8
\$\begingroup\$

Reading this answer I was surprised to hear that there are "excellent" kinds of coaxial cable with a propagation speed of 0.9c; 90% the speed of light.

The bargain basement number is about 2/3c, and coaxial cable with a faster propagation velocity would have to have a lower dielectric constant. Assuming \$v/c\$ scales as \$1/\sqrt{\mu_r \epsilon_r}\$ as it would in free space, that would mean the cable would have a relative dielectric constant of 1.2 for example.

Does this exist as a standard product? If so, are there applications where having "excellent cable" with such a high propagation velocity would be important? Or would it be a side-effect of other desirable properties of the dielectric?

\$\endgroup\$
8
  • 6
    \$\begingroup\$ Signals in a hurry? I have read of using line-of-sight tower to tower laser links from NY to Chicago to shave a few ms off the fiber transit time to give a competitive edge to stock exchange trading companies. (I've also read of a company adding a 32 km coil of fibre in their switchroom to add a few ns delay to users to limit the rate of transactions.) These thinks matter to some folks. \$\endgroup\$
    – Transistor
    Commented Sep 24, 2017 at 16:00
  • 1
    \$\begingroup\$ The "best" I have seen is around 0.8x and I have never seen a case where it is necessary to have a special factor, just some where you need to know it. \$\endgroup\$
    – PlasmaHH
    Commented Sep 24, 2017 at 16:01
  • 1
    \$\begingroup\$ @Transistor more like a ms than a ns. So this cable was invented for high speed stock trading? \$\endgroup\$
    – uhoh
    Commented Sep 24, 2017 at 16:02
  • 5
    \$\begingroup\$ I haven't a clue. I don't think coax is used for long distances anymore. Fiber took over a long time ago. \$\endgroup\$
    – Transistor
    Commented Sep 24, 2017 at 16:04
  • 1
    \$\begingroup\$ The high velocity specs translates into lower loss, more air than plastic and more copper $$ more rigid and more bandwidth were all standard rigid coaxial trunks for 300+MHz CATV in the 1980’s before fibre was more cost-effective. Reducing C from lower mu was the goal while maintaining impedance error tolerances from O/I diameter ratio damage \$\endgroup\$
    – Hoagie
    Commented Feb 27, 2023 at 15:05

5 Answers 5

14
\$\begingroup\$

Air lines still exist, with velocity factor very close to 1.0. These are AFAIK mainly used in old-fashioned VSWR measurements. The advantages are that the dielectric constant of air is fairly stable and well-known, and that you can insert a probe (a tiny antenna) into the middle of the transmission line without damaging the dielectric.

ePTFE (aka "Teflon foam") dielectric typically gives a velocity factor of about 0.85. These cables are, in my experience, used because they maintain low loss to fairly high frequencies and their phase delay is quite stable under variations of temperature and flexure, not specifically because of the high phase velocity. I've used them in test and measurement applications, and I imagine they're also used in things like radar and avionics.

I found a reference saying that "foam polystyrene" dielectric gives a velocity factor of 0.91, but I have no experience with such cables, and I don't know what applications they're favored in. In fact I couldn't (with 2 minutes googling) find any vendor actually selling them.

\$\endgroup\$
3
  • \$\begingroup\$ This is a great explanation of the situation, thank you! \$\endgroup\$
    – uhoh
    Commented Sep 24, 2017 at 20:45
  • 1
    \$\begingroup\$ LMR 1700 timesmicrowave.com/documents/resources/LMR-1700.pdf has a velocity factor of 0.89. The insulation material is gas-injected polyethylene. \$\endgroup\$ Commented Sep 25, 2017 at 15:05
  • \$\begingroup\$ @davidmneedham, thanks. One of my references also said ePTFE can reach 0.9 (presumably when made with higher air content). But I find 0.85 is much more common. \$\endgroup\$
    – The Photon
    Commented Sep 25, 2017 at 16:17
7
\$\begingroup\$

The high propagation velocity would be a "side effect" of more desirable property - signal loss/dissipation along a cable. Losses are dependent on properties of isolation material. If you already have the best low-loss material, the next thing is to make it porous, so it will mostly contain dry air. The higher propagation velocity is a side effect of this.

