Shape Memory Alloy Actuator
The present techniques generally relate to shape memory alloy actuators, and for methods of manufacturing such SMA actuators.
Consumer electronics devices, such as laptops and smartphones, may employ different types of controls to give users of the devices some feedback indicating that they have successfully pressed a button on the device. This is generally known as haptic feedback, and haptic buttons or controls on a device may provide a tactile sensation to the user to confirm that they have successfully pressed the button/control/switch. To generate a good haptic sensation, it is desired that the actuator moves with a high enough force to provide a sufficient displacement. It will be appreciated that there are other applications, for example latches or medical devices where a high force is also desired to achieve a relatively large displacement.
The present applicant has identified the need for an improved shape memory alloy actuator.
According to a first approach of the present techniques, there is provided a shape memory alloy (SMA) actuator comprising a static element; a moveable element which is moveable relative to the static element; a plurality of SMA wires which are each coupled to one or both of the static element and the moveable element and which on contraction cause movement of the moveable element; and a coupling element which couples at least two wires from the plurality of SMA wires to one of the static element and the moveable element.
Preferably, the coupling element comprises a crimp connector which holds the at least two wires.
According to a second approach of the present techniques, there is provided a haptic assembly comprising a touchable component (e.g. a button) and the actuator described above, wherein when a user presses or releases the touchable component, the actuator assembly is activated to provide haptic feedback to the user by moving the touchable component using the moveable element.
According to another approach of the present techniques, there is provided a method of manufacturing an SMA actuator (e.g . one as described above), the method comprising : feeding a plurality of SMA wires into an open crimp connector; closing the crimp connector; and trimming any excess wire.
According to a further approach of the present techniques, there is provided an apparatus comprising any of the SMA actuators described herein . The apparatus may be a smartphone, a camera, a foldable smartphone, a foldable image capture device, a foldable smartphone camera, a foldable consumer electronics device, a camera with folded optics, an image capture device, an array camera, a 3D sensing device or system, a servomotor, a consumer electronic device (including domestic appliances such as vacuum cleaners, washing machines and lawnmowers), a mobile or portable computing device, a mobile or portable electronic device, a laptop, a tablet computing device, an e-reader (also known as an e-book reader or e-book device), a computing accessory or computing peripheral device (e.g. mouse, keyboard, headphones, earphones, earbuds, etc.), an audio device (e.g. headphones, headset, earphones, etc.), a security system, a gaming system, a gaming accessory (e.g. controller, headset, a wearable controller, joystick, etc.), a robot or robotics device, a medical device (e.g . an endoscope), an augmented reality system, an augmented reality device, a virtual reality system, a virtual reality device, a wearable device (e.g. a watch, a smartwatch, a fitness tracker, etc.), an autonomous vehicle (e.g . a driverless car), a vehicle, a tool, a surgical tool, a remote controller (e.g . for a drone or a consumer electronics device), clothing (e.g. a garment, shoes, etc.), a switch, dial or button (e.g . a light switch, a thermostat dial, etc.), a display screen, a touchscreen, and a near-field communication (N FC) device. It will be understood that this is a non-exhaustive list of possible apparatus.
Preferred features are set out in the appended dependent claims.
Implementations of the present techniques will now be described, by way of example only, with reference to the accompanying drawings, in which :
Figure la is a schematic cross-sectional view of a first haptic assembly;
Figure lb is a schematic cross-sectional view of a second haptic assembly;
Figure lc is a schematic enlarged view of a detail within Figure lb;
Figures 2a and 2b are schematic block diagrams of the functional arrangement of the haptic assemblies shown in Figures la to lc;
Figure 3a is a graph plotting the variation in temperature against time for a single wire having a diameter of 60mhi which is being heated, for each of the central wire (W2) and an outer wire (Wl) in a set of three adjacent touching wires which each have a diameter of 35mGh and which are being heated and for each of the central wire (W2) and an external wire (Wl) in a set of three wires which are being heated and which each have a diameter of 35mhi but are spaced apart so that the wires are adjacent but not touching;
Figure 3b is a graph plotting the variation in temperature against time when the arrangements of wires of Figure 3a are cooled;
Figure 4a is a schematic cross-sectional view of another haptic assembly which is similar to that shown in Figure lb;
Figure 4b is a graph plotting the variation in temperature against time for various distances (e.g. 420pm to 40 pm) between a heatsink and a wire as the wire is cooled;
Figure 4c is a graph plotting the variation in temperature against distance between a heatsink and a wire as the wire cools, wherein the temperature is the value after 100ms or 200ms of cooling respectively;
Figure 4d is a graph plotting the variation in temperature against time for various distances (e.g. 420pm to 40 pm) between a heatsink and a wire as the wire is heated;
Figure 5 is a graph plotting the variation in time to recover for different displacement for each of three different arrangements;
Figures 6a and 6b are schematic partial illustrations of different arrangements of crimps for use in an assembly;
Figure 7 is a schematic cross-sectional view of a third haptic assembly; and
Figure 8 is a schematic illustration of a method of crimping wires in an SMA actuator.
The embodiments described below illustrate a shape memory alloy (SMA) actuator. Such an SMA actuator may be any type of device that comprises a static part (or element - the words may be used interchangeably) and a moveable part which is moveable with respect to the static part. The moveable part is moved by a plurality of SMA wires which are coupled (or connected - the words may be used interchangeably) between the static part and the moveable part. A coupling element couples at least two wires to at least one of the static part and the moveable part. The coupling element may couple any number (N) of wires, and merely as an example between two to six wires.
The coupling element may provide a direct connection between the at least two SMA wires and one of the static element and the moveable element. In an arrangement having a direct connection at both ends of the SMA wires, "coupled between" means that only the SMA wires and the coupling elements are between the static element and the moveable element. Alternatively, the coupling element may couple the at least two wires to an intermediate component to provide an indirect connection between the at least two wires and one of the static element and the moveable element. The intermediate element itself may be connected to one of the static element and the moveable element by a direct or indirect connection . For an indirect connection, the intermediate element may itself be connected to at least one other intermediate element and for a direct connection, the intermediate element is connected direct to one of the static element and the moveable element.
