3D Printing Electrically Small Spherical Antennas, l1 k6 k7 K- Y
/ z4 x, ~7 N$ M7 f+ P% r. f3 C
Abstract—3D printing is applied for rapid prototyping of an electrically small spherical wire antenna. The model is first printed in plastic and subsequently covered with several layers of conductive paint. Measured results are in good agreement with simulations.
: [8 L2 \: {8 q) }' K4 o; R/ c% p- w J6 p
I. INTRODUCTION$ `" G# n" H3 x1 A" [( u. c
Spherical wire antennas composed of thin conducting wires wounded on the surface of a spherical core are known to exhibit the lowest radiation quality factor Q for a given electrical size ka, where k is the free-space propagation constant and a is the radius of the minimum sphere enclosing; X. \+ s ?% y: l3 \
the antenna. At the limit, with no stored energy inside the core, the spherical antenna can closely approach the Chu lower bound [1]–[3]
+ K5 V+ e6 q6 l* k( M+ A: C5 z& fQChu =1/(ka)3 +1/ka.
( n7 d: d0 v( C' a
M7 }8 m% u8 ?# bThis was proved possible, e.g., for spherical antennas with metal cores coated with a high-permeability magnetic material [4]–[6], for both electric and magnetic dipole modes excited.
/ s1 v1 ]( }0 b: l6 P4 h5 f$ Y+ s# J( h/ \5 Y5 |6 C
In the case of air core, the lower bounds for the Q are Q ! 1.5QChu and Q ! 3.0QChu, as ka ! 0, for electric and magnetic dipole mode antennas, respectively. And again, spherical wire antennas can approach these bounds very closely, as shown by numerous numerical simulations and experimental results, e.g. [7]–[9].
' b, |5 I8 q7 |$ b/ b, v" ?* r8 h# w/ r' u: s* A0 H% Z6 `
While the properties of spherical wire antennas are appealing, their physical implementation is not a trivial task. Straightforward bending of wires requires substantial skills in handwork [7], [9]; and a satisfactory result is not always guaranteed. More sophisticated methods include, e.g., direct
" ~$ |( l/ w( s0 N. [$ d5 ~& k" ^writing with conductive ink [10] or direct transfer patterning [11] on a curved surface; in both cases, the technology is
2 Q4 a0 G2 ^5 @/ ncostly and is not generally available.( M6 G2 Z _2 A& N
On the other hand, there exists a relatively mature technology, which allows virtually any shape to be physically reproduced in material. The technology is quick and cheap; it has been already commercialized to the level, where an order can be placed on-line and the product delivered next day. It is3D printing.
/ o; {6 ^1 C. B9 o) G( JIn this contribution, it is shown how the technology can be used for rapid prototyping of electrically small spherical wire antennas. Although 3D printing is available nowadays both in dielectric and metal, the former is more common, and thus a model printed in plastic with subsequent conductive coating is presented here. U+ t6 Z- p6 p6 A
9 L0 }" Q6 S/ g& s% ^' u; f
II. PRINTING AND PAINTING
& @ H+ Z0 d; gFor the first test, a well-known folded spherical helix antenna [7] is selected. The antenna is fed by a coaxial cable through the ground plane; the wire radius is chosen to be the same as the radius of the central pin in the standard SMA connector (0.64 mm). At 750 MHz, the four-arm configuration
3 d. }1 S, I R8 o. s9 c- r, W, mis tuned to the resonance by adjusting the number turns in each helical arm, while matching to 50 ohms is achieved by changing the radius of the antenna. The resulting radius is 25 mm, which corresponds to a minimum sphere a = 25.64 mm and ka = 0.403.5 V8 S$ r2 K0 f4 k8 F5 M: M
From a great variety of materials available for 3D printing, Nylon PA 2200 was deemed to be the most suitable. The antenna was printed on a small support, which was subsequently covered by several layers of copper paint, together with the rest of the antenna. Finally, the structure was mounted on a 770 mm circular aluminum ground plane, and a plastic support was introduced along the axis of the antenna to fix its height. The result is shown in Fig. 1.
* q! C3 u3 F* \% |* F; G5 @
. H! D2 n1 [$ GSince electrically small antennas are very sensitive to the losses in their structures, stringent requirements are imposed on the conductivity of the antenna surface. To ensure good electrical conductivity, several layers of copper paint were applied; the first layer was applied with a brash, the other two were sprayed. The reflection coefficients measured after each painting are presented in Fig. 2. It is observed, how the conductivity increasing with each consecutive layer of paint improves the matching of the antenna. The obtained resonance frequency is 736.3 MHz, which is less than 2% below the expected 750 MHz. The difference is attributed to the gravityinduced deformations of the helical arms.
, L8 m+ i5 c3 g+ R2 {" xThe radiation efficiency measured in the anechoic chamber of the DTU-ESA Spherical Near-Field Antenna Test Facility is 80%, whereas the efficiency predicted by simulations (CST) for the antenna made of solid copper wires on an infinite copper ground plane is 97%. This indicates that the efficiency
! q/ \" }4 d: q; B+ `- o4 hof the fabricated antenna can be further improved by extra layers of copper paint. The radiation patten is omnidirectional with a deep null in the broadside direction. The measured cut in the elevation plane is shown in Fig. 3.
9 m* u; o# O% m+ C uIII. CONCLUSIONS
. ~% h L1 P6 H; V H% [It is demonstrated that modern 3D printing technologies can be successfully applied for rapid prototyping of electrically small spherical wire antennas. As an example, a self-resonant folded spherical helix antenna of radius 25 mm was designed, printed in plastic, painted with conductive paint, and finally measured. The antenna exhibits characteristics, which are in good agreement with the predictions. The improved designs are being fabricated and tested, and further results will be presented at the conference.
3 G9 M6 |7 h) ~! fACKNOWLEDGEMENT) {; j( k4 ]" d5 {* J
Mr. Martin Nielsen is acknowledged for painting and assembling the antenna, Dr. Sergey Pivnenko for measurements.
/ z5 R) f3 Q' S" `) ?7 u |