US2167735A - Antenna - Google Patents

Description

Aug. 1, 1939.

A. ALFORD' 2,167,735

ANTENNA 7 Filed March 17, 1936 2 Sheets-Sheet 1 nnv/x MAM) INVENTOR ANDREW ALFORD ATTORNEY Au .1,1939. L R 2,167,735

ANTENNA Filed March 17, 1936 2 Sheets-Sheet 2 lNVENTOR 1 ANDREW ALFORD BY g I ATTO RN EY til Patented Aug. 1, 1939 UNITED STATES PATENT OFFICE ANTENNA Application March 1'7, 1936, Serial No. 69,292

7 Claims.

This invention relates to antenna structures for radio communication and pertains more particularly to lei-directional antennas adapted for short wave communication.

An object of the present invention is a bidirectional antenna which has a nearly constant input impedance.

Another object of the present invention is a pi-directional antenna which may be conveniently installed on a ship, and which will require only relatively simple arrangements for efliciently transferring the power from the antenna to the transmission line, and from there to the receiver when this antenna is used for receiving, and from the transmitter to the transmission line and from there to the antenna when the latter is used for transmitting.

.In the past, two types of long wire antennas have been used; the terminated antennas and the open-ended antennas. The terminated antennas have always been unidirectional, while the open-ended antennas have always been bidirectional, unless provided with suitable refiectors.

The terminated antennas, when properly terminated, have input impedances which remain nearly constant as the frequency is varied and for this reason they are well suited for use at a number of different frequencies.

The open-ended antennas have input impedances which vary with frequency within rather wide limits and consequently they are not suited for use at a number of different frequencies except when used in conjunction with relatively complicated matching devices.

Thus, generally speaking, the terminated antennas are superior to the open-ended antennas in those instances in which the constancy of input impedance is one of the primary requirements.

Now, there are applications which require bidirectional rather than uni-directional antennas. For instance, aboard a ship, which sails back and forth approximately along the great circle between two ports, antennas should preferably be, in certain instances bi-directional rather than uni-directional, so that the ship may communicate with both ports.

Under these circumstances, in the past, the open-ended type antennas would have been used, since the terminated antennas of the prior art were uni-directional unless provided with objectionable switching arrangements and transmission lines extending from the transmitter or receiver, as the case may be, 1' 100th n s Of the antenna. Since in practice it is often found that a transmission line from the translating device to the far end of the antenna cannot be installed, particularly on a ship, it is clear that in the past it would have been necessary to use bi-directional antennas having end impedances which vary with frequency over a considerable range, an undesirable characteristic where communication is to be carried out at several different frequencies.

In accordance with the present invention it is possible to construct a bi-directional antenna which has a nearly constant near-end impedance within a considerable range of frequencies.

The above mentioned and further objects and advantages of my invention, and the manner of attaining them will be more fully explained in the following description taken in conjunction with the drawings.

In the drawings Fig. 1 shows a simple form of antenna constructed in accordance with my invention.

Fig. 2 shows a modified form of antenna embodying my invention.

Fig. 3 shows the antenna of Fig. 1 installed on a ship in the immediate vicinity of a stay and means for detuning the stay, and

Figs. 4 and 5 show other arrangements for detuning stays.

In Fig. 1, reference numerals H and i2 indicate two inclined wires joined together at their far ends by jumper l3 and connected to a concentric tube transmission line l5 through an aperiodic coupling device It which matches the impedance of the antenna to the surge impedance of the transmission line l5. A resistance It the value of which is equal to the surge impedance of the wire I2 is connected between the lower end of wire l2 and the ground. Seventeen (ll) is a translating device, that is, either a receiver or a transmitter, 18 is a tower, and i9, 20, 2| are insulators.

