Friday, November 27, 2020

Reflection on the Sweepstakes Contest

With the year's premier event -- CQ WW CW -- rapidly approaching I thought it worthwhile to say a few brief words about ARRL Sweepstakes. I have some warm feelings about these contests because of the central role they played in my early contesting career going back 45 years ago when I was a teenager and a new ham. I don't feel so enthusiastic about it now. Times change.

Once again I entered the CW weekend in the QRP category. I do this for the experience of using skill over brawn to put contacts in the log. In contrast I will be burning up the watts in CQ WW. I resurrected the 3 db attenuator so that the FTdx5000 could be used at 5 watts. Luckily the resistor bundle was still around, having been stored for just such an eventuality. I soldered it back into the since repurposed box and it was ready to go.

Sunday can be a drag in Sweepstakes because there are a limited number of stations to work. It is less so with QRP since, due to the lower rate, there are still stations to work, mostly those that haven't yet been able to pull my tiny signal out of the noise. Chasing rare multipliers is more thrilling with a power handicap. I am one of those contesters who likes to play both ends of the power range -- QRP and QRO -- each with its unique challenges and benefits.

I operated SO2R as before to maximize the use of my many antennas. QRP is a good introduction to SO2R since the rates are lower. Stress is minimized as you learn the ropes. Despite this it does get quite boring by Sunday evening when I can be CQ'ing on two bands and the QSOs still come several minutes apart. Indeed, I quit a little early because my rate had slowed so much that I probably only sacrificed 3 or 4 contacts.

New this year was the SO2R Mini to switch rig audio. Keyboard control of receiver audio was a great convenience compared to the manual switch I used for my first foray into SO2R. The kit is inexpensive and takes care of the most important switching responsibilities, while eschewing sophisticated features you might never use. I recommend it.

Although I did 15% better this year than the last two years I was squashed by the QRP big guns. Last year it was N0NI in Iowa. Unfortunately that super-station was flattened by a major storm and was absent this year. From reports they are rebuilding and will be back in action before too long.

This year I was outgunned by W2GD and VY2ZM despite a capable station. As with real estate the secret is location, location, location; rospects in Sweepstakes are very location dependent. For example, as I type these words I am hearing W2 stations working Europeans on 10 meters that I cannot hear. Even without an ongoing geomagnetic storm, 500 km further north it is a different world with respect to propagation. 

Of course W2GD is a top notch contester and, to my knowledge, has good antennas. I have no way of knowing whether the problem was me, my northerly locale or antennas. A few decibels can make a big difference at the QRP level. It just seemed there weren't any more contacts to be had. Would a different strategy have helped? I am doubtful. It is more likely that I am a victim of circumstance.

The over 1000 QSO log from VY2ZM is spectacular. Here we have a top notch station and operator in a location well placed to work the large midwest population on 20 meters, where many casual operators congregate, and from a section that, for the first time, is a new multiplier and he had little competition. Back in the beginning I had a similar advantage from VE4. However there were no assisted classes in or the supporting technology in the 70s so I still had to work hard to attract attention.

I did have technical difficulties leading up to the start of Sweepstakes. Since I hadn't done SO2R in a while I was unaware that the remove antenna switch had a problem. Relays were misbehaving on the second radio/operating position. Less than 1 hour before the contest started I discovered that the problem was earwigs building a nest inside the DB25 connectors to the switch. Moist eggs are a conductance path between the connector pins. Once they were evicted the problem was eliminated. 

Earwigs have been a big problem this year in all my close to the ground electronics at the towers, antennas and control systems. It takes a very small hole to permit them entry. They, of course, love weatherproof enclosures.

Unlike the CW weekend I had no interest in the SSB version of the contest. I got on 80 meters briefly Sunday evening and made 28 contacts in 13 minutes. When stations stopped calling I quit. The motivation to continue "for fun" wasn't there. Phone contests don't attract me the way they once did unless I do it as part of a multi-op. But that's me and not the fault of the contest.

