RELEASABLE COUPLING FOR JAMMING HANDLING RELATED APPLICATIONS

This application claims priority and benefit from Swedish patent application applications Nos. 0700576 0, filed March 18, 2007, 0700690-1, filed March 19, 2007, 07006576-0, filed March 25, 5 2007, 0700916-0, filed April 12, 2007, 0701949-0, filed August 29, 2007, 0702014-2, filed September 10, 2007, 0702088-6, filed September 16, 2007, 0702451-6, filed November 2, 2007, 0702690-9, filed December 4, 2007, 0702799-8, filed December 7, 2007, 0800294-1, filed February 11, and 0800362-6, filed February 18, 2008, the entire teachings of which are incorporated herein by reference. 0 FIELD OF THE INVENTION

The present invention relates to coupling arrangements, in particular cog based coupling arrangements, between one element for which jamming is intolerable and another element that when required can rotate or move the first element but that should be disconnected if a jamming between the first and second elements or within the second element occurs, where in particular the coupling 5 arrangements can also be used to disconnect the two elements from each other when shock loads are expected to occur. The present invention also relates to methods of detecting jamming conditions in coupling arrangements and of using the result of such detecting to disconnect the coupling between two mechanical elements.

BACKGROUND The permitted failure rate for safety critical components in aircraft are extremely low, for example one fault during 1000 million flight hours. This puts very stringent requirements on the system design. This can be arranged so that failing components can be disconnected. This should be done as close to the essential system as possible.

A device for preventing overload is disclosed in the published Japanese patent application 06- 158667. However, this device cannot be used for protecting mechanical systems for faults in driving mechanical parts. SUMMARY

It is an object of the invention to provide an actuator or power device in which jamming in mechanical driving systems and/or transmission parts can be detected and the result of the detecting used to protect an output member from being mechanically blocked..

It is another object of the invention to provide methods to locate a jam in a mechanical transmission.

It is yet another object of the invention to provide methods to locate a jam in a mechanical transmission that can operate with a minimum of extra components.

Thus, an actuator or power device may generally include a movable output member having cogs. The movements of the movable output member must not be blocked by jamming in any component in the actuator or power device. A power transfer part is connected to the movable output member and has cogs or threads for engagement with the cogs of the movable output member. A driving system interfaces to the power transfer part for driving it. In the actuator or power device e.g. the driving system is apt to becoming jammed. A releasing device is arranged to cause the movable output member to be separated from said at least one power transfer part so that the cogs or threads, respectively, of the power transfer part do not any longer engage the cogs of the movable output member. A control system is connected to the releasing device for controlling the operation thereof.

The control system is arranged to detect jamming in the driving system and to control the releasing device in accordance with the result of the detecting, hi particular, the control system may include a control unit and a position sensor for a motor included in the driving system and then the control unit may perform, when a malfunction of the actuator indicates a jam, first issuing commands to the driving system to produce suitable torques or forces, then evaluating signals from the position sensor obtained as response to these produced forces or torques, and finally activating the releasing device in accordance with the result of the evaluation.

Generally then, in order to detect jamming in an actuator or power device that as above includes a movable output member, at least one power transfer part connected to the movable output member, at least one driving system interfacing said at least one power transfer part, and a releasing device arranged to cause the movable output member to be separated from said at least one power transfer part, the at least one driving system can first be controlled to produce suitable torques or forces, and then signals representing the movements of the at least one driving system obtained as response to these forces or torques are evaluated to decide whether a jamming condition has occurred or not.

Thus, as discussed above, means for reducing the damage caused by jamming in mechanical transmissions can be provided, such a mechanical transmission having e.g. a cog transmission that in one embodiment can be designed to permit a controlled change of the device is such a way that the cog transmission is opened, either by an actuator that forces the cog bearing parts to separate, or by removing forces that normally stops the forces between the teeth in the cog transmission to open itself if under load.

Additional objects and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objects and advantages of the invention may be realized and obtained by

means of the methods, processes, instrumentalities and combinations particularly pointed out in the appended claims.

