I was curious if the performance of our 4O3A high power Bandpass filters could be improved with additional coax stubs in parallel. We were operating at ED1R for some time now with the added stubs, and we are pleased with the results. Check out the details!

In an earlier post I explained how to build coaxial stubs for 20m and 40m. Why 20m and 40m? A decade of hardcore contesting from stations all around the world has shown me that these two bands are the most critical ones, when it comes to In-Station interferences.

High Power Filters with additional coax stubs

The picture above shows the bank of 4O3A high power bandpass filters we use are ED1R. One filter per band, from 10m to 80m. On 20m and 40m a coaxial stub is connected in parallel to the filter’s entrance, using a PL259 T-Connector.

Results 40m

Click the diagram to enlarge

To be honest, I was very positively surprised about the result. In the diagram above you can see the S21 attenuation of the 4O3A 40m bandpass filter (red), the coax stub only (green) and the coax stub in parallel to the Bandpass filter (blue). The coax stub adds through its sharp notch another 30db(!) on the second harmonic. The attenuation from 7MHz to 14 MHz increases from 52dB to 82dB!

Results 20m

Click the diagram to enlarge

This diagram shows the S21 attenuation of the 4O3A 20m bandpass filter (red), the 20m coax stub only (green) and the coax stub in parallel to the bandpass filter (blue). Again, the stub’s notches on 7MHz and 21MHz clearly improve the overall attenuation on these bands. Attenuation of 40m is kicked up from 63dB to 97dB. As a side effect, 15m gets also additional 25dB, resulting in a total attenuation from 20m to 15m by 95dB.

Conclusions

The 4O3A bandpass filters are doing a tremendous job at ED1R. But unfortunately due to the direct harmonic relation of the two bands, the 20m station was still suffering severely when 40m was operating at the same time. The coax stubs give us additional attenuation between the two bands at a very inexpensive price point.
With the stubs in parallel to the bandpass filters, the interferences are still not completely gone, but at least have been reduced by another 30dB.

I can unconditionally recommend the addition of coaxial stubs in parallel to your filters if you also suffer from interferences between 20m and 40m.

At our contest station ED1R we have no space for permanent Lowband Verticals. Each Contest, the verticals have to be installed temporarily in a field nearly. Especially the deployment and removal of the antenna radials is very time consuming. Over time we optimized our system which reduced the required time from 2 hours down to 30 minutes.

Fastron is key

Screws & nuts are a bad choose when things have to be done fast. Changing to Fastron terminals improved significantly the speed of deployment. We mounted on the ends of each radial a male and female terminal.

We also added to male Fastron Terminals on each position of the radials plates.

The female connector of the radials are plugged to the male connectors on the plate. No screwdriver is needed. Just rapid Plug & Play.

Radial carts

When we started using Vertical antennas on 160m and 80m, especially the removal and storage of the radials was extremely time intensive. We had to roll up each radial individually, fix it with a piece of tape or a tie-wrap. The picture below gives you an idea how much work it was.

During my stay in Qatar, where I was invited to operate CQWW SSB 2011 as A73A Dave, K5GN showed me a smart & inexpensive way to deploy & store coax cable – Garden hose carts!
They are available in all DIY stores for less than 20€ a piece. At ED1R we applied this to the radials. Look how they are stored now:

Now you might get an idea why we crimp beside the female Fastron connector, also a male connecter. Whenever the radials are removed from the plate, they are connected in chain to the form one single long wire on the cart.

Deployment & Removal

Radials are now very easy to deploy. In practice we found that radials are fastest deployed with a team of 3-4 persons. While on person stays at the cart and connects the ends of the radials to the plate, the other 2-3 deploy the radials in the field, directly from the cart.

Radial removal is best done by two persons. One disconnecting the radials from the plate and connecting them into a consecutive chain on the cart, while the other person turns the winch of the radial cart.

Resumé

The time needed for deployment and removal of radials (approx 40 – 60 radials per Vertical) could be reduced from 2 hours to less than 30 minutes with this technique. It takes some initial time to crimp the fastron terminals on the cables, but once the solution is in place, signifiant time is saved whenever the radials have to be deployed.

This question bugged me after trying to exchange the connectors of a 25m long Aircell7 coax cable. During preparation I discovered a black inner conductor. I seems water entered the cable. Do I have to throw it away? Read more to discover a probably unexpected answer!

