US2418389A - Automatic limiter system - Google Patents

Description

April 1, 1947- 'GQ J. c. ANDRESEN $4 AUTOMATIC 'LIMITER SYSTEM Filed July 21, 1945 I 2 Sheets-Sheet 1 ATTORNEYS 1, CVANDRESEN AUTOMATIC LIMITER SYSTEM Filed July 21, 1945 2 Sheets-Sheet 2 6' 0/005 oklmva ill/0m 12 r: 10 0/005 INPUT Hui/m 12E lNPl/T v INVENT OR lynx/reach ATTORNEY5 PatentedApr. 1, 1947 UNITED STATES PATENT OFFICE 2,418,389 AUTOMATIC LIMITER sYs'rEM Gilbert J. C. Andrcsen, Cuyahoga Falls, Ohio Application July 21, 1945, Serial No. 606,415

12 Claims. 1 v

This invention relates to an automatic limiter system for effectively reducing static and noise and the like interference, in the reception of radio signals.

The principal object of the invention is to provide an automatic carrier controlled limiter with which discrimination may be attained, within reasonable limits, between signal and noise in periods of strong static, particularly during the reception of relatively weak signals.

The invention is particularly adapted for use in the reception of voice modulated carrier signals to improve the clarity of speech which otherwise would be distorted, or partially or wholly masked, by the interference.

In accordance with the invention, an improved system is provided comprising in combination several component electron tube circuits and arrangements, some of which are broadly old in the art, others specifically new.

A feature of the invention is a limiter whose cutoff levels are automatically controlled in proportion to the carrier amplitude of the signal, with provision for inverse coupling of strong static pulses in such a manner that their peaks are inverted and are limited at another level.

A preferred embodiment of the invention includes a pair of electron tube rectifiers which are adapted to supply half-wave rectification components of radio-frequency carrier and modulation to an electron-tube voltage regulator, which tube also acts as a reversing repeater for producing an opposite-polarity replica of said components, a supplemental electron tube in the voltage regulator for establishing a gate voltage determining the cutoff leads in a pair of series-limiter electron tubes, and means for coupling the final output to the audio system of a radio receiver.

A complete understanding of the improved system and the mode of operation of the several component electron tubes and circuits of the invention will be had by reference to the followingdescription and to the drawings to be considered in conjunction therewith.

In the drawings: Fig. 1 is'a circuit diagram illustrating the several component electron tubes and the circuit,

arrangements of a preferred embodiment of the improved system;

Figs. 2A-D show schematically certain current paths in relation to the rectifier; and

Fig. 3 shows certain characteristic curves relating to the regulator amplifier and the signal rectifier.

Referring now to Fig. 1, the invention is illusthe transformer secondary l0 and the tuning capacitor Illa, with its cathode ll connected to the point i2, and its grid l3, which serves as one diode anode, coupled to the point I through capacitance 22. The plate l5 of tube VIA, which serves as another diode anode, is conected to ground l6 through capacitance Ila and parallel resistance I1. The electron tube rectifer Via, a diode used to facilitate neutralization, is connected to the signal input circuit l0, Illa, with its anode it connected to the point l4 and its cathode is connected to ground i6.

The tube VIA supplies half-waves of radio-frequency carrier and modulation of the signal as a negative potential Eg, to the control element 20 of electron tube Vzfa regulator and radio-frequency limiter and amplifier which serves also as a reversing repeater. A supplemental regulator electron tube Va, operates in response to tube V: to establish a gate voltage determining the cutoff levels in the series limiter electron tubes V4 and V5. The output of the system may be applied to the audio system of the radio receiver.

Directing attention specifically to the tube VIA of Fig. l, the cathode-to-plate and cathode-togrid load resistors l1 and 24a. thereof provide leakage paths of high resistance for the signal and static voltages which are rectified in this tube. The rectified half-waves appear across these resistors and are preserved in some degree by making the shunt capacitance Ila and 20:: rather small, perhaps only the inherent capacitance of the circuit elements.