BRIEF DESCRIPTION OF THE DRAWINGS

While the novel features of the invention are set forth with particularly in the appended claims, a complete understanding of the invention, both as to organization and content, and of the above and other features thereof may be gained from and the invention will be better appreciated from a consideration of the following detailed description of non-limiting embodiments presented hereinbelow with reference to the accompanying drawings, in which:

- Fig. 1 is a view of a redundant rotary actuator with inherent risk for jamming of the output member,

- Fig. 2a is a view of a cog wheel transmission that is jammed by a foreign part that has been placed between two teeth in a cog transmission,

- Fig. 2b is a view of a cog wheel transmission that is released thus preventing jamming caused by a foreign part, - Fig. 3 is a view of a releasable coupling intended to power an aircraft landing tyre,

- Fig. 4 is an overview over a landing gear tyre transmission,

- Fig. 5 is a view of an actuator having two releasable worm gears,

- Fig. 6 is a view of a force balanced actuator including a releasable rack,

- Fig. 7 is a detail view of the force balanced actuator of Fig. 6 seen in another direction, - Fig. 8 is a perspective view of an encased version of the actuator of Fig. 6,

- Fig. 9 is a perspective view of an electromechanical release device,

- Fig. 10a is principle view of an actuator or power device,

- Fig. 10b is similar to Fig. 10a for an actuator or power device having also a position sensor in an output member, and - Figs. 10c and 1Od are similar to Figs. 10a and 10b, respectively, for an actuator or power device having redundant driving and intermediate stages. DETAILED DESCRIPTION

Fig. 10a is a schematic of a general actuator or power device 1. It includes at least one driving system 3, a mechanical output device 5, also called a movable output member, and an intermediate mechanical system 7, also called a power transfer part, connecting the driving system to the output mechanical device. Mechanical interfaces 9, 11 connect the driving system to the intermediate mechanical system and the intermediate mechanical system to the mechanical output device. Such mechanical devices can generally have a mechanical play, such a between cooperating cogs. The driving system includes a motor 4.

In the actuator or power device 1 the movements of the mechanical output device 5 must not be blocked by jamming. Hence, the interface 11 between the intermediate mechanical system 7 and the mechanical output device includes or can be controlled by a releasing device 13, also called release device, the releasing device e.g. including a device for separating cogs from each other or a 5 mechanical clutch. The releasing device can make the mechanical output device 5 be separated from the intermediate mechanical system, thereby allowing the mechanical output device to move. Also, if suitable, the degree of separation may be controlled. The releasing device can be controlled by or include a control system 15, also called a control device or controller.

Jams can occur e.g. internally in the driving system 3 and in the two mechanical interfaces 9,

10 11. To detect possible jams a position sensor (PS) 17 is arranged in the driving system such as in the motor 4 and outputs position signals to the control system 15. The control system evaluates the received position signals and controls the releasing device 13 accordingly, such as making it totally disengage the intermediate mechanical system 7 from the output mechanical device 5. In response to the result of the evaluation the control system may be connected to also control the driving system

15 and to issue, to make a more safe decision that a jamming condition has occurred, commands to the driving system 3 and in particular the motor 4 therein to produce suitable torques or forces from the driving system over the mechanical interface 9 and then to evaluate the resulting position signals from the position sensor 17 obtained as a response to these torques or forces.

In order to make the decision that a jam has occurred even more safe, a position sensor (PS)

20 19 may be arranged for the output mechanical device, as illustrated in Fig. 10b. This position sensor outputs positions signals to the control system 15. The control system then uses also these received position signal to evaluate whether a jamming condition has occurred.

In some cases, in order to provide redundancy, such as in those cases where a very high degree of reliable operation for the actuator or power system is required, a plurality of driving systems and

25 intermediate mechanical devices can be provided. Hence, as seen in Figs. 10c and 1Od two driving systems 3, 3' are arranged that can make the single output mechanical device 5 move through individual intermediate mechanical systems 7, T and mechanical interfaces 9, 9' and 11, 11', respectively.

