DE69127786T2 - Active elevator suspension system - Google Patents

Description

The invention relates to elevators, in particular to improved ride quality.

Conventional elevator system suspensions can be characterized by the mechanical properties of the transmitting elements that connect three main elevator components: the car platform, the support frame and the guide rails. The conventional elevator car platform is typically attached to the support frame by means of hard rubber cushions. The support frame, in turn, runs along the guide rails and is supported by either hard-sprung wheels or by sliding wedges at four attachment points.

The movement of the car platform in these conventional elevator systems is influenced by forces acting directly on the car, e.g. reaction forces due to passenger movement or wind forces, and by forces acting indirectly, particularly from guide rail irregularities, e.g. butt joint misalignments or undulations due to building subsidence. These conventional elevator suspension systems can be classified as "passive" in the sense that no energy is applied to the suspension system to counteract the direct or rail induced forces. In passive systems there is a system-specific compromise in ride quality. Rigid transmission elements mitigate the effects of direct car forces, while compliant transmission elements (low stiffness) mitigate the effects of guide rail irregularities.

In US Patent 4,899,852 to Salmon et al. a passive suspension configuration with a mechanically compliant attachment between the car platform and the frame is disclosed. The mechanically compliant attachment is realized by suspending the car platform from the frame by means of long steel rods. This elevator configuration, hereinafter referred to as a "pendulum car", is a passive construction in which the attenuation of the influences of the Rail irregularities are maximized at the expense of increased sensitivity to direct car forces. In a disclosure concerning a non-pendulum car, U.S. Patent 4,754,849 to Hiroshi Ando, electromagnets are shown arranged outside the car symmetrically with respect to the guide rails in a control system which uses opposing forces from the electromagnets to keep the car stable, the rails serving as the required ferromagnetic mass, but the rails are not used as a straight reference line, but instead a cable stretched between the ceiling and floor of the elevator shaft serves as the straight reference line. The position of the car with respect to the cable is controlled by means of detectors in a control loop system. Serious questions arise as to whether such a cable can be successfully used as a reliable straight guide, particularly in tall buildings.

In another disclosure relating to a non-pendulum car, U.S. Patent No. 4,750,590 to Matti Otala, there is shown what appears to be a substantially open loop control system with solenoid operated guide shoes which makes use of the concept of storing the deviations from the straightness of the guide rails in a computer memory for later sensing the position of the car in the hoistway with a view to retrieving the relevant information from the memory and correcting the guide rail shoe positions accordingly. An acceleration sensor is mentioned in claim 6, but does not appear to be otherwise disclosed as to its purpose in the description or drawing therein. It may be for determining the acceleration of the car within the hoistway. Such an acceleration signal would presumably be necessary to determine which data point to retrieve from the memory as stated in claim 2. The Ottala approach suffers from the problem of changes in the deviations from the straightness before a correction run is made. and the accuracy of the stored information can be brought into line with the current position of the car.

A suspension arrangement for a pendulum-type or suspended car is shown in U.S. Patent 4,113,064 to Shigeta et al., where the car is suspended within an external car frame at the top thereof by means of a plurality of rods coupled to the floor of the car. A plurality of stabilizing stops are located between the bottom of the suspended car and the floor of the car frame. Each stop includes a cylinder extending downward from the bottom of the suspended car to enclose a rubber torus located on a rod raised from the floor of the car frame. The clearance between the cylinder and the suspended car is sufficient to allow movement, but not large enough to allow the suspended car to strike the car frame. Another embodiment using a "cushion" device incorporating ball bearings allows movement in any direction within the horizontal plane.

Another approach is disclosed by Luinstra et al in U.S. Patent 4,660,682, wherein a pair of parallel rails are arranged horizontally in a parallelogram between the suspended car and the car frame, with followers arranged to roll or slide on the rails such that the suspended car can move in any horizontal direction relative to the car frame.

The last two pendulum or supported cabin designs discussed use passive restraints for a motion that is inherently reactive rather than active.

AU-A-41565/89 by Kahkipuro (from which the two-part claim 1 is formed) discloses a procedure for damping vertical vibrations of an elevator car which is suspended by means of elastic suspension elements comprising the steps of measuring acceleration and using the output of the accelerometer to control at least one vibration damper in the vertical direction. A high pass filter serves to block the passage of low frequency vertical accelerations which are part of the usual starting and stopping accelerations of the elevator. Only vibration frequencies above 1 hertz pass through the high pass filter to be effectively absorbed by the damper. On page 8 of Kahkipuro, lines 24 to 34, it is stated that vibrations in all three dimensions can be suppressed, but that for the other dimensions (which are not disclosed in detail) transducers for measuring acceleration in the other two dimensions do not need to be filtered. The Kahkipuro disclosure does not go into further detail as to the purpose of measuring acceleration in the horizontal direction, nor as to the type of signal conditioning required for the horizontal plane. In fact, Kahkiputo teaches that no horizontal plane signal conditioning, not even filters, is needed. But even if a purpose were given, no teaching emerges on the problem of how to use an acceleration controller as a soft spring against vibrations while at the same time creating a hard spring for the purpose of leveling or centering. Both of these problems must be addressed in order to successfully control vibrations in the horizontal plane.

Active suspension systems are well known in automotive engineering. Tunable impedances are used in particular in the form of so-called "tunable shock absorbers". They contain a device that undergoes a relative displacement and, from the point of view of the "system", consists of a mechanical impedance (here defined as the frequency-dependent ratio of deflection to an applied force) of a stiffener in parallel with a damper. The stiffening and damping elements are adjusted during different conditions adjusted. For example, during cornering, as detected by accelerometers, increased stiffening of selected shock absorbers is desirable. Similarly, during braking, the two front shock absorbers are stiffened. This is done by software by detecting the displacement of the vehicle in relation to the chassis and setting a target offset. There is an opposite effect to simply adjusting stiffening and damping: if the mechanical impedance of the shock absorbers is increased, the vehicle becomes more sensitive to a bumpy road. If the mechanical impedance of the shock absorbers is decreased, the vehicle becomes more sensitive to direct forces other than the bumpiness.

In our study of improved ride quality for elevators, we compared the frequencies of disturbances caused by rail bumps in a pendulum car with frequency distributions for direct forces and found that, at least for a pendulum car, there is a critical region between 2 to 10 Hz where we could not simultaneously satisfy the desire for reduced mechanical impedance to eliminate the severe bumps and our desire for increased mechanical impedance to mitigate direct forces. At least for a pendulum car, this problem limits to a very considerable extent the effectiveness of the impedance-tunable active suspension used in automobiles.

Some specific examples of active suspension systems of interest are given below:

US Patent 4,809,179 by Klinger et al

Klinger et al discloses an acceleration sensor 26 that feeds a microprocessor which in turn appears to control an actuator for a motor vehicle suspension unit.

US Patent 4,892,238 to Kurtzman et al

Kurtzman et al disclose an active suspension system for controlling the orientation of the chassis of a motor vehicle with respect to the frame. Figure 2 shows an accelerometer feedback signal between a control arm processor 20 and a control arm assembly 10 between each wheel 14 and the chassis, i.e., the frame 12 of a motor vehicle.

US Patent 4,621,833 by Soltis

Soltis discloses an acceleration sensor 16 in Fig. 2 that provides a signal to a suspension control module for a multi-stable suspension.

US Patent 3,939,778 by Ross et al

Ross et al disclose a lateral accelerometer 40' in Fig. 6, which has interconnect components insertable at the Z-Z inputs in Fig. 4 to obtain lateral stability. Reference is also made to Fig. 4 for an illustration of the electromagnetic side frame 1 and a ferromagnetic side frame 2 as part of a block diagram of a complete electrical control system with an active suspension for a rail vehicle according to Fig. 2.

US Patent 4,625,993 to Williams et al

Williams et al show control signals that can be modified by signals representing vehicle speed as well as lateral and longitudinal acceleration.

US Patent 3,871,309 to Kolm et al

Kolm et al disclose the active damping of vibrations of a magnetically lifted vehicle using inertial and position sensors on the vehicle.

US Patent 4,770,438 to Sugasawa et al

Sugasawa et al disclose an automotive suspension control system with vibration sensors for detecting road surface conditions, as well as corresponding countermeasures.

US Patent 4,215,403 by Pollard et al

Pollard et al disclose an active suspension for a vehicle using an accelerometer 3.

US Patent 4,909,535 to Klag et al

Klag et al disclose an "active" suspension system between the body and wheel of a vehicle. A high gain closed loop position-velocity control system is disclosed.

US Patent 4 898 257 by Brandstatter

Brandstatter discloses an active hydropneumatic suspension system for a heavy combat vehicle using vertical accelerometers 188a, 188b, 188c and 200 as shown in Fig. 3.

NASA TECH BRIEFS, July 1990 discloses a "Flux-Feedback Magnetic-Suspension Actuator" in which the flux density is kept substantially constant and Hall effect devices serve as sensors for an electronic feedback circuit which controls currents that in the electromagnetic windings to maintain the flux binding the suspended element at a substantially constant value, independent of changes in the gap width. Reference is made to further information found in NASA - TEM 100672.

One aim of the present invention is to improve the ride quality of elevators.

As used herein, the term "cab" refers to a passenger platform that is suspended or movably supported within an external frame. "Car" refers to a passenger platform that is not free to move within a support frame or on another support platform, or "car" is alternatively used to refer to a frame for a movably supported or suspended car platform (e.g., "car frame"). According to the invention, there is provided an elevator car with a device for stabilizing the car, the device comprising: an accelerometer responsive to a horizontal acceleration of the car for providing a measurement signal having a magnitude corresponding to the horizontal acceleration; and a control device responsive to the measured acceleration signal for controlling an actuator device for setting the car in horizontal motion against the acceleration, characterized in that the device further comprises: a position sensing device responsive to the degree of centering of the car for providing a sensing signal whose magnitude corresponds to a horizontal position of the car; and that when the acceleration signal is applied to the actuator device, the control device is also responsive to the signal from the position sensing device to thereby compensate for components of the acceleration signal which are not representative of the actual horizontal acceleration acting on the car and to compensate for forces acting on the car to decenter it.

According to an embodiment of the present invention, the controller is responsive to the read signals to process the read signals to drive the actuator with a frequency response tailored to provide a stable, drift-free, high performance system.

According to a further embodiment of the present invention, at least in shuttle car embodiments, in those zones where car motion is to be minimized, the loop gain is set to a value greater than 1. At high frequencies, the loop gain is reduced to meet safe stability requirements. At low frequencies, the loop gain is "washed out" to reduce the effects of sensor noise and drift.

The active elevator suspension system disclosed here is a combination of sensors, a controller and an actuator applied to one or more axes to minimize the effects of rail-induced disturbances and direct car forces.

An elevator traveling up and down a hoistway is subject to disturbances due to rail irregularities, the frequency content of which is not dissimilar to disturbances caused by other forces we call direct forces. Unfortunately, at least in a pendulum car, and unlike automobiles, the frequency content of the rail-induced disturbances is in the same range as frequencies simultaneously encountered in disturbances caused by direct forces. Direct forces can be most effectively counteracted by high mechanical impedance, while rail irregularities can be effectively counteracted by low mechanical impedance. In automobiles, there are well-defined demonstrable modes of operation, for example, cornering, acceleration and deceleration, which can be effectively counteracted using a tunable impedance as explained above. In our method, according to the teaching of the invention, we directly generate forces depending on the acceleration sensed, and only the restoring force of the pendulum controls the relative displacement between the car and the frame or elevator shaft. The same method applies to the active control of conventional elevator cars.

According to a further embodiment of the present invention, in an embodiment for conventional elevator cars the actuators can be arranged to counteract horizontal transverse forces acting on the car moving in the elevator shaft, while in non-conventional suspended (pendulum) or supported car embodiments they are arranged to counteract horizontal forces acting on the suspended or supported car moving in the car frame as the frame moves in the elevator shaft. When such a concept is applied to a conventional elevator car, only four active actuators are needed near the bottom of the car, two each for setting in motion relative to rails on opposite sides of the car. Four conventional or passive guides could additionally be used near the top of the car. Such an arrangement could be advantageously used, for example, with a non-conventional "V" shaped rail, as disclosed for other purposes, for example, by Charles R. Otis in U.S. Patent 134,698 (issued January 7, 1873). Similarly, if such a concept is used in a pendulum or supported car, it could, without limitation, require only four actuators by employing an active actuator assembly at or near the bottom of the car to move the car frame, using, again without limitation, conventional rails to guide the frame.

According to yet another embodiment of the invention, the actuators can be arranged to counteract rotational forces about the vertical acting on the conventional car in a hoistway or on a car suspended or supported by a car frame. Furthermore, if such a concept is used to control a conventional car in a hoistway, this will also only require four actuators, using the same new rail form for active control. Similarly, if such a concept is used on a pendulum or supported car, this will only require four actuators using an active actuator arrangement and, without limitation, the use of conventional rails to guide the frame.

In yet another embodiment of the invention, the actuators may be arranged to counteract rotational forces acting on a conventional elevator car or a car suspended or supported on a frame about one or more axes in a horizontal plane. Such axes may, but need not, be defined for control purposes as orthogonal axes in such a horizontal plane, possibly parallel to the elevator shaft walls. In this case, multiple actuators may be used on the ceiling and floor of the elevator car, with those on the ceiling controlling horizontal accelerations of the ceiling while those on the floor controlling horizontal accelerations of the floor. By independently controlling horizontal accelerations on the ceiling and floor, rotations are automatically taken into account. If the four-actuator method described above is implemented for a non-pendulum car (in connection with the control of horizontal translations and vertical rotations), this may (but does not have to) use only eight actuators (four above and four below) using a new rail shape for active control. If such a concept is to be used for a Similarly, if the system is implemented as a pendulum- or frame-supported cabin (in conjunction with a control of horizontal translations and vertical rotations), it may, but need not, use only eight actuators, placing four on the ceiling of the cabin and four on its floor.

According to yet another embodiment of the present invention, the actuators may be of the electromagnetic type.

According to yet another embodiment of the invention, the actuators can be electromechanical, e.g. solenoid-operated wheels.

According to a further embodiment of the invention, four electromagnetic actuators are used, each of which operates along an axis which, in the case of a car which is not designed as a pendulum or not supported by a frame, runs such that forces act at an angle of 45 degrees with respect to a shaft wall, e.g. opposing shaft walls equipped with rails, while in the embodiment of the suspended car the axis runs such that forces act along axes which form an angle of 45 degrees with the planes of the walls of the suspended or supported car.

The present embodiment teaches, among other things, that accelerometers, rather than position sensors, are most advantageously used in a closed loop to actively control an elevator car or cabin. It also teaches that in a car that is guided by rails attached to shaft walls, the twelve electromagnets for controlling horizontal translations of an elevator car according to Ando can be replaced by a smaller number of actuators. According to a preferred embodiment of an aspect of the present invention, four actuators are sufficient to achieve such translational forces in the horizontal plane. As a further teaching, the same four actuators can also be used to control torsional forces about the vertical. In the non-pendulum cabin embodiment, although conventionally designed rails similar to the Ando configuration can be used, a new rail design can be advantageously applied to an active system based on acceleration feedback, where four or more actuators can be properly arranged to control translational forces in the horizontal plane. And also, in a similar manner, as a further teaching, the same actuators can be used to control torsional forces about the vertical.

These approaches have the additional advantage of greatly simplifying the design. By using accelerometers, there is also no need to use the Andor cable, which is potentially subject to straight-line-destroying forces due to numerous factors such as fluctuations, expansions and contractions of the building due to temperature changes, vibrations due to air currents in the elevator shaft, and other causes. Such a structure can be replaced, according to a preferred embodiment of the invention, by accelerometers which serve to provide signals which are also indicative of position in a closed loop control system.

Although we teach that a position control system based on an accelerometer output is a superior method, we also recognize that accelerometers are subject to drift, which we teach to correct according to the invention based on a slow control loop for controlling the mean car or cab position with respect to a fixed reference.

Thus, in accordance with one aspect of the present invention, a preferred embodiment comprises a relatively fast, simple, analog control loop responsive to accelerometers with one or more relatively slower, but more accurate digital control loops responsive to position or acceleration sensors or both types of sensors.

Another preferred embodiment of the invention arose from the need to improve the combined strength-to-weight ratio of the guide rails without compromising their control characteristics. This embodiment of the invention achieves this goal by using a "Y"-shaped rail in cross-section instead of the usual "T"-shaped guide rail

The guide rail may form an interface with three wheels attached to the elevator car, the three wheels rolling on surfaces comparable to those of the "T" shape, namely on two opposite sides of the base, but with the bearing surface passing through the top of the "Y" and through the legs of the "Y" forming a "V", rather than flat and vertical "at the top" of the "T".

This can be implemented in the form of two or three actively controlled wheels or other actuators such as electromagnets using the upright section of the "T" or the upper part of the "Y" as ferromagnetic masses.

The Y-shape has better cross-sectional properties than the corresponding T-shape, yet weighs considerably less. This weight saving is achieved while maintaining the cross-sectional properties by spreading the material out in the two legs of the "Y". The T-section has more material below the web, which is concentrated near the centerline and thus contributes less to the area moment of inertia with respect to the centerline.

In the Y-shaped, multi-web rail, the material below the web is displaced into the two legs of the "Y" and further away from the centerline. These contribute more to the area moment of inertia with respect to the centerline. Consequently, a Y-shape with the same cross-sectional properties is lighter than the corresponding T-shape.

