Tuning coax traps

I’m interesting in building some trapped wire dipoles using coax traps.

I don’t have an antenna analyser like an MFJ 259 or a dip meter, but can I not do something with a VNA (Vector Network Analyser) and or a Spectrum Analyser?
(I have access to those :wink: )

I’ve used the VNA in the past to fine tune an open ¼wave stub coaxial notch filter, nice and simple task as all coaxial, but wondering if I could do the same (use VNA to tune) with a trap?

Also, I’m finding lots of designs for trap dipoles for 10, 15, 20, 30, 40, 80, but I’m thinking less (at least dropping 80) something like 15, 20, and 40. Presumably I can just use 2 traps for three bands, how do I re-calculate the antenna sections between them?

Yes, you will have done a S11 measurement for a stub to effectively measure the return loss at a given frequency.

For a trap you want to measure on a two port VNA the S21 parameters of the trap. This again will give you the parameters of the trap at the wanted frequency but instead you are measuring through loss from port to port. Calibration is important with VNA’s otherwise they tell you very big lies.

You don’t necessarily need a VNA on the lower HF bands to setup traps, a simple Sig gen and a scope will do if they are setup properly.

I don’t have a VNA at home but I do have a HP 141 with a HB return loss bridge it is very useful for making measurements on a multitude of different stuff. There are many ways to achieve what you are trying to do.


Agree a VNA is overkill for an HF trap, but it is something I have access to :smile:
So would I be looking for, a good S11 and S22, and what response on S21?


You might find other designs of trap more appropriate for SOTA use (activating), being less bulky, sized for the lower power levels we have to handle. For example David, VK3IL, has a low-L design here:

Trapped five band EFHW SOTA antenna

There is also an interesting kit from SOTABeams for light-weight traps.

73, Simon

1 Like

Hi Simon.

Thanks for that. The SOTA kit certainly looks interesting, and looks likely to be easier to tune than coax!

I tried coax traps about ten years ago after a design appeared in PW. They were surprisingly heavy and detuned when it rained so I abandoned that idea. I also tried an OCF dipole but the weight of the balun at the feedpoint made it necessary to position it well down the pole, so I decided that it wasn’t suitable for SOTA.


Hi Brian,
I had considered seeing what power I could get with using RD316 which is lighter than say RG58 but not as puny as RG178, but I haven’t looked for how to work out the power handling yet.
Also I was planning for any outside stuff to use the liquid insulation tape.
I am looking at the trapped wire dipoles for both SOTA and an attic antenna for home.
For SOTA I will have more of a free hand so more bands probably better, at home space restricted so looking at less.

Have a look at the SOTA Beams web site and see their trick for tuning traps.


1 Like

Hi Ron. Do you mean their traps with the test points, and moving the coils on the ferrite to tune?
I do like the look of that I must say.

Your measuring loss. S22 is the same as S11, but on the opposite port.

You are looking for the lowest loss on a S21 for your desired bandwidth, which is specific to your calibration. If you change the bandwidth by increasing it, you loose your calibration.

To be perfectly honest with a dipole, you can not beat a good set of links to make it multi-band. That is what many of us do, and you will find its easy enough to reach up and disconnect the necessary links to achieve the desired band.

Traps are heavy and for the ease of changing band it doesn’t seem to make sense. If anything it gives you an excuse to move about and relieve cramp which is another problem with SOTA especially after a big walk.


1 Like

For a SOTA activator’s dipole, Jonathan is right, links on the dipole are the way to go as that gives the least loss.

You could make the links smart, use a tiny Arduino type single chip computer with Bluetooth and a latching relay driven by an output pin. Then you can string up your dipole and with a quick bit of software, use your mobile phone to send the messages to the smart links to open or close the relay. In Bluetooth Parkmode and with latching relays, the batteries in the links will last a long time. Make the links work in pairs (i.e. your program sends the commands to one link and it commands its sibling link in the other leg of the dipole) and you can swap bands quickly with just a swipe. If you address the other links then you can setup OCF style antennas by having links open in one side and closed in the other. Hopefully, the link board and relay will weigh less than a trap. It’s a winner in my book!

1 Like

I have had the exact same idea but using ZigBee. One day :smile: You could even modify RRT to do it in-sync with the spot.


Woooo! You are as warped as I am… scary! :wink:

1 Like

I assume you mean RG316 - if so avoid using soldered connectors if possible and use crimps instead. RG316 goes brittle when soldered and will snap after some use.

No RD316, or as some makers call it RG316D, the double screened version of RG316. We use the RD316 term at work to ensure purchasing buy the right thing!
Good point on the less than robust core though, though should be OK inside the confines of a trap I’d think.
I am leaning heavily towards the SOTABeam PCB based traps now. I could layout my own, and buy the bits, but to be honest, why bother, wouldn’t save me any money in the long run I doubt. I think I’ll use those for sure for my internal base antenna, but I am not swaying towards the lighter link dipoles for activations. Makes sense. And cheaper.

Agree Jim, it’s rubbish when you solder it, breaks in no time. But with RG-316 being Teflon unlike RG-174 it doesn’t melt when you try to solder it with aging hands and worn out eyes. Which is of course brilliant, until the blighter breaks. Putting you in a wonderful Kafkaesque scenario: either solder the wire that doesn’t melt but will break or try and solder the wire that doesn’t break before it melts.

1 Like

Surely it would be possible to use the CAT port on the transceiver to automatically tell the uProc which band you’re on - no need for fidgeting with a phone then.

I really think you should have a go at it Andy :smile:


The intriguing bit would be using DC to feed the relays over the feeder without using batteries on the links.

This has been done before on other systems, the hard part would be the return feed. Micro code would be easy. Rs232 catch and a few case statement commands.

Would do nothing to relieve cramp though so I can not see the point :wink:


It always seems to be the way that just after I’ve got up to change the links and settled back down again my phone (which up to this point has probably been sullenly refusing to talk to the network) chimes in with a spot on the band I’ve just moved off!

1 Like