The operation of the antenna shown in Fig. 1 may be described as follows: electric currents generated in transmitter ll proceed through transmission line [5 into matching device M. Emerging from l4 these currents proceed in the form of a travelling wave toward jumper I3. After passing through jumper it these waves proceed along wire l2 toward the terminal resistor I6. When the value of resistor I6 is properly chosen, that is, when it is equal to the surge impedance of wire l2, the waves travelling along 12 are not reflected at It.

For this reason there is no primary reflected wave on 12 which, travelling through l3, could cause standing waves along I I. Consequently wire II acts in approximately the same manner as though it were terminated into its surge impedance at the point of junction with jumper I3.

In operation wire II induces currents in wire I2, and vice versa wire I2 induces currents in wire II, this results in the formation of a certain amount of standing waves on both wires. The amount of these standing waves, however, is much less than one might expect. Thus, for example, I find by actual experiment at frequencies of the order of 10 to 20 megacycles that when wires II and I2 are the order of 1 /2 wave-lengths long and are placed a foot or two from each other the amount of reflected wave on portion I 0 of wire II adjacent to the coupling device I4 is only about 10%. This result is somewhat surprising particularly in view of the fact that one might expect some reflected waves due to reflection at the junction points of I3 with wires II and I2, in addition to the reflected waves produced by the mutual inter-action of wires II and I2.

Because of the relative absence of reflected waves in portion II] of wire II, the input impedance of the antenna at I4 is nearly independent of the frequency and consequently matching device I 4, capable of matching two fixed impedances, namely the impedance of the antenna and the surge impedance of I5 at various frequencies within a certain band, is all that is required for properly terminating transmission line I5. When line I5 is so terminated transmitter I! will work into nearly the same impedance at all frequencies within the band in which matching device I4 is operative.

When it is desired to carry on communication between two points separated by a distance greater than about 400 miles and when a horizontal reflecting surface such as grounder sea is located at a distance of a fraction of a wavelength below the lower end of the antenna it is found that the following table gives approximately the relation between the length of each of the wires II and I 21/). in wavelengths and the angle 0 to which they should be inclined to the horizontal.

Degrees The values of 0 for Z/x not given in the table may be interpolated or extrapolated.

In a few cases when the lower end of the antenna is located at a considerable height above the reflecting surface it may be found that the reflected wave is detrimentally out of phase with the direct wave, in which case the height, or angle of inclination or other constants of the antenna may be adjusted to correct this difliculty, in a manner which will be obvious to one skilled in the art.

Fig. 2 shows a modification of the antenna shown inFig. 1. The difference between these antennas of Figs. 1 and 2 is that wires I I and I2 in the antenna of Fig. 2 do not end near the tower but continue downward on the other side of the tower. The various parts and devices which were already described in connection with Fig. 1 have been designated by the same numbers inFig. 2 in order to save repetition. In Fig.

,lengthaswires III and II2.

2 wires III and I I2 are continuations of wires II and I2. Jumper I3 in Fig. 2 connects the ends of I I I and II2 instead of the ends of I I and I2.

The operation of the antenna of Fig. 2 is quite similar to the operation of the antenna in Fig. 1. The gain which is obtainable with the antenna of Fig. 2 is somewhat greater than that obtainable with the simpler structure of Fig. 1. The interaction between wires II, I2, III, and II2 is also somewhat greater than the interaction between wires of the simpler structure of Fig. 1.

For example, when wires II, I2, I I I, II2 are made each about two wavelengths long and the distance between II and I2, and III and II2 is 6 feet, I find that any variation in the antenna impedance at I4 is but a small percentage of the variation which would occur with an open-ended antenna.

The angle at which wires II and I2 as well as the angle which wires III and II 2 should make to the horizontal may be determined from the table. Wires II and I2 need not be of the same In some cases, however, it may be necessary to install phase correcting means at the point of junction of wires II and III and wires I2 and II2. The operation of such phase correcting means has been described in my copending application, Serial No. 18,995, filed April 11, 1935.

