Directional coaxial cable (twisted stranded center), a myth?

[kj7e]

I heard someone on 40m comment on how he found his coax was directional, that is, he tested the loss both directions and found one direction had more loss than the other.  I don't know the details, freq, power or test methodology.  I am extremely skeptical on his claims, matter of fact I'm going to put a length of RG8x I have at home on the VNA just to see if there is ANY possible validity to this.  The claim is, the twist of the center conductor can affect the flow of RF. 

Thoughts?

[Grandchuck]

Damaged connectors perhaps.

[ConKbot]

When your twist length, or bumps in the conductors are measured in mm, and wavelength in meters, it would be a hard sale to get me to believe the electrically small features would have an effect that isnt something you could reduce to a lumped element model of RCL.

Getting differing S21 and S12 amplitude measurements indicates cable damage to me. If you have damage that is 1/4 wave from one end and 1/2 wave from the other, you will get different insertion losses. If you're measuring with just a spec-an/tracking gen, or fixed power source and power meter(scalar measurements) you won't notice the swr being weird.  Sweeping a wider frequency range, or using a VNA may help identify this.

[kj7e]

Damaged connectors perhaps.

Thats what I suspect as well, but the idea of twist to the center conductor did get me to think about it.

[joeqsmith]

Maybe he didn't tighten the connectors the same when flipping the cable.   Maybe they were making a fuss out of a difference of 0.001dB which was really just the VNA warming up.   

[drvtech]

Damaged connectors perhaps.

Thats what I suspect as well, but the idea of twist to the center conductor did get me to think about it.

If you think about it a little more you'll see that a twist doesn't make something directional. That's why you can screw a nut onto a piece of threaded rod from either end!

[G0HZU]

If you measure s21 and s12 with a perfect 2 port VNA and you assume the coax is a passive reciprocal device, then the forward s21 should be the same as the reverse s12 measurement. I think it's a reasonable assumption that coax is reciprocal. Some passive devices contain ferrite and this can be non-reciprocal.

If it helps any further here's a recent 2 port VNA plot of some RG213 coax I borrowed from work. I've plotted s21 and s12 for you. The markers show the same attenuation in both directions. This cable doesn't have a solid core, the centre is made up of twisted wires. I've also added a plot of about 22m of RG316 cable. The centre of this cable is also twisted strands, probably copper plated steel. You can see the s21 and s12 measurements are also the same.

[VK5RC]

I would punt that it is related to those 'lovely' PL259s, 😂

[themadhippy]

Maybe they picked up the wrong reel of coax and are using audiofool grade,not rf grade cable

[radiolistener]

I would punt that it is related to those 'lovely' PL259s, 😂

I'm agree, I had one PL259 which had VSWR about 1.4 at 28 MHz, just for connector... 😂

It's better to replace it with SMA or N connector.

[ejeffrey]

Bad VNA calibration alongside a damaged cable or connector. 

[TERRA Operative]

Damaged connectors perhaps.

Thats what I suspect as well, but the idea of twist to the center conductor did get me to think about it.

If you think about it a little more you'll see that a twist doesn't make something directional. That's why you can screw a nut onto a piece of threaded rod from either end!

Yep, that was my first thought too. It don't matter which way the threaded rod is flipped around, the nut still tightens the same way!

[Bud]

I don't know the details, freq, power or test methodology.

We should start with that. 

[fourfathom]

Damaged connectors perhaps.

Thats what I suspect as well, but the idea of twist to the center conductor did get me to think about it.

If you think about it a little more you'll see that a twist doesn't make something directional. That's why you can screw a nut onto a piece of threaded rod from either end!

Yep, that was my first thought too. It don't matter which way the threaded rod is flipped around, the nut still tightens the same way!

As for the OP's question, no, the twist on the stranded coax center conductor makes absolutely no difference here.  Even if the twisted strands were insulated from each other (and they're not) it would have no directional effect.

But twist direction in RF structures definitely has an effect, if the twist dimensions are a significant fraction of a wavelength.  I'm thinking of a circularly-polarized helical antenna. where left-hand and right-hand polarizations can be had.

[kj7e]

Just for clarification, the guy who made the claim was stating the twist of the inner conductor in a coaxial cable would have an impact of the flow of RF energy due to the right hand rotation of the EM field.  Where if the rotation aligned with the EM field rotation with the direction of flow, this would alter the path loss.   I agree, smells Audiofoolery, but it did get me to think about it some.

[PA0PBZ]

Just for clarification, the guy who made the claim was stating the twist of the inner conductor in a coaxial cable would have an impact of the flow of RF energy due to the right hand rotation of the EM field.  Where if the rotation aligned with the EM field rotation with the direction of flow, this would alter the path loss.   I agree, smells Audiofoolery, but it did get me to think about it some.

I guess that's reversed when used in Australia?

[kj7e]

Need to market special cable for northern and southern hemispheres.   I guess it was a silly thought, if there were any validity to this, it would either effect the RX or TX and I know that is not the case, at least not to any degree that can be practically measured.

[ejeffrey]

Just for clarification, the guy who made the claim was stating the twist of the inner conductor in a coaxial cable would have an impact of the flow of RF energy due to the right hand rotation of the EM field.  Where if the rotation aligned with the EM field rotation with the direction of flow, this would alter the path loss.   I agree, smells Audiofoolery, but it did get me to think about it some.

The EM field in a coaxial cable doesn't rotate in the right handed direction and that's also not what "right hand rule" means in electromagnetism.

The right hand rule or right handedness of EM fields in ordinary matter refers to the direction of propogation (Poynting vector) being the cross product of E and B.  If E&M were left handed (as in some metamaterials for a narrow frequency band at least) the propogation would be in the opposite direction, -E x B. It doesn't have anything to do with fields rotating.

It is possible to have rotating fields.  This is what you get with circular polarization in free space.  Here both E and B rotate as the wave propagates through space.  But there is no preferred direction, left and right circular polarization are equally possible and it has nothing to do with right handedness of E & M.

In any case, you don't get circular polarization in coax, or really any polarization.  Coaxial cable is operated below the cutoff for the waveguide modes of the outer ground, only a single mode exists.

[Hamelec]

Just for clarification, the guy who made the claim was stating the twist of the inner conductor in a coaxial cable would have an impact of the flow of RF energy due to the right hand rotation of the EM field.  Where if the rotation aligned with the EM field rotation with the direction of flow, this would alter the path loss.   I agree, smells Audiofoolery, but it did get me to think about it some.

but this only happens only 3 days +- 5 hours after the full moon

[SeanB]

My bet is the cable has water ingress. When at the outer end the loss there is going to appear smaller, and when at the inside the loss is going to make the TX pull power down, because of the mismatch, and thus the lower power at the other side.