As everyone knows Sweepstakes is approaching a crisis. Participants are old and the contest risks passing as they pass. There are more new hams to be found on SSB than CW though not enough to keep the contest going for the long term. That's a shame. 

Sweepstakes is not an indicator that this is the end of contesting as a global sport. Contests will be around for as long as amateur radio is around. But those contests won't be Sweepstakes. 

 It'll be a very different story in CQ WW CW this weekend. For this contest I will be running a kilowatt and the new 15 and 20 meter stacks are online and working. And, boy, do they work! More about that in a future article.

Sunday, November 22, 2020

Small Pile Ups

Many hams have noted that activity on the bands changed as the pandemic continued and lock downs came in waves in countries around the world. My own antenna plans were altered in the spring when it became impossible to bring a crew together. Eventually we did resume but the lost time could not be recovered. Work on my new antennas has extended into the cold of late autumn.

Travel restrictions put a brake on almost all DXpeditions. By mid-spring there were almost no new countries to be had by ardent DXers. That seemed to bring a lull to the HF bands that is unusual. Ordinary person-to-person QSOs continued as before, for the most part, so that non-DXers didn't see much that was different. For non-DXers it was an opportunity to be more active, on the air or concentrating on projects indoors.

Despite the increased leisure time there has been a psychological toll. We have the time but we may not have the same degree of motivation. With the worries over family, income and life as we've known it we haven't used the time as productively as we might. I noticed a similar effect after I retired: I have all the time in the world, which removes urgency and often causes a loss of focus on doing things now. An external spur is needed, whether it be an upcoming contest or a DXpedition to finish antenna projects.

The lack of DXpeditions removed a major spur. Outside of the summer sporadic E season on 6 meters my own on-air activity has lagged. Summer is also the low season for major contests. I did what I could on the ground to progress antenna projects but for the most part my attention was diverted to other pursuits.

When several DXpeditions appeared in recent weeks I was surprised at the small the pile ups. You would think that after months of no rare DX the pile ups would be fierce. Lethargy does not evaporate instantly. For a time I could get through the pile ups very quickly running low power since there was little competition. 

The small pile ups are becoming bigger as DXers are drawn back to the shack. There is a stark change between the number of callers to 7Q6M than to the more recent 7Q7RU DXpedition. The first was easy to work and the latter more difficult. The pile ups are quite a bit bigger and the competition more fierce. That and their QRN made working them on 80 and 160 a challenge.

Although DX chasing has been subdued this year it has not been difficult for me to drum up activity with just a CQ. With cycle 25 beginning to flex its muscles I find it quite easy to draw a lot of callers on 15 meters even without high power. Contests have also seen plenty of action. Contest organizers have noticed a sharp year over year increase in entries. Small contests have also been the beneficiaries.

Clearly there are things that will spur activity and overcome our pandemic lethargy. Perhaps it's the novelty of a higher MUF, the lure of contest competition or DX on 6 meters. Whatever it is I am enjoying the activity when it occurs. After all, what's the point of building all these towers and antennas if I can't put them to good use? So I get on when I can. 

Soon I'll have more time to operate. The weather is turning cold and the year's antenna projects are coming to an end, and mostly to a successful end. With the end of the pandemic in sight we should enjoy the winter season and the coming major contests. DXpeditions will not be common until well into 2021 so jump in and pursue the ones that do appear during the interim.

A little luck with the weather in the coming days should see me well equipped for CQ WW CW next weekend. The new 15 and 20 meter stacks should be largely operational and the new 160 meter antenna appears to be working very well. 

Helped by new sunspots expect the largest turnout ever for this major contest. Let's see some big pile ups on the DX that is able to get on and pursue those points, multipliers and, for the casual contester, new band-countries. It should be a lot of fun. It'll also help us to shake free of the pandemic blues.

