GB1585321A - Brake control systems for vehicles - Google Patents

Description

(54) IMPROVEMENTS IN OR RELATING TO BRAKE CONTROL SYSTEMS FOR VEHICLES (71) We, THE BOEING COMPANY, a Company organised and existing under the Laws of the State of Delaware, United States of America, of P.O. Box 3707, Seattle, Washington 98124, United States of America, do hereby declare the invention, for which we pray that a patent may be granted to us, and the method by which it is to be performed, to be particularly described in and by the following statement:- This invention generally relates to brake control systems for vehicles and, more particularly, to limited-slip brake control systems, that is, those brake control systems which seek to limit the brake force developed by a braked vehicular wheel so that the brake force remains on the positive slope side of a characteristic brake force-percent slip curve for the vehicle.

For a better understanding of the limitedslip brake control system of this invention, it is important to understand the distinctions between it and a known conventional antiskid braking system with and without automatic braking, such as described for example in U.S.

Patent No. 3,669,508, issued June 13, 1972, entitled "Wheelspeed Information Signal Processing System." In the known conventional antiskid braking systems without automatic braking, the rate of deceleration is controlled by the pilot; and when the pilot meters full brake pressure, the antiskid system operates to maximize stopping capability. However, under some circumstances, the tires will be forced to skid; and when they do, the antiskid system then functions to quickly relieve the skidding by decreasing the brake pressure. With this type of system, it is difficult, if not impossible, to achieve a constant rate of deceleration during the stopping operation; and the percent-slip inherently exceeds the desired range of values in which the tire wear is minimized and the steering control optimized.

In the known conventional antiskid braking systems with automatic braking, the rate of deceleration is held constant by the system; and it is not necessary that the pilot use the brake pedals except to override the automatic braking. The antiskid portion of the system functions under certain operating conditions; particularly, when momentary wheel lockups occur.

In the limited-slip brake control system of this invention, the selected rate of airplane deceleration is held relatively constant and the braking system is actuated and operated by the pilot using the brake pedals. This system functions to prevent skids from occurring, as opposed to the known conventional antiskid braking systems which function to relieve skids as rapidly as possible. In order for the system of this invention to function properly, it will be necessary to predetermine the percent-slip range within which the wheels are to operate so that adequate stopping capability is achieved without the wheels constantly going into and being rescued from skids.This can be done by analysis, design, and testing of the type of tires to be used and should also take into account the physical factors of the tires; such as, tire size, rolling radius variation during braking, tire to ground friction factors, and further include the aircraft characteristics.

With these results, the full range of values for the percent-slip range of that particular aircraft can be determined and the system can then be made to operate in that particular range.

The limited-slip brake control system functions to limit the brake force developed by the tires of a vehicle to the nonslip portion of a tire brake force and wheel velocity characteristic curve, i.e., the front or positive slope side of the characteristic Mu/slip curve for the vehicle, to thereby minimize both tire and brake wear. The system includes a basic deceleration control circuit; and a large deviation control circuit.

The basic deceleration control circuit produces a wheel deceleration signal by filtering and differentiating a signal representing measured wheel speed. The wheel deceleration signal is then compared with a selected refer ence deceleration signal representing a reference or desired deceleration and a first deceleration error signal is generated which is applied to a hysteresis circuit which outputs a constant level, positive or negative signal.

The output signal from the hysteresis circuit is integrated and supplied to a valve driver which in turn controls a brake valve to modulate the metered brake pressure applied to the wheel to be braked. The reference deceleration represented by the selected reference deceleration signal is chosen so that brake force developed as a result of the application of brake pressure is limited to the positive slope or nonslip portion of a characteristic Mu/slip curve for the vehicle, and the basic deceleration control circuit functions to cycle brake pressure, and therefore brake force, about a value that will produce the desired deceleration.

A situation may be encountered, however, where the coefficient of friction between the braked wheel and the ground surface is abruptly lowered, e.g., where a patch of ice exists on a runway. If the reference deceleration has been set at a level corresponding to the brake force that can be developed for a ground surface having a higher coefficient of friction, e.g., a dry runaway, then the basic deceleration control circuit will command a brake pressure that will seek to develop a higher brake force than can be obtained when the lower coefficient of friction surface condition is encountered, thus resulting in an actual brake force which lies on the negative slope or slip portion of the Mu/slip curve and which causes the braked wheel to skid.The large deviation control circuit accordingly provides a second deceleration error signal when the wheel deceleration exceeds the reference deceleration by a predetermined amount.

This second deceleration error signal is supplied to a lag circuit which is controlled to obtain an output signal which, when supplied to the valve driver, results in brake pressure being removed from the braked wheel in an amount and for a time related to the amount and time that the wheel deceleration exceeds the reference deceleration. Typically, the output signal from the lag circuit in the large deviation control circuit, and the output signal from the hysteresis circuit in the basic deceleration control circuit, are summed at the input to the valve driver so that the large deviation control circuit may control brake pressure when a skid is encountered and so that the basic deceleration control circuit may control brake pressure at all other times.

In order that all brakes of a multiwheeled vehicle perform properly with even wear of the tires of the braked wheels, an energy balance system may be used for supplying the same brake pressure to all brakes, such as by using a common brake valve. Alternatively, the brake energy being put into each wheel may be determined by measuring, for each wheel, the actual brake torque and wheel speed, by multiplying the measured brake torque and wheel speed, and by intergrating the resultant product. The measured brake energies are then compared, and any difference between brake energies is integrated and used to reduce, through a valve driver for each wheel, the brake pressure applied to the wheels that are working more and to increase the brake pressure applied to the wheels that the wheels that are working less.

An object of this invention is to control braking, i.e., to keep the braked wheel speed decrement from the free rolling wheel speed decrement, in the range in which the wheel decrement is produced by tire deformation and out of the range in which tire slippage on the ground contributes to the wheel decrement; thereby providing improved steering capacility and reduced tire wear.

Another object is to use a deceleration rate signal input as the primary control of the braking system.

Another object is to control the amount of wheel braking so as to produce a rate of deceleration which matches the manually or automatically set value.

Another object is to limit the operation of the braked wheels to the front or positiveslope side of the brake force to wheel velocity curve, or the characteristic tire Mu/slip curve.

The limited-slip brake control system of this invention has several advantages over the current braking systems which incorporate automatic braking. These advantages are possible mainly because this system has been devised from the beginning to do everything that a conventional system would do, plus operate mostly in the beneficial slip range, whereas the conventional system evolved as a series of add-ons to the basic hydraulic braking system.

An advantage of this invention is that it provides an optimum passenger comfort level, relative to the average rate of deceleration, by the gradual onset and release of brake force and by maintaining a constant and smoothly modulated deceleration.

Another advantage is that it relieves the pilot's workload and skill requirements, particularly combined steering and braking, by optimizing ground control qualities in terms of achieving as much wheel braking as possible under existing conditions, without jeopardizing directional steering control Another advantage is that the system automatically adjusts the upper limit of angular velocity difference between the braked wheel and the free-rolling wheel, so as to operate the tire where there is no actual slippage between the tire and the ground.

Another advantage is that the pilot can take full control of the brakes without shutting off the automatic braking system.

Another advantage is the improved operational economies of the aircraft in terms of reduced tire wear, reduced brake wear, and a more simplified and reliable braking system.

The present invention also provides a limited-slip brake control system which incorporates certain improvements over that previously described.

As a first example, the basic deceleration control circuit and the large deviation control circuit previously described each function to integrate a deceleration error signal obtained from a comparison of wheel deceleration with a reference deceleration. Since the output signals obtained from each integration in the large deviation control circuit and the basic deceleration circuit are summed before application to the valve driver, the integration functions provided in those circuits tend to oppose each other so that a resultant control signal applied to the valve driver does not precisely respond to desired changes in brake pressure commanded by either the large deviation control circuit or the basic deceleration control circuit.

It is therefore an additional object of the present invention to provide a simpler system than that previously described, which simpler system includes both a large deviation control circuit and a basic deceleration control circuit, and which system permits both of these circuits to more precisely control brake pressure than is possible with the limited-slip brake control system previously described.

- As a second example, the limited-slip brake control svstem nreviously described is not capable of satisfactorily compensating for a ground surface condition having a low co efficient of friction that is encountered imme diately upon application of brake pressure. In particular, the limited-slip brake control system previously described permits brake pressure to build up to the maximum value established bv the basic deceleration control circuit upon initial annlication of brake pres sure. If a low coefficient of friction ground surface condition is encountered upon this initial application of brake pressure, the basic deceleration control circuit forces the vehicle into an initial skid, as previously described, which skid is eventually compensated for by the large deviation control circuit.It is desirable in certain circumstances to minimize the effect of this initial skid, particularly in the case where the vehicle is an aircraft and the initial skid occurs immediately upon touch down.

It is therefore another additional object of this invention to provide an improved limited slip brake control system which very quickly reduces brake pressure upon the occurrence of an initial skid.

As a third example, the braked wheels of an aircraft are supported from the aircraft fuselage by a lightly-damped landing gear strut assembly. Application of brake pressure by the limited-slip brake control system previously described results in fore-and-aft oscillation of the landing gear strut assembly as the aircraft touches down and proceeds down the runway. This fore-and-aft oscillation produces an apparent variation in the velocity of the braked wheel which is sensed by the limited-slip brake control system as an apparent variation in wheel deceleration. As a result, the limited-slip brake control system functions to erroneously vary brake pressure about the value otherwise commanded by the basic deceleration control circuit.

It is therefore a further additional object of this invention to provide an improved limited-slip brake control system which is relatively insensitive to apparent changes in wheel deceleration occasioned by landing gear strut assembly oscillation.

These additional objects are achieved, briefly, by a limited-slip brake control system which controls the brake force to be developed by a braking means for a wheel of a vehicle.

The limited-slip brake control system of the present invention includes a first means for comparing measured deceleration of the wheel with a selected reference deceleration representing a level of brake force generally lying in the nonslip portion of a characteristic Mu/ slip curve for the vehicle, and operative to provide a fixed-level, first output signal gener ally having a first polarity representing a decrease in brake force when the measured wheel deceleration exceeds the selected refer ence deceleration and generally having a second polarity representing an increase in brake force when the selected reference deceleration ex ceeds a predetermined amount representing an incremental deceleration less than the selected reference deceleration.A second means compares the measured wheel deceleration with a second reference deceleration representing a level of brake force generally lying in a slip portion of the characteristic Mu/slip curve and is operative to provide a second output signal when the measured wheel deceleration exceeds the second reference deceleration, the second output signal having the first polarity representing a decrease in brake force and having a level proportional to the amount by which the measured wheel deceleration exceeds the second reference deceleration. A third means additively combines the first and second output signals to obtain a summed output signal. A fourth means integrates the summed output signal to obtain a brake control signal representing a desired level of brake force, the brake control signal increasing and decreasing at a rate depending upon the level of the summed output signal.The brake control signal is applied to a fifth means which controls brake force in inverse proportion to the level of the braking control signal.

In order to quickly correct for deep skids encountered during the application of brake pressure, a sixth means is provided for comparing the measured wheel deceleration with a third reference deceleration representing a level of brake force lying in the slip portion of the characteristic Mu/slip curve, the third reference deceleration being higher than the second reference deceleration, and operative to provide a second brake control signal when the measured deceleration exceeds the third reference deceleration, the second brake control signal having the first polarity representing a decrease in brake force and having a level proportional to the amount by which the measured wheel deceleration exceeds the third reference deceleration.A gating means is provided for coupling to the fifth means that one of the brake control or second brake control signals whose level represents a greater reduction in brake force.

An initial skid circuit may be provided for minimizing the extent of an initial skid encountered upon the initial application of brake force, the initial skid circuit being coupled with the second means and operative to produce a fixed-level, third output signal having the first polarity representing a decrease in brake force, but only in response to the first time that the wheel deceleration exceeds the second reference deceleration, the third output signal having a duration substantially equal to the expected duration of the initial skid.

