Archive - September 2009

Friends,

The UK Scout Contest Team is putting on a high power Multi-Two station for CQWW from Dorridge (24th / 25th October 2009).

Would you like to operate a mega-station? The station details are genuinely very impressive and would blow away anyone’s capability they might have at home; from full sized 160m dipoles at 70 feet to 3-element arrays on 40m to the USA. Both multi-band stations have an ACOM 2000 fed by either an FT2000 or a K3. We have station filtering to allow simultaneous high powered operation by two stations just feet from each other on adjacent bands and each band will have at least two antennas available. It’s pile-up heaven.

Stub tuning completed

21st September 2009

Finished tuning the stubs filtering system yesterday, they are quite easy to do however i should probably have done this at the main station shack as my own shack is a bit small for hundreds of meters of Coax

Filtering

20th September 2009

Overview

Very early on in the development of the station we realised that filtering was going to be essential to our station, it would be almost impossible to put together a multi transmitter station without adequite filtering. Because our antenna’s are in very close proximity we set about designing a filtering system that would cope with this.

We decided to address the filtering issue in two stages. The first would be through the use of Band Pass Filters on the low power side (between transmitter and amplifier) and the second stage would be coaxial stub filters on the high power side (between aplifier and antennas). One we bagan to research this we realised quite quickly that there is very little practical information available on the real world use of filtering in a M/2 or M/M contest station environment.

Low Power Stage – Band Pass Filters

After quite a bit of searching for a suitable BPF solution it appeared that there were only really two main options for a Multi-2 setup where each station must have switchable filtering for each band, these are either the Dunestar 600 or the I.C.E Band pass filters. Both of these options are quite expensive and didn’t really help with our understanding of filters as they are essentially off the shelf solutions. It then came to our attention that Bob Henderson 5B4AGN had designed a switchable multi band BPF system based on the W3NQN filter design and had produced a number of professional boards & cases for these.

The W3NQN filter design is very widely used as it provides excellent out of band rejection and is quite simple to construct. See below for the schamtic of a W3NQN filter.

What Bob had done is combine the excellent filter design of W3NQN, with a motherboard & daughter board system that allowed filters for all the HF contest bands to be combined together in one box and be easily switched. We purchased two sets of boards and cases along with the various components needed, these filters also incorporate a band decoder and relay drivers, the relay drivers especially would be essential for our plans for the second stage filters. The filters also allow us to automatically bypass the BPF’s if we are using a band where a supported filter does not exists, WARC bands for example. Below is a picture of one of the completed BPF boxes.

Performance and Frequency Response

Performance of the filters is excellent, these had their first real test on SSB Field Day in 2009 and they performed superbly, however due to the close proximity of our antenna’s there were still some problems on the exact harmonics, these will be dealt with in the second stage filter system.

I will be posting frequency responses when i am able to, however due to the failure of my VNA i am unable to do this currently. Once the new VNA is completed i will update this article.

SWR Response

SWR Response was measured with an MFJ-260C Dummy Load and where taken using 100W from an FT-2000 on an external SWR Meter

BPF1

	160	80	40	30	20	17	15	12	10	6
Control (no filter in line)	1.05	1.05	1.07	1.09	1.09	1.08	1.09	1.08	1.09	1.1
Pass Through	1.03	1.03	1.05	1.05	1.04	1.05	1.06	1.06	1.07	1.04
160	1.19			N/A		N/A		N/A		N/A
80		1.15		N/A		N/A		N/A		N/A
40			1.14	N/A		N/A		N/A		N/A
20				N/A	1.18	N/A		N/A		N/A
15				N/A		N/A	1.38	N/A		N/A
10				N/A		N/A		N/A	1.3	N/A

BPF2

	160	80	40	30	20	17	15	12	10	6
Control (no filter in line)	1.05	1.05	1.07	1.09	1.09	1.08	1.09	1.08	1.09	1.1
Pass Through	1.06	1.05	1.06	1.08	1.07	1.08	1.08	1.06	1.05	1.19
160	1.13			N/A		N/A		N/A		N/A
80		1.1		N/A		N/A		N/A		N/A
40			1.08	N/A		N/A		N/A		N/A
20				N/A	1.14	N/A		N/A		N/A
15				N/A		N/A	1.2	N/A		N/A
10				N/A		N/A		N/A	1.4	N/A

High Power Stage – Coax Stub Filters

Coaxial stubs are very versatile things, they have many uses from antenna matching to phasing lines for antenna array’s, however they are also extremely effective for filtering of harmonics. They are widely used but there seems to be little practical information on constructing a stub filtering system.