Water in the Cable

It’s the nightmare of all hams – water inside the coax cable. Over years, I heard over and over that once water has entered the cable, it becomes useless and has to be thrown away. From my times at university I remembered that due to the Skin Effect, RF energy tends to flow on the skin, rather than the inner core of the conductor. Since copper oxide conducts much worse than copper, theory seems to confirm the country saying.

But considering the price of 2,50€/m I felt a stich in my heart when thinking about throwing away this 25m role of Aircell7. Instead of accepting defeat, I pulled out my Vector Network Analyzer and measured the attenuation from 1…500MHz.

Oxidized coax cable

It seems water entered from one side of the coax cable. While I had to cut off 2 meters to reach the part of the cable where the shield was not oxidized anymore, the inner conductor was completely oxidized from the beginning to the end.

Measuring attenuation

Click to enlarge the diagram

Before measuring S21 attenuation I scratched of the copper oxide and put tin-solder on the ends to ensure a proper contact. Surprisingly, the results where much better than expected!

Results

Will I throw the cable in the trash can? No! Below 30MHz, the attenuation increased with 0,2dB only marginally. Even on 144MHz it could still be used in an emergency situation. Only above, the skin effect really starts to be noticeable. With 7dB additional attenuation of 432MHz, it’s definitely not suitable for UHF anymore.

I hope this measurement might also helps you to take the right decision when you find next time a coax cable with an oxidized inner conductor. I didn’t expect this result. I just saved 50€ and will continue using the cable on Shortwave.

Until today, I’m fascinated what can be achieved with good filters. At our Contest Station ED1R we bought a Triplexor (10m/15m/20m) in order to be able to use our new Optibeam OB11-3 on the three bands simultaneously.

Motivation

At ED1R we only have one tall tower with 20m height. While we are able to finish in contests within the Top10 of our category, results showed that the highbands need to be boosted to get a permanent seat in the Top3 ranks. Since Vertical steel equity is restricted, we thought that a big Tribander on top of the tallest tower with a 4O3A Triplexor would give us the biggest bang for the buck.

After the successful installation of the Triplexor, each of the three highbands (10m / 15m / 20m) has now 3 Antennas available:

1. Rotatable Monobander
2. Monoband Yagi fixed towards USA or EU
3. OB11-3, connected through the triplexor

Schematic

The schematic below shows the setup:

A prerequisite for the Triplexor is a set of High Power Bandfilters. Achieving >100dB attenuation on a harmonic band is quite and engineering challenge. In order to ensure proper operation, the input terminals of the Triplexor have to be terminated at any time with 50Ohm. Therefore I built three switchable dummy loads.

The picture above shows the Triplexor mounted on a timber panel with the three switchable dummy loads

The switchable dummy load consists mainly of a 50Ohm/100W RF resistor, connected to a heatsink and a 16A/230V relais. If no power is applied, the input terminal from the Triplexor is always connected to the load. For wiring high power Teflon cable has been used. An LED indicates the usage.

This shows the new antenna switching system of ED1R. Four homebrew Stackmatches, a 10×2 Antenna switch, the 4O3A High power Bandfilters and the 4O3A triplexor.

Same setup, just with all wires connected

I updated and included the Dummy-Load switching into the software of our Microcontroller based Stackmatch Controllers. Now the relays inside the load will be triggered whenever the OB11-3 is selected.

Triplexor in Action

After the successful installation, Javi, EC4DX recorded a short video to demonstrate the performance of the Triplexor. Unfortunately we missed to recored the harmonics on 10m while transmitting on 20m. But the video gives you anyhow a good idea about how smooth it works.

Conclusion

I invested in design, hardware and software roughly 40 hours during various evenings and weekends. Ranko’s Triplexor is a truly great invention. Except of a small portion (+-20kHz) of the 20m’s harmonic on 10m, almost no interference can be noted. It’s almost a bit spooky to send 1,5kW on 10m and 15m though a coax cable while listening listening on 20m with the same antenna.

A few days ago I had the time to build and try a Double Half Delta Loop (DHDL) at our Contest Station ED1R. The Antenna was built within 3 hours and can be deployed / errected within 20 minutes.

The Antenna

The Double Half Delta Loop (DHDL) was developed by George, AA7JV for his Lowband Expedition to Chesterfield (TX3A). It’s a ground independent Receiving antenna which only needs two 10m support poles. Once the wire’s are cut and the Balun build, it takes less than 20 minutes to deploy the antenna.

