US2983874A - Phase responsive alternating current networks - Google Patents

Description

y 1961 HENRI-GEORGES DOLL 2,983,874

PHASE RESPONSIVE ALTERNATING CURRENT NETWORKS Original Filed March 23, 1946 4 Sheets-Sheet 1 Z3--- AMPLIFIER Fl5.|.

26 F L. i

I NETWORK 32 AMPLIFIER W H V a VARIOMETER DELAYING 4 EXC'TAT'ON TUBE NETWORK 4 RECTIFIER F I G- 4 lNVENTOR H ENRl-GEORGES DOLL W W jwM HIS ATTORNEYS i y 1961 HENRI-GEORGES 001.1. 2,983,874

PHASE RESPONSIVE ALTERNATING CURRENT NETWORKS Original Filed March23, 1946 4 Sheets-Sheet 2 S X Wm E mm v E I EB m HL NN N EM .MA T E A R0 N N N E A C RG .m *F U% E8 L MM H R w w P U E J o P R n Y O o O I D G E W X T E E S W E G E A T. v N X N N IN A IE N A" U0 E H n H UP M NM. ES u M H EM R 5 C WW P R0 U 0 C P Y T FIGZA.

INDUCTlVE RESISTIVE INDUCTIVE RESISTIVE FIG.3A.

l I I I I I PERMANENT RESI STIVE COMPONENT (0 P) I 'I l I I I I I I l 511* H IS ATTORNEYS PERMANENT RESISTIVE COMPONENT (OF) May 9, 1961 HENRI-GEORGES DOLL PHASE RESPONSIVE ALTERNATING CURRENT NETWORKS Original Filed March 23, 1946 FIG.5.

4 Sheets$heet 3 AMPLIFIER x SE RECTIFIER 36 40 '49 TRANSMITTER con. P 4/ 37 48 3 EXCITATION A 414 n j OSCILLATOR AXIS RESISTIVE AXIS INDUCTIVE AXIS INDUCTIVE INVENTOR HENRI-GEORGES DOLL HIS ATTORNEYS May 9, 1961 HENRI-GEORGES DOLL 3,983,874-

PHASE RESPONSIVE ALTERNATING CURRENT NETWORKS 4 Sheets-Sheet 4 Original Filed March 23, 1946 HIS ATTORNEYS United States Patent vided and thisapplication Dec; 20, 1955, Ser. No. 554,217

sciaims: Grazia-166 This invention relates to manners of and systems for rendering. alternating current networks phase selective and, therefore, insensitive to components in quadrature to the selected phase. It relates further to systems for automatically maintaining such networks, or a part thereof, in a reference condition. y I

This application is a division ofthe applicants copending application Serial-No, 656,574, filed March 23, 1946, now Patent No. 2, Z53,520,lfor Methods'a'n'd Sys. tems for Maintaining Alternating Current Networks in a Reference Condition.

Infour-terminalnetworks, curren s supplied to," the two input terminals and indications or an output are obtained at or across-the other two terminals. I 7

An alternating current bridge is a typical form of four-terminal network in which changes in-the characteristics of components ofthebridge are reflected in the output of the bridge. These bridges are outgrowthsor developments of the Wheatstone bridge which has been used heretofore to determine unknownquantitiesby balancing the bridge initially, inserting the unknown, and rebalancing by means of calibrated standards or by inserting an unknown and observing a deflection of a suitable meter which has beencalibrated to indicate quantitles in accordance with: the amount of unbalanceof the bridge. v v 2 I Alternating current networksbased upon the principles of the alternating-current bridge, and particularly mutual impedance bridges,have been used for such purposes as the detection; of land mines and the detection of metal in lumber and other materials. These prior bridges have not been entirely satisfactory for the reason that they must be sensitive, and, for that reason, are difficult to maintain in proper balance. Under laboratory conditions; it is reasonably easy to maintain such bridges in a balanced condition, inasmuch as the various factors which tend to create'bridge unbalance are not subject to wide variations'and can be controlled. However, under .the conditions" of commercial or military usage, quite different conditions exist. The devices are subjected to wide variations in humidity, temperature, to rain and shockyand asa result, the-electrical characteristics and, in particular, the mutual inductance tends to vary and render the bridge unbalanced. Such unbalanced conditions usually can b corrected by provision of suitable manually operated adjusting devices, but in unskilled hands, errors may occur which render the indications obtained with thesystem undependable.- Moreover, the time required for=rebalancing the bridge manually is undesirable in military operations, In many cases, the bridges used in minedetection hadto-berebalanced aboutevery two'minutes, requiring 'about oneminute 'to rebalance the bridge.

In such devices asland mine detectors, the tendency of -the bridge circuitto become unbalanced is greatly erhphasizedfor -the=-reason that-- the"'-detectors-"must be 2,983,874 Patented May 9, 196} 'ice 2, used underall sorts of climatic conditions, and the opportunity for exactadjustmentof the detecting bridge network is, at best, limited. Moreover, the prior systems used in mine detecting have been sensitive primarily to metallic bodies rather than nonmetallic' bodies", and, therefore, they are not useful in detecting the nonmetallic type of mine or for that matter any other non metallic substance. The sensitivity of the bridge canbe increased to such anexteht that it might be possibleto detect nonmetallic mines buried in conductive ground, but if its sensitivity: were increased to this'extent, its rate of; drift would be increased so greatly that it would requirerebalancing' every few seconds, which is pro hibitive;

An object of the present invention is' to provide a detector system which has" sensitivity to" conditions or objects providing a' signal of one phase and which discriminates strongly against objects or conditions providing a signal of" adifferent phase.

Another object'of the invention is to provide means A further object'of the invention is to provide'alternatin'g current networks which are capable of automatically compensating for driftsinthe magnitude'and/or phase of-currentsin the network;

A-still: further object of'the invention is to provide means for impressing an unbalance of predetermined phase and magnitude on-asystem to render the system phase-sensitive.

