CQWW CW 2010 - the story

Result certificate

4 November 2010 - deciding which category to enter

I'm making plans to enter the CQWW CW contest over the weekend of 27-28 November 2010. I've entered this contest a few times before and currently (2010) hold the GM record scores for the Low Power 1.8 MHz Single Band category (2006) and also the Low Power Assisted 14 MHz Single Band category (2009). The first decision I need to make is which category to go for. I could enter either of the categories for which I already hold the GM record score but that seems a bit pointless, so I'll look for another category where I might have a chance of claiming another GM record score.

Looking through the 2010 CQWW CW All Time Records I see that the GM QRP 7 MHz Single Band category record is currently at 924 points based on 36 contacts...in a 48 hour contest this should be an easy target to beat even with QRP (5W)! I've never been much of a QRP operator, so entering a major contest in this category would be a new experience for me. In this year's RSGB 80m CW Sprint contest series (2010) I decided to try the QRP section and have been very surprised at the results that can be achieved in a contest using low power. I even managed to come first equal in one of the contests (click here to see this result). This has given me an interest in doing some more serious QRP contesting.

CQWW contests are all about working DX so, if I am only using 5W, it will be essential to have the antenna performing as well as it possibly can. 40 metres is a band I can easily put a decent DX antenna up for (see the section on this site '40m quarter wave vertical' and 'ARRL 2010 CW Contest'). The GM record score in the QRP 7 MHz Single Band category is so low that it's not really very much of a challenge - but looking at the Europe All Time Records for CQWW CW in this category the current holder is S54AA with 152,950 points from 565 contacts. That would be much more of a challenge! So the final decision is to enter the QRP 7 MHz Single Band category with the first target being to take the GM record score (easy) and the second target being to take the European All Time Record (hard).

16 November 2010 - deciding which antenna to use

As mentioned earlier, it will be simple for me to erect my 40m quarter wave vertical again. This is a good DX antenna and has performed well for me in the past. The unknown factor is how it will do with QRP. Never being content to just do things they same way all the time, I've been looking through the ARRL Antenna Book to see if there is any kind of wire antenna that I could build to give me some gain on 40m. In the past I've considered building a 2 element yagi for 40m using 4 x 10m fishing poles - but I can't get enough height (60 feet) at my home QTH to make this option worth trying. That idea will need to wait for another day. There is a type of 2 element phased vertical wire array called a 'half square' that might be worth investigating. N6LF has described a wire 80m half square. I'll do some modelling of a 40m version of the N6LF 80m design and see if it's worth trying.

19 November 2010 - investigating the 40m half square

Clicking on any of the images below will display a larger version. Based on the 80m design described by N6LF a 40m half square was designed by simply scaling down to 7MHz.

Scaled down dimensions for 7 MHz.

Before building any new antenna, it can be very informative and interesting to model it using one of the excellent modelling software programs we now have. The modelling software I use is 4nec2x which is available as freeware. It would be very useful to compare the new 40m half square to a simple low dipole and also to a quarter wave vertical. After creating a model of the half square from the dimensions above, the following results were obtained.

The physical layout of the model. The wires and ground can be seen and the purple blob on the top right is the feedpoint.

SWR plot. The lowest SWR is slightly below 7 MHz but to save time it was decided not to bother fiddling with the precise model dimensions. The performance is close enough to 7 MHz to allow the model to be compared to a simple 40m low dipole - which is what I want to do.

The 3D radiation pattern. The directivity in the broadside directions is clearly seen. The low elevation angle of maximum radiation is also shown. Both of these characteristics are potentially useful to help me get my QRP signal out to DX locations.

Vertical cross section of radiation pattern in the direction of maximum radiation - comparison of half square to a low dipole. The better low angle performance of the half square is clear. The low dipole would be better for non-DX working (UK and near Europe). It should be remembered that low dipoles are more or less omnidirectional.

Horizontal cross section of half square radiation pattern with 3dB beamwidth angles shown by red lines. The beamwidth is about 70 degrees. I will need to orient the half square so that the DX regions I'm most likely to be able to work on QRP are within the 3dB beamwidth angles.

Vertical cross section comparing the half square with a quarter wave vertical. The gain provided by the half square is clearly visible.

Horizontal cross section comparing the half square with a quarter wave vertical. Notice that the quarter wave vertical is better in some directions due to it being omnidirectional.

So - what are the regions I have a fighting chance of working with 5W of CW on 40m? Well, it's impossible to be precise about this because there are a lot of variables involved. Some of the key variables are: propagation conditions; quality of TX antenna (mine); quality of RX antennas (the stations I'm working); operator ability (me and them). Based on my experience, it would be optimistic to expect to work lots of DX with QRP - but it should be possible to get into north America, Asiatic Russia and possibly Africa. Let's see if the half square could be put up at my QTH in a way to get my strongest signals in any of these directions.