There are many different types of stubs that can be used for filtering however we will only be using 1/4 wavelength shorted stubs and 1/2 wavelength open stubs. A 1/2 wavelength open stub, that is a stub that is open circuit at one end will null harmonics on F/2, 3F/2 & 5F/2 where F is the frequency for which the stub is 1/2 wavelength. A 1/4 wavelength shorted stub, that is a stub that is a short circuit at one end will null harmonics on 2F, 4F, 6F… where F is the frequency for which the stub is 1/4 wavelength.

The type of coax to use for stubs is not particularly important, however i would suggest using RG-213 or better (avoid RG-58 if you are using any power at all). The impedance of the stub coax is also not important, for our stubs we use 75Ohm coax. The one factor that will affect the performance of the stubs is the loss of the coax, the lower the loss the sharper and narrower the null, coax with higher losses will present a shallower but wider null.

When building stubs, it is useful but not essential to know the velocity factor of the coax being used as this will allow the rough cutting of the coax to be more accurate. Stubs cannot be cut by measuring only, it is essential to use some form of analyser to tune the stubs correctly. I would advise that any stubs be rough cut with sufficient spare left for the final tuning. Stubs should also be well insulated at their ends as high voltages can be produced here.

Single stubs can be very effective, however they often leave harmonics outside of their null’s, because of this our filtering system was designed to utilise multiple stubs on each band to reduce the harmonics as much as possible. As the null’s can be quite sharp they will often not cover enough of the band, to address this two stubs for the same band can be connected together, this will provide much greater attenuation, however if the tuning of the stubs is offset then this will also widen the null to the extent that it will cover a sufficient partion of the band.

There are many ways of switching stubs, however we decided that we would utilise two ArraySolutions SixPak switches for this purpose as this would allow us to switch in feed lines with our multi stub arrangement attached for each band, it would also allow us to share the one stub system between the two operating positions.

The SixPak’s have been modified to take some nice lockable DIN connectors, this allows us to easily dismantle and re-assemble the system as needed, it also makes things a lot easier to fault find if a piece of equipment needs to be removed. The two SixPak’s are driven by the band pass filters which contain current sourcing relay drivers, these are connected to the two SixPak’s via an interface box that allows us to drive the two SixPak’s in tandem. The Stub’s are then connected to between what would normally be the antenna outputs.

160m Stub Arrangement

160m uses a simple two stub arrangement, these are 1/4 wave length on 160m

Center of nulls

Cable	160m	80m	40m	20m	15m	10m
HPF160S1A		3.510	7.035	14.095	21.155	28.230
HPF160S1B		3.550	7.105	14.210	21.315	28.478

HPF160CC is the phasing line between the two sets of stubs and is cut to 1/24 wavelength at 1.900Mhz

80m Stub Arrangement

80m uses a simple two stub arrangement, these are 1/4 wave length on 80m

Center of nulls

Cable	160m	80m	40m	20m	15m	10m
HPF80S1A			7.044	14.093	21.160	28.244
HPF80S1B			7.108	14.223	21.300	28.550

HPF80CC is the phasing line between the two sets of stubs and is cut to 1/32 wavelength at 3.550Mhz

40m Stub Arrangement

40m uses a six stub arrangement, HPF40S1A & HPF40S1B are both 1/4 wavelength shorted stubs which will null 20m & 10m, however this still leaves a harmonic on 15m to deal with. This is addressed with stubs HPF40S2A & HPF40S2B, these are both 1/4 wavelength at 10.5Mhz or 1/6 wavelength shorted on 40m, however this presents a problem as it will affect the SWR on 40m. To compensate for this stubs HPF40S3A & HPF40S3B are inserted in parallel as compensation stubs, these are both 1/12 wavelength on 40m.