The coupling element may comprise a fixed connector which provides a permanent (i.e. fixed) connection between the SMA wires and the static element or the moveable element. Such a fixed connector may be in the form of a crimp connector, a welded component that is welded to the at least two wires to form a weld, or other similar connectors. When crimp connectors are used, the coupling element may comprise a single crimp connector which holds multiple wires or may comprise multiple adjacent crimp connectors, each of which hold a single wire. The or each crimp connector may have a width of between 500mGh to 750mGh. The adjacent crimp connectors may be vertically aligned, i.e. may be in a stack. Alternatively, the adjacent crimp connectors may be laterally offset from each other, or vertically offset from each other, or laterally and vertically offset from each other.
The coupling element may alternatively comprise a connector which provides a non-fixed connection between the plurality of SMA wires and the static element or the moveable element. Such a non-fixed connector may be in the form of a protruding element such as a hook, dowel pin or similar element around which the SMA wires are looped or similarly held in place.
There may be a first coupling element coupling one end of the SMA wires to the static element and a second coupling element coupling the opposed end of the SMA wires to the moveable element. The first coupling element may be the same type as the second coupling element, i.e. both the first and second coupling elements may be crimp connectors which hold multiple wires, may comprise a plurality of adjacent crimp connectors each of which hold a single wire or may be welds. Alternatively, the first coupling element may be a different type to the second coupling element, i.e. the first coupling element may comprise one or more crimp connectors and the second coupling element may be a protruding element or the first coupling element may be a crimp connector holding multiple wires and the second coupling element may comprise a plurality of adjacent crimp connectors each holding a single wire.
A first coupling element and a second coupling element which are coupled to the same wires may be considered to form a pair of coupling elements. There may be multiple pairs of first and second coupling elements with each pair of
coupling elements coupling at least two different wires to both the static element and the moveable element. In this way, the plurality of wires may be divided between the pairs of coupling elements. For example, there may be six wires with a first set of three wires coupled between a first pair of coupling elements and a second set of three wires coupled between a second pair of coupling elements.
When the plurality of wires are activated, e.g. in a haptic assembly in response to detecting a requirement for haptic feedback, the SMA wires contract which moves the moveable part. A first pair of coupling elements may couple a first set of at least two wires to the moveable element and a second pair of coupling elements may couple a second set of at least two wires to the moveable element, wherein when the wires coupled to the first pair and the second pair of coupling elements contract the moveable element is moved in one direction. For example the first pair of coupling elements may couple the at least two wires to one side of the moveable element and the second pair of coupling elements may couple the at least two wires to the opposed side of the moveable element and the direction of movement may be generally parallel to the opposed sides of the moveable element. Arranging the pairs of coupling elements on opposite sides of the moveable element may lead to a balanced application of the force from the contraction of the wires, i.e. may avoid rotation of the moveable element. A similar result may be achieved by coupling at locations which are close to the opposed edges of an end of the moveable element. These are just examples and any suitable coupling locations on the moveable element may be used.
The moveable part may be restored to its original position by a restoring element which provides a restoring force. The SMA wires may also be returned to their original length by the restoring element when the activation of the SMA wires ceases, e.g. the power is removed. The restoring element may be a resilient element, e.g. a spring, flexure, other SMA wires, or a force applied at the surface of the touchable component by the user's finger. For example, there may be a first pair of coupling elements which couple a first set of at least two wires to the moveable element and a second pair of coupling elements which couple a second set of at least two wires to the moveable element, wherein when the first set of at least two wires contract the moveable element is moved in a first direction and when the second set of at least two wires contract the moveable element is moved
in the opposite direction to the first direction. For example, the first pair of coupling elements may couple the first set of at least two wires to one end of the moveable element and the second pair of coupling elements may couple the second set of at least two wires to the opposed end of the moveable element. This is just one example and any suitable coupling location on the moveable element may be used.
Embodiments of the present techniques describe SMA actuators which are designed to deliver a high force, e.g. between 1.2 to 3N, more preferably between 1.2 to 10N, whilst maintaining the strain in the wire within safe limits (e.g. 2-3% reduction in length over original length). The force will be dependent on the target displacement required. A plurality of relatively thin wires (e.g. approximately 25mGh or 35mGh in diameter) are used in combination to provide a desired force. As explained in more detail below, the use of a plurality of wires provides an overall cross-section which is designed to deliver the desired force. Furthermore, the use of a plurality of wires allows the wires to cool more quickly than a single wire of similar cross-section. Accordingly, the plurality of thinner SMA wires is ready to be reactivated more quickly than a single wire with equivalent cross- sectional area.
The SMA actuators may be incorporated in a haptic assembly to move a button or other touchable element which is contacted by a user (moveable element) relative to a casing or housing (static element) to deliver a haptic sensation to a user pressing on the button (or other touchable element). Embodiments of the present techniques describe haptic assemblies which may be arranged to move the button laterally along the edge of the device, perpendicularly with respect to the edge of the device, helically around an axis perpendicular to the edge of the device or in any other suitable direction, e.g. in plane rotation parallel to the edge of the device or perpendicular to the device. Examples of actuators which move the button in a lateral direction with respect to the contact by the user are described for example in WO2018/046937 and GB2551657. Examples of actuators which generate vertical movement are described in GB1803084.1 and GB1813008.8 to the present applicant. Examples of arrangements of crimps which may be used to connect the SMA actuators into
the haptic assembly are described in WO2016/189314, GBI800484.6,
G8I80I291.4 and GB1815673.7.
The present techniques may provide a local haptic sensation caused by a direct impulse, rather than through inertial effects. For example, smartphones comprise inertial haptic actuators - a significant mass is moved when a haptic effect is required. Movement of the mass causes the whole smartphone to shake or vibrate. Thus, the haptic effect is general and is not localised. The present techniques provide a localised haptic feedback. Further still, the haptic feedback provided by the present techniques may be customisable by a user by modifying software parameters. This allows different types of haptic feedback to be provided for different purposes or to suit different users.
Each of the haptic assemblies described herein may be incorporated into any device in which it may be useful to provide a user of the device with haptic feedback. For example, the haptic assemblies may be incorporated into any of the electronic devices or consumer electronics devices listed previously, including but not limited to a computer, laptop, portable computing device, smartphone, computer keyboard, gaming system, portable gaming device, gaming equipment/accessory (e.g. controllers, wearable controllers, etc.), medical device, user input device, etc. It will be understood that this is a non-limiting, non- exhaustive list of possible devices, which may incorporate any of the haptic assemblies described herein. The haptic assemblies described herein may be, for example, incorporated into or otherwise provided along an edge of a smartphone or on a surface of a smartphone.
