1
$\begingroup$

I want to understand what factors determine acceptable turning radius of RF cable. are the losses high for low frequency or at high frequency if cable is turned or twisted ? the cable i am referring is an coaxial cable. i want to know how cable core conducting metal diameter, sheath metal, insulator characteristics etc in RF cable effect cable turning radius.

$\endgroup$
  • $\begingroup$ Turning radius? Are you asking about a single bend like a 90 degree corner or are you thinking of a coil (even if a single turn). Also, there are a lot of different coaxial cables with differing materials, outer conductor style (braid, metal sheath, etc). If you are planning on some type of project, you might want to choose a specific coax cable. $\endgroup$ – K7PEH Nov 22 '16 at 1:02
  • $\begingroup$ Conduit & thin wall, if I remember right, is 7 times its diameter equals minimum radius. $\endgroup$ – Optionparty Nov 22 '16 at 20:41
4
$\begingroup$

The problem with bending a coaxial cable too sharply is that the cross-section of the cable is crushed out of round, which would change the characteristic impedance of that section. Just as with most other defects in transmission lines, the loss will increase with frequency.

The bending radius has no direct electrical significance; it just determines how much stress there is on the materials that make up the cable, and the effect on the cable will depend on exactly what those materials are. The particular materials will also affect whether the cable is permanently damaged or it will recover when unbent.

I doubt you will find any general rule for calculating the bend radius; it's a fairly complex interaction between the material properties of the at least four different elements of the cable.

$\endgroup$
  • $\begingroup$ I was told that the usual problem with coax being bent too tightly is that the center conductor tends to migrate when the dialectric gets hot (either from handling power or from being outside on a hot day), which would also result in a change in the characteristic impedance. Either way the result is the same ;) $\endgroup$ – rclocher3 Nov 22 '16 at 2:29
  • $\begingroup$ @rclocher3 heating up of the dielectrict, to the point where that loses its mechanical rigidity?! You're not using the right cable for your powers. Dielectric should, by design, have the least losses possible, and even if you push through let's say 1kW of RF power (at "few GHz", I'd like to see anyone outside an air traffic control or military radar operation do that), the losses would, in a proper cable, be spread across the full length of cable. So even if your cable has 10dB of signal loss over 100m (that's, at 1kW in, 100W), that means just 10mW/cm heat. A normal cable shouldn't get hot!! $\endgroup$ – Marcus Müller Nov 22 '16 at 12:26
  • $\begingroup$ (in fact, that is a gross oversimplification, because obviously, power density decreases exponentially with length, but it still gives you the order of magnitude we're operating at) $\endgroup$ – Marcus Müller Nov 22 '16 at 12:27
  • $\begingroup$ So, if that heating up happens only at the bend, than that's a result of increased losses, and that is the result of deformed geometry. The central conductor migration is just a subsequent fault. $\endgroup$ – Marcus Müller Nov 22 '16 at 12:28
  • $\begingroup$ (errrr 90mW/cm, not 10. brainfart) $\endgroup$ – Marcus Müller Nov 22 '16 at 12:39

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

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