P40L-P49Y Contest Summary Information

Comments:

Surely the conditions for RTTY Roundup couldn’t be any worse than what we experienced last year, right? Well, here in Aruba they were! While there were periods of exciting rate, there were several hours where only one band was productive at all, rendering SO2R ineffective. This is the 14th consecutive year I’ve done this contest from P4 and this weekend ranks third from the bottom score wise.

10 never opened. 15 was marginal and 40/80 fell far short of making up for the high band deficiency. 88% of P49X QSOs were with North America and only 9% with Europe. Partly this divide results from Europe’s relative lower interest in this contest compared to other contests where a typical split from here might be 60-40 NA-EU.

But wait! We had this new-fangled digital mode: FT8. How would that impact the contest? More on this later.

So, I enjoyed similar exhilarating QSO rate the first three hours to the best years from here. Then, activity dropped off faster and further than ever before. I was 500+ QSOs down halfway through when I took a sleep break, after Europe daybreak. Sunday morning I feared 15 might never open. The band scope was empty but I CQ’d anyway until the first call came in at 1410z. Signals were surprisingly strong, just few and far between until around 15z. I had exactly one EU QSO on 15m, a relatively strong French station. I’m sure many others could have been worked, had they tried the band. Still, there wouldn’t be the tremendous openings to smaller stations that we have in better years.

Usually, we can take solace in the low bands at the bottom of the cycle, but 80 and 40 were unimpressive to say the least. They finally opened late Saturday night for a taste of how we typically find the bands. I had been on 160m each night the last week and it was in very poor condition also.

Local Issues

Maintaining a remote station subject to constant trade winds and salt air is a challenge as I found last week. Arriving late Monday, I didn’t see the antennas until Tuesday morning when I noticed the skewed reflector on the 20m Yagi and the same for one T-bar loading element on the 80m dipole. I checked SWR on all antennas and they were normal. Then, in the middle of operating 80m Tuesday night, the SWR shot way up. Wednesday morning showed me that the skewed T-bar was now missing from the 80m dipole. Not able to find the T-bar in the tangled mass of thorn bushes in the open space behind the property, I set out to add a wire coil at the feed point to compensate for the missing loading on one side of the dipole element. That worked great to bring the resonance back to where it was and our switched relay box of other coils gave us the same 4 band segments in CW and Phone. As I was climbing the tower to install the coil, I noticed that the 20m reflector was now also on the ground! At the top of the tower, I could see the reason for both failures: severely corroded U-bolts had disintegrated. Of course, that doesn’t bode well for all the other U-bolt and hardware fasteners on all the antennas. I wondered how much more of the overall antenna system would be raining down soon.

The 20m Yagi actually still works fine and doesn’t seem to be too hampered by the missing reflector. However, Thursday night the 80m SWR shot up again and Friday morning I discovered severe RF arcing had blown a hole through the PVC element-mount insulator. I cleaned it up as best I could, knowing that once RF carbon has been created in material, its impossible to return it to the original insulating ability. I planned to run low power and see if I could sneak by. My initial idea was, heck, I’ll run FT8 on 80m and take advantage of the weak signal capability. I soon realized that isn’t currently possible with SO2R running RTTY on the other radio (to efficiently maintain only one signal at a time). It turned out that I was able to run 600 watts with no apparent arcing and it ran fine all weekend like that.

FT8

The telegram is that I planned to run RTTY only this year, but wound up making about 10% of the total contacts on FT8 during the last 3 hours of the contest. Although I choose to not use not to use spotting assistance, I had to enter SOU because of using the WSJT-X "multi-channel decoder" to operate FT8.

FT8 is not ready for serious high-rate contesting as we’ve known it on the HF bands. The theoretical maximum FT8 rate is currently less than half what good SO2R RTTY operators can achieve in practice. And there are many reasons why the minimum 4 transmit cycles per FT8 QSO (one-minute) often stretch to more cycles, or never completes at all.

OTOH, FT8 has significant advantages for a much larger group of contest participants who are constrained by geography, housing limitations, solar conditions, power, noise, etc. The amazing explosive growth of FT8 activity since its introduction in mid-2017 also means that there are many more participants for all of us to work in contests. In the first-ever FT8 (only) contest, the FT8 Roundup last month, fully 35% of the logs received were call signs not in Super Check Partial! It was the first contest for the winner (who has only been licensed 2 years!) and one other Top Ten finisher. Contests don’t exist without participants for everyone to work, so these new “contesters” are great news.