QTH GM3YEH. North is vertically upwards. The yellow line shows how I could fit the half square into my garden.

Red shaded areas show the approximate 3db beamwidth coverage. North America is well covered, Africa is partially covered but Asiatic Russia is not well covered.

I'll need to decide whether to go for the half square on it's own or if I should maybe consider also putting up my quarter wave vertical (which is omnidirectional). Another interesting idea would be to think about finding a way to orient the half square as close to N-S as possible. If that could be done then better coverage of Asiatic Russia would result and there might even be a slight chance of getting into South America. This is because the 3dB beamwidth coverage areas would rotate anti-clockwise by about 20 degrees.

21 November 2010 - some propagation predictions

HF propagation prediction is nototiously difficult but a great deal of work has gone into trying to develop reasonably reliable models to use in simulation software. Some of the software used for this purpose is difficult to understand and use. Because of the combined issues of predictions being of limited accuracy and most software being difficult to use, many radio amateurs prefer simply to use their general knowledge of propagation along with live monitoring of the bands. I have to concede that this latter approach has a lot going for it. I also feel, however, that having sophisticated tools at our disposal for HF propagation predictions at least makes it of some interest to run predictions when planning any kind of major operation such as entering a contest. I never expect the real conditions to comply precisely with the predictions but there is often a general and approximate relationship between the real and predicted band conditions. One of the easiest ways to run HF propagation predictions is to use the online tool created by Jari OH6BG (click here). I've run some predictions based on my 5W of QRP on 40m using a quarter wave vertical. It has been assumed that the DX station has a similar type of antenna. The results are below.

GM3YEH to Europe

Looking along the 7MHz line we can see that very high probabilities of good propagation exist for almost all times of day and night. The path to Europe only becomes poor between about 9am and 2pm. This information could give a good indication about when to have a rest period.

GM3YEH to east coast North America

The horizontal 7MHz line predicts possible reasonable propagation from about 8pm until 9am with possible good openings at around 5am and around 9pm.

GM3YEH to Asiatic Russia

The 7MHz line indicates possible openings between about 4am and 8am and aslo between 3pm and 6pm.

GM3YEH to Africa

No propagation is predicted on this path.

22 November 2010 - deciding on QSO rates required to hit the target.

To have any chance of taking the European QRP All Time Record it will be necessary to make at least 600 QSOs. The final score depends heavily on the number of zones and DXCC countries worked since these are used as multipliers. Using QRP I will just need to focus on the number of QSOs since it is unlikely that I'll have a lot of joy trying to catch rare zones/DXCC entities. When using 100W or more on CW it's not too unusual to be running at well over 100 QSOs per hour...sometimes even double that. With QRP this kind of rate is extremely unlikely. To accommodate rest breaks it would be reasonable to plan for around 38 hours of operating over the 48 hours of the contest. To get to the target QSO count this means an average of around 16 QSOs per hour. Looking at the propagation predictions makes it possible to make some guesstimates of QSO rates that will result in achievement of the 16 per hour average. Below is a graph that would do the trick if it can be achieved. If these targets are met it will give around 700 QSOs.

QSOs per hour targets for both days.

26 November 2010 - 15:00 - Half square is built and is under test

The 80m dipole I use for the RSGB CW Sprint contests uses the same supports that I would need to use for the new half square so it needed to come down before the half square could go up. The November sprint was on the evening of Wednesday 24 November so I left the half square build until this morning (Friday)...which is cutting things fine - CQWW CW starts at midnight UTC tonight! Most hams are well used to flying by the seat of our pants so this is fairly normal 'last minute' behaviour. The full build took about 3 hours with a tea break included. Below are a couple of scans from my antenna analyser. I only needed to make one trimming adjustment - that involved shortening each of the lower horizontal short legs by 1m. I managed to get the orientation around anticlockwise a bit, but not by the 20 degrees I wanted.

Antenna analyser plot showing SWR on initial build and the effect of reducing the short horizontal legs by 1m each. Initially the antenna was resonant a bit too low in frequency and this was the reason for shortening the elements.

This is a more detailed scan of the final build. Note that the resonant frequency is 7.0268 MHz but that the minimum SWR is 1.16:1 on 6.9900 MHz. In other words, minimum SWR does not indicate the resonant frequency. This is not a freak result - it's perfectly normal. It shows that the common belief that the frequency of lowest SWR is the resonant frequency is wrong.

To get an idea of how the antenna is working on receive, I've set up my CW Skimmer system and interfaced it to the Reverse Beacon Network (RBN). The idea is to allow my new antenna to pass all CW CQs it hears to the RBN server where I will be able to inspect the results over the rest of today. This will happen automatically while I'm at work. Signals are being monitored between 7.000 MHz and 7.020 MHz. The RBN lets you see a screen like the one below. It will be interesting to see if any DX starts appearing as darkness approaches. The only DX spotted so far is 4L8 in Georgia. Daylight hours have already been predicted as the least productive for DX so it's no surprise that there's not much DX being spotted at the moment.

