US 1643323 A
Description (OCR text may contain errors)
Sept. 27, 1927. 3, 3
J. S. STONE- DIRECTIVE ANTENNA ARRAY Filed Jan. 4. 1921 4 Sheets-Sheet 1 g P x Q 5 V 1' I. m Q m v INVENTOR.
I Jafin J3me Jim 2e L W TTORNEYS.
Se t. 1
p J. 5. STONE DIRECTIVE ANTENNA ARRAY Filed Jan. 4. 1921 4 Sheets-Sheet 2 flwumtoi f0ht Stone 510110 W Sept. 27, I Q I J. S. STONE DIRECTIVE ANTENNA ARRAY Filed Jan. 4., 1921 4 Sheets-Sheet 3 wvmtoz $2, Wwm W Se t 2 192 1,643,323
p 7 J. s. STONE DIRECTIVE ANTENNA ARRAY Filed Jan. 4. 1921' 4 Sheets-Sheet 4 IN VEN TOR.
.fa/Wa Jib/26550188 ATTORNEY DIBEGTIVE AENNA ARRAY.
Application filed dauuary t, 1921. serial Ito. 484,847.
The principal object of m I invention is to provide a new and improve directive radio transmitting or receiving station. Another object of my invention is to provide s. radio station with o. plurality of interconnected ontonnes, so errenged as to transmit or receive efiectively in e certain desired direction but not in other directions. Uther oh iects of my invention hove to do with such matters as so curing a convenient and compact distribution oi the sntenum at such 9. station. exciting them iii proper emplitudo end phesc relotion, end get-thug on angular distribution of intensity for the array. corresponding substantially to e poler diagram of 2. single loop. A directive entemie erruy constructed end operated sccording to my invention and used as s trsnsmitter will rediete power only upproximsteiy in the direction in which it is desired to transmit end thus it will give economy or power or in other words, the
" o transmitter,
evsilslole power of the station will he radieted efiectively instead of lorgely in non cfiective directions. My directive array, es
secures non-interference to stations in other directions than that direction to which it is desired to trensmit; and
for receiving, it prevents interference from stations lying in any other directions than that direction from which it is desired to receive. In short, my system greatly reduces the annoyance of cross talk. In the same way that it eliminates interference from other stations for receiving, it also eliminates interference from static to a large extent.
Directive systems have been devised which were directive in a plurality of directions, sometimes with the same intensity in each of these difierent directions, sometimes with different intensities in difierent directions. I am referring to cases in which polar disgrams of intensity show maximum radii for a plurality of'difi'erent angular positions.
My improved antenna array may be designed to give substantially only one maximum at the particular frequency for which the system is designed aswill be made apparent in the following disclosure. These and other objects and advantages'of my invention will become apparent on consideration of the disclosure of s few specific embodiments given in the following specification. The inventwo is defined in the appended claims, and I now proceed to describe the particular forms thereof which I have chosen to disclose by Way of illustrotionz' Referring to the drawings, Fi ure 1 is s diagram showing an elevation o o. pair of antennae which may be regarded as constituting an elementary component of certain types of my improved antenna. arrays. Fig. 2 s e plan diagram showing the angular distribution of intensity of radiation or reception; in other words, this is e polor diogrem for the intensity of radiation ii-om or reception hy the antenna. pair of Fig. i. Fi 3 is a diagram showing an elevation t snot 1e1- entenna pair which may enter in u diderent way in certsin types of my improved antenna.