Appropriately designed "Y"-shaped guide rails can replace the T-shapes, achieving a weight saving of, for example, ten to twenty percent (10% to 20%).

As suggested above, the passive limiters used by Shigeta et al are not as effective as the present invention in that they do not actively counteract the undesirable translational forces to which the car is subjected and thus do not provide as smooth a ride for passengers as is achieved by the present invention. Furthermore, they do not actively counteract the undesirable twisting forces to which the car is subjected and thus, for similar reasons, do not provide as smooth a ride for passengers as is achieved by the present invention. And certainly they do not provide for passive limiters?? and certainly not for active countermeasures of any kind with respect to the axes of twist other than the vertical as disclosed here.

These and other objects and advantages of the present invention will become more apparent in light of the following detailed description, which is given by way of example only with reference to the accompanying drawings, in which:

Fig. 1 is a block diagram of an active control system for an elevator car or cabin;

Fig. 2 is a representation of a car or cabin of an elevator with a coordinate system shown;

Fig. 3 shows the coordinate system of Fig. 2 in greater detail;

Fig. 4 to 7 different active rail configurations equipped with multiple webs;

Fig. 8 shows a prior art active rail configuration;

Fig. 9 to 13 different active rail configurations equipped with multiple webs;

Fig. 14A is a plan view of a shuttle car shown from the side in Fig. 14B.

Fig. 14B is a side view of the shuttle car of Fig. 14A with an active suspension shown;

Fig. 14C is a sectional plan view of the floor of an elevator car with a five degree of freedom platform suspended or supported on a frame which in turn is suspended by a cable or supported by a piston, showing a new actuator arrangement (which may be similarly constructed on the ceiling of the car) which uses only four electromagnets between the bottom of the suspended or supported car and the floor of the frame;

Fig. 14D is another similar illustration (as in Fig. 14C) of a group of electromagnetic actuators located between the underside of the suspended or supported cabin and the floor of the car frame, except that six electromagnets are used here;

Fig. 15 is a plan view of the bottom (the top may be similar) of an elevator car platform with an active control using "V"-shaped or triangular-shaped rails;

Fig. 16 is an illustration of a signal processor used as the means 20 shown in Fig. 1 for determining the amount of response necessary to counteract noise;

Fig. 17 is an illustration of a series of steps that may be performed by the processor of Fig. 17 or its equivalent to determine the amount of response needed to counteract disturbances;

Fig. 18 is a mathematical abstract block diagram of a preferred control arrangement for carrying out the active control of Fig. 1 with an inner loop with acceleration feedback and slower outer loops with position and acceleration feedback in accordance with the present invention;

Fig. 19 shows concrete means for implementing the abstracted control of Fig. 18 for the control of Fig. 14C, in which a fast, analog control is used in the inner loop and a relatively slow, but more accurate, digital control in the outer loop;

Fig. 20 the analog control in Fig. 19 in more detail;

Fig. 21 shows a power controller;

Fig. 22 an ignition board for controlling controlled silicon rectifiers in the power regulators according to Fig. 21;

Fig. 23 the concept of non-linearity and offset in an exaggerated form;

Fig. 24 shows the theory of operation according to Fig. 18 in a reduced block diagram representation;

Fig. 25 an even more reduced model, which is valid for all frequencies except the lowest frequencies;

Fig. 26 a graphical representation of the current as a function of the air gap;

Fig. 27 a graphical representation of the power as a function of the air gap;

Fig. 28 a graphical representation of the time constants as a function of the air gap;

Fig. 29 shows a pair of coils for use with the core shown in Fig. 30;

Fig. 30 a laminated core for use with the coils according to Fig. 29;

Fig. 31 is an illustration of a single-axis side vibration stabilization system as may be used with the actuator arrangement of Fig. 14D;

Fig. 32 shows the force summation method used in the system of Fig. 31;

Fig. 33 shows a negative rectifier inverter as used in the system of Fig. 31;

Fig. 34 shows a positive rectifier inverter as used in the system of Fig. 31;

Fig. 35 shows an FC-controlled clamp circuit as used in the system of Fig. 31;

Fig. 36A is another system level diagram of a specific implementation of an active suspension system for a special pendulum cabin (according to US-A 4 899 852);

Fig. 36B is a qualitative illustration of design requirements for the total loop gain for a specific shuttle car (e.g., as described in US-A-4,899,852);

Fig. 37A and 37B are graphical representations of the open loop gain and the phase angle of the product (L(s) of C(s) and G(s) in Fig. 36A);

Fig. 38A and B are graphical representations of the feedback control gain and phase angle for that particular shuttle car design;

Fig. 39 to 41 are a summary of the results of a simulation study of an active suspension system using the compensator of Fig. 36A and 36B, 37A and 37B and 38A and 38B;

Fig. 42 is an illustration of a digital controller that can be used to implement the controller of Fig. 36A;

Fig. 43 is an illustration of a series of steps that may be performed by the processor of Fig. 42;

Fig. 44 shows the results of a test to verify the effectiveness of the active system of Fig. 36A for a pendulum car in mitigating direct car forces, illustrating a ratio of the measured car acceleration to the magnitude of a sinusoidal input force in a frequency sweep range;

Fig. 45 a comparison of the predicted temporal response (by simulation) and the achieved response for the direct pendulum car force attenuation;

Fig. 46 the response of the pendulum cabin system to a rail irregularity, simulated on a test bench with a rotating unbalance;

Fig. 47 is a detailed schematic representation of a circuit for controlling one of the electromagnets according to Fig. 14D in dependence of a force command signal 106 from a signal processor, as shown in Fig. 42, for example;

Fig. 48 shows a preferred device for carrying out the preferred control scheme of Fig. 15;

Fig. 49 is a representation of an active guidance system comprising three wheels;

Fig. 50 shows a solenoid-operated wheel for use in the active system, for example as shown in Fig. 49;

Fig. 51 Steps that can be performed using actuators to bring a suspended or supported platform to rest on a threshold;

Fig. 52 is a front sectional view of a standardized and in practice introduced guide rail with a "T"-shaped cross section, with its interfaces shown with respect to three wheels;

Fig. 53 is an end sectional view of the "Y"-shaped cross-section guide rail that can be used in active elevator systems, showing its interface with three wheels;

Fig. 54 is a front sectional view of the cross-sectionally "Y"-shaped guide rail according to Fig. 52 with two electromagnetic actuators;

Fig. 55A, 55B and 55C are detailed comparison sketches with tables 1 and 2 for comparing the characteristic features of the T-shaped guide rail according to Fig. 55A and the Y-shaped guide rail according to Fig. 55B, with definitions for the elementary parts and dimensions of the two shapes to be compared being given in the upper part, in conjunction with two corresponding tables below the illustration and below that again a legend of definitions used in the tables;

Fig. 56 is a perspective view of a guide roller group;

Fig. 57 is a side elevational view of the guide roller assemblies of Fig. 56, showing details of the lateral roller adjustment mechanism for the secondary suspension;

Fig. 58 is an exploded, schematic view of the angle lever used for adjusting the roller in the forward-backward direction, to which the spring according to Fig. 59 is coupled;

Fig. 59 is a plan view of the flat spiral spring used in the forward-rear guide for damping and adjusting the forward-rear rollers in the group;

Fig. 60 is a front plan view of the forward-rearward guide rollers of the group;

Fig. 61 is a partial plan view of a guide and one of the rollers of the guide rail group of the guide according to Fig. 56, showing the positioning of the electromagnets of an actuator designed for a relatively small force;

Fig. 62 a gap sensor;

Fig. 63 shows a flow sensor used in the acceleration loop according to Fig. 69;

Fig. 64 is a side view of an electromagnet core;

Fig. 65 a top view of the core according to Fig. 64 with indicated coils;

Fig. 66 is a simplified block diagram of a steering circuit for controlling two active guides arranged on opposite sides of an elevator car for lateral control, but which can be used for forward-backward control of the guides on opposite sides of a rail web;

Fig. 67 is a graphical representation of a biasing method for controlling a pair of opposing electromagnets, for example, where the force command for the right active guide of Fig. 66 is biased in the positive direction and the force command for the left guide is biased in the negative direction to achieve a composite response which avoids abrupt switching between the pair;

Fig. 68 is a more detailed representation of the discrete signal processor according to Fig. 66;

Fig. 69 is a control scheme for a pair of active guides as shown in Fig. 66, including control of both the small actuators and the large actuators, including a steering arrangement for the large actuators;

Fig. 70 is a diagram showing some of the parameters shown in the control scheme of Fig. 69;

Fig. 71 is a diagram showing the response of a single position transducer associated with each of the position transducers shown in Fig. 62; and

Fig. 72 is a representation of the composition of two such converter characteristics, as may possibly occur on line 698 in Fig. 69.

According to Fig. 1, a passenger platform 10 for an elevator car or cabin is suspended or supported by a device 12. In the present context, "cabin" refers to a passenger platform which is suspended or movably supported within an outer frame (not shown in Fig. 1). "Car" refers to refers to a passenger platform which is not movably supported within a frame or, alternatively, to a frame for a suspended or supported cabin platform (in some cases also referred to as a 'car frame'). Some examples of the use of each term are: (i) a car suspended from a cable passed over a rotating pulley, (ii) a ("swing") cab suspended by a cable, one or more rods in a car frame, (iii) a car supported on a movable platform which in turn is supported on a hydraulically operated piston, (iv) a cab supported on a movable platform in a supported or suspended car frame, etc. In all cases, the elevator car or car frame is moved up and down in an elevator shaft (not shown in Fig. 1) guided by means such as vertical rails (not shown) attached to the shaft walls.

In accordance with the invention, one or more disturbances 14 (e.g., an air current in the shaft acting on the car or car frame, a disturbance in the form of a bumpy ride imparted to the car or car as a result of a non-straight condition of a rail cross section, etc.) may be sensed by a sensor 16 mounted in or on the car or car platform 10. The sensor 16 typically senses an effect of the disturbance 14 to provide a signal on line 18 having a magnitude corresponding to the magnitude of the effect. A device 20 is responsive to the signal on line 18 for determining the amount of response required to counteract the sensed effect of the disturbance and for providing a signal on line 22 which causes an actuator 24 to move the platform 10 as directed by an actuation signal on line 26. The actuator 24 may, without limitation, be arranged between the car or the car frame and the shaft, or it may are located between the car frame and the cabin inside the frame in order to allow forces to act therebetween depending on the control signal on the line 22.

A plurality of sensors similar to sensor 16 may be responsive to one or more selected parameters indicative of translational or rotational movements of the car or cab which cause the latter to deviate from a perfectly centered rest position on an imaginary vertical line through the center of the elevator shaft. Such sensors may be responsive to any one or any number of selected parameters, such as the position of the car or cab relative to the elevator shaft, the translational accelerations acting on the car or cab, etc., as sensed in accordance with the invention. Such sensors may provide one or more readings to device 20 or similar device to complete a closed loop control system for automatic control in accordance with the invention.

As suggested above, one way of considering a preferred embodiment of the invention is to think of the control system as causing the vertical centerline of the elevator car (or the vertical centerline of a frame-suspended or frame-supported elevator cab) to consistently coincide with an imaginary fixed reference line extending up the center of the elevator shaft, without the centerline of the suspended car or cab deviating from the reference centerline of the shaft, or without the car or cab, whose centerline coincides with the fixed centerline, rotating about the fixed centerline.

Fig. 2 shows accelerometers 16a, 16b and 16c mounted on the car, which together form an example of a sensor arrangement that can be used to measure horizontal accelerations to sense small horizontal translations which cause deviations of the car or cab centerline from the shaft centerline, and in addition - without limiting the above - to sense accelerations which represent small rotations of the car or cab about the shaft centerline. An additional set of similar sensors 16d, 16e and 16f may - but not necessarily - be located near the ceiling of the car or cab. The selective use of one or more groups of actuators, e.g. actuator groups 24a, 24b, 24c and 24d, allows the application of forces to maintain the desired conformity of the car or cab centerline on the one hand and, on the other hand, and if desired, provides this with freedom of rotation with respect to the vertical or even about one or more axes in the horizontal plane. A preferred embodiment of the present invention uses two groups of actuators 24a, 24b, e.g. one group each comprising a pair of actuators. Thus, although two groups of actuators are shown near both the ceiling and the floor of the car or cab, it is to be understood that this is only shown to indicate that actuators operating from any position or in any grouping, i.e. other groupings in other positions, are certainly encompassed by the claims of the invention. As mentioned, the preferred embodiment uses only floor actuators. The fact that the actuators are shown detached from the platform, which are electromagnetic actuators of the non-contact type (air gap type), in no way excludes actuators which are mechanically attached to the platform.

An arbitrarily configured three-dimensional coordinate system 44 in Fig. 2 lies with its XZ plane in the drawing plane, with its origin being considered to coincide with the center of gravity of the elevator car or cabin 10, while the minus Y axis protrudes vertically from the drawing plane in the direction of the observer. The coordinate system 44 according to Fig. 2 is shown in more detail in Fig. 3. There it can be seen that in addition to rotations around the vertical Z axis, rotations around the X and Y axes are also possible, which according to the invention can also be controlled if desired. According to the invention, translations in the horizontal plane can be controlled using the device shown. This can also - but does not have to - be used to control twists. Thus, the present invention can be used to control twists in the horizontal plane and can be extended to two or more axes, including an additional horizontal and a vertical axis.

It will also be appreciated that the accelerometers cannot be located at the center of gravity, which might be desirable. A floor or ceiling of a passenger compartment is illustrated here without limitation as an acceptable compromise. Of course, the selected positioning of the illustrated sensors is arbitrary. The symmetry of this positioning with respect to the illustrated coordinate system or with respect to each other should not be inferred that the selected relation is required to practice the claimed invention. In other words, for example, sensors for measuring accelerations can be aligned along axes that are parallel to each other or that coincide with the axes of actuation, i.e. 45 degrees with respect to the shaft walls. In any event, it may sometimes be advantageous to use a coordinate system whose axes are similarly aligned with the directions of force action of the actuators. It should also be understood that the orientation of the actuators at angles of 45 degrees with respect to the shaft walls is not absolutely essential. In fact, the relations of the actuators with respect to the car or cab are not critical. However, it is preferable to align the actuators orthogonally in order to obtain the possibility of universal force vectors and to allow a distance between paths of opposing force and thus the creation of torque. Thus, the actuators could be mounted in each corner so that they are aligned in the diagonal direction. and not perpendicular thereto. Although such a configuration is not particularly preferred since it would exclude the possibility of counteracting vertical twists, it would nevertheless fall within the scope of the present claims.

It can also be seen from the locations of the illustrated accelerometers near the floor that lateral accelerations along the X axis can be sensed by accelerometer 16a, while those along the Y axis can be sensed by accelerometers 16b and 16c. A mismatch of the outputs of the two Y-sensitive accelerometers indicates rotation about the Z axis. Clockwise or counterclockwise rotation is indicated depending on whether the Y accelerometer 16b or 16c provides the reading signal with the larger magnitude. The magnitude of the difference is indicative of the magnitude of the angle of rotation relative to a reference position. A similar situation exists for sensors 16d, 16e and 16f on the ceiling.

Although no guide rails are shown, they are typically located at opposite locations on two of the four shaft walls. These could then serve, for example, for a car as ferromagnetic masses used by, for example, the actuators 24a, 24b, 24c and 24d if the actuators are of the electromagnetic type. In such cases, the actuators 24a and 24b may be located near the floor and the actuators 24c and 24d near the top of the platform 10 to create magnetic induction for non-contact interaction across air gaps with respect to the rails. Electromechanical, ie contact-operating active actuators may also be used, as will be discussed below. Conventional, passive wheel guides may be used instead of active actuators 24c, 24d on opposite sides in the upper part of the car to provide additional stability without the need for to provide additional active control systems, as is required for example by Ando, but for its more limited purposes.

For example, in a suspended (swing) car, electromagnetic non-contact actuators 24a and 24b may be mounted on the bottom of the car in conjunction with suitable ferromagnetic reaction plates which extend up from the bottom of the car frame to provide a path for the magnetic flux created by the actuators. In this case, no additional passive guides for the top of the car are necessary.

In an example of a supported car or cab using a horizontally sliding platform for support, such as shown in Luinstra et al.'s US-A-4,660,682, but supported on a hydraulic piston or in a suspended car rail (as shown in Luinstra et al.), electromagnetic non-contact actuators 24a and 24b may be mounted on the underside of the sliding platform with suitable ferromagnetic reaction plates raised below the sliding platform on a non-sliding horizontal platform attached to the top of the piston or, in a supported car, on the floor of the car frame, to provide a path for magnetic flux originating from the actuators.

From the above it should be apparent that a preferred embodiment of the present invention can be used to increase the ride comfort in an elevator car or cabin. The preferred embodiment of the present invention will be described first for a cabin and then for a lift car. It will be clear that the same approach can be used for both the lift car and the cabin, differing in detail only to the extent necessary to take into account the fact that the actuators for a Car actuators act against a rail on a shaft wall, as shown in Fig. 15, while the actuators for a cabin act on a frame, as shown in Figs. 14B, 14C and 14D.