Various SMA actuators are now described with respect to the Figures. It will be understood that elements or features described with respect to one particular Figure may equally apply to any of the Figures described herein, for example, crimps holding multiple wires may be used in combination with crimps holding single wires, the overall number of SMA wires may be selected to provide the desired force and the heatsink(s) may be incorporated in any embodiment. Furthermore, although the Figures show the incorporation of the SMA actuators in a haptic assembly it will be appreciated that the SMA actuators may be
incorporated in other devices requiring a high force such as latches or medical devices.
Figure la shows a cross-sectional view of a first arrangement of an SMA actuator within a haptic assembly 100. The haptic assembly 100 comprises a button 102 (but it will be appreciated that other touchable components, surfaces or elements may be used interchangeably). The button 102 may be pressed by a user to perform a particular operation, such as making a selection, turning a device on/off, entering data (e.g. typing on a keyboard), scrolling, turning a function of the device in which the assembly 100 is located on/off or adjusting the function (e.g. adjusting volume of audio output from the device), etc. Pressing or releasing the button 102 may cause haptic feedback or a haptic sensation to be delivered to the user, so that the user is provided with some sensory feedback (particularly touch-based feedback) to indicate that the operation has been performed.
In many of the arrangements described herein, the button 102 may be a surface feature on a device/apparatus that incorporates the haptic assembly. Such a surface may be pressed in a similar way to a button and pressing or releasing the surface may be detected by a sensor and the haptic feedback may be triggered. However, instead of a press or release of a button (or surface) triggering haptic feedback, the haptic feedback may be triggered by software in response to another event. For example, if a user makes a selection on a screen of their smartphone, the selection may cause haptic feedback to be triggered, where the feedback is provided by the button or surface feature. (Software triggered haptic feedback may occur in particular applications, such as in gaming and/or virtual/augmented reality devices). Thus, in many of the arrangements and embodiments described herein, direct pressing of the haptic button 102 may not be required in order for haptic feedback to be delivered. However, in each case, the mechanism will determine which sensor registered a need to provide haptic feedback and where there is a plurality of actuators to provide the feedback, the mechanism may determine which is the appropriate actuator(s) to be activated to provide the haptic feedback.
In the arrangement shown in Figure la, the haptic assembly 100 may comprise a housing 104 (also referred to herein as "support", "chassis", "casework" and "casing"). The housing 104 may comprise a cavity or recess. The button 102 may be provided within the cavity of the housing 104. As shown, the button may be arranged with the cavity such that a contact surface (may also be referred to as an outer surface, external surface or upper surface) of the button is substantially level with/flush with an external surface of the housing 104. Alternatively, the button may protrude from the housing. It will be understood that the housing 104 surrounds and encases the button 102, such that only the contact surface 106 of the button is visible/contactable by the user.
The SMA actuator may comprise an intermediate, moveable element 106 which is provided within the cavity below the button 102. The moveable element 106 is moveable relative to the static part (i.e. the housing) in a first direction that is perpendicular to the external surface of the housing 104. Contact of a user's finger on the contact surface of the button may cause the button to move into the housing, e.g. in the direction of arrow 116, or out of the housing as appropriate. A sensor (not shown) may be mounted in the housing below the button 102 and the moveable element 106. The sensor is any suitable sensor for determining that haptic feedback is required, e.g. by detecting depression of the button. The sensor may be coupled to control circuitry (not shown), and the sensor may be configured to communicate with the control circuitry when the haptic feedback is required. For example, the sensor may detect when the force on the sensor changes, or when the force on the sensor has been applied for a minimum duration. As an example, the detection by the sensor of a user pressing the button causes the haptic feedback to be generated and applied by the haptic assembly.
The moveable element 106 is also moveable relative to the housing in a second different direction which may be substantially perpendicular to the first direction, e.g. substantially parallel to the external surface of the housing 102. Movement of the moveable element 106 in the second direction may cause movement of the button 102 in the first direction, so as to provide a haptic sensation to the user. The concept of moving the intermediate moveable element
106 in one direction to cause movement of the button 102 in another direction may be implemented in a number of ways.
There are a plurality of bearings 128 positioned between the sloped side of the moveable element 106 and the adjacent sloped side of the button to reduce friction between the moveable element and the button. Similarly, there are a plurality of bearings between the moveable element 106 and the housing to facilitate movement of the moveable element by reducing friction.
The term "bearing" is used interchangeably herein with the terms "sliding bearing", "plain bearing", "rolling bearing", "ball bearing", "flexure", and "roller bearing". The term "bearing" is used herein to generally mean any element or combination of elements that functions to constrain motion to only the desired motion and reduce friction between moving parts. The term "sliding bearing" is used to mean a bearing in which a bearing element slides on a bearing surface, and includes a "plain bearing". The term "rolling bearing" is used to mean a bearing in which a rolling bearing element, for example a ball or roller, rolls on a bearing surface. The bearing may be provided on, or may comprise, non-linear bearing surfaces. In some embodiments of the present techniques, more than one type of bearing element may be used in combination to provide the bearing functionality. Accordingly, the term "bearing" used herein includes any combination of, for example, plain bearings, ball bearings, roller bearings and flexures. In embodiments, a suspension system may be used to suspend the intermediate moveable element and/or the button within the haptic assembly and to constrain motion to only the desired motion. For example, a suspension system of the type described in W02011/104518 may be used. Thus, it will be understood that the term "bearing" used herein also means "suspension system". In embodiments, the bearing may be provided on, or may comprise, non-linear bearing surfaces. The bearing may be formed from any suitable material, e.g. ceramic.
In this arrangement, the button 102 and the moveable element 106 are wedge shaped such that a wider end of the wedge-shaped button 102 is in proximity to a narrower end of the wedge shaped moveable element 106. This arrangement means that when the moveable element 106 is caused to move
within the housing 104 in the second direction (i.e. generally parallel to the surface of the housing), the button is caused to move in the first direction (i.e. generally perpendicular to the surface of the housing). In this arrangement, the intermediate moveable element 110 is a "single wedge", as only one surface of the element is sloped/inclined.