The third means correspondingly provides an additive combination of the first, second, and third output signals to obtain the summed output signal.

Alternatively, the initial skid circuit may coupled with the second means and operative to increase the level of the second output signal at the first time that the wheel deceleration exceeds the second reference deceleration and for a period of time thereafter substantially equal to the expected duration of the initial skid.

In order to compensate for undesirable lowfrequency oscillation in the measured wheel deceleration which typically occurs when brake force is developed by a wheel supported from an aircraft by a lightly-damped landing gear strut assembly, a dynamic compensation circuit may be provided which is responsive to the measured wheel deceleration for providing a compensating control signal having the first polarity representing a decrease in braking pressure, a 90" or greater phase lead for a range of frequencies in the measured wheel deceleration surrounding th e expected frequency of the undesirable low-frequency oscillation, and a lesser phase lead for all other frequencies in the measured wheel deceleration.Means are provided for additively combining the compensating brake control signal with the brake control signal to compensate the brake control signal for oscillation therein resulting from the undesired low-frequency oscillation in the measured wheel deceleration.

The invention can perhaps best be understood by reference to the following portion of the specification, taken in conjunction with the accompanying drawings in which: FIGURE 1 is a characteristic tire coefficient of friction graph depicting by three curves the relationship of the brake force, the tire side force and the tire wear to the percentage of slip of the tire; FIGURE 2 is a simplified illustration in a functional block diagram of the limitedslip brake control system of the invention; FIGURE 3 is a detailed illustration in functional block diagram of the preferred embodiment of the invention depicting three control circuits, i.e., a basic deceleration control, a large deviation control, and an energy balance control; FIGURE 3A comprises three graphs depicting the function of the selected deceleration control circuit shown in FIGURE 3;; FIGURE 4 shows a triangular wave form produced from a plot of the time history of the hysteresis circuit shown in FIGURE 3; FIGURE 5 is another embodiment of the basic deceleration control circuit shown in FIGURE 3, wherein a threshold and singleshot multivibrator circuit is used instead of the hysteresis circuit; FIGURE SA shows a triangular wave form produced from a plot of the time history of the circuit in FIGURE 5; FIGURE 6 shows in functional block diagram a brake energy balance control system; FIGURE 7 is a functional block diagram of the improved limited-slip brake control system of the present invention including: a basic deceleration control circuit comprising a hysteresis circuit; a large deviation sensing circuit; an initial skid circuit; an integrating circuit responsive to output signals from the hysteresis circuit, the large deviation sensing circuit, and the initial skid circuit; and, a dynamic compensation circuit; FIGURE 8 is a combined schematic and block diagram of a preferred embodiment of a pulse generator forming part of the initial skid circuit; and FIGURE 9 is a functional block diagram of an alternate embodiment of the initial skid circuit.

FIGURE 1 is a graph showing the relationship between three curves: a brake force curve 11, in solid line; a tire wear curve 12, in dash-dot line; and a tire cornering or side force curve 13, in dashed line. These three curves are shown plotted on a characteristic graph of the coefficient of friction (Mu, or Wu) between a tire and the ground, relative to the function of percent-slip, or as generally referred to as the Mu/slip curve. The maximum coefficients of friction are achieved (depending on many factors) when the apparent braked wheel speed is a value less than the free rolling speed. The decrease in braked wheel speed is the result of a rolling radius increase caused by the dynamic deformation of the tire, rather than by the slippage between the tire and the ground. The term "percent-slip" is not rigorously accurate.

However, it is used because of the difficulty in measuring therelative velocity between the tire footprint and the ground; e.g., if the number of tire revolutions down the runway are multiplied by the deflected radius of the tire, this calculated distance will be somewhat less than the actual distance traveled down the runway.

Therefore, a distinction is made between the calculated or measured slip and the actual slippage between the tire footprint and the ground. In the percent-slip equation, shown in FIGURE 1, the rotational speed * is calculated for a given radius tire that will equal the actual distance traveled; and from this, the actual measured rotational speed of the tire is subtracted, and this difference is then normalized or put into percent by dividing the calculated rotational speed w.

and multiplying by 100. The zero percentslip value, when the calculations are based on measurements from a wheel speed transducer, occurs at the point of zero braking, as shown in the graph of FIGURE 1. There is in fact no actual slippage of the tire footprint relative to the bearing surface of the runway in the percent-slip range, under the positive slope or front-side portion of the brake force curve 11 or Mu/slip curve, or that portion thereof which is shown to the left of the vertical dashdot-dot line 14.Actual tire slippage does not occur at the initial application of the brake force, but would begin somewhere around the peak of the Mu/slip curve, or near the vertical dash-dot-dot line 14, the basic premise being that for percent-slip below this level, there is no actual tire to ground slippage or tire footprint contact area slippage; however, above this level, the tire does begin to slide relative to the ground. Also, the frictional brake force or Mu is increaseed, up to a point where the tire starts to slide, and then it begins to decrease as the percent-slip is increased.

With respect to the tire cornering or sideforce curve 13: the maximum tire side-force is obtained when there is no brake force being applied and there is zero percent tire slip.

The tire side-force decreases fairly rapidly with the application of brake force. After the peak in the brake force curve is reached, or the tire starts slipping, the side-force curve 13 flattens out to a very low value.

With respect to the tire wear curve 12: initially, the tire wear remains at a fairly low value until the peak of the brake force 11 is reached and the tire starts slipping, then the tire wear increases very rapidly.

One of the objects of a limited-slip brake control system is to provide an essentially better compromise on these three aspects: tire wear, tire cornering or side-force capability, and brake force; by keeping the operation of the braking system on the front side of the brake force curve 11, so that a fair amount of side-force capability remains with very low tire wear.

The approach used to accomplish these objectives is to limit the rate of deceleration of the aircraft to such a level that the tires are operating on the front side of the brake force or Mu/slip curve 11, or to the left of the vertical dash-dot-dot line 14. This is in essence limiting the tire brake force to a level that can be generated without slippage between the ground and the tire.

In general, the well-known detailed apparatus of a braking system (not shown) would comprise: pilot operated brake pedals; metering valves; a control system; control valves; brake system hydraulics; and brakes. In the present system, the pilot would actuate a deceleration rate selector switch and apply pressure to the brake pedals, controlling the metering valves that supply pressure to the brakes through the control valves; and the control valves would then, through the present system, modulate the metered pressure to keep the operation of the tires on the front side of the Mu/slip curve.

In a further modified configuration of the present system, incorporating the fundamental braking apparatus, the actuation of the brake pedals would produce an electrical signal which would control the operational range of the system, rather than the metered brake pressure. The operational range would still be limited to the front side of the Mu/ slip curve 11 as in the previously described configuration; however, no metering valves would be required. Differential braking for steering with brake forces would be achieved by making the appropriate pedals control the range of the wheels on respective sides of the aircraft.

On the backside or negative slope side of the Mu/slip curve 11, or to the right of the vertical dash-dot-dot line 14, slipoage between the tire footprint and the ground does occur.

This slippage or skidding results in a rapid increase in tire wear rate as shown by the positive sloping line 12, and a deterioration in the tire's ability to react to a side load as shown by the negative sloning line 13. The generally known type of antiskid systems continually force the time into operation on the backside of the Mu/slip curve 11, or to the right of the vertical line 14, in their continual functional determination to locate the peak of the Mll/sliD curve 11.

FIGURE 2 is a simplified illustration in block diagram of a limited-slip brake control system wherein a function of aircraft deceleration is measured and compared to a selected value of that function, which will maintain the tire brake force on the front side or positive sloping side of the Mu/slip curve shown in FIGURE 1. In general, the system comprises a function of aircraft deceleration, F(X), which is measured and compared to a selected value of that function; and from a comparison of the function of aircraft deceleration, with the desired value of that function, an error signal is generated and fed to the controller. The controller then produces a brake valve current in response to the information contained in the error signal, and this brake valve current then produces the brake force in the wheel brake system.The wheel brake system then produces the brake torque and ground force that results in the deceleration of the aircraft and the function of aircraft deceleration. A function of aircraft deceleration is indicated since the actual deceleration of the aircraft may not be the most convenient value to measure and use.

Another function that can be used in place of the function of aircraft deceleration is the function of wheel deceleration. The wheel deceleration for a nonslipping tire is related to the deceleration of the aircraft by the rolling radius of the tire; whereas, the wheel deceleration for a slipping tire is not related to the deceleration of the aircraft. This discrepancy can often be used to the system's advantage. Wheel deceleration has the advantage that it can be easily derived from wheel speed information, and it also indicates horizontal deceleration. In contrast, direct aircraft deceleration measurement produces a deceleration component referenced to the aircraft which therefore varies with aircraft attitude.

FIGURE 3 is a more detailed block diagram of the limited-slip brake control system of this invention and depicts all three circuit aspects, i.e., basic deceleration control loop circuit 15, large deviation control loop circuit 16, and energy balance control which is provided by supplying the same deceleration control signal to all of the brakes.

FIGURE 3, as well as the remaining FIGURES 4 to 9, will be described in combination with the presentation of this invention through these three distinct system aspects.

In general, the basic deceleration control 15 is a primary circuit control loop which maintains the deceleration of the aircraft at the required level; the large deviation control 16 is a backup circuit control loop which provides for wheel recovery, should the wheel be forced into a slip condition exceeding the desired limits, e.g., a wheel suddenly entering an ice path resulting in a sudden drop of the available frictional coefficient or Mu; and the energy balance control is a mean of assuring that all of the brakes are doing their share of the work, since the system which is operating on a deceleration basis cannot determine the force causing the deceleration.

More specifically, the basic deceleration control, which is shown within the dashed outline 15 in FIGURE 3, utilizes "across the aircraft pairing" of the deceleration control, and this provides a passive energy balance by supplying an essential equal brake valve current to each brake system. The desired rate of deceleration comes from the deceleration rate selector switch 18, which has three selectable values: Dry, Wet and Icy. It will be understood, of course, that this could be a multiposition switch or an infinitely variable one.The operation of the braking system on the front side of the Mu/slip curve 11 (shown in FIGURE 1) is effected by the pilot's selection of a deceleration rate value that is low enough to prohibit developing an increase in the coefficient of friction that would drive the system over the critical peak Mu for the existing ground conditions, and by the large deviation control circuitry 16, as described infra, which senses deceleration rates greater than that selected and quickly reduces brake pressure.

Since the location of the coefficient of friction peak changes with runway surface conditions, the deceleration rate selector switch 18 is included to allow the pilot to optimize the operation of the braking system to the prevailing runway surface conditions.

It is conceivable that in a more sophisticated braking system the selector switch 18 could be replaced by an electrical circuit that would sample the surface frictional conditions of the runway and then automatically set the operational range of the deceleration rate.

The frictional characteristics of the tires on an aircraft are such that each has a certain maximum available frictional force, depending upon runway surface conditions, etc., and if all of the available ground friction force is used for braking operation, there will be none left for the side frictional loads on the tire to control steering. This is one of the reasons why the known antiskid braking svstems, which function at the crest of the Mu/slip curve 11 shown in FIGURE 1, permit the aircraft to drift off the runway when braking in a crosswind condition. The peak of the Mu/slip curve 11 or the maximum obtainable rate of deceleration under ideal dry runway surface conditions, would be approximately 12 to 13 ft/sec2 (feet per second squared). However, in order to have a sufficient margin of frictional force in reserve for tire side loads, yaw correction, steering loads, and for passenger comfort, the positioning of the selector switch 18 to the Dry mark would produce a deceleration rate of approximately 10 ft/sec2; the Wet position would produce a deceleration rate of approximately 7 to 8 ft/sec2; and the Icy position would produce a deceleration rate of approximately 5 to 6 ft/sec2.