Another object of; the invention is to provide fourterminal networks-*which are automatically stabilized to a predetermined signal 'level'in order to eliminate amplitude drift. 7

Another object of the invention is to compensate for any phase drift automatically and continuously. I

A still further: objectv of this invention' is to provide systems whereby the phase and magnitude of a voltage or-current in analternating current network can be effectively controlled. g I

Other objects of theinventio will become apparent from the following description of typical forms of systerns, embodying the present invention. I p v The foregoing phase selective objects have been attained by providing four-terminal alternating current networks in the output of which an unbalancing comonem, generally'either resistive or inductive, is introd'uced toflprovide a permanent unbalance, level or refer ence value which rendersthe network more sensitive to factors tendingto unbalancethe output in thesame phase, and less sensitive to eifects tending to unbalancethe out put in a phase in quadrature with the phase of the introduced component. By controlliugthe phase of the unbalance, it is possible to locate metallic objects while discriminating strongly against spurious eifects produced, for example, by saltwater or magnetic ground.

The terms resistive and"inductive ,as used herein have the following meaning. 7

An electromotive force su'ch as appears in the output from a 1 four-terminal network will be referred to as "in ductive" if its phase with respect" to' the energizing "input currentis either 270". I

Similarly, an output ele'c't'romotive force will be re ferred to" as"resistiv if *its 'phasewith' respect to the inputcur'rentiseither-0 or The above definitions were chosen because they con form to the action in mama; impedance bridges such as will be"describedhereinafter? If,"for'ei{arnple," a purely inductive material, such as powdered iron,is brought into themagntic fieldofisuchmutual impedance bridge, it produces a signal of the inductive 'phas'ei" Since tliis aeeaen 3 electromotive force is produced by an inductive material, it is convenient to refer to this as the inductive component of the signal.

' In the same way, if a purely resistive medium is introduced into the magnetic field, is produces a signal of the resistive phase.

In general, it may be said that, at the frequencies generally used, most metal objects give both inductive and resistive components which are of the same order of magnitude so that either one can be used for detection.

Most of the other materials which provide spurious effects give signals having only one of the two components; for example, sea water gives signals that are substantially purely resistive, while magnetic ground gives a substantially purely inductive signal. Thus, in accordance with the present invention, by introducing a resistive component into the network, it is possible to discriminate against the effects of materials giving inductive signals such as magnetic ground and thereby permit the location of mines in such ground by detection of the resistive component of their signals. Conversely, it is possible to discriminate against the resistive signal of sea water and detect mines in such water by the inductive component of their signals.

Further, in accordance with the invention, four-terminal networks are provided with means for returning the system to a substantially constant reference value or level of unbalance to compensate for drifts. Preferably, the automatic balancing system is arranged to have a delayed action so that sudden variations of the magnitude of a signal such as those obtained when passing over a mine can be indicated in the output of the system. Such automatic balancing of the system is particularly desirable in case of vehicular-mounted detectors which have to operate continuously for a long period and the efiiciency of which would be greatly reduced if the vehicle had to be stopped every few minutes for rebalancing. Vehicles equipped with such an automatic system can be operated for as long as a day or even longer without requiring an adjustment, whereas earlier models had to be rebalanced every few minutes. Similarly, with detectors of very high sensitivity used for nonmetallic mine detection, the drifts become so fast that field detection is impossible, unless the detectors include automatic balancing systems of the type embodying the present invention.

Moreover, in accordance with the invention, the system may be arranged to nullify automatically any unbalance in quadrature with the phase selected for permanently unbalancing the system. Inasmuch as such quadrature unbalance is unimportant from a detection or signalling standpoint, usually the automatic balancing system therefor will not purposely be delayed, but instead will be arranged to nullify as quickly as practical the effects of such quadrature unbalance.

Systems of the type described generally above are useful for many different purposes such as, for example, in the detection of land mines, the inspection of materials formed of electrically nonconductive materials such as wood which might contain metallic objects or viceversa, detection of mines in salt water or other objects in conductive or magnetic media. Moreover, it may be used in order to overcome changing relationships between the elements of delicately balanced electrical systems wherein the characteristics of the system tend to be altered continuously by climatic or other conditions.

For a better understanding of the present invention, reference may be had to the accompanying drawings in which:

Fig. l is a schematic circuit diagram of a mutual impedance bridge provided with means for manually compensating for drift;

Figs. 2A to 2F are vector diagrams showing vector additions and combinations;

Figs. 3A and 3B are diagrams illustrating vectorially the use of a permanent signal or unbalance to make an electrical indicator phase selective;

Figs. 4 and 5 are circuit diagrams of typical devices embodying the invention including means for automatically balancing the circuit;

Figs. 6 and 7 are diagrammatic illustrations of the principle relied upon for automatically balancing drifts in quadrature to the phase of the selected component; and

Fig. 8 is a combination block and schematic diagram for an automatic network balance system which maintains constant both the phase and magnitude of signals provided by the system.

While the invention will be illustrated primarily in connection with a mutual impedance bridge as typical of a four-terminal network, it will be understood that the in vention has many other applications.

An example of such a four-terminal network is the arrangement disclosed in Fig. 1 which is a form of mutual impedance bridge that may be used for detecting metallic or nonmetallic objects. The device includes an oscillator 11 which supplies energy to the terminals 12 and 13 of a network having four terminals 12, 13, 14, and 15.

The oscillator 11 is connected in series with the primary 16a of a detecting coil system 16, the primary 17a of a bucking air core transformer 17, the primary 18a of variometer 18 and a tuning condenser 19. The primary 20a of a second variometer 20 is connected in series with a resistance 21 across the primaries 16a, 17a, and 18a. The secondary 16b of the detector coil 16 is connected in series with a secondary 17b of the transformer 17, the secondary 18b of the variometer 18, the secondary 20b of the variometer 20 and a tuning condenser 22 and to the input of an amplifier 23. The output of the amplifier 23 is fed to voltmeter 24 or other indicating device.

