Shakespeare and Coax Stub Filters

Discussion in 'Amateur Radio News' started by VK6FLAB, Apr 14, 2018.

ad: L-HROutlet
ad: l-rl
ad: Left-2
ad: Subscribe
ad: Left-3
ad: L-MFJ
ad: abrind-2
  1. VK6FLAB

    VK6FLAB Ham Member QRZ Page

    Foundations of Amateur Radio

    Shakespeare and Coax Stub Filters

    If you read it on the Internet, it must be true. but what happens if you read it and there are 700 different answers?

    In my day job I search countless times a day for answers to problems. Based on my experience I can look at a list of responses to a question and tell myself what the skill-set is of the poster, they don't know what they're talking about, they're guessing, they've got no clue, they tried it, ah, this one knows what they're talking about.

    As an aside, a company once advocated that we should use social media as a way to provide support to customers, but based on my experience, seeing the correct answer in a series of posts being voted down into oblivion and seeing the wrong answer being promoted is a fantastic example of why that won't work, ever. Infinite monkeys with typewriters might eventually write Shakespeare, but it will take an infinite amount of time and before they succeed there will be a whole lot of rubbish.

    When I started researching magnetic loop antennas several years ago I went through the same process, search for answers online. I found lots of different stories, opinions, measurements, contradictory statements and formulas. I spent some weeks reading everything I could on the subject and after a while a picture started emerging that started to explain to me how a magnetic loop antenna works. I'm no expert, my foray into this died when two ADSL modem transformers died within seconds of me hitting the PTT on my radio and I sort of lost interest. I have a magnetic loop antenna standing behind me, on loan from a friend and I use it to scan the bands. It's compact, easy to tune and one day I'll make more than a single contact on it.

    All this to say that I've been investigating coax stub filters.

    If you're not familiar with the notion, you can cut a piece of coax cable to a specific length and connect it with a t-piece to your antenna feed line. If you do that, depending on the length of the coax you cut, you get interesting effects. These effects include filtering, or notching out frequencies, passing other frequencies and all in all affecting what your radio is able to receive and transmit.

    If you've ever set up your radio with some friends nearby for a field day, a contest or a camp-out, you might be surprised to learn that even across the space of a field or a caravan park you'll be able to hear the other station, even if they're not on the same band. You might hear their actual voice, or more likely, you'll hear all manner of overload sounds that essentially show up as noise blocking out the station you're really trying to hear and work.

    Of course, the reverse is also true. When you're transmitting, your friend is hearing the same horrible gunk coming out of your radio.

    One of the ways that you can manage this is to set up filters, either notch filters which reduce the strength of undesired transmissions or pass filters which only allow certain frequencies to get to your radio. Combining these will make your life much easier.

    Coax stub filters are a tried and true method to achieve this and the Internet is full of expert opinion on how to exactly do this. With infinite budget and time, you can try them all out and with your trusty network analyser you can find the combination that works just for you, but in the real world you have a limited amount of coax, money and your lottery winning didn't cover that network analyser.

    I started this process in earnest two or so weeks ago and frankly I'm no closer now than I was then. I'm still in the reading articles stages. I think I'd like to create two sets of band pass filters and connect each set to a radio. If I'm on 10m, I set my radio to use a 10m band pass and set the other radio to use a pass filter for their band. I figure between the two of them I have something that resembles what I'm looking for.

    If wishing made it so.

    To make matters more interesting, I have two rolls of Quad Shield RG6 coax left over from my days climbing roofs and installing satellite dishes for broadband Internet in the bush. If you're familiar with that you already know that this is 70 Ohm coax. Of course none of the documentation I've been reading even talks about using 70 Ohm coax. I like it because it's strong, low loss and I have loads of it.

    I've been reading everything I can get my hands on. Friends have been sending me emails with copies of copies of copies of a hand-scribbled note written by a guy 30 years ago with the magic combination. I have several of these, all different. This morning I came across some electrical circuits that purport to do the same thing with a handful of components.

    One thing I do know, if the Internet has a thousand answers on how to do something, no-one really knows, or one of them does and is buried among the not-yet correct copies of the works of Shakespeare written by monkeys.