The SMA actuator comprises a plurality (e.g. N) of shape memory alloy (SMA) actuator wires 108, for example there may be between 2 to 6 wires. As shown the SMA wires 108 extend into a further cavity 112 in the housing 104. Both ends of the SMA actuator wires 108 are connected to the housing 104 using a coupling element in the form of a pair of connectors/crimps 110a, 110b which are electrical and mechanical connectors (to connect the SMA actuator wires to a power supply). Each connector holds the ends of the multiple wires. Each of the SMA wires 108 is hooked at their midpoint over a hook 120 provided on a side of the moveable element 106. The hook is thus another coupling element which couples and connects the multiple SMA wires to the moveable element. Each half of each wire may be considered to form an active wire section and the two active wire sections of each SMA actuator wire mechanically act in parallel and therefore, each looped SMA actuator wire may provide twice the force of a single wire which only spans once from the movable element to the housing.
In this arrangement, there is a coupling element which couples each SMA wire at its midpoint to the moveable element using a hook and two other coupling elements in the form of crimps each of which couples the plurality of SMA wires at one end to the static element. Thus, the coupling element which couples each SMA wire to the moveable element has a different structure (i.e. is a different type) to the coupling element which couples each SMA wire to the static element. Furthermore, in this arrangement, the crimps are between each SMA wire and the static element and thus each SMA wire has a direct connection to the static element but it will be appreciated that the connection may also be indirect and thus there may be one or more intermediate elements between each SMA wire and the static element.
When a haptic sensation is required, this requirement is communicated to control circuitry (not shown). Power is then delivered to each SMA actuator wire
108. When each SMA actuator wire 108 is powered, it becomes hot and contracts. The contraction of each SMA actuator wire 108 causes the intermediate moveable element 106 to move laterally within the cavity and towards the further cavity 112. As the moveable element 106 moves sideways, the wedge-shape of the moveable element 106 forces the button 102 to move upwards. In other arrangements, the wedges may be arranged so that the button moves down or moves laterally. The intermediate moveable element may cause the button to move by, for example, between 20mGh to 0.5mm. In some embodiments, the button may move by as much as 1mm.
The haptic assembly may comprise a restoring element 126 which opposes the force of the SMA actuator wires 108. The restoring element 126 may be provided within the further cavity 112 and may be coupled at one end to the housing 104 and at the other end to the moveable element 106. The restoring element 126 (e.g. a return spring or any suitable biasing resilient element) may be arranged to oppose the contraction of the SMA wires 108 and thereby move the moveable element in the opposite direction when the SMA wire 108 is not powered. It will also be appreciated that the restoring element 126 may comprise one or more additional SMA wires which on contraction pull the moveable element 104 in the opposite direction to the plurality of SMA wires 108.
There may be an endstop 114 which may be formed as part of the housing within the cavity or may be a separate element within the cavity. The endstop may be provided at a location in the cavity to restrict movement of the moveable element. Generally speaking if the SMA actuator wires are stretched too far (i.e. a certain tension is exceeded), the SMA actuator wires may weaken or become damaged, or even break. The force of the restoring element 126 on the intermediate moveable element 110 may cause the SMA actuator wires to become overstretched. Therefore, the endstop 114 may restrict the movement of the intermediate moveable element 110 so that the SMA actuator wires 108 do not overstretch. Similarly, a force applied to the button surface by the user's finger may cause the wires to overstretch if there is no endstop.
Figure lb shows a plan view of another haptic assembly 100' comprising three pairs of parallel SMA actuator wires. The function and structure of the haptic
assembly 100' is the same as that shown in Figure la, except for the different arrangement of wires, and thus for conciseness, like features are not described. In this arrangement, the actuator wires in each pair are coupled to opposite sides of the intermediate moveable element 106. In some arrangements, by opposite sides it is meant that the actuator wires may be coupled to the opposed sides of the intermediate moveable element 106 which are parallel to the direction of the movement of the intermediate moveable element. In other arrangements, opposite sides can mean that the actuators wires may be coupled to the same end face of the intermediate moveable element but at either side of the end face. In both arrangements, the pairs of wires act in the same direction (i .e. may apply a force to the intermediate moveable element in the same direction) to provide double the force compared to a single wire coupled to one side of the intermediate moveable element.
Three SMA actuator wires, one from each pair of wires, are coupled at one end to the intermediate moveable element 106 via a coupling element in the form of a crimp 110c (or a crimp connector - the words can be used interchangeably) and at the other end to the housing 104 via a coupling element in the form of a crimp 110a . The pair of crimps 110a, 110c may be considered to form a first pair of coupling elements which couples three wires to both the intermediate moveable element 106 and the housing 104. Each crimp 110a, 110c holds the opposed ends of three wires. In this arrangement, each SMA wire has a direct connection to the housing but it will be appreciated that the connection may also be indirect and thus there may be one or more intermediate elements between each SMA wire and the housing.
Similarly, on the opposite side of the moveable element (i .e. on the side which is not visible as shown in Figure l b), three actuator wires; one from each pair (not shown), are coupled at one end to the intermediate moveable element 106 via a connector or crimp and at the other end to the housing 104 via a connector or crimp. Thus, again three SMA actuator wires are coupled by a pair of coupling members to both the intermediate moveable element 106 and the housing.
Figure lc shows schematically the detail of the crimp 110a which connects the ends of each of the three wires to the housing. Although an arrow is shown from Figure lb, it will be appreciated that the detail of the crimp in Figure lc also applies to Figure la. In each individual crimp, there are the ends of three wires 108a, 108b, 108c. There may be spacing between each wire and such spacing may be uniform. However, there may also be no or little spacing between the wires.
The width w of each crimp may be sufficient to hold and to connect to each of the three wires. The crimp width w is the size of the crimp parallel to the SMA wires within the crimps. The width thus defines how much of each SMA wire is held in the crimp. The width may be between 400mGh to 750mGh with a standard crimp typically having a width of 500mGh. For example, a crimp which holds three wires may have the standard width. A larger crimp may form a better mechanical connection between the crimp and wires because there is more crimp material. The length is the dimension of the crimp which is perpendicular to the width and which is defined after the crimp has been folded over the wires. The length is typically 450mGh, i.e. an unfolded crimp is typically 900mGh. The crimp may have a thickness which is dependent on the material. Any suitable material which forms a mechanical and electrical connection may be used for the crimp, e.g. phosphor bronze or stainless steel. The crimp may be coated, e.g. with gold or another suitable material, to reduce corrosion and/or reduce the resistance of the electrical connection to the wires. A very thick or thin piece of material may be more difficult to fold, or may not form a good mechanical connection with the wire. For example, the folded crimp may have a total thickness of IOOmGh. The dimensions of the crimp may be selected to balance the requirements in relation to spacing between the wires, as well as providing an acceptable mechanical and electrical connection.