Fig. 14A is a top view and Fig. 14D is a side view of a shuttle car 46. It includes an active suspension system with an actuator 45 as shown in Fig. 14B which is driven by a controller (not shown) to process read data regarding the movement of a platform 10b. Sensors 50, 52 and 54, which are accelerometers, measure the movement of the platform. The platform 10b is suspended by rods 56, 58 from a frame 60 which in turn hangs from a cable 62 to move the car 46 up and down in an elevator shaft having walls 64 and 66 and rails 68 and 70 supported thereon. A plurality of wheels 72, 74, 76 and 78 are mounted on the frame 60 by means of springs 80, 82, 84 and 86. A passive shuttle car of this type (without the actuator 45) is shown in detail in U.S. Patent 4,899,852.

As the wheels move up and down the shaft, they are subject to rail irregularities caused, for example, by a rail joint misalignment 88 or by a waviness 90. Such joint irregularities cause relatively high frequency car vibrations, while a waviness typically causes lower frequency vibrations. In addition to vibrations caused by rail irregularities through the frame 60 to the platform 10b, the platform 10b is also subjected to what we call direct forces 92, which can have a large number of different causes, including wind forces, movement of persons within the car standing on the platform, and numerous other similar "direct" forces.

In both Figures 14A and 14B (similar to Figure 2) a coordinate system is shown which represents an X-Y plane in the floor of the platform 10b with the Z axis pointing vertically. The present invention mitigates both direct forces and forces induced by track irregularities acting on the platform 10b through the frame 60. It does this by counteracting lateral forces in both the X and Y directions. Twisting with respect to the Z axis is also mitigated by lateral control along the X and Y axes as discussed below.

In Fig. 16C, a floor 200 of a passenger platform similar to the cabin 10b in Figs. 14A and 14B and a floor 202 of a frame similar to the frame 60 in Figs. 16A and 16B are arranged one above the other and shown in a plan view showing the two parts in substantially alignment in the resting state. For purposes of description and without limitation, assuming a rectangular or, for further simplicity, a square configuration of the cabin floor 200 and the frame floor, one can imagine a pair of reaction planes perpendicular to the cabin floor and the frame base intersecting each other at a vertical cabin centerline which in turn intersects the center of the square perpendicularly. The reaction planes may or may not intersect the floor and the base along the diagonals of the base (and the floor).

As mentioned, one way to view the preferred embodiment of the invention is to think of the control system as causing the centerline of the elevator car to constantly coincide with an imaginary reference line running down the center of the elevator shaft, without the suspended or supported car rotating about the coincident centerlines of the car and the shaft.

It can achieve this by using accelerometers 204, 206 and 208 mounted on the cabin, which together serve to to measure accelerations manifested as small transverse deviations of the car centerline from the shaft centerline, additionally measuring accelerations representing small rotations of the car about the shaft centerline, and also making selective use of actuators 210, 212, 214 and 216 which apply forces perpendicular to the reaction planes to maintain the desired coincidence of the centerlines without vertical rotation of the car about the shaft centerline. These actuators correspond to actuator 45 shown in Fig. 14B. A three-dimensional coordinate system 218 shown in Fig. 14C has its XY plane in the plane of the drawing and should be thought of as having its origin at the center of squares 200 and 202 while its Z axis points perpendicularly from the plane of the drawing toward the viewer, similar to the coordinate system of Fig. 14B. From the locations of the accelerometers, it can be seen that transverse accelerations along the Y axis can be detected by the accelerometer 206, while those along the X axis can be measured by the accelerometer 204 or 208. A mismatch of the output signals of the two X-sensitive accelerometers indicates a rotation around the Z axis. A clockwise or counterclockwise rotation is displayed depending on which X accelerometer 204 or 208 supplies the reading signal with the larger magnitude. This means that the magnitude and sign of the mismatch are indicators for the magnitude and direction of the angle of rotation. In the exemplary embodiment considered, however, ie during control, the measured movements are not perceptible to the passenger.

Ferromagnetic reaction plates 218, 220, 222 and 224 of equal size may rise symmetrically with respect to the center of the frame bottom near each corner along the diagonals so that they lie in the reaction planes. Four electromagnet cores 226, 228, 230 and 232 with coils may be mounted on the bottom surface of a suspended or supported platform in such a way that they each face one of the reaction plates. Attractive forces generated by the control system with the help of the four electromagnet core coils are effective in such a way that the core-coil pairs move away from or closer to the associated reaction plates.

The positioning of the core-coil assemblies with respect to the reaction planes can of course vary. For example, as shown in Fig. 14C, electromagnet core-coils located along the same orthogonal at opposite corners, i.e., pair 226, 232 or pair 228, 230, are arranged to exert attractive forces on opposite sides of the reaction plane, whereby a pair of electromagnets belonging to one of the reaction planes jointly opposes the clockwise rotational forces, while the other pair opposes the counterclockwise rotational forces. Electromagnetic actuators acting along axes intersecting in the same cabin wall, for example 230, 232 or 226, 228, may be located between that wall and its respective reaction plates so that they can work together to displace transverse forces.

However, it should be noted that the electromagnets in Fig. 14C could all be located on the opposite sides of the reaction plates from those shown, the only change being that all control actions would be reversed. In addition, the core-coil pairs for cooperating against disturbing forces in a particular direction of rotation could be assigned to adjacent corners of the cabin so that they are on the same side of the reaction plate with respect to the diagonals, and thus the diagonally assigned pairs no longer cooperating. In this case, of course, the equations given below would have to be rewritten, although the same basic principles disclosed here would still apply.

It should also be noted that alternatively the reaction plates could be supported on the underside of the cabin while the electromagnet core coils would be supported on the floor of the frame.

It should also be recognized that an "X" or diagonal concept with "planes of response" has been introduced here as an educational tool, but that this is merely a conceptual aid to describe a preferred cabin embodiment, but there is no requirement for practical implementation, nor even for application of the concept to all practical cases of the invention.

While conceptually applicable to all or part of the other embodiments, although not necessarily so as will be apparent from the embodiment of Figure 4D to be described below, it should be appreciated that the orientation of the "X" need not be from corner to corner, but may be in any suitable orientation. Similarly, the actuators and reaction plates need not be located between the floor of the car and the floor of the frame. Nor do they necessarily all have to be at the same level, although such a configuration could potentially result in unnecessary complexity. Needless to say, the invention is not limited to the use of four actuators; three, four, five or even more actuators may be used. The number four was chosen as a suitable number because it fits well with the symmetry of a typical elevator car and elevator shaft. An "X" orientation was first disclosed in Applicant's U.S. Patent 4,899,852 to Salmon et al. in connection with a passive stabilization system.

In the car embodiment to be disclosed in greater detail below, Figs. 4 to 7 and Figs. 9 to 13 serve to illustrate different embodiments of several webs rail configurations for use in active control systems, all of which have multiple web rails that are different from the prior art single web rail shown in Figure 8 that is used in at least one conventional active system (see U.S. Patent 4,754,849 to Ando et al.).

In Figs. 4 to 7 and 9 to 13, more than one "web" is used in each case to interact with two or more corresponding actuators. In Fig. 8, however, a single web 40 is used by all three actuators 42, 44 and 46. It should be appreciated that in all of the multi-web rails shown below, the associated actuators may be arranged in a manner other than exactly as drawn.

In Fig. 4, a rectangular rail 48 has three webs 50, 52 and 54 which serve as ferromagnetic paths or masses for three separate electromagnetic actuators 56, 58 and 60, respectively. As an example of how an associated actuator could be arranged differently than shown, the actuator 58 could instead be arranged between the web 52 and the shaft wall to save space and make the overall arrangement more compact.

In Fig. 5, a rail 62 is shown with two webs; it has a V-shape with a web 64 and a web 66. The triangular configuration was previously disclosed for a passive system by Charles A. Otis in US Patent 134,698. However, according to the invention, multiple webs are used in an active system, for example, with web 64 serving as a ferromagnetic mass for an electromagnetic actuator 68, while web 66 performs a similar function for actuator 70. It should be noted that rail 62 may have foot pieces 72 and 74 for easy attachment of the rail to the elevator shaft wall 76. Rail 62 may also be formed with a full triangular cross-section, with or without feet (not shown). Returning to Figure 4, the three-web embodiment may similarly be a four-web box-shaped rail without feet. As another example of how an associated actuator could be arranged differently than shown in the drawing, actuator 70 could be arranged opposite actuator 68 on the other side of web 64, while web 66 could be used as a counter projection for a safety brake (not shown).

In Fig. 6, an I-beam 78 is used. One web 80 serves a pair of opposing electromagnetic actuators 82 and 84, while a second web 86 provides a third actuator 88. A third web 90 does not serve as a ferromagnetic mass or path for another actuator but can be used to attach the remaining two webs to a shaft wall 92.

Fig. 7 shows a variation of the two-webbed, V-shaped rail 62 shown in Fig. 5. The rail 94 includes a pair of webs 96, 98 facing respective actuators 100 and 102. The rail also includes a projecting web 104 which can serve as an easily accessible handle to which a safety brake (not shown) is applied.

Fig. 9 shows an inverted V-shaped rail 106 with a web 108 for cooperation with an electromagnetic coil 110 and a web 112 for a coil 114. Webs 116 and 118 provide structural strength.

Fig. 10 shows a C-shaped rail 120 with a web 122 and a web 124 which form a ferromagnetic path for a coil 126 and 128, respectively. A coil 130 uses a web 132 as its ferromagnetic mass. The web 132 can also be used to attach the rail 120 to a shaft wall 134.

Fig. 11 shows a rail 136 supported on a shaft wall by means of a front base 140. The rail 136 includes a curved section 142 which in effect consists of two "webs", one on each side of a projecting web 144 for safety braking purposes. One side of the curved cross-section is for interacting with a coil 146, while the other is for interacting with a coil 14B.

Fig. 12 is an illustration of a rail 150 attached to the shaft wall 152 by means of a foot piece 154. The active part of the rail 150 includes a circular rail 156 which in effect has two semicircles on each side of a projection 158. Coils 160 and 162 use the respective halves of the circle 156 as ferromagnetic masses. Thus, the rail 150 is in effect a two-web rail.

Fig. 13 is an illustration of a rail 164 attached to a shaft wall 166 by means of feet 168 and 170. A curved section 172 is effectively divided into two sections, one section on either side of a projection 174. Each section is used by an actuator, i.e., an actuator 176 and an actuator 178. The rail 164 is similar in concept to the rail 136 in Fig. 11, except that it is in the shape of an "omega" rather than a "D".

The rail of Fig. 7 represents the preferred embodiment, which allows the use of only eight electromagnets, as shown below in connection with stabilization within the horizontal plane and about three axes of rotation.

In a suspended (pendulum) cabin, there is little or no need for stabilizing the top of the cabin in relation to the frame on which it is suspended, because there is no significant twisting about any horizontal axis. In a However, in a supported car, e.g. tiltably supported at a point located on a transversely displaceable platform within the frame, twists about horizontal axes can be considerable. In such a case, it may be desirable to use a control system similar or identical to that described above in connection with Fig. 14C for the roof of the car, and which operates completely independently of the control system providing stabilization of the floor. In the problem of tilt stabilization, it might at first glance be considered necessary to measure the tilt of the car in order to directly counteract rotations about any horizontal axis or axes. According to the teachings of the present invention, two independent control systems are used for both cars and elevator cars to stabilize horizontal translations in the ceiling and floor; surprisingly, any twists about any horizontal axes are automatically taken into account. Although applicable to both cabins (and in particular to supported cabins) and lift cars, the following description will refer to the case of a lift car. The person skilled in the art will have no difficulty in using the teaching given below to manufacture and use a cabin with horizontal torsion stabilization.

Fig. 14D is a plan view of the shuttle car of Fig. 14B looking down from the bottom surface of the platform 10b of an actuator assembly in the space between the floor of the car frame and the bottom of the suspended car. In this case we have shown three actuators 240, 242 and 244 instead of the single actuator 45 shown in Fig. 14B. The actuators shown in Fig. 14D are of the electromagnetic type which will be discussed in more detail below. The orientation of the platform 10b is the same as shown in Fig. 14A and the same coordinate system applies. Thus, actuator 142 generates forces along the X axis while actuators 240 and 244 generate forces along axes parallel to the Y axis. Each of the actuators 240, 242 and 244 has a separate control loop independent of the others, at least in the sense that they do not depend on sensors in any other axis. It is of course understood that the various control axes are mechanically coupled.

Several active suspension inventions are disclosed herein that describe embodiments of the present invention that include separate, single-axis controls as disclosed herein in connection with Figure 14D, with several of these embodiments describing combined multi-axis control channels as shown in Figures 14C and 15. It will be appreciated that the separate, single-axis controls disclosed herein have the advantage of simplified design and the advantage of being electronically decoupled for the various control axes. However, it should be noted that the separate, single-axis approach is somewhat more expensive than the combined multi-axis approach because an additional number of electromagnets are required. On the other hand, there are only three electronic channels for the separate channels, while the combined, multi-axis approach requires a minimum of four electronic channels.

It has been stated above, it should be understood, that a preferred embodiment of the invention may be used to increase the comfort of travel in an elevator car or cabin. The preferred embodiment of the present invention has been described in part initially for a cabin, it will be described below for an elevator car. Again, it will be seen that the same procedure is used for both the elevator car and the cabin, with differences in detail only being necessary to take into account the fact that the actuators for a car operate against a rail on a shaft wall, while the actuators for a cabin operate on a frame, as shown in Fig. 14C.

Referring to Figure 15, the floor of a suspended or supported car 250 is shown in plan view showing the car at rest. Again, similar to the above illustration for a car, for purposes of description and not limitation, a rectangular or, for even greater simplicity, a square layout of a passenger platform or car floor is assumed, where one can imagine a pair of reaction planes perpendicular to the floor of the car 250 intersecting each other along a vertical car centerline which in turn passes perpendicularly through the center of the square. The reaction planes may, but need not, intersect the floor along the floor diagonals.

As above, again, one way of considering the preferred embodiment is to imagine a control system as causing the centerline of the elevator car to constantly coincide with an imaginary reference line running down the center of the shaft, without the suspended or supported car rotating about the coincident centerlines of the car and the shaft.

It does this by using car-mounted accelerometers 252, 254 and 258 (analogous to sensors 16b, 16a and 16c in Fig. 2, respectively) which together serve to sense accelerations manifested as small transverse displacements of the car centerline relative to the shaft centerline and to further measure accelerations manifested as small rotations of the car about the shaft centerline, making selective use of actuators 258, 260, 262 and 264 to apply forces perpendicular to the reaction planes to maintain the desired coincidence of the centerlines without rotation. A three-dimensional coordinate system 266 in Fig. 15 has its XY plane in the plane of the drawing and should be viewed as its origin lies in the middle of the square 250, while its Z axis points perpendicular to the plane of the drawing in the direction of the observer. From the mounting locations of the accelerometers it can be seen that transverse accelerations along the Y axis can be detected by the accelerometer 254, while those along the X axis can be measured by either the accelerometer 252 or 256. A mismatch of the output signals of the two X-sensitive accelerometers indicates a rotation about the Z axis. A clockwise or counterclockwise rotation is indicated depending on which of the X accelerometers 252 or 256 supplies the measurement signal with the larger magnitude. This means: the magnitude and sign of the mismatch are characteristic of the magnitude and direction of the angle of rotation.

V-shaped rails 267 and 268 similar to the rail shown in Figs. 5 and 7 or similar to the rail suggested by C.R. Otis, attached to the two opposing shaft walls 267a and 268a, represent ferromagnetic reaction plates 270, 272, 274 and 276. Four electromagnet cores 280, 282, 284 and 286 with associated coils may be mounted on the sides near the bottom of a suspended or supported platform so that each faces one of the reaction plates. Attractive forces generated by the control system by means of the four electromagnet core coils act to move the core coils apart or closer to their associated reaction plates. The positioning of the core coils in relation to the reaction plates can of course vary as in the example of the cabin, with the special exception here regarding the selected rail shape.

Referring now to Fig. 16, the device 20 of Fig. 1 is shown in its embodiment as a digital signal processor which may include an input/output (I/O) device 280 which may include an analog-to-digital (A/D) converter (not shown). responsive to an analog signal from sensor 16. The sensor may take the form of accelerometers 204, 206 and 208 as shown in Fig. 14C or accelerometers 252, 254 and 256 as shown in Fig. 15. I/O device 280 may further include a digital to analog (D/A) converter (not shown) for providing force command signals on line 22 to an analog actuator 24 which may include actuators 210, 212, 214 and 216 as shown in Fig. 14C, actuators 258, 260, 262 and 264 as shown in Fig. 15, or other suitable actuators. Also within the controller 20 of Fig. 16 is a control, data and address bus 282 which interconnects a central processing unit (CPU) 264, a random access memory (RAM) 286 and a read only memory (ROM) 288. The CPU executes a step-by-step program stored in the ROM, stores within the RAM input signals having magnitudes indicative of the value of the parameter being controlled as presented on line 18, as well as signals having magnitudes corresponding to the results of intermediate calculations, and output signals having magnitudes corresponding to the value of the parameter being controlled as presented as the output signal on line 22.

