NanoVNA cable measurements

Having a VNA opens up so many possibilities for RF measurements. Typical measurements are antenna matching, RF filters, impedance matching and cable loss. But it an do more. You can use it even to measure the coaxial cable length – a task that is normally performed with time domain reflectometer (TRD), an instrument I have some knowledge about as I have designed/built few such circuits (Time Domain Reflectometer pulse generator, TDR kit, Avalanche pulse generator and Potato chip to TDR circuit).

Accurately measuring cable length with NanoVNA article will go through some mathematics on deriving TDR response with VNA data. Remember that the VNA does its measurements in the frequency domain. If we transform the frequency domain data into the time domain, we should see the time domain nature of our measurement (with certain limitations though). By using the magnitude and phase of the signal measured throughout the frequency sweep, we can compute the distance from where reflection occurred.

NanoVNA To Test The Loss & Length Of Coax Cables by Jim W6LG YouTube Elmer for Ham Radio Basics

The NanoVNA does many things. Two of which are testing coax cables for loss and length. Jim W6LG shows how to do both in this first video about the NanoVNA. In a subsequent video, Jim will show how the NanoVNA can be used to check antennas for resonance, impedance and reactance. The software that Jim uses is NanoVNA Saver. That software is free and easy to use.
Here is the link to the software :

#316: Use NanoVNA to measure coax length – BONUS Transmission Lines and Smith Charts, SWR and more

nanoVNA – Coaxial Cable Measurement Methods (Characteristic Impedance and Cable Loss) – VE6WGM

1. Calibrating the nanoVNA
2. The Smith Chart – Briefly (Begins at 8:24)
3. Measuring the characteristic impedance of coax cable. (Begins at 11:46)
4. Measuring coax cable loss. (Begins at 22:34) *Correction: At 23:34 I stated that the nanoVNA has a directional coupler. I have since discovered that this is not accurate information. The nanoVNA actually uses a bridge to make it’s measurements. My apologies. For more information on how this works, please see this excellent video here:

#502 NANOVNA Demystified

#37: Use a scope to measure the length and impedance of coax

Cable Basics; Transmission, Reflection, Impedance Matching, TDR

Reflected waves on a cable


  1. Tomi Engdahl says:

    nanoVNA – Coaxial Cable Measurement Methods (Characteristic Impedance and Cable Loss) – VE6WGM

  2. Tomi Engdahl says:

    1EZ53September2004-Subject to change J. SimonProduct: Vector Network Analyzer ZVBMeasuring Balanced Components with Vector Network Analyzer ZVB

    Balanced RF components are advantageous compared to traditional single-ended components, since they cause less EMI, and are less susceptible to EMI. This application note describes the fundamental concepts of differential and common mode signals and of mixed-mode parameters, which are essential for balanced components. Techniques for the measurement of mixed-mode parameters are presented.

  3. Tomi Engdahl says:

    nanoVNA – measuring cable velocity factor – demonstration – open wire line

  4. Tomi Engdahl says:

    nanoVNA-H – measure ferrite transformer – Noelec balun

  5. Tomi Engdahl says:

    #316​: Use NanoVNA to measure coax length – BONUS Transmission Lines and Smith Charts, SWR and more

  6. Tomi Engdahl says:

    Has anyone in the group used a nanovna to run TDR tests on twisted-pair cables to measure length of cable or distance to a fault? Although not a coaxial cable, most references list a characteristic impedance of about 100 ohms and velocity factor at about 0.65. Does anyone have tips for using a nanovna calibrated for 50 ohms to measure such twisted pair cable?

    For CAT5/5e certification we produced plots (on each of the four wire pairs) of attenuation, RL, NEXT, FEXT as well as measurements of delay/m, capacitance/m, resistance, length, and impedance. The Lantek had a differential driver sourced through a RL bridge.

    I was contemplating using a BNC-to-SMA adapter, along with a BNC-to-Banana Jack adapter to connect an individual pair. Obviously not ideal, but if “calibrated” to a known length of Cat-5e or Cat-6 cable, it might give a ballpark TDR measurement.

    I hadn’t thought of using a 50:100 ohm balun as the interface. Would one of the CCTV video baluns possibly provide a better match to the nanovna? Here’s one example of what one can get on the web for relatively cheap money:

    Since most of these top out at about 8 to 10 MHz, you’d have to limit the test frequency to something like 5 MHz.

    Don’t bother, try clip leads or banana jacks and see what happens.
    Sure, you get a bump from the mismatch at the beginning of the cable, but what you want is the length of the cable and whether there’s any damage, and that will reflect, regardless of the impedance of the cable.

    Make two and you can put a 100-120 ohm termination on the other end.

    On another aspect of TSP: TP by itself is nominally 90-100 ohms
    characteristic impedance (somewhat dependent on the dielectric, dielectric
    thickness, and twist tightness). This is without shield, just plain ol’
    TP. Now add the shield. In an EMC course I once took from Howard Johnson,
    himself, one of his demos clearly showed that 80+ % of the fields of each
    conductor of the TSP close onto the shield. That leaves only 20% of the
    induced fields to close between the TP conductors, as intended and
    modelled. So, what’s the deal with 90 to 100 ohms characteristic impedance
    between the conductors of the TSP?