Figures 2a and 2b schematically illustrate the functional arrangement of the SMA wires of Figures la and lb. Figure 2a schematically illustrates an example of three looped SMA wires 208a, 208b, 208c. Each of the three looped SMA wires may be located within the SMA actuator of the haptic assembly as described in relation to Figure la to provide the functionality described above. One end of each wire is connected to the housing 205 by a coupling element 210a (e.g. a crimp connector or a weld). Each wire loops around a hook as described
above and the other end of each looped wire is connected to the housing 205 by a coupling element 210b (e.g. a crimp connector or a weld). Each coupling element 210a, 210b thus contains three wires and each coupling element 210a, 210b is connected to the static part (i.e. the housing). As explained above, this coupling may be direct as illustrated or may be indirect, e.g. via an intermediate element (not shown).
Figure 2b schematically illustrates an example of SMA actuator of a haptic assembly using three pairs of SMA wires (218a, 218f), (218b, 218e), (218c, 218d). Each of the three pairs of SMA wires may be located within the SMA actuator of the haptic assembly as described in relation to Figure lb to provide the functionality described above. Ends of each pair of SMA wires are connected to coupling members which may be permanent connectors such as welds or crimps. Coupling members 210a, 210b are connected to the housing (i.e. static part) and may be termed static coupling members. Similarly, coupling members 210c, 210d are connected to the moveable element 206 and may be termed moveable coupling members because they move together with the moveable element but do not separately move. The static and moveable coupling members form two pairs of coupling members (210a, 210c), (210b, 210d) which couple the SMA wires to both the static and the moveable elements. As explained above, this coupling may be direct as illustrated or may be indirect, e.g. via an intermediate element (not shown).
A first or outer pair of SMA actuator wires comprises a first outer SMA actuator wire 218a and a second outer SMA actuator wire 218f. The first outer SMA actuator wire 218a is coupled at one end to the intermediate moveable element 206 via a first moveable coupling member 210c and at the other end to the housing via a first static coupling member 210a. The second outer SMA actuator wire 218f is coupled at one end to the intermediate moveable element 206 via a second moveable coupling member 210d and at the other end to the housing via a second static coupling member 210b. Similarly, the second or central pair of SMA actuator wires comprises a first central SMA actuator wire 218b coupled to the first moveable coupling member 210c and the first static coupling member 210a and a second central SMA actuator wire 218e coupled to the second moveable coupling member 210d and the second static coupling member 210b.
The third or inner pair of SMA actuator wires comprises a first inner SMA actuator wire 218c coupled to the first moveable coupling member 210c and the first static coupling member 210a and a second inner SMA actuator wire 218d coupled to the second moveable coupling member 210d and the second static coupling member 210b. Thus each of the first wires 218a, 218b, 218c in each pair of wires is connected to both the static and moveable parts by a first pair of coupling member 210a, 210c and each of the second wires 218d, 218e, 218f in each pair of wires is connected by a second pair of second coupling members 210b, 210d.
In Figure la (as schematically illustrated in Figure 2a), the halves of the wire loop mechanically act in parallel and therefore, each half of each wire may be considered to form an active wire section. Thus, the arrangement in Figure la has effectively six wires acting in parallel. Similarly, in Figure lb (and Figure 2b), using three pairs of wires also provides six wires acting in parallel. Thus both arrangements provide the potential to achieve six times the force when compared to a single wire. This is because the maximum force which is available is proportional to the diameter of the wire. It will be appreciated that the SMA force will vary with the load applied to the wires but some values are provided merely for illustration of the increased force generated by the arrangements. For example, a wire having a cross-section of 25mGh typically generates a maximum force of between 120mN to 200mN and thus six wires (or six active wire sections) provide a force of approximately 720mN to 1.2N. Increasing the diameter of each wire from 25mGh to 35mGh approximately doubles the cross-sectional area of each wire and thus approximately doubles the force provided by each wire. The maximum total force which can be provided by six wires having a 36mGh diameter (or six active wire sections) is in the range of 1.5N to 3N.
The arrangements of Figures la and lb may thus provide a significantly higher force than known SMA actuators without compromising other properties of the assembly as explained in more detail below. Typically, SMA wires are operated so that the strain on the SMA wire is kept within a low limit (e.g. 2-3%) to prevent damage to the wire. Providing a higher force with a SMA wire may increase the stress within the wire unless other factors of the SMA wire are changed. For example, the force provided by an SMA wire is related to the cross- section of the wire. Increasing the cross-section of the wire means that the total
force available is increased. However the volume of material that needs to be heated to activate the wire is also increased and to gain access to the higher force, the power also needs to be adjusted. There is a risk that using a higher power with a smaller diameter wire may damage the wire due to excessive strain. Another issue to be addressed is that a wire having a larger cross-section has a slower cooling rate than one with a smaller cross-section which may affect performance as explained below.
Figures 3a and 3b illustrate the simulated heating and cooling rates for three different arrangements: a single wire having a diameter of bqmhh, three adjacent touching wires each having a diameter of 35mGh and three wires each having a diameter of 35mGh but spaced apart so that the wires are adjacent but not touching. A single wire of diameter 60mhi has a cross-sectional area of 2.8xl0 9m2 which is similar to the combined cross-sectional area of 2.9xl0 9m2 for the three wires. For each of the three wire arrangements, the temperature of the centre wire W2 is simulated separately from the temperature of the outer wire Wl. The simulation parameters are ambient temperature of 25 degrees Celsius and a cooling starting point of 150 degrees Celsius.
As shown in Figure 3a, there is minimal difference in the temperature of both of the three wire arrangements over the first few milliseconds. The temperature of the centre wire W2 is also similar to the temperature of the outer wire Wl . Initially, there is also minimal difference over time in the temperature of the single wire arrangement when compared to both of the three wire arrangements. Nevertheless, after 10ms, the three wire arrangements are approximately 6 degrees cooler than the single wire arrangement. Beyond 10ms, the single wire arrangement begins to increase in temperature more rapidly than each of the three wire arrangements so that at 20ms, there is approximately an 18-20 degree difference in temperature.