The instantaneous relationship of the various voltages and currents are afiected by the anode and grid resistance-capacitance (RC) time constants which, if greater than one-half the carrier period, accumulate substantial negative bias potentials on the diode anodes of tube Vm, of magnitude proportional to the amplitude of the volt-- age from the input circuit.

To obtain the fastest response in the rectification of the carrier-frequency voltage from the input circuit, the RC time constant of the load of equipped with shielded primary and 3 each diode is made less than the time interval or pulse width of the shortest pulse of noise, as determined by the preceding selective circuits in the receiver. The stray capacitance of the wiri'ng connections and the tubes, and additional capacitance, if present, will not seriously affect the performance, provided the time constant does not exceed the minimumpulse width.

The preferred RC time constant for one-half cycle would be RC= /2f, in which I is the frequency 'of the voltage applied to the rectifiers.

Referring again to Fig. 1, and in particular to the tube Vu. thereof, it is to be understood that the stray capacitance of wiring and the internal capacitance of the tube provide a radio-frequency coupling path across the rectifier, which is the principal path eilective during the non-conductive phase.

The diode rectifier tube Vl'B is connected in,

series with the ground return of the signal input circuit IO-and is poled the same as the diodes joined by tube VIA to conduct current on the same phase as said diodes. Thus, during the nonconductive phase, both the tubes VIA and V118 and their wiring connections still provide capacitive coupling to the output circuits of the diode rectifiers, namely, resistors I1 and 24a and the subsequent circuits coupled thereto.

The small capacitors 2|, 22, 23. are provided to neutralize or balance out this capacitive coupling to the output circuit. During the non-conductive phase, the capacitor 2| and the capacitance of tube vlB are connected in series (through ground leads) across the input circuit, and their junction is connected to ground. This forms a bridge circuit such that alternating voltages of opposite polarity appear at points 12 and 14 on opposite sides of the input circuit l0. By virtue of this bridge circuit, capacitor 22 neutralizes the cathode-to-grid capacitance of tube VIA, and 23 the cathode-to-plate capacitance of tube VIA.

Figs. 2A-D are simplified diagrams of those parts of Fig. 1 that are involved in the neutralization, and are shown to facilitate the explanation of this function.

Fig. 2A shows schematically the current path, and by arrows. the direction of electron flow, during the conductive phase of the rectifier, referring to either of the diodes in tube VIA. Figs. 2B-D show the capacitive current paths effective during the non-conductive phase of the rectifier and make clear the means of effecting complete or approximate balance of the capacitive coupling as above described.

In Fig. 2B the flow of capacitive current during the non-conductive phase is shown, the cathode-grid diode of tube VIA and the tube V13 restored to a low value.

stored in capacitor 26 is discharged at a rate acting as capacitive coupling means Cg and CB,

pling paths results in the required approximate neutralization of the tendency for radio-frequencycurrent to be coupled to the diode loads during thenon-conductive phase.

Referring again to Fig. 1, and in particular to the rectifier load 24a, there is sometimes developed at the point -E an undesired amount of negative potential caused by electron flow. To counteract that undesired condition, a divided load resistor 24a, 24b is provided and a small opposing current is supplied in resistor 24a through the resistor 25 from the power supply voltage source +E1.

If it is desired to provide the usual grid bias on the tube V2, a ca'thode resistor 60 may be employed as shown in Fig. 1. If a resistor 60 of proper value is used, resistors 24b and 2 5 may not be required. The negative feedback action of the voltage drop across resistor 60 serves to reduce dstortion and to permit a higher input voltage to be applied to the control element 26 before overloading occurs.

As pointed out above, the negative half-waves of radio-frequency voltage of the rectifier output are supplied to the control electrode 20 of the voltage regulator electron tube V2. Thereupon the output'current of this tube will decrease in proportion to the voltage from the rectifier.