In all of the figures, the LaPlacian operator s is used to denote d dt dt Thus a block with an enclosed s indicates differentiation, and s indicates integration. A first order time constant is indicated by 1 ras + 1 In the operation of the braking system shown in FIGURE 3, the pilot positions the deceleration rate selector switch 18 to one of the settings Dry, Wet, or Icy, and this provides a pilot selected deceleration signal on line 19 to both the large deviation control loop 16, shown in dash-dot outline, and to the basic deceleration control loop 15, shown in dashed outline.

The measured wheel rotational velocity signal 01 is generally derived from the use of an Ac wheel speed transducer which puts out a sine wave that is squared, and then filtered or rectified to produce a DC voltage that is proportional to the speed of the wheel. The wheel speed signal l enters a differentator 21. In the differentiator block 21, the letter s designates a differentiation, i.e., a derivative with respect to time. For a low frequency sinusoidal signal input wi, the signal on line 22 will be the differentation of the sinusoidal signal such that with a sine wave signal on line 22 will be a cosine wave. The expression s rs + I represents a differentation with a lag circuit wherein the expression T S + I stops or cuts off the differentation after a certain frequency.This would be equivalent to taking any wave form on the wheel signal w,, and then converting it to a Fourier series. At the higher frequencies, this term r s + 1 functions as a filter to attenuate the higher frequencies so that if there is the presence of noise, like radio frequency interference, this term will function to filter or attenuate it so that it will not feed through the system.

From the differentiator block 21, the wheel deceleration signal on line 22 goes to both the large deviation control loop 16 and to the basic deceleration control loop 15.

With respect to the basic deceleration control circuit 15, the signal on line 22 enters a summer 23. In the case of a four wheeled aircraft, the wheel signals 24 from the other three wheels would also be summed into the summer 23. The summed signal on line 25 would enter an averaging circuit 26 and be divided bv the number of wheels on the aircraft, which in this case is four, and an average aircraft wheel deceleration signal on line 27 will enter the summer 28. Also entering summer 28 is the pilot selected deceleration signal on line 20. In the summer 28, the average aircraft wheel deceleration signal on line 27 is compared to the pilot selected deceleration signal on line 20, and a deceleration error signal is generated on line 29. This error signal on line 29 enters a hysteresis circuit 15A comprising a summer 30, a saturated amplifier 32, a gain circuit 34 and an integrator 36.The hysteresis circuit 15A provides a signal on line 37 to a summer 38 where it is summed with signal on line 52 from the large deviation control loop 16 and the output signal on line 39 from the summer 38 is fed to the brake valve driver 40. The signal on line 39 will cause the brake pressure and the ground deceleration force to increase when the rate of wheel deceleration is low, and to decrease when the rate of wheel deceleration exceeds the selected value. The integrator 36 is essentially the inverse of a diffentiator, and the term K stands for the gain. The integrator 36 will cause the brake pressure to increase if the rate of wheel deceleration is too low and cause the brake pressure to decrease if it is too high.

The basic deceleration control circuit 15 functions to adjust the brake pressure and thereby the rate of wheel deceleration, until it satisfies or drives the error signal on line 33 to zero, in which case the average wheel deceleration signal would be equal to the pilot selected deceleration signal, thereby maintaining the rate of aircraft deceleration on the front side of the brake force curve 11, shown in FIGURE 1 The basic deceleration control circuit 15 wil]- work fine for pretty much stabilized conditions, but if the aircraft tires happen to hit an ice patch or water puddle on an otherwise dry runway with the selector switch 18 set for dry runway braking conditions, this will cause the peak of the brake force curve 11 to be lowered to the point where the tire frictional force may not be able to sustain the applied brake torque and thereby result in the tire going into a skid.

Therefore, in order to provide recovery from the skid, the large deviation control circuit 16 enclosed within the dash-dot outline is utilized. Also, it is conceivable that the basic deceleration control 15 may not have sufficient authority to maintain the selected deceleration rate and braking control over the wheel, and therefore the large deviation control circuit 16 would be necessary. The large deviation control circuit 16 also receives the wheel deceleration signal on line 22, which enters into the summer 45 where the signal on line 22 is compared to a signal on line 44 coming from block 43. From block 43, a second reference deceleration signal on line 44 is provided which is slightly greater than the amount of the pilot selected deceleration signal on line 19 entering block 43, and is shown as approximately 110% thereof.Basically, the large deviation control circuit 16 controls brake force when deceleration exceeds the second reference deceleration, or, approximately 110% of the signal on line 19. From the summer 45, the deceleration error signal on line 46, or the difference between the wheel deceleration signal on line 22 and the second reference deceleraion signal on line 44, enters the limiter 47.

The graph shown in the limiter 47, as well as those graphs shown in other blocks in the figures, are of the standard x, y coordinate type, wherein x is the input and y is the output. The input signal is shown as coming in on the horizontal line, and the output signal is indicated on the vertical line. The upper right quadrant is positive, and the lower left quadrant is negative.

The limiter 47 limits the output signal on line 48 to positive values, i.e., if the output signal on line 46 from summer 45 is negative, then from the limiter 47 there will be a zero output signal on line 48; and if the output signal on line 46 from summer 45 is positive, then there will be essentially a one-to-one positive output signal 48. From the summer 45 the output signal on line 46 will not become positive until the wheel deceleration signal on line 22 is a predetermined amount greater than the pilot selected deceleration signal on line 19; i.e., when the signal on line 22 is greater than 110% of the pilot selected deceleration signal on line 19, or greater than the signal on line 44, then there will be a positive output signal from the limiter 47 on line 48.For example, assuming that, by use of switch 18, the pilot selected a deceleration rate of 10 ft/sec, then, before there would be any signal output from the summer 45, there would have to be a signal on line 22 equivalent to a wheel deceleration greater than 11 ft/sec2.

The output of limiter 47, which is a positive output error signal on line 48, enters block 49 and through line 50 to summer 51. In block 49, the term Kr denotes a constant gain, and the term KDS denotes a gain KD times the derivative s of the error signal on line 48.

Therefore, the output signal on line 50 is derived from a constant times the amount that the wheel deceleration is exceeding the value of the signal on line 44 plus a signal that is proportional to a change of that signal.

In other words, for example, if the change in the wheel deceleration is rapidly increasing, the K,s term will produce an output signal on line 50 which is proportional to that rate of - change. Essentially, block 49 exaggerates the error signal on line 48 so as to provide an output signal on line 50 of a greater signal strength or amount in order to produce a faster rate of change. The output of limiter 47, which is a positive output error signal on line 48, also enters the summer 54 where a predetermined discharge value signal on line 53 is subtracted from it.From summer 54, the signal on line 55 enters block 56 which is a lag circuit or essentially a time constant circuit R rs + 1 The block 56 functions somewhat like an integrator, in that with a signal input it will cause the signal to be integrated or come up to the signal value that there is on a time constant basis, i.e., -T Cr The block 56 provides a controlled duration memory so that if an error signal is developed on line 48, it will be fed into block 56 and the output signal on line 57 will come up slowly.However, if the error signal on line 48 goes away, which would be equivalent to gettmg a wheel deceleration signal on line 22 that would be less than the signal on line 44, then there will be an output remaining from block 56 on line 57, that will not disappear as soon as the signal on line 48 disappears.

In general, the function of the lag circuit, comprising the summer 54 and block 56, is to provide a signal to the brake valve driver 40 that will cause the brake pressure to remain off for a little while longer than the length of time that the error signal on line 48 exists.

The discharge signal input on line 53 to the summer 54 will control the rate at which the error signal on line 48 decays, in addition to the normal time constant of the circuit in block 56. The signal outputs from blocks 49 and 56 through lines 50 and 57, respectively, are summed in summer 51 and fed through line 52 to summer 38. From summer 38, the signal through line 39 is applied to the brake valve driver 40. When there are signal outputs from blocks 49 and 56, they are applied to the valve driver 40 so that the brake pressure is reduced, thereby driving the wheel deceleration signal on line 22 down to a value less than the value on line 44, in which case, the basic deceleration control circuit 15 will again assume primary control of the braking operation.

Referring to FIGURE 1, during dry run way conditions the pilot would normally select a rate of deceleration at point A on the brake force curve 11, and the basic deceleration control circuit 15 would then maintain the rate of deceleration of the aircraft at the selected position. Although the actual rate of deceleration of the aircraft on the brake force curve 11 will vary somewhat due to aerodynamic drag and other forces that operate on the aircraft, e.g., if the wheels go through a water puddle or over an ice patch, or something that causes the value of the coefficient of braking friction Mu to decrease, then the aircraft deceleration rate will not be able to be maitnained at point A.The tire slippage will result in the braking operation moving over the peak of the brake force curve 11 to some point B on the back side of the brake force curve, in which case the excess rate of deceleration of the wheel due to tire slippage will increase the signal on line 22 shown in FIGURE 3, thereby causing in the large deviation control circuit 16 an error signal output on line 48 from block 47. This error signal on line 48 is fed through blocks 49 and 56, lines 50 and 57, respectively, into summer 51, through line 52 and into summer 38, and through line 39 to the brake valve driver 40.This excess rate of wheel deceleration due to tire slippage causes the large deviation control circuit 16 of the braking system to call for less brake pressure, which will cause the operation of the system, referring to FIGURE 1, to move back onto the front side of the brake force curve 11, because the brake pressure has been reduced. When the operation of the system moves back onto the front side of the brake force curve, the wheel deceleration signal on line 22 in FIGURE 3, will become less than that on line 44, and the error signal on line 48 will disappear. Now, the brake pressure applied by the valve driver 40 will remain at a lower value due to the signal output of the lag circuit until the normal decay of block 56 causes the brake pressure to come back up.

This will then revert the braking control back to the basic deceleration control circuit 15.

The function of the selected deceleration control circuit 60 enclosed within the dashdot-dot outline, between the large deviation control circuit 16 and the basic deceleration control circuit 15, is to accommodate for possible pilot error in selecting a deceleration rate that is greater than the available runway coefficient of friction conditions, in which case, the tires will go into a skid, causing the large deviation control circuit 16 to be activated.

When the wheels begin to skid, the wheel deceleration signal on line 22 becomes positive and exceeds the second reference deceleration signal on line 44 to produce a deceleration error signal of positive sign on line 46 from summer 45. As a result, block 61 of the selected deceleration control circuit 60 provides a step signal on line 62 as shown in the upper graph of FIGURE 3A which is fed to a time constant or lag circuit 63. This circuit 63 provides a controlled duration memory which will reduce the brake torque of the wheel to a level consistent with the runway condition.In lag circuit 63, the step signal on line 62 will oroduce an output signal on line 64 that will increase exponentially on the basis of rT, to the value of the step signal on line 62 as shown in the center graph of FIGURE 3A; and if the step signal on line 62 goes away to a zero or a minus value, the output signal on line 64 will decay exponentially back to a zero value. The output signal on line 64 goes to summer 65 and functions to lower the pilot selected deceleration signal coming into the summer from line 19; i.e., the output signal on line 64 from time constant or lag circuit 63 is subtracted from the pilot selected deceleration signal on line 19 in summer 65 in order to reduce the pilot selected deceleration signal to a value more closely matching the runway operating conditions as shown in the lower graph of FIGURE 3A.The output of summer 65 on line 20 is compared in the summer 28 with the signal on line 27 in order to produce the error signal on line 29. The selected deceleration control circuit 60 has the net effect of reducing the signal level that is being compared in the summer 28 with the signal on line 27 and then bringing it back to the level that the pilot has selected in switch 18.

This can be more clearly seen in FIGURE 3A. When the step signal on line 62 from block 61 has disappeared, the output signal on line 64 will return to zero, and the signal on line 20 will begin to increase back up to the value that the pilot has selected in switch 18. Therefore, if the pilot selects a rate of deceleration that is too high, the selected deceleration control circuit 60 will function to adjust the selected rate of wheel deceleration downward to a level that the tires can sustain without slippage. At some point, if the operating conditions continue poorer than the system is set for, another step signal will be generated on line 62 and the cycle will repeat.Without the selected deceleration control circuit 60 interconnecting the large deviation control loop 16 with that of the basic deceleration control loop 15, the large deviation control circuit 16 would cycle rapidly, thereby keeping the deeceleration rate lower than necessary and detracting from the overall efficiency of the system.