Sunday, November 15, 2020

160 Meter Shunt Fed Tower

I am trying something different for 160 meters this winter. The T-top wire vertical hung from the 150' tower the past few years is a great antenna but not without its negatives attributes. These include:

  • Potential for interactions with yagis, of which I now have more
  • Tower interactions affecting directivity and efficiency, up to 3 or 5 db in some directions and down to -2 db, respectively
  • Low radiation resistance that reduces efficiency by, perhaps, -1 db

A few possible alternatives were reviewed in an article last month. My intention is to do the best I can within the constraints I have. My constraints are atypical in that I have two towers that are approximately λ/4 tall on 160 meters, and that makes many things possible that are unattainable for most hams.

After more analysis, research and modelling I decided to shunt feed the 140' tower. Shunt feeding a tower with yagis on top has long been a popular choice for hams with limited space and height since it exploits the modest height tower they have, very often with relatively efficient top loading provided by a tri-band yagi on top. 

In my case the electrical length of the tower is far greater than λ/4 and while that's impressive and potentially very effective it does introduce a few unique problems. These I knew in advance because they are covered in depth in ON4UN's Low-Band DXing book. 

This is a timely reflection on the book and John DeVoldere's remarkable legacy since he died on November 9. If you have an interest in the low bands you must have this book on your bookshelf. The antenna described in this article behaves almost exactly according to the chart and equations to be found in that book. It is the marriage of theory with the practical that makes the book a treasure.

A shunt fed tower is pretty simple: run a wire parallel to the tower, attach it to the tower where the resistance part of the impedance is 50 Ω and use one (gamma match) or two (omega match) capacitors to tune out the net inductive reactance at the base of the gamma rod (or wire). 

It is straight-forward to model the antenna in NEC2 (I use EZNEC), with a few cautions:

  • Align the segments of the tower and gamma rod (wire)
  • Use MININEC ground and place resistance loads in a one segment wire connected to ground to emulate the loss due to the tower ground and radial system
  • Rather than include yagis in the model it is easier and equally effective to measure the electrical length of the tower and make the tower that height in the model
  • The short wires connecting the tower and gamma rod should be one segment
  • Put the gamma capacitor and source at the bottom of the gamma rod
  • Connect the bottoms of the gamma rod and tower with a one segment wire, and do connect them in the built antenna

The guidelines on how to proceed are addressed in ON4UN's book. Recall that I measured the resonant frequency of the tower to be in the vicinity of 1200 kHz. This is an electrical length of about 62 meters, which is 135° or ⅜λ at the low end of top band. Although this is quite a bit more than 90° (λ/4) the pattern does not sprout any lobes at high elevation angles.

I was successful replicate the tabulated data from the book with the model I developed in EZNEC. The initial height estimate for tapping the tower with a wire spaced 1.5 meters is about 25 meters. Perhaps the largest uncertainty is the diameter of the tower which, as a lattice structure, is typically equivalent to a wire somewhat less than the tower face dimension and depends on the structural details. Gamma wire diameter and insulation are other variables with an effect.

For simplicity in getting to a first measurement I used the boom of the lower 5-element 20 meter yagi that is ~22 meters high. A clamp holds the wire in position along the boom and the wire is bonded to the tower metal by scraping away the paint around a hole in the adjacent tower girt. I did not rely on continuity between boom and tower because there's a layer of paint between them. I scraped the paint off the galvanizing at the attachment point.

It was no fun to lean out from the tower on a face with no climbing horizontals to clamp the wire out 1.5 meters (5') from the tower centre. I couldn't get it quite that far without additional acrobatics, an unwarranted effort since the tap would almost certainly have to be moved.

That was indeed the case. After cancelling the inductive reactance with a capacitor (low voltage capacitors are suitable for analyzer measurement) the resistance, at 122 Ω it was far wide of the mark. Part of that was due to the mess of clip leads but no where near 150% error's worth. Apart from the high impedance the antenna seemed to work okay. I am using the same set of 8 × 30 meter radials from my earlier antenna.

It was a windy during the first trials and the periodic billowing of the long gamma wire during the analyzer's scan made for peculiarly wavy SWR curves. It is worthwhile to place insulating arms at several points on the tower to hold the wire in place to avoid impedance swings and wire fatigue. I have not yet done so since with the antenna incomplete the positions were subject to change.