' It is not necessary to control tube V: with the resultant sum of the half-waves. It may be preferable to actuate the control electrode 20 thereof by the voltage of each increment. In that case, the internal resistance of V2 should be restored quickly to its no-signal value between the half-wave conducting phases of the rectifier to assure the fast operation which is required to follow static pulses.

Assuming the presence of a strong pulse of static, the internal resistance of tube V: will become infinitely .high during the first negative half-wave of the series of half-waves. If the internal resistance is infinite, the capacitor 26 will charge to a higher potential through resistor 21 from voltage source +E1. The voltage of +E1 is assigned the least value which permits the regulator to function over a wide range of signal strengths; its small voltage prevents the static pulse amplitude from exceeding a limited amplitude preferably not much greater than that of the signal.

During a static pulse the capacitor 26 is momentarily charged at a more rapid rate, but the first rectified half-wave during the conductive phase is followed by the non-conductive phase of the rectifier and the internal resistance of V2 is Whereupon the energy several times the charging rate. That effects a rapid decay of the static charge on capacitor 26 during the non-conductive phase of the rectifier. T e signal half-wave increments which immediately follow the static are therefore more efiective in restoring the internal resistance of the tube V: to its normal value determined by the signal carrier intensity.

The regulator tube V2 may be considered as a resistance in a voltage-divider network including also resistors 21, 28, 29, 30. The voltage source E1 being constant, any change in voltage in the network is caused byvariation of the internal resistance of the tube V2 determined by the instantaneous voltage on the control electrode 20. As explained hereinbefore, in the absence of signals the preferred voltage on the control electrode 20 is zero. During the negative half-cycle of the signal, the control voltage is negative and of a value directly proportional to the amplitude of the radio-frequency voltage applied to the rectifiers. During the positive half-cycle of the siglinear relation with point may attain is limited by the voltage +E1' 01' the source.

If it is desired to obtain regulated voltage free of the signal half-cycle ripple,-a capacitive filter may be employed at any point along the voltage divider network. The capacitor 26 associated with the voltage-divider network is connected at the common terminal 01' the resistors 21, 28, 28, 30. Then the voltage E: at this junction will not fluctuate with the instantaneous value oi the signal and the static modulation, but will have an average value resulting from the operation just explained.

The voltage regulation at E: is'such that the efl'ects of static pulses are largely overcome, insofar as their peaks are limited by cutoff in tube V2, and the energy introduced in capacitor 26 by static pulses is reduced to a negligible amount.

The rectified voltage output+Es from the regulator tube V2 is limited by the voltage Es, since that part of E1 is held uniform during pulse modulation by capacitor 26.

The output of the hali-cycle radio-frequency voltage +Es includes the rectified modulation of the carrier, and when the halt-wave increments thereof are impressed upon an audio system the original modulations are reproduced.

The regulator tube V2 is preferably of the variable-ma type and is, therefore, adapted to operate with a high voltage input. However, a conventional triode tube, screen-grid tube, pentode or beam tube can be used. The pentode or beam tube, for example, would provide satisfactory limiter or cut-off characteristics it the input and element voltages are held within limited values. Ordinarily the output current 01' the tube Vs connected as a triode, and having a negative potential impressed upon its control electrode 20, would have a characteristic curve like A oi Fig. 3. The rectified output voltage of the rectifier is plotted along the abscissas to represent the negative, voltage g applied to the control electrode 20 of tube V2. The output current 19 of tube V2 is plotted along the ordinates.- It the rectifier characteristic is adjusted to be substantially as in curve B 01 Fig. 6, the resultant negative voltage impressed upon control element 20 of tube V2 will produce an output current characteristic substantially as in C, a substantially respect to the radio-frequency input voltage Eg. In other words, the curvature of rectification in tube VIA will compensate for the curvature of amplification in tube V2 to efiect a linear response to modulation of the input signal. Y

Electron tube Va, Fig. l, operates as a supplementary carrier-controlled voltage regulator and develops a differential voltage between the points E3 and E4 as a limit or gate voltage for the series-limiter electron tubes V4 and V5.