    Also, from practice, after some 10-lamnda (my own estimate) in TSP, most of
    the energy becomes differential mode between the TP conductors and the
    shield – much like coaxial cable, which is considered a common mode cable
    (NOT TP or open wire feeders in the case of antennas). However, the shield
    embodied in the TSP of today is no where near of the integrity of the
    shield on a good grade of 50 (or 75) ohm coaxial cable. So, the stuff
    leaks like a sieve.

  7. Tomi Engdahl says:

    . say you have a coax choke … then connect the
    shields to the hot ends of your test jig …

    if you have a two wire (twisted pair) then connect only ONE of the two
    to your test jig (does not matter if the “white” or “black” wire used
    cause they should be identical)

    for trough loss you connect both wires on input and output of your balun

    for common mode surpression only connect ONE on input and output
    (normally shield used if you have a coax choke)

    a good choke not only has good match on input and output and low trough
    loss … a good choke also has a high isolation (common mode
    surpression) .

  8. Tomi Engdahl says:

    NanoVNA – Testing the Common Mode Attenuation of a DG0SA 1:1 Current Balun by VE6WGM

    NanoVNA and FT240-43 1:1 50 Ohm Current Balun

  9. Tomi Engdahl says:

    Calculate the Common Mode Rejection Ratio (CMRR) of a Balun

  10. Tomi Engdahl says:

    #316​: Use NanoVNA to measure coax length – BONUS Transmission Lines and Smith Charts, SWR and more

  11. Tomi Engdahl says:

    #95: Three Methods to Measure Impedance with the NanoVNA

    Are you aware that there is more than one method to measure impedance with a VNA? You might find this video interesting.

  12. Tomi Engdahl says:

    A surge of energy on a finite transmission line will see an impedance of Z o {\displaystyle Z_{\text{o}}} {\displaystyle Z_{\text{o}}} prior to any reflections returning; hence surge impedance is an alternative name for characteristic impedance.

    The analysis of lossless lines provides an accurate approximation for real transmission lines that simplifies the mathematics considered in modeling transmission lines. A lossless line is defined as a transmission line that has no line resistance and no dielectric loss.

    Z = sqrt ( L / C)

    In particular, Z o {\displaystyle Z_{\text{o}}} {\displaystyle Z_{\text{o}}} does not depend any more upon the frequency. The above expression is wholly real, since the imaginary term j has canceled out, implying that Z o {\displaystyle Z_{\text{o}}} {\displaystyle Z_{\text{o}}} is purely resistive. For a lossless line terminated in Z o {\displaystyle Z_{\text{o}}} {\displaystyle Z_{\text{o}}}, there is no loss of current across the line, and so the voltage remains the same along the line. The lossless line model is a useful approximation for many practical cases, such as low-loss transmission lines and transmission lines with high frequency. For both of these cases, R and G are much smaller than ωL and ωC, respectively, and can thus be ignored.

  13. Tomi Engdahl says:

    Cable Impedance Calculator

    Coaxial Cable Impedance Calculator

    Pasternack’s Coaxial Cable Impedance Calculator allows you to enter the Outer Diameter Dielectric width, Inner conductor Diameter width and either the Dielectric Constant or Velocity of Propagation (VoP) values in order to calculate the impedance of the coax.

  14. Tomi Engdahl says:

    Twisted-Pair Impedance Calculator
    A tool designed to calculate the characteristic impedance of a twisted-pair cable

  15. Tomi Engdahl says:

    #564​ NANOVNA Coax Loss Measurement

  16. Tomi Engdahl says:

    I almost never post because I try to manage myself, but I don’t know how to achieve a measure and my search on the net does not give me much explicit results.

    How can we measure losses in a system already installed (coaxial cable, connectors, CMC, filters, etc.) and where it is not possible to access both ends with nanoVna?

    Thank you in advance for your help


    One testing strategy is to have RF noise generator on one end and spectrum analyzer on other end. You get idea of amplitude response and attenuation.

    You can measure the return loss. You have the remote end of the cable either open or short circuit, which reflects back all the RF. Short circuit is preferred as a short is more definite than an open circuit at RF. You then measure the amount of signal coming back. If the return loss is say 14dB, then the cable loss is 7dB i.e. the signal has travelled there and back.
    Hope this helps

  17. Tomi Engdahl says:

    If you’re testing a 75 ohm cable, nanoVNA uses a 50 ohm bridge, so allow for two 0.177 dB mismatch losses at the VNA in addition to the cable loss. Not a big error in a long cable run.

  18. Tomi Engdahl says:

    #564​ NANOVNA Coax Loss Measurement

    #583​ NANOVNA Not all 50ohm loads are 50ohms

  19. Tomi Engdahl says:

    #316: Use NanoVNA to measure coax length – BONUS Transmission Lines and Smith Charts, SWR and more


Leave a Comment

Your email address will not be published. Required fields are marked *