The coils 17a and 17b may constitute a closely coupled air core transformer which is situated at a suflicient distance from the coils 16a and 16b to prevent any appreciable interaction between their magnetic fields.

Physically the transformer 17 may be positioned in the center of the coils 16a and 16b, the axes of the two sets of coils being perpendicular.

For convenience, the circuit including the oscillator 11 will be referred to as the transmitter circuit and the circuit including the coils 16b, 17b, 18b, and 20b will be referred to as the receiver circuit.

In the transmitter circuit, the oscillator 11 causes current to flow through the primary coils 16a, 17a, 18a, and 20a, each of which induces a corresponding electromotive force in the secondaries 16b, 17b, 18b and 20b, in the receiver circuit.

In order for any alternating current bridge to be perfectly balanced, it is necessary that it be balanced for both amplitude and phase. The variometers 18 and 20 are included in the circuit for this purpose.

Inasmuch as the primary 18a of the variometer 18 is in series with a primary 16a and 17a, current of the same phase passes through these three primaries. The electromotive force induced in the secondary 18b of the vairometer 18, therefore, is in phase with the electromotive forces induced in the secondaries 16b and 17b. The variometer 18, therefore, may be used to adjust the inductive component in the receiver circuit.

The primary 20a of the variometer 20 is in series with the relatively high resistance 21 across the inductive part of the transmitter circuit. Because of the presence of the resistance 21 and because the coils 16a and 1711 have a reactance much larger than their resistance, the current through the primary 20a is nearly ahead of current flowing through the primaries 16a, 17a and 18a. The electromotive force induced in the secondary 20b of the variometer 20, therefore, is nearly 90 ahead of the electromotive force induced in the variometer secondary 18b. The variometer 20, therefore, may be used to adjust the resistive component of the signal.

.5 Combined operation of thevariometers. 18 and 20 permits balancing the bridge in amplitude and' phase.

While the variometers 18 and 20 have been used heretofore to balance the bridge to zero, in accordance with this invention, they are preferably used to adjust the magnitude and phase of a permanent bridge signal. Thus, if it is desired to detect metallic articles in salt water, for example, a permanent inductive component or unbalance of suitable magnitude would be introduced into the receiver circuit and adjusted by means of the variometer 18. The variometer 20 is used to balance out any resistive component that may be present.

Conversely, if it is desired to detect metallic materials in magnetic ground, a permanent resistive component or unbalance of suitable magnitude would be introduced into the receiver circuit and adjusted by means of the variometer 20, while the variometer 18 is used to balance out any residual inductive component that may be present.

This feature will be better understood by reference to Figs. 2A to 2F and 3A and 3B of the drawings.

Figs. 2A to 2F show various components and combinations of signal electromotive forces for purely inductive signals, purely resistive signals and combinations of the two. If the signal vectors are plotted in a system of rectangular coordinates, with the resistive components plotted along the OX axis and the inductive components plotted along the OY axis, a purely inductive signal or the inductive components of a signal, such as OB will be plottedalong the OY axis, as shown in Fig. 2A. A purely resistive signal OA will lie along the X axis as shown in 23. c

When two inductive signals are impressed simultaneously in the receiving circuit, they add their effect if they both have a phase of 90 or if they both have a phase of 270 with respect to the OX axis. One vector subtracts from the other if the phase of one is 90 and the phase of the other is 270.

Similarly, the resistive component vectors OC and CD in Fig. 2C have the same phase so that the resultant of the two is OD. Fig. 2D shows the combination of two similar resistive components of opposite phase. Thus, if OE having a phase of 0 is combined with vector EF having a phase of 180, the resultant is OF.

When an inductive component and a resistive component are supplied simultaneously to the receiving circuit, they combine as rectangular vectors as shown in Figs. 2E and 2F As shown in Fig. 2E if a resistive signal 06 is combined with an inductive signal GH of equal magnitude, the resultant OH has a phase of 45 and a magnitude 1.4 times as large as either of the component vectors. When one of the component facts is appreciably larger than the other, for example, the resistive component OJ of Fig. 2F combines with a substantially smaller inductive component JK, the resultant component OK has about the same magnitude as the component OJ. For example, if the inductive component JK is one-fifth as large as the resistive component OJ, then the magnitude of the resultant OK is only 2% larger than the larger component OJ. If the magnitude of the vector OJ is ten times larger than the magnitude of the vector JK, then the resultant vector OK is only one-half percent larger than the vector OJ.

If a conventional system were'used to locate metallic mines, for example, in a magnetic ground, a large number of false signals would result. In order to discriminate against these unwanted signals without impairing the resistive signals due to the mines,. the detector may be made much more sensitive to the resistive than to the inductive components. To accomplish this, a strong signal of the resistive phase is maintained in the receiver circuit, This resistive signal may be represented by the vector OP in Fig. 3A; The amplifier 23 or the meter 24 in Fig. 1 would then be biased by an amount almost corresponding to the permanent signal OP so that when a signal of magnitude O]? is flowing in the receiver circuit, there will be only a small deflection on the' meter 24'. If a metallic mine is encountered, it will cause an additional resistive signal PA (Fig. 3A) to flow in the receiver circuit. The composite signal would then be represented by the vector 0A. The amplifier and detector are so arranged that they respond only to the magnitude of the resultant signal. Thus, when the signal is increased from OP to 0A, a change in the deflection corresponding to the vector PA is noted on the meter. I

If the detector coil system 16 (Fig. 1) is brought close to magnetic ground which causes an inductive signal PB (Fig. 3B) to appear in the receiver circuit, the resultant vector OB has very nearly the same magnitude as the permanent signal OP and the meter 24would then be deflected only by a small amount corresponding to BQ. The signal change is so small, so long as the permanent signal is large, that the deflection of the meter 24 would be negligible.