With the selected deceleration control circuit 60 included in the system, the deceleration rate selector switch 18 could be eliminated; however, there are some advantages in retaining it. With the selector switch 18 left in the Dry position at all times, at the start of braking the system would adjust to the prevailing runway surface conditions within a few cycles of the large deviation control circuit 16 and this would sacrifice some stopping time and distance. Therefore, it would be advisable, whenever stopping distance was critical, for the pilot to select as accurately as possible the prevailing ground surface conditions and thereby the rate of deceleration, prior to touchdown. In any case, the brake force is kept on the positive side of the Mu/slip curve 11, thus providing optimum steering capability and reduced tire wear that are both due to the limited-slip function.

With respect to FIGURE 3, although four wheels are summed in summer 23, it will be understood that the basic deceleration control circuit 15 could be operated without averaging the wheel dceleration rate and deleting summer 23 and block 26; and then the system would function on an individual wheel control basis, utilizing a pressure balance or energy balance system interposed for maintaining equal pressure or equal braking torque on all of the wheels, Also, the system as shown in FIGURE 3 utilizes wheel rotational velocity as an input signal 1 which is converted in differentiator 21 to a wheel deceleration signal on line 22; however, another source for an input signal to the basic deceleration control circuit 16 could be from an air data system or an inertial navigation system and, if used, would be an input signal through line 70 coupled with the deletion of the circuit comprising line 22, summer 23, and block 26.

Basically, it is desirable to control the rate of aircraft deceleration. Therefore, if there were an aircraft decleration input signal on line 70 to the summer 28, from an air data system or an inertial navigation system, it could be utilized as the controlling parameter instead of the input signal on line 27 relating to the rotational velocity of the wheels.

In FIGURE 3, the basic deceleration control circuit 15 includes a hysteresis circuit 15A in heavy outline comprising summer 30, saturated amplifier 32, and gain circuit 34.

This hysteresis circuit operates as a deceleration rate limit searching circuit system, wherein the object or function of the circuit 15A is to make each wheel increase its rate of deceleration slightly above the pilot's selected deceleration rate in order to insure that each wheel can develop the torque necessary to produce the desired aircraft deceleration rate.

The signal inputs to the deceleration control circuit 15 are obtained from the measured wheel rotational vlocitv signal w, which is put through the lagged differentiator block 21 to become the wheel deceleration signal on line 22, and the pilot selected deceleration signal on line 19. In the summer 28, the wheel deceleration signal on line 22, which is now the signal on line 27, is subtractively combined with the pilot selected deceleration signal on line 19, which is now the signal on line 20 and the difference produces an error signal output on line 29 which enters the summer 30. Also entering summer 30 is a signal on a line 35 which is a percentage of the output of the hysteresis circuit 15A on a line 33.The signal on line 35 is algebraically summed with the signal on line 29 in the summer 30, and the difference produces an error signal output on line 31 which is transmitted to the saturated amplifier 32.

If the signal on line 31 is of a positive sign, then saturated amplifier 32 produces a step-negative output signal on line 33; and if the signal on line 31 is of a low value or of a negative sign, then saturated amplifier 32 produces a step-positive output signal on line 33.

The signal on line 33 leaving the saturated amplifier block 32 is of a constant level, that is, either of a low negative value for a positive input signal on line 31, or of a high positive value for a negative input signal on line 31.

The signal on line 33 of the hysteresis circuit 15A enters the integrator circuit 36 where it is integrated and produces an output signal on line 37 which changes at a rate proportional to the amplitude of the signal on line 33, and its sign is in the direction determined by the sign of the signal on line 33. From the integrator 36, the signal on line 37 enters the summer 38 and then through line 39 to the brake valve driver 40 of the braked wheel (not shown).

Since the signal on line 33 from the saturated amplifier block 32 is either a plus or a minus voltage signal of a constant amplitude, there will be a constant rate output from the integrator 36 through line 37 to the valve driver 40; and in a sense, the constant rate output of an increasing and decreasing deceleration, working around the loop in the hysteresis circuit 1SA, produces the triangular wave from shown in FIGURE 4. A signal on line 37 which is going negative, leaving the integrator 36, produces an increase in the brake pressure, which results in an increase in the rate of wheel deceleration; i.e., along the positive slope side of the Mu/slip curve shown in FIGURE 1, or on the positive slope side of the graph lines shown in FIGURE 4.

Likewise, a signal on line 37 which is going positive, leaving the integrator 36, porduces a decrease in the rate of wheel deceleration, or on the negative slope side of the graph lines shown in FIGURE 4.

Part of the signal on line 33 is fed back to summer 30 through the gain circuit 34.

Should the signal leaving circuit 34 be negative, then the signal entering the summer 30 will have a negative value; and this negative value, when summed in summer 30 with a positive value of deceleration error signal on line 29, will produce a greater strength positive signal on line 31 leaving summer 30, thereby continuing to produce a constant strength step negative signal on line 33 for an increased brake force. In order to get the saturated amplifier 32 to switch to a step-positive signal on line 33 to decrease the brake force, the measured wheel deceleration signal 27 entering the summer 28 must increase until it exceeds the pilot selected deceleration signal on line 20, by an amount A1, as shown in FIGURE 4, which is equal to the signal strength level on line 35 that is determined by the value of the constant K in gain circuit 34.When the brake force is increased so that the amount of the wheel decleration signal on line 27 does exceed the sum of the signal amounts on lines 20 and 35, the summer 30 will then produce an error output signal on line 31 of a negative sign. This negative error signal on line 31, when fed into the saturated amplifier 32, will produce a constant strength step-positive output signal on line 33; and a part of this signal on line 33 is fed back to the summer 30.Now, the signal on line 35 entering the summer 30 is positive, and is therefore subtracted in summer 30 from the signal on line 29 which is the summation of the pilot selected deceleration signal on line 19 and the wheel deceleration signal on line 22; whereby, the wheel deceleration signal on line 22 now has to be less than the pilot selected deceleration signal on line 19 by an amount 2, or by the amount equal to the gain constant K times the amplitude of the negative signal out of saturated amplifier 32, in order for the brake force on the wheel to be increased.The amount of signal feedback to summer 30 through line 35 is aDproximately 10% of the pilot selected deceleration signal on line 19, e.g., assume that the pilot selected deceleration signal on line 19 is equal to 10 ft/sec2, then in Figure 4 51 and 2 would each be equal to 1 ft/sec2.The following description of operation of the hysteresis circuit system 1SA shown in FIGURE 3 will be more clearly understood when considered along with FIGURE 4 and with the following assumptions: that the wheel deceleration signal on line 22 is zero; that the pilot selected deceleration signal on line 19 is 10 ft/sec2; that the high level of the saturated amplifier 32 is +1 and the low level -1; and that the gain constant K in gain circuit produces a signal on line 35 representing a deceleration of 1 ft/sec2. When the wheel deceleration signal on line 27 is zero, then the output from summer 28 will be a positive signal on line 29 into summer 30.From summer 30 a positive output signal on line 31 will enter the saturated amplifier 32 to produce a constant strength negative signal on line 33 to the integrator 36 whose output is applied through summer 38 to the brake valve driver 40 to apply a brake force. The signal on line 33 into the gain circuit block 34 will produce a signal of negative 1 ft./sec2 input to summer 30. Referring to FIGURE 4, point A, at the initial application of the brake force, the rate of aircraft deceleration is low and for a wheel deceleration signal on line 22 which is less than the pilot selected deceleration signal on line 19 by an amount a2, the output of the hysteresis circuit 15A will be negative.

This negative signal will increase the brake pressure, resulting in an increase in the rate of aircraft deceleration as shown by the positive slope of the line from point A to B.

Referring to FIGURE 3, as brake force is applied, the wheel deceleration signal on line 22 will increase, and this will reduce the amount of the positive error signal from summer 30 on line 31 to the saturated amplifier 32. The brake force will increase at a constant rate until the wheel deceleration signal on line 27 exceeds the pilot selected deceleration signal on line 20 by an amount , as indicated by point B in FIGURE 4, or until the measured wheel deceleration equals or exceeds 11 ft/sec2. At this time, referring to FIGURE 3, the output signal from summer 30 on line 31 becomes zero or negative, and this signal then causes the saturated amplifier 32 to produce a constant strength positive output signal on line 33.

This positive signal on line 33 then causes a decrease in the brake force and a corresponding reduction in the rate of aircraft deceleration, as shown in FIGURE 4 by the negative slope of the line from point B to C. Referring to FIGURE 3, this positive output signal on line 33 fed back through gain circuit 34 produces a positive 1 ft/sec2 signal on line 35 to summer 30. In summer 28, the summation of the pilot selected deceleration signal on line 20, of 10 ft/sec2, and the wheel decleration signal on line 27, of 11 ft/sec2 or more, will produce a negative error signal on line 29 to summer 30 of more than a minus 1 ft/sec2 which, when summed with the positive 1 ft/sec2 signal from line 35 will produce a minus input signal 31 of 2 ft/sec2 to the saturated amplifier 32.This will result in the saturated amplifier 32 continuing the output of a constant strength positive signal on line 33, to further decrease the brake force. This will continue until the output of summer 30 into line 31 becomes positive again, or referring to FIGURE 4, until the rate of aircraft deceleration has decreased the specified amount of A1 + A, or reached point C. The output from the hysteresis circuit 15A then becomes negative again and the rate of aircraft deceleration then increases. This cyclic operation thus continues within the hysteresis circuit 1SA in a counterclockwise direction around the loop.

However, between the two limit values of A1 and A,, the sign of the output signal on line 33 depends on its previous state, i.e., if its output were negative, it will remain negative until the wheel deceleration signal on line 22 exceeds the pilot selected deceleration signal on line 19 by the amount A1, at which time the signal on line 33 will become positive; and if the output of the hysteresis circuit 15A is positive, it will remain positive until the wheel deceleration is less than the selected decelera tion by an amount 2, at which time it will again become negative.

FIGURES 5 and SA depict a second em bodiment of the basic deceleration control circuit 15 which also uses a deceleration rate limit searching circuit system. The operation of this second embodiment system is similar to that previously described with reference to FIGURE 3, in that in each of these systems, each wheel is forced to a deceleration value slightly higher than the selected or aircraft deceleration level in order to obtain assurance that each wheel has sufficient torque to produce the selected deceleration rate.This second embodiment system utilizes a threshold and a single-shot multivibrator circuit instead of the hysteresis circuit 15A and the signal inputs are again the measured wheel rotational velocity signal w1 which is put through a lagged differentiator 21 to become a wheel deceleration signal on line 22 and the pilot selected deceleration signal on line 19. These two input signals on lines 22 and 19 enter a summer 71 which could be equivalent to summer 28 in FIGURE 3, wherein they are algebraically summed and their difference produces an error signal output on line 72 which enters the threshold circuit 73.

When the measured wheel deceleration rate signal on line 22 is greater than the pilot selected deceleration signal on line 19, a negative error signal is generated on line 72 to the threshold circuit 73. When this negative error signal on line 72 is of a zero or low amount, i.e., below a preset A amount shown in FIGURE 5A, there will be no output signal on line 74 from threshold circuit 73 of suffi cient magnitude produced to trigger the single shot multivibrator 75. The A amount, or tuning parameter of the threshold circuit 73, is approximately 10% of the selected deceleration rate or of some fixed constant amount, e.g., 1 ft/sec2. However, when this negative error signal on line 72 exceeds the preset A amount, the threshold circuit 73 will produce a steppositive signal on line 74, which will enter the single-shot multivibrator 75.This steppositive signal on line 74 will cause the single shot multivibrator 75 to produce a single positive pulse of constant amplitude and of fixed duration, which through line 76 will enter the summer 77. This positive pulse on line 76 is summed with a constant level, negative discharge signal on line 78, and from summer 77 the output signal on line 79 enters the integrator 80. When the multivibrator 75 fires a single-pulse through summer 77 to the integrator 80, it will integrate the pulse and produce a constant rate signal on line 81 to the brake valve driver 40 which will result in a fixed, incremental decrease in the decelera tion to a level below the threshold; and the sequence will then repeat.