Returning to the car platform design of Figs. 14C and 15, and referring to Fig. 17, a simplified step-by-step program will be explained which the CPU of Fig. 16 executes in performing the control loop function described above in connection with the device 20 of Fig. 1 and the embodiment thereof shown in Fig. 16. After entering step 300, an input step 302 is executed in which the magnitude(s) of the signal(s) on line 18 is received by the I/O unit 280. With regard to Figs. 14C and 15, these shall be referred to as signals Ax1, Ax2 and Ay, respectively, for the signals from accelerometers 204, 208 and 206 of Fig. 14C (or from accelerometers 252, 256 and 254 in Fig. 15 (with appropriate coordinate transformation)), and they are stored in the RAM 286 in Fig. 16. One or the other of the two X-axis accelerometers 204, 208 (or 252, 256) may be used in a step 304 to calculate the magnitude of a positive or negative signal Ax, or both accelerometers may be used to check each other and then average, or they may be used in some other similar redundancy method. (Of course, it should be seen that steps 302 and 304 may be combined into a single reading step when a rotary sensor is used in conjunction with two translational (X and Y) sensors.) From a comparison of the two signals provided by the accelerometers 204, 208 (or 252, 256), a value Aθ may be calculated in step 304. The magnitude of the signal Aθ depends on the degree to which the magnitude of the signals from accelerometers 204 and 208 (or 252 and 256) differ. The sign of their sum determines the direction of rotation. The values of Ax, Ay and Aθ are temporarily stored in RAM 286.

Next, a step 306 is executed in which a calculation is made of the forces required to counteract the influence(s) of the disturbance(s) as manifested in the form of one or more measured parameters (accelerations). This can be done using the known mass of the suspended or supported car or lift cage and the formula F = ma, where "F" is the required counterforce, "m" is the mass of the suspended or supported car or lift cage and "a" is the value of the determined acceleration. Thus, from the signals Ax, Ay and Aθ stored in step 304 within the RAM 286, the values Fx, Fy and Fθ are calculated. These calculated values are given in the form of force command signals over the line 22, which is indicated by a step 308. It should be noted that the orientation of the actuators according to Fig. 14C and 15 is such that a command signal requiring a counterforce in the positive X direction is received by the electromagnets 210 and 214 (or 258, 262) in conjunction must be applied, with each electromagnet providing half of the required counterforce by providing a force equal to the specified force in the X direction multiplied by cos (45 degrees). Similar distributions of counterforces are made for the Y direction and for twists. A set of formulas covering all possibilities is given below (in the following equations, subscripts 1, 2, 3 and 4 correspond to the electromagnetic actuators 210, 212, 214 and 216 of Fig. 14C (or actuators 258, 260, 262 and 264 of Fig. 15)):

Fx+: F1; = (KCS) (Fx+) Fx-: F₂ = (KCS) (Fx-)

F3; = (KCS) (Fx+) F₄ = (KCS) (Fx-)

Fy+: F1; = (KCS) (Fy+) Fy-: F₃ = (KCS) (Fy-)

F2; = (KCS) (Fy+) F₄ = (KCS) (Fy-)

Fθ+: F₂ = (KCS) (Fθ+) Fθ: F₁ = (KCS) (Fθ-)

F3; = (KCS) (Fθ+) F₄ = (KCS) (Fθ-)

where F = force and

KCS = cos(45º) = sin(45º) = 0.707.

After the necessary calculations are made and the required counterforce command signals are provided, the program exits at step 310. In accordance with the invention, additional steps are taken to overlay a system that compensates for improperly leveled accelerometers and also compensates for changing offset in the accelerometers. Accelerometers have two main errors: (i) offset drift and (ii) picking up unwanted gravity components because they are not perfectly level. Also, but of lesser importance, there are (iii) linearity errors. A non-level accelerometer senses accelerations due to gravity in proportion to the sine of the angle it makes with the actual vertical. Correction of non-linearity is not normally important in embodiments of the present invention, but correction can be made if desired (see Fig. 23).

Assuming that the nonlinearity maintains its basic relationship with the actual linearity as adjusted for offset changes, such linearity can be corrected at any stage of a measured acceleration by consulting a look-up table designed to provide a correction factor. If the offset were constant over time, it could be easily corrected with a constant correction factor. However, since the offset can change over time due to temperature fluctuations, aging, etc., corrections should be made in a dynamic manner. Offset and offset changes as well as accelerations due to gravity can be corrected by providing a relatively slow-moving control system to control the position of the car or cab with respect to the shaft centerline. This can be done by recognizing that the mean lateral acceleration must be zero (otherwise the car or cab would wander into free space). The slow-speed control loop offsets the average accelerometer output signal. Averaging can be done using, for example, an analog low-pass filter or a digital filter.

So if we consider a single axis control, for example a control such as the x-axis shown in Fig. 2, the theory of operation for such a system for controlling the car with both acceleration and position sensors is shown in Fig. 18. In its elementary form, the system includes the mass of the car, represented by a block 320. The mass of the car is subjected to a force via a line 224. A disturbance force is shown schematically as a signal on a line 326 which is summed in a "summer" 328 (an abstract representation for the fact that the disturbing force is physically counteracted by a counterforce) with a counterforce signal on a line 330 which is supplied on a line 330 in proportion (Kc) to the acceleration (A) on the line 324 sensed by an accelerometer 332 which supplies a sensed acceleration signal on a line 334 to a summer 336. The scale factor (Ka) of the accelerometer is (volts/m²/s). (As previously stated, the acceleration on line 324 is produced by the control force on line 326 interacting with the mass of the suspended or supported car or cab according to the relationship F/M as shown in block 332, where F is the control force and M is the mass of the car or cab. Summer 328 represents the summation of the control force on line 326 and the counterforce on line 330 to form a net force on line 322 acting on mass 320.) Summer 336 may provide a signal on line 338 to a force generator 340 which may have a transfer function of 1.0 Newton/volt. Summer 336 is used to connect an inner acceleration loop signal on line 334 to signals from an outer acceleration and position loop, discussed below, before merging on line 338 leading to force generator 340. The inner acceleration loop comprising elements 320, 332 and 340 and the associated summers form the primary control loop used for the "mass augmentation" defined herein.

The description of Fig. 18 up to this point covers the theory of the control system, which was previously explained in connection with Figs. 1 to 17. Secondary control loops can also be added in the manner shown in the summary of Fig. 18.

Two secondary control loops are shown that can be used to eliminate offsets in the accelerometer 332 caused, for example, by misalignment with respect to gravity and by manufacturing inaccuracies. The first of these secondary loops corrects based on position offsets. A position transducer that indicates the car position is abstractly represented by an integrator block 342 and an integrator block 344. The integrator 342 supplies a speed signal via a line 346 to the integrator 344, which in turn generates a position signal on a line 348. The car position signal on the line 338 is compared in a summing element 350 with a reference signal supplied via a line 352. The signal on line 352 would actually be a fixed DC level scaled to represent, for example, the X position (in the car coordinate system 218 shown in Fig. 14C or in the car coordinator system 266 of Fig. 15) of a selected reference, such as the elevator shaft centerline (which substantially coincides with the true vertical, i.e., a line along which the earth's gravity acts). This entire process is carried out in practice by using a position sensor which indicates the relative position between the car and the car frame. The summing element 250 forms a signal on line 354 which represents the relative position of the car with respect to the frame, and it may be designated as a relative position signal or a position error signal. After being summed in a summing element 360 with a signal on a line 362, it is passed via a line 356 to a low-pass filter 358. The low-pass filter 358 supplies a filtered signal 364 which causes the force on the line 330 to be passed via the line 322 to the car or cab 320 until the position error signal is driven to zero or a value close to zero.

A second secondary control loop can be introduced if the position signal is not suitably available, or if the stability of the position correction control loop. The position error signal on line 354 can thus be modified in the summer 360 by adding the signal on line 362 provided by a gain block 366 which in turn is responsive to a signal on line 338 representative of the acceleration measured in the primary loop.

An extraneous signal on line 338 appears directly on line 322 when G₁ = 0 and G₂ = 0. Assuming that no position error is indicated on line 354 and the gains G₁ and G₂ are different from zero, a disturbance caused by an acceleration signal on line 334 appears on line 322, reduced by a dynamic factor

Traffic jam; + 1 / Sτ + (1 + G₁ G₂)

This factor approaches unity at higher frequencies, which means no effectiveness. However, at lower frequencies, this factor approaches 1/(1 + G1 x G2). Typically, G1 x G2 could be chosen to be nine (9) to reduce accelerometer offsets by a factor of ten (10).

Position control provides the advantage of very low error. Without the acceleration control loop 366, 360, 358 and 336 and/or the presence of practical control elements, this loop may not be as stable. Assuming a gain G2 = 0, the only possibility for a stable position control loop is that the mass of the car or cab is acted upon by damping, friction or an inherent spring constant due to pendulum action, either individually or in conjunction. One or more of these elements are present in a practical system.

Using an acceleration loop by making G2 nonzero can improve the operation of the position loop.

The control shown in abstract form in Fig. 18 can be implemented in numerous different ways, but Fig. 19 shows a preferred procedure.

There, a fast analog loop for quickly counteracting disturbing forces is combined with a slower but more accurate digital loop for compensating for gravity components and drift in the accelerometers. Several such fast analog control loops may take the form of analog controllers 370, 372, 374 and 376 as shown, one for each of the respective actuators 210 or 358, 212 or 260, 214 or 262, 216 or 264 of Fig. 14C or 15. With suitable interfacing (not shown), a single digital controller 380 may handle the signals going to and from all four analog controllers in a manner to be described. Each analog control responds to a force command signal on the lines 382, 384, 386 and 388 from the digital controller 380. The force command signals have different amounts depending on the transverse and rotational forces that are to be counteracted. The digital controller 380 in turn responds to acceleration signals on lines 99, 392 and 394 from the accelerometers 204 or 252, 206 or 254, 208 or 256 (the accelerometers of Figs. 14C and 15, respectively), and to position signals on lines 396, 398, 400 and 402 indicative of the size of the air gaps between the coil cores 226 or 280, 228 or 282, 230 or 284, 232 or 286 and the corresponding associated plates 218 or 270, 220 or 272, 222 or 274, 224 or 276.

In response to the force command signals on lines 382, 384, 386 and 388, the analog controls 370, 372, 374 and 376 deliver control signals via lines 404, 406, 408 and 410 to the coils of the coil cores 226 or 280, 228 or 282, 230 or 284, 232 or 286 to produce more or less strong attractive forces between the respective core coils 226 or 280, 228 or 282, 230 or 284, 232 or 286 and their associated reaction plates. The return current through the coils is monitored by current monitoring devices 412 or 420, 414 or 422, 416 or 424, 418 or 426 which supply current signals over lines 428, 430, 432 and 434 to the respective analog controls 370, 372, 374 and 376. The current sensors can be, for example, of the Beil IHA-150 type with multiple loops of the "through" line.

A plurality of sensors 440 or 448, 442 or 450, 446 or 452, which may be Hall cells (e.g. of the Bell GH-600 type), are associated with each core 226 or 280, 228 or 282, 230 or 284, 232 or 286 for the purpose of providing an indication of the flux density or of the magnetic induction (volt-sec/m²) in the gap, i.e. between the faces of the cores and the associated plates, or, in other words, of the flux density in the air gaps between these parts. The sensors 440 or 448, 442 or 450, 444 or 452, 446 or 452 provide reading signals to the analog controls 370, 372, 374 and 376 via lines 460, 462, 464 and 466, respectively.

Referring now to Fig. 20, the analog controller 370 is shown among the plurality of analog controllers 370, 372, 374 and 376 of Fig. 20 to illustrate one embodiment in detail. The remaining analog controllers 372, 374 and 376 may be identical or similar. The force command signal on line 382 from the digital controller 380 in Fig. 19 is fed to a summing circuit 470 where it is summed with a signal on line 472 from a multiplier 474 which is designed as a squaring circuit (for control linearization) with a gain dimensionally equivalent to magnetization (amps/meter) and properly scaled to convert a signal on line 476 indicative of flux density into a signal indicative of force The flux density signal on line 476 is provided by a Hall cell amplifier 478 which serves to raise the level of the signal on a line 480 from the Hall cell 440 or 448.

The summer 470 provides a force error signal on a line 484 to a proportional-integral (PI) amplifier 486 which provides a PI amplified signal on a line 488 to a firing angle compensator 490. The compensator 490 provides a firing angle signal on a line 492 which controls the firing angle of a plurality of silicon controlled rectifiers (SCRs) in a controller 494 after filtering in a filter 496 which provides a filtered firing angle signal on a line 498 to the controller 494, which can be more fully described as a single-phase, two-quadrant, full-wave power controller with SCRs. This type of controller is preferred over single-quadrant and half-wave converters. The least preferred combination would be a single-quadrant, half-wave converter. There would be a slight cost saving in taking these non-preferred approaches, but dynamic performance would be significantly degraded. A low-cost, single-quadrant system is possible using a DC rectifier and a transistor PWM chopper. The highest performance is achieved with a full-wave, two-quadrant, three-phase converter, but this approach is not the preferred approach for cost reasons. For example, the two-quadrant, full-wave converter 494 shown in Fig. 20 may consist of a pair of Powerex CD4A1240 dual SCRs and a circuit configuration as partially shown in Fig. 21 (RC attenuators in parallel with the SCRs are not shown) and a commercially available ignition board 253, such as the Phasetronics PTR1209 shown in Fig. 22. Gate signals on several lines 253a for the SCRs are provided by the ignition board 253. The power divider 252 is supplied with 120 VAC via a line 254, as is the ignition board, and it provides the appropriate current level on line 180 in response to the filtered firing angle signal on line 250.

The signal on line 428 from the current sensor 412 is applied to an analog multiplier/divider 504 (e.g., Analog Devices AD534) which is also responsive to the flux density signal on line 476 and divides the magnitude of the current signal on line 428 by the magnitude of the flux density signal on line 476, and multiplies the result by a proportionality factor to form the signal on line 296 (returned to the digital controller 380 shown in Fig. 19) which is indicative of the amount of gap (g1) between the face of the core of the face coil 226 and the plate 218.

As mentioned above, the digital controller 380 is responsive to the gap signals on lines 396, 398, 400, 402 and to the acceleration signals on lines 390, 392 and 394 to perform the control functions of Fig. 18 in conjunction with analog controls of the type shown in Fig. 20. Instead of generating force signals on lines 382, 384, 386 and 388 in exactly the same manner as explained above in connection with Figs. 16 and 17, these signals, although generated in a similar manner, are modified by summing corrective force signals calculated to correct position imbalances detected by position sensor 440, or 448 and similar sensors 442 or 450, 444 or 452, 446 or 452 associated with actuators 212 or 260, 214 or 262, 216 or 264 of Fig. 14C or 15. (Note that these are Hall sensors used to detect flux density.) The signals from the position sensors such as sensor 440 or 448 and from the current sensor C1 are passed on to line 396 as signal SPALT1 after processing by the divider circuit 504. Similar processing in other channels provides signals SPALT1, SPALT3 and APALT4 on lines 398, 400 and 402.) Such correction Force signals can be generated, for example, by first resolving the sensed position signals into components along the axes of the Cartesian coordinate system 218 or 266 in Fig. 14C or 15 according to the following equations:

Px+ = (P₁ + P₃) / (2KCS), Px- = (P₂ + P₄) / (2KCS),

Py+ (P₁ + P₂) / (2KCS), Py- = (P₃ + P₄) / (2KCS),

Pθ+=(P₂+P₃)/2, Pθ- =(Pi+P₄)/2,

and then, based on the above, calculate or select Px, Py and Pθ (which together specify the absolute position of the car) from the values Px- and Px+, Py- and Py+ and Pθ+ and Pθ-. For example, Px can be calculated as follows:

Px = (Px+ - Px-)/2.

One can also select Px+ or Px- depending on which is smaller. (Note: for large air gaps, e.g., large values of Px+ or Px-, the value is likely to be inaccurate and can be ignored.) The resulting components are used to determine the disturbance control force components Fpx, Fpy, Fpθ as shown in Fig. 18 on a single axis basis ("P" stands for position feedback). For example, Px on line 348 is compared to a reference quantity on line 352 to produce an X position error signal on line 354. This in turn passes through a low pass filter, such as filter 358. This forms a signal Fpx. To resolve the required X reaction force, if a positive force is required, Fp1=Fp3=(0.5)(Fpx)/(cos45º). For a negative force, Fp2=Fp4=(0.5)(Fpx)/(cos45º) applies. The same procedure can be followed for Fpy and Fpθ, using the corresponding equations. This allows the force components Fpx, Fpy, and Fpθ can be resolved into correction signals Fp1, Fp2, Fp3, Fp4 according to the following complete set of equations

Fpx+: F1; = (KCS) (Fpx+) Fpx-: F₂ = (KCS) (Fpx-)

F3; = (KCS) (Fpx+) F₄ = (KCS) (Fpx-)

Fpy+: F1; = (KCS) (Fpy+) Fpy-: F₃ = (KCS) (Fpy-)

F2; = (KCS) (Fpy+) F₄ = (KCS) (Fpy-)

Fpθ+: F₂ = (KCS) (Fpθ+) Fpθ: F₁ = (KCS) (Fpθ-)

F3; = (KCS) (Fpθ+) F₄ = (KCS) (Fpθ-)

where F= force and

KCS = cos(45º) = sin(45º) = 0.707,

which values are then summed with the acceleration feedback signals F₁, F₂, F₃, F₄ (e.g. with the signal on line 364 or line 382), which are formed in the manner explained above in connection with Figs. 1 to 20.

20 It should be understood that a valid position reading is only available from the flux sensors of the specified type when the associated force actuator is driven. This means that any processing algorithm must depend on whether or not there are currently solenoid activation currents.