The differences between the single wire arrangement and the three wire arrangements are more significant on cooling. As shown in Figure 3b, there is minimal difference over time in the temperature of both of the three wire arrangements as they cool. The temperature of the centre wire W2 is also similar to the temperature of the outer wire Wl with only a relatively small difference
(e.g. 5 degrees) between the centre and outer wires in the spaced arrangement after 10ms and a smaller still difference (e.g. 2 degrees) between the centre and outer wires in the touching arrangement after 10ms. The three wire arrangements cool significantly quicker than the single wire arrangement and are approximately 50 degrees cooler after 200ms. For the same overall cross-section, the single wire arrangement takes approximately 3.2 times longer than the three wire arrangement to cool (i.e. 0.12 seconds versus 0.38 seconds to cool from 150 degrees to 60 degrees).
Figures 3a and 3b thus show that the three wire arrangement will heat almost as quickly as a single wire arrangement so the heating event which activates the actuator and provides the contraction to move the moveable element is similar for both arrangements. However, the three wire arrangement provides a significant advantage over the single wire arrangement for cooling. The cooling is significantly quicker and thus the wires in the three wire arrangement will cool and return to their original shape much quicker than the thicker, single wire. The three wire arrangement is thus ready for reactivation more quickly that the single wire arrangement. The simulations in Figures 3a and 3b also illustrate that the heating and cooling of the three wire arrangement when the wires touch is similar to when the wires are spaced apart which is perhaps counter-intuitive. Furthermore, each of the wires within the multiple wire arrangement is at approximately the same temperature and thus there is no loss of performance which may have been expected if the wires were at different temperatures.
Thus in summary, the heating and cooling issues associated with a larger diameter wire may be addressed by using several smaller diameter wires giving approximately the same total volume of material and hence the same maximum total force available. In addition to being able to provide more force, the use of multiple wires may improve the reliability of the device. This is because if one wire breaks in a crimp holding multiple wires, there is still at least one other wire in the crimp which is connected. There is a desire in some industries to not have single wires because of the reliability issue.
Heatsink arrangement
Figure 4a shows a variant of the arrangement of Figure lb. All the unchanged elements retain the same reference numbers and for the sake of conciseness, the function and structure of these unchanged elements is not repeated. The haptic assembly 100" comprises a heatsink 130 which in this arrangement is a separate element on the housing 104 near the SMA wires. A single heatsink is illustrated but it will be appreciated that multiple heatsinks may be incorporated. The heatsink may be at a distance which is less than 5 times the diameter of each wire from the wires. More preferably the heatsink may be at a distance of less than 3 diameters, more preferably still less than 2 diameters. The heatsink may touch the wires. The heatsink 130 may be made of any suitable material, e.g. aluminium, phosphor bronze or steel, which enhances the cooling rate of the wires. The heatsink 130 may have a substantial thermal mass and/or high thermal conductivity to achieve the increased cooling.
In the arrangement, schematically illustrated in Figure 4a, the heatsink is a small distance from the wires but it will be appreciated that an arrangement in which the heatsink is in contact with the wires may also be used. The direct contact is likely to increase the rate of cooling but may lead to an increase in the power required to instantaneously heat the wire. Nevertheless, such an arrangement may be desirable for tolerance reasons as it means that the wire does not have to be positioned relative to the heatsink with high precision.
The heatsink and/or the wires may also be configured to be moveable relative to one another between an activated position in which the heatsink is adjacent (close to or even touching) the wires and an equilibrium position in which the heatsink is further from the wires. Before actuation of the SMA wire, the SMA wire is at room temperature and the heatsink is in the equilibrium position. In this way, the SMA wires may be heated without requiring any additional power. Once the SMA wires are heated, the wires are at an elevated temperature and the heatsink may be moved from the equilibrium position to the activated position to ensure good cooling rates for the wires. Additional cooling mechanisms, e.g. air flow around the wires, may also be triggered as the wire cools.
Figures 4b to 4d are graphs from a simulation to show the effects of a heatsink. The cooling and heating rates are calculated for a single wire of 60mhi diameter with a heatsink positioned at a distance ranging from 40mGh to 420mGh from the wire. It will be appreciated that similar results are likely to be achieved with the multiple wire arrangements.
Figure 4b plots the change of temperature over time when cooling the single wire with the heatsink at different distances from the wire. As expected, the presence of the heatsink increases the cooling rate particularly when the heatsink is closer to the wire.
Figure 4c plots the temperature change with distance between the wire and the heatsink at 100ms and 200ms after cooling has started. As expected, after 100ms of cooling, the temperature is higher for each position of the heatsink than after 200ms of cooling. However, both line graphs follow a similar curve with the temperature after the fixed amount of time increasing as the distance between the wire and the heatsink increases. After 200ms of cooling, a wire with a heatsink situated 250mGh from the wire will have cooled from 150 degrees Celsius to 83 degrees Celsius. After 200ms of cooling, a wire with a heatsink situated IOOmhh from the wire will have cooled from 150 degrees Celsius to 70 degrees Celsius. Thus, there is less cooling for the heatsink which is further away. A similar result is achieved after 100ms of cooling. Both Figures thus show that the proximity of a heatsink changes the cooling rate significantly, with a closer heatsink improving cooling.
Figure 4d plots the change of temperature over time when heating the single wire. As shown, the presence of the heatsink does not have a significant effect on short-term heating rates (e.g. before 30ms) until it is very close, e.g. just 40mGh away. Moreover, the period for which the actuator is driven is very short for some applications such as haptics or latches, typically less than 10ms and as shown, there is virtually no effect on the heating caused by the heatsink. Accordingly, any increase in power required is very small in such applications because there is no requirement to hold the wire at a higher temperature for a long time. Thus, the benefit in cooling rates is not adversely affected by a detrimental effect on heating. This is in contrast to other actuators where the
wire is driven for longer and so a heatsink is less desirable due to the increased power requirement to heat the wire.