In the second embodiment of the basic deceleration control circuit, the following description of operation of the circuit system shown in FIGURE 5 will be more clearly understood when considered in combination with FIGURE 5A and with the following assumptions: that the aircraft is rolling unbraked down the runway at a constant speed; that the wheel deceleration signal on line 22 is zero; and that the pilot selected deceleration signal on line 19 is 10 ft/sec2. The output from summer 71 will be a negative value error signal equivalent to a minus 10 ft/sec2 into line 72 which enters the threshold circuit 73. The function of the threshold circuit 73 is depicted by an x, y, axis graph. The x axis represents a zero value of the output signal represented by the y axis, which in the circuit shown could be a constant, low-level reference signal.To the left of the y axis represents the negative or minus sign of the input signal. The negative value error signal on line 72, equivalent to a minus 10 ft/sec2, is shown on the graph as point A on the left side of the y axis and also directly along the x axis and causes threshold circuit 73 to produce a constant, low-level reference signal output which will not be enough to trigger the single-shot multivibrator 75 into producing a step-positive signal on line 76. Due to the negative discharge signal on line 78, the output from summer 77 through line 79 to the integrator 80 will be of a negative value. This will cause the integrator 80 to integrate downward in a negative direction to produce a signal value becoming more negative, which will cause an increase in the brake pressure as depicted in FIGURE 5A by the positive slope of the line from point A to C.As the brake pressure is increased, the wheel deceleration is also increased along the line A-C. When the wheel deceleration signal on line 22 and the pilot selected deceleration signal on line 19 coming into summer 71 are equal, the positive slope line A-C in FIGURE SB is intersecting, at point B, the solid horizontal line repre- senting the pilot selected deceleration signal on line 19 of 10 ft/sec2. As the wheel deceleration continues to increase along the line A-C, at some point indicated at C, it will reach the A amount of 1 ft/sec2 over the selected deceleration of 10 ft/sec2, or the measured wheel deceleration rate of 11 ft/sec2 which is represented by the upper dashed horizontal line. When this point is reached or exceeded, the threshold of circuit 73 is crossed and circuit 73 will send out a step-positive signal on line 74 to the single-shot multivibrator 75.

This will produce a constant duration pulse, e.g., of one-half of a second, with a constant amplitude positive signal of approximately one.

During the duration of this pulse, there will be a positive signal output from the summer 77 which will cause the integrator 80 to integrate upward in a positive direction, and this will result in a decrease in brake pressure which, as shown in FIGURE 5-A, will correspond to the negative slope of the line from point C to D. As the brake pressure is decreased, the measured wheel deceleration will decrease along the line G--D until it goes below the selected deceleration and the amount that it does go below will depend upon the amplitude of the pulse and the duration of the pulse or system gain, e.g., for a very long duration pulse or a higher amplitude pulse, more braking pressure will be taken off.The amplitude and width of the pulse are tuning parameters of the system for decreasing brake pressure along the negative slope line C--D.

When the measured wheel deceleration has reached point D, the output signal on line 74 from threshold circuit 73 will again be a constant, low-level reference signal and the cycle will start over again.

FIGURE 6 is a block diagram of a brake equalization or a brake energy balance control system, and the circuit depicted is for equalizing the brake energy between two wheels.

A brake energy balance control system is necessary to insure that all brakes are performing their assigned work load and to assure even wear of all the brakes, since an operation problem can result because a wheel can decelerate at the same rate that the aircraft de celerates. without any brake torque or brake pressure being applied. This problem manifests itself in that some of the brakes do all of the work in decelerating the aircraft, which results in excessive wear and overheating of these brakes.

Brake energy balance control can be accomplished by either a passive method or an active method, as shown in FIGURE 6, which essentially comprises a subcontrol in which some function of brake energy is measured; and this measured brake energy is then compared to a reference signal for producing a control response action.

In controlling the deceleration of an aircraft, the brake energy balance control system shown in FIGURE 6 will function satisfactorily regardless of the method used for achieving the selected deceleration rate; such as thrust reverses, wing spoilers and flaps, or other means by which the selected deceleration rate can be accomplished without the use of - brakes. Also, there is the possibility that the deceleration requirement may be met by braking only one wheel, or through wheels carrying unequal braking loads; and this could cause brake overheating and excessive wear of the wheels carrying the braking load.

In FIGURE 6, the measured rotational velocity of the wheel w1 or wheel speed signal of a first wheel on a line 91 and its measured braking torque Tbl signal on line 90 are input to a multiplier circuit 92, wherein the brake torque is multiplied by the wheel speed, and the product thereof is the rate at which the brake is doing work, or the rate at which the brake energy is being put into the wheel. This rate of brake energy signal on a line 93 is sent to an integrator circuit 94, and the integral thereof is the brake energy Ebl, which is the input signal on line 95 to the summer 97. Also entering the summer 97 is a similar brake energy signal Eb2 from a second wheel.

As an alternative, the first wheel brake energy signal Ebl on line 95 could be compared in the summer 97 with an average brake energy input signal from all of the wheels.

The output signal on line 98 from summer 97 will be the difference between the two input signals on lines 95 and 96; and if the brake energy Eb2 from the second wheel is greater than the brake energy Ebi from the first wheel, then from summer 97 the signal output on line 98 will be of a positive sign.

This signal on line 98 will enter block 99 which depicts a standard x, y coordinate graph, the x axis being the input, with the right side of the y axis being the positive input side, and the y axis being the output, with the upper side of the x axis being the positive output. For a positive signal input to block 99 which would be on the right side of the vertical y axis, there will be a fixed amplitude or constant negative signal output, as indicated by the short horizontal line in the lower right quadrant of the y axis, and for a negative signal input, which would be on the left side of the vertical y axis, there will be a zero signal value output.With a negative signal of constant amplitude output from block 99 through line 100 to the integrator K s in block 101 for the first wheel, the integrator signal output on line 102 to the brake valve driver 40 will cause the brake pressure to increase. And in a similar manner, when the brake energy of the first wheel is greater than the brake energy of the second wheel, there will be no signal output from the block 99 to the integrator 101, and hence no increase in brake pressure. A similar circuit, though, for the second wheel will cause its brake pressure to increase, thus equalizing brake energies.

An alternative to equalizing the brake energy as described in FIGURE 6 would be to equa lize the power or the rate at which energy is being put into the brake and this can be done by eliminating the integrator circuit 94, so that instead of comparing, in summer 97, the brake energies from the two wheels or of one wheel and the average brake energy for the aircraft, the rate of energy input to the brake would be compared.

Since the brake torque is difficult to measure, the brake pressure or the amount of current flow to the brake valve driver 40 could be used with sufficient accuracy to insure the proper operation of the system. Also, brake temperature could be used as an indication of brake energy.

The active methods of brake energy balance control comprise the use of a searching system, as shown in the hysteresis circuit 15A of FIGURE 3, and the circuit shown in FIGURE 5. In each of these systems the wheel is forced into a deceleration rate value slightly higher than the selected deceleration rate level; and in this way, assurance is obtained that each wheel has sufficient torque to produce the selected deceleration rate.

A passive method of brake energy balance control involves producing approximately the same torque in all of the brakes, and this can be done by supplying the same pressure to all of the brakes through the use of a common valve for all brakes, or through the use of a common control circuit for all of the wheels including, as shown in FIGURE 3, the large deviation control circuit 16, on an individual wheel basis, thereby forcing the low pressure level wheels toward a higher pressure.

Now referring to FIGURE 7, the limitedslip brake control system illustrated therein includes a basic deceleration control circuit responsive to a wheel deceleration signal and a selected reference deceleration signal to provide a first brake control signal which is used to modulate brake pressure applied to the braked wheel in order to maintain the vehicle deceleration at the selected rate, and a large deviation control circuit which is also responsive to the wheel deceleration signal and which functions to reduce brake pressure in the situation where the braked wheel encounters a sudden decrease in the available coefficient of friction and is forced into a skid by the basic deceleration control circuit.

The basic deceleration control circuit includes: a summing junction 122 which compares a wheel deceleration signal on a line 121 from a differentiator 120 with a selected reference deceleration signal on a line 129 from a decel select circuit 128; a hysteresis circuit 130 which is responsive to a first deceleration error signal on a line 123 from summing junction 122; a summing junction 138 responsive to an output signal on a line 135 from hysteresis circuit 130; and, an integrating circuit 162 which is responsive to an output signal on a line 139 from summing junction 138, with the first brake control signal appearing on an output line 167 from integrating circuit 162.The large deviation control circuit includes: a large deviation sens ing circuit 124 which is responsive to the wheel deceleration signal on line 121 from differentiator 120 to provide an output signal on a line 145 when wheel deceleration exceeds a second reference deceleration, preferably higher than the selected reference deceleration; the summing junction 138 which is also responsive to the output signal from large deviation sensing circuit 124 on line 145; and, the integrating circuit 162.

The first brake control signal on line 167 is applied to a summing junction 168. Neglecting for a moment the effect of the other signals supplied to summing junction 168 as illustrated in FIGURE 7, the first brake control signal appears at the output of summing junction 168 and is coupled by a diode 169 and a line 159 to the input of a brake valve driver 160 which in turn provides a brake control signal to control the position of a brake valve (not illustrated) located in the hydraulic braking system. As previously described, a typical hydraulic braking system would provide metered brake pressure to the brakes of the braked wheel in response to pilot-applied pressure to the brake pedals of the brake system. The brake valve functions to modulate the metered brake pressure in inverse proportion to the first brake control signal applied to the brake valve driver 160.

The first brake control signal includes a component, obtained from the basic deceleration control circuit, which will cause the brake pressure to increase when the actual wheel de ceferation is lower than the selected reference deceleration, and to decrease when the actual wheel deceleration is greater than the selected reference deceleration. The first brake control signal will also includ a component, obtained from the large deviation control circuit, which will cause the brake pressure to decrease as long as the actual wheel deceleration exceeds the second reference deceleration.

The basic deceleration control circuit and the large deviation control circuit illustrated in FIGURE 7 therefore function in a manner similar to corresponding circuits in the limitedslip brake control system previously described.

The limited slip brake control system illustrated in FIGURE 7 also provides a second brake control signal on a line 157 from large deviation sensing circuit 124 when the actual wheel deceleration exceeds a third reference deceleration, higher than the second reference deceleration (and therefore the selected reference deceleration), signifying that the braked wheel has gone into a deep skid, which second-b"rake control signal is applied directly through a diode 158 and line 159 to brake valve driver 160 to immediately reduce brake pressure; an initial skid circuit 146 responsive to an output signal on a line 143 from large deviation sensing circuit 124, which output signal occurs when the actual wheel deceleration has exceeded the second reference deceleration, and with the initial skid circuit 146 functioning to provide an output signal on a line 151 to the summing junction 138 so as to modify the first brake control signal to minimize or eliminate the tendency of the basic deceleration control circuit to place the braked wheel into a skid upon the initial application of brake pressure; and, a dynamic compensation circuit 126 responsive to the wheel deceleration signal on line 121 and providing a compensating brake control signal on a line 195 to the summing junction 168 in order to compensate the brake pressure command represented by the first brake control signal applied to brake valve driver 160 for apparent changes in wheel deceleration occasioned by landing gear strut assembly oscillation.

Considering the improved limited-slip brake control system in more detail, a measured wheel speed signal W2 is derived from an AC wheel speed transducer which puts out a sine wave that is squared, and then filtered or rectified to produce a DC voltage that is proportional to wheel speed. The wheel speed signal w1 is applied to a differentiator 120 whose transfer function is represented by the LaPlacian operator s 1 fry in which the operator s in the numerator represents a differentiation, and in which the operator 1 +rips in the denominator represents a lag circuit whose frequency is represented by the value of the constant r.Typically, r is chosen so that the operator 1 +ras functions as a filter to attenuate higher frequencies, such as noise, radio frequency interference, and the like. The resultant output on line 121 from differentiator 120 is therefore proportionaT to differentiated wheel speed, or wheel deceleration.