I returned to the computer to re-calibrate the model based on what I learned. It turned out that the error was not as bad as I feared. I estimated that the 50 Ω tap could be found 3 to 5 meters lower. I split the difference and used the available tower girt about 3.5 meters lower (~60' above ground).

I built an extendible arm to suspend the gamma wire from the tap point. A length of angle steel and an ABS pipe are joined with hose clamps. The outer end of the ABS pipe is notched to hold the wire in place. The bolt that connects the arm to the tower doubles as the electrical bond to the tower. The end of the wire is a tinned loop that fits on the bolt and minimizes galvanic corrosion when sandwiched within the galvanized hardware.

Minimal acrobatics are required to adjust the telescoping arm. To change the distance I disconnect the wire on the ground, loosen the hose clamps and slide the pipe to its new position. A retractable steel tape measure can be extended out to the end of the pipe to set the wire separation.

The separation between the gamma wire and tower was not constant. The wire was farther at the bottom than at the top. Although this is not ordinarily discussed in the literature there is no reason for the two to be exactly parallel. All that does is improve predictability. The effect of the angled wire is that of a parallel wire with a separation between those of the top and bottom.

The reasons for the non-parallel wire were to avoid more tower acrobatics and because the radial hub is not easily moved. I only moved the radial hub, along with all 8 radials, when the antenna was approaching completion. The general layout of the antenna base is shown in the photo below.

Pictured are the final positions of the components, after all the experiments and final adjustment. A wood stake has a wood platform for the plastic container with the gamma capacitor and an insulator for the gamma wire.  This keeps the capacitor and feed point off the ground and safe from snow, puddles and even small animals.

This is a high voltage point so beware of the presence of incidental conductors. Insulated wire is recommended for the vertical gamma rod (above the capacitor). A wire snakes along the ground between the tower lightning ground rod and the radial hub (ABS pipe next to the stake) to prevent the high loss of a return path via the soil. A short length of RG213 runs from the main Heliax transmission line to the feed point adjacent to the capacitor. The Heliax is buried in a trench along with all the other transmission lines and control cables.

It is important to check for continuity between the gamma wire and the tower ground. Despite the many splice bolts between tower sections it is possible for paint to insulate tower sections. A friend in the business once explained that the metal ridges I found on several of my used commercial tower sections are evidence of the tower having been the radiator of an AM broadcast station. After installation all the legs at each section splice are spot welded to ensure continuity and no resistance loss due to high RF currents.

Returning to the tuning discussion I have to say that we got lucky with the second trial using the lower tower tap. After tuning out the inductive reactance with the capacitor the impedance was almost exactly 50 Ω. This is what I call hitting the bullseye! Unfortunately after tidying the site -- changing connecting wire lengths and location of the radial hub -- the impedance dropped to 38 Ω. Further adjustment of the gamma wire got us back to 50 Ω.

I drew a diagram of the adjustment process for the gamma wire. Coarse tuning to get to 50 Ω is done by changing the height of the tower tap. Higher is higher impedance and lower is lower impedance. Once you're close -- 35 Ω to 65 Ω range -- you can fine tune the impedance by changing the separation of the wire and tower. Closer is lower and farther is higher impedance. There is an effect on bandwidth in different combinations of the two, however it is small enough that I solely focussed on achieving a match.

Since the radials are difficult to move you can either run a wire from the bottom to the radial hub or fix the gamma wire position at the bottom and extend or retract the tower arm. I performed trials using both methods. Do whatever works best in your situation. The extra wire on the ground for the former method changes the tuning and that can be tuned out with the gamma capacitor.

I centred the match at 1840 kHz. This is 10 kHz higher than for general DXing, for which most top band operators aim for 1830 kHz. Because contests are a primary interest of mine and the activity in top band contests can spread quite a lot higher in the band I centred the antenna higher. Alternatively the antenna centre can be moved higher just for contests, easily accomplished in a few minutes by adjusting the gamma capacitor.

The 2:1 SWR bandwidth is a little more than 75 kHz, as predicted in ON4UN's book and in the software model. This is narrow though typical of an electrically long shunt fed tower. Better than 100 kHz bandwidth can be had with a wire cage for the gamma wire. I'll consider doing that next year if I decide to keep this antenna.