Fig. 3-shows the antenna which was discussed in connection with Fig. 1 installed aboard a ship. In this figure numbers III to 2I refer to the same parts and apparatus which have already been described in connection with Fig. 1.

In practice, when an antenna is installed aboard a ship it is often found that stays, low frequency antennas and other wires or metal masts may interfere with the proper operation of the antenna. Thus, for example, stay 30 in Fig. 3 may sometimes interferewith the proper action of the main radiating wires II and I2 of the antenna.

The mechanism of the interfering action is briefly as follows: the radiating portions of the antenna located in the vicinity of a stay produce an electric field which has a component along the stay. This component of the electric field produces an electromotive force which sets up currents in the stay. The magnitude of the current so produced in the stay depends on two factors; namely, (1) the magnitude of electromotive force along the stay and (2) the self impedance of the stay. When high frequency currents flow through a long conductor some energy is always radiated into space. In this respect a stay does not differ from an antenna wire. Since the phase of the radiation from a stay may have any relation whatsoever to the phase of radiation from the inducing antenna, this parasitic radiation from the stay may either increase or decrease the field produced by the antenna in a given direction. The more sharply directional is the inducing antenna, the greater is the probability that a stay or any other haphazardly placed wires in the vicinity of the antenna will decrease, rather than increase, the total radiation in the desired direction.

In addition to being able to scatter energy by radiation a stay may dissipate a substantial portion of the energy by turning it into heat. This occurs when the internal resistance of the stay is fairly large in comparison with its radiation resistance.

When the internal resistance of a stay is very high so that the major portion of the energy which it picks from the antenna is dissipated as heat, the small radiated portion does not affect the directional characteristic of the antenna as seriously as in the case when the internal losses in the stay are relatively small and when most of the picked up power is radiated. Thus, a high resistance stay tends to decrease the total radiated power without distorting the directional characteristic of the antenna while a low resistance stay tends to distort the directional characteristic without changing the total radiated power.

Since both the power radiated from, as well as the power dissipated in a stay, is proportional to the square of the current which is induced in it, it is clear that the effect of a stay on an antenna may be controlled by controlling the induced current. It has already been pointed out that the magnitude of this current depends among other things on the self impedance of the stay. By making the self impedance of a stay as large as possible the induced current in it may be reduced to a minimum.

If a stay may be cut and insulators inserted at frequent intervals, the self impedance of each section may be made so high that the stay becomes entirely inactive either as a radiator or a dissipator. This procedure is well known in prior art.

When, however, a stay can not be cut and broken up by insulators the problem of eliminating the effect of the stay on an antenna is much more diiiicult. In the first place, since the self impedance of a stay of a given length varies with frequency and at a given frequency may have any value lying between the radiation resistance, which may be of 100 ohms, up to 10 or 15 times the radiation resistance, it is clear that, in general, a given stay may not interfere with an antenna at one frequency and still cause a considerable distortion of the directional characteristic at another frequency.

Let us, therefore, assume that at the particular frequency at which antenna of Fig. 3 is to be operated, stay 3i] has a low self impedance so that it picks up and scatters a considerable amount of power which would normally be radiated by the antenna.

Under these circumstances the induced cur rent in the stay will be distributed in the form of standing waves. This distribution of current is diagrammatically indicated in Fig. 3 by the dotted wavy line i.