It should be noted that the air gap signals on lines 396, 398, 400 and 402 can be provided by a simple position sensor.

An additional advantage of our invention is that the electromagnets can be used to control the position of the car or cab at stops, for example to bring the suspended or supported car or cab into a rest position with respect to the frame while passengers are boarding and alighting. Of course, the signal processor of Fig. 16, the digital controller 380 of Fig. 19 or any additional signal processor may perform additional control functions such as starting and stopping cars and dispatching cars. In the case of stopping at a floor, the processor 20 of Fig. 16 may receive a read signal on line 18 or an algorithmically determined but similar signal indicating that the car is in a vertical rest position and then provide a signal on line 22 to control the position of the suspended or supported car or cab. For example, if the car platform 200 in Fig. 14C is oriented in the elevator shaft such that the left vertical edge of the car means that the threshold of the car is flush with the elevator shaft door threshold 7, the signal processor 20 of Fig. 16 may be programmed to provide force command signals to the actuators 210 and 214 to provide the attractive forces which urge the suspended car upward against, for example, stops 702 and 704 mounted in the car frame 202 to urge the car threshold into its rest position relative to and in close alignment with the shaft entrance threshold 700 after the signal processor 20 has caused the frame 202 to assume the rest position. The stops may also be part of the actuators themselves, for example they may be mounted on the tops of the legs 304 and 306 of the core shown in Fig. 30, so that supplying a selected current to the coils produces an attractive force sufficient to press the stops against the reaction plates. In a double door platform, as another example, if the platform 200 of Fig. 14C is oriented in the shaft such that the left vertical edge of the car means that the car threshold is aligned with the elevator shaft door threshold 700, and the right vertical edge is simultaneously aligned with a threshold 514, the signal processor 20 of Fig. 16 may be programmed to provide force command signals to the actuators 212 and 216 to provide the attractive forces necessary to move the suspended car upwards against, for example, stops 513, 513a mounted in the car frame 202, thereby forcing the car threshold into its rest position relative to and in alignment with the elevator shaft entrance threshold 514 after the frame 202 has assumed the rest position.

The method for achieving this is shown in Fig. 51 and is explained further below in connection with Fig. 15.

In the above description of a best mode active three-axis control for a suspended car, considerable attention has been devoted to the details of this particular mode and the teachings of its practical implementation. However, it should be recalled that we have previously indicated that there are any number of additional different approaches to implementing the subject invention, which is active control of a suspended car. The basic principle of active control can be implemented in a plurality of coordinated single-axis controls, as discussed above. Recall that Fig. 18 illustrates the theory of operation of single-axis stabilization of the horizontal motion of a car suspended in an elevator frame. In connection with this, it has been proposed to use an accelerometer in a feedback loop to effectively increase the car mass by electromechanical means. One can use slow position and acceleration control loops to compensate for accelerometer offsets and the like. Fig. 24 shows a reduced block diagram of the same concept, and Fig. 25 shows an even further reduced model which is valid everywhere except for the lowest frequencies.

The diagram in Fig. 25 can be expressed in the following dimensions:

Acceleration of the cabin = [FD/G] [1/(M+Ka)],

where FD is the disturbance force,

M is the mass of the suspended cabin,

Ka the counter mass "added" by the actuator and

FD/G is the mass equivalent to the perturbing force using the acceleration due to gravity (G) on the Earth's surface.

If in the above equation we assume Ka=0, i.e. if we assume the absence of active control, and if M = 1,000 kg and FD/G = 25 kg, we obtain an acceleration due to the disturbance force (FD) of 25/1,000 = 25mG. If we now want to introduce active control, we can assume Ka = 9,000 kg and we now obtain a tenfold reduction in the acceleration due to the disturbance, namely 25/1,000 + 9,000) = 2.5 mG. From this we can conclude that if we follow the path specified here we will achieve at least an order of magnitude improvement in driving comfort.

If we now assume that the value Ka is to be aimed at being 9,000 kg, we can assign an acceleration scale factor (ASF) of 100 volts and a force generator scale factor (FGSF) of Ka/ASF, which has the value 9,000 kg/100 volts/G = 90 kg (force)/volt or 882 Newtons/volt.

An electromagnetic actuator as described above may be formed with a U-shape as shown in Figs. 29 and 30. In Fig. 29, dual coils 300, 302 are shown which fit onto legs 304 and 306 as shown in Fig. 30. Coils 300 and 302 form a continuous winding and are shown in Fig. 30 as an isometric section. Coil 300 and coil 302 may each comprise, for example, 936 turns of #11AWG magnet wire with a fill factor of 0.500. The U-shaped core may, for example, comprise a nested structure of 29 GA M6 laminations of 3.81 cm strip material with vacuum impregnation. The dimensions shown in Fig. 30 are, for example, A = 10.16 cm, B = 3.81 cm, C = 7.62 cm and D = 7.62 cm. The resistance is 6.7 ohms and the inductance 213 mH. Such a component weighs 22.2 kg and can generate 578 Newtons.

If we use an electromagnetic actuator in a control system as described above, we can expect an average command response delay of, for example, 4.2 msec. The time delay until full force of, for example, 578 Newtons is developed at a maximum air gap of 20 mm can be estimated as lsmsec as follows (based on the relation V = Ldi/dt):

Δt = LΔi/Δv = (0,3)(8,6)/(170) = 15 msec.

The time to develop the full force (578 Newtons) with a minimal air gap (5 mm) would be:

Δt = LΔi/Δv = (1,2)(2,15)/(170) = 15 msec,

so just as much.

The time to develop half the force would of course be half as long. An accuracy of the air gap signal of 10% of the full range can be tolerated. We can show the relationship between the air gap and various other factors in graphic form as shown in Figs. 26, 27 and 28. The maximum power is 500 watts with a maximum permissible air gap of 20 mm. The average power is expected to be 125 watts. From a short-term thermal point of view, the mass of copper in such an electromagnet is 14.86 kg with a specific heat of 0.092 cal/g-ºC (=3853/kgºC). The temperature change for feeding 500 watts of energy for sixty seconds is therefore:

ΔT = Watt-sec/(385)(14.86)

= (500)(60)/(385)(14.86)

ΔT = 5.24ºC.

So there is only a slight temperature increase even with maximum power input for one minute.

Fig. 31 shows a single axis side vibration stabilization system for use in a system as shown in Fig. 14D. The concept is the same as that shown in Figs. 18, 24 and 25. The implementation is shown as analog, but it will be understood that it can be done digitally. In this case, without limitation, a platform 352 is attached to the suspended car, while a pair of electromagnetic actuators 354 and 356 are attached to the elevator car frame. The accelerometer 358 measures accelerations of the suspended car and provides a measurement signal on a line 360 to an amplifier 362 which in turn provides an amplified measured acceleration signal on a line 364 to a summing node 366 where it is "summed" with a disturbance force signal on a line 368 and a position loop correction or "error" signal on a line 370. A resulting summed signal on a line 372 is fed to a pair of rectifiers 374 and 376 shown in Figs. 33 and 34. The rectifier 374 provides a signal on a line 378 to a signal inverter 380, also shown in Fig. 34, which produces a signal on a line 382 which may be referred to as a negative control signal. In a similar manner, the rectifier 376 delivers a signal on a line 384 which can be referred to as a positive control signal. Both signals on lines 382 and 384 are summed at summing nodes 388 and 390 respectively with a preset signal on a line 392. These deliver biased control signals via lines 394 and 396 to electromagnetic actuators 398 and 390 respectively. 400. The actuators 398 and 400 are or may be similar to those shown in Fig. 20.

The effect of the command signal on line 392 is illustrated in Fig. 32, which shows the composite resultant force (in dashed lines) on two forces on each side of reaction plate 352 (solid lines) versus the control signal (FC) for the system of Fig. 31.

This method is used to avoid discontinuities in control around the zero position. Without preloading, one magnet could be turning off and the other on at the same time. The method shown using preloading results in reduced control gain at or near zero force. The advantage is that only one magnet changes from on to off at any given time, and vice versa. Preloading helps to ensure that the cabin does not "jitter" during periods when little or no correction is required.

The signals on lines 394 and 396 can be thought of as force command signals similar to the force command signal on line 382 in Fig. 20. Similarly, actuators 398 and 400 provide actuator output signals over lines 402 and 404 to actuators 356 and 354, respectively, in a manner similar to the output signal on line 500 in Fig. 20.

Similarly, each actuator 398 and 400 provides position output signals 406 and 408 corresponding to the air gap signal on line 396 in Fig. 20.

For the position loop, both position signals on lines 406 and 408 and the corresponding, but oppositely located Rectification signal on lines 384 and 382 is applied to a pair of FC- controlled clamp circuits 410 and 412 to obtain a selection of the allowable position signal (both P+ and P represent position signals, however only the position signal corresponding to the driven force generator is valid).

The output signals of the clamp circuits are fed to a summing node 414 for the purpose of obtaining the valid position signal. The summing node 414 also receives an attenuated acceleration signal on line 415 from an attenuator 415a which is responsive to the amplified acceleration signal on line 364.

Both the FC controlled clamps 410 and 412 and the summing node 414 are shown in greater detail in Fig. 35. The output of the summing node is a composite position and acceleration signal on line 416 which is fed to a low pass filter 418 having a time constant in the range of 1 to 10 seconds. The low pass filter 418 in turn provides the correction signal to the summing node 366 on line 370 as described above.

For the actuators in Fig. 14D, instead of the combined three-axis schemes discussed above in connection with Figs. 14C and 15, one can provide three single-axis controls as just described. However, the three-axis scheme has numerous advantages. Among them is the stabilization of all sensitive axes using a minimum number of electromagnets. In addition, the gap movement carries cosine 45º = 0.707 for movement in the x or y direction. Thus, a gap variation of plus or minus 15 mm for a single-axis or multi-axis system is reduced to a variation of plus or minus 10.5 mm for a three-axis scheme. In Figs. 14C and 15, only four electromagnets are used, and in addition only four power electronics actuators are needed. Since magnets in Figs. 14C and 15 are used two at a time, the magnet size can be reduced. Using magnets that are only half the size of those needed for the single axis method is sufficient for a commercially viable system.

Fig. 36A is another block diagram of a specific implementation of an active suspension system according to the invention. There is a separate feedback loop for one axis (e.g., side-to-side along the X-axis passing through the vertical centerline of the car) and for the other axes (e.g., front-back along two axes parallel to the Z-axis on, for example, equidistantly opposite sides of the vertical centerline), but only a single axis is shown in Fig. 36A. An acceleration reference signal can be input via line 100 and set to zero. The difference between the reference signal on line 100 and a measured car acceleration signal on line 102 forms an error signal on line 104, which in turn is input to a feedback capacitor 106, designated C(s).

The summing node 109 is shown to be responsive to the capacitor output signal and to indirect type disturbances, such as rail disturbances, as provided by the system dynamics at 109a, and direct type disturbances, such as wind forces, as provided by the system dynamics 109b.

C(s) can be characterized as follows:

The first term takes into account the DC component cancellation, the second stands for the pendulum compensation and the third for the control.

The feedback capacitor in each axis processes the car acceleration error signal for that axis to generate actuator commands. In this case, a force command signal is generated on line 110. These compensators can be viewed as dynamic filters whose characteristics (gain and phase versus frequency) are set to meet the requirements of the elevator system. The design of C(s) can be translated into design requirements for the total loop gain, denoted L(s), which is the product of C(s) and the system dynamics, represented by block 108, denoted G(s).

Conceptually, the force command signal is given on line 110 to the system dynamics 108. As shown in Fig. 36B, in those areas where the car acceleration is to be minimized, the loop gain is chosen to be greater than one, ie in a zone around a frequency ω0. At higher frequencies above ω2, the loop gain is "gradually lowered" to meet stability-robustness requirements. At low frequencies below ω1, the loop gain is "cancelled" to reduce the effects of sensor noise and drift.

Based on the model of the shuttle car according to US Patent 4,899,852, we have performed an analysis of the behavior of an active suspension concept. Figures 37A and 37B are a graphical representation of the designed open loop transfer function for an active system, L(s), that resulted from this analysis. A graphical representation of the feedback compensator gain and travel angle is shown in Figures 38A and 38B, respectively, for this particular design.

Figures 39 through 41 summarize the results of a simulation study of the performance of the resulting active suspension system. The plots at the top left of each of these figures (labeled a) is the specific disturbance input (direct car force or slew profile). The remaining plots in each of these figures show the car acceleration response for three configurations: (1) shuttle car (top right, labeled (13)), (2) conventional car (bottom left, labeled (c)), and (3) active suspension using the control arrangement of Figures 36A, 36B, 37A, 37B, 38A, and 38B (bottom right, labeled (d)). It can be seen that this system reduces the levels of car platform acceleration compared to conventional systems and even to a shuttle car that does not have active control.

Fig. 39 is the predicted response to butt joint misalignment.

Fig. 40 is the predicted response to rail undulations.

Fig. 41 is the predicted response to car force disturbances.

Tests were conducted using a half-scale model of the elevator system shown in Fig. 14A and 14B. The effectiveness of the concept was demonstrated using a rotating unbalance supported on the frame to simulate rail-induced disturbances and by simulating a direct disturbance force using an actuator between the frame and the car.

The feedback capacitor 106 shown in Figure 36A was implemented using a digital computer 100 as shown in Figure 42. Data from the sensor 102 is provided over a line 103 to a 12-bit A/D converter (not shown) for processing after passing through an input/output port 104. A digital form of the feedback capacitor 106 was used to generate a command signal on a line 106 for an actuator 108 capable of applying forces to a car 110.

The signal processor 100 includes a central processing unit 104a, a read-only memory 104b, a read/write memory 104c, and a data, address, and control bus 104d, via which the aforementioned components communicate.

Fig. 43 is an illustration of a series of steps that may be performed by the signal processor 100 of Fig. 42. For example, after entering step 112, an acceleration is measured using an accelerometer 102 in step 114. The processor then calculates the magnitude of a reaction force by implementing the dynamic compensation 106. Next, a step 118 is performed in which the signal processor 100 provides the reaction signal on line 106, which is a reaction force signal, for example, the signal on line 106 as calculated in step 116. Exit occurs in step 120.

Fig. 44 shows the results of a test to verify the effectiveness of the active system in mitigating direct car forces. The plot shows the relationship of the measured car acceleration to the magnitude of the sinusoidal input force over a range of frequencies. The upper plot corresponds to the response of the pendulum car system without active suspension control. The lower plot, the response of the active suspension system, verifies the expected 80% to 90% reduction. Fig. 45 shows a comparison of the predicted time-dependent response (by simulation in Figs. 45B and D) and the actually achieved (experimentally) time-dependent response (Figs. 45A and C) in mitigating direct car forces.

This reduction in system sensitivity to direct car forces was achieved without sacrificing system performance in the presence of rail-induced disturbances. Figure 46 illustrates the system response to rail irregularities simulated on a rotating imbalance test rig. Figure 46A shows the unmagnified shuttle car response at a 3 Hz application frequency. Figure 46B is the response of the active suspension system. It can be seen that the magnitude of car acceleration was reduced. Thus, the active suspension system improves ride quality using car acceleration as the measure of performance in the presence of both rail-induced disturbances and direct car forces.

In any event, the actuator command signal on line 106 of Fig. 42 may be a force command signal, as already explained above, and as shown in greater detail in Fig. 47, to be applied to a specific actuator which may, for example, include the core coils and ferromagnetic plates of Figs. 29 and 30.

In Fig. 47, the force command signal on line 106 is applied from the signal processor 100 in Fig. 42 to a PWM amplifier, which may be a device manufactured by Copley Controls Corp. of 375 Elliot Street, Newton, Massachusetts, U.S.A. and designated "Class B PWM Servo Aplifier Model 218A" as described in the published specification entitled "Model 215A, 218 Servo Amplifier". The PWM amplifier 150 is also responsive to a force feedback signal on a line 152. The PWM amplifier serves as a double-pole, double-position electronic switch to apply a selected voltage to lines 154 and 156 with a polarity reversal at a selected duty cycle. A pair of steering diodes 158 and 160 direct the output current on line 154 or 156 into the correct coil of the corresponding magnet 130 and 132. It will be seen that either the ferromagnetic mass 134 or the electromagnets stand on the bottom of the frame 126 while the other element stands on the bottom surface of the platform 14. In a preferred embodiment, the electromagnets 130 and 133 stand on the bottom of the frame 126 while the ferromagnetic mass 134 is fixedly attached in a downwardly hanging manner to the bottom of the platform 10b. This also applies to the remaining actuators 240, 244 in Fig. 14D.

To obtain an effective feedback loop, Hall cells 170 and 172 may be placed in the magnetic circuit to sense the magnetic flux in the air gap between the ferromagnetic plate 134 and the associated cores of the core coils 130, 132. Hall cells 170 and 172 provide measured flux density signals over lines 174 and 176 to a device 182, which may be a multiplier IC, e.g. an AD-534, which squares the magnitudes of the flux signals on lines 174 and 176 and provides a difference signal indicative of the difference between these signals as a force feedback signal on line 152.

Here too, the control shown in abstract form in Fig. 36A and 36B can be implemented in numerous different ways, including the fully digital version similar to Fig. 42; however, a preferred version is shown in Fig. 48. The embodiment of Fig. 48 includes two such independent control systems, each for controlling actuators on opposite sides of the car, identical or similar to those shown in Fig. 15, one system for the floor or near the floor of the car and the other system for the ceiling or near the top of the car.