With the gamma capacitor value and range determined once the 50 Ω impedance was found for the final antenna layout I proceeded to select a gamma capacitor from my ample junk box. The gamma capacitor has a high voltage across it due to the large inductive reactance it must cancel. By large I mean thousands of volts. The greater the inductive reactance the smaller the capacitor value required to cancel it and the higher the voltage it must withstand. For the 145 pf capacitor measured and modelled it is more than 2500 volts for 1000 watts.

A variable capacitor with the required range and voltage rating is large. It is better and cheaper to use a small value variable capacitor in parallel with a high voltage fixed capacitor. The smaller range of the variable capacitor is also easier to tune and this is important because the capacitor value is critical: a tiny change has a large impact on the feed point impedance.

My first attempt was a split stator capacitor with 50 pf per section in parallel with a 100 pf ceramic doorknob capacitor. The approximate range is 100 to 200 pf, plus stray capacitance in the wires connecting it to the gamma wire and feed point. 

The plate spacing is not enough to withstand 3000 volts, which is the minimum recommended in this application. Air requires at least 0.12" to survive 3000 volts, and that is for dry air with no margin for humidity and particulate contaminants. To determine the actual flash over voltage I followed the advice from an old joke:

To determine the load rating of a bridge drive heavier and heavier trucks over it until the bridge collapses. Rebuild the bridge as before and put up a sign with the load limit as the weight of the last truck that made it safely across.

I gradually increased power until the capacitor arced. This occurred around 650 watts. I had no difficulty operating at 600 watts. The load data in EZNEC tells the story.

Load #3 is the capacitor. At 600 watts it sees a little over 2000 volts. It is reasonable to assume from this (and a measurement of the plate spacing) that the capacitor is rated for 2000 volts. The box the capacitor comes in does not specify the voltage rating.

I rebuilt the capacitor by adding a 30 pf ceramic doorknob and wiring the split stators in series. The range of the assembly is now 130 to 155 pf. Recall that series capacitors of equal value result in a capacitance half that of each, which is 25 pf for the two 50 pf section. These act as a voltage divider with a combined rating of 4000 volts. Should you use different value capacitors in series the voltage division will be unequal and you may not eliminate arcing.

I added less than 10 pf to the range with a few inches of scrap RG213, which is 2.1 pf/inch and has a voltage rating of at least 3000 volts. The braid is trimmed near the edges to prevent arcing through the air.

For weather protection I put the capacitor in a margarine container. There are holes for the external connections and bottom holes to weep water and moisture. A scrap piece of wood on top reduces UV damage and, with a stone, keeps it from blowing away in a high wind!

Returning to the load data you'll see the 3 resistance loads for the radial system and the lightning grounds of the shunt fed tower and the similar tower 60 meters away. Ground loss has been reduced compared to my previous wire vertical and that of a wire vertical adjacent to one tower or between the two towers. 

It is difficult to know whether the efficiency increase is true in practice. For now all I can say is that the antenna is working very well and is certainly at least the equal of my previous winter 160 meter antenna. Cumulative experience over the coming winter's DXing and contests will provide additional performance data.

I'll close with noting a couple of issues, one that I ran into and one that may arise. After tuning the antenna and testing it in the shack I found that the resonant frequency shifted from 1840 to 1855 kHz. That may not seem a lot but for a narrow bandwidth antenna on 160 meters it is an almost 1% change that pushed the SWR over 2 at the bottom of the band. Scale this to 20 meters and it's equivalent to shifting resonance from 14.100 MHz to 14.220 MHz. Most hams would notice that!

The cause was the connection to the main transmission line. It is buried 20 cm below grade and behaves as an additional radial, one that is longer than the other 8. There is no common mode choke on the transmission line to defeat this behaviour. The gamma match provides a modest amount of common mode attenuation and the ground around the Heliax provides a degree of common mode choking because, by being buried, the common mode field loses strength along the way. However this choking effect is not enough to prevent it acting as a radial. With a small radial count there can be a current imbalance due to its different length and depth but imbalance is almost certain in any case with just 8 radials.

