Using RG-6 / U as a feedline?

Discussion in 'Antennas, Feedlines, Towers & Rotors' started by KK4NSF, Jul 26, 2021.

ad: L-HROutlet
ad: l-rl
ad: abrind-2
ad: L-Geochron
ad: MessiPaoloni-1
ad: HRDLLC-2
ad: Left-2
ad: L-MFJ
ad: Left-3
  1. K7JOE

    K7JOE Subscriber QRZ Page

    Good explanation!

    I have a selectable 50 and 75 ohm watt meter so presumably I can use this to take some practical readings.
    Easy to terminate the far end of the load with a 75 ohm load...then measure loss... 50 ohm system and 75 ohm system ... all using correct impedance for measurements.

    Could one simply re- calibrate the Nano-VNA to a 75 ohm load ?
  2. W9IQ

    W9IQ Ham Member QRZ Page

    Thanks, I am glad you liked it.

    Yes it is easily done with a nanoVNA when used with the nanoVNA-saver software. The software allows you to specify the SOLT parameters. All reported values, including the pertinent S21 (loss) parameter, will then be correctly reported.

    - Glenn W9IQ
  3. K7JOE

    K7JOE Subscriber QRZ Page

    Great question.

    It so happens that a half wavelength (electrically speaking) feedline of any impedance is magic -- it will show the exact antenna impedance at the other end of the feedline (at the radio end ).

    This is frequency dependent of course, but works well to cover a single ham band with non-50 ohm impedance coax.

    The feed line must be exact electrical length - for this replication scenario. The length (half lambda x velocity factor of feed line) is needed for this special case. It can also be any multiples of a half-wavelength too...

    Great way to use non-50 ohm feed line without the need for antenna system impedance matching device (antenna tuner).

    Having a good match at the radio end, now days, is important since most solid state transmitters begin to fold back even with the slightest VSWR. So this trick becomes more handy now than in the past.
    Last edited: Jul 27, 2021
  4. W9IQ

    W9IQ Ham Member QRZ Page

    Sorry to be a buzz kill but that isn't correct.

    First, an even or odd multiple of a 1/2 λ doesn't matter. Any multiple of a 1/2λ has the general property discussed. You were probably thinking of 1/4 λ impedance transformation sections.

    A 1/2 λ piece of coax when used in this fashion has loss due to the standing wave in its length. So it is far from "lossless". Due to this loss, the impedance transformation is also not exact.

    Here is an example: Use a 1/2 λ piece of RG-8x on 80 meters with a 300 ohm load impedance (300+j0). The physical length is 107.4 feet at 3.5 MHz. If this were attached to a matched load, this section of coax would have a loss of 0.74 dB. But due to the elevated SWR caused by the 300 ohm load, the loss is 1.9 dB. The input impedance is 203-j1.8 ohms. That is a far cry from 300+j0 ohms.

    It is also worth noting that while the load would suggest an SWR50 of 6, the input SWR50 is 4. Again this is due to the losses in the 1/2 λ piece of coax.

    For any given type of coax, the lower the frequency, the greater the loss in a 1/2 λ section of coax and the mismatch further increases the loss.

    Let's repeat the above scenario at 28 MHz: The 1/2 λ section is now 13.6 feet long. Its matched loss is 0.24 dB. With the 300 ohm load, the loss is 0.7 dB. The input impedance is 258-j0.3 and the input SWR50 is 5.1. So the input impedance is closer to that of the load but still quite a ways off from exactly repeating the load impedance.

    - Glenn W9IQ
    Last edited: Jul 27, 2021
  5. K7JOE

    K7JOE Subscriber QRZ Page

    Check this out. Plug in RG6... plug in RG8X... almost identical.

    Loss of 0.523 vs 0.534 dB... guaranteed not to make a bit of difference at 3.5 mhz despite the 50 ohm coax having a 1:1 assumed match and the RG6 having a 1.5:1 assumed match.
    It's mice nuts...
  6. K7JOE

    K7JOE Subscriber QRZ Page

    Ya... I fixed the mistype on the odd-multiples. It's every multiple of a half wave. And I guess I'm so used to using this method with high above earth dipoles that the antenna impedance is 72 ohms, close enought to 75 ohms that there is no mismatch between the load and the line.
    Last edited: Jul 27, 2021
  7. W9IQ

    W9IQ Ham Member QRZ Page

    First, that coax calculator is not the best. But more importantly, it is calculating those losses on the presumption that the center conductor is solid copper. None of the coax calculators that I have seen do a correction for copper coated steel. Even the excellent Transmission Line Details calculator from Dan, which uses K0, K1 and K2 parameters to calculate loss does not adjust for non-homogeneous conductors with insufficient skin depth.

    This is the basis for most of the discussion - the calculators could very well be wrong in the case of CCS wires at low frequencies. This is what Owen was highlighting in his post.

    - Glenn W9IQ
    W6JJZ likes this.
  8. K7JOE

    K7JOE Subscriber QRZ Page

    Well..the spec they use for 8215 Belden cable, is very clearly CCS...
    Even if it was off by a magnitude of 2X...that's still a mere tenth of a dB !!!

    Belden 8215 Specifications:
    Coax Construction COAXIAL
    Coax Type RG 6
    Conductor Material COPPER CLAD STEEL
    Conductor Type SOLID CONDUCTOR
    Dielectric Material PE-POLYETHYLENE (SOLID)
  9. K7JOE

    K7JOE Subscriber QRZ Page

    PS...i totally understand the issue with steel as a (non) conductor. It's the very reason those wolf river coils are terrible compared to a real copper or silver plated coil...
  10. W9IQ

    W9IQ Ham Member QRZ Page

    They state CCS but they do not rate the losses of that cable for the 160 meter to 40 meter range so you are not able to calculate the plating thickness. The center conductor accounts for the majority of the losses in a coax cable on HF so a change in its RF resistance can significantly alter the loss of the cable. Certainly doubling the RF resistance would result in more than a 1/10 dB additional loss per 100 feet.

    The point of this entire thread is that with a CCS conductor, you must take care at lower frequencies to determine its effect on attenuation. If there is sufficient copper thickness then the effect is minimal. If not, the effect could significantly affect applications.

    - Glenn W9IQ
    KK4NSF and W6JJZ like this.

Share This Page