Title: INTELLIGENT BELT DRIVE SYSTEM AND METHOD CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit of U.S. Provisional Patent Application No. 61/531 ,021 , filed September 5, 2011 , U.S. Provisional Patent Application No. 61/584,314, filed January 9, 2012, U.S. Provisional Patent Application No. 61/588,076, filed January 18, 2012 and PCT application PCT/CA2011/001032, filed September 12, 2011 , all of which are incorporated herein by reference in their entirety. FIELD

[0002] The present disclosure relates to tensioners for tensioning engine driven elements such as timing belts, timing chains and accessory drive belts. In particular, the present invention is directed to belt tensioners that have the capability to actively adjust the belt tension in the belt.

BACKGROUND

[0003] Tensioners for timing belts, timing chains and accessory drive belts are well known. Some tensioners are adjustable in terms of the amount of tension is generated in the belt during operation of the engine. Such tensioners react to certain sensed conditions that indicate a potential for, or occurrence of, belt slip. While such systems may reduce the frequency of belt slip, they are, by design, inherently permitting some belt slip to occur. In those systems that sense the actual occurrence of belt slip, belt slip has already necessarily occurred before the tensioner will increase belt tension. In systems where the conditions for belt slip are sensed, the conditions that create the potential for belt slip must already occur for them to be sensed by the tensioner, which inherently means therefore that belt slip can occur before the tensioner has reacted to increase belt tension.

[0004] While these systems have some advantage in terms of reducing the frequency of belt slip, they still permit some belt slip to occur, which may be detrimental to the belt, to certain accessories driven by the belt, and to the perception of quality of a person hearing the chirp that can accompany belt slip.

[0005] Furthermore, the tension adjustment mechanisms employed by such tensioning systems may be expensive, and/or unreliable.

[0006] It would be beneficial to provide a tensioning system that has at least partially addresses one or more of these problems.

SUMMARY

[0007] In one aspect, a tensioner is provided that is capable of precognitively anticipating when it is desirable to increase the tension in a belt or other endless drive element so as to prevent belt slip prior to events that would raise the risk of it.

[0008] In another aspect, a system and method for controlling a tensioner that has adjustable tension, in such a way as to avoid the tendency of the tensioner to cycle between high and low tension settings.

[0009] In another aspect, the invention relates to the use of one or more of the parameters described herein for the purpose of preventing belt slip.

[0010] In another aspect the invention relates to the embodiments of tensioners shown and described herein.

[0011] In another aspect, a tensioner is provided that incorporates a hydraulic cylinder to adjust the tension in a belt or other endless drive member is provided. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] The foregoing and other aspects will be more readily appreciated having reference to the drawings, wherein:

[0013] Figures 1a-14b are views of tensioners that include systems for mechanically and electrically driving changes in the tension of an endless drive member;

[0014] Figures 15-43 are views of tensioners that include systems that incorporate at least a fluid actuated cylinder for driving changes in the tension of an endless drive member.

[0015] Some aspects of Figures 1 a-14b are relevant to the tensioners shown in Figures 15-43. For example some of the sensors shown in Figures 1a-14b can be incorporated on the tensioners shown in Figures 15-43. Some aspects of Figures 15-43 are relevant to the tensioners shown in Figures 1a- 14b.

DETAILED DESCRIPTION OF EMBODIMENTS

[0016] In this specification and in the claims, the use of the article "a", "an", or "the" in reference to an item is not intended to exclude the possibility of including a plurality of the item in some embodiments. It will be apparent to one skilled in the art in at least some instances in this specification and the attached claims that it would be possible to include a plurality of the item in at least some embodiments.

[0017] Reference is made to Figure 1a, which shows a crankshaft 10 from an engine 3 from a vehicle. The crankshaft 10 has a crankshaft pulley 12 thereon. The crankshaft pulley 12 drives one or more vehicle accessories via a belt 14. The accessories may include an alternator 16, an air conditioning compressor 18, a water pump (not shown in Figure 1a), a power steering pump (not shown) and/or any other suitable accessory. Each of the driven accessories has a shaft, and a pulley that is connectable and disconnectable from the shaft via a clutch. The alternator shaft, clutch and pulley are shown at 50, 52 and 54 respectively. The air conditioning compressor shaft, clutch and pulley are shown at 56, 58 and 60 respectively. This permits each of the accessories to be shut off when not needed even though the belt 14 itself is still being driven by the crankshaft 10.