To compensate for the shift I tuned the antenna to resonance at 1825 kHz. With the transmission line attached the resonance shifted 15 kHz to 1840 kHz, which is where I want it.

The other concern, one common with shunt fed towers, is RF into electronics on the tower and heating of common mode chokes on yagis. At a minimum all control lines requires bypass capacitor or RF chokes, and even then there may be effects. I will be watching for problems with my home brew stack switches when they are installed in the next week or two. Since I use air core coax chokes there are no ferrite cores to overheat and damage. At least that's the theory. Again, I will have to see what happens.

Most often these problems do not manifest themselves. When they do occur they can be difficult to cure. Opting for a wire vertical adjacent to the tower rather than shunt feeding the tower itself can help but not as much as you might think since the coupled energy is almost as great as in the wire itself. More separation is needed to lower the risk.

There is no easy way to predict which installations will suffer from these problems. I will try it and see and take action if necessary. One reason I chose the tower with the 15 and 20 meter stacks is that it less likely I'll be on those bands at the same time as 160 meters than 40 or 80 meters.

Saturday, November 7, 2020

160 Meter Antenna Arcing

A few weeks ago my "small" 160 meter antenna malfunctioned. At first it was only with a kilowatt, then it deteriorated until I was only able to run QRP without the problem occurring. It pretty well took me off top band other than playing around as a QRP entry in the Stew Perry contest in October.

The antenna is the 160 meter mode of my 80 meter vertical yagi. It's base matched by an L-network that is switched in when 160 meters is selected. Although only half height the antenna exploits the extensive radial system of the 80 meter array. In comparative testing it appears to be -6 db relative to my usual full size wire vertical. While not a wonderful antenna it allows me to operate year-round on top band. The full size wire vertical must be removed in spring to keep radial wires are out of the way of farm equipment used for haying. 

The first symptom of trouble was erratic SWR changes while transmitting with high power. At lower power, although the SWR was more steady, the rig detected a problem and reduced power. A low volume sizzle was heard in the headphones. This was my first hint of arcing.

Troubleshooting the antenna was . Since it is so far from the shack I could not do a transmit test while also being present at the antenna. Visual inspections uncovered nothing amiss.

My suspicion first fell on the L-network capacitors. There is a 2000 pf vintage transmitting mica in parallel with several 100 pf 1 kV low loss capacitors. Old style mica high current capacitors are known to occasionally suffer internal connection failures after several decades. Although the small ceramic capacitors are calculated (current, voltage, Q) as more than adequate to the application and have not failed in other networks I've built, well, you never know.

I first substituted an identical 2000 pf mica capacitor. The problem persisted. Cutting the 100 pf ceramic capacitors out of the circuit again had no effect. I turned elsewhere to find the problem.

The DPDT relay that switches the tower connection between the 80 meter L-networks and the 160 meter L-network has its 8 A contacts wired in parallel for greater current capacity. I disconnected the 80 meter connection and shorted the relay contacts with a wire. Again, no change. I didn't really expect the problem to be the relay since it continued to work properly on receive and low power which would not be the case had the contacts been ablated by arcing.

My attention next shifted to the L-network coil. It is large, to maximize Q, and is exposed to the weather. It is wound with weathered THHN wire I scavenged from an old antenna. The high voltage across its terminals could possibly arc through the weathered wire insulation to an adjacent turn or to a nearby conductor. A close inspection showed no evidence of arcing.

Having failed to diagnose the problem visually or by component substitution and bypass I asked a friend to help. We communicated by handheld. He transmitted at power just high enough to elicit the arcing while I poked around the antenna. Despite the audible sizzle of arcing it took a few minutes to pinpoint the location. It can be difficult to localize a quiet sizzle outdoors with components of the antenna bound together and hidden by weather covers.

The problem was not at all what I expected. The arc was from the tower to a few of the 12 VDC control wires. There is a bundle of Cat5 cables that connect the central control box to the 80 meter parasitic elements and to the shack. The cables are buried and are only exposed at the tower base.