In the same way, the detector can be made highly discriminative against resistive signals, while remaining sensitive to inductive signals, by introducing a permanent signal that is purely inductive.

In the circuit described above, or in any alternating current network of general or special application, it may be desired to maintain the circuit in a reference condition. In the embodiment of the invention described, it is convenient to maintain the bridge unbalanced to the selected degree, or to maintain a permanent unbalance or level.

Circuits for accomplishing this result are shown schematically in Figs. 4 and 5.

Referring to Fig. 4, an oscillator or other source of alternating current 26 supplies energy through a series tuning condenser 33 to input terminals 27a and 27b of a network 27. The output terminals 27c and 27d of network 27 are connected through one winding of a transformer 32 to an amplifier 28. The output of amplifier 28 is rectified and filtered by rectifier 29. The output of rectifier 29 passes through a delaying network 30 to a variometer tube 31 which maintains the magnitude of the permanent signal constant. The variometer tube 31 is excited by the voltage appearing across condenser'33. The output of variometer tube 31 is fed back' into the network output by means of transformer 32 or in any other suitable way.

In operation, the output of oscillator 26, and hence the current input to the network 27 is maintained constant. When something is altered in the network 27, for instance, when an electric circuit constant may be changed, or the coupling between parts is altered, the output signal appearing at terminals 27c and 27d changes. The change in signal may be either an increase or a decrease, depending upon the conditions. The change may be in magnitude, in phase, or both.

When the network'27 is at its standard or'reference status, a signal is fed to amplifier 28, rectifier 29, delaying network 30 and the variometer tube 31, which introduces a permanent signal of given phase into the output circuit of the network. The action of the tube 31 is such that conditions will be stable at this point. When any condition or conditions are changed in network 27, the magnitude or phase or both of the signals appearing at terminals change. If, for example, the signal suddenly increases, the increased signal will be fed to amplifier 28, rectifier 29, and through delaying network 30, to the tube 31. The action of the tube 31 is such that it then decreases the amount of permanent signal it has been introducing into the amplifier circuit. Due to the presence of delaying network 30, the compensation is not instantaneous, but has any characteristic delay required. The compensation is progressive, following an exponential law and it may take a few seconds, for example, for the bridge to be rebalanced. The action of the system,- and particularly the action of the automatic variometer tube will become more evident from the following description of Fig. 5.

Referring to Fig, 5, the four-terminal network may comprise a mutual impedance bridge having a primary or transmitter coil 34 and a secondary or receiving coil 35. In this bridge the two coils may be mounted at right angles so that normally no voltage is induced in the receiving coil 35. The transmitter coil 34 is energized by an oscillator 44, the transmitter circuit being tuned to series resonance by a condenser 38. The receiving coil 35 is connected to an amplifier 42, the output of which is rectified by rectifier 43. The output of rectifier 43 is applied through a bucking bias battery 49 to a potential indicating meter 48. The direct current output of rectifier 43 also passes through a conventional resistance capacitance delaying network which may comprise, for example, resistors 39 and 40 and a shunt condenser 41. After passing through this delaying network, the direct current potential is applied to the control grid of variometer tube 36 which may be, for example, a variable a pentode. The control grid of tube 36 is also excited by the alternating appearing across the condenser 38, the DC. being blocked by condenser 37. The remainder of the connections to the tube 36 are conventional. The plate of tube 36 is connected through a blocking condenser 45 to the primary winding of a transformer 46, the secondary of which is connected between one end of the coil 35 and the amplifier 42.

In operation, the oscillator 44 energizes the transmitter coil 34, the circuit being tuned by condenser 38. The A.C. voltage appearing across condenser 38 is applied through the blocking condenser 37 to the control grid of tube 36; thus, the output of tube 36 is modulated in accordance with the alternating voltage appearing across the condenser 38. The DC. bias on the control grid controls the amplification of the tube. The variometer tube 36, therefore, always has an A.C. output which is fed through a blocking condenser and the transformer 46 to the input of the amplifier 42. The reactance of the condenser 45 is small and can be neglected; the selfimpedance of the primary winding of the transformer 46 is small compared with the resistance which it sees in looking toward the tube 36; while the input impedance of the amplifier 42 is high compared with the impedances in its input circuit. The phase of the applied to the input of the amplifier is then at 90 or 270 with respect to the current flowing in the transmitter circuit. The action of tube 36 is, therefore, to insert an inductive component of signal voltage in the receiver circuit, the magnitude of this added signal being controlled by the DC. bias applied to the control grid. The action of the variometer tube 36 is similar to the action of the conventional variometer 18 or 20 of Fig. 1; hence the name variometer tube or electronic variometer. It differs from the ordinary variometer in that instead of requiring manual control, it may be controlled by varying its DC. bias and hence may be rendered automatic. The axes of transmitter coil 34 and receiving coil 35 are perpendicular so no normally is induced in the receiving coil due to current flowing in the transmitter coil. In this instance, for example, when no mine is detected, the only appearing in the receiver circuit is the inductive component induced by pentode 36.

The bias voltage supplied by battery or other suitable means 49 is adjusted so that a given deflection, such as half-scale, for example, to a false zero position is present on the meter 48 when no mine is present.