In Fig. 1 a specific embodiment of such a redundant actuator 100, for example a device for

30 turning a front wheel of a landing gear of an aircraft, is shown. The output shaft includes a pinion 101 that is driven by two worm gears 102 and 112. The first worm gear 102 is supported by a radial and trust bearing 103 and a radial bearing 106. A reducer 104 is driven by a conventional Y- connected electric motor 105 for driving the first worm gear. In the same way the second worm gear 112 is supported by a radial and trust bearing 113 and a radial bearing 116 and is driven by a reducer

114 that in turn is driven by a conventional Y-connected electric motor 115. By having two systems

102 - 106 and 112 - 116, some of the possible single faults can be handled without loss of the basic actuator function. The electric motor 105 for the first worm gear can become useless e.g. if one of its windings becomes short-circuited to ground, if one of the conductors in the cables connecting it to a control computer, not shown, breaks or if its position transducer, not shown, becomes in- operational. Assuming that the worm gears are not self-locking and that each of the systems like 102 - 106 is powerful enough to drive the load, i.e. rotating the pinion 101, and also a malfunctioning system 112 - 116, the single faults listed above can be handled without affecting the basic function of the actuator 100. Other single faults can cause one of the driving systems 102 - 106 or 112 - 116 to jam and they will however also jam the actuator output pinion 101. For example, if the reducer 104 has jammed, its output shaft 102, i.e. the respective worm gear, will be also be jammed. Thereby the electric motor 105 will be useless as it cannot force a jammed reducer to turn. The redundancy given by the two electric motors 105, 115 is in such a case useless, as the jammed reducer 104 will lock the position of worm gear shaft 102, hence the output pinion 101, the other worm gear shaft 112, the reducer 114 and the electric motor 115 for the other worm gear.

By inserting clutches, not shown, between the reducers 104, 114 and the worm gear shafts 102, 112 jamming in for example the motor 105 or the reducer 104 for the first worm gear 102 can be handled by opening the clutch between the reducer 104 and the worm gear shaft 102. However, jamming downstream this clutch, for example of the bearings 103 or 106, will still cause a total system jam.

Furthermore, a cog transfer may cause a jam, as illustrated in the principle drawing of Fig. 2a. A cog wheel transmission 200 is jammed by a foreign part 201 that is located between two teeth in a rack 202 included in the cog transmission. The pinion 203 is safety critical. The foreign object 201 blocks the whole transmission for rotation in the indicated direction.

In Fig. 2b a solution to this problem is illustrated. By arranging a possibility for the pinion 203 and rack 201 to be separated from each other in the case of a jam and activating this possibility when a jam is detected, the jamming can be released by moving the rack 202 and the pinion 203 away from each other. This will not only solve the problem of a foreign object 201 located in the cogs of the pinion 203 or rack 202, but will also permit the pinion 203 to move freely if the rack 202 would be jammed for another reason.

Fig. 3 is a sectional view, the section taken in a plane parallel to the tyre shaft of a transmission intended for an aircraft tyre, not shown. The purpose of the transmission is to give the aircraft the ability to move on the ground without a need for a tractor. The transmission could also

be used to speed up the aircraft tyre to a rotational speed close to the ground speed before landing, thereby reducing the wear on the aircraft tyre.

The aircraft tyre is mounted to the shaft 301 on which a cog wheel 302 is assembled. It is during power transfer connected to a pinion 303 that is rotating in a housing 304 which can move towards and away from the shaft 301 along linear bearings, not shown. The pinion 303 can thereby either be in functional contact with the output cog wheel 302 or be mechanically disconnected therefrom, hi the housing 304 the pinion 303 is assembled together with a bevel gear transmission with an angle of 90° between its shafts which are integrated with the cooperating toothed wheels

305 and 306. In the embodiment shown, the movement of the assembly 303-304-305-306 towards or away from the cog wheel 302 is controlled by an electric motor 307 acting on a roller screw 308.

This short movement can obviously be arranged in many other ways, such as using a gear motor turning an eccentric, by a solenoid, by a hydraulic cylinder, etc.

The shaft 309 from the bevel gear wheel 306 is to be connected to a traction motor, not shown, by mechanical devices adapted to the properties of the landing gear, not shown, its geometry and its shock absorber system.

Fig. 4 shows an example of a drivetrain for connecting the shaft 309 of the bevel gear wheel

306 to a traction motor 406. The figure shows a section selected in such a way that one of the tyres and its brakes are hidden. The visible tyre 401 is mounted to the shaft 301. This is either connected or disconnected from the shaft 309, seen only in Fig. 3, that is engaged through splines, not shown, to a shaft 402. This shaft is connected through an angularily flexible coupling 403 to a length-wise flexible dual spine shaft 404 in turn connected to a reducer 405 and the traction motor 406.

The motors 307 and 406 and their control power stages and control computers, not shown, could be designed e.g. in accordance with U.S. patents 7,098,619 and/or 6,885,162 issued for the applicant of the present patent application. To permit the traction motor 406 to spin the tyres such as 401 before landing, a planetary gear such as the reducer 405 can be designed as a two speed gear box. Configured with a 1 :1 ratio, the traction motor can spin up the tyre to the expected landing speed, hi this way, the acceleration shock at landing can be reduced during normal conditions as the tyre is already running at a speed close to the correct speed, and the inertia of the traction motor 406 on the transmission is much less than when the planetary gear 405 is active, hi the landing phase, the planetary gear is in a 1 :1 ratio. If the traction motor be unable to make the tyre reach the expected speed, the cogwheel housing 304 will be retracted and the landing gear will operate in the same way as in currently used aircrafts, without any motor inertia connected.

Alternatively, the cogwheel housing 304 will always be removed to its passive position after

having set the tyre speed to the expected landing speed. Thereby, there will be no landing shock loads on the traction system 302-303-304-305-306-309-402-403-405-406, and any jams that could occur during the landing phase can not affect the tyres such as 401.

By this arrangement single faults can be tolerated. A jamming of the housing 304 in its active 5 position can be tolerated if the reducer 405 can be set in its 1 : 1 gear ratio and the traction motor 406 operates as this will set the tyre at a speed close to landing speed.

During ground operation, the planetary gear such as 405 is in a reduction ratio like 8:1 permitting a ground speed of 1/8 of the landing speed. This will increase the inertia of traction motor 406 as seen from the tyre by the square of the gear box ratio, for example by 82 or 64 times.

10 Fig. 5 is a schematic of an actuator similar to that illustrated in Fig. 1 including two releasable worm gears 501, 502 acting on a pinion or gear 503. hi the left end each of the worm gears is assembled in an axial bearing house 506, 507 that can turn around a shaft 504, 505. The other end can be moved from the engaged position shown for the worm gear 502 to a not engaged position shown for worm gear 501. The movement can be arranged by an EED ("Electrically initiated

15 Explosive Device") that removes a structural part, by a solenoid latch as shown for a releasable rack system in Fig 9 below, or, as shown, by a roller screw 508, 509 and a roller nut 510, 511 driven by motors, not shown. The arrangement is shown with the worm gear 501 disconnected from the main cogwheel 503. In this way jamming in the worm gear 501, the reducer 514 or the main motor 516 can be disconnected. Another function of the possibility given by the arrangements 508-510-512 and

20 509-511-513 is to adjust the play between the worm gears 501, 502 and the main cog wheel 503. The tolerances for these plays given by some manufacturers is so narrow that the change in play caused by thermal expansion from -50° to +50° C of the main cog wheel 503, the casing, not shown, and the worm gears is more than twice the totally permitted play variation. This can in principle be compensated for by a movement of the roller nuts 510, 511 as a function of temperature.

25 Fig. 6 is a schematic of a force balanced half actuator including a releasable rack, and Fig. 7 is a detail view of the same actuator in another direction. The line and arrows at "Fig. 7" in Fig. 6 illustrate the viewing direction for Fig. 7.

The half actuator has a roller screw 601 that is moving a roller nut 602. Two racks 603a, 603b are attached to the roller nut where only rack 603b is visible in Fig. 7. The racks are engaged with

30 pinions 604a, 604b mounted to or integrated with the same output shaft 701, this forming a movable output member. The forces between each rack and the pinion will create a radial force that will act to move the pinion to the left as seen in Fig. 6 and the rack 603 a, 603b to the right. These forces are balanced by rollers 605a, 605b supporting the racks.

Rollers and ball nuts are sensitive to radial forces and torques acting to bend the nut relative to

the screw. To reduce these forces, the embodiment of Figs. 6 and 7 has two racks 603 a, 603b with the roller nut 602 located centered between the two racks in such a way that the contact line between the pinion cogs and the rack cogs are in the same plane as the centre line of the roller screw 601. Thereby, the axial forces created by the roller screw and the racks will create no bending torque over the roller nut.

The roller nut 602 is axially rigidly connected to an axial bearing house 606. This bearing house is supported in the actuator frame, not shown, over two cylindrical bolts 607 permitting the roller screw 601 to move slightly to adjust to minor misalignments and flexing in the actuator frame and the structure, such as a landing gear, in which the actuator is inserted. A gear set of four gears such as 608 will permit the motor 609 to turn the roller screw 601.

The radial force absorbing rollers 605a, 605b are kept in place in the actuator frame by fixation studs 610 in which a bolt 611 is kept.

A control system is used that comprises a controller or control unit 614 that is common for the two half actuators, only one of which is shown, can control the motor 609 using information from a first position sensor such as an angular transducer 613 mounted to the motor. The controller can also control the release device seen as 702 in Fig. 7 and hidden by the bolt 611 in Fig. 6. A second position sensor, such as a transducer 612, can be connected to the controller 614 to measure the position, in this case the angle, of the movable output member 701 or the equivalently of the pinions 604a, 604b. Fig. 7 is another view of the actuator seen in Fig. 6. The main actuator output shaft 701 includes or is integrated with the two pinions 604a and 604b. The diameter of the output shaft is locally reduced to permit the roller screw 601 to pass close to the output shaft, thereby reducing the space required.

The supporting rollers, such as 605b, are shown in sections. The frame stud is shown as 610 and the releasable bolts as 611a and 611b. An explosive charge 702 can be used to eject the bolt 611a. Also the other bolt 61 Ib can be released. If the bolts are released, the radial forces from the pinions 604a, 604b will turn the roller screw 601 - rack set 603 a, 604b counterclockwise as seen in Fig. 6 around the shaft 607, thereby releasing them. Springs, not shown, can be added to ensure that this releasing movement occurs. Fig. 8 is a perspective view of the actuator illustrated in Figs. 6 and 7. The motor has been removed to show the cogwheel 608. In the embodiment shown in Fig. 8, the shafts of the rollers 605 are permanently fixed in the case 801.

The whole case 801 is assembled to the frame of the output pinion bearing structure, not shown, through a shaft, not shown, along the axis 804. The forces from the rack and pinion cogs

must be balanced by a torque shown schematically as 805. This torque may be arranged by joints between the upper part of the case 801 and the frame of the output pinion bearing structure. These joints should normally be static but can be disconnected in the case of jamming. This can be performed by EEDs similar to the arrangement of the bolts 611a, 611b shown in Fig. 7. It can also be arranged by bolts withdrawn by a slower device such as a screw actuated by an additional gear box motor, not shown, that is used only for jam handling. The forces from the racks 803a, 803b and the pinions 604a, 604b will create a torque around the axis 804 that acts in the same direction as any device designed to disconnect the racks from the pinions, thus reducing the power for example required to unwind an anti-jam screw connection. Some features shown in Fig. 8 are arrangements to handle buckling forces. A support bearing

802 is to support the roller screw 601 against buckling forces and the U-shaped cross section of the racks 603 a, 603b has the purpose to make the racks stiffer. (A thin section of the upper part of the rack 603 a has been removed in the shown view; the rack reinforcement shown as 803 has a symmetric counterpart on the upper side of the rack, but only part thereof is visible). Two half actuators as the one shown in Fig. 8 can be assembled to act on the same pinions

604a, 604b. As long as none of the two half actuators has jammed, they deliver half the load each, hi the case where one thereof has jammed, it can be disconnected by using a very fast EED design or a slower disconnection device e.g. as described above. Each of the two half actuators maybe designed to handle the full load for a very limited number of strokes. Alternatively they can be designed to handle a reduced set of conditions.

Fig. 9 is a perspective of a mechanical release device assembled on a half actuator like the one shown in Fig. 8. It consists of a linkage arm 901. The lower part of the arm is connected to the frame, such as the landing gear, in which the pinion 701 is rotating. This arm creates when required the torque illustrated as 805 in Fig. 8. The upper part of the arm is kept in its shown position by a lever 902 that can rotate around the shaft 903. It is kept in the position shown by a rod, not shown, that acts against the upper side of the two needle bearings 904. The rod is preferably part of a linear actuator or linear motor, not shown, that can be energized to remove the rod, thereby releasing the lever 904. The tension in the arm 901 will then cause the lever 902 to turn. This will move the arm 901 and tilt the half actuator around the shaft shown as 804 in Fig. 8, thereby disengaging the teeth of the racks 603a, 603b from the pinions and the output shaft 701. Springs can be added to ensure that the lever 902 and the half actuator move as described even it there are no forces between the teeth of the pinions and the racks.

Alternatively, the release mechanism can consist of a linear actuator, for example a motor having an internal roller screw and a brake similar to those shown in Fig. 5. The linear actuator

roller screw end could replace the lower end 906 of beam 901 that keeps the half actuator 907 engaged to the pinion of the output shaft 701. The linear actuator motor end could be assembled on the upper surface 905 of the half actuator 907 using some bearings to permit the motor to tilt a few degrees when the linear actuator is moving to the non-engaged position for the half actuator 907. If a jam occurs in a dual driver motor system, it is necessary for the actuator controller to locate the position of the jam so that the jammed driver can be released and the not jammed driver can drive the actuator output member. The actuator of Fig. 6 can be taken as an example. Two embodiments will be described, one embodiment including a position transducer for the output members 604a, 604b and another embodiment having no such position transducer. To find the position of the jam, the position sensors that are necessary for ordinary operation can be used. Each motor must have at least a medium resolution position sensor such as the position sensor 613 to achieve a good servo response and reliable commutation. For jam detection it is irrelevant whether the position sensor is incremental or absolute. There is also in some cases an encoder such as 612 on the output member such as 701, for example on the tyre carrying tube of a landing gear. The actuator shown in Fig. 6 is one half of a system containing two half actuators, hereinafter denoted A and B. The total system has two driving systems, each including a motor 609, gears 608 and a roller screw 601. There is also a power transfer part consisting of the racks 603 a, 603b that transfers the driving system force to the output members 604a, 604b and 701. Assume as a numerical example that there is a cog play between the power transfer parts 603 a, 604b and the output members 604a, 604b of some 70 μm, a roller screw pitch of 4 mm and a screw to motor ratio of 7.4 in the reduction gear 608. The 70 μm cog play between the rack 603a, 604b and the respective pinion 604a, 604b will then correspond to some 0.13 turns peak to peak of the motor 609. The reduction unit 608 has at least two cog interfaces between the motor rotor or output shaft and the roller screw 601. Assume as an example that the play in these cogs is selected to permit some 0.003 turns peak to peak of the motor, which is a play internal to the driving system 609, 608. Finally, assume that there is no output member position transducer 612.

- A jam in the "A" motor 609 would show up as a no movement of the motor rotor or motor output shaft at all even with torques applied in both directions.

- A jam in the "A" roller screw 601 would show up as a motor 609 rotor movement of some 0.003 turns peak to peak with small torques issued in both directions.

- A jam in the output member pinions 604a, 604b shaft 701 would show up as an A motor 609 rotor movement of some 0.13 turns peak to peak with torques applied in both directions.

- A jam in the roller screw 601 of the opposite half B actuator would show up as an A motor 609 rotor movement of some 0.26 turns peak to peak when torques are applied in both directions.

- A jam between a rack 603a, 603b and the respective pinion 604a, 604b caused by a foreign object or part of a broken cog could be searched by first commanding the A half actuator motor 609 to give full force in the jamming direction and command the B half actuator, not shown, to apply torques in both directions. If the jam is caused by a foreign object between the A rack 603a, 603b and the respective pinion 604a, 604b, the B motor can move freely over a play corresponding to some 0.13 turns peak to peak when using a small torque in both directions. Making this test again with the actuator halves reversed so that the B motor gives the full force would cause the force from the B motor 609 over the B rack 603a, 603b and the pinion 604a, 604b to reach the A rack 603a, 603b and the foreign object. The A motor 609 can therefore not move more than some 0.003 turns peak to peak when using a small torque in both directions.

Now, assume that also an output member position transducer 612 is arranged.

- A jam in the "A" motor 609 would show up as a no rotor movement at all even with torques applied from the "A" motor in both directions. The transducer 612 would not be affected.

- A jam in the "A" screw 601 would show up as an A motor 609 rotor movement of some 0.003 turns peak to peak with small torques issued in both directions, and the transducer 612 would not be affected.

- A jam in the output member pinion shaft would show up as an A motor 609 rotor movement of some 0.13 turns peak to peak with torques applied in both directions. The transducer 612 would not be affected. - A jam in the screw of the opposite half B actuator would show up as an A motor 609 rotor movement of some 0.26 turns peak to peak with torques applied in both directions. The transducer 612 would register an angular movement representing a 70 μm linear movement of the output member 604.

- A jam between a rack 603a, 603b and the respective pinion 604a, 604b caused by a foreign object or part of a broken cog could be searched by first commanding the A half actuator motor 609 to give full force in the jamming direction and command the B half actuator, not shown, to apply torques in both directions. If the jam is caused by a foreign object between the A rack 603a, 603b and the pinion 604a, 604b, the B motor can move freely over a play corresponding to some 0.13 turns peak to peak when using a small torque in both directions and the transducer 612 would not register any significant movement. Making this test again with the actuator halves reversed so that the B motor gives the full force would cause the force from the B motor over the B rack and the pinion 604a, 604b to reach the A rack 603a, 603b and the foreign object. The A motor 609 can therefore not move more than some 0.003 turns peak to peak when using a small torque in both directions as the rack 603 is locked by the large force given by B motor 609.

In more general terms, the identification of the position of a jam should permit to select which releasing device that should be activated, i.e. whether the releasing device for the A half actuator or the releasing device for the B half actuator should be activated. For this purpose, it is irrelevant if the jam has occurred in the power transfer part such as a rack 603a, 603b or in the driving system such as 601-602-608-609.

For systems having release devices that are not reversible, such as the EED device 702 or the system shown in Fig. 9, it is essential that this identification can be correctly made in the first attempt, hi such systems, a play between the output member such as a pinion 604a, 604b and the power transfer part such as 603a, 603b is very useful. Assuming that the play as seen from the driving system position sensor 613 between the driving system position sensor 613 and the power transfer part such as 603a, 603b is significantly smaller than the play as seen from the driving system position sensor 613 between the power transfer part 603 a, 603b and the output member 604a, 604b, the play between the power transfer part and the output member permits clear indications of the position of the jam. If the play inside the driving system and between the driving system and the power transfer part is negligible, even a small play between the output member and the power transfer part will be sufficient. If the motor of Fig 6 should have been connected directly to the roller screw 601 and if the roller nut was designed with a preload, there would be no play between the transducer 613 and the power transfer part 603a, 603b.

For systems including reversible release devices, such as shown in Fig. 3 and 5, a jam can be identified by activating each release device, one after another. In the case where there always is a play between the output member 503 and the power transfer parts 501, 502, a play measuring method as described above is probably faster. In the case where there is no useful play in normal operation, a useful play can be arranged by releasing both release devices 508-51 Ia small distance.

For systems having no play, high resolution position transducers can be used to measure the elasticity of different links in the system. A jam in the bearing set 113 of Fig. 1 can be taken as an example. Once such a jam has occurred, the controller similar to 614 will find that the output member 101 does not move in the normal way and that it does not move more than marginally even at maximum permitted torques from motors 105 and 115. A search procedure can then be initiated. For example, both motors can first be set to 100 % of maximum permitted torque and the positions of the transducers for motors 105, 115 and for output member 101 can be recorded. The torque of motor 105 can then be reduced to 20 % of full torque. This will reduce the tension on the reducer 104, the bearing set 103, the worm gear shaft 102, the output member 101, the worm gear shaft 112 and bearing set 103. Both the transducer for motor 105 and the transducer on output member 101 will therefore register a considerable displacement. (It is assumed that the worm gears and the

reducers are not self-locking.) The transducer of motor 115 might register a minor displacement as the elasticity of the jam is less deformed by the reduced torque from motor 105 giving more elastic angle to the torque from motor 115. Reversing the torques so that motor 115 gives 20 % and motor 105 100 % of permitted value might cause the transducer on motor 105 to register a minor 5 displacement relative to the initially stored values as the elasticity of the jam is less deformed by the reduced torque from motor 115 giving more elastic angle to the torque from motor 105. The transducer of motor 115 might register a minor displacement as the elasticity of the jam is less deformed by the reduced torque from motor 115 giving less elastic angle to the torque from motor 115. This torque sequence and the resulting transducer displacements indicate a jam in the system 10 112-113-114-115. If the operation had been made with the system of Fig 5, this system should have been released by activating the release device 509 and 511.

For highly reliable systems, it is common to periodically verify that all safety relevant system components are operational. For the system shown in Fig. 5, this could be arranged by moving the roller screws 509-510 a full stroke to a no-load condition, such as for aircraft's standing with then- is brakes engaged. For the system shown in Figs. 3 - 4 it is foreseen that the cogwheels 305 and 306 will be disconnected before a landing and that they will be connected after a landing. The electric motors will almost always have angular feedback devices such as 613 to permit a good speed or position response, and the output members will also in many cases need angular feedback devices such as 612. Assuming that such devices are in place, the position of the transducers like 612 - 613 0 before the cog separation can be used as an initial value. When the cogs are to be engaged, the rotation of the output member like 604 after the separation of the cogs can be calculated from the change in reading from its transducer like 612. The required movement of the motors like 609 to get their racks 603 a, 603b in the correct position for engagement to the pinions 604a, 604b can then be easily calculated. 5 As is obvious for those skilled in the art, the basic principle of the invention can be modified in many ways. The methods for detecting jam and for identifying where the jam is located can be made in many ways depending on available position feed-back devices, system stiffness, cog plays and eventual extra sensors. The release devices shown and described are based on separation of cog carrying parts. The same principles of detecting and identifying jams can be applied to systems 0 having clutches as release devices. However, the safety level from such systems is lower as jams in the downstream part of the coupling and following downstream components will cause a jammed output member also after that the clutch is opened.

While specific embodiments of the invention have been illustrated and described herein, it is realized that numerous other embodiments may be envisaged and that numerous additional

advantages, modifications and changes will readily occur to those skilled in the art without departing from the spirit and scope of the invention. Therefore, the invention in its broader aspects is not limited to the specific details, representative devices and illustrated examples shown and described herein. Accordingly, various modifications may be made without departing from the spirit or scope of the general inventive concept as defined by the appended claims and their equivalents. It is therefore to be understood that the appended claims are intended to cover all such modifications and changes as fall within a true spirit and scope of the invention. Numerous other embodiments may be envisaged without departing from the spirit and scope of the invention.