When I completed the 80 meter array, long before adding the 160 meter L-network, I made a temporary weather cover for the control cables. There's a lot of them! I bundled and wrapped them in a black polyethylene sheet that was handy. I taped the wrap closed and taped that to the lowest tower brace.

On 80 meters this isn't a problem because the tower base is at a low RF potential, where current is maximum and voltage is minimum. The base of a vertical is like the centre of a dipole. This is not the case on 160 meters where the relatively large transformation ratio of the 160 meter L-network is a pussycat at the 50 Ω port but a monster with bloody fangs where it connects to the antenna.

TLW shows 1200 volts across the coil, and therefore on the antenna port, with 1000 watts applied. The control cables were an easily accessible path to RF ground through the inadequate layers of thin plastic. It punched a hole through the polyethylene sheet (left side, above) and then burned through the thin plastic covers on the two crimp connectors closest to the tower brace.

Therein lies the problem with temporary solutions: you fail to follow up to replace it with a permanent solution. This is a lesson I never seem to learn. For those following the blog you'll know I have far too many projects on the go. I forget or I choose to focus on other things. Small maintenance items like this tend to be put off and finally forgotten. Sometimes these oversights come back to bite me.

A more robust weather cover for the control cables is now in place. It's simply a short length of PVC pipe with the cable bundles stuffed inside and taped top and bottom to protect against rain and insects, respectively. I cleaned and painted over the scorched area of the tower brace (visible in the picture). This time it ought to last.

RF is weird. You might expect that high RF power leaking into the control wires would damage the control system components. It didn't. All the 80 and 160 meter modes of the array continue to perform as before. It probably helped that all the control lines entering and exiting the central control box have RF chokes. However the parasitic element boxes do not have chokes on their end of the control lines since they did not appear to be necessary. Exactly where the RF decided to travel is unclear. 

Soon I'll have my winter 160 meter antenna ready and this lesser antenna will lie idle until next summer. Among the alternatives available to me, I'm trying something different this year for 160 meters. Its RF arc potential is even greater than for the antenna just discussed. More on the new antenna in a future article.

Wednesday, November 4, 2020

15 Meter Stack Switch

The 15 meter stack switch that I recently completed is almost a duplicate of the one for 20 meters. Both are single band, utilizing an L-network. There are a few differences that I'll describe in this article. For background on the design, design objectives, construction and testing you must read that article.

The L-network is very similar, only the L and C value are different. As before it is a low pass design. For 15 and 10 meters a high pass design may be preferable in a multi-operator or SO2R station, however I took this route for convenience in the choice of components. In any case, adjacent band attenuation is modest (no better than -10 db), so the benefit is not critical.

The major change to the 15 meter version is the layout of the input port and relays surrounding the L-network. There was stray inductance in the 20 meter stack switch that required a significant reduction of the coil inductance. The effectively longer lead length -- wire distance from port to port via the relays -- was responsible. At a higher frequency and lower inductance a better layout is required.

To arrive at the optimum layout I spent an hour with pen and paper to see how different component positions affected the lengths and paths of interconnecting wires. For the layout I settled on the two ports and relays for the yagis are unchanged. The input port was moved to the side of the box with the L-network relays beside it, between the yagi port relays and the barrier connector for the control wires. As before the L-network is situated above the other components, supported by conductive spacers and a terminal strip. This style of layering keeps wires short.

For comparison the 20 meter stack switch is shown on the left. On the right is the complete 15 meter stack and in the centre it is shown without the L-network so that the layout is clearly visible. The L-network connects between the NC (normally closed) pin of the input port SPDT relay and the common pin of the other. 

The (unpowered) default position is BIP (both in phase) so the L-network is in the circuit and connected to both yagi ports. For upper and lower yagi selections the L-network is bypassed and the input port is directly connected to the upper or lower port.