\$\endgroup\$
2
  • \$\begingroup\$ Makes sense; the only thing better than having the best is having less of it :-) \$\endgroup\$
    – uhoh
    Commented Sep 24, 2017 at 20:46
  • 2
    \$\begingroup\$ Just as an extra comment, if you end up working with higher frequencies such as 10 or 20 GHz, the loss per foot for regular coax starts to be very large. \$\endgroup\$
    – user57037
    Commented Sep 24, 2017 at 21:08
4
\$\begingroup\$

Some (older?) coaxes suspend the center conductor with glass beads every so often. Thus most of the dielectric is air. Propagation velocity is almost 1.0 \$c\$.

\$\endgroup\$
5
  • \$\begingroup\$ That's really interesting! I'd love to read about that or see a picture if possible. As long as the bead-lumping distance was way smaller than the wavelength, I suppose it's no different than the air pockets in foam or the dielectric spacing in ladder lines. \$\endgroup\$
    – uhoh
    Commented Sep 25, 2017 at 4:12
  • 1
    \$\begingroup\$ Show me a coaxial cable with a velocity factor of 1.0. Even "air core" coax has a sub 1.0 vf. \$\endgroup\$
    – W5VO
    Commented Sep 25, 2017 at 6:33
  • 2
    \$\begingroup\$ @W5VO even a beam of light six feet wide traveling in a vacuum will travel slightly slower than the speed of light. See Wouldn't any structured beam of light be expected to travel slower than a plane wave?. Somehow I missed that last sentence-fragment in the answer, thanks for pointing it out. Indeed Nothing really travels at exactly 1.0 c other than "infinite plane waves". \$\endgroup\$
    – uhoh
    Commented Sep 25, 2017 at 10:26
  • 1
    \$\begingroup\$ @uhoh Some applications use rigid line coax, which is essentially a copper tube suspended in a copper tube (example image). EEVblog #569 - Tour of an Analog TV Transmission Facility is a fascinating video about high-power analog RF devices, including rigid coax. \$\endgroup\$
    – marcelm
    Commented Sep 25, 2017 at 19:00
  • \$\begingroup\$ @marcelm thanks for the links! I will watch the video tomorrow, I got distracted by all of the cool coax your linked image led me to! \$\endgroup\$
    – uhoh
    Commented Sep 25, 2017 at 19:56
3
\$\begingroup\$

Thanks to the link to the image from the eriinc.com site in @marcelm's comment (from which another page links to commscope.com) I've been enjoying some amazing pics of coax instead of doing work.

Here's the rigid line coax and dielectric space from the comment and a similar one:

rigid line coax 1 rigid line coax 2

but they also have a link for coaxial cable at Commscope where there are some very nice low volume dielectric spacers; HJ11-50, H5-50 and HJ9HP-50 with propagation speeds of 92%, 92% and 96% the speed of light!

HJ11-50

H5-50

HJ9HP-50

\$\endgroup\$
1
  • 1
    \$\begingroup\$ Those spiral spacers are where the "heliax" name comes from. Although nowadays it also encompasses foamed PE dielectric which is more conventional, but still manages 88-91% Vf by being mostly air. \$\endgroup\$
    – hobbs
    Commented Feb 27, 2023 at 15:29
1
\$\begingroup\$

The high velocity specs translates into lower loss, more air-core than plastic and more copper $$ more rigid and more bandwidth were all standard rigid coaxial trunks for 300+MHz CATV in the 1980’s before fibre was more cost-effective.

Reducing C from lower mu was the goal while maintaining impedance error tolerances from O/I diameter ratio damage.

This also raises impedance from higher L/C suitable for some apps unless the centre diameter is increased to compensate with lower L thus more $$ to get back to 75 ohms for CATV or 50 ohms for almost everything else.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.