The decel select circuit 128 provides a selected reference deceleration signal on line 129 which preferably comprises a DC signal representing a desired deceleration. The selected reference deceleration signal may be infinitely variable or may vary in discrete increments, in which case the decel select cricuit 128 may include a selector switch having, for example, three positions corresponding to Dry, Wet and Icy runway conditions and accordingly representing high, intermediate and low decelerations.

i;ewheel deceleration signal and selected reference deceleration signal are subtractively combined in summing junction 122, with the result that the first deceleration error signal on line 123 is proportional to the difference therebetween. The first deceleration error signal is positive if the wheel deceleration is greater than the reference deceleration and is negative if the wheel deceleration is lower than the reference deceleration. The first deceleration error signal is applied to a summing junction 132 where it is additively combined with a portion of the output of the hysteresis circuit 130 appearing on a line 137. An error signal appearing on the output of summing junction 5132 is applied by a line 133 to the input of a limited amplifier 134.If the error signal on line 133 has a positive sign, then the limited amplifier 134 will provide, on its output line 135, a signal having a predetermined, positive value; and if the error signal on line 133 has a negative sign, then the signal on the output line 135 will have a predetermined, negative value. The signal on line 135 passes through the-summing junction 138 and is applied by a line 139 to the input of the integrating circuit 162 and, more particularly, to an integrator 164 therein whose LaPlacian operator is represented as 1 The output of the integrator 164 is coupled by a line 165 to the input of a positive-value, limiting circuit 166 to whose output the line 167 upon which the first brake control signal previously described appears.As previously noted, the first brake control signal passes through the summing junction 168, the diode 169, and the line 159 to the brake valve driver 160.

~The operation of the basic deceleration control circuit may be understood by considering the steps that take place when the system is installed on an aircraft and the aircraft touches down. At the time of touchdown, full brake pressure is applied to the braked wheels by the hydraulic braking system (not illustrated).

As each braked wheel begins to turn upon frictional contact with the runway, the wheel speed will first increase, and then decrease due to the application of brake pressure. The rate of decrease of wheel speed, or wheel deceleration, is represented by the wheel deceleration signal on line 121. Previous to touchdown, a pilot will have selected the desired reference deceleration whereby a representative selected reference deceleration signal is being supplied on line 129. If the rate of wheel slowdown, or wheel deceleration, is lower than the reference deceleration, the first deceleration error signal on line 123 will have a negative sign and a magnitude proportional to the difference between the wheel deceleration and the reference deceleration.

The negative first deceleration error signal will result in a negative, fixed-value output signal on line 135 from the limited amplifier 134.

As a result, the output of the integrator 164 appearing on line 165 begins to decrease at a predetermined rate towards a negative value.

The positive-value, limiting circuit 166 will only produce an output signal whose value is zero in response to an input signal whose sign is negative. Therefore, the first brake control signal on line 167 goes to and remains at a zero value.

The brake valves typically used to modulate the brake pressure in a hydraulic braking system have a deadband, in which the application to the valve of a brake control signal having a magnitude varying from zero to a first predetermined value results in no modulation of brake pressure. Above this first predetermined value, a further increase in the magnitude of the applied brake control signal results in an inversely proportional decrease in brake pressure. When the applied brake control signal reaches a second, predetermined value, the brake pressure is reduced to zero, i.e., the valve is completely closed.

In the summing junction 168, the first brake control signal, now at a zero value, is summed with a positive, fixed value signal 8 which produces an offset in the brake control signal provided to the brake valve by the brake valve driver 160. This offset is equal to the first predetermined value representing the brake valve deadband, as a result of which the brake valve normally rests, when the first brake control signal has a zero value, at the start of its inversely proportional operating range. At this time, the brake valve driver 160 therefore provides a small positive brake control signal to the brake valve so that brake pressure is not reduced.

Due to the application of full brake pressure, the aircraft, and therefore the braked wheels, continues to decelerate. When the wheel deceleration exceeds the reference deceleration, the first deceleration error signal on line 123 becomes positive so that the output signal from limited amplifier 134 on line 135 switches to a positive, fixed value. As a result, the output signal from integrator 164 on line 165 eventually begins to rise to a positive value at a rate determined by the magnitude of the positive, fixed value output from limited amplifier 134. This positivegoing output from integrator 164 is coupled through positive-value, limiting circuit 166 to appear as the first brake control signal on line 167 which accordingly causes brake valve driver 160 to provide an increasing, positive brake control signal to the brake valve to reduce brake pressure.The rate of brake pressure reduction is determined by the rate of increase of the first brake control signal, and thus by the magnitude of the positive, fixed value output from limited amplifier 134.

In the hysteresis circuit 130, the output signal appearing on line 135 is applied to the input of a gain circuit 136 appearing (having a gain constant K2), and the output signal from gain circuit 136 on a line 137 is coupled to and additively combined in summing junction 132 with the first deceleration error signal present on line 123. The gain constant K2 is chosen so that the magnitude of the output signal on line 137 represents a certain incremental deceleration, e.g. 1 ft/sec.2. As the wheel deceleration begins to decrease in response to the decrease in brake pressure, the wheel deceleration signal on line 121 decreases. At some point, the wheel deceleration decreases to a point where it is equal to and then lower than the reference deceleration. At this point, the first deceleration error signal on line 123 goes negative.Due to the positive output signal from gain circuit 136 present on line 137, the output signal on line 133 from summing junction 132 does not go negative until the wheel deceleration has decreased below the reference deceleration by the incremental deceleration, e.g., 1 ft/sec2.

Limited amplifier 134 then switches to its negative, fixed value output, with the result that integrator 164 begins to ramp down to provide a corresponding ramp decrease in the first brake control signal, and to therefore again increase the brake pressure applied to the braked wheel at a rate determined by the magnitude of the negative, fixed value output from limited amplifier 134.

As brake pressure increases, wheel deceleration again increases. When wheel deceleration equals and exceeds the reference deceleration, the first deceleration error signal on line 123 goes positive. At this time, gain circuit 136 is supplying to the summing junction 132 a negative output signal representing the incremental deceleration, e.g., 1 ft/sec2. Therefore, the output signal on line 133 does not go positive until the wheel deceleration equals and then exceeds the incremental deceleration.

It will therefore be appreciated that the basic deceleration control circuit allows the limited-slip brake control system to function as a deceleration rate searching system in which brake pressure, and therefore wheel deceleration, is constantly cycling between an amount A1 above the reference deceleration and an amount 2 below the reference deceleration, and at a rate determined by the fixed value outputs of the limited amplifier 134, as illustrated in the graphs superimposed on FIGURE 1 for Dry, Wet and Icy settings of the decel select circuit 128.

Where a sudden change in the coefficient of friction of the runway is encountered, or when the pilot has selected too high a reference deceleration for the runway condition, the basic deceleration control circuit cannot maintain control of the braked wheel and will force the braked wheel into a skid. To permit the braked wheel to recover quickly from the skid, the large devaition control circuit is provided for modifying the first brake control signal to reduce brake pressure for the duration of and in response to each skid.

Specifically, the large deviation sensing circuit 124 includes a summing junction 140 which receives the wheel deceleration signal on line 121 from differentiator 120. A second reference deceleration signal 6, is supplied to and subtractively combined in summing juntion 140 with the wheel deceleration signal.

Preferably, the second reference deceleration signal is set to represent a second reference deceleration higher than the reference deceleration represented by the selected reference deceleration signal from decel select circuit 128 and therefore would typically be in the range of 5-20 ft/sec2. The second deceleration reference signal may be obtained from the selected predetermined increment above the selected reference deceleration, or may be independently set. The output from summing junction 140 comprises a second deceleration error signal which is coupled by a line 141 to the input of a positive-value, limiting circuit 142 whose output on a line 143 is coupled through a gain circuit 144 and a line 145 to a second input of the summing junction 138.

The positive-value, limiting circuit 142 has a transfer function similar to that of positivevalue, limiting circuit 166. That is, the positive-value, limiting circuit 142 provides no output for negative values of the second deceleration error signal and provides a positive output signal proportional to positive values of the second deceleration error signal up to a predetermined value, after which the output signal of the circuit 142 is limited at that predetermined value.

When a skid occurs, i.e., when wheel deceleration exceeds the second reference deceleration represented by signal d,, the second deceleration error signal on the line 141 is coupled through circuit 142 to the input of gain circuit 144, and thence to the summing junction 138. The resultant output signal on line 139 causes the integrator 164 to ramp the first brake control signal up from the value established by the basic deceleration control circuit at a rate determined by the magnitude of the second deceleration error signal and the gain constant (K3) of gain circuit 144. Accordingly, the signal supplied to the brake valve from brake valve driver 160 begins to increase to lower brake pressure.

As brake pressure decreases, wheel deceleration decreases until a point where the wheel deceleration signal on line 121 equals and then goes below the signal l82, at which time the output signal on line 145 from the large deviation sensing circuit 124 is terminated (due to the operation of positive-value, limiting circuit 142). Deceleration control then reverts to the basic deceleration control circuit which causes the integrator 164 to ramp the first brake control signal back down to increase brake pressure and thus wheel deceleration to the selected reference deceleration represented by the signal on line 129.Of course, if the ground surface conditions do not permit that selected reference deceleration to be maintained, the large deviation control circuit will again be operative to reduce brake pressure Where very deep skids are encountered, c that is, those skids in which very high wheel decelerations occur, the portion of the large deviation control circuit previously described may not function to relieve the skid, inas much as the integrating circuit 162 requires a certain amount of time to reduce brake pressure. Accordingly, the large deviation sensing circuit 124 includes a summing junc tion 152 which is provided with the wheel deceleration signal on line 121 from differen tiator 120. A third reference deceleration signal ,83 is subtractively combined in sum ming junction 152 with the wheel decelera tion signal.Preferably, the third reference deceleration signal '83 represents a third or high reference deceleration, e.g., 50 ft/sec2, denoting a deep skid condition. The output of summing junction 152 comprises a third deceleration error signal which is coupled by a line 153 to the input of a positive-value, limiting circuit 154 whose output on a line 155 is applied to the input of a gain circuit 156. A second brake control signal appears on the output of gain circuit 156 and is coupled by a line 157 and diode 158 to the input line 159 of the valve driver 160. Positive value, limiting circuit 154 is similar to posi tive-value, limiting circuits 142 and 166.

Accordingly, when wheel deceleration exceeds the third reference deceleration represented by the signal '8,3, the third deceleration error signal on line 153 is coupled through circuit 154 to the input of gain circuit 156, as a result of which the second brake control signal, proportional to the third deceleration error signal, appears on line 157. Diodes 158 and 169 function as an exclusive OR-gate, that is, they couple to input line 159 only that one of the signals present on the output of summing junction 168 or the output line 157 which has a higher value. In the situation being discussed, the first brake control signal on line 167, as coupled through summing junction 168, will of necessity represent a commanded rate pressure that is higher than that obtainable with the existing ground sur face condition.If the gain of circuit 156 (represented by the gain constant K9) is correctly chosen, then diode 169 will be back biased and diode 158 will be forward biased so that only the second brake control signal on line 157 will be coupled to brake valve driver 160 to immediately reduce brake pressure to relieve the deep skid being en countered. As brake pressure is reduced, wheel deceleration will correspondingly decrease. As wheel deceleration decreases, the magnitude of the second brake control signal on line 157 will also decrease. A point will eventually be reached where diode 169 is forward biased and diode 158 is back biased, thereby return ing control of brake pressure to the first brake control signal from integrating circuit 162.

It will be noted that the positive-value limits in positive-value, limiting circuits 166 and 154 are chosen to produce a signal from valve driver 160 corresponding to the zero pressure or closed position of the brake valve.