A detailed description can be found at OK1RR’s website, the DHDL simulation files can be downloaded at TX3A website and IV3PRK’s report about two broadside phased DHDLs is also worth reading.

3D View on the Simulation Model

DHDL in the backyard of ED1R

DHDL from the feedpoint towards the second pole

DHDL 1:18 Transformer & Resistor in their enclosures

Supporting fibre pole in the center at 1,5m

RF Choke at the base (25 ferrite beats) is crucial

With the specified dimensions, this non-resonant loop is optimized for 80m. On 40m, the Front/Back is too low and on 160m the signals will be very weak. The diagrams below show the patterns for 40m, 80m and 160m.

The antenna’s feedpoint impedance alters between 1000 – 1200Ohm (see below). I built a 1:18 transformer on a Binocular core (Fairrite #2873000202) and terminated the loop with 1000Ohm.

Results

Despite another receiving antenna for comparison, it eliminated very well the QRM we have at ED1R on 80m Fullsize Deltaloop. On 80m I noted a very good F/B while switching between the Transmission antenna and the DHDL. As expected, on 40m the F/B was only marginal and on 160m the Signals were too weak.

Bottom line: The DHDL is a solid performer for 80m. On 160m a preamp is definitely required and on 40m the antenna is not that useful anymore.

A few words of precaution

• Make sure you don’t listen on the coax cable. Put sufficient ferrite beats on the coax, close to the feedpoint and close to the shack. Ground the cable in 3m distance to the feedpoint.
• Make sure that metalic structures don’t disturb the antenna. Fences, Towers, Verticals in the close vicinity will create problems and couple into the DHDL. Make sure that these metal objects are at least half a wavelength away. If that’s not possible you will have to de-tune your transceiving antennas – but that’s another chapter

At our Contest Station ED1R we have on each highband 3 Yagis available. In 2012 I built 4 high power Stackmatches in order to switch and combine them properly. Since I never liked the way of WX0B’s stackmatch controllers with the rotative knob, a microcontroller and button based solution was the answer. After a few contests the solution has proven it’s reliability and comfort.

The platform

My good friend Pablo, EA4TX suggested to reuse his ARS Interface box, instead of developing everything from scratch. A welcome offer! Pablo’s ARS box, which is a USB powered, enhanced replacement/addon control for any kind of rotator, brings everything I needed:

• Microcontroller with USB Interface
• 5 Relay based output ports
• Available Input for Pins
• LCD Display
• 4 Buttons

So all I had to do was “just” to write a new Software, adding two plugs and connecting my cables.

Here are the main features of the ED1R Stackmatch Controller:

• Names can be loaded for all antennas
• PTT Sensitive to select different RX and TX antennas
• Control through buttons or USB messages
• Ready for 4O3A Triplexor
• Comfortable Configuration through Windows application

Configuration Software

This little Windows program allows a comfortable configuration of the individual Stackmatches through the USB port.

Pablo, EA4TX has already developed an TCP/IP based client/server which allows us to control the Stackmatches from any operator position. This is another great advantage. In a competitive M/S or M/2 environment all three Stations need access to the Antennas.

One of last years projects was the improvement of our Contest Station ED1R. For the various Yagis we needed smart ways to combine antennas. Instead of buying commercial stackmatches (antenna combiners) I decided to build them up buy myself and adjust them to our needs.

HF part

A great resource is Mike’s, SM2WMV (SJ2W) website. Mike is a professional RF engineer and he designed beautiful Stackmatch PCB’s which he also makes available for purchase. Instead of spending hours in designing my own stackmatch PCB, I preferred to buy Mike’s PCBs – especially at this price – a no-brainer.

Basically all I had to do is purchasing the missing components, drilling the holes in the box and installing everything. I used two stacked FT240-61 ferrit cores for the transformers. All the transformers consist of 4 (10m & 15m) or 5 (20m & 40m) trifilar windings of 2mm enameled copper wire. Since impedance is very low (17 – 50 Ohm), no high voltages are expected. That’s why no additional (e.g. Telfon) insulated wires are necessary.

In order to fix the Balun properly on the PCB I purchased from Hans, DK3YD high temperature resistant Delin washers.

Above you can see the aluminium box I bought at the local electronic store. The PCB fits perfectly into the box.

and for outside.