The L-network has a 150 pf shunt on the 50 Ω (input) port and a series 0.19 μH coil to the 25 Ω port. The output port connects the two nominal 50 Ω antennas in parallel. After testing with a thin wire coil I wound the permanent coil from AWG 12 wire. Its inductance is calculated as approximately 0.14 μH, which implies 0.05 μH of stray inductance. The stray inductance is equivalent to ~2" (5 cm) leads, and that visually corresponds well with the layout. The revised layout proved its effectiveness by halving the stray inductance.

Testing was done with my VNWA3 by DG8SAQ. With this instrument I could accurately measure insertion loss, port imbalance and input SWR. The final smoke test with a kilowatt is done only after the insertion loss is determined to be acceptably low at no more than -0.05 db. This is achieved with high Q coils and capacitors to minimize ESR (equivalent series resistance). 

The VNA was recalibrated before measurements were done since I discovered that insertion loss (S21) was at least 0.04 db better than expected. The drift from the original calibration, tiny as it is, is enough to make the stack switch appear to have unity or better gain, and that's impossible. Accuracy is worth the few minutes it takes to re-calibrate.

Let's start with the default BIP position. There was little difficulty adjusting the coil to centre the pass band. The excellent match is indicative of the capacitance value being correct. As for the 20 meter unit one yagi port was connected to the VNA to measure the insertion loss (S21) and the other had a precise 50 Ω load.

There is a small imbalance between the ports. The insertion loss with the yagi ports reversed is -3.08 db. Since the exact value per port should be -3.01 db for a perfect and lossless power division the net loss is very good. As tuned the BIP position is still not too bad on 20 meters despite no attempt to make this a multi-band unit. When selecting either the upper or lower yagi the L-network is bypassed so that the unit can be used on any band.

The insertion loss is effectively nil when one yagi is selected. It shows -0.01 db but that is variable with the measurement trial and port and other times shows as 0 db. If it is -0.01 db that is only 2 watts of loss when transmitting at 1000 watts. As the frequency climbs conductor loss increases due to skin effect. This is evident in the slight upward slope of the blue trace. The stack switch uses a mix of AWG 12 and 18 copper wire, and the relay internal wires are smaller. However, as we'll see later in the "smoke test" none of these parts exhibited measurable heating.

Another frequency effect is the input port SWR. As the frequency climbs the internal wire lengths are a larger fraction of a wavelength and increasingly exhibit transmission line effects. The "dead bug" wiring method I use is beginning to show its limits at 21 MHz. A similar unit for 10 meters would be worse though probably still acceptable. Different techniques are required at VHF and above.

An SWR of 1.07 is quite good and is typically dominated by the yagi impedance which is higher over most of the 15 meter band. To alleviate the problem it is necessary to use coax for the internal connection or PCB with parallel conductors (one at ground potential) to maintain the internal connection at 50 Ω and thereby lower the SWR. This is typical of the construction method in commercial stack switches.

Of course the final step is the smoke test, just as it was for the 20 meter unit. For this test a legal limit signal is passed through the stack switch in BIP mode to determine how much power is dissipated, which is the true indication of the insertion loss. This is done in the shack with both sides of the 2 8 switch connected to the yagi ports and each connected to different 15 meter antennas.

Since the new stacks are not yet connected I use the TH6 and TH7, the same as I did for the 20 meter test. I chose a time when the band was dead so there is no inconvenience to others. I picked a frequency on 15 meters where both have a low and near equal SWR.

Driving the two yagis at full power the coil got warm. Everything else stayed cool to the touch. With one minute at a kilowatt the coil was uncomfortable to touch for more than a second but there was no danger of a burn. Coil Q is not as high as I'd prefer: it is calculated by K6STI's Coil to be ~300. This isn't great but neither is it bad. A higher Q coil would be wider and difficult to fit in the space without causing other proble.

With the smoke test completed I am confident that the 15 and 20 meter stack switches are ready to be installed on the tower. That could happen as soon as a week from now. The weather is improving so there is time to catch up on tower work. 

I am looking forward to having 15 and 20 meter stacks fully operational later this month. All 4 yagis are on the tower, and that is a major step forward. More on that in a future article (or two).