The positive-value limit in the positive-value, limiting circuit 142, on the other hand, is chosen to determine the maximum rate at which brake pressure may be removed by the corresponding portion of the large deviation control circuit.

Upon touchdown of the aircraft on the runway, it wil be remembered that full brake pressure is initially applied to the braked wheels, inasmuch as the basic deceleration control circuit takes a certain amount of time to reduce brake pressure to a level supportable by the existing runway conditions. As a result, a braked wheel is oftentimes forced into an initial skid which eventually will be corrected by the large deviation control circuit.

It is desirable to minimize the extent of such an initial skid, particularly in the case of aircraft, and for this purpose, the initial skid circuit 146 is provided so that brake pressure may be more quickly reduced than is possible given the operation of the large deviation control circuit alone.

Specifically, the output signal on line 143 from the positive-value, limiting circuit 142 in the large deviation sensing circuit 124 is supplied to the input of a pulse generator 148 in the initial skid circuit 146. Pulse generator 148 is of a type, more completely disclosed with reference to FIGURE 8, that provides a constant duration output signal on a line 149 in response to the first application of a signal to its input. The output signal on line 149 is coupled through a gain circuit 150 and a line 151 to a third input of the summing junction 138.

In operation, the first appearance of an output signal on line 143, which occurs when the wheel deceleration first exceeds the second reference deceleration established by the signal 82, triggers pulse generator 148 to provide its constant duration output signal. The magnitude of this output signal is adjusted by the gain circuit 150 (having gain constant K ) and is applied through summing junction 138 to the input of integrator 164, whereby the integrator 164 causes the first brake control signal on line 167 to increase at a greater rate than is commanded by the output signal on line 145 from the large deviation sensing circuit 124.Accordingly, brake pressure is more quickly reduced than would be the case if the large deviation control circuit alone were used, with such an increased rate of pressure reduction being in effect for the duration of the output signal from pulse generator 148. Once the output signal from pulse generator 148 has been terminated, pulse generator 148 is inhibited from providing another output signal for a period of time sufficient to allow the vehicle to be braked to a complete stop. Therefore, the initial skid circuit 146 is operative to increase the rate of brake pressure reduction only in response to the first skid that is encountered, and to thereafter be disabled until the vehicle has come to a complete stop.

With reference to FIGURE 8, the output signal from positive-value, limiting circuit 142 on line 143 (which is the second deceleradeceleration error signal) is applied to a summing junction 170. A bias signal -6,, is subtractively combined with the signal on line 143 in summing junction 170 and the output thereof is coupled by a line 171 to the input of a positive-value circuit 172. For negative input signals, circuit 172 provides a zero output signal, and for positive input signals, circuit 172 provides a fixed, positive-value output.The output signal from circuit 172 is coupled by a line 173 to a summing junction 174 and subtractively combined therein with a fixed bias voltage Vs The output from summing junction 174 is coupled by a line 175 to the set (S) input of an RS flip flop 176 whose Q output is connected to the output line 149 of the pulse generator 148 and therefore supplied to the gain circuit 150 (FIGURE 7). The Q output of flip flip 176 is coupled to the base of a transistor 178 whose emitter is coupled to reference potential and whose collector is coupled by a line 179 to one input of a summing junction 180, to one side of a capacitor C1, and to one side of a resistor Rl. The other sides of capacitor C1 and R1 are respectively connected to reference potential and to a supply voltage Ve.

The fixed bias voltage V is also supplied to the summing junction 180 and subtractively combined therein with the signal on line 179.

The output of the summing junction is coupled by a line 181 back to the reset (R) input of flip flop 176.

The output signal on line 149 of the pulse generator 148 is also fed back through a diode 182 to the input of a lag circuit 183 whose LaPlacian operator is represented as 1 ~ , , I + rS s The output of lag circuit 283 is coupled by a line 184 to the input of a gain circuit 185 (having gain constant K6), and the output of gain circuit 185 is coupled by a line 186 to the summing junction 170 and subtractively combined therein with the signal on line 143 from positive-value, limiting circuit 142.

In operation, the presence of a signal on line 143 signifies that the wheel deceleration has exceeded the second reference deceleration established by the signal 6, and therefore signifies that the braked wheel is in a skid.

The value of the bias signal 6,, is chosen to be some arbitrary, small amount so that summing junction 170 provides a positive output signal on line 171 only when the wheel deceleration has exceeded the second reference deceleration by a small amount. In response to the positive output signal on line 171, circuit 172 provides its fixed, positivevalue output whose magnitude is chosen to be greater than that of the bias voltage V,.

As a result, summing junction 174 provides a positive output signal on line 175 to place flip flop 176 in its set state whereby a positive output signal is provided on line 149 and therefore supplied to gain circuit 150 to cause brake pressure to be reduced as previously described. When the output signal on the Q output of flip flop 176 goes positive, the output signal on the Q output of flip flop 176 goes negative, thereby turning off transistor 178 (which has been previously maintained in a conducting condition by a previous positive output signal occurring on the Q output.

When transistor 178 turns off, a shunt provided thereby around the capacitor C1 is removed and capacitor C1 begins to charge to the supply voltage V at a rate determined by the time constant of capacitor C2 and resistor Rl. When the voltage across capacitor C1 equals and then exceeds the bias voltage Vs, the output from summing junction 180 becomes positive and is applied through line 181 to the reset input of flip-flop 176 to switch flip flop 176 to its reset state whereby the output signal on the Q output thereof is negative and the output signal on the Q output thereof is positive. At this time, the positive output signal on line 149 is removed.

It will therefore be appreciated that the time constant associated with capacitor C1 and resistor R2 determines the length of the pulse from pulse generator 148, and therefore determines the time during which brake pressure is being reduced at an increasing rate by the initial skid circuit 146.

Only positive output signals on line 149 are coupled through diode 182 to the input of the lag circuit 183. Lag circuit 183 essentially functions as a delay circuit with a very long time constant, established by the value of the constant rS, so that lag circuit 183 provides an output on line 184 for a very long period of time after the initial appearance of a positive output signal on line 149.

Preferably, r is chosen so that the output signal on line 184 persists for a period of time sufficient to allow the vehicle to come to a complete stop, e.g., two to three minutes.

The gain constant K6 in gain circuit 185 is chosen so that the output signal on line 186 from gain circuit 185 has a magnitude which is greater than any expected deceleration error during the time that brake pressure is being applied. Since the signal on line 186 is subtractively combined in summing junction 170 with the second deceleration error signal present on line 143, it will be seen that circuit 172 provides a zero output on line 173 until the vehicle has come to a full stop so as to inhibit flip flop 176 from again being set and to therefore inhibit the production of another positive output signal from pulse generator 148 until the vehicle has stopped.

The initial skid circuit 146 (FIGURE 7) increases the rate of brake pressure reduction by a fixed amount as determined by the gain constant K of gain circuit 150. It may be desirable in certain circumstances to modulate this rate of brake pressure reduction in accordance with the magnitude of the initial skid, that is, in accordance with the magnitude of the deceleration error initially encountered.

In such situations, the alternate embodiment illustrated in FIGURE 9 may be used.

The second deceleration error signal on the 143 is applied, as in the embodiment of FIGURE 7, to the gain circuit 144 (having gain constant K3), with the output of gain circuit 144 being coupled by line 145 to the summing junction 138. The second deceleration error signal on line 143 is also coupled to the input of a second gain circuit 144' (having a gain constant R3t), whose output is coupled by a line 145' to summing junction 138. A switch 190 is provided having normally closed contacts 190A interposed in line 145 and normally open contacts 190B interposed in line 145'.In situations where an initial skid is not encountered or where the initial skid has been compensated for, the gain circuit 144 is coupled to summing junction 138 by normally closed contacts 190A and therefore functions as part of the large deviation control circuit in the manner as previously described.

The second deceleration error signal on line 143 is also applied to the input of a pulse generator 148' which functions identically to pulse generator 148. The output of pulse generator 148' is counled by a line 149' to the innut of switch 190. Accordingly, when an initial skid is encountered, pulse generator 148' provides an output signal on line 149' for a predetermined period of time which causes switch 190 to open contacts 190A and close contacts 190B. It will therefore be appreciated that, during the time of the output signal from pulse generator 148', brake pressure reduction will be controlled by the portion of the circuit in FIGURE 9 including gain circuit 144' and that the rate of brake pressure reduction will be dependent not only upon the gain constant K.!, but also upon the magnitude of the second deceleration error signal.

It will also be recognized that switch contacts 190A may be eliminated, in which case the rate of brake pressure reduction in response to an initial skid will be denendent on both of the gain constants K2 and K,'.

When a limited-slip brake control system is used to modulate the brake in a hydraulic brake system for an aircraft in which the braked wheels are supported from the aircraft fuselage bv a landing gear strut assembly, the initial application of brake pressure causes the landing gear strut assembly, which is lightly damned, to move in a direction onnosite the direction of travel of the braked wheel along the runway. As a result of this movement, the braked wheel appears to slow down, resulting in a decrease in the measured wheel speed signal w,. The decrease in the measured wheel speed signal X provides a corresponding in the wheel deceleration signal on line 121 which will cause the limited-slip brake control system to reduce brake pressure.As brake pressure is reduced, the lightly damped landing gear strut assembly moves in the direction of travel of the braked wheel down the runway, producing an apparent increase in wheel speed which appears as a decrease in the wheel deceleration signal on line 121. The limitedslip brake control system will therefore cause brake pressure to increase so that the landing gear strut assembly again moves in a direction opposite the direction of aircraft travel.

The landing gear strut assembly oscillation therefore causes brake pressure to constantly oscillate upwardly and downwardly in an undesirable manner. One solution to this problem would be to introduce lag into the differentiator 120 so that the limited-slip brake control system is essentially nonresponsive to the frequency of landing gear strut oscillation.

However, since this frequency (typically 10Hz) is very low, the inclusion of such a lag in the differentiator 120 would significantly reduce the responsiveness of the limited-slip brake control system to actual changes in aircraft deceleration.

Therefore, the wheel deceleration signal on line 121 is applied to the dynamic compensation circuit 126 and specifically to the input of a double lead-double lag circuit 192 therein whose transfer function is represented by the LaPlacian operator s(l + r2S) (1 + r,3S) (1 + r4S) The output of circuit 192 is coupled by a line 193 to the input of a gain circuit 194 (having a gain constant Kl) and the output of gain circuit 194 comprises a compensating brake control signal which is coupled by line 195 to the summing junction 168.The constants , 2, r3, and r! of circuit 192 are chosen so that the output signal on line 193 has approximately a 90" phase lead for a range of frequencies in the wheel deceleration signal from zero up to and through the expected landing gear strut assembly oscillation frequency, and lesser decreasing phase lead for increasingly higher frequencies in the wheel deceleration signal. As an example, for an expected landing gear strut oscillation frequency of approximately 10Hz (about 64 radians per second), values of the constants 2, r3 and r4 of approximately 30 radians per second, 90 radians per second, and 280 radians per second, respectively, will produce the required 90" phase lead.As a result, the output signal on line 193 (and thus the compensating brake control signal on line 195) will be increasing and decreasing at the frequency of landing gear strut assembly oscillation represented in the wheel deceleration signal, but 90e ahead of such oscillation. Those skilled in the art will appreciate that the intergrating circuit 162 will introduce a substantially 900 phase lag to the landing gear strut assembly oscillation present in the wheel deceleration signal, and that the first brake control signal on line 167 will accordingly be increasing and decreasing at the frequency of landing gear strut assembly oscillation but delayed therefrom by 900.

Accordingly, the summation afforded by the summing junction 168 effects cancellation of the changes in the first brake control signal occasioned by landing gear strut assembly oscillation, provided that the value of the gain constant K1 of gain circuit 194 is properly chosen.