When a metallic mine is detected, for example, or when a signal is induced in receiver coil 35 as a result of the current flowing in transmitter coil 34, the input to amplifier 42 is altered. For the purpose of illustration, the signal produced by the detection of the mine may be considered to be in phase with the permanent signal produced by the tube 36. Accordingly, the input to amplifier 42 is increased suddenly and the output of rectifier 43 increases. When the output of rectifier 43 increases, due to the signal produced by the mine, a corresponding deflection will be noted on the meter 48. The output of rectifier 43 is also fed through the delaying network to the control grid of tube 36 so that the DC, bias on the control grid gradually becomes more negative, which in turn, decreases the amplification of tube 36 which reduces the magnitude of the inductive signal being introduced into the receiver circuit. The action continues until the total signal strength is approximately equal to the original permanent signal and the meter has returned to its original indication.

The action of the delaying network is purposely slow and may require a few seconds to restore the network to its reference condition. The reason for this delay is to permit signals produced by a mine to be observed.

Ordinarily, the coils 34 and 35 are swept parallel with the surface of the earth so that when they pass over a mine, a momentary signal is produced.

On the other hand, drifts that occur are gradual, normally, and the slow action of rebalancing or restoring the system is sufficient to compensate for these drifts without production of an observable signal.

In most cases, an apparatus incorporating the above salient features is sufficient. The only case not completely controlled and compensated for is the one where a network is subject to a large drift, the phase of which is in quadrature to the phase of the Wanted signal, that is, the permanent signal.

A typical example of this case is the detection of metallic and nonmetallic mines in magnetic ground. The magnetic ground introduces an inductive signal and also temperature changes cause a change in the relation of the coils which causes an inductive drift. It is, therefore, desirable to detect the mines by their resistive signal and discriminate against inductive signals or drifts.

This invention also provides means for compensating for quadrature drift, the principle being indicated in Figs. 6, 7 and 8.

Referring to Fig. 8, the transmitter circuit 50 including transmitter coil 50b and tuning condenser 51 is excited by an R-F oscillator 52. The appearing across condenser 51 is fed through a phase shifting network and voltage divider 53 to the control grid of a variometer tube 54. Tube 54 may, for example, produce a permanent resistive signal which is introduced into the receiver circuit by any conventional means such as transformer 55. The receiver circuit signal in the winding 50:: is amplified by amplifier 56 and rectified by rectifier 57 which also includes the delaying network. The output of rectifier 57 is used to bias the tube 54, thus maintaining the magnitude of the permanent resistive signal constant in precisely the same manner and for the same purposes as explained in connection with Figs. 4 and 5.

The E.M.F. appearing across condenser 51 is also fed through a phase shifting network and voltage divider 58 to a second variometer tube 59 which, with the aid of other circuit components, compensates for any quadrature signal, so its phase is adjusted so that it injects a purely inductive signal into the receiver circuit through a transformer 60. The bias on tube 59 is controlled in a different manner, however. To control the bias on tube 59, a third tube 61 and associated circuit is employed. The across condenser 51 is applied through a phase shifting network 62 to a grid of a tube 61, so a radio frequency potential appears in its plate circuit. The R-F potential is amplitude modulated at a much lower frequency, for example, by a 1,000 cycle derived from a 1,000 cycle oscillator 63. The tube 61 may, therefore, be referred to as a modulator tube. The modulator tube 61 introduces the 1,000 cycle modulated R-F signal of inductive phase containing no resistive comaasasva ponents through transformerv 64 into the receiver circuit. When the mutual' impedance bridge isbal'anced, the receiver circuit signal may contain:

(a) A permanent R-F resistive signal produced by the electronic variometer tube 54;

(b) An inductive R-F signal produced by the variometer tube 59;

(c) A constant R-F inductive signal (carrier) produced by tube 61;

(d) Two R-F signals of inductive phase'(sidebands) also produced by tube 61 and spaced 1,000 cycles per second (the frequency of the modulating signal) above and below the carrier frequency of signal 0.

The connections to transformers 60 and 64 are polarized so that in the receiver circuit the inductive R-F signal of b above is opposite in phase to the inductive R-F signal of above.

Fig. 6 shows graphically the composite signal appearing in the receiver circuit. In the case of Fig. 6, b is equal and opposite to c inasmuch as the result is zero, there is no inductive signal from these two combined sources. The modulated output of modulator 61 is constant both in phase and magnitude. The phase of the output of tubes 54 and 59 is constant, but the magnitudes may change as the result of the automatically controlled D.C. biases fed to their grids.

Fig. 6 illustrates graphically the case Where the permanent signal is purely. resistive, that is, when there has been no drift in the inductive component. Fig. 6, therefore, portrays the conditions when there is'no need for compensation of the quadrature (in this case inductive) component 0A is the permanent resistive signal. AB A132; A13 and AR, withmaximum amplitude at AB and represent some of the inductive components successively applied during one modulating period to the receiver circuit by modulator tube 61, which is amplitude modulated at a 1,000 cycle rate, the center of the swing being at A. Ifthe arc C-C represents the cutoff level of the amplifier, then the R-F output at difierent intervals during an audio cycle may berepresented as C A, C B CIBI, COA, C3133, C4134, C3B3, and back to COA. X1, X2, X1, X0, X3, X4, X3, andX d, then the radio frequency output may be represented by the area under the curve at the right. Inspection of this curve will reveal that during a single audio cycle, two maximums, X and X and two mini- 'mums Xb, appear. in this case, X =X The radio frequency signal in the receiver circuit is, therefore, modulatedfat 2,000 cycles and higher harmonics. If in one branch of' the receiver circuit, it is possible to separate the audio frequency from the R-F, and if the-audio signal is-"then fed-througha filter which blocks frequencies of 2,000 {cycles and above, then there would be no output from this branch of the'circuit for the conditions assumed rang. 6. Toaccomplish this, the output of R-F amplifieii'56 in'Fig. Sis connected to a detector 65, the output of'whi'ch is the audio signal. The output of detector 65 is connected tothe input of a low-pass filter 66 which passes frequencies of 1,000 cycles and below. The outp'utof filter 66*would, therefore, be zero when the composite permanent signal is purely resistive as shown in Fig. 6." This is the condition it is desired tomaintain. In order to 'see how this conditionis maintained, it is necessaryto consider the case where an inductive drift 'has' occurred; that is, where the permanent signal contains a'quadrature component. Such a situation is depicted in Fig. 7. Referring to Fig. 7, the permanent signal 0A the same magnitude as 0A in Fig. 6, but it an angle with the OX or resistive axis. The audio .modulation then produces a swing from E to B and the; radio; frequency signal inthe receiver circuit during an audio cycle is shown by the curve on the right of Fig. 7. X is no longer equal-to X as was the case of the corresponding points on Fig. 8A. The R-F output increases from the} v lue X0 99 a Xz' Whichis above,