This is how the stackmatch looks assembled

Measurements

S11 between the Input and the Output Ports is always better than -20dB (click to enlarge)

Isolation between the ports is between -60dB (7MHz) and -40dB (28MHz) (click to enlarge)

S11 and S21 of two antennas combined (3dB Splitting) (click to enlarge)

S11 and S21 of three antennas combined (6dB Splitting) (click to enlarge)

Stackmatches installed

And finally all the stackmatches installed at ED1R above the 8-Pack and the 4O3A High Power Bandpass filters

At the second HamRadioWeb convention in Bolognia, Italy Claudio, I4LEC has given a great presentation about the current status of SDR. In his presentation he starts with SDR principals then analyzes then the strength and weaknesses of available Software Defined Radios (mainly FlexRadio Products) and finishes with a comparison against TS-590 and Elecraft K3. The presentation is entertaining and informative at the same time. Definitely worth having a look at!

Claudio was so generous the grant the permission to show it on my website!

With his low cost, high precision Vector Network Analyzer, Tom (DG8SAQ) started a revolution in Ham Radio. Thanks to his efforts, amateur (and professional!) enthusiasts have now the possibility to perform measurements which were reserved to well funded HF laboratories before. In this video, Tom explains the evolution and capabilities of his VNWA and why even NASA relies on it now as well!

The presentation was held during the Ham Radio Hamfest in Friedrichshafen, Germany in June 2012. Tom is speaking in German, but Kurt Poulsen, OZ7OU has translated it into English!

Enjoy!

During the summer break it became quite here on my blog. While Instead of chilling at the beach, I spend a lot of time on new Hardware and Software projects. Most of them are done, and now it’s time to document them.
In this blogpost I want to share the construction of a versatile 4:1 Balun. In combination with ladder line and an automatic Antenna tuner you can use any dipol on almost all frequencies.

Why do you need a Balun?

Depending on the length of your dipol, the impedances at the feedpoint can get quite high. Built-in Automatic Tuners usually cover only impedances between 20…200Ohm. With the 4:1 Balun, we transform the high impedances down into the range of the automatic tuner.

Design

I decided to go with the old “Ruthroff” Design which can be seen in the picture below. It’s a transmission transformer and does not provide galvanic isolation. My BalUn (Balanced-Unbalanced Transformer) is broadband and can be used from 1..30MHz.

10 bifilar turns are mount on an Amidon FT-240-61 ferrite core. I used 1,5mm PTFE (Teflon) insulated wire. With high impedances, high voltages can appear on the wires, so it’s better to have a proper insulation!
The Balun is mounted in a sheet metal housing which I bought at the local electronic store. The parallel wire (450 Ohm ladder line is connected through two Banana jacks.

Assembly

The picture above and below show the bifilar turns on the FT-240-61 ferrite core. Note: This was the first prototype which only had 7 turns!

In order to fix the core properly, I used on both sides of the balun a Delrin (POM – Polyoxymethyline) washer, which was mounted with a long M4 screw on the chassis.

The Balun in action is shown below

…with my 2x5m long Inverted-V Dipol

Some Considerations

As mentioned in the footnote of the pictures I built two versions of the 4:1 BalUn. The first one with 7 bifilar turns and the second one with 10 bifilar turns. As it turned out, the 7 turn version was only usable from 7…30MHz. But since I wanted to create a universal, broadband BalUn covering 1.8MHz … 30MHz I had to add another three turns.
When winding the turns on the ferrite core, make sure the wires are always in parallel. Avoid any gap between the wires, since this results in an undesired impedance variation. I always fixate the two wires every 2-3cm with a small strip of adhesive tape.

Performance

I measured the Performance with my DG8SAQ Vector Network Analyzer (VNWA). Since the BalUn has a ratio of 4:1 I had to terminate the antenna-end with a 200Ohm resistor. The plot below shows the S11 / VSWR curve from 1..30MHz which does not exceed 1:1,36.

I’m satisfied with the result.

S11 / VSWR of the 4:1 Broadband BalUn – Click to enlarge

Resources

• Ferrite Cores, PTFE insulated wire and Delrin washers can be bought from DK3YD / Bausch-Gall GmbH.
• DK3YD also sells professionally manufactured 4:1 BalUns BR-41-100-A (low power) and BR-41-750-A (high power)
• Martin Ehrenfried, G8JNJ published a nice article about the construction of Ruthroff Transformers