    I'm Onno VK6FLAB

    To listen to the podcast, visit the website: and scroll to the bottom for the latest episode. You can also use your podcast tool of choice and search for my callsign, VK6FLAB, or you can read the book, look for my callsign on your local Amazon store, or visit my author page:

    If you'd like to participate in discussion about the podcast or about amateur radio, you can visit the Facebook group:

    Feel free to get in touch directly via email:, or follow on twitter: @VK6FLAB (

    If you'd like to join the weekly net for new and returning amateurs, check out the details at, the net runs every week on Saturday, from 00:00 to 01:00 UTC on Echolink, IRLP, AllStar Link and 2m FM via various repeaters.
  2. VK3KTO

    VK3KTO Ham Member QRZ Page

    I always thought Will Shakespeare must have known a thing or two about inductance, as I seem to recall he made mention of mortal coils.
    73 de Mike VK3KTO
    N9PBD and VK6FLAB like this.
  3. K5VOU

    K5VOU Ham Member QRZ Page

    I did something like this BFI (before the internet) and just cut up a ton of coax before I gave up. The application was harmonic suck-outs for a two-meter amplifier. I did learn that the 1/4 wave stubs were pretty good at reducing the harmonics but they were also very sharp and not very frequency agile so not good for a third, fifth etc notch filter to be used with a two meter radio changing frequencies all the time. They were not so good for creating a band pass or a band stop filter but they realy did reduce the thirds in a fixed frequency power amplifier.

    Your RG-6 should work just fine as the impedance of the coax is not nearly as important as knowing the actual velocity factor of the coax. An MFJ-269C ought to let you measure everything you need to measure like coax length in wave lengths and a long tape measure will tell you how long the coax is physically thus giving the data for true velocity factor.

    Of course your trusty vector analyzer would come in handy too as would a spectrum analyzer with a tracking generator.

    Tom K5VOU
    N9PBD and VK6FLAB like this.
  4. K7RHT

    K7RHT Platinum Subscriber Platinum Subscriber QRZ Page

    And thus the need for an unlimited budget :)
    VK6FLAB likes this.
  5. KQ6XA

    KQ6XA Ham Member QRZ Page

    As Good Luck Would Have It...
    To Break The Ice, the voice of a professional RF playwright speaks, steeped in the Brave New World of such shorted stub coaxial devices; yet I wear not My Heart On My Sleeve.

    I offer but a few humble suggestions for The Mind's Eye of the budding ham to gain a stunning performance within the Full Circle of the theatre of electomagnetic excellence:
    • When The Game is Afoot, a vector network analyzer provides a Look at What is Best: a Tool Thou Shall Draw, for both the final test and the initial characterization of the coaxial element, For Goodness' Sake.
    • It is yet a Forgone Conclusion: the impedance of the coax is critical, and mismatch is not for the Faint Hearted.
    • Give the Devil His Due, for upon the velocity factor of the coaxial cable: Thou Must Be Counted!
    • Playing Fast and Loose, with a mere Fancy Free inline coaxial T and shorted stub, Doth Provide No useful odd harmonic suppression.
    • RF Computer Modeling can be quite helpful, but 'tis not the Be-All and The End-All.
    Since Brevity Is The Soul of Wit, I bid you Adieu, adieu, adieu.
    WE4E, N9PBD, VK6FLAB and 3 others like this.
  6. AG2AA

    AG2AA Ham Member QRZ Page

    so to translate what bonnie said, and based on my VERY limited experience with coax stubs,

    1 - Impedance matters.
    2 - velocity factor REALLY matters
    3 - where you put the "T" matters.
    4 - you didn't mention, but the differences between open and shorted stubs matters
    5 - 1/4 wave and 1/2 wave stubs do different things

    see also smarter than me at
    KQ6XA likes this.
  7. W1VT

    W1VT Ham Member QRZ Page

    Last edited: Apr 16, 2018
  8. KD6LOQ

    KD6LOQ Ham Member QRZ Page

    AG2AA's list is excellent info. And don't forget temperature and humidity. What you finally get to work in Florida summer, probably will react differently at that vacation spot in Nebraska during winter.
    As for Internet advice, remember engineering is making the best guess possible based on insufficient information from people of questionable knowledge.

    Experiment, document, redesign, rinse, wash, repeat.
  9. K1FBI

    K1FBI Ham Member QRZ Page

    I had an excellent marine antenna I used for 10 meters...made by Shakespeare.
  10. KA9UCE

    KA9UCE Ham Member QRZ Page

    I love swapmeet duplexers and cavity filters.
    I check them for condition, frequency range and tuning range, including notch, if so-designed.
    These I tune to offending signals, notch the TX, sweep the pass.
    I build comb filters from the old DB flat packs...remove the internal cables and test each cable and cavity.
    Singles should be able to tune/notch no less than 8dB.
    Never expect more separation from a single cavity than the design specs provide, it will never happen.
    Interconnecting cables are not used, as you have already changed the selectivity and notch depth when you pulled it apart....start from scratch.
    Personal preference is type 'N', but in small systems, BNC has uses, but not for high power (+75W).
    Use a return loss bridge, vector analyzer, or a service monitor with track-gen.
    Cavity filters are a pain if you need to use more than one or two, but they sure do help when you need to drop an offender's signal by several dB, and you can easily add more for a deeper notch.

    Just my take on this.. Stub tuned filters were used by Wacom in their duplexer designs, but the application fell out of favor due to these being so narrow-tuned, and prone to damage because they were exposed.

Share This Page