passed through only once on each complete audio cycle. The radiofrequency output, therefore, contains. an appreciable 1,000 cycle modulation. It also contains a 2,000 cycle modulation and higher harmonics as before. It should be noted at this point that the greater the drift in the inductive signal becomes, that is, the greater the angle beween 0A and the OX axis, the greater becomes the proportion of 1,000 cycle modulation in the output.

Accordingly, since the percentage of 1,000 cycle modulation is directly related to the inductive drift, detecting the 1,000 cycle modulation in the signal will furnish an indication of the relative magnitude of the inductive component. Further, the 1,000 cycle modulaton may be used to control the automatic compensation for the inductive component. This is what is accomplished by the circuit of Fig. 8. The grid bias and, therefore, the gain of the inductive phase variometer tube 59 is controlled by the 1,000 cycle modulation appearing in the output signal.

When an inductive component is present in the permanent signal, a corresponding amount of 1,000 cycle energy is put out by filter 66. This 1,000 cycle signal may be amplified by audio amplifier 67 and its phase adjusted if necessary by a phase adjustment network 68. The output of network 68 is connected to the primary 73a of an audio transformer. In order to control the grid bias of tube 59 by this 1,000 cycle signal, it is necessary to change it to D.C. Also, as the inductive drift may be in either direction, the relative polarity of the D.C. must also be controlled; that is, the grid bias may have to be greater or smaller than standard. This will be better understood by again referring to Fig. 7. While the point A has been shown above the OX axis, the drift might have been such as to cause the point A to lie below the OX axis. If the point A is above the X axis, the 1,000 cycle modulationof the total RF signalwill have one phase relation with the audio oscillator 63. If the inductive drift is in the opposite direction, that is, if the point A? is below the X axis, the phase relationship between the 1,000 cycle modulation of the total RF signal and the oscillator 63 will be changed 180. Therefore, an indication of the relative phase of the 1,000 cycle modulation of the RF signal will tell the direction of the drift. This phase relationship is utilized to control the direction of change at the grid bias on tube 59. The intensity of the 1,000 cycle modulation of the total RF signal is utilized to control the amount of change of the grid bias.

It is, therefore, possible to provide a D.C. bias which may be changed in both magnitude and direction with respect to the reference values. This is accomplished by the rectifiers 69 and 70 and associated elements as shown in Fig. 8.

Two diode rectifiers 69 and 70 are connected in opposition, with resistances 71 and 72 between their cathodes,

as shown. The two plates are connected together through the secondary 73b of the audio transformer 73. A 1,000 cycle signal from oscillator 63 is fed to the com- 'mon point of resistors 71 and 72 and the center tap of transformer secondary 73b. The output 74, 75 is taken ofi the cathodes of the two diodes.

This portion of the circuit functions similarly to a wattmeter. There is no output on leads 74, 75 if the diode circuit is energized only by oscillator 63. Likewise, there is no output if the diode circuit is energized only by a 1,000 cycle signal appearing in primary 73d of the transformer 73. When energized simultaneously by signals from both sources, any one of three conditions may result:

(1) There may be no output.

(2) The D.C. output may become relatively more positive, increasing the bias on tube 59.

(3) The D.C. output may become relatively more negative, decreasing the bias on tube 59.

The degree of the change in output in steps 2 and 3 depends upon the amount of 1,000 cycle modulation; that is, it is a function of the intensity of the 1,000 cycle signal appearing in primary 73a.

In order to better understand the operation of the diode circuit, let it first be assumed that the signal is derived only from oscillator 63 (hereinafter referred to as the oscillator signal); secondly, that the signal is derived only from the bridge drift and is the signal appearing in primary 73a (hereinafter referred to as the drift signal); and third, that both an oscillator signal and a drift signal are applied simultaneously. The third case is the one corresponding to actual operation of the equipment.

When only an oscillator signal is applied to the diode circuit, the circuit constitutes a balanced circuit or bridge as the signal encounters an identical circuit whether it follows the top loop including diode 69, or the bottom loop including diode 70. Suppose at one instant of time the point 77 is positive. The plates of both diodes are then positive and both diodes are conductive. A current then flows in both resistors 71 and 72, these currents being equal and opposite. Each may be volts for example. If resistors 71 and 72 are shunted by large enough condensers 78 and 79, respectively, then leads 74 and 75 are maintained at their peak positive potential, which may be assumed +10 volts. During the other half cycle, the point 77 is negative, so both plates are negative, so both diodes are nonconductive. Condensers 78 and 79 maintain the output potential during this half cycle. As the grid bias applied to tube 59 is the difference in potential between leads 74 and 75, there is no output; that is, there is no grid bias applied, when only an oscillator signal is applied to the diode circuit.

