frequency independent antennas....
in [DSP]
|
Prev: FFT Windowing and Transmit Pulse Shaping
Next: Mavenz Kenlighten - a social knowledge network for teachers, educators, job seekers, students
From: Tim Wescott on 11 Aug 2010 20:09 On 08/11/2010 04:35 PM, glen herrmannsfeldt wrote: > Tim Wescott<tim(a)seemywebsite.com> wrote: > (snip, I wrote) > >>> I hadn't thought of it that way before. Though the range of the >>> whole UHF TV band now (with higher channels removed) is less than >>> a factor of 2:1. The VHF TV channels cover a range from 54MHz >>> to 216MHz, a factor of 4:1. > >>> A low-Q UHF antenna near the middle of the band won't be so far >>> off on either end. > >> A log-periodic array will have much more consistent directional >> properties, though. > > How about a two-dimensional array of bow-ties? > > Usually only two across, but one could do more than that. > I believe I used to know some that were sold with optional > cross-coupler such that you could connect them in parallel. > (The spacing carefully chosen to impedance match the combination.) That may work, but only if the bowties were bent across the middle, i.e. if you looked at them from above it'd go \ / \ / \ / \ / X / \ / \ / \ / \ I'm not sure how you'd get the pattern to have good backwards rejection, though. -- Tim Wescott Wescott Design Services http://www.wescottdesign.com Do you need to implement control loops in software? "Applied Control Theory for Embedded Systems" was written for you. See details at http://www.wescottdesign.com/actfes/actfes.html
From: Jerry Avins on 11 Aug 2010 22:11 On 8/11/2010 5:41 PM, Clay wrote: > On Aug 11, 4:18 pm, "fisico32"<marcoscipioni1(a)n_o_s_p_a_m.gmail.com> > wrote: >> Hello Forum, >> >> a common wire antenna works the best when its size is equal to half the >> incident radiation wavelength. >> That means that the physical size is a constraint in terms of the operating >> bandwidth and performance. >> >> There are frequency independent antennas that have a finite size but can >> work on a much larger bandwidth...How can they do that? >> Those antennas are self-scaling....in what sense? >> The flat spiral antenna is an example. >> >> Does it means that by simple rotating the antenna we get the smaller and >> larger scale of the antenna itself? So what? >> >> A wire antenna that is 1 meter long is also made of smaller straight wire >> sections which are smaller versions of the antenna, but that is not a >> frequency independent antenna.... >> thanks! >> fisico32 > > Hohlfeld& Cohen of course worked out the exact two requitrements for > frequency independence. > > 1) You need self similarity > > 2) The feed point needs to be a point of symmetry of the antenna. > > Using just these two constraints and Maxwell's equations you can prove > frequency independence. Details in this paper: > > SELF-SIMILARITY AND THE GEOMETRIC > REQUIREMENTS FOR FREQUENCY > INDEPENDENCE IN ANTENNAE > > ROBERT G. HOHLFELD& NATHAN COHEN In the case of a log-periodic antennas (and other agglomerations of dipoles) a condition of self similarity often overlooked id the diameter of the radiator. The driving-point impedance is much more constant when the diameters are proportional to the lengths. Jerry -- Engineering is the art of making what you want from things you can get. �����������������������������������������������������������������������
From: glen herrmannsfeldt on 11 Aug 2010 22:52 Tim Wescott <tim(a)seemywebsite.com> wrote: (snip) >>> A log-periodic array will have much more consistent directional >>> properties, though. >> How about a two-dimensional array of bow-ties? >> Usually only two across, but one could do more than that. >> I believe I used to know some that were sold with optional >> cross-coupler such that you could connect them in parallel. >> (The spacing carefully chosen to impedance match the combination.) > That may work, but only if the bowties were bent across the middle, i.e. > if you looked at them from above it'd go > > \ / > \ / > \ / > \ / > X > / \ > / \ > / \ > / \ > I'm not sure how you'd get the pattern to have good > backwards rejection, though. Backwards rejection is usually from a screen reflector behind the bow-ties. The directional properties come from the phase difference between the different parts of the array. Pretty much the same as any diffraction problem, where narrow beam requires a wider source. I don't see how the X works, though. -- glen
From: fisico32 on 11 Aug 2010 23:13 >Tim Wescott <tim(a)seemywebsite.com> wrote: >(snip) > >>>> A log-periodic array will have much more consistent directional >>>> properties, though. > >>> How about a two-dimensional array of bow-ties? > >>> Usually only two across, but one could do more than that. >>> I believe I used to know some that were sold with optional >>> cross-coupler such that you could connect them in parallel. >>> (The spacing carefully chosen to impedance match the combination.) > >> That may work, but only if the bowties were bent across the middle, i.e. >> if you looked at them from above it'd go > >> >> \ / >> \ / >> \ / >> \ / >> X >> / \ >> / \ >> / \ >> / \ > >> I'm not sure how you'd get the pattern to have good >> backwards rejection, though. > >Backwards rejection is usually from a screen reflector behind >the bow-ties. > >The directional properties come from the phase difference >between the different parts of the array. Pretty much the >same as any diffraction problem, where narrow beam requires >a wider source. > >I don't see how the X works, though. > >-- glen > Could anyone please define self-similarity requirement? I know what a fractal is. It is self-similar: its small scales look the same as the large scales... But a wire antenna is a straight wire made of small straight wires...however it is not self-similar....Why? >
From: glen herrmannsfeldt on 12 Aug 2010 00:07
fisico32 <marcoscipioni1(a)n_o_s_p_a_m.gmail.com> wrote: (snip) > Could anyone please define self-similarity requirement? > I know what a fractal is. It is self-similar: its small scales look the > same as the large scales... > But a wire antenna is a straight wire made of small straight > wires...however it is not self-similar....Why? Look at the usual TV antenna, sort of shaped like an arrow. There are rods of different lengths, over the whole range of wavelengths. The active rods, and the function of those rods, changes with wavelength. If you start with a simple half-wave dipole, it radiates (or recieves, by reciprocity) equally in both directions. (Ignore up and down, for now.) A longer than half-wave rod acts, mostly, like a reflector, so add one of those behind the active dipole. A shorter rod, called a director, in front of the dipole increases the signal on the dipole. In an array of different length rods, at a given frequency some will act as reflectors, some as directors. The active dipoles are connected together with appropriate phase delay (transmission lines) in between. (Many have only one active dipole.) A single half-wave dipole, center fed, has an impedance of about 75 ohms. The popular 300 ohms comes from a folded dipole which, electrically, is about like two half-wave dipoles in series. There are also complicated combinations of dipoles and transmission lines (often thick metal wire) appropriately spaced (which sets the impedance) to properly match a 300 ohm lead wire. (Or ferrite transformer to 75 ohm coax.) A good source of detailed information about antenna design as the ARRL handbook, available in many public libraries. -- glen |