The grid voltage applied to grid 3| of tube V: is directly proportional to the voltage E: and to the signal carrier voltage. The voltage applied to the grid 3| oi the tube V: is adjusted 6 along the voltage-divider resistor lll until the plate current of the tube causes the voltage at.

age (Es-Ea) is in proper adjustment for the.

desired amplitude limiting as thesignal carrier intensity varies. The grid-cathode voltage also may be adjusted on which the cathode is tapped.

The cathode potential of tube V: is preferably to be adjusted by resistor l3 when the switch 35 shunts resistor 32. The cathode potential should be adjusted during absence or signal to make the input voltage to the grid 3| such that the voltage at point E4 is just sumcient to cut oil the series limiter tubes V4 and V5. The purpose of providing switch 35 is that with resistor 32 in circuit, the carrier signal strength must exceed a certain minimum before the control voltage applied toelement ll of tube V: will bring the limit voltage at point E4 to a value which will permitthe signal to pass through the limiters V4, Va

when no static or noise is present the voltages at points II and 42 are, permitted to vary up to the value of the voltage Ea before the point 42 becomes negativ with respect to point 4| and the limiter rectifier V4 opens up. Thereafter for all greater values 01' rectified and amplified voltage Es the limiter is non-conductive. With proper adjustment as described above, the signal carrier actuates the regulator .in such a manner that the limiter V4 stays closed for the normal range of modulation, but opens up on peaks of greater amplitude such as strong pulses of static.

While the clipping of static pulses is known to be a great aid in reception, this invention gains a major'advantage by a further action, namely, inversion of the strong pulses which otherwise would cause the loudest residual clicks. The inverson requires two coupling paths for rectified modulation from the detector to the limiter. The first is'a direct coupling path, whichpredominates in the amplitude range of normal modulation but which is cut oil at greater amplitudes. The second is an inverse coupling path which predominates at greater amplitudes and. causes inversion of the peaks of strong pulses.

In Fig. 1, the direct path is through the repeater V2, which conveys the desired modulation. The inverse path is through the capacitor 45 and resistor 44, which detour around the limiter V4 so this path is not subject to its clipping action.

Now considering the performance toward a signal accompanied by strong pulses of static, the pulse peaks coupled to point 42 for example, by

resistor 44 and capacitor 45 from Ep predominate at instants when the tube V4 is-cut oil. The

' static pulse voltage is weakly coupled to point 42 so that the anode voltage of V4 becomes less positive, cutting oil. further passage or the static pulse through V4. The value oi. resistor 44 should be such that strong static will not modulate the voltage at point 2 beyond the zero axis value of the signal while unmodulated. The capacitor should be of sufllcient value to prevent any substantial-phase shift in the passage of the pulse current from -Ep to point 42.

Asv insurance that the inverted pulse peak will not recede past the zero axis of the modulation, there is added a second electron tube limiter Vs coupled by capacitor TI to Vi and having a further by resistors 32, 22, 34,

ceived from electrode l3 of the detector.

7 The limiter Va is adapted separate gate voltage.

on the inverted peak of a .coupling of the inverse peak of the static pulse can-be increased to obtain a more precise limiting thereof, and that compensation may be effected during the static pulse for the biasing of points 4| and 42, respectively.

The inverse peak of a strong static pulse at point 42' is coupled'to point 43 making tube Va non-conductive beyond a certain point, for the reason thatthe time constant 48, 49, eflective at point 52 will not permit the voltage on the cathode of Vs to follow the sudden inverse peak at point 43. Before equilibrium can be established between point 62 endpoint 43, tube V4 has cut off in the manner explained hereinbefore. This mode of operation effects a very sharp cut-off of the static pulse because the inverse coupling may be increased to permit the effective use of rapid-charging of the small interelectrode capacitance, wiring connection capacitance, etc., of the limiter tubes V4 and V5.

During the .static pulse interval the limiter between points 43 and 52 prevents the inverted peaks from passing the zero axis value or the signal modulation. This is a desirable mode of operation because the inverted peak amplitude is within the range of normal modulation.

