0

How to I design an antenna which produces a strong field within the antenna, but doesn't radiate out or reflect back to the source? (as much as possible)

I'd like ideally to have a planar structure (manufacturable with PCB traces), matched to 50 ohm, and reasonably broadband at microwave frequencies (2.4-5.8GHz).

I know about the standard PCB trace antenna types eg PIFA, meander, etc; and I can understand what makes them radiate; but it is hard to think of how to do the opposite. It seems what is needed is essentially a capacitor (where the output field is between the plates) plus some discrete components or transmission line to match it to 50ohm and prevent reflections. But how to broaden the bandwidth?

UPDATE A schematic using lumped elements would look something like this:

schematic

simulate this circuit – Schematic created using CircuitLab

Everything on the right is the antenna: a small inductance and tiny capacitance which is resonant at 2GHz as shown, with a resistive component a lot less than 50 ohms. The transformer is "something" which matches the source 50ohms to the (much smaller) resistance of the antenna.

Alex I
  • 3,341
  • 2
  • 36
  • 59
  • 4
    What is the purpose of this antenna if it doesn't radiate? Could you just use a terminated transmission line? – The Photon Sep 16 '20 at 22:47
  • 1
    Or an ordinary antenna surrounded by a Faraday cage? – The Photon Sep 16 '20 at 22:50
  • @ThePhoton An ordinary antenna surrounded by a Faraday cage would produce a lot of reflections back into the feedline. A terminated transmission line - sure, with a resistive termination that would work, but most of the energy would just be dumped into the termination. I'd like something which is mostly reactive - the only resistances should be those inherent in the structure (eg the resistance of the traces of the transmission line) so that the local field is fairly strong relative to the input power. – Alex I Sep 17 '20 at 00:24
  • 3
    The energy had to go somewhere. Either it will heat up the structure or it will radiate or it will reflect back into the feed line. You could take your microwave oven and line it with absorbent material if you want to minimize reflections. – The Photon Sep 17 '20 at 00:27
  • Any of your input energy that doesn't heat the structure must radiate or reflect, and you've said you don't want those things either. So I don't see any way to give you an answer that will satisfy you. – The Photon Sep 17 '20 at 00:55
  • @ThePhoton I'm sorry for not explaining better. What I'd like is straightforward: moderate input power feeds into a structure acting as a resonator => very high fields build up in the structure, so ideally there shouldn't be anything acting as load other than the resistance of the conductive elements of the structure itself (no load resistors, absorbent materials etc). Eventally the fields/standing waves would be high enough that the power gets dissipated in the structure, but that would happen at high circulating currents. – Alex I Sep 18 '20 at 05:56
  • 1
    By reciprocity, the antenna will be just as effective at coupling fields from the structure into the feed line as it is at coupling fields from the feed line into the structure. So there will always be strong reflections from this system. – The Photon Sep 18 '20 at 14:52
  • 1
    @AlexI *why* do you want this? What *useful purpose* do you hope it will accomplish? It's been explained that things don't work the way you'd been thinking they do, to get any further you're really going to need to explain what the motivation for this project is, so that people can help you come up with an alternative that actually *could* work. – Chris Stratton Sep 22 '20 at 01:57
  • @ChrisStratton Thanks - the goal is to produce as strong an E field in a limited volume of space as possible, from a relatively small power input. The effects of the field on "stuff" inside the volume is the purpose. A microwave oven is not a bad analogy, except in this case for it's for small objects which wouldn't absorb much energy even in strong fields. Roughly, I need <1W to produce fields >>1000V/m in a volume of 100cm^3; the "stuff" in the volume (slightly lossy dielectric) would absorb <0.1W in that case. This seems physically possible, at least at low frequencies. – Alex I Sep 22 '20 at 23:43

0 Answers0