2014 Ten Meter Contest

Since moving to the current QTH, it has now become something of a tradition for me to operate in the ARRL 10 Meter Contest. The last few years of the contest have been enjoyable since we're on the peak of the solar cycle, and I don't have to stay up all night to catch stations, meaning I can still sleep and have some family time. I think Radiosport is fun, but I just don't have time for much of it with my other obligations. This is one of the few times I get to indulge and spend a significant portion of a weekend sequestered in the ham shack, clutching a mug of coffee.

I always enter the SO SSB QRP category of the 10 Meter Contest, which is a pretty lonely category, presumably because of the relative difficulty. One nice thing about entering the category is that there may only be one or two entrants from your section (or even division!), so the chances of scoring some wallpaper is pretty good, even if you put in a fairly minimal effort.

Since I have no illusions about competing with the LP and HP entrants, my main goal each year is to beat my previous best score. Since my score from last year was 7,490, I figured with a bit of effort, I should be able to get to at least 10,000. So that was my goal for 2014.

Over the last few years, I've just used my stock station equipment for the contest, which means an Icom IC-718 (turned down to 5 W PEP of course) and a ZS6BKW doublet up about 30 feet (probably not the best pattern on 10 meters). I figured that I was going to have to up my game a bit in the equipment department in order to make a big jump in scoring, especially since we are now past the peak of Cycle 24. So I decided on a two-pronged attack to the problem: I needed a directional antenna and a way to process my speech to give me more readability for the same RF output power.

I'm on a limited budget, so purchasing a brand new commercial beam wasn't in the cards, but fortunately it's fairly easy to homebrew a decent 10 meter antenna. After putting out the #lazyweb call on Twitter for some antenna plans, Robin G7VKQ pointed me to some simple plans for a 10 meter Moxon.

These plans looked like they would be just about perfect for me, since I already had a lot of 3/4" Sch 40 PVC pipe and fittings. A Moxon doesn't have quite the raw gain of a 3-element (or more) Yagi, but it does have a very nice front/back ratio, which means a lot of my 5 watts should be only going where I want it to.

So a trip to Lowe's a few days before the contest secured me the remaining supplies that I needed (mostly the 1-1/4" PVC pipe and fittings) and I was able to construct most of the antenna in my garage in one afternoon. I had to extrapolate the PVC measurements from the plans on WB5CXC's web page a bit, since the fittings I purchased no doubt had different dimensions from the ones he used. I also ended up using some spare stiff steel wire that I had left over from a previous project and some Lexan as the spacers between the driven element and reflector.

Here's the completed Moxon up on my 6 foot ladder for testing and tuning. Using the DSA815-TG and my HFRLB return loss bridge, I was able to see that the initial resonance was around 27 MHz, and quickly got it trimmed up to a center frequency of about 28.7 MHz.

Once tuning was complete, I put the antenna up on 15 feet of 1-1/2" Sch 40 PVC mast, secured to a 2x4" support screwed into the eaves of the house and another one on the ground held down with a sandbag.

Some quick checking of the efficacy of the antenna by tuning in a JA station, then moving the antenna off-axis with my Armstrong rotor, indicated that the front/back ratio of the Moxon was indeed very impressive.

WIth the antenna situation well in hand, I just needed to get my speech processing in place. For that, I went to my Elmer, Dave W8NF, in order to borrow one of his LogiKlipper LK-1 prototypes. The LK-1 provides an adjustable amount of RF clipping (not audio processing) and the ability to interface with just about any commercial amateur radio imaginable. With the addition of a headset and footswitch, I was able to confirm on a second receiver that the LogiKlipper was working and that the adjustable clipping settings had an effect on my readability.


By the time that 13 December 0000Z rolled around, I was still attending to some family business, but I was able to get into the shack at around 0030Z to try to pick off a few stations before the band closed here at dark at around 0100Z. I pointed the Moxon to about 300 degrees and worked a KH6 right off the bat, but then something awful happened. A horrible, very bad, no good bit of wideband QRM popped up on the entirety of 10 meters. There seemed to be two components to it: a wide rushing noise at a constant S8 or S9, then a pulsing buzz (at about maybe 0.5 Hz) peaking at about 20 dB over S9.

I whinged about this on Twitter and some suggested that I go DFing for the QRM. That's a worthy suggestion, but I've never done that before, and I probably would have spent the entire contest trying to figure out the how to do it, which didn't seem like a trade-off I wanted. I could still hear the strongest stations (although the pulsing would sometimes even wipe them out on peaks), and I was really only interested in working the strongest stations any way, as a SSB QRP station. So I decided to just grit my teeth and press on.

I woke up at daybreak on Saturday and was able to get in the shack right away. Usually if there's propagation to Europe, I'll hear them on 10 meters first thing in the morning, but there was almost nothing I could hear, and certainly nothing I could work.  However, there was good propagation to New England, so I was able to make quite a few QSOs there; usually able to make a contact with anyone who was loud with a call or two. Later in the day, propagation opened up to the Midwest, followed by the South (and a bit of the Caribbean), then the Plains and Mountain West states (but no Dakotas!). In the early evening, I wrapped up with a run of JAs.

The first full day ended with a score of 8282 and 101 QSOs, which was already better than my previous best score. It didn't seem like it would take much effort to beat my 10,000 point goal at this point, barring something catastrophic like an equipment failure or solar blackout.

Up again at dawn on Sunday (one of the "benefits" of having young kids), the bands seemed to start off a bit slower. I did manage a few QSOs to the islands off the western edge of Europe, but still nothing in Europe proper. But by about 10 AM local, things seemed to pick up significantly, at least for North American propagation. Since I was hearing a lot of the same big contest stations as the previous day, my strategy was to spend a bit more time calling new mults. The entire time, I was still dealing with that terrible QRM, which was not that much of an impediment to hearing most of the time, but was awfully fatiguing.

There was also a bit of assistance and moral support from my two sons, which was much appreciated.

The second day of contesting went much like the first, at least as far as propagation went. I was able to get more DX mults on Sunday, as the pileups for those stations died off and my small signal was able to compete a bit better in the smaller piles.

When all was said and done, I ended up with just a bit over 20,000 points; more than double my initial goal!

My QSO total was 182, which indicates that I didn't quite get as many QSOs on day 2 as I did on day 1, probably because I spent more time chasing mults. I think that was a good strategy overall, but I probably could have chased non-mult QSOs a bit harder on the 2nd day had I been super-motivated.

This is obviously a subjective view, but propagation for this contest didn't seem quite as good as last year. I didn't hear nearly the amount of EU stations that I did in 2013. However, domestic propagation was still pretty great, and I was able to make plentiful QSOs from stateside stations. Those mults are just as valuable as DX mults, so it wasn't terrible that I didn't get much in the way of EU QSOs. I did manage to work almost every state except for the states immediately surrounding Oregon, with the exception of Rhode Island and North Dakota. I didn't hear as many Canadian stations as I would have expected, but I did seem to have a pipeline into Manitoba. I posted my score to 3830, but that's just a condensed version of what is here.

I believe it's safe to say that equipment upgrades played a significant role in my much-improved score. How much credit goes to each is probably impossible to precisely define, but I'm content to call it 50/50. As always, the contest was a ton of fun, and allowed me to hone my station equipment and operating skills. I've set the bar pretty high if I plan on beating this score next year. As the solar cycle continues to decline, I'm going to have to do even more on the antenna front to give me a fighting chance to beat 20k. But for now, I'm content to have done so well and had such a good time.

Wideband Transmission #5

Latest CC1 Progress


As you can see from the above photo, I have finished a significant portion of the digital side of the newest CC1 prototype and now I'm on to the receiver section. This weekend I finished my first pass of the audio chain and characterized the gain and frequency response of the chain. Next up is the design of the IF and front end of the receiver. This time I plan to do a much better job of characterizing the performance of entire radio, designing for specific critical receiver specifications, and iterating the design as necessary instead of holding on to dodgy performance from circuits.

Mixer Investigations and the Search for Better Dynamic Range

Since I decided to ditch the dual-gate MOSFET mixer front end, I've been considering what to replace it with. At first, I was thinking about using the ADE-1 for the mixer and product detector, but I've been intrigued with reading about H-Mode mixers over the last few weeks, which led me to the similar, but simpler KISS mixer by Chris Trask. That seemed like a good candidate for the CC1, with relative simplicity and better-than-average performance. Since good IP3 performance is the main characteristic of this mixer, I wanted to try measuring IIP3 at my own bench to see how it looked in a home made circuit with less than optimal parts and layout.

To get warmed up, I first attempted to measure the IIP3 of a few parts that I had on hand where I already knew IIP3 values to expect: the SBL-1 and the ADE-1. Using a DG1022 as the signal generators, my HFRLB as a hybrid combiner, and the DSA815TG, I was able to measure an IIP3 of +13 dBm for the SBL-1 and +17 dBm for the ADE-1, which is pretty much right on what other people have published.


Here is my test setup for measuring the KISS mixer performance. I deviated from the circuit described in the KISS mixer white paper in a few ways. First, I used a TI TS5A3157 analog switch, as I didn't have any Fairchild FST3157 on hand. I also used a hand-wound trifilar transformer on a BN2402-43 core instead of a nice transfomer from a company like Mini-Circuits. I drove the KISS mixer with +3 dBm from a Si5351. My measurement of IIP3 for this variant of the KISS mixer came out to +27 dBm, which seems reasonable given the poorer components I was using. Conversion loss was 7 dB. I'm going to try to measure it again with an actual FST3157 and a Mini-Circuits transformer in the near future, so it will be interesting to how much that will improve the IMD performance.

But honestly, I probably won't need better than +27 dBm performance if this mixer is used in the CC1. Since the CC1 is meant to be a trail-friendly radio with modest current consumption, I don't think I want to include the high current amplifier needed after the KISS mixer to get maximum performance out of it. Which is kind of a shame, but I figure that I should be able to keep RX current to around 50 to 60 mA and still have a receiver with better IMD performance than your typical level 7 diode ring mixer receiver. Stay tuned for more details on the CC1 front end as they are worked out in the NT7S shack.

10 Meter Contest!

Yes, it's almost time for my favorite contest of the year: the ARRL 10 Meter Contest. Ever since I moved into the current QTH, it has been a bit of a tradition for me to operate the contest as SSB QRP only. By virtue of entering that least-liked category, it has been no problem to collect some modest wallpaper from this contest. That's fun, but my real goal is to beat my previous score. Last year, I think I did fairly well with 7490 using a stock IC-718 and my ZS6BKW doublet. So this year, I'm going to have to step up my equipment game in order to have a good chance of besting last years score. I'm thinking some kind of gain antenna is going to be a must. If I can get a Moxon or small Yagi up around 20 feet and use an Armstrong rotor, that should help give me a little more oomph than last time. We'll see if I can get something built in the less than 3 weeks before the contest.