RBN screenshot showing spots and a map with the location of the received stations. Notice the spots from Georgia and Kazakhstan. The Angola spot is not real...there is no D2AKJ so this has been a decoding error.

26 November 2010 - 20:53 - The half square is hearing DX!

Having been running for 6 hours, it's clear that the new half square is hearing some good DX to the east and to the west. After sunset some east coast North American signals have been coming in.

RBN screenshot at 20:53 - 3 hours before the start of the contest. Note the continued good coverage of Europe and the signals from the east coast of Canada and the USA.

The testing done today of how well the new antenna is hearing confirms that I can pick up signals from Europe, Asiatic Russia and North America. What remains to be proven is if I can get a good enough signal out to these regions with only 5 watts. Time will tell...we'll know in about 3 hours when the contest gets underway.

Sat. 27 November 2010 - 09:15 - Quite a good first session

That's the first operating session done and it's time for a rest now. I had a bit of a scare about an hour after the start when my old faithful IC751A started playing up with a problem I've never had before. Sometimes when I started turning the tuning dial the receiver went dead and the only thing that brought it back to life was a power off-on cycle. This happened a few times and my heart was sinking fast at the thought that I might be on the verge of a rig failure. It occurred to me that there was an unusually thick pile of paper, books, boxes and other sundry stuff around and on top of the rig - so it might be getting a bit hot. I cleared all this stuff away to let it cool down and in 15 minutes I was back in business with no further rig problems. For the first operating session from Friday midnight until 9am on Saturday morning the QSO target set was 210. I managed 297 so I'm ahead of schedule at this stage. I've worked 71 countries and 18 of the 40 zones. The points total just now is 44,767. The best DX was JA and I was surprised to be able to work him. A third of the contacts so far have been outside Europe. That is a lot of DX for 5W. Europe has been easy to work as expected. Lots of Ws and VEs are in the log and I managed a few South Americans and Asiatic Russians. It's time for a sleep now that 40m has gone quiet. As I've been writing this update at 10:30am I heard another JA on the QRG the rig is sitting on. I managed to work him after about 4 calls. The band is very quiet now. Time for a sleep. The next shift will start at 3pm.

Sat. 27 November 2010 - 22:00 - Start of second session

I was back on the air just after 3pm on Saturday afternoon. Mostly Europeans were worked with the occasional Asian contact. Some of the DX contacts were a real struggle with multiple repeats being required. The H on the end of my callsign gives a lot of operators problems. Many of the difficult DX contacts were because my callsign was being read as GM3YES. This is not an unusual problem for me but it's probably worse this weekend because my QRP signal will be marginal at DX QTHs. Just before 9pm I started hearing and working east coast Ws and VEs. This opening lasted until just before 10pm and it corresponds exactly to the propagation predictions. I'm not getting many new multipliers now. I was hoping for a DX opening to the east around sunset and I did hear VK, HS, BY and a few others - they were very weak. I didn't even get a 'dit dit dah dah dit dit' from any of them. I haven't worked a new zone since the first session. Looking at the S54AA record score it's clear at this stage that I'm going to be nowhere near his tally of zones. I'll probably creep towards 100 countries by the end but my multiplier count is going to fall far short of what would be needed to take the record. It's possible that I might manage more QSOs that the S54AA, record but without a similar number of multipliers my points score will fall short. The best I can do is to watch for the DX openings and pick up as many 3 pointers outside Europe as I can (contacts within your own continent get 1 point but contacts outwith your continent get 3 points).

Sun. 28 November 2010 - 09:30 - End of second session

All the zones likely to be worked on QRP on 40m were picked up in the first session so I was on the lookout for any rare ones last night. I eventually managed to get zone 3 at 08:45 and within 15 minutes I had worked another five of them! A propagation window had obviously opened. This is another example of the propagation predictions being quite close to what has been happening. There are a few stations I keep hearing but can't get them to reply. In particular, JT5DX has cropped up many times but no matter how long I sit and call him he can't copy me. That's the second session finished with the band now fairly quiet so I'll have another sleep until mid-afternoon. I'm not expecting to work any more new zones and I'll concentrate on trying to get to 100 countries. The QSO count is exactly on target to be bigger than in S54AAs European record score. Below are the current zone multipliers and country multipliers from Europe and in North America.

Zone mults worked by Sunday at 09:30. There are no easy ones left!

EU country mults by Sunday 09:30. Not many left that are likely. The only ones I might still get are the missing G prefixes (no short skip on 40m so far) and possibly OJ0 - I've heard one of those already but couldn't raise him.