orreys. Fig. i is the polar diagram correspending to Fig. 3. Fi 5 is a plan diagram indicating the combinetion of two pairs, such as shown inFigs. 1 or 3. Fig. 6 is it plain diagram showing this combination of two of the elementary pairs of Fig. l with polar diagrams for intensity of rodiotion or reception. Fig. 7 is o corresponding diagram based on component pairs of the type of Fig. 3. Fig. 8 is s. diagram showing the combination of two sets of antennas, each like those of Fig. 5. Fig. 9 is a diagram showmg the combination of two sets like those of Fig. 8. Fig. 10 is e diagram showing at rectangulor array built up along one dimension according to the development indicated in connection with Figs. 1, 2, 5, 6, 8 and 9 and along the other dimension as indicated in Figs, 8, t, 5., 7, 8 and 9. Fig. 11 is s. combination of elemental and resultant polar diugrams for Fig. 10. Fig. 12 is a. polar diagram for a pair of antennae separated of a wave length and energized by currents whose phase is of a. period apart. Fig. 13 is a. polar diagram for a pair of antennas separated by of a wave length and excited lFJy currents whose phase is period apart. ig. 14 is a diagram for a certain antenna array based on the systems of Fig. 12 and 13 as components. Fig. 15 is a diagram showing the amplitude and phase relation of the currents in the antennae of Fig. 14 when they serve for transmitting. Fig. 16 is a diagram for a linear array of nine antennae, located wave length apart and excited in a phase relation that will be disclosed in the exp anation that follows: Figs. 17 and 18 are perspective diagrams showing how the currents may be supplied to the antennae of one of my arrays in pro er phase and lntensity, and Fig. 19 is a iagram illustrating an array whose contour is not rectangular.
The two antennae of Fig. 1 are located a quarter wave length apart and supplied with exciting currents which are in quarter phase relation. This phase displacement is indicated by the insertion of artificial lines at A to secure the desired diflerence of phase. At considerable distances from these two antennee, where their respective fields are practically parallel, the resultant intensity w ll 'difier accordin to the direction. Only in the direction 0 the line joining the two antennae and from the leading antenna through the lagging antenna will the intensities from both of them add and give a maximum. In
other directions the intensity will be less,.
according to the polar diagram given in Fig. 2.
Thus, it follows that the antenna pan of Fig. 1 has the directive property indicated by the diagram of Fig. 2, and also the energy radiation from the pair is less than twice what it would be from a single antenna, as indicated (not necessarily quanti-' tatively) by the proportion of the cardioidlike curve ofFig. 2 to the dotted circle.
In Fig. 3, I have shown an antenna pair with the members spaced a half wave length apart and excited in the same phase. The angular distribution of intensity of radiation for such a pair is indicated by the polar'diagram of Fig. 4.
Comparing Figs. 2 and 4, it will be seen that the antenna pair o f'Fig. 1 gives a maximum radiation in the direction of the line joining the two antennae and from the leading one toward the lagging one; on the other hand, the antenna pair of Fig. 3 gives a maximum radiation equally in either direction along the normal to the line joining the two antennae.
While I have discussed the diagrams of Figs. 1 and 3 as if they represented transmitters, it is true that they act as receivers according to direction. as indicated in the diagrams of Figs. 2 and 4. In other words, the directive properties are the same whether the antenna pairs of Figs. 1 and 3 are employed for transmitting or receiving.
At a great distance from the antenna pair of Fig. 1 or Fig. 3, its effect is equivalent to a single source half way between the two sources, but having a olar diagram like that given in Fig. 2 or ig. 4, respectively.
venience I call this equivalent source a consequent so rce of the first order.
If two 0 nsequent sources of the first or-'- der are placed at P and Q, in Fig. 5, a quarter wave length apart and with their axes in the same direction, this will give an arrangement of four antennae, two of which will be positioned close together, as shown by the adjacent crosses in Fig. 5. Instead of adding the intensities at the intermediate position by placing two antennae there, it will be more convenient to add these intensities on a single antenna. This gives the arrangement shown schematically in Fig. 6, where the numerals in parentheses indicate relative intensities on the respective antenna, thus in order (1), (2) and (1). The polar diagram for the combination of the two consequent sources of the first order is shown in a full line and for the purpose of comparison a polar diagram of t e same-maximum radius, such as given by a single conse uent source of the first order, is shown in otted lines. It will be seen that by combining the two consequent sources of the first order, as in Figs. 5 and 6, I have secured greater directivity. This is shown by the fact that the polar diagram has shorter radii for all -directions except the direction of maximum intensity. The combination of the two consequent sources of the first order shown in Figs. 5 and 6 gives what I call a. consequent source of the second order.
