RF non-resonant feedback antenna

[Drederick]

resonant antenna

Hi all,

I am trying to design a non-resonant RF antenna which is basically a radiating 50 Ohm microstrip transmission line (on a Teflon substrate) that operates at approx. 1 GHz. When I feed the microstrip with a coax and terminate the other end with a 50 Ohms resistor, I get the antenna that I want (no reflections, no standing wave pattern, 50 Ohms input impedance), except for the fact that I am wasting a lot of power in the resistor.

Now I would like loop the microstrip back to its beginning and feed the signal from the end of the antenna again to the its input via a power combiner. The loop is exactly one wavelength long so that the two signals that will be combined should be in phase.
Do you think this is this possible with a Wilkinson combiner? I tried, but unfortunately I get a very poor isolation between the two input ports and therefore a standing-wave pattern on the antenna.

I am thankful for any help on this...

Drederick

 

[FvM]

feedbac antenna

The basic concept sounds like a travelling wave antenna. They are usually designed in a way, that all (respectively most) of the input power is radiated along the antenna, so no power can be wasted in a termination resistor.

In my opinion, the problem with your microstrip "antenna" is, that it's only poorly radiating, so you may want to use a more effectively radiating waveguide type.

 

[Drederick]

feedback antenna

You're right, it is just a traveling wave antenna or a terminated microstrip transmission line which is poorly radiating.

The possible length of the antenna is limited to not more than one wavelength, say 30cm. I don't know if there is a kind of transmission line antenna which radiates most of its energy on this short path. Is there a way to improve radiation from the line or does somebody have a suggestion which kind of line to use?

 

[FvM]

resonant vs non-resonant antenna

I guess, most structures (e.g. meander) are radiating better than a straight micro strip. But I'm no antenna expert, I would consult respective literature.

Just another comment on your original "feedback" concept. To make it working, the signal level at the strip line must be increased to compensate for the low radiation factor. Not impossible generally, but hard to achieve with a "non-resonant" design, I think.

 

[Drederick]

non resonant antenna

Thanks for your comments.

One thing is not really clear for me: Are there general rules how to improve the radiation from a microstrip transmission line by changing the eps of the substrate, its height h and the stripwidth?

Unfortunately I have two papers here, one stating that high eps (Al2O3, GaAs), low h microstrip transmission lines radiate most, the other (which is about RF microstrip resonators) recommends eps=1 and a big distance between strip and ground for the strongest B in the nearfield.

 

[biff44]

resonance antenna radiate

I would look at antennas that the UWB guys are using (vivaldi, etc). I did a low frequency (<1 GHz) ground penetrating pulse radar antenna with just an open-top microstrip structure, but with a very wide strip width (similar to what you would find in a GTEM cell), and it seemed to work well. The far end was terminated in a number of chip resistors equalling 50 ohms when put in parallel. It seemed to have a pretty broadband radiation, and little pulse reflection from the feed port, although it was not designed for radiation efficiency. It is the square black thing called "GPR human target simulator"

Rich