When the second condition obtains and only a drift signal is applied to the diode circuit, one diode or the other is conductive depending upon polarity of the induced in secondary 73b. If at one instant the plate of diode 69 is positive, then the plate of diode 70 is negative. Diode 69 is then conductive causing current to flow through resistor 71 and charging condenser 78. The circuit is completed through the low impedance of oscillator 63. At this instant, lead 75 becames positive with respect to the reference point 76; it may be +6 volts, for example. During the opposite half cycle, diode 70 conducts, causing current to flow through resistor 72 and charging condenser 79 to the peak potential. A positive potential, therefore, appears on lead 74. This potential is equal to the potential previously applied to lead 75. Although diode 69 is nonconductive, the positive potential on lead 75 has been maintained by condenser 78. Accordingly, over a period of one cycle or more, both leads 74 and 75 are positive and their potentials are equal. Therefore, the action of a drift signal alone does not change the grid bias applied to tube 59.

In operation, the oscillator signal is continuously applied to the diode circuit. A drift signal may or may not be applied to the circuit, depending upon whether or not a drift of the inductive phase has occurred in the detector unit. If there is no drift, then obviously there is no need to correct for one. If a drift has occurred, a 1,000 cycle signal will appear in primary 73a, as previously explained. The relative phase of this 1,000 cycle signal with respect to the 1,000 cycle modulation of the modulator tube will depend upon the direction of drift.

For the sake of explanation, let it be assumed that at one instant of time the potential of the point 77 is positive due to the oscillator signal. At the same time the drift signal is such that it tends to make the plate of diode 69 positive and the plate of diode 70 negative. The plate of diode 69 is, therefore, made positive by two additive sources, so the potential of lead 75 may be +14 volts. Also at this same time, two potentials are being applied to the plate of diode 70; the potential due to the oscillator signal is positive and the potential due to the drift signal is negative. The instantaneous difference between the two is then the plate potential with the result that diode 70 passes less current and the potential of lead 74 will be less than +10 volts; it may be +6 volts. Lead 75 is then 8 volts positive with respect to lead 74,

i 12 so the grid bias on tube 59 is changed by 8 volts. when the bias on tube 59 becomes more positive, it passes more current, so it injects an inductive component signal through transformer 60 into the composite signal.

Inasmuch as this drift represents an unwanted signal, there is no need to delay the action of the automatic balance.- If, for some reason, it would be desired to delay the compensation for quadrature (in this case, inductive) drift, it would only be necessary to insert a delaying network between the diode circuit and the tube 59.

To complete the operational analysis set forth above, if the drift at that same instant of time had been in the other direction, then the phase of the drift signal would have been 180 different, as was explained in connection with Fig. 7. Then under these circumstances, when the point 77 was positive due to the oscillator signal, the induced in secondary 73b due to the drift signal would have been such as to tend to make the plate of diode 69 negative, and the plate of diode 70 positive. Diode 70 would then have drawn more current and diode 69 less current, with the result that lead 74 would have been positive with respect to lead 75, thereby just reversing the action. If the same values are taken as before, then lead 75 is +8 and lead 74 is +14 volts. The grid bias on tube 59 is reduced, reducing the quadrature component signal produced by tube 59.

An instrument may be provided, if desired, for indieating and/or recording the grid bias supplied through the leads 74, 75 to denote the amount of induction component that is compensated (which might be due to magnetic ground, drift due to thermal expansion or metal).

The above described system is useful for detecting nonmetallic mines in conductive ground. However, for the detection of metallic mines or other metallic objects, the oscillator 52 may provide a signal of lower frequency, for example, 1,000 cycles and modulation frequency may be substantially lower than that, for example, between 10 and 50 cycles.

It is believed evident from the foregoing how the status of any part or parts of any network may be maintained, and how an indication may be obtained of the changes that would have occurred had the automatic balancing features not been present.

While representative embodiments of the invention have been described in detail hereinabove, the invention is not intended to be restricted thereto, as those embodiments are susceptible of numerous changes in form and detail. For instance, the control tubes have been described as pentodes. Triodes or other tubes could be used just as well. Several tubes may be used in place of each one shown. Also control has been achieved by varying the D.C. bias on the control tubes, but the varying D.C. may be used to control the amplifier gain if desired. The means shown for impressing a signal on the circuits are conventional and any other suitable means may be used. For instance, transformers 55, 60, and 64 in Fig. 8 have been shown sepaartely for the sake of clarity. A common transformer may be used; in fact, no transformer is necessary as winding 50a may serve as an autotransformer. The frequencies cited are only typical; the higher frequencies work better when detecting nonmetallics, and lower frequencies are preferable when searching for metallic objects. Preferably a frequency is selected that will cause metallic mine signals to have both inductive and resistive components.

When it is desired to change the phase derived from a tube, it may be done by changing the phase of its excitation (control grid) or the phase shift in the coupling from its plate to the receiver circuit, as is well known in the electronics art.

Therefore, the above-described embodiments 'of the invention should be considered illustrative and not as limiting the scope of the following claims.

I claim: r l ln apparatus for investigating a signaloutput to determine its signal component content along 'a given phase axis, the combination of means for combining with said signal output a reference signalin phase quadrature to said given phase axis and phase modulated at a predetermined frequency, means producing an electrical signal representative of any modulation component of said predetermined frequency in said combined signal output and reference signal, and means responsive to said electrical signal for combining with said signal output an opposite polarity signal along said given phase axis to balance out said modulation component in the combined output and reference signal. 7

2. In apparatus for investigating a signal output that may have in phase and quadrature phase components with respect to'alternating current of a given phase, the combination of means for modifying said signal output by combining therewith reference signals having the same phases as said in phase and quadrature phase components, respectively, means for combining with said modified signal a modulated signal of the same frequency as one of said reference signals but of opposite phase, means for demodulating said combined modified signal and modulate'd signal to reproduce the modulations impressed upon said modulated signal whenever said signal output includesa component of the phase of said one reference signal, and means responsive to said reproduced modulation. a

3. In apparatus for investigating earth formations, the

icombination of means providing an alternating current signal representative of at least one property of the formations, means for combining with said alternating current signal a reference signal of given phase and phase modulated atiapredetermined frequency, means for producing an electrical signal representative of any modulation component of said predetermined frequency in said combined alternating current signal and reference signal, as a measure of signal component content along a phase axis in phase quadrature to said given phase in said alternating current signal, and means responsive to said electrical signal for combining with said alternating current signal an opposite polarity signal along said quadrature phase axis to balance outsaid modulation component in the combined alternating current signal and reference signal.