When the static pulse voltage has decreased from its peak to a value where the regulator V2 output voltage E. combined with the accumulated charge on capacitors 40 and 41 makes point 4! negative with respect to point 42, the limiterVl becomes conductive and V5 is already conductive so the normal response to modulation is resumed...

yin the absence of a static pulse the limitersVr and V5 remain conductive to signal amplitudes of the full range of the differential voltage of E: and E4. 7

The operation of the limiter as a; whole will be summarized briefly with reference to the essential functions of some parts of the circuit as described above. The principal parts are the signal detector including tubes VIA and Via, the reversing amplifier includinga tube V2, the inverse coupling path including resistor 44, the limiter including tubes V4 and V5, and the gatevoltage regulator including tubes V2 and V3.

The signal detector utilizes the rectifier tube vlli to deliver rectified signal voltages of negative polarity from electrodes l3 and I5, both acting as anodes. The other rectifier tube V n is needed only to preserve the potential at point I4 during the non-conductive phase of VM and thus permit the neutralizing capacity 22 to be effectlVe. i

The reversing amplifier utilizes the tube V: to reverse and amplify the rectified signal re- The amplified signal is coupled from the anode of tube V: through the capacitor 40 to the limiter. In the usual manner, the signal is reversed in this amplifier, so the amplified signal coupled throughsthe capacitor 40 may be regarded as positive, meaning that outward modulation of the signal causes a change of voltage in the positive direction. since this amplifier provides the principal coupling from the detector to the limiters. it is termed the direct coupling circuit in contradistinction to the inverse coupling circuit now to be identified.

The inverse coupling circuit utilizes resistor 44 to couple to the limiter therectified signal received from electrode Ill of the detector. This signal is negative in polarity, and is therefore inverse relative to the positive signal output from the amplifieia. The purpose of this coupling will be further summarized below.

' The limiter includes two rectifier tubes connected in series opposition for limiting the signal in both directions corresponding to outward and inward modulation. The first tube V4 is-poled to cut oil! beyond a certain level of outward modulation. The second tube V5 is poled oppositely to cut of! beyond a certain level of inward modulation.

When. a peak of the signal envelope, for example a static pulse, has a peak voltage exceeding the cutofl. level for outward modulation, as defined above, it is applied with positive polarity to point 4| and the cathode of limiter tube V4, and causes this tube to become non-conductive. Therefore the peak of this pulse is not coupled in this tube through the limiter: However, the inverse coupling path through resistor 44 is still operative on the peak of such a pulse and couples the peak with reverse polarity to the point 42 in the limiter, detouring around the tube V4 which is cut ofif. Therefore thepeak of the pulse appears inverted in the output of the limiter so should not indent the "signal too far. The second limiter tube Vs'achieves this purpose by lim- -iting any excessive peak values of inverse pulses. Furthermore, by relating the time constant of resistor 48 and capacitor 49, in the limiter output circuit, to the highest modulation frequency, the limitin'glevel for pulses of short duration of either signal or static can be reduced without detrimentto the desired modulation components of lower frequencies and maximum amplitude. The clipping of short'pulses 01' modulation at a level lower than low-frequency components of modulation is Permissible because the desired modulation seldom includes short pulses of strong modulation. l

The gate-voltage regulator circuit utilizes tubes V: and V3 to establish in the limiter certain cutoff levels automatically controlled in response to the carrier amplitude of the signal. These tubes are connected in the manner of a direct-current amplifier so that the rectified carrier voltage from electrode I3 01 the detector causes a proportional change of voltage E4 across the capacitor 58. The latter voltage is applied to the limiter tubes V4 and V5, in conjunction with the voltage Es across capacitor 5|. This voltage E: is obtained primarily from the constant source E1 and in- .creases somewhat, under the influence of tube V2, with increasing carrier amplitude. The dif- -ference of these two voltages determines the cut- I the output of the limiter.