The improved limited-slip brake control system of the present invention operates by measuring the wheel speed of a single braked wheel and by providing a control signal to control the brake pressure applied to the brakes for that wheel. In the case where the vehicle has a plurality of braked wheels, it will be necessary to provide a plurality of systems as illustrated in FIGURE 7, one for each braked wheel, and to provide some means for energy balance between the braked wheels, such as the passive and active methods of energy balance previously described.

While the invention has been described with respect to a preferred embodiment, it will be clearly understood by those skilled in the art that the invention is not limited thereto, but that the scope of the invention is to be interpreted only in conjunction with the appended

Claims (24)

claims. WHAT WE CLAIM IS:

1. A limited-slip brake control system for controlling the brake force to be developed by a braking means for a wheel of a vehicle, said limited-slip brake control system comprising: means providing a wheel deceleration signal related to the measured deceleration of the wheel; a basic decelerating control circuit responsive to said wheel deceleration signal and providing a brake control signal for application to the braking means, said brake control signal being adapted to increase and decrease brake force at a predetermined rate to maintain the average value of said measured wheel deceleration at a first reference deceleration representing a level of brake force generally lying in a nonslip portion of a characteristic Mu/slip curve for the vehicle;; a large deviation control circuit responsive to said wheel deceleration signal for modifying said brake control signal to decrease brake force at a rate greater than said predetermined rate for as long as said measured wheel deceleration exceeds a second reference deceleration higher than said first reference deceleration and representing a level of brake force generally lying in the slip portion of said characteristic Mu/slip curve, said greater rate at which said brake force is decreased being related to the amount by which said measured wheel deceleration exceeds said second reference deceleration; and, means for applying said brake control signal to the braking means.

2. A limited-slip brake control system as recited in Claim 1, wherein said basic deceleration control circuit includes: a deceleration select circuit for providing a first reference deceleration signal representative of said first reference deceleration; means for subtractively combining said wheel deceleration signal and said first reference signal to produce a first deceleration error signal; a hysteresis circuit responsive to said first decelertation error signal for producing a fixed-level output signal generally having a first polarity representing a decrease in brake force when said measured wheel deceleration exceeds said first reference decleration, and generally having a second polarity representing an increase in brake force when said reference deceleration exceeds said measured wheel deceleration, said output signal switching between said first and said second polarities only when the difference between said measured wheel deceleration and said first rference deceleration exceeds a predetermined amount representing an incremental deceleration less than said first reference deceleration; and, means for integrating said output signal to produce said brake control signal.

3. A limited-slip brake control system as recited in Claim 1, wherein said basic deceleration control circuit includes: a deceleration select circuit for producing a first reference deceleration signal representative of said first reference deceleration; means for subtractively combining said wheel deceleration signal and said first reference deceleration signal to produce a first deceleration error signal; threshold circuit means for producing a trigger signal when said first deceleration error signal exceeds a predetermined level representing an incremental deceleration less than said first reference deceleration; singleshot multivibrator means for producing a constant level, constant duration output pulse in response to said trigger signal; and, means for integrating said output pulse to produce said brake control signal to decrease brake force at a rate dependent on the level of said output pulse for the duration of said output pulse, and to gradually increase brake force in the absence of said output pulse.

4. A limited-slip brake control system as recited in Claim 1, wherein said large deviation control circuit includes: a source of a second reference deceleration signal representative of said second reference deceleration; means subtractively combining said wheel deceleration signal and said second reference deceleration signal to produce a second deceleration error signal; means responsive to said second deceleration error signal to modify said brake control signal to cause an exponential decrease in brake force below the level controlled by said basic deceleration control circuit when said measured wheel deceleration exceeds said second reference deceleration, and to cause an exponential increase brake force back to the level controlled by said basic deceleration control circuit when said second reference deceleration exceeds said measured wheel deceleration.

5. A limited-slip brake control system as claimed in Claim 4, wherein said second reference deceleration is a predetermined amount above said first reference deceleration.

6. A limited-slip brake control system as recited in Claim 1, wherein said large deviation control circuit includes: a source of a second reference deceleration signal representative of said second reference deceleration; means subtractively combining said deceleration signal and said second reference deceleration signal to produce a second deceleration error signal; means responsive to said second deceleration error signal to produce an output signal representing a decrease in brake force and having a level proportional to the amount by which said measured wheel deceleration exceeds said second reference deceleration; and, means for integrating said output signal to modify said brake control signal to cause brake force to decrease below and to increase back to the level controlled by said basic deceleration control circuit when said measured wheel deceleration exceeds ,-aid second reference deceleration at rates dependent upon the level of said output signal.

7. A limited-slip brake control system ss recited in Claim 6, wherein said second reference deceleration is a predetermined amount above said first reference deceleration.

8. A limited-slip brake control system as recited in Claim 1, wherein said basic deceleration control circuit includes means responsive to said wheel deceleration signal for providing a first output signal generally having either a fixed positive level or a fixed negative level, respectively depending on whether said measured wheel deceleration is greater or lesser than said first reference deceleration, said first output signal switching between said positive and negative levels only when th- difference between said measured wheel de celeration and said first reference deceleration exceeds a predetermined amount representing an incremental deceleration less than said first reference deceleration; wherein said large deviation control circuit includes means responsive to said wheel deceleration signal for providing a second output signal as long as said measured wheel deceleration exceeds said second reference deceleration, said second output signal having a positive level determined by the amount that said measured wheel deceleration exceeds said second reference deceleration; and further comprising means summing said first and second output signals to provide a composite output signal, and integrating circuit means for integrating only positive levels of said composite output signal to provide said brake control signal which increases and decreases at a rate determined by the level of said composite output signal.

9. A limited-slip brake control system as recited in Claim 1, for controlling the brake force to be developed by first and second braking means for first and second wheels of the vehicle, and further comprising: means for averaging the measured wheel decelerations of the first and second wheels to provide said wheel deceleration signal; and, means for providing said brake control signal to the first and second braking means.

10. A limited-slip brake control system as recited in Claim 1, for controlling the brake force to be developed by first and second braking means for first and second wheels of the vehicle, said limited-slip brake control system further comprising: first and second means providing first and second wheel deceleration signals respectively related to the measured wheel deceleration of the first and second wheels; first and second basic deceleration control circuits, and first and second large deviation control circuits, respectively operable to provide first and second brake control signals; first means for applying said first brake control signal to the first braking means and second means for applying said second brake control signal to the first braking means and second means for applying said second brake control signal to the second braking means; and, an energy balance system for modifying said first and said second brake control signals to equalize the brake energy expanded by the first and second braking means.

11. A limited-slip brake control system as recited in Claim 10, wherein said energy balance system comprises: means providing first and second wheel speed signals respectively related to the measured rotational speeds of the first and second wheels; means providing first and second brake torque signals respectively related to the measured brake forces developed by the first and second braking means; first and second energy balance circuits respectively associated with the first and second braking means, for respectively modifying said first and second brake control signals, each of said first and second energy balance circuits including: first multiplier circuit means receiving said first wheel speed signal and said first brake torque signal and producing therefrom a first brake energy signal; second multiplier circuit means receiving said second wheel speed signal and said second brake torque signal and producing therefrom a second brake energy signal; a first integrator circuit means coupled to receive said first brake energy signal and producing therefrom a first brake energy integral signal; a second integrator circuit means coupled to receive said second brake energy signal and producing therefrom a second brake energy integral signal; means subtractively combining said first and said second brake energy integral signals to produce a brake energy error signal whose polarity is representative of the difference between said first and second brake energy integral signals; a limiting circuit responsive to said brake energy error signal to provide a constant level output signal only when said brake energy error signal has a polarity representing a lesser brake energy in the associated one of the first and second braking means; and, third integrating circuit means responsive to said output signal for modifying the associated one of said first and second brake control signals to cause an increase in brake force in said associated one of the first and second braking means.

12. A limited-slip brake control system as recited in Claim 1, further comprising a selected deceleration control circuit connected between said large deviation control circuit and said basic deceleration control circuit for decreasing said first reference deceleration for a time related to the time that said measured wheel deceleration exceeds said second reference deceleration.

13. A limited-slip brake control system as recited in Claim 12, wherein said selected deceleration control circuit includes: means providing a fixed level output signal whenever said measured wheel deceleration exceeds said second reference deceleration; a lag circuit responsive to the presence and absence of said fixed level output signal to respectively provide an exponentially increasing and decreasing output signal; and, means responsive to said exponentially increasing and decreasing output signal to respectively decrease and increase said first reference deceleration.

14. A limited-slip brake control system as recited in Claim 1, further comprising an initial skid circuit for minimizing the extent of an initial skid encountered upon the initial application of brake force by the braking means, said initial skid circuit being connected with said large deviation control circuit and operative to further decrease brake force at the first time that said wheel deceleration ex ceeds said second reference deceleration and for a period of time thereafter substantially equal to the respective duration of the initial skid.

15. A limited-slip brake control system as recited in Claim 14, wherein said initial skid circuits includes: a pulse generator providing a constant-duration, pulse generator output signal only when said wheel deceleration initially exceeds said second reference deceleration; and, means for integrating said pulse generator output signal to modify said brake control signal to cause brake force to decrease below the level controlled by said basic deceleration control circuit at a rate dependent upon the level of said pulse generator output signal.

16. A limited-slip brake control system as recited in Claim 1, further comprising: means responsive to said wheel deceleration signal for providing a second brake control signal as long as said measured wheel deceleration exceeds a third reference deceleration higher than said second reference deceleration, said second brake control signal having a level representing a decrease in brake force which is determined by the amount that said measured wheel deceleration exceeds said third reference deceleration; and, means for applying said second brake control signal to the braking means in pdace of said brake control signal whenever said second brake control signal represents a greater reduction in brake force than does said brake control signal.

17. A limited-slip brake control system as recited in Claim 1, further comprising a dynamic compensation circuit responsive to said wheel deceleration signal for compensating for undesirable low-frequency oscillation occurring in said measured wheel deceleration, said dynamic compensation circuit providing a compensating brake control signal representing a decrease in brake force and having a 900 or greater phase lead for a range of frequencies in said measured wheel deceleration surrounding the expected frequency of said undesirable, low-frequency oscillation, and a lesser phase lead for all other frequencies in said measured wheel deceleration; and, means additively combining said compensating brake control signal with said brake control signal to compensate said brake control signal for oscillation therein resulting from said undesirable, low-frequency oscillation.

18. A limited-slip brake control system as recited in Claim 17, wherein said dynamic compensation circuit includes a double-lead double-lag circuit.

19. A method for controlling the brake.

pressure applied to a brake means for a wheel of a vehicle, comprising the steps of: causing said brake pressure to increase and decrease at a predetermined rate to produce a wheel deceleration whose average value is substantially equal to a first reference deceleration representing a level of brake force to be developed by the brake means which generally lies in a nonslip portion of a characteristic Mu/slip curve for the vehicle; and, causing brake pressure to be reduced at a rate greater than said predetermined rate as long as said wheel deceleration exceeds a second reference deceleration representing a level of brake force generally lying in the slip portion of said characteristic Mu/slip curve, said greater rate of brake pressure reduction being related to the amount by which said wheel deceleration exceeds said second reference deceleration.

20. A method as recited in Claim 19, further comprising the step of causing brake pressure to be reduced in inverse proportion to the amount by which said wheel deceleration exceeds a third reference deceleration, said third reference deceleration being greater than said second reference deceleration, as said wheel deceleration exceeds said third reference deceleration.

21. A method as recited in Claim 19, comprising the additional step of further increasing the rate of brake pressure reduction for a predetermined period of time subsequent to the first time that said wheel deceleration exceeds said second reference deceleration after the application of brake pressure to said braking meres

22. A method as recited in Claim 18, comprising the additional step of decreasing said first reference deceleration for a period of time related to the time that said measured wheel deceleration exceeds said second reference deceleration.

23. A limited-slip brake control system substantially as described hereinbefore with reference to and as illustrated in the accompanying drawings.

24. A method for controlling the brake pressure applied to at brake means for a wheel of a vehicle substantially as described hereinbefore with reference to the accompanying drawings.