4. An automatic stabilizing system for four-terminal alternating current networks having an input circuit and an output circuit, comprising means for energizing said input circuit with alternating current at a first frequency, means forintroducing into said output circuit a first current component at said first frequency, having a predetermined phase with respect to said alternating current and a substantially constant amplitude, means for further introducing into said output circuit a second current component at said first frequency, said second current component being in phase opposition to said first component and having its amplitude modulated at a second frequency lower than said first frequency in order to provide a combined current component of said first frequency, means to detect in said combined current component the amplitude of the second frequency modulation and its phase with respect to the modulation of said second current component, and means responsive to said amplitude and phase of said detected second frequency modulation to adjust the amplitude of said first frequency component to maintain the amplitude of any signal of the same phase as said first frequency component in said output circuit substantially at zero. I a

5. An automatic stabilizing system for four-terminal alternating current networks having an input circuit and an output circuit, comprising means for energizing said input circuit with alternating currentiat a first frequency, means for introducing into said outputcircuit a first current component at said first frequency having 'a' predetermined phase with respect to said alternating current and a substantially constant amplitude, means for further introducing into said output circuit a second current component at said first frequency, said second current component being in phase opposition to said first component and substantially equal in amplitude to said first current component with its amplitude modulated at a second frequency lower than said first frequency in order to provide a combined current component of said first frequency, means to detect in said combined current component the amplitude of the second frequency modulation and its phase with respect to the modulation of said second current component, means responsive to said amplitude and phase of said detected second frequency modulation to maintain the total signal content of the phase of said first'frequency in said output circuit substantially constant, means for further introducing into said output circuit a third current component at said first frequency, said third current component being in phase quadrature to said first current component and of sufficient magnitude to form with the total signal content in said output circuit a resultant signal that is relatively unaffected by the total signal content of the phase of said first current component in said output circuit, and delayed action means responsive to said resultant signal for maintaining the amplitude thereof substantially constant.

6. An automatic stabilizing system for alternating current bridges, comprising a transmitter circuit having a first coil system, means for energizing said first coil system with alternating current at a first frequency, a receiver circuit having a second coil system disposed adjacent to said first coil system for inductive coupling therewith, means for indicating the output of said re ceiver circuit, means for introducing into said receiver circuit a first current component at said first frequency having a predetermined phase with respect to said alternating current and a substantially constant amplitude, means for introducing into said receiver circuit a second current component at said first frequency, said second component being in phase opposition to said first component and substantially equal in amplitude to said first current component with its amplitude modulated at a second frequency lower than said first frequency in order to provide a combined current component of said first frequency, means to detect in said combined current component the amplitude of the second frequency modulation and its phase with respect to the modulation of said second current component, means responsive to said amplitude and phase of said detected second frequency modulation to maintain the total signal content of the phase of said first frequency in said circuit substantially constant, means for further introducing into said receiver circuit a third current component at said first frequency, said third current component being in phase quadrature to said first current component and of sufficient magnitude to form with the total signal content in said output circuit a resultant signal that is relatively unaifected by the total signal content of the phase of said first current component in said output circuit, and delayed action means responsive to said resultant signal in said receiver circuit for maintaining the amplitude thereof substantially constant. I

automatic stabilizing system for alternating current n'etworks having an input circuit and an output circuit, comprising means for energizing said input circuit with alternating current at a first frequency, means including an electronic variometer tube having a gain variable in response to a variation of a direct current biasing potential applied thereto'for' introducing into said output circuit a first current component at said first frequency and of predetermined phase with respect to said alternating current, means for exciting said tube at said first frequency with the desired phase, means for introducing a second current component along the same 5 a a phase axis as said first current component but of opposite polarity and modulated at a second frequency lower than said first frequency to provide with said first current component a combined current component of said first frequency, means for detecting in said combined current component the amplitude of any modulation component of said second frequency and its phase with respect to the modulation of said second current component and means for converting said detected modulation component into a direct current and supplying the latter to said variometer tube for varying the bias thereon.

8. An automatic stabilizing system for alternating current networks having an input circuit and an output circuit, comprising means for energizing said input circuit with alternating current at a first frequency, means including an electronic variometer tube having a gain variable in response to variation of a direct current biasing potential applied thereto for introducing into said output circuit a first current component at said first frequency of a predetermined phase with respect to said alternating current, delayed action means responsive to variation in the output of said output circuit for varying the direct current bias on said variometer tube to restore the output to a substantially constant amplitude, means including a second electronic variometer tube having a gain variable in response to a variation of a direct current biasing potential applied thereto for introducing into said output circuit a second current component at said first frequency and in phase quadrature to said first current component, means for introducing a third current component opposite in phase to said second current component but modulated at a lower frequency than said first frequency to provide a combined current component of said first frequency, means for detecting in said combined current component the amplitude of said second frequency modulation and its phase with respect to the modulation of said third current component, and means for converting said detected second frequency modulation into a direct current and for supplying the latter to said second variometer tube for varying the bias thereon.

References Cited in the file of this patent UNITED STATES PATENTS 1,812,392 Zuschlag June 30, 1931 2,160,356 Fore et al May 20, 1939 2,225,653 Monk Dec. 24, 1940 2,250,571 Buschbeck July 29, 1941 2,377,500 Johnson June 5, 1945 2,413,396 Weagant Dec. 31, 1946 2,431,569 Labin Nov. 25, 1947 2,436,834 Stodola Mar. 2, 1948 2,551,348 Sunstein May 1, 1951

Images