Since the position of current maxima and minima with respect to the ends of a stay depends on the relation of terminal impedances to the surge impedance of the stay, the distance of the first current maximum from one of the ends may have any value between minus A, wavelength and plus wavelength depending on the terminal impedance. When a stay is terminated by a good insulator the first current maximum occurs approximately at A; wavelength from the insulated end. When, as in Fig. 3, a stay is terminated by a very large metal object such as funnel 34, without insulation, the first current maximum occurs but a small fraction of a wavelength from the end of the stay. For this reason if we imagine that stay 30 were cut at a point J the impedance of the short position of the stay between J and funnel 34 as seen at J would gradually increase to a maximum as J is moved from 34 toward a point A; wave from 34. This maximum impedance seen at J when looking into J-34 is usually of the order of 1500 ohms. If then at J, which is located at a fraction of a wavelength from a current maximum, there were connected an auxiliary wire 31 there would be directed a substantial portion of the current into this auxiliary wire. The shelf impedance of the whole stay would, of course, change and the effect produced by varying the length of 3| would be approximately the same as the effect which would be produced by altering the length of stay 30 when 3| is disconnected. Thus, the self impedance of the stay may be controlled by varying the length of the auxiliary wire 3|. By controlling the self impedance in this manner the current in the stay may be reduced to a small fraction of what it is when the stay has a very low self impedance. The wire 3| may be terminated by an insulator 32, or may be connected directly to ground or to a large metal object. In the latter case the wire 3| should preferably be about A; wavelength shorter or longer than when it is terminated at an insulator.

It sometimes happens that after a stay has been detuned at one frequency, it acquires a low self impedance at another frequency at which it may be desired to use the antenna. In such a case an additional degree of freedom is required. Fig. 4 illustrates how this additional degree of freedom may be secured in practice. In this figure 3| and 35 are two auxiliary wires connected at two points J1 and J 2 to stay 30. By adjusting the length of these wires one can detune stay 30 at two frequencies. The points of junction of the auxiliary wires with the stay may either incide or be separated by a substantial distance.

In Fig. 4 there is shown still another arrangement for the same purpose. This: arrangement consists of the auxiliary wire 36 to which there is connected another auxiliary wire 31. By adjusting the length of wires 36 and 3'! as well as the points of junction J3 and J4 it is possible to detune stay 33 at two different frequencies. All of the auxiliary wires in Fig. 4 may be either terminated by insulators or connected to large masses of metal depending upon which of the various possible arrangements results in greater detuning effect and is simpler to erect.

In Fig. 5 another method is illustrated for re ducing the undesirable effects of a stay. This method is particularly useful when. the antenna with which the stay is interfering is a receiving antenna.

It has already been pointed out that a high resistance stay has relatively little effect on the directional characteristic of an antenna and that it can merely affect the total radiated or received power. Since the signal to static ratio is not affected when the gain of an antenna is reduced equally in all directions but is usually seriously affected when the shape of radiation characteristic is distorted, it is clear that the presence of a high resistance stay in the neighborhood of a receiving antenna is not nearly as objectionable as is the presence of av low resistance radiating stay.

In Fig. 5 is shown an arrangement which enables one to convert a low resistance stay into a high resistance stay. The arrangement in its simplest form consists of an auxiliary wire 50 and a resistor 5!. The value of resistor 5| is not critical but it should be of the same order of magnitude as the surge impedance of the stay and of the auxiliary wire 50. The length of the auxiliary wire 50 is likewise not critical. This length may be a small fraction of the wavelength or a whole wave length or even longer depending on mechanical convenience. The distance between the point of junction of auxiliary wire 50 and stay 30 and the end of the stay, however, should be chosen with reasonable care. The proper position of junction J is fixed by the following considerations. If at some frequency F, point J falls at a current maximum along stay 30 the addition of wire 50 and resistor 5| has little effect and little current is diverted into 50 under these circumstances. On the contrary when point J falls at a current minimum the major portion of the stay current is diverted into wire 50 and hence into resistor 5|. In this latter case a large portion of the power in the stay is dissipated in resistor 5| and the stay acts as a high resistance stay. When point J is a small fraction of a wave length from the current minimum the effect of network 50, 5! is still quite considerable. For this reason the effect of this network is not confined to a single frequency but is spread out over a band of frequencies. In fact by using two such networks, one on each end of the stay, it is often possible substantially to suppress the effects of a stay within the whole range of frequencies received by an aperiodic antenna.