Crimping arrangements
In the arrangements shown above, three wires are held together in each individual coupling member (e.g. crimp) although as set out above, there may be any number of wires in each coupling member, e.g. between 2 to 6 (or more). Figure 5 is a graph showing the time to recover against displacement for a single bOmGP diameter wire in a crimp, a single 35mGh diameter wire in a crimp and three 35mGh diameter wires in a crimp. The results in Figure 5 are obtained from experimental data rather than simulated data. As shown in Figure 5 and in line with the results above, the recovery time for the single 60mhi diameter wire is significantly worse than the other arrangements. Accordingly, more time will be needed between actuation events to allow the wire to recover to produce the next contraction. The experimental data suggests that the recovery time for the single bOmGP diameter wire is approximately 1.75 times slower than the three 35mGh diameter wire arrangement. This is a smaller difference than the simulation data but nevertheless the three wire arrangement provides a significantly improved cooling rate compared to the single wire arrangement having a similar cross- section.
Figure 5 also shows the cooling time for a single 35mGh diameter wire. As expected, the single thinner wire has a significantly shorter recovery time than each of the arrangements having a larger cross-section. For example, if in each arrangement the contraction of the wire(s) results in a displacement of 40mGh, the bOmGP diameter wire takes approximately 275ms to recover, the 35mGh diameter wire takes approximately 115ms to recover and the three wire arrangement is between the two extremes at approximately 160ms of recovery time.
Another variation of the arrangements shown in Figures la and lb is to have each wire held in its own separate crimp. Figure 6a shows one variation with a single crimp for each SMA wire. The arrangement of the moveable and static elements is not shown for simplicity but the components are supported on a chassis 600 for ease of assembly. In this arrangement, there are two pairs of
wires (608a, 608c), (608b, 608d). A first upper pair of wires 608a, 608c is arranged above a second lower pair of wires 608b, 608d. There are thus two wires on each side of the chassis 600.
Upper crimps 610a, 610b, 610c, 610d each have a lower crimp 610a', 610b', 610c', 610d' beneath them, i.e. the crimps are vertically aligned. Each stacked pair of crimps at either end of a wire (610a, 610a', 610c, 610c'), (610b, 610b', 610d, 610d') may be considered to form a coupling element and thus each coupling element is coupled to multiple wires albeit by using different parts, e.g. individual crimps at each end of each wire. The stacked crimps (610a, 610a'), (610c, 610c') from a first pair of coupling elements with a first coupling element (610c, 610c') coupling one end of the two wires to the static element and a second coupling element (610a, 610a') coupling the opposed end of the two wires to the moveable element. Similarly, stacked crimps (610b, 610b'), (610d, 610d') from a second pair of coupling elements with a first coupling element (610d, 610d') coupling one end of the two wires to the static element and a second coupling element (610b, 610b') coupling the opposed end of the two wires to the moveable element.
In this arrangement there are four wires in total but it will be appreciated that the six wire arrangement of Figure lb could be achieved by having three wires arranged on either side of the chassis with the corresponding sets of three pairs of crimps arranged on either side of the chassis. If the wires are arranged vertically, one on top of the other as in Figure 6a, the crimps can also be arranged in a vertical stack. Similarly, if two wires are desired, there could be either one wire on each side or a pair of wires on one side. The arrangements may have between 2 to 6 wires in each coupling element which comprises a plurality of crimps.
A first upper wire 608a is connected at one end to a first upper moveable crimp 610a which when the assembly is installed will be connected to the moveable portion, e.g. an intermediate moveable element as in the arrangement of Figure lb. The first upper wire 608a is connected at the other end to a first upper static crimp 610c which when the assembly is installed will be connected to the static portion, e.g. a housing as in the arrangement of Figure la. Similarly,
a second upper wire 608c is connected at one end to a second upper moveable crimp 610b and at the other end to a second upper static crimp 610d. For the lower wires, the first lower wire 608b is connected at one end to a first lower moveable crimp 610a' and at the other end to a first lower static crimp 610c' and the second lower wire 608d is connected at one end to a second lower moveable crimp 610b' and at the other end to a second lower static crimp 610d'.
Arranging the wires and the crimps in a vertical stack achieves good separation of the wires which as explained above may be beneficial for cooling rates. However, the overall height is increased by the use of the vertical stack. Furthermore there are extra parts (i.e. more crimps) and additional assembly steps. Some of the crimps, e.g. the moveable crimps 610a, 610b, 610a', 610b', may be integrally formed with the chassis 600 to reduce the overall number of components. To assist in the assembly some of the crimps, e.g. the static crimps 610c, 610d, 610c', 610d' may be formed on separate tabs 602 which are attached, e.g. welded, to the chassis 600 during the assembly process.
Figure 6b shows a variation of the arrangement in Figure 6a with a single crimp for each SMA wire in two pairs of wires (618e, 618c), (618b, 618d). In this arrangement, the wires are still arranged in a vertical stack on either side of the chassis 600 but the upper and lower crimps are offset laterally as well as vertically and are thus not arranged in a vertical stack as in the previous arrangement. It will be appreciated that this is just one arrangement and the crimps could just be laterally offset or just vertically offset, not necessarily both as shown.
The crimps (620a, 620a'), (620c, 620c') from a first pair of coupling elements with a first coupling element (620c, 620c') coupling one end of the two wires to the static element and a second coupling element (620a, 620a') coupling the opposed end of the two wires to the moveable element. Similarly, crimps (620b, 620b'), (620d, 620d') from a second pair of coupling elements with a first coupling element (620d, 620d') coupling one end of the two wires to the static element and a second coupling element (620b, 620b') coupling the opposed end of the two wires to the moveable element.
A first upper wire 618a is connected at one end to a first upper moveable crimp 620a and at the other end to a first upper static crimp 620c which when the assembly is installed will be connected to the moveable portion. Similarly, a second upper wire 618c is connected at one end to a second upper moveable crimp 620b and at the other end to a second upper static crimp 620d. For the lower wires, the first lower wire 618b is connected at one end to a first lower moveable crimp 620a' and at the other end to a first lower static crimp 620c' and the second lower wire 618d is connected at one end to a second lower moveable crimp 620b' and at the other end to a second lower static crimp 620d'.
As shown, the upper wires are shorter than the lower wires because the upper crimps are located closer together than the corresponding lower crimps. It will be appreciated that if each of the upper crimps were offset laterally from the lower crimp in the same direction, a similar spacing between upper and lower crimps could be achieved. Accordingly, a similar offset arrangement can be used for wires of the same length. However, it may be useful to have different length wires in some designs.
