My Cheap Coax is Junk, but Why?

Discussion in 'Antennas, Feedlines, Towers & Rotors' started by N0TZU, Apr 26, 2020.

ad: L-HROutlet
ad: l-rl
ad: Left-2
ad: L-Geochron
ad: HRDLLC-2
ad: MessiPaoloni-1
ad: Left-3
ad: L-MFJ
ad: abrind-2
  1. N0TZU

    N0TZU Platinum Subscriber Platinum Subscriber QRZ Page

    Here are some new photos of the cladding. I wasn't happy with the previous cross sections because there was smearing and pull-outs which obscured the actual layer thickness. Despite many attempts I was unable to get a better 90 degree cross section. After further experimentation I found that by taking a section at an angle the problem could be minimized.

    The Carol cable has an obvious well defined copper layer, which we know is highly likely to be 30 um from LU6PSG's information above. The other photo shows the Logico cable with essentially just a "flash" layer of copper cladding that catches some light in the photo. Even by eye in the stereo inspection scope (40x) it's difficult to discern any significant thickness. (Note, the copper at the top of the samples is smeared there from the cutting pliers)

    Logico still has not answered my email request for attenuation specifications as of this post. I doubt they ever will.

    IMG_3507.jpeg
    Carol C5775

    IMG_3508.jpeg
    Logico COX3520
     
    Last edited: Apr 30, 2020
    WD4IGX, KA2RRK, K3RLD and 2 others like this.
  2. W3PX

    W3PX Ham Member QRZ Page

    My guess is that the high loss is due to dissipation in low-quality dielectric.

    73 Frank W3PX
     
    AE8W likes this.
  3. KM1H

    KM1H Ham Member QRZ Page


    Its mighty hard to have poor foam dielectric at 10 MHz. OTOH if most of the RF is well into the steel core then....... Sort of like using MIG welding wire at RF.

    Carl
     
    W3PX likes this.
  4. WB2WIK

    WB2WIK Platinum Subscriber Platinum Subscriber QRZ Page

    I sure agree.

    The dielectric is polyethylene and air bubbles. More bubbles mean lower dielectric constant and higher velocity factor, fewer mean higher d.c. and lower v.f., but loss tangent either way is almost immeasurably low at frequencies lower than a few hundred MHz.

    I have a very difficult time measuring dielectric loss even at 1 GHz in conventional coaxial cables.

    In this particular case, it's quite clear the loss is due to very thin copper cladding, making the steel core part of the conductor at 10 MHz; it would not be at VHF-UHF where these "TV" cables are intended for use.
     
    K7JEM and W3PX like this.
  5. W3PX

    W3PX Ham Member QRZ Page

    Carl and Steve,

    Thanks for your posts, I think you have it right.

    73 Frank W3PX
     
  6. K3XR

    K3XR Ham Member QRZ Page

    "My Cheap Coax is Junk, but Why?"
    Looking at the thread title this comes to mind.
    [​IMG]
     
  7. AE8W

    AE8W Ham Member QRZ Page

    No one knows what the dielectric is. It is one thing to section and remove the Cu cladding, but the dielectric requires a material analysis and at a minimum is dk measurement ... not trivial.

    I am not trying to assert that the dielectric is the definitive problem. But conductivity of Cu cladded steel (what alloy?) is also not trivial.
    And is a definite maybe. To put it another way, if I as a knock off manufacturer is going to go cheap ... I will go all the way.
    The manufacturing of cable is in the flow system. One way is to measure capacitance as the cable is being extruded. The feed back control
    system (remember that there are some very intelligent Chinese educated in top US univesities) controls the foaming of whatever the dielectric is for a nominal 50 Ohms.
    Quality product uses PTFE which is very low loss as in nonexistant at HF. I see no reason to make such an assumption except that the dielectric is (probably) a white-ish gray and that
    is all we know about it.

    In short, I see what is presented but I do not believe my eyes. I need real numbers but isn't this way out of the context of the OP?
     
    Last edited: May 2, 2020
    N0TZU likes this.
  8. WB2WIK

    WB2WIK Platinum Subscriber Platinum Subscriber QRZ Page

    There isn't any kind of RG6 that uses Teflon as a dielectric, and actually polyethylene is exactly as low loss as PTFE; they both have a d.c. of 2.3 and create the same velocity factor in coax.

    PTFE is used not because it's lower loss, but for its high temperature rating which allows the cable to get hot if need be without damage, and sometimes for its high dielectric breakdown rating which allows it to be used under extreme mismatch conditions without flashing over.

    The "lowest loss" dielectric is dry gas (air, nitrogen...) but its d.c. is so low the ratio of inner to outer conductor diameter in coax must be dramatically increased to maintain the same impedance.
     
    N0TZU likes this.
  9. LU6PSG

    LU6PSG Ham Member QRZ Page

    Here you have a close up of the center conductor of a locally made CNA-660RW (CCS RG6 for DirecTV installs, standard quality 60% Al braid coverage+Al foil). This particular cladding is on the thin side with 28um of copper, but well within specs regarding attenuation and DC resistance showing 29 ohm/100ft.

    Then, as promised, a logarithmic chart showing the measured attenuation (interpolated and normalized) of one piece of the afforementioned coaxial cable measuring 1/2 WL at 3.56MHz loaded with R=50 ohm , and how it compares to some common 50 ohm lines. The selected length of 1/2WL at 3.56 MHz is also integer multiple of 1/2WL at 7.12, 14.24 and 28.48 MHz among others so the transmitter sees an impedance very close to the 50 ohm load or SWR very close to 1. This is only necessary when the characteristic impedance of the line (RG6 is 75 ohm) is different from the nominal impedance of the transmitter and the load (50 ohm), so the impedance transforrmation within the transmission line is practically nulled out (although the VSWR within the coaxial is 1.5)
    It's not hard to see that it performs a bit better than an RG8X (100% copper) on 40m, and matches the performance of LMR240 above 14MHz. On 80m it's just 0.2/0.3 dB above LMR-240 and RG213 respectively, both being 100%copper too.

    Unbeatable performance for being a CCS coaxial with Aluminum shielding and it's small price tag.

    CNA_RG6_3.jpg CNA_RG6_2.jpg RG6-COMP.jpg
     
    Last edited: May 2, 2020
    N0TZU likes this.
  10. LU6PSG

    LU6PSG Ham Member QRZ Page

    So, your thin cladding is definitely causing higher than normal resistive loses. I would not be surprised it the PE foam dielectric comes from some sort of recycled poor quality polyethylene or irregular foaming, also lowering overall efficiency.

    China produces many levels of quality. Don't forget that they manufacture the best smartphones on the planet, but also the worst knockoffs. Same goes for cables. I have a spool of chinese made PCT trishield RG6 showing attenuation and resistance levels better than specified.
     
    Last edited: May 2, 2020
    AK5B, AE8W and N0TZU like this.

Share This Page