So far we have assumed that there is in the vicinity of the stay some large metal object to which resistor 5| may be connected. When this is not the case either an auxiliary counterpoise may be constructed for this purpose or else resistor 5| may be placed between wire 50 and the point of junction J. In this latter case wire 5!] will function as a good counterpoise as long as it is roughly A; wavelength long and is insulated at its far end.

While I have described particular embodiments of my invention for purposes of illustration, it should be understood that various modifications and adaptations thereof, occurring to one skilled in the art, may be made within the spirit of the invention as set forth in the appended claims.

What is claimed is:

1. A bi-directional antenna comprising two substantially parallel radiant acting conductors spaced a short distance apart, each of said conductors being at least about a wavelength of the operating frequency long, inclined at an angle to the horizontal and having their upper ends connected together, a transmission line connected to one of the lower ends and an impedance equal to the surge impedance of the antenna connected between the other of the lower ends and ground whereby traveling waves are caused to fiow in opposite directions in said two conductors so as to render the radiant action pattern of the complete antenna bidirectional.

2. A bi-directional antenna comprising two substantially parallel radiant acting conductors spaced a short distance apart, and each arranged in the form of an inverted V, the ends of said conductors at one side of the V being connected together and the ends of said conductors at the other side of the V being connected to a transmission line, and through an impedanceto ground, respectively.

3. A- bi-directional antenna comprising two parallel radiant acting conductors of a length in the order of at least a wavelength at the operating frequency of said antenna inclined with respect to the horizontal at an angle so related to the length of said conductors that the radiant action pattern of each of said conductors is effected in the desired direction and connected together at adjacent ends, a transmission line connected to one of the remaining ends and means for suppressing reflections connected to the other of the remaining ends whereby a bi-directional radiant action pattern is produced for the complete antenna by the addition of the unidirectional patterns of radiant action of all the conductors comprised in said antenna.

4. A lei-directional antenna comprising two parallel radiant acting conductors of a length in the order of at least a wavelength at the operating frequency of said antenna inclined with respect to the horizontal at an angle so related to the length of said conductors that the radiant action pattern of each of said conductors is efiected along the desired direction and having their upper ends connected together, a transmission line connected to one of the lower ends and means for suppressing reflections connected to the other of the lower ends whereby the radiant action pattern of the whole antenna is bi-directional.

5. A bi-directional antenna comprising two parallel radiant acting conductors each arranged in the form of a. V, the ends of said conductors at one side of the V being connected together and the ends of the conductors at the other side of the V being connected to a transmission line and to means for suppressing reflections of waves traveling along the antenna, respectively.

6. A bi-directional radiant acting system comprising a wive translator, a terminating impedance, a plurality of substantially parallel radiant acting conductors each of a length in the order of at least a wave length at the operating frequency of said antenna serially connected between said wave translator and said terminating impedance the said conductor connected to saidterminating impedance having a surge impedance substantially equal to said terminating impedance, at least two of said radiant acting conductors being inclined with respect to the horizontal at angles so related to their length and elevation above ground, that the radiant action patterns thereof are substantially unidirectional for waves traveling-in one direction therealong, and all said conductors being so oriented and spaced that the radiant action pattern of the whole system. is predominantly bi-directional.

7. A lei-directional radiant acting system comprising a wave translator, a terminating impedance, a plurality of substantially parallel radiant acting conductors each in the order of at least a wavelength long serially connected between said wave translator and said terminating impedance, said conductor connected to said terminating impedance having a surge impedance substantially equal to said terminating impedance, at least two of said radiant acting conductors being inclined with respect to the horizontal at angles related to their lengths substantially in accordance with a smooth curve defined by the appended tabulated relationship, and all said conductors being so oriented and spaced that the radiant action pattern of the whole system is predominantly bi-directional Length in wave- 1 lengths 1 2 3 4 5 6 Angle of inclination 47 34 27 24 22 20 ANDREW ALFORD.

Images