As with Figure 6a, arranging the wires and the crimps in this way achieves good separation of the wires. Again, the overall height may be increased by separating the wires vertically but the offsetting laterally means that the crimps may be formed simultaneously. The trade off may be an overall increase in the actuator width. As before, some of the crimps, e.g. the moveable crimps 620a, 620b, 620a', 620b' may be integrally formed with the chassis 600 to reduce the overall number of components. To assist in the assembly some of the crimps, e.g. the static crimps 620c, 620d, 620c', 620d' may be formed on separate tabs 602 which are attached to the chassis 600 during the assembly process.
It will be appreciated that the choice of the number of wires in the Figures shown above is merely exemplary and additional or fewer wires may be used. Using additional wires, particularly if they are in individual crimps, is likely to increase the size of the assembly and thus there is a balance between the number of wires and overall size to consider when designing the assembly The arrangements having multiple crimps and multiple wires may have a similar performance to an arrangement having the same number of wires held in a
smaller number of wires in terms of force generated but may have increased cooling rates as suggested by Figure 5. However, the multiple crimp arrangements are likely to have increased costs due to the additional material, increased assembly steps if the crimps are formed sequentially and an increased size of the assembly. Space is typically constrained in devices incorporating such SMA actuators and thus an increase in size may not be desirable. However the design may be chosen to balance the desired cooling rate with the other factors.
Third haptic assembly
Figure 7 shows a ("third") haptic assembly which operates in a different way to the above-described haptic assemblies.
The third haptic assembly includes a moveable element 706 mounted above a static element 704. The adjacent faces of the moveable element and the static element have complementary shapes.
The third haptic assembly includes a plurality of wires 708 (only one of which is visible in the drawing). Each wire is attached at its ends to the static element 704. The ends of each wire are both attached using a static coupling element 712a in the form of a crimp connector, wherein the same crimp connector preferably holds two or more of the plurality of wires 708. Each wire 708 defines a plurality of wire sections 708a, 708b (in this case ten). A first plurality of the wire sections 708a are generally parallel to one another and a second plurality of the wire sections 708b are generally parallel to one another and set at e.g. right angles to the other wire sections 708a. Thus, the wire sections 708a, 708b may be considered to form V-shaped pairs of wire sections 708a, 708b.
Each of the wire sections 708a, 708b engage with the static element 704 and the moveable element 706. A first end of the first wire section 708a is attached to the static element 704 using the crimp connector 712a and the other end of the first wire section 708a engages with the moveable element 704 via e.g. a dowel pin 714. Similarly, a first end of the final wire section 708b engages with the moveable element 706 via a dowel pin 714 and the other end of the final wire section 708b is attached to the static element 704 using the crimp connector
712a. Each of the other wire sections 708a, 708b engage with the static element 704 via a dowel pin 712b (or other type of non-fixed connector, e.g. a hook) at one end and with the moveable element 706 via a dowel pin 714 at the other end.
The V-shapes are located within a channel having a zig-zag cross-section. It will be appreciated that if the wire sections are designed with different shapes, the channel may similarly be designed with a complementary matching shape.
The plurality of wires 708 may be spaced uniformly apart, may be parallel with one another and/or aligned in separate channels (or a single channel) between the static and moveable elements 704, 706.
As will be appreciated, when the plurality of wires 708 contract, the moveable element 706 moves in a direction (upwards) that is at e.g. 90° to the direction in which the wires 708 generally extend (horizontally).
Manufacture
Figure 8 is a schematic illustration of one method which may be used for crimping multiple wires in a single crimp in an SMA actuator described above. A plurality of wire spools 70, 72, 74 simultaneously feed wire over a guide wheel 76 and into the crimp 78. The guide wheel may help to maintain a spacing between wires in the crimp. Each wire may be made from any suitable shape memory alloy and may be coated (e.g. with polyimide or similar material) to reduce the risk shorting with other components in the SMA actuator if the wires are in contact with them. The number of wires is illustrated as three but it will be appreciated that this is merely exemplary and other numbers of wires can be used.
For optimal performance, the axis of each wire is preferably parallel to the axis of the crimp but due to imperfections in crimps, each wire is often seen to exit the crimp at an angle. A bend in the wire at the crimp point can increase fatigue. The guide wheel may also help to maintain each wire at a desired angle relative to the crimp. A small deviation from parallel may be acceptable. The
size of the deviation is dependent on the thickness of the wire but may be up to 8 degrees from parallel for a 25pm diameter wire. Moreover, the variation in angles between each of the wires may be within this range. The guide wheel may be omitted if the wires can each be fed into the crimp at an acceptable angle.
Once the wires are within the crimp, the crimp can be crimped (i.e. folded over or closed) to create the mechanical and electrical connection between the wires and the crimp. Any excess wire protruding from the crimp can then be trimmed. The crimping steps can be incorporated at any suitable point in the assembly process.
The same process can be used to connect the same wires to a second crimp (as in the haptic assemblies of Figures 6 and 7, for example).
While closing the crimp(s), the tension in each of the wires and/or the length of each of the plurality of SMA wires between the crimps is controlled so that each wire has substantially the same tension and/or length between the crimps in the resulting assembly. This may be achieved by arranging the plurality of SMA wires to each follow an equivalent, parallel path, as described above.
Generally, a more reliable haptic assembly can be manufactured if the wires do not cross (i.e. pass over each other). However, in some instances (e.g. in high volume manufacturing), it may not be practical to fully prevent crossing. In such instances, the inventors have found that satisfactory reliability can be achieved even if the wires cross in between the first and second crimps, provided that the wires do not cross inside the crimps.
As an alternative to crimping, the ends of the wires may be connected in place (e.g. direct to the static element or the moveable element or to an intermediate element) using welding (e.g. arc welding, welding using a weld bar, laser/heat-based welding) the ends of the wires in place. By welding the wires in place, it may be possible to more accurately control the spacing between the wires. During the welding process, care needs to be taken to control the welding so that damage to the wire, e.g. melting or loss of material, is minimised.
Those skilled in the art will appreciate that while the foregoing has described what is considered to be the best mode and where appropriate other modes of performing present techniques, the present techniques should not be limited to the specific configurations and methods disclosed in this description of the preferred embodiment. Those skilled in the art will recognise that present techniques have a broad range of applications, and that the embodiments may take a wide range of modifications without departing from any inventive concept as defined in the appended claims.