NA country mults worked by Sunday 09:30. There are still a few that could crop up from NA so I'll need to be on my toes this evening when propagation opens to this area. This is where I'm most likely to pick up any new countries to try and get me to 100.

Sun. 28 November 2010 - 19:00 - Start of third session

I started the third session at around 5pm on Sunday. Other than some Asiatic Russians, the only DX I heard was a BY. Unfortunately he had been on his QRG for a while when I came across him so there was an almighty pileup. I didn't hang around too long with my 5W. The only way to get rare DX with QRP is to catch them on their first or second call - before the pileup starts. I heard JT5DX again but couldn't raise him. I settled for picking up a fairly regular stream of Europeans to keep the QSO rate up. I managed to catch a GJ who was calling someone that I also had called. The station we were both calling went silent so I called the GJ and he replied to give me a much needed country multiplier. I still haven't heard/worked GM! Most contacts are taking about a couple of minutes to complete today...sometimes even longer. I frequently need to give repeats until my callsign is correct. I still need another 12 countries to get to 100 but the only new one I've had all day was the GJ. I might get a few more Caribbean countries later this evening, but it's not looking good for getting to 100. I've already gone well past S54AA's QSO record but even with a lot more contacts my lower multiplier tally means I'll come nowhere near his total points score at the end.

Sun. 28 November 2010 - midnight - End of third session

That's it for another year...the mightiest CW contest of them all! Sunday evening provided excellent propagation to North America and certral Europe, but not a lot else. The possibility of openings to the Caribbean and South America that I was hoping for to get my country count up to 100 didn't materialise. Although the North Americans were strong I had already worked most of them so it was hard work finding the ones I still needed. I managed another zone when I got VK6AA....an epic struggle! The 3 new countries in this session were VK, OD and A45. I tried to raise JT5DX again without success. The QSO count achieved was significantly higher than the S54AA record but the final points score is lower due to him getting a lot more multipliers. Getting 34 of the 40 zones using 5W is incredible. I'll write some final comments and conclusions on Monday after I've had a decent rest.

callsign points QSOs zones countries

S54AA 152,950 565 34 127 Current Europe All Time Record

GM3YEH 44,767 297 18 71 9 am Saturday

GM3YEH 66,639 411 19 78 10 pm Saturday

GM3YEH 96,036 550 20 86 9:30am Sunday

GM3YEH 109,944 650 20 88 7 pm Sunday

GM3YEH 127,568 721 21 91 midnight Sunday

29 November 2010 - Final comments and conclusions

Here are some graphs showing the analysis of how I got on. I used Athena by PC5M to generate the graphs.

QSOs per hour and comparison of the running QSO total against the goals set before the contest. The QSO count was greatest just befor the band started to close for DX after sunrise on Saturday morning. The predicted QSO rate was almost exactly what happened in reality.

Points per hour

Multipliers per hour

There were a lot of successes achieved by GM3YEH in the 2010 CQWW CW contest. The majority of the pre-contest predictions and analyses turned out to be relatively accurate. The propagation predictions were close to what was experienced. The half square antenna performed very well. Some more testing will be done at a later time on this antenna to check it's real-life performance against some other antennas such as a quarter wave vertical and a dipole. This work will need to wait until the snow and sub-zero temperatures go away though! One major target was to beat the European All Time QRP 40m Record, but this was not achieved in spite of the QSO count being much higher than the record. It's points that count in the end and a few more zone and DXCC country multipliers would have been needed to take the record. If conditions had allowed me to get some more Central and South American multipliers then the record attempt would have definitely been on. My antenna didn't favour these directions so this is food for thought if I go for this record again another year. I've got the feeling that once we have an increase in sunspot numbers the record could be a more realistic target. It has clearly been demonstrated that it's possible to rack up enough QSOs using only 5W to challenge the record. A few highlights are worth mentioning. I don't know the full reason why, but with only the QRP signal I was really impressed by the high proportion of Stateside stations who copied me on my first call. I know that I have a good terrain profile from my QTH towards the USA, but there must be more to it than that. Their big antennas and extremely sensitive receivers will help as well but I also think that there are some really great CW operators over there. All of this qualifies them as having 'good ears'. I managed to get a VK and this was unbelievable. I had to call for a while but when he eventually heard me it was a simple QSO. Late on Sunday night when I was resigned to just picking up a couple of Qs per hour I was surprised to find an OD and worked him with no trouble. So - it's now a bit of a wait to see where GM3YEH is placed in the 2010 CQWW CW QRP 7MHz Single Band Category. The results won't be out until the autumn of 2011 - I'll finish the story then.

I hope you've enjoyed following my exploits. If you have then drop me an email at gm3yeh@gmail.com to let me know that someone is actually reading this stuff. I'll then know whether it's worth telling more amateur radio stories online this way.

Final result:

4th in Europe

5th in the world