' system in conjunction with Vhsingle' section auasao when there isno carrier voltage. The result of the gate-voltage regulator is to insure that the beneficial action of the limiter is secured automatically for any signal strength within the useful range of the receiver.

As mentioned above, the re .t'ified half-waves of the signal are present in the output of the detector VIA and also in'the output of the tube V2. By the usual properties of rectification. the voltage half-waves carry an alternating-voltage component of modulation frequencies and a direct-voltage component of carrier amplitude. For purposes of regulation, the latter is obtained acre-:s capacitor 20 and finally corresponding direct potentials across capacitors 5| and 58, all of these capacitors being so large that they cannot becharged and discharged at pulse modulation frequencies.

In the foregoing description, it is presumed that a strong pulse of static superimposed on a modulated carrier has all the appearances of a pulse of modulation, that is, an excessive peak in the modulation envelope. A static pulse is distinguished only if its peak amplitude exceeds the modulation peaks of the signal. the present invention minimizes its fect by reducing its peak Value 88 disturbing efit remains in By adjusting automatically the gate voltage of the limiter in proportion to the carrier amplitude, the limiter is enabled to identify static peaks Just above the signal peaks, regardless of what that level happens to be for a signal of any given strength.

The preferred embodiment of the invention has been successfully employed as an automatic carrier-radio-frequency controlled limiter in con junction with commercial receivers, RCA type CRV-46151, among others, voice modulated carrier signals. The following tabulation of circuit constant values and types of electron tubes identifies those employed in the the RCA receiver type CRV-46151 for the successful reception of voice modulated signals through both atmospheric static and locally produced noises. It is as follows:

Via-Single section of RCA 65117.

Via-Diode rectifier-may be contained in the V2 envelope such as an RCA 6825'].

of an RCA GS L'I.

V4, Va-RCA 6H6.

24a,'b..Section a, 100 K ohms; section b, 500 to 1000 ohms. a I

32-0 to i K ohms.

20, "-100 K ohms.

04-150 K ohms.

25, 21-250 K ohms.

50, 40, lit-500K ohms.

".55, 40, 50-1 mesohm.

51-1-10 megohms (high values preferred).

' slices), 2: (on), 2: waimwmnzm capacitors 3-10 a d.

for the reception of In that case,

' l0 lla (Op), Ci-Load capacitances usually inherent strays-approximately 10 mil. 5|, l9.001 #1.

45-001 to .01 at. a

- 26, i8.001 to .05 Id.

41-.01 pf. 40, 59-.05 at. 80-0-5000 ohms.

prior art should be considered inrelation to the appended claims.

I claim:

1. A signal coupling and amplitude limiting system comprising, an input circuit including a rectifier providing rectified voltage in response to a modulated carrier signal, an output circuit, direct and inverse circuits coupling said input and output circuits with opposite polarities, a rectifier connected in said direct circuit, and voltage regulating means including integrating means responsive to the carrier amplitude of said niL clulated-carrier signal for applying to. said rectifier in said direct circuit a gate voltage to limit the peaks of amplitude in said direct circuit at a cutof! level proportional to the carrier amplitude, said direct and inverse circuits each providing substantial coupling and being so proportioned that the direct circuit provides the greater. coupling for. amplitudes below said cut-oil? level. whereby the peak of any pulse exceeding said cuton level is inverted.

2. A system in accordance with claim 1 including a rectifier connected in the inverse coupling circuit and means applying a gate voltage to said rectifier causing said rectifier to limit the peaks of amplitude in the inverse circuit at a cut-off level determined by the gat voltage on said rectifier, said direct and inverse circuits being so proportioned-that the inverse cut-oiif level is beyond the direct cut-oil level and the direct circuit provides the greater coupling for amplitudes below itscut-off level, whereby the peak of any pulse exceeding the direct cut-off level is inverted to the extent of the limit imposed by the inverse cut-ofi level.