In like manner, the combination of the two consequent sources of the-first order as shown in Figs. 5 and 7 gives a consequent source of the second order. Thus, it will be seen that a source of the second orderis more narrowly directive than the first order sources of which it is built, though the general character of the transmission is. un-
changed, that is, it continues to be uni-directional or duo-directional, as the case may be.
Two second-order consequent sources, with their centres a quarter wave length apart and their directions of maximum intensity coinciding, are shown assembled in Fig. 8.
The component intensities on the four an-' tennae are indicated and the addition is shown by which the resultant intensities are obtained. In order, the latter are (1), (3), (3) and (1). The result of the assembly indicated in Fig. 8 is a consequent source of plitudes in order are respectively (1), (4),
(6), (4) and (1). Each time that I combine two consequent sources of the nth order to form a consequent source of the (n+1)th order, I secure a further narrowin of the polar diagram, just as I explained in detail for the transition of Figs. 6 and 7. The illustration has been carried far enough to show that the law by which the intensities are created in the successive antennae of the row, is the law of the coeflicients in the biformula.
nomial expansion. In general, the formula for the (r+1)th coeflicient is nl rim-r)! where n is the order of the expansion. In terms of the present discussion, the intensity of the (r+1)th antenna in a consequent source of the nth order is given by this Thus, for Fig. 9, if n is made equal to 4 and 1' is successively 0, l, 2, 3 and 4, the results will give the numbers (1), (4:),
Returning now to Figs. 3 and 4 and comparing with Figs. 1 and 2, it will be seen that each loop of the polar diagram in Fig. 4 is decidedly narrower than the diagram of Fig. 2. This is an advantage in directivity but with it comes the disadvantage that radiatlon in this case is equal in opposite directions. I will now show how the two component types may be combined to join their advantages and eliminate their disadvantages.
Fig. 10 may be looked upon as an assembl of consequent sources of the nth order (see source corresponding to a column of the figure), combined according to the rules for building up a consequent source of the mth order, or vice versa. Hence, I call it a consequent source of the math order, or equally well, it might be called a consequent source of the math order. in this particular example, l have made at equal 4. and an equal 5. The numbers along the marginal column, s (e (a. st t give mensof a wave length and which are energized for transmitting by currents diflerin in phase by of a com lete period. It will 0 seen that even in t c direction of maximum resultant intensit the intensity is less than the scalar sum 0 the maximum intensitles on the two antennae. Otherwise, the
polar diagram is of the general character shown in Fig. 2. Com ared with Fig. 2, the geographical extent 0 the array of two antennee 15 only half as great. i
Fig. 13 gives the polar diagram for two antennaespaced of a Wave length apart and excited by currents which differ in phase by half a complete period. .It will be seen that this polar diagram 'ves maximum intensities an two opposite irections lying along the line determined by the two antennse.
The antenna pairs of Figs. 12 and 13 each constitute a consequent source. For convenience I represent Fig. 12 by the formula [asl where the expression preceding the one gives the spacing between the two elemental sources in terms of'the Wave length A, and the expression following the comma gives the phase difference in terms of the time F for a complete cycle. In the same way, the d1agram of Fig. 13 corresponds to the formula A l, I
ties on the respective antennas for the single consequent source of the fourth order. For the consequent source of the fifth order, the numbers run (1), (5), (10), (15)}, (5), (l), as given in the upper row of the iagram of Fig. 10. The numbers for the interior positions are obtained by multiplication, as will be readily apparent.