Cable	160m	80m	40m	20m	15m	10m
HPF40S1A				14.090		28.238
HPF40S1B				14.245		28.560
HPF40S2A					21.220
HPF40S2B					21.350
HPF40S3A					21.220
HPF40S3B					21.350

HPF40CC is the phasing line between the two sets of stubs and is cut to 1/16 wavelength at 7.100Mhz

20m Stub Arrangement

20m uses a four stub arrangement, these are 1/4 wave length on 20m, two are shorted and two are open

Center of nulls

Cable	160m	80m	40m	20m	15m	10m
HPF20S1A			7.044		21.160
HPF20S1B			7.110		21.400
HPF20S2A						28.263
HPF20S2B						28.580

HPF20CC is the phasing line between the two sets of stubs and is cut to 1/24 wavelength at 14.225Mhz

15m Stub Arrangement

15m uses a simple two stub arrangement, these are 1/4 wave length on 15m

Center of nulls

Cable	160m	80m	40m	20m	15m	10m
HPF15S1A				14.090		28.230
HPF15S1B				14.245		28.560

HPF15CC is the phasing line between the two sets of stubs and is cut to 1/24 wavelength at 21.300Mhz

10m Stub Arrangement

10m uses a four stub arrangement, two are 1/2 wavelength on 10m, the second two are 1 wavelength on 10m or 1/4 wavelength on 40m

Center of nulls

Cable	160m	80m	40m	20m	15m	10m
HPF10S1A				14.090
HPF10S1B				14.230
HPF10S2A			7.058		21.215
HPF10S2A			7.110		21.375

HPF10CC is the phasing line between the two sets of stubs and is cut to 1/6 wavelength at 28.500Mhz

SSB Field Day 2009

18th September 2009

Field days seem to whoosh by in a seamless tirade of bodily abuse that starts in the balls of the feet, grows through your hands and wind blown cheeks before reminding you that you are starving hungry and you’ve only just taken over the driving seat.

A very dissapointed 1012s QSO this year and we still can’t work out why we lost ground so fast with the other leading stations, even on 80m where we should have held our own. Certainly having a tribander at only 35 feet is a hell of a penalty to have to suffer every year, but why our 40m and 80m runs were not producing the numbers that the other high-flying entrants were managing? I just don’t know. High probably does mean might. Last year’s 1,000 foot entry was very different on the higher bands to this year. Mostly, we were firing right through a heavy forest. I wonder how much attenuation there is in 500 yards of dense oak trees that are about 80 feet tall? Well, you get my drift.

Filtering this year was amazing. We used a second rig as a dedicated 2nd receiver using James’s new Band Pass Filters, made to Bob Henderson’s designs and we had negligable breakthrough. I can’t wait to use the new permanent multi-two station when we have the stub filters in place too.

Dorridge Day 2009

18th September 2009

Fan Dipoles at 50 feet, 400 watts and lots of atmosphere. We scored just under 100 contacts mostly UK based throughout the UK.

After bumping into my friend Alan (G3LUA) on 80m, he jumped in his car and came to see us with his wife. Alan was one of my instructors at Solihull Amateur Radio Society when I first applied for a license and realised that I couldn’t do the RAE anymore, I had to do a Foundation exam. I did finally achieve my aim though, I took the last ever RAE ever held in Dec 2004 (I think). and passed..!

Lovely day, great fun and good excuse to roll out the toys again.

Wendy and her team were also out in force with Girl Guiding. Having both Guiding AND Scouting in the one family can be hilariously funny as we compete with each other. I won the tallest structure, she won the most colourful!

Pictured is James (2E0YOM) running the pile on 80m.
Callum.

The Station

17th September 2009

The M0XXT Contest station is located in the Dorridge scout hut.

We are currently undergoing a shack move and major station re-design, below is the current systems overview schematic for the planned station

we will be documenting the various elements of the new station as it is designed and constructed, so check back often for updates.

Great Circle Map Centered on the M0XXT shack

Band Pass Filters Complete

17th September 2009

Following the discovery of a design problem when using current sourcing drivers in the Decoder boards within the BPF’s a permanent fix has now been applied.

The SWR characteristics of the Filters were also measured for future reference. These have been used twice now and have performed excellently.

Full Size Megaloop

14th September 2009

Firing supports over trees

Supported Bands

- 160m
- 80m
- 40m
- 30m
- 20m
- 17m
- 15m
- 12m
- 10m
- 6m

Description

Nearly two years ago, we decided that we needed a better antenna for all band operation to replace our traditional use of wire dipoles. Unfortunatly we do not have the option of putting up many towers with a large quantity of yagi’s so the antenna had to be visually low impact, cheap, all band and allow us to be competative in contests.

