quasi” vertical “random” wire readily tunes from 10m to 80m but not in 160m, so I knew I was going to be missing the opportunity for additional multipliers.

The logging software was the latest version of RumPed running in the MacBook Pro and the key was the Bencher Iambic paddle. RumPed could read the frequency and the mode from the KX3 but did not PTT the rig. For this I did made use of the memories in the KX3, but most exchanges were transmitted manually. Also I search and tune with the VFO knob as I do not use a Panadapter or any of the SDR capabilities of the KX3.

I chose to use my primary callsign (VA3PCJ). VA3PCJ is longer to transmit in CW than VE3DTI. However, the “J” at the end is easier to pick than the “I”. The exchange to be reported was “name and province”: transmitting “JOSE ON” seemed to have caused some confusion. I made sure to correct those who asked but a few may have logged my name as “JOE”, “JOS”, “JOSH” or even “JOHN”.

Roughly, I operated in 20m the first third of the contest, in 40m the second and in 80m the last third. Under current propagation conditions 10m was out of the question. I did hear some faint signals in 15m but at my location the background noise in this band was high. 20m was fine with some QSB though it faded early. 40m was the busiest but the signals had some flutter (an effect of the unusually high Electron Flux?). Towards the end 80m was the only band that remained open.

Here is a map of all the QSOs (always nice to confirm how the range of the contacts diminishes with increasing band wave-length) and their time-line during the 12 hours of the contest:



My logging software (RumPed) estimated my operating time at 5.2 hours. However, I estimate my “sitting time” significantly longer than that.

Fortunately – and as expected – my experienced teammates did much better than myself and my effort ended contributing just 10% of the Team Total Score. I wish it had been more but having surpassed 100 QSOs and logged 35 S&P’s I did not feel too displeased with my performance. Overall, a very enjoyable radio experience, with added camaraderie and time flying-by quite unnoticeably…

3,000 QSL’s in LoTW

Within less than 24 hours of the NAQP CW, over 20 of the participants had confirmed their contact in the “Log of The World” (LoTW). This has had an interesting and immediate consequence: I have just surpassed 3,000 QSL’s in LoTW (imagine if they were QSL cards).

As of today (August 7, 2016), considering all the callsigns with all the prefixes and suffixes that I have used in the nine years that I have been active in radio since being licensed in January 2004 (I was not active in radio 2007, 08 and 09), LoTW shows 3,021 independent confirmations (“QSL records”) from 1,639 different callsigns or stations in 111 DXCC entities out of some 6,000 QSO’s (the QSL number is exact but the number of QSO’s may be slightly overestimated due to occasional duplication of erroneous dupes).

Most of those contacts (~90%) were made after May 2010, when I started operating QRP.


How to extract QSL’s unique callsigns and unique DXCC’s is fairly simple. Here is how I did it:

  • Log into your LoTW account
  • Go to “Your QSO’s”
  • On the left margin click on “Download Report”
  • Select the desired parameters for the QSL’s to download and click on “Download”
  • An .adif file named”lotwreport.adi” will be downloaded to your computer
  • Input this file in an empty log using any logging software (I use RumPed)
  • Highlight the entire data and copy and paste it into an empty Excel spreadsheet
  • Use Excel menus and functions (mainly “sort” and “advanced filtering”) to extract unique records and manipulate the QSL data in any way you wish.

Here is a chart with the 3,021 LoTW QSL’s plotted per DXCC entity in a semi-log bar-graph (click on it to enlarge):









A Frustrating QRP Weekend in Hell and Chasing Bumblebees

A deep QSB was noticeable in the two HF bands that seemed open during this last weekend (40m and 20m): a signal would be 559-579 in one exchange and disappear in the next. I imagine that the very low SFI (071) and relative high SW (~400) and EF (~900) may have accounted for this. However, the background noise was low and the Geomagneitic Field remained “Very Quiet” (K=1).

On Saturday I set off to work “All Africa, Asia, Europe and Oceania” in the Feld Hell Club WAAAEO Sprint… My rig at the QTH was the Icom 703 at 5W; the antenna was the end-fed long-wire and the T-60 IBM was communicating with the Icom via Fldigi and the Signalink USB. Half-a dozen stations were heard and, quite literally, seen in the bands, none of which was from outside the North American continent. The only one to acknowledge my QRP signal was Steve KI4GGJ/P (FH 5120) operating portable from Alabama (EM62).

On Sunday, for the ARS FOBB and as BB Nr. 91, I returned to the front gardens of the Science and Technology Museum to operate portable with the KX3 at 5W and the PAR EndFedz “Trail-Friendly” Tribander in a “number one” (an asymmetric inverted “V”) configuration. The QSB seemed even deeper than on Saturday and I was barely able to contact a couple of stations. However, the reward was to log two QRP-QRP CW contacts: one with Steve WA2AF in NJ (in 20m) and another one in 40m with N3AQC, Bumblebee Nr. 0 (zero), the official FOBB station of the NAQCC, operated from the Kittanning Pennsylvania Community Park by John K3WWP and Mike KC2EGL.

The following composite shows the Feld Hell QSO captured from the Fldigi window and the conditions during the FOBB on Sunday:


DX QSOs and a NPOTA while /P from Rockland, ON

Yesterday, for less than an hour before dinner and a a few minutes after, I was able to operate /P from Rockland, Ontario (FN25im). The conditions for HF were dim as have been lately. However the K-index was low and the geomagnetic field was quiet. From Rockland, ON, the background noise has been usually less than from my QTH in Ottawa and yesterday there was almost no noise in the 20m band.

I operated the KX3 at 10W (not quite QRP…) and the antenna was the 20m PAR EndFedZ (held up by the DK9SQ 10-meter mast). The PS was a P-BOX battery and the key was the Palm Radio single paddle. This was the log:

21:33  14006.0        UA3KW            Vlad                 Voronezh
21:39  14010.0        EA8TL               Jorge               Tenerife (IOTA AF-004)
21:44  14033.0        CT8/W6NY      Thurston        Azores (IOTA EU-149)
21:53  14033.0         DR160TESLA  Special Stn.   Erbach
21:58  14033.0         S59AA              Franc               Ljubljana
23:28  14025.5         N0AC/M          Bill                   NPOTA MN28, Colorado

Also heard: UT5UJO, EA7GX and IT9SSI

This was a nice fishing harvest: Two IOTA’s in two different continents (CT8 was a first in my logs), Russia, Slovenia, a German station commemorating the 160th birthday of Nikola Tesla and a portable to mobile contact with a /M station operating from NPOTA MN28 (Florissant Fossil Beds Park in Teller County, Colorado).

The site of operation in Rockland is in the SW part of town, which is not particularly high ground. However, about 800m towards the S there is a large promontory ~100m high, which likely contributed to the name of the town and whose role, if any, in favouring N-E over S-W HF communications can only be speculated upon.