Of course, the basic principle of active control can be implemented in a plurality of coordinated single-axis controls, as suggested above. Consequently, the control shown in Fig. 36A in an abstract single-axis version can be implemented in many different ways, but a preferred version follows the same principles as disclosed for the three-axis control shown in Fig. 15 to arrive at the five-axis control shown in Fig. 48. Although shown for the elevator car, the principles shown in Fig. 48 can also be extended to a pendulum or supported car, as will be apparent to those skilled in the art.

In Fig. 48, fast analog loops for rapidly counteracting disturbing forces are combined with slower but more accurate digital loops for compensating for gravity components and drift in the accelerometers. Several such fast analog loops can be implemented as analog controllers 500, 502, 504 and 506 to independently control the top of the car, while analog controllers 508, 510, 512 and 514 are used to independently control the bottom of the car, one controller for each of the eight actuators 516, 518, 520, 522 and 524, 526, 528 and 530, respectively. With suitable interfaces (not shown), a single digital controller 532 can handle the signals to be described going to and coming from the analog controllers. Each analog control responds to a force command signal on lines 534, 536, 538 and 540 as well as 542, 544, 546 and 548 from the digital controller 532. The force command signals have in the Typically different amounts depending on the transverse and rotational forces to be counteracted. The digital controller 532 in turn responds to acceleration signals on lines 552, 558, 559 and 550, 554, 556 from the accelerometers 562, 568, 569 and 560, 564 and 566, as well as to position signals on lines 570, 572,574, 576 and 578, 580, 582 and 584, respectively, which are indicative of the size of the air gaps between the cores of the actuators 516, 518, 520, 522 and 524, 526, 528 and 530 on the one hand and the associated, opposite ferromagnetic webs on the other.

In response to the force command signals on lines 534, 536, 538, 540 and 542, 544, 546, 548, the respective analog controls 500, 502, 504, 506 and 508, 510, 512, 514 provide actuation signals on lines 586, 588, 590, 592 and 594, 596, 598, 600 to the coils of the actuators 516, 518, 520, 522 and 524, 526, 528, 530 in order to generate more or less strong attractive forces between the respective actuator cores and the ferromagnetic webs associated with them. The return current through the coils is monitored by current monitoring components 602, 604, 606, 608 and 610, 612, 614, 616, which supply current signals via lines 618, 620, 622, 624 and 626, 628, 630, 632 to the respective analog controls 500, 502, 504, 506 and 508, 510, 512, 514. The current sensors can be, for example, of the type Beil IHA-150.

A plurality of sensors 634, 636, 638, 640 and 642, 644, 646, 648, which may be Hall cells (e.g. of the type Beil GH-600), are associated with the actuator cores accordingly to provide an indication of the flux density or magnetic induction (volt-sec/m²) in the air gap, i.e. between the end faces of the cores and the associated webs or, in other words, the flux density in the air gaps between these parts. The sensors 634, 636, 638, 640 and 642, 644, 646, 648 provide read signals via lines 650, 652, 654, 656 and 658, 660, 662, 664 to the analog controls 500, 502, 504, 506 and 508, 510, 512, 514.

The analog controller 500 among the several analog controllers shown in Fig. 48 is similar or identical to that shown in more detail in Fig. 20. The remaining analog controllers may be of the same or similar construction.

As explained above, the digital controller 532 is responsive to the air gap signals on lines 570, 572, 574, 576 and 578, 580, 582, 584 and to the acceleration signals on lines 552, 558, 559 and 550, 554, 556, in conjunction with an analog controller as shown in Fig. 20, to perform the single axis control functions of Fig. 36A in five axes, i.e., translations along two horizontal axes at both the floor and ceiling, rotations about the same two axes at the floor and ceiling, and rotations at the floor and ceiling about a vertical axis.

To be more precise, to best carry out this aspect of the invention, we describe control operations with respect to nine axes, namely, two translational axes and two rotational axes on both the floor and the ceiling, and one rotational axis with respect to the vertical, common to both the floor and the ceiling. However, if the horizontal axes in the floor and ceiling are approximately represented for the purposes of illustration by a single set of horizontal axes in a plane midway between the top and the floor of the car or cab, we can speak of "five-axis" control. For the purpose of simplifying the description, but not by way of limitation, we can therefore consider the car or cabin, regardless of the actual rigidity or lack thereof in the structure between the floor and the ceiling, as a solid or rigid cube with a three-axis Cartesian coordinate system, whose origin is at the center and is subjected to translations along and rotations around the horizontal axes as well as rotations around the vertical axis.

The force command signals both at the floor and at the top of the car or cab can be generated, for example, by first resolving the sensed position (gap) signals into components along the axes of the Cartesian coordinate system 30 shown in Fig. 3 (which have their origin in the plane of the floor or ceiling, depending on which independent control system is considered), according to the following equations:

Px+ = (P₁ + P₃) / (2KCS), Px- = (P₂ + P₄) / (2KCS),

Py+ (P₁ + P₂) / (2KCS), Py- = (P₃ + P₄) / (2KCS),

Pθ+=(P₂+P₃)/2, Pθ- =(P₁+P₄)/2,

and then, based on the above, calculate the values Px, Py and Pθ (which together indicate the absolute position of the car or cabin) from Px- and Px+, Px- and Px+ and Pθ+ and Pθ-. For example, Px can be calculated as follows:

Px = (Px+ - Px-)/2.

One may also select Px+ or Px- depending on which is smaller. (Note: for large air gaps, i.e. large values of Px+ or Px, the value is likely to be inaccurate and can be ignored.) The resulting components are used to determine the position control force components Fpx, Fpy, Fpθ as shown in Fig. 48 on a single axis basis ("P" stands for position feedback). Px, for example on line 348, is compared with a reference quantity on line 352 to produce an X position error signal on line 354. This in turn is passed through a low pass filter, for example filter 358. This provides a signal Fpx for the purpose of resolving the required X reaction force; if a positive force is required, Fp1 = Fp3 = (0.5)(Fpx)/(cos 45º). For a negative force, Fp2 = Fp4 = (0.5) (Fpx)/(cos 45º) applies. The same procedure can be followed for Fpy and Fpθ, using the corresponding equations. This allows the force components Fpx, Fpy and Fpθ are resolved into correction signals Fp1, Fp2 and Fp3 and Fp4 according to the following complete set of equations:

Fpx+: F1; = (KCS) (Fpx+) Fpx-: F₂ = (KCS) (Fpx-)

F3; = (KCS) (Fpx+) F₄ = (KCS) (Fpx-)

Fpy+: F1; = (KCS) (Fpy+) Fpy-: F₃ = (KCS) (Fpy-)

F2; = (KCS) (Fpy+) F₄ = (KCS) (Fpy-)

Fpθ+: F₂ = (KCS) (Fpθ+) Fpθ: F₁ = (KCS) (Fpθ-)

F3; = (KCS) (Fpθ+) F₄ = (KCS) (Fpθ-)

where F = force and

KCS = cos(45º) = sin (45º) = 0.707,

which are then summed with the acceleration signals F₁, F₂, F₃, F₄ (e.g., the signal on line 364 or line 382 in Fig. 20) generated in the manner described above.

It should be recognized that a valid position measurement is only available from the flux sensors of the type described when the associated force actuator is driven. This means that any processing algorithm must depend on whether or not there are any magnetic gap currents.

An additional teaching of our invention is that the electromagnets are used to control the position of the car or cab at stops, for example to bring the suspended or supported car or cab into a rest position relative to the frame while passengers are boarding and alighting. Of course, the signal processor according to Fig. 16, the digital controls 380, 532 according to Figs. 19 and 48 or an additional signal processor can perform additional control functions, for example starting and stopping cars and dispatching cars. In the case When stopping at a floor, it can receive a reading signal via line 18 or an algorithmically determined but similar signal indicating the rest position of the car, and then emit a signal via line 22 to control the position of the suspended or supported car or cabin. To repeat the previous example, if the car platform 200 of Fig. 14C is oriented in the elevator shaft such that the left vertical edge of the car means that the threshold of the car is aligned with a door threshold of the elevator shaft 700, the signal processor 20 of Fig. 16 may be programmed to provide force command signals to the actuators 210 and 214 to provide the attractive forces needed to urge the suspended car against, for example, stops 702 and 704 mounted on the car frame 202, thereby bringing the threshold of the car into a rest position relative to and in close alignment with the entrance threshold of the elevator shaft after the frame 202 has assumed its rest position.

The method by which this is accomplished is illustrated in Figure 51, where in a step 720, a stop signal is provided by means 722 (which may be included in processor 532 as an additional means for controlling a car or group of cars) to indicate that the car frame is at rest to provide a stop or a stop command signal, so that in response an actuator 724 (which may be the jointly acting actuators 210 and 214) provides an actuation signal in a step 726 to cause a suspended car 728 (which may be the car 200) to come to rest with respect to the car frame (which may be the frame 202) so that the car threshold is adjacent to and motionless with respect to the hall threshold.

A similar set of stops 730, 732 may be provided at each landing for the car according to Fig. 15 to serve as a thrust abutment, and a similar procedure as in Fig. 51 may follow.

It should be understood that although a preferred embodiment of the invention uses electromagnetic non-contact actuators, particularly in conjunction with a suspended or supported elevator car, employs electromagnetic actuators such as those shown in Figure 15 in conjunction with shaft rails, it is also possible to use active contact actuators. For example, Figure 49 shows a standard rail 750 attached to an elevator wall 752 and having three contact actuators with wheels 754, 756 and 758 in contact with the rail for guiding an elevator car. Figure 50 shows one of the actuators 760 in detail. It has a wheel 754 which is actuated by an electromagnet 762 having a coil 764 similar to a coil that would be used in an electromagnetic actuator of the non-contact type disclosed above. In other wheels 756 and 758 similar electromagnets are associated.

Returning to Fig. 59, it is of course common practice to provide one or more guide rails for each elevator car in an elevator system to guide and stabilize the car as it moves up and down between floors. Such guide rails typically carry spring-mounted wheels on the car for the purpose of guiding and controlling the movement of the car.

Since an elevator car must be guided so that it remains on a given path of movement, the guide rails must withstand all the forces that may arise from the normal operation of the elevator and during emergency stops. In general, the guide rail can have any shape that is capable of keeping the elevator on its path. The problem then is to find the most economical shape.

Currently, it is common for such rails to have an inverted T-section, with the elevator car having three wheels that different parts of the guide rail, typically the arrangement shown in Fig. 49. As can be seen from this figure, two of the wheels are directly opposite each other, while the third rolls on the web side of the "T" between the other two wheels.

However, one problem with this standard inverted "T" rail is that it is relatively heavy compared to its strength and therefore does not have the high strength to weight ratio that one would like.

Preferably, therefore, an improved guide rail profile is used which provides at least the same level of strength at a relatively low weight or at least a greater strength-to-weight ratio than the standard inverted "T" rail cross-section, but still allows a comparable (if not equal or greater) level of guidance and stabilization control to that achieved with the conventional standard profile. Furthermore, the improved profile is particularly suitable for the use suggested in Figs. 5 and 7.

Although other people, including a relative of the actual "father" of the elevator, have proposed other configurations for the guide rails, such as one with a "V" cross-section in which two wheels roll on the outer, diverging sides of the "V" (see, for example, US Patent No. 134,698, dated January 7, 1873, by Charles R. Otis), it is believed that these profiles are not capable of achieving the desired combination of control with as good a strength-to-weight ratio as the present profile.

Another known configuration is a round tube shape, similar to a pipe, but this profile was basically limited to cases where an emergency stop by safety brakes is not required. Other possible shapes for guide rails included, for example, the shape of a "hat", similar to an inverted and flat-bottomed "U" with end flanges, a "bell" shape and a rectangular hollow shape.

For further background and general information on general elevator car guidance systems and their layouts and relationships to other parts of an elevator system using the generally adopted standard T-section, see, for example, U.S. Patent 4,793,441 issued December 27, 1988 to Cilderman et al, entitled "Elevator Car System with Three Guide Rails." See also, for example, U.S. Patent 3,669,222 issued June 13, 1972, entitled "Guiding and Dampening Device" to Takamura et al, and U.S. Patent 4,754,849 issued July 5, 1988, entitled "Control System for Elevator Cage Guide Magnets" to Ando.

As can be seen from Figs. 53 and 54, the preferred embodiment of the elongated guide rail 1 contains a base or web section 2 with two preferably integrally formed, diverging webs or legs 3 which end in end flanges 4. The end flanges serve to attach and fasten the guide rail 1 to the building. The two diverging legs form a cavity 3A between them.

As can be seen from Fig. 53, three guide wheels 5A, 5B and 5C cooperate with the base or web section 2 in a similar manner and in a similar configuration as they did with the "T"-shaped guide rail, as can be seen from Fig. 52. The two opposing wheels 5A and 5B rest on the opposite sides of the web section 2, while the intermediate wheel 5C rests on the free end of the web section.

The characteristics of the T-shape and the Y-shape to be compared are compared in detail using the sketch in Fig. 55C, whereby a sixty-six and seventy-two/100 Newton (66.72- NT) T-rail was used as an example for comparison purposes. The tables 1 and 2 of Fig. 55 show the respective dimensions and characteristics of the two shapes, from which it can be seen that the corresponding Y-shape can replace the present state of the art "T" rail with a weight saving of approximately eighteen (18%) percent.

As further shown in Table 2, the Y-shape is slightly better in cross-sectional properties than the corresponding T-shape, yet weighs eighteen (18%) percent less. The overall dimensions of both shapes are essentially the same (Table 1), twelve and seven-tenths of a centimeter by eight and 89/100 of a centimeter (12.7 cm x 8.89 cm) for the T-shape and thirteen and 02/100 of a centimeter by nine and 53/100 of a centimeter (3.02 cm x 9.53 cm) for the Y-shape.

It should be noted that all of the current equipment, i.e. roller guides, sliding surfaces, safety features, etc., used in conjunction with the T-shaped rail can also be used in the Y-shape without any modification. It should also be noted that the legs or webs 3 according to Fig. 53 can be used to support wheels or to provide electromagnetic flux, as can be seen from Fig. 54.

The Y-shaped rails can be rolled in steel mills into pieces of desired lengths, for example four and 88/100 meters (4.88 m). Machining of the webs can be easily done in rolling or finishing machines, similar to that used for the "T" rail. Cutting appropriate slots at the end and drilling holes in the "Y" guide rail is typically done in the same way and just as easily as for the "T" rail. Figures 56 and 57 are still further illustrations of an embodiment of an "active" roller guide device for carrying out the present invention, showing details of a roller group 1000. Although one of the rollers (the side-mounted one) is raised relative to the other two, it will be appreciated that the roller group 1000 is a relatively conventional configuration of rollers on a rail 1001.

However, we are only aware of groups being used passively and we are not aware of a state-of-the-art roller group being used with actuators.

The assembly 1000 includes a side guide roller 1002 and fore-and-aft guide rollers 1004 and 1006. The roller assembly 1000 is supported on a base plate 1008 which in turn is fixed to an elevator car frame crosshead (not shown). The guide rail 1001 is a conventional generally T-shaped construction having base flanges 1010 for attachment to the shaft walls 1012 and a web portion 1014 projecting into the shaft toward the rollers 1002, 1004 and 1006. The web 1014 has a free end surface 1016 which cooperates with the side roller 1001 and side surfaces 1018 which cooperate with the fore-and-aft rollers 1004 and 1006. The guide rail web 1014 extends through a slot 1002 in the roller group base plate 1008 so that the rollers 1002, 1004 and 1006 can engage the web 1014.

As best seen in Fig. 57, the side roller 1002 is pivotally mounted on a link 1022 which is pivotally mounted on a base 1024 by a pivot pin 1026. The base 1024 is attached to the base plate 1008. The link 1022 includes a cup 1028 which receives one end of a coil spring 1030. The other end of the spring 1030 engages a spring guide 1032 which is coupled by a bolt 1036 to the end of a telescoping ball screw adjuster 1034. The adjuster 1034 can be extended or retracted to vary the force exerted by the spring 1030 on the link 1022 and thereby the roller 1002. The ball screw device 1034 is mounted in a fork 1038 attached to the platform 1040 with bolts, whereby the platform in turn is attached to the base plate 1008 by means of supports 1042 and 1044. The use of the platform 1040 and the supports 1042 and 1044 enables the arrangement to be retrofitted to a conventional roller guide arrangement directly on the existing Base plate 1008. The ball screw assembly 1034 is driven by an electric motor 1046. A ball screw actuator useful in the practice of this invention is available from Motion Systems Corporation, Box 11, Shrewsbury, New Jersey 07702. The actuator motor 1046 may be an AC or DC motor, both of which may be purchased from Motion Systems Corporation. The Motion Systems Model 85151/85152 actuator has been found to be particularly suitable for use in the practice of the invention. These components have the AC or DC motor 1046 attached to a reduction gear 1048 which serves to reduce the motor speed and drive the ball drive actuator, which is a recirculating ball gear 1034, only the cover of which is shown. Alternatively, a brushless DC motor may be provided. Although shown only schematically, a position sensor 1049, such as a potentiometer or an optical sensor, may be mounted on the car frame by attachment to the reduction gear 1048 on a lip on the back of the spring holder 1032 to measure the linear extension of the screw. Of course, any other position sensor may be used.