Callum M0MCX and James 2E0YOM (now M0YOM) began researching various options but nothing appeared to fit the bill. The basic 1 wavelength loop was the obvious starting point as this provided the potential for all band operation.

After many hours of research, modelling and experimentation we discovered that the use of horizontal triangular loops would provide us with the antenna we needed. Traditionally this shape of loop is used in the “Delta Loop” configuration and is widely believed to be a good NVIS antenna when mounted horizontally and a DX antenna when mounted vertically. We discovered that this is not always the case, when the loop is small <=1 w/l and low to the ground it behaves as you would expect and is a reasonable NVIS antenna, one you start getting to sizes of >=2 w/l at over 1/4 w/l above ground then the loop starts to exhibit some very high gain at low angles.

After building smaller versions of the Megaloop we realised that our location contained 3 tree’s perfect for the support of a full top band megaloop at around 20-22 meters.

The loop is constructed from approximatly 174 meters of 14AWG hard drawn enamelled copper wire and is fed via a 2Kw 4:1 Balun using Aircell 7 coax.

For further details of the Megaloop and it’s construction go to the following links

The basic 80m loop

The 160m Loop

160m Analysis

The megaloop is a little over a full wave length on 160m and only 1/8th of a wave length above ground so as you would expect it behaves as a traditional NVIS loop, this works excellently for inter-G or close in european stations with some occasional DX. The modelled SWR curve shows a slightly high SWR throughout the band however the real Megaloop exhibits a much better SWR of around 2:1 although this is probably due to losses in the balun we are currently using as this is not rated for 160m.

Modelled Properties

Modelled Far Fields

80m Analysis

Performance on 80m is excellent for inter-G and EU with excellent gain for single hop central europe contacts. Low angle gain is also reasonable with similar performance to a high dipole at around 15 degress, NA contacts are not unusual with some occasional other DX. Modelling of SWR shows a fairly usable curve however the real world curve is much sharper than this and the antenna becomes unusable above 3.65, this currently requires a tuner for operation across the whole of 80m and we are working on an auto switching matching network for 80m to remove the need for a tuner. The megaloop starts to exhibit some directivity on 80m, with a slight tendancy towards the side oposite from the corner being fed.

Modelled Properties

Modelled Far Fields

40m Analysis

40m is where the Megaloop really starts to come into it’s own, gain on this band peaks at 10.4dBi with a relativly low angle of radiation. The horizontal pattern shows that although there are some slight nulls in places these are not too deep and still very workable. Contacts to anywhere on the globe can be expected. The modelled SWR shows the curve to be slightly high in the band, in practice the SWR curve is less than 2:1 over the entire band, this is probably due to slight modelling inacuracies. The antenna is excellent on 40m for smaller contests/DX/Special events however for larger contests its omnidirectional pattern and lack of FB can prove problematic and a directional receive antenna may be usefull in these situations.

Modelled Properties

Modelled Far Fields

20m Analysis

The Megaloop performs very well on 20m, with a maximum gain of 13.56dBi at only 13 degrees the DX potential is superb. The modelled 2:1 SWR curve covers the whole band and is confirmed by real world measurements. The far field pattern begins to exhibit some deeper nulls on 20m and it does have some dead spots. It is excellent for smaller contests/DX/Special events however the larger contests can be problematic because of the lack of FB, using a directional receive antenna would be beneficial.

Modelled Properties

Modelled Far Fields

15m Analysis

On 15m the pattern starts to become more unusual with some extremely deep nulls, the SWR curve is good for the whole band however there is some discrepancy between the measured SWR and the software models, probably due to minor modelling errors. When propegation exists in the direction of the lobes performance is excellent offering very low noise levels and giving a high S/N ratio. However due to the deep nulls a small Tri-Bander at 10 meters will sometimes outperform the loop if the propogation falls within its very deep nulls.

Modelled Properties

Modelled Far Fields

10m Analysis

The modelled SWR curve on 10m is slightly high, however in practice the 2:1 curve is good for the whole of the CW & SSB portions of the band. The far field pattern begins to get very messy on 10m, probably due to the size of the antenna, however gain of 14.4dBi is present at only 7 degrees with a double peaked front lobe. The potential for excellent DX is good however not much real world experience of the Megaloop on 10m exists due to poor propogation over the last 18months.

Modelled Properties

Modelled Far Fields

Why antenna maintenance is important

12th September 2009

Amateur Radio