I did not think RTTY and FT8 would work well coexisting in the same contest. I feared that there would be a drain of participation from RTTY by folks who are enjoying FT8 now, thereby dispersing participation for both modes. I worried about conflict between FT8 and RTTY easily sharing the same sub-bands and whether all participants would “get the word” about recommended locations to operate.

Of course, like many things we can debate ad infinitum, a little bit of actual experience goes a long way to inform our imagined opinions about these things. My observation this weekend was that RTTY participation was more affected by the poor solar conditions than any number of “defectors” to the FT8 mode. And, the FT8 activity was confined to 1-2 very small 2-3 kHz sections of each band. The RTTY ops steered clear, mainly because those few kHz were “occupied” by density of signal strengths that would make RTTY operation unpleasant and ineffective. Ironically, I think there was nearly as much FT8 activity in 4 kHz as there was RTTY activity in the other 50 kHz of, say, 20 meters. Another nice attribute of this new mode.

As Sunday wore on and I watched my rate dropping in the afternoon, plus consulting my rate sheets from prior years, it occurred to me that maybe the RTTY rate could get low enough that spending the last few hours on FT8 might be more productive. In the end, I think it was probably a wash in terms of score. I also picked up 5 DX mults I doubt would have come via RTTY. Finally, it gave me actual operating experience in this new dual-digital mode contest.

In theory, the maximum SO2R FT8 rate would be 120/hour and when I switched modes, my RTTY rate was in the 60s. My goal was to maintain an actual FT8 rate of 90, which I was able to do for the first hour. Maybe that was dumb luck or maybe my attempts to “game” the WSJT-X software to maximize rate worked. The last two hours, the rate dropped off to 65. I don’t think I would have done much better, if at all, by staying on RTTY and I’m pretty sure I wouldn’t have worked as many as 5 new mults.

On the plus side, I figured a lot of the FT8 stations would not have been on RTTY, giving me new blood to work. However, offsetting this advantage might be an excessive number of dupes by stations I had already worked on RTTY but with no good way to flag dupes between RTTY and FT8 (no software duping between software programs and modes), or they didn’t care that a QSO only counts once per band regardless of mode. I was happy that there were only 10 dupes in the 270 FT8 QSOs, higher than my overall dupe rate, but much less than I feared.

On the down side, I depended on each of the other operators being skilled enough with WSJT-X to avoid wasted 15-second cycles. Unnecessary FT8 cycles would kill rate, so I decided to use this 3-hour opportunity to figure out what I could do on my side to reduce wasted cycles. For example, how long should I let message re-sends go on before aborting the attempted QSO? What is the best way to do that? What specific features, configurations and checkboxes should I use to accomplish this?

I tried a number of things to “game” the software for higher rate. Most of them worked great at times and not at all at other times. Several times I thought I found the magic, only to discover several QSOs later that it didn’t always work. By the end, I had little to show for take-aways. I did learn that it is a very complex problem that involves WSJT-X software that was designed without consideration for contest usability and that my QSO partners make any number of decisions that affect the outcome. And, frankly, I suspect many of them don’t really care much about “my” rate!

In classic ham spirit, I was excited and honored to be able to participate in an emerging technology that contesters had their first peek at operating. I’m confident that over time improvements will be made in operability and performance of new modes like FT8. For the first dual-mode Roundup, I’m very impressed with how it all worked out.

Rigs: Elecraft K3 (2), K-Pod (2), P3 (2)
Amps: Alpha 86, Alpha 91B
FilterMax III low power band pass filters (2)
4O3A high power band pass filters
SixPak, StackMatch (2), BandMaster III decoder (2)

Tower 1: C31XR at 43 feet
Tower 2: 2 elements on 10 meters / 5 elements 15 meters at 55 feet
Tower 3: 4 elements 20 meter at 68 feet
         2 elements 40 meter at 76 feet
         1 element 80 meter Sigma 80 at 64 feet
         160 meter \"Double L\" vertical at 67 feet
Four 400-500 foot beverages using K9AY switching box/preamp (JA/W6, W1, EU and
Africa/VK-ZL)

Logging software: WriteLog 12.24E on three networked PCs

RTTY Decoders (each K3): MMTTY, 2Tone(3) on main receiver
                         MMTTY, 2Tone on second receiver
                         (setting both receivers on same frequency yields
                         6 parallel decoders)

73,

Ed - P49X (W0YK)