3. A'system inv which the means for applying gate voltage to the rectifier in the inverse circuit is a voltage regulating means including integrating means responsive to the carrier amplitude of said modulated carrier signal whereby the peaks of amplitude in the inverse circuit are cut oil at a level proportional to the carrier amplitude.

4. A system in accordance with claim 3 in which each of the rectifiers is poled to limit outaccordance with claim in l in the inverse circuit and inward modulation in the direct circuit.

6. A system in accordance with claim 5 in which the rectifier which provides rectified voltage in response'to the modulated carrier signal provides rectified voltage of one polarity and in which-one of the coupling circuits includes a reversing repeater providing voltage of opposite polarity. a

7. A system in accordance with claim 1 in to limit outward modulation.

8, A system in accordance with claim 7 in,

which the rectifier which provides rectified voltage in response to the modulated carrier signal provides rectified voltage of one polarity and in which one or the coupling circuits include a reversing repeater providing voltage of opposite polarity.

9. A system in accordance with claim 1 in which the rectifier which provides rectified voltage in response to the modulated carrier signal provides a rectified voltage of one polarity and in which one of said coupling circuits include a reversing repeater providing an opposite polarity of voltage.

10. A signal coupling and amplitude limiting system comprising, an input circuit including a rectifier providing rectified voltage in response to a modulated carrier signal, an output circuit, di-

rect and inverse circuits coupling said input and output circuits with opposite polarities, a rectifier connected in each of said coupling circuits and poled to limit outward modulation in its respective circuit, voltage regulating means including integrating means responsive to the carrier amplitude of said modulated-carrier signal for applying to the rectifier in one of said coupling circuits a gate voltage to limit the peaks of amplitude in the circuitof said rectifier at a cutoi! level proportional to the carrier amplitude, means for applying a gate voltage to the rectifier in the other coupling circuit causing said rectifier to limit the peaks of amplitude in the said circuit at a cut-oil level determined by the gate voltage on said rectifier, said direct and inverse circuits each providing substantial coupling and being so proportioned that the inverse cutoil level is beyond the direct cut-oi! level and the direct circuit provides the greater coupling for amplitudes below its cut-oft level, whereby Number the peak of any pulse exceeding the direct cutoil? level is inverted to the extent otthe limit imposed by the inverse cut-ofl level.

I 11. A system in accordance with claim 10 in which the rectifier which provides rectified voltage in response to the modulated carrier signal provides rectified voltage of one polarity and in" which one oi. the coupling circuits includes a reversing repeater providing voltage of opposite polarity.

12. A signal coupling and amplitude limiting system comprising, an input circuit including a rectifier'providing rectified voltage of one polarity in response to a modulated carrier signal, an output circuit, direct and inverse circuits coupling said input and output circuits'with opposite polarities, a reversing repeater in one of said coupling circuits providing the voltage of opposite polarity for said circuit, a rectifier connected in each oi! said coupling circuits, voltage regulating means including integrating means responsive to the carrier amplitude of said modulated-carrier signal for applying to the rectifier in one of said coupling circuits a gate voltage to limit the peaks of amplitude in the circuitof said rectifier at a cut-oi! level proportional to the carrier amplitude, means for applying a gate voltage to the rectifier in the other coupling circuit causing said rectifier'to limit the peaks of amplitude in the said circuit at a cut-01f level determined by the gate voltage on said rectifier, said direct and inverse circults each providing substantial coupling and being so proportioned that the inverse cut-oil level is beyond the direct cut-off level and the direct circuit provides the greater coupling for amplitudes below its cut-off level, whereby the peak of any pulse exceeding the direct cut-ofi level is inverted to the extent of the limit imposed by the inverse cut-oil level. I

' GILBERT J. C. ANDRESEN.

REFERENCES CITED The following references are of record in the file of this patent:

UNITED STATES PATENTS Name Date Starrett i Dec. 10, 1935 Van Loon i Dec. 2'7, 1938 Anderson June 18, 1940 Dome July 25, 1944 FOREIGN PATENTS Country Date Italy Mar. 1'7, 1938 Number

Images