The guide roller 1002 is rotatably mounted on an axle 1050 which in turn is mounted in an adjustable receptacle 1052 in the upper end of the link 1022. A pivot stop 1054 is mounted on a threaded rod 1056 which extends through a passage 1058 in the upper end 1060 of the platform 1024. The rod 1056 is threaded into a bore 1062 within the link 1022. The stop 1054 is designed to limit the travel of the link 1022 in a counterclockwise direction with respect to the pin 1024 by selectively engaging the platform 1024 and thereby limit the travel of the roller 1002 away from the rail; this direction is indicated by the arrow D. The pedestal 1024 is equipped with a recess 1064 which contains a magnetic button 1066 which is made of a rare earth compound. Samarium-cobalt is a rare earth compound which is used for the magnetic button 1066. A steel tube 1068 containing a Hall effect detector (not shown) at its end 1070 is mounted in a passageway through the connector 1022. The magnetic button 1066 and the Hall effect detector together form a proximity sensor which is operatively coupled to a switch which controls the power supply to the electric motor 1046. The proximity sensor senses the distance between the magnetic button 1066 and the steel tube 1068, which distance reflects the distance between the pivot stop 1054 and the pedestal 1024. As the tube 1068 and its Hall effect detector move away from the magnet 1066, the pivot stop 1054 moves toward the pedestal 1024. The detector produces a signal proportional to the size of the gap between the detector and the magnetic button 1066, and this signal is used to control the electric motor 1046, causing the ball jack 1034 to move the link 1022 and the roller 1002 toward and away from the rail as is the case. Depending on the type of control system used, the stop 1054 can be prevented from touching, or at least prevented from having prolonged contact with the platform 1024. This ensures that the roller 1022 is constantly dampened by the spring 1030 and does not rub against the base plate 1008 by the stop 1054 and the platform 1024. Side tilting of the car due to asymmetric passenger loading or other direct ride forces is also corrected. As stated, the electric motors 1046 can be reversible motors, allowing adjustments on each side of the car to be coordinated in both directions, both toward and away from the rails.

Referring now to Figs. 56, 57 and 58, the mounting of the forward and backward rollers 1004 and 1006 on the base plate 1008 will be explained. Each roller 1004 and 1006 is mounted on a connecting member 1070 which is coupled to a pivot pin 1072 which carries a crank arm 1074 at its end facing away from the roller 1004, 1006. Axles 1076 of the rollers 1004 and 1006 are mounted in adjustable recesses 1078 in the connecting members 1070. The Pivot pin 1072 is journalled in two-piece bearing bushes 1080 which in turn are seated in grooves 1082 formed in a base block 1084 and a cover plate 1086 which are together bolted to the base plate 1008. A flat spiral spring 1088 (see Fig. 49) is journalled in a space 1089 (see Fig. 56) and has its outer end 1090 connected to a rotatable collar (not shown) which is rotated by a gear set (not shown) in a gear box 1094. The gear set is rotated in one direction or the other by means of a reversible electric motor 1096. The spiral spring 1088 is the suspension spring for the roller 1006 and provides the spring preload which urges the roller 1006 against the rail web 1018. The spiral spring 1088, when rotated by the electric motor 196, also provides the return impulse to the roller 1006 via the crank arm 1074 and the pivot pin 1072 to offset tilting of the car in the forward-backward direction caused by direct forward and backward car forces, such as asymmetric passenger loading of the car.

A rotary position sensor (not shown), such as an RVDT, rotary potentiometer, or the like, may be present to perform the function of sensor 1049 in Figure 57. Such a sensor may be coupled at one end to crank arm 1074 and at the other end to base 1008.

Each roller 1004 and 1006 may be independently controlled by associated electric motors and coil springs as shown below in Fig. 69, if desired, or the rollers may be mechanically coupled and controlled by just one motor/spring assembly as shown in Figs. 56 and 60. Details of an operative connection of the rollers 1004 and 1006 are shown in Fig. 60. In Figs. 58 and 60, it can be seen that the links 1070 include a downwardly directed fork 1098 with bolt holes 1100 formed therein. The link fork 1098 extends downwardly through a gap 1102 in the bearing plate 1008. A collar 1104 is connected to the fork 1098 via a bolt 1106. A connecting rod 1108 is telescoped through the collar 1104 and secured therein by a pair of nuts 1109 which are screwed onto the threaded end portions of the rod 1108. A coil spring 1110 is mounted on the rod 1108 to bias the collar 1104 and hence the connecting member 1070 counterclockwise about the pivot pin 1072 as shown in Fig. 60. It will be understood that the opposite roller 1004 has an identical connecting member and collar arrangement which is coupled to the other end of the rod 1108 and biased clockwise by the spring. It can be seen that the clockwise movement of the link 1070 caused by the electric motor 1096 also results in a counterclockwise movement of the opposite link due to the connecting rod 1108. At the same time, the spring 1110 allows both links to pivot in opposite directions if this is necessary due to discontinuities in the rail web 1018. A flexible and smooth ride is achieved even when the two roller links are coupled together by a connecting rod.

As shown in Fig. 60, a stop and position sensor assembly similar to that described above is mounted on the link 1070. A block 1112 is bolted to the base plate 1008 below an arm 1114 formed on the link 1070. Mounted on the block 1112 is a cup 1116 containing a magnetic button 1116 made of a rare earth element, such as samarium cobalt. Mounted in the passage 1120 of the link arm 1114 is a steel tube 1118 which receives a Hall effect detector in its lower end, thereby completing the proximity sensor which monitors the position of the link 1070. A pivot stop 1122 is mounted on the end of the link arm 1114 opposite the block 1112 to limit the extent of possible pivotal movement of the link 1070 and the roller 1006 away from the rail web 1014. The distance between the pivot stop 1122 and block 1112 is proportional to the distance between the Hall effect detector and the magnetic button 1116. The Hall effect detector provides a feedback signal to activate the electric motor 1096, for example, whenever the stop 1122 approaches block 1114 to within a preset distance, whereupon the motor 1096 pivots the link 1070 via the coil spring 1088 to move the stop 1122 away from block 1114, or, to give another example, to be compared to a reference quantity in a proportional, proportional-integral or proportional-integral-derivative control loop, so that the difference therebetween is then more or less continuously returned to zero by the control loop. The position sensor 1049 in Fig. 57 can also be used to track the position of the actuator relative to the base 1008, as described below in connection with Fig. 69. In any case, this movement presses the roller 1006 against the rail web 1014 and pulls the roller 1004 via the connecting rod 1108 in the direction of arrow E in Fig. 60. The simultaneous displacement of the rollers 1004 and 1006 results in any canting or tilting of the elevator car in the forward-backward direction, caused for example by an asymmetrical passenger distribution, being compensated for.

Referring to Figures 56, 57 and 61, an electromagnet having coils 1130, 1132 is mounted on a U-shaped core 1134 which in turn is mounted on the carrier 1044. The carrier 1044 itself is mounted on the base plate 1008. As previously described, the ball drive shaft 1034 exerts forces along the axis of the ball screw against the pivoted link 1022. The link 1022 pivots at point 1026 and extends below the pivot point down to the electromagnet coils 1130 and 1132 and has an end face 1138 which is separated from the core faces of the electromagnet core 1134 to receive electromagnetic flux across the air gap therebetween.

Fig. 62 is an illustration of the cup 1064, which should be made of ferromagnetic material, and in which the rare earth magnet 1066 is housed. The depth of the cup may be 15 mm, with an inner diameter of 25 mm and an outer diameter of 30 mm, as shown by way of example. The sleeve 1068 may have a length of 45 mm, with an inner diameter of 12 mm and an outer diameter of 16 mm, for example. A Hall cell 1140 is shown to be located near the opening of the tube 1068 to assume a position in which the flux of the magnet 1066 can be detected. The composition of the tube is ferromagnetic in accordance with the teachings of the present invention to enhance the ability of the Hall cell to sense flux from the magnet and also to provide a shield from flux created by electromagnets otherwise attached to the roller guide.

Specification for position transducers

1. Magnetic transducers can be used.

2. Operating range: 10 mm

3. Repeatability : 0.1 mm

4. Temperature range: 0-55C

5. Temperature coefficient: < 0.02 % / C

6. Sensitivity to Magnetic field: 100 Gauss at a distance of 30 mm should not affect the transducer output by more than 0.5%.

7. Operating voltage: 9 - 15 VDC

8. Cables: Use separate signal and power grounds and use twisted shielded pairs.

Fig. 63 shows a Hall cell 1140a mounted on one side of the reaction plate 1138, with a projection 1134a of the electromagnet core 1134 on the plate 1138 facing the coil 1130 (also in a projection 1130a 56, 57 and 61. The sensor can also be placed on the face of the core itself, but in this position it could become overheated.

Specification for the Hall sensor arrangement

1. It is attached to or opposite the front of the electromagnet.

2. Operating range: 0.5 to 1.0 Tesla

3. Accuracy: 5% tolerable, 2% desirable

4. Scale factor: 10 V/Tesla

5. Temperature range: 0 - 55 C

6. Temperature coefficient: < 0.02 % / C

7. Thickness: must not exceed 2.0 mm

8. Supply voltage: ± 12 to 15 VDC

9. Cables: use separate signal and power grounds and use twisted shielded pairs.

Referring now again to the forward-reverse roller 1006, a pair of electromagnets 1144 and 1146 are shown in Fig. 57. A block portion 1148 of the link 1070, shown in perspective in Fig. 58 and in section in Fig. 60, has an extension 1150 shown in Figs. 57 and 60 (not in Fig. 58) having an end face 1152 opposite a pair of core faces associated with a core 1156 on which coils 1144 and 1146 are supported, only one end face 1154 being shown in Fig. 60.

Fig. 64 is a side view of a ferromagnetic core as used to support the coils 1130, 1132 of Fig. 56 or the coils 1144 and 1146 of Fig. 57. Dimensions are in millimeters. Fig. 65 shows a top view of the same core with the depth dimensions shown together with a pair of coils shown in phantom. The core of Figs. 64 and 65 may be made from a grain oriented (M6-)-29 steel, supported on an angle iron by welding, for example. For example, coils 1130 and 1132 are required in pairs, each having 1050 turns of wire of 1.15 mm diameter. The coil connection should be a series connection with the possibility of parallel reconnection. The wire insulation can be of heavy (double) construction GP200 or equivalent rated for 20ºC. The impregnation can be vacuum resistant at 18ºC or above. The coil voltage rating can be of the order of about 250 volts, the coil itself can be tested at high potential to earth at 2.5 kilovolts or similar as required. The coil conductors can be of stranded wire of 1.29 mm diameter and about 50 centimeters long for the preliminary connection. The weight is about 2.0 kilograms, of which 0.8 kg is iron and 1.2 kg is copper. With an air gap of 2-10 mm and a flux density of about 0.6 Tesla, a force of about two hundred Newtons is achieved. Such a design is appropriate for the active roller guide disclosed above. It has a force reserve of more than twice the required value.

Fig. 66 illustrates a pair of active roller guides 1140 and 1142 mounted on the floor of the elevator car 1144 for lateral control. Fig. 66 also shows a control for a corresponding pair of electromagnets 1146, 1148. Acceleration feedback is used in the electromagnet control circuit described, although other means of control may be used. Acceleration control is discussed in detail in connection with the position control of the high force actuators in connection with Fig. 69. An accelerometer 1150 measures the lateral acceleration at the floor of the platform, which may be located between two active roller guides 1140 and 1142. The direction of sensitivity of the accelerometer is indicated by an arrow SS and is perpendicular to the shaft walls. A measurement signal on a line 1152 is applied to a signal processor 1154 which in response thereto produces a force command signal via line 1156 to a second signal processor 1158, which may be comprised of discrete components for faster response. The force command signal on line 1156 is summed in a summer 1160 with a force feedback signal on line 1158 to provide a force error signal via line 1162 to a steering circuit comprised of a pair of diodes 1164, 1166. A positive force error signal causes diode 1164 to open, whereas a negative force error signal causes diode 1166 to open. To avoid abrupt turning on and off of the two electromagnets 1186 and 1184 near the junction between positive and negative force response as shown in Fig. 67, a bias voltage is provided to bias the left and right signals given to the PWM controller. This is done by means of a pair of summing elements 1168 and 1170 from a potentiometer 1172 which is biased with an appropriate voltage to achieve the force summation shown in Fig. 67. This allows a smooth transition between the two electromagnets. A pair of pulse width modulation controllers 1174 and 1176 are responsive to the summed signals from summers 1168 and 1170 and provide signals over lines 1178 and 1180 having variable duty cycles that depend on the magnitudes of the signals on lines 1182 and 1186, respectively, from summers 1168 and 1170, respectively.

The force feedback signal on line 1158 comes from a summer 1186 which is responsive to a first force signal on line 1188 and a second force signal on line 1190. A squaring circuit 1192 is responsive to a flux signal read by a Hall cell 1196 on line 1194 and provides a first force signal on line 1188 by squaring and scaling the flux signal on line 1194. Similarly, a squaring circuit 1198 is responsive to a flux signal read by a Hall cell 1202 on line 1200. The pair of Hall cells 1196, 1202 are mounted on or opposite one of the core faces of the associated electromagnets to be in a position to direct the flux between the electromagnet and the corresponding arms 1204, 1206 of the roller guides 1140, 1142.

The signal processor 1154 of Fig. 66 is programmed to perform the compensation described in detail in connection with Figs. 18 and 24.

The signal processor 1158 of Fig. 66 is shown in greater detail in Fig. 68, where an integrated circuit 1230, which may be an Analog Device AD534 device, is responsive to the force command signal on line 1156, the first flux signal on line 1194, and the second flux signal on line 1200 to produce the force error signal on line 1162 as shown in Fig. 66. A PI controller 1252 amplifies the force error signal and provides an amplified signal on line 1254 having a gain of 100 volts per volt (gain 100) to precision rectifier or diode control circuits 1164 and 1166, similar to that shown in simplified form in Fig. 66. An inverter 1258 inverts the output of steering circuit 1164 so that the signals on lines 1260, 1262 that are applied to summers 1168 and 1170 have corresponding polarities. The summed signals on lines 1182 and 1186 are applied to PWM controllers, which may be Signetics NE/SE 5560 controllers. These apply variable duty cycle signals to lines 1168 and 1180, which in turn are applied to high voltage gate driver circuits 1260 and 1262, which in turn provide gate signals to bridge circuits 1264 and 1266 that apply current to electromagnets 1146 and 1148.

Amplifiers 1268 and 1270 monitor the current to the bridge and, in response to an overcurrent, provide a shutdown signal to the PWM controllers 1174 and 1776.

In addition, a potentiometer 1272 can supply a reference signal to a comparator 1274, which calculates the output signal of the current sensor 1270 with the reference signal and provides an output signal over a line 1276 to an OR gate 1278 which provides the signal on line 1276 as a signal on a line 1280 to the high voltage gate driver 1272 when the signal from the current sensor 1270 exceeds the reference value provided by the reference potentiometer 1272. In addition, a thermistor or thermocouple on the heat sink of the illustrated circuit can be used to be compared with an overtemperature reference signal on line 1284 in a comparator 1286. The comparator 1286 provides an output signal on a line 1288 to the OR gate 1278 when the temperature of the heat sink exceeds an overtemperature reference value. In this case, the signal on line 1280 is provided to the high voltage gate driver to disable the H-bridge. Although most of the overcurrent or overtemperature protection circuitry described above is not shown for the H-bridge of magnet 1 (1146), it should be appreciated that the same arrangement may be provided for this bridge as well, but is not shown for simplicity of illustration.

Referring now to Fig. 69, a system level diagram is shown to illustrate a control scheme for a pair of opposed guides, such as the side active roller guides 1140, 1142 of Fig. 66. The diagram includes both the acceleration feedback described, for example, as described in detail above for a pair of small actuators 1146 and 1148, and position feedback for a pair of large force actuators, such as the spindle actuators 1300 and 1302. It should be appreciated that the scheme of Fig. 69 is also applicable to independently controlled, opposed (located on opposite sides of the same rail web) forward-backward suspensions, i.e., those suspensions which are not mechanically coupled, as shown in Fig. 60. The elevator car mass 1304 is shown in Fig. 69 as being acted upon by a net force signal on line 1306 from a summing element 1308 which responds to a disturbance force on a line 1310 and several forces represented on lines 1312, 1314, 1316, 1318, 1320 and 1322, all of which are summed in the summing element 1308. The disturbance force on line 1310 may represent several disturbance forces, all of which are represented on a line 1310. These disturbance forces may include direct car forces or rail-induced forces. The distinction between the two types of forces is that direct car forces result in a larger force that acts more slowly, such as wind, or is even static, such as load imbalance, while rail-induced forces are low-force disturbances at higher frequencies. The forces represented on lines 1312-1322 represent forces that counteract the disturbance forces represented on line 1310. In any event, a net force on line 1306 causes elevator mass 1304 to accelerate as indicated by an acceleration on line 1324. The elevator system integrates the acceleration as indicated by an integrator 1396 reflecting the car motion at a certain speed as indicated by line 1328, with the integrated value in turn being integrated by the elevator system as indicated by integrator 1330 to obtain a position change for the elevator car mass as indicated by line 1332.

