Connect with us

VSWR Terminator

Discussion in 'Electronic Equipment' started by rjkfsm, May 26, 2005.

Scroll to continue with content
  1. rjkfsm

    rjkfsm Guest

    Hi,

    I need to test some antenna's. The customer wants them tested by
    sweeping with the operating frequencies (1MHz to 3GHz) by using a
    tracking generator, bi-diectional coupler, a spectrum analyzer and VSWR
    terminators. The procedure is to hook up all equipment through the
    production cable with the test terminator at the end and use that to set
    the spec an to mid scale and mark. Then attach the actual antenna and
    ensure that VSWR is less than the terminator. This is the procedure the
    customer wants and the customer is a 400lb gorilla. The customer gets
    what the customer wants.

    Does anyone know where I can get relatively high precision VSWR
    terminators with 'N' or 'BNC' connectors? I need: 1.8:1 (90 Ohms), 1.9:1
    (95 Ohms), 2.0:1 (100 Ohms), 3.0:1 (150 Ohms) and 3.5:1 (175 Ohms). 'N'
    connector is prefered, but BNC is acceptable.

    RK
     
  2. Robert Baer

    Robert Baer Guest

    BNC connectors are rather indecent going up to three giggles; use H,
    HN, or GR.
     
  3. bart

    bart Guest

    I'd like to see the 1 MHz antenna next to the 3 Gig unit!
    That sort of testing sounds a bit weird.
    A quick & dirty way is just using a Bird antenna analyser, a bit
    pricey but they get results fast. ( nothing up to 3 gig though)
    http://www.tessco.com/products/displaySkus.do?groupId=490&subgroupId=10
     
  4. bart

    bart Guest

  5. Russ

    Russ Guest


    OK, these are for a ship, so yes all the antenna's will be closely placed on
    a mast, however great thought by minds sharper than mine have devoted
    themselves to a layout that utilizes harmonics, dead zones, shadowing and
    ground planes to prevent interference.

    The antenna's are fed by heliac and most of the connectors are 'N' which
    would eliminate the need for adaptors. There are some H and HG connectors
    as well, but using adaptors for testing these is acceptable to the
    customer. Most of the adaptors we have are to or from 'N' style, so that is
    the preferred style of connector in our lab.

    Again, this customer is a 400lb gorilla. They have decided already how they
    want the antenna's tested. There is no altering what they want. I cannot
    use an SWR meter. Period. I wish I could. I already have several and it
    would make my life so much easier.

    Don't get me wrong: Thank you for your time in responding, however it does
    not meet what I need.

    RK
     
  6. The carbonbased lifeform rjkfsm inspired sci.electronics.equipment with:
    With the setup you describe the return loss is directly read from the SA
    when you attach the antennae, no need for an extra calibration.
    All you need to care about is the directivity of the coupler, which
    needs to better than the return loss you expect.
    Nop, sorry.

    Theo
     
  7. Fred Abse

    Fred Abse Guest

    Am I missing something here? Is your 400lb gorilla expecting his antennas
    to exhibit a lower VSWR than a precision termination?

    I'd be more worried about finding directional coupler(s) with sufficient
    isolation over that frequency range than finding precision terminations,
    which are fairly plentiful.

    Dunno how long your cables are, but their attenuation will mask out small
    mismatches, anyway, especially at the top end.
     
  8. Glenn

    Glenn Guest

    rjkfsm

    How about using some known attenuators with either shorts or opens on
    the far end to generate a precision reference R and therby a known VSWR,
    assuming Heliax, connectors and all are OK.
    Errors in this value along with directivity of the dual directional
    coupler reduce margin but it's unlikely that they are bad compared to
    antennas in the environment you describe.

    If you don't want to do the math, you should be able to measure the
    input resistance of a decent attenuator terminated in a short or open
    (keep it the same for all measurements) and safely assume that the R is
    constant with frequency and the the parallel Z is very high.

    If you don't mind doing the math, you can validate your method for the
    gorilla too.

    Glenn
     
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day

-