The efiect of combining consequent sources in the manner here illustrated in Fig. 11, is to gain the advantage of uni-direction over duo-direction, inherent in the one species, and the advantage of narrower angular distribution of intensity, inherent in the other species. In Fig. 11, curve 1 is the olar diaam for the antenna pair MN. urve 2 is t e polar diagram for the antenna pair RS, and curve 3 is the polar diagram for the combined array of Fig. 10.
Here as elsewhere in this scification, I
To arrive at the antenna array whose polar diagram appears in-Fig. 14, l first combine two of the -consequent sources of Fig. 13 to get a consequent source of the next higher order and again I combine two of these consequent sources to t another consequent source of the next higher order. Each such combination adds one antenna to the array and thus I get a row of four antennw constituting a consequent source of the third order which I represent by the formula X1 3 he l The general effect is to give a polar diagram 1 its llll
A 3 is's i' This adds one more antenna, making the complete array of five, and substantially converts the double loop of Fig. 13 into a single loop, as shown in Fig. 14. With a larger number of antennae, the diagram can be narrowed to any extent that is desired. The actual resultant currents on the five antennae are shown in magnitude and phase relation in the diagram of Fig. 15. In each antenna there is an alternatin current which may be represented accordlng to the well known convention by the projection of a rotating vector. Fig. 15 shows the five vectors whose projections as they rotate simultaneously will give the five currents in the respective antennae. Calling the angle 0 for the antenna designated I, and the current magnitude on this same antenna being represented by unity, the currents and angles for transmitting are given in the following table:
A l 3 A 3 sa (as?) 1 1, 0 II 3.96 169 13' 111 4.24 22= 30' IV 3.96 445 47' The resultant magnitude of intensity in the direction for a maximum is less than the magnitude on some of the individual antennae. Comparing the array of Fig. 14 with those of Figs. 1 to 11, it will be seen that the currents in the individual antennae of Fig. 14 buck one another to a certain extent more than for the earlier figures. On the other hand, the consecutive antennae are nearer together in Fig. 14 and hence the array is of comparatively less geographical extent. Of course, for substantially the same polar diagrams in the two cases, the energy radiated (in transmission) would be substantially the same, but for the array of Fig. 14 the current magnitudes in the apparatus would be greater and thus, to some extent, the true resistance losses would be greater. This slight disadvantage might be much more than compensated by the gain in having a compact geographical array.
For receiving purposes, the loss of energy in resistance would be immaterial because amplifiers could be employed to magnify the received energy. In receiving, the actual currents excited in the antennae would not have the phase and magnitude relations that they would have for transmitting, but the phase shifters and amplifiers shown in Fig.
14 would operate to make the current on the receiver correspond to the polar diagram. In Fig. 16, I have shown an antenna array of'nme antennae in a row, separated by intervals of only of a wave length. The
design of this arrany is reached as follows:
I start with a simple consequent source of the first order represented to two antennae.
wave length apart and differing in phase by half a complete period and corresponding to the formula- WE I build two of these into a consequent source of the next higher order, two of the resultants into a consequent source of one step higher-order, and so on until I have seven antennae in an array represented by the formula Then I take two such consequent sources, each represented by the foregoing formula and combine them into a consequent source, in accordance with the formula Since the antennae up to this point were wave length apart and since the two equal consequent sources last mentioned are to be.
wave length apart, this last operation adds two more antennae to the array, making the nine shown in the diagram. The details of a further discussion of Fig. 16 would correspond to those for Figs. 14 and 15 and .I will merely point out that the polar diaoff on the ground with their intersection at one antenna, then a: and y are the rectangular coordinates of the other antenna, and the currents in these two antennae are both of the same frequency 39/2, but they differ in phase by the angle 6. At a great distance from such an antenna pair, the intensity Will be different in different directions. Assuming polar coordinates. the intensity may be represented by a radius vector with origin near the antenna pair. The expression which gives the value of this radius vector in terms of the angle is called the interference pattern. In the present case the interference .pattern is given -by the radius vecmeasles tor whose value as a function of the independent variable angle b 1S which, when referred to the origin may be written as source located at sin pLF (ct) where is the resultant interference pattern. It follows at once that the two simple radiators or sources are equivalent to a single compound source, located at and having an interference pattern F(). This equivalent compound radiator or source is what I refer to as a consequent source of the first order.