Both electromagnets 1146 and 1148, as well as the drivers represented by the signal processor 1158 in Fig. 66, are shown together in Fig. 69 as a block 1334 which is responsive to a signal on a line 1336 from a summing element 1338 which in turn is responsive to the force command signal on line 1156 from the digital signal processor 1154 of Fig. 66, shown in Fig. 69 as a "filter &compensation" block, similarly designated 1154. This block performs the compensation and filtering explained in detail in connection with Figs. 4 and 5. A position control acceleration signal on line 1340 can be formed from the gap error signal on line 1398. Suffice it to note that this acceleration signal can be used to assist the fast control with the slow control. Such assistance is also inherently provided by direct measurement by the accelerometer. The accelerometer 1150 of Fig. 66 is shown in Fig. 69 as responsive to the acceleration of the elevator car, represented by line 1354, but also influenced by a vertical component of the acceleration, indicated by line 1350, which is summed with the actual acceleration by a summer 1352. Thus, the lateral acceleration of Fig. 66 on the line labeled SS can be corrupted by a small vertical component so that the signal on line 1152 is not a completely pure lateral acceleration signal. Similarly, the accelerometer is subject to drift as indicated on a signal line 1354, which can be represented as summing the accelerometer output signal 1150 with a model of a scattered acceleration signal in a summer 1356. Finally, a measured acceleration signal is provided to the processor 1154 on a line 1358. This concludes the description of the acceleration loop.

It can be seen that the two electromagnets 1146 and 1148 of Fig. 66 do not present a problem of "opposing" or "conflicting" because the control between the two is directed or guided. In the case of two opposing large-sized actuators, e.g. the two ball screw actuators 1300 and 1302, we have a similar problem with their independent operation because it may end in "conflicting". In the following, a concept of controlling the two high-force actuators 1300 and 1302 of Fig. 66 will be presented, according to which the operation is directed or guided to one or the other actuator.

The method of developing a centering command signal and directing this signal to control the two opposing actuators as shown in Fig. 69 will now be explained with reference to Fig. 70.

Reference points are marked with zeros. A pair of elevator shaft walls 1360 and 1362 contain a corresponding pair of rails 1364 and 1366 attached thereto. On the surface of each rail rolls a primary suspension, such as a roller 1368, 1370, which rolls on a surface of the corresponding rail at a distance XRAIL 2 or XRAIL 1, respectively. A spring constant K2, indicated by a block 1371a in Fig. 69, acts between the rollers 1368 and the actuator 1300, while a spring constant K1, shown as block 1371b in Fig. 69, acts between the roller 1370 and the actuator 1302. The position of the actuator 1300 with respect to the car 1304 is given by a distance X2, while the distance between the car 1304 and the centered position 1371 is given by a distance POS with a positive sign for right and a negative sign for left with respect to the center. The distance between the elevator car 1304 and the surface of the rail 1364 is given by a distance GAP2, and thus the distance between the actuator 1300 and the surface of the machine is GAP2 - X2. GAP20 represents the distance between the shaft wall 1360 and the car 1304 when the car is centered. Similar quantities are also shown on the other side of the car.

Returning now to Fig. 69, a position sensor similar to the sensor 1066, 1070 in Fig. 67 is shown as block 1376 for measuring the distance GAP1 in Fig. 70. Similarly, a position sensor 1378 measures the size of GAP2 in Fig. 70. It should be seen that, although a pair of sensors 1376, 1378 are shown in Figs. 66 and 69, such a function of measuring the gaps (GAP1 and GAP2) can be performed with a single sensor, albeit without the self-centering capability of the signal obtained by determining the difference between the two GAP signals. It will be seen by examining Fig. 69 that the measured quantities are related to the quantities shown in Fig. 70 by the following equations:

GAP1 = -POS - XRAIL1 + GAP10, and

GAP2 = POS - XRAIL2 + GAP20.

It will be seen that Fig. 69 is similar to Fig. 18 in many respects, except that there are two position sensors 1376 and 1378 responsive to the position (POS) of the car as indicated on line 1332, and the additional inner loops of position sensors for retracting the large actuators to the rest or zero position whenever they are not actively acting as actuators. In Fig. 70, the two gap position lines (GAP10 and GAP20) represent the distances between the car and the shaft ends when the car is centered. These values are additionally represented as "signals" which are input to "summers" 1384 and 1386 to produce the physical gaps indicated on lines 1388 and 1390 as GAP1 and GAP2. They serve to better understand the system.

Output signals from position sensors 1376 and 1378 are provided via associated signal lines 1392 and 1394 to a summer 1396 which takes the difference between the magnitudes of the two signals and provides a difference signal (centering control signal) via line 1398 to a lag filter 1400 which provides a filtered centering control signal via line 1402 to a connecting node 1404 which provides the filtered difference signal to a respective pair of precision rectifiers 1406, 1408 which together with the connecting nodes 1404 form a steering control 1409 for directing the filtered centering signal on line 1402 to one or the other, but not to both simultaneously. A pair of geared motor controllers 1410, 1412 are shown, one of which is responsive to the directed centering command signal by moving an actuator position (X1 or X2) indicated on a line 1412 or 1414, respectively, with integration by the system according to integrating blocks 1416 or 1418, respectively, at a relatively low speed indicated on a line 1420 or 1422 to cause a spring constant 1371d or 1371c to provide the force indicated by a line 1316 or 1314, respectively. It should be seen that in this control diagram, the spring constants 1371b and 1371d belong to the same spring actuated by actuator 1410. Similarly, spring constants 1371a and 1371c belong to the same spring, in this case actuated by actuator 1412. A pair of position feedback blocks 1424, 1426 are responsive to the actuator positions indicated by lines 1420, 1422, which include position sensors to provide feedback position signals over lines 1428, 1430 indicative of the position of the actuator relative to the car. These position signals may be subject to signal conditioning which may include providing a low gain feedback path. A pair of summers 1432, 1434 are responsive to the feedback signals on lines 1428 and 1430 and the centering command signal on line 1402 directed by the master controller for forming difference signals on lines 1436, 1438 representative of a difference between those signals. It should be understood that one of the pair of output signals on lines 1440, 1442 from precision rectifiers 1406 and 1408 represents the directed centering command signal on line 1402, while the other signal has a value of zero. By the value of zero is meant a command whose magnitude is equal to the amount required to return the actuator to its zero position, which is the position required to maintain at least the desired preload on the primary suspension.

Referring now to Fig. 71, there is shown the response of a position transducer such as that shown in Fig. 62. This is an experimentally determined response. Although the response is shown for a specific transducer, it will be appreciated that any other suitable type of position sensor may be used, including linear position sensors. The summation of the two signals on lines 1392 and 1394 is shown in Fig. 72 over the entire range of displacement of the elevator car (scaled to the specific sensor arrangement we have shown). The positioning of the links of the active guides according to the embodiment shown is shown so that no more than ten millimeters of displacement can be expected. It can therefore be seen that the two position sensors for the corresponding two roller guides can be combined into a seamless response as shown in Fig. 72 to be presented to the lag filter 1400 in Fig. 68.

Although the present invention has been shown and described with respect to at least one detailed illustrative embodiment, it should be apparent to those skilled in the art that numerous modifications in form, detail, methodology and/or procedure are possible without departing from the scope of the invention as defined by the appended claims.

Claims (28)

1. Elevator car with a device for stabilizing the elevator car (10; 46; 250; 1144; 1304) in an elevator shaft, wherein the device comprises: an accelerometer (16; 50-54; 204-208; 252-256; 1150) responsive to a horizontal acceleration of the elevator car to provide a measurement signal with a magnitude corresponding to the horizontal acceleration; and a control device (20; 380; 532; 1154) which responds to the measured acceleration signal in order to control an actuator device (24; 210-216; 240-244; 1146, 1148, 1300, 1302) for setting the car in horizontal motion against the acceleration, characterized in that the device further comprises: a position sensing device (440-454; 634-648; 1376, 1378) responsive to the degree of centering of the car to provide a sensing signal having a magnitude corresponding to a horizontal position of the car; and that, when the acceleration signal is given to the actuator device, the control device also responds to the signal coming from the position sensing device in order to compensate for components of the acceleration signal which are not representative for the actual horizontal acceleration acting on the car and to compensate for forces acting on the car to decenter it. 2. Car according to claim 1, wherein the control means is responsive to the sensed centering signal for integrating or lag-filtering the sensed centering signal to thereby provide an integrated or lag-filtered signal having a magnitude corresponding thereto; and wherein the control means is responsive to the acceleration signal and to the integrated or lag filter signal for controlling the actuator means such that the actuator means applies counterforces to the car in a direction opposite to the direction of the measured horizontal acceleration, effectively adding mass to the car in proportion to the magnitude of the read horizontal acceleration signal and centering the car in proportion to the magnitude of the integrated or lag filter signal. 3. Car according to claim 1, wherein the control means includes a comparator means (53) responsive to the centering signal (348) and to a comparison signal (352) of selected magnitude for comparing the magnitude of the sensed signal with the selected magnitude of the comparison signal and providing a difference signal (354) having a magnitude equal to the difference between the signals; and wherein the actuator device responds to the difference signal to set the car in motion. 4. Car according to claim 1, 2 or 3, in which the actuator device comprises a pair of opposing actuators (24a, 24b; 354, 356; 1140, 1142). 5. A car according to claim 4, in which only one actuator of the actuator pair is active at any given time, except in a central traversing zone. 6. A car according to claim 4 or 5, wherein the sensed centering signal is directed between one or the other of the opposing actuators. 7. Car according to claim 6, in which the controlled signal is provided with a default value in order to prevent simultaneous switching between the opposing actuators. 8. A car according to any preceding claim, wherein the sensing device comprises: means (16a-16c) for measuring an acceleration along a lower horizontal car axis to provide a car floor acceleration signal having a corresponding magnitude; means (16d-16f) for measuring an acceleration along an upper horizontal car axis parallel to the lower horizontal car axis to provide a car ceiling acceleration signal having a corresponding magnitude, and wherein the actuator means comprises: means (24a, 24b) responsive to the car floor acceleration signal for applying a counterforce to the car floor in a direction opposite to that of the measured car floor acceleration; and means (24c, 24d) responsive to the car ceiling acceleration signal for applying a counterforce to the car ceiling in a direction opposite to that of the measured car ceiling acceleration. 9. A car according to any preceding claim, in which the apparatus comprises: a plurality of accelerometers (16a-16c; 16d-16f) arranged to provide for independently measuring the horizontal position of the ceiling and floor of the car in an elevator shaft to enable determination of an amount of rotation of the car about a horizontal axis and to provide one or more corresponding read car rotation signals, the magnitudes of which correspond to the car rotation; and wherein the control means is responsive to the one or more read car rotation signals for providing a plurality of rotation control signals, and wherein the actuator means comprises two-level actuator means (24a, 24b; 24c, 24d) respectively responsive to the plurality of rotation control signals for independently moving the car at each level in a direction opposite to the indicated rotation. 10. A car according to any preceding claim, wherein the actuator device (1300, 1146; 1302, 1148) is part of an elevator-operable secondary horizontal suspension which is inserted between a primary suspension (1368, 1370) and the elevator car (1144) to guide the elevator car in the elevator shaft along a elevator shaft rail (1364, 1366), and wherein the control device comprises: a first control loop responsive to the sensed centering signal for controlling the elevator car with respect to the rail so that the elevator car remains centered in the shaft; and a second control loop responsive to a read signal having an amount indicative of a parameter of an actuator portion of the secondary horizontal suspension to maintain at least a selected force on the primary suspension. 11. A car according to any preceding claim, wherein the actuator means (1300, 1146; 1302, 1148) is an actuable part of an elevator secondary horizontal suspension for guiding the elevator car (1144) in the elevator shaft along an elevator shaft rail (1344,1366), and wherein the control means comprises: an outer control loop responsive to the sensed signal with an amount representative of the centering of the elevator car in the shaft for controlling the actuable part and hence the elevator car with respect to the rail so that the elevator car remains centered in the shaft; and an inner control loop responsive to the measured acceleration signal to control the actuatable part of the secondary suspension so as to reduce the horizontal acceleration of the car. 12. The car of claim 10 or 11, wherein the sensed centering signal has a magnitude indicative of a position of the actuator portion of the secondary horizontal suspension with respect to a selected reference. 13. A car according to any preceding claim, wherein the sensed centering signal is a position signal, the magnitude of which is representative of a horizontal position of the car within the shaft. 14. Car according to claim 1, wherein the actuator device is part of a pair of roller guides (1140, 1142) which are mounted opposite one another on the elevator car in order to guide the elevator car along vertical guide rails (1364, 1366) in the elevator shaft; and wherein the means for sensing the degree of centering of the car senses the centering of the car relative to one or more actuatable rollers (1368, 1370) of the roller guides to provide a sensed centering signal having a magnitude representative thereof, and wherein the control means includes: means for comparing the magnitude of the sensed centering signal with the magnitude of a reference signal to form a difference signal whose magnitude is representative of a difference between the magnitudes; and a steering device (1409) responsive to the differential signal for steering the car relative to the rollers of the roller guides by controlling the differential signal between controlling one or the other of the oppositely mounted roller guides. 15. The elevator car of claim 1, wherein the actuator means includes an electromagnetic force actuator (226-232, 280-286) responsive to a coil current signal for generating inductive flux, the control means comprising: summing means (470) responsive to a force command signal and a force feedback signal for generating a force difference signal having a magnitude proportional to a difference in magnitude between the force command signal and the force feedback signal; a current control device (486-494) responsive to the difference signal for providing the coil current signal; a current sensor (412-426) responsive to the coil current signal for providing a sensed current signal; a flux density sensing device (444-454) responsive to the magnetic induction and providing a sensed flux density signal having a corresponding magnitude; means (504) responsive to the sensed current signal and the sensed flux density signal for providing an air gap signal (396); and means (474) responsive to the flux density signal for multiplying its magnitude by itself and by a scale factor signal to form the force feedback signal; and wherein the Control device (380) has a force setpoint generator responsive to the air gap signal and the acceleration signal in order to provide the force setpoint signal in proportion to the magnitude of the acceleration signal. 16. Car according to claim 15, wherein the current control device comprises: a compensating device (486) responsive to the force difference signal for providing a proportionally compensated signal; a firing angle compensator (490) responsive to the proportionally compensated force difference signal to provide a firing signal; and a two-quadrant full-wave power controller (494) responsive to the ignition signal to form the coil current signal. 17. Car according to claim 16, wherein the compensation device (486) serves to provide a proportionally and integrally compensated signal. 18. Car according to claim 15, 16 or 17, in which the force setpoint generator is a digital signal processor (83), and the summing device (470), the device for multiplying (474), the device for providing the air gap signal (504) and the current control device (486-494) operate analogously. 19. Elevator car according to claim 1, in which the actuator device is an electromagnet (516-530) for guiding the elevator car vertically along a rail (267, 268) mounted vertically in the elevator shaft, and wherein the device for sensing the degree of centering of the elevator car comprises: means (634, 648) for sensing the flux density between the rail (267, 268) and the electromagnet (516-530) and for providing a sensed signal having a corresponding magnitude; means (602-616) for sensing the current of the electromagnet to provide a sensed signal having a corresponding magnitude ; means (500-514) responsive to the sensed current and flux density signals for dividing the magnitude of the sensed current signal by the magnitude of the sensed flux density signal and for providing a position signal whose magnitude is indicative of a horizontal position of the car within the shaft. 20. Car according to claim 1, in which the elevator car (46) comprises a cab (lob) suspended in an elevator car frame (60) within the elevator shaft, the accelerometer and the centering sensing device being responsive to respective parameters associated with the suspended cab to provide respective measurement signals; and the Actuator means (45) is mounted between the cabin and the frame, the actuator means responsive to the control means to move the suspended cabin horizontally with respect to the frame. 21. The elevator car of claim 1, wherein the elevator car comprises a cab suspended in an elevator car frame in a shaft, and the control means is responsive to the sensed acceleration signal for providing a control signal to the actuator means having a magnitude selected in accordance with the magnitude and frequency of the sensed acceleration signal. 22. The elevator car of claim 1, wherein the control means is responsive to the sensed acceleration signal for providing a control signal to the actuator means having a magnitude selected in accordance with the magnitude and frequency of the sensed acceleration signal. 23. Car according to claim 21 or 22, wherein the magnitude of the control signal is selected in accordance with a compensation device which has a negative gain or low gain below a first selected frequency (ω1) and also above a second selected frequency (ω2), and which has a positive gain or a gain which is greater than the low gain between ω1 and ω2. 24. The elevator car of claim 23, wherein the first selected frequency (ω1) is less than 1 Hertz. 25. Car according to claim 23 or 24, wherein the second selected frequency (ω2) is less than 10 Hertz. 26. A car according to any preceding claim, wherein the accelerometer (16) comprises at least three accelerometers (16a-16c) for detecting horizontal accelerations of the car and for providing three corresponding measurement signals to the control device for calculating corresponding forces required to counteract the measured accelerations. 27. The elevator car of claim 26, wherein two (16b, 16c) of the three accelerometers are oriented to sense horizontal accelerations along parallel axes perpendicular to a third horizontal axis. 28. Car according to claim 1, in which the accelerometer serves to provide a measurement signal which is representative of the horizontal acceleration and which contains component signals with different magnitudes at different frequencies, the control device comprising: a compensation device responsive to the read acceleration signal for supplying to the actuator device a control signal the magnitude of which comprises a sum of the component signals each increased or decreased according to a gain and phase shift function dependent on the component frequency, said function providing maximum gain in a middle region of a selected frequency range and delivers maximum phase shift.