Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

High frequency through cable?

  1. Nov 20, 2008 #1
    Hi, recent EE graduate, I'm working on a project for myself. I dealt with mostly controls, signals, and digital stuff, and not enough electromagnetics or RF so I am pretty clueless here.

    If I were to radiate the full electromagnetic spectrum from dc to visible light onto a copper cable, would it pick up all the frequencies or would it kill off the higher frequencies entirely? Mainly I am interested in a few Terahertz (infrared).

    I know I'm missing a lot of info here so sorry if sound ignorant but this wasn't my area of study and its hard to pick this stuff up on your own. I know there are such things as waveguides and something with the ground plane when you are at really high frequencies but all I have to work on is what I learned in my electromagnetic interference class.

    With skin effect, I calculated that at that frequency a 20 mil copper cable would have

    R per meter = 447 ohms
    L per meter = 2.4 pH

    They both get smaller as the thickness of the cable decreases

    I made a lumped circuit approximation (not sure if this is correct to do with a 1meter cable and 3 THz signal) where I put 1 of each in series on each side of a load resistor and do AC analysis in LTSpice, which gives me an LP filter with w_0 of roughly 10 THz. So it seems like a 1m cable like this cable would be able to pass a signal at the frequencies I am interested in, but this doesn't really seem feasible because its saying that a really thick copper cable could pass visible light. I know nothing about antennas so I don't know if it could pick up the signal either. Is it at all possible to pick up a THz signal with a copper wire with proper filtering and gain or am I wasting my time?
  2. jcsd
  3. Nov 21, 2008 #2
    There are modes where a single wire can carry radio frequency but not without a lot of radiation and therefore loss.

    RF is normally carried by coaxial cable or waveguide. Coax has a 'cut-off' because when the inner diameter of the outer conductor is greater than half a wavelenth it also behaves like a waveguide and if the frequency is even higher waveguide higher modes start to appear.

    Waveguide tend to be used over a limited range of frequencies. ( less than an ocatave). There is a minimum cut-off where the half wavelength is less than the largest dimension of the waveguide cross section (e.g longest side in a rectangular guide). The usually mode is TE10. Using a waveguide at a higher frequencies outside its normal band while result in multiple higher order modes appearing.

    You needs to read up on waveguide and modes.. TE10 .. TE11. TM10. Transverse Electric. Transverse Magnetic.

    I think in theory light would propagate through plain copper tube if it was slightly bigger than half a wavelength. It might be rather difficult to manufacture.
  4. Nov 21, 2008 #3


    User Avatar
    Science Advisor
    Gold Member

    For frequencies in the range 300-1000 GHz it is usually more efficient to use a dilelectric waveguide (basically a plastic pipe) than a copper waveguide.

    However, it is actually possible to use differential copper lines for frequencies in the THz regime (there was a talk about numerical simulations of this at a Comsol conference, I have the paper on a CD somewhere).
  5. Nov 21, 2008 #4
    Waveguides also act as receivers/antennas for higher frequencies?

    I am very interested in that paper.
  6. Nov 21, 2008 #5


    User Avatar

    Staff: Mentor

    The waveguide itself is just a conduit to guide the EM waves; it is not an antenna itself. The waveguide will end in a horn of some sort, which acts as the TX/RX "antenna". It will often be placed at the focus of a parabolic reflector, for example, to increase the gain of the antenna dramatically...
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook