[0053] FIG. 1 shows the cross-section of an actuator, e.g. an actuating device for brakes, including the basic machine elements into which the mentioned magnetically effective components are integrated and with which they form the arrangement or device of the present invention. The components are a stator 1 , 2 , 3 and an actuating element 4 which is a rod, for example, which are displaced to and fro by the distance x in relation to each other during the actuating or resetting operation. In this example, the actuating element 4 is actuated by an outside force that acts on the actuating element, for example, by way of another rod 5 , and both rods are mechanically coupled by way of a ball-and-socket joint 6 . Rod 4 is mechanically coupled to a shank 7 , and rod 4 and shank 7 may also be manufactured from one joint piece. In the present example, the shank is configured as a tube. The manufacture from one piece and partial design as a tube is preferred in this arrangement. In this case, only the piece with a tubular design is referred to as shank. It is also possible according to the present invention to use a continuous rod 4 which accommodates the encoder(s). The cross-section of the rod 4 and the shank 7 and the opening cross-section of the stator 1 , 2 , 3 which comprises the rod are favorably of a circular design, what is not imperative though. Thus, it is also possible to use cross-sections that are designed like a square, triangle, or trapeze or dovetail, respectively. Advantageously, the stator is a compound made up of several housing parts, but it may also be of integral design. The sensor carrier 2 in or on the sensor S is stationarily connected to the housing part 1 , favorably, including associated integrated circuits for the purpose of signal conditioning SC which are embedded, clipped, screwed or form-locking clamped additionally. The sensor carrier preferably represents also the mounting support for a plug 8 or, alternatively, for a cable for the propagation of sensor signals S(x) representative of the displacement x to an electronic unit further processing the signals. Housing part 1 and sensor carrier 2 are favorably slipped onto each other in a non-rotatable manner. In the present example, sensor carrier 2 is press fitted on the housing part 1 by way of a resilient cap 3 . However, it is also possible that the sensor carrier 2 is screwed, cemented or press fitted to the housing part 1 by any other resilient holding mechanism. It is especially favorable that there is a means which prevents twisting of the sensor carrier and the housing part.

[0054] The arrangement of the present invention is protected against dirt, humidity (corrosion), ferrous particles, small parts, etc. To this end, a sealing means 10 which is a rubber sleeve in this case is provided in the area of the shank between the housing and the rod 5 . A seal 11 close to the rod also provides protection against contaminants.

[0055] FIG. 1 shows a variant that is especially favorable with respect to the shank/field-generating means (encoder). A tubular encoder (see FIG. 5 , types 28 a to 33 a and 28 b to 33 b ) is inserted into a tubular thin-walled shank 7 . An iron return path may be employed advantageously also in this case. The advantage is that no magnetic attractive forces act on particles at the tubular encoder inside. The encoder is mechanically protected in the annular construction and the shank can be guided by way of a pressure-loaded seal.

[0056] A linear distance sensor largely corresponding to FIG. 1 is illustrated in FIG. 2 . The arrangement has a simple redundant design in contrast to FIG. 1 . Two sensors S, S′ including sensor circuits SC, SC′ respectively associated with the sensor are provided in the sensor carrier 2 . Two encoders 9 and 9 ′ are embedded in the shank. Position-synchronous signals S 1 (x) and S 2 (x) are produced by interaction of encoder, sensor, and sensor circuit.

[0057] Further field-generating means according to the present invention are illustrated in FIGS. 4 and 5 . Variants 17 a to 39 a are encoders without iron return path 281 , 282 , and variants 17 b to 39 b are encoders with iron return path 281 , 282 . In the case of tubular encoders, the iron return path is a thin inside tube, without an air slot towards the magnetic material. Preferably, this tube may additionally be a mechanical support for especially thin-walled tubes. With encoders shaped as a round rod, the iron return path is a wire-shaped iron core.

[0058] According to the present invention, the encoders can be provided with the examples of magnetization patterns listed hereinbelow:

[0059] A magnetic encoder period (illustrated in partial pictures 18 a, 18 b, 33 a, 33 b, 34 a, 34 b ),

[0060] Several magnetic equidistant encoder periods (illustrated in partial pictures 17 a, 17 b, 25 a, 25 b, 26 a, 26 b, 27 a, 27 b, 28 a, 35 b ),

[0061] Several magnetic non-equidistant encoder periods (illustrated in partial pictures 19 a, 19 b, 32 a, 32 b, 39 a, 39 b ),

[0062] Multi-track magnetic rulers (illustrated in partial pictures 20 a, 20 b ),

[0063] Individual linear magnetic pitch periods (illustrated in partial pictures 21 a, 21 b ),

[0064] Several linear magnetic pitch periods (illustrated in partial pictures 22 a, 22 b ),

[0065] Individual non-linear magnetic pitch periods (illustrated in partial pictures 24 a, 24 b ),

[0066] Several non-linear magnetic pitch periods (illustrated in partial pictures 23 a, 23 b ),

[0067] Screw lines with individual linear magnetic pitch period (illustrated in partial pictures 29 a, 29 b, 36 a, 36 b ),

[0068] Screw line with several linear magnetic pitch periods (illustrated in partial pictures 30 a, 30 b, 37 a, 37 b ),

[0069] Screw line with several non-linear magnetic pitch periods (illustrated in partial pictures 31 a, 31 b, 38 a, 38 b ).

[0070] Examples for encoder shanks are shown in FIG. 6 . The material of shank 7 is desired to have a minimum possible magnetic conductivity as is e.g. the case in magnetically non-conductive steel or hard aluminum. The encoders 9 , 9 ′, 9 ″ can be embedded in different regions 74 , 75 of the shank body.

[0071] Partial picture a of FIG. 6 shows a tubular shank 7 having an outside periphery 72 wherein a ruler-shaped encoder 9 is embedded. The illustrated ruler is a ruler of the type 25 a according to FIG. 4 . The said has a trapezoidal profile and is thereby kept captive in the shank, provided the recess 74 has a matingly trapezoidal shape. Rulers with a rectangular profile ( FIG. 4, e .g. type 17 a ) may also be embedded in the outside periphery. When there is integration into the outside periphery, it is especially favorable to provide the encoders with an iron return path.

[0072] Partial picture b of FIG. 6 shows another tubular shank having several encoders of different length embedded in the inside periphery. This variant is advantageous when the shank is guided by means of a pressure-loaded seal.

[0073] Partial picture c shows another tubular shank having in its wall bores 92 , 93 ′ into which rod-shaped encoders may be slipped. Advantageously, the encoder is completely protected mechanically in this case, and the shank can be guided by means of a pressure-loaded seal.

[0074] Partial picture d shows another shank which, compared to the embodiment in partial picture e, can be designed as a tube with very thin walls, while the wall thickness is locally reinforced to incorporate the bores.

[0075] Partial picture e in FIG. 6 shows another shank with the special feature that the bores of the wall are designed eccentrically, and the diameter of the bores is larger than the wall thickness of the shank. Favorably, this permits captively mounting encoders (e.g. types 34 a to 39 a and 34 b to 39 b ) with a relatively large diameter and an accordingly higher magnetic field strength. This is e.g. the case if the radius and the residual wall thickness remaining after drilling are smaller than the wall thickness of the shank.

[0076] Sensor assemblies with a different orientation of the sensors with respect to the shank are illustrated in FIG. 3 . Encoder 9 generates a periodic course of lines of flux 33 in the longitudinal direction of the encoder. The course of the lines of flux may also be referred to as encoder track. The illustrated course of the lines of flux is characteristic of magnetic material with zones 31 , 32 of a homogeneous magnetic material, e.g. a ferrite, which are alternatingly magnetized in north/south-pole alignment. The encoder track extends two-dimensionally in the image plane. In the first variant illustrated in the left partial picture of the Figure, AMR sensors are connected in a bridge circuit 13 , with the sensors all being placed in one joint plane 131 . In the left partial picture, the plane is aligned vertically to the encoder plane, that is in parallel to the surface normal of the displaceable element and in parallel to the moving direction or the longitudinal axis of the displaceable element. In addition, the lamellar structure of the sensors in this orientation extends vertically or in parallel to the surface normal. When this structure is moved in the direction X along the encoder track, the vector of the magnetic field strength will rotate through the bridge plane, and the bridge produces an outlet signal with two signal periods 14 per encoder period λ. This effect will be referred to as 2α-effect hereinbelow.

[0077] In the right-hand partial picture, the joint plane of the bridge circuit is aligned in parallel to the encoder plane, that is, vertically to the surface normal of the displaceable element. When the displaceable element is moved in the direction X along the encoder track, only a partial vector of the magnetic field strength acts on the bridge layer so that an output signal is produced with only one signal period 16 per encoder period λ. This effect will be referred to as 1α-effect hereinbelow. Both effects can be converted into specific sensor elements which either are configured to detect a moving direction or to detect no moving direction. When the 2α-effect is used, the sensor elements especially include two bridges which are rotated by 45° in relation to each other and supply SIN/COS signals that permit deriving the moving direction according to known methods. In the 1α-effect, the phase shift of the bridge branches with respect to the encoder period λ is used to detect the moving direction.

[0078] A particularly simple arrangement is achieved when employing only one of the sensor bridges of FIG. 3 . In this case, however, the possibility of detection of the moving direction must be dispensed with.

[0079] When sensors are employed on the basis of the 2α-effect, a high distance resolution may be achieved especially if per se known interpolation networks are inserted after the sensors and sensitively quantize the encoder period λ and, thus, the distance x.

[0080] Examples for sensor/encoder combinations with a high position resolution are illustrated in FIG. 7 . The sensors shown operate according to the AMR principle and utilize the principle of the 2α-effect. In partial picture a, the combination of a sensor 40 with an electronic network 41 connected downstream of the sensor is shown. The network serves for interpolation and signal conditioning. The network and the sensor are preferably integrated in one joint sensor assembly A. The network has an internal buildup so that an interpolation factor with a value of at least (λ/8)/Δx is used, and Δx corresponds to the distance increment with the minimum resolution. In partial picture a, the sensor plane of the sensor 40 is aligned in the direction of the moving direction X and in the direction of the surface normal of the encoder so that, with the encoder used, the field vector of the generated magnetic field rotates periodically through the plane of the AMR layer.

[0081] According to partial picture b, the sensor assembly A furnishes as a sign for the moving direction a binary signal and in the distance Δx an incremental signal pulse train V*(n*Δx).

[0082] In partial pictures c and d of FIG. 7 , examples for sensor/encoder combinations are shown which are different from the embodiment illustrated in partial picture a with respect to the orientation of the sensor plane. The sensor plane is oriented vertically to the moving direction X herein, which is in contrast to partial picture a. The encoders of the type 29 a or 31 a illustrated herein generate a magnetic field which, similar to partial picture a, rotates through the plane of the AMR-layer.

[0083] Sensors 40 are connected with a component for the signal conditioning SC so that a position-variable signal S(x) is produced which corresponds in its effect to the signal of sensor A.

[0084] Examples for sensor/encoder combinations are shown in FIG. 8 which operate according to the 1α-effect by using the AMR principle. However, they generate different output signals at a like encoder.

[0085] Partial picture a shows a sensor assembly B in combination with an encoder of type 21 a. When the sensor assembly B is displaced in a positive or negative X-direction, the sensor will react depending on the direction either with an increase or a decrease of the output signal. The analog signal can be quantized almost as sensitively as desired by way of an analog-digital converter in order to reach a high distance resolution. Partial picture b of FIG. 8 shows a sensor assembly C in combination with an encoder of type 17 a. The output signal of the assembly is a pulse train having a frequency which corresponds to the number of the scanned poles of the encoder.

[0086] A bit sequence with additional pieces of information is impressed between the pulses directly after a pulse, making use of the not required pulse pauses. It makes sense to configure the additional information so that the moving direction of the encoder may be taken from it.

[0087] Partial picture c shows an embodiment with two sensor assemblies of type D (reference numerals D 1 and D 2 ). However, sensor assemblies of type B may be used just as well (reference numerals B 1 and B 2 ). An appropriate encoder is e.g. one of type 17 a. The components D 1 (or B 1 ) and D 2 (or B 2 ) are offset in position with respect to each other by half a pole width and generate two signals which are orthogonal relative to each other. SIN or COS signals are produced when two sensor assemblies of types B are used. The output signal can be converted into a high-resolution quantized position signal with direction information corresponding to the fashion described hereinabove with respect to FIG. 8 (2α-sensor). Square signals are produced when two sensors of type D are used, from which signals the direction information may be derived likewise in a similar manner. In this respect, however, the attainable position resolution of the total distance is limited to the number of poles.