Now consider two similar and equal consequent sources of the first order having in/ terference pattern F and relatwe displacement 40,, y,, Q9 It follows at once that they are equivalent to a single compound 2 ya a and having an interference pattern F,()
type T (2) Two consequent sources of 1st order source of 1st order of of ty e T displaced in accordance with type forming a consequent source of 2nd order of type T T,.
(3) Two consequent sources of 2nd order of type T T displaced in accordance with it it ti 1 secure this relation by a proper design of the antennae, and to indicate this, I have shown the antennas with their overhead structures of different size, somewhat in accordance with the foregoing table. It will be readily understood that those antennae that have larger overhead portions, and hence, larger capacity to ound, will take more current when supplied with a given electromotive force, and will radiate more energy.
To secure the desired quarter-phase lead for each transverse row to the rear, relatively to the direction of radiation, I introduce artificial lines or similar devices A, as indicated in the diagram. Of course, it will he understood that to keep all the antennw in the same transverse row in the same phase, it may be necessary to em loy hase restoring devices, which can he intro uced into the transverse feeders accordingly.
in Fig. l8 l have shown overhead connections between the antennn of the array b which certain antennae are fed throng others. The artificial lines or' other devices A are intended to secure the proper phase shift for each transverse row to the rear, relatively to the direction of mam'mum radiation. The devices A are intended to secure such an adjustment that all the antennas oi the same transverse row shall oscillate to.- gether in the same phase. The design of the several antennae may be varied to contribute to securing the proper hase relation, as well as the'proper amp 'tude relation.
The disclosure in connection with Figs. 17 and 18 has had reference more particularly to transmission, but the structure and operation for receiving will be evident, in accordance with the principles heretofore disare , understood how the consecutive antennae asshown in the dia-' grams. Also the antennae may be energized by respective local sources, controlled by pilot currents from a central station as disclosed in the Buckley Patent No. 1,301,644 of April 22, 1919.
It is not necessary that an array be merely linear or rectagular to realize the principle of my invention. Fig. 19 shows a non-rec tangular contour, obtained in this case. by superposing one rectagle on another. -The intensities on the antennae common to both rectangles will be the respective sums of the intensities they would have if they belonged to the rectan'g s apart.
What I claim is:
1. A plurality of antennae distributed in rectangular array in two dimensions with more than two antennae along each dimension, a common station conductively connected with said antennae, and means interposed in the connections to get the currents in progressive orderly phase relation along one direction and in progressive orderly phase relation alon the other direction.
2. A plurality o antennae distributed in rectangular array vin two dimensions with more than two antennae alon each dimension, a common station con uctively connected withsaid antennae, and means interposed in the connections to get the currents in progressive orderly phase relation along one direction and in progressive orderly phase relation along the other direction, said antennae being spaced a half-wave length along one dimension and a quarter-wave length along the other dimension.
3. The method of securing uni-directional transmission or reception with a rectangular antenna array having more than two antennae along each dimension, which comprises grading the intensities of excitation along the rows and columns of the array according to the coeflicients of the binomial expansion.
4;. The method of securing unidirectional transmission with a rectangular antenna array, which consists in grading the intensities of excitation along the rows and colunms of the array according to the coeflicients of the binomial ex ansion and excitin the antennae a quarter p iase ahead for eadli interval in the direction of maximum intensity and in the same phase across that direction.
In testimony whereof, I have signed my name to this specification this 23rd day of December, 1920.
JOHN STONE STONE.