[0088] Further examples for encoder/sensor combinations are shown in FIGS. 9 a to 9 d.

[0089] The shank 7 with embedded encoder 9 (type 21 a ) illustrated in FIG. 9 a is combined with a 2α sensor 40 . The sensor's output signal is sent to an electronic circuit SC which generates the signal S(x).

[0090] FIG. 9 b shows a shank 7 with two embedded encoders 9 , 9 ′ of type 17 a. Sensor assemblies of type A which achieve high-resolution signals are referred to in this case. The said signals are then further processed in another electronic circuit SC 3 . The said circuit monitors the function of both sensor assemblies according to per se known redundancy principles and generates at the output the signal S(x) which is freed from disturbances, if any.

[0091] A shank 7 with two encoders of the type 17 a and 18 a with a different length is shown in FIG. 9 c. Encoder type 17 a is combined with a high-resolution sensor assembly of type A, and encoder type 18 a is combined with a sensor assembly D of low position resolution. In the local area where both sensor assemblies are in the range of the respectively associated encoder (redundancy range) the sensor signals are processed in another electronic circuit SC 3 in a per se known fashion according to redundancy principles to form an output signal S(x).

[0092] The embodiment in FIG. 9 d shows a tubular shank with a likewise tubular encoder slipped into the shank, and the encoder has an iron return path with respect to type 28 b. A redundant operation in connection with several sensors is this way especially simple. As sensors 40 two equal types with a high position resolution are apt for use, and as sensor D a type with a comparatively low position resolution is used. The sensors 40 along with the circuits SC 1 and SC 2 form sensor assemblies of the type A also in this arrangement. All three sensor signals are processed in an electronic circuit SC 3 according to per se known redundancy principles to form an output signal S(x). The cylinder-symmetrical shape of the encoder shank is especially favorable because the sensors can be arranged at any desired angle at the periphery of the shank. Iron particles are prevented from adhering in the interior of the shank due to the iron return path.

[0093] The encoder of type 28 b used is made of a plastics-bonded magnetic material.

[0094] The sensors or sensor assemblies which can be used according to the present invention are mainly commercially available.

[0095] Examples for commercial sensors and sensor assemblies will be listed hereinbelow:

[0096] Angular sensor LK28, IMO Wetzlar (2α)

[0097] Angular sensor LK15, IMO Wetzlar (2α)

[0098] AMR bridge KMZB34, Philips Hamburg (1α)

[0099] ASIC UA1272, Philips Hamburg (1α)

[0100] Active wheel sensor OH223, Philips Hamburg (1α)

[0101] Active wheel sensor OH243, Philips Hamburg (1α)

[0102] Active wheel sensor OH191, Philips Hamburg (1α)

[0103] Active wheel sensor OH203, Philips Hamburg (1α)

[0104] All listed 1α types are offered with a uniform housing shape and have a two-wire connection. The output signal is available in the form of a current signal.

[0105] Some dimensional information with respect to the embodiments is specified in the following:

[0106] thickness of the magnetic film of the encoder:

[0107] >=1 mm (preferred: 0.2 to 4 mm)

[0108] length of an encoder:

[0109] 54 mm (preferred: 10 to 100 mm)

[0110] Example for the Realization of a High Resolution in the Range of Roughly Δx=0.1 mm:

[0111] air slot(encoder/sensor element):

[0112] roughly 2 mm

[0113] (preferred: roughly 0.5 larger and roughly 5 mm smaller)

[0114] pole period λ:

[0115] 6 mm (preferred: less than or equal 6 mm)

[0116] sensor element:

[0117] angular sensor LK28, IMO, Wetzlar (2α)

[0118] network:

[0119] eightfold interpolation network type 601.3028.02, IMO, Wetzlar

[0120] Example for the Realization of a Resolution in the Range of Roughly Δx=2.0 mm:

[0121] air slot (encoder/sensor element):

[0122] 2 mm (preferred: roughly 0.5 to roughly 2.5 mm)

[0123] pole period λ:

[0124] 4 mm (preferred: 0.5 to 5 mm)

[0125] sensor element:

[0126] active wheel speed sensor OH223, PHILIPS Hamburg (1α)