[0018] Providing at least a certain amount of tension in the belt 14 is beneficial in that it reduces the amount of slip that can occur between the belt 14 and the driven accessories or even between the belt and the crankshaft 10. However, providing an unnecessarily high tension in the belt 14 has many detrimental effects. For example, it causes more power from the engine 13 to be consumed in driving the accessories, leaving less power for use in driving the vehicle. Thus, to achieve a particular level of performance from the vehicle, a greater amount of fuel would be consumed than would be consumed if the power loss were smaller. Additionally, a high belt tension generates greater hub loads on the pulleys for the driven accessories, which necessitates the use of relatively larger shafts on the accessories, larger bearings to support the shafts, heavier brackets to hold the accessories in place, all of which add to the weight of the vehicle and thereby negatively impact fuel economy for the vehicle.

[0019] A belt tensioner 20 is shown which keeps the belt 14 tensioned so as to reduce belt slip, and to keep the belt on the pulleys of the crankshaft 10 and accessories. The tensioner 20 includes a tensioner pulley 22 which engages the belt 14, a tensioner arm 24 which holds the tensioner pulley 22, a tensioner biasing member 26 for biasing the tensioner arm 24 towards a free arm stop position (the position the arm 24 would reach if it was not pressing into the belt 14), which would represent the position of reduced stored mechanical potential energy within the elastomeric biasing spring within the tensioner travel range, and a load stop position (the position the arm 24 would reach if the arm 24 were pushed all the way throughout its range of travel away from the free arm stop position), which would represent the position of increased stored mechanical potential energy within the elastomeric biasing spring within the tensioner travel range. The tensioner 20 further includes a tensioner actuator 28 which is operatively connected to the tensioner pulley 22 and tensioner arm 24 to move the pulley 22 and arm 24 between the free arm stop and load stop positions. The tensioner 20 may include structure that dampens its motion.

[0020] The belt tensioner 20 may have any suitable structure that provides the aforementioned capabilities. For example, the arm 24 may telescope linearly and may be biased outwards by a compression spring, which may be the tensioner biasing member 26. The first end of the spring 26 engages the arm 24, and a second end of the spring 24 engages a base 30. The base 30 is movable along the arm 24 by the tensioner actuator 28. For example, the tensioner actuator 28 may include an electric motor that turns a lead screw (not shown), whose rotation drives a traveler (also not shown) forwards or backwards. The base 30 is connected to the traveler and moves therewith. The lead screw may be hollow and the arm 24 may telescope outwards from it. Thus, the arm 24 is movable independently from the base 30 and from the lead screw. The tensioner actuator 28 is controllable to drive the base 30 forwards or backwards. Driving the base 30 forwards increases the amount of compression in the spring 26 thereby increasing the biasing force exerted by the spring 26 on the arm 24 and in turn on the pulley 22 in the free arm stop direction, thereby increasing the tensioning force exerted by they pulley on the belt 14. Driving the base 30 backwards decreases the amount of compression in the spring 26, thereby reducing the tensioning force exerted by the pulley 22 in the free arm stop direction on the belt 14. The biasing force of the spring 26 may be referred to as an actuator force since the actuator 28 controls the magnitude of this force. The tensioner actuator 28 may have any suitable motive means with which to drive the movement of the base 30 instead of an electric motor, such as, for example, pneumatic or hydraulic pressure, a shape memory metal actuator, or some other means. [0021] » In an alternative embodiment shown in Figure 1 b, the tensioner arm 24 is pivotable about a tensioner arm pivot axis. The tensioner biasing member may be a torsion spring which is engaged at a first end with the arm to bias the arm towards the free arm stop. The second end of the torsion spring may be engaged with the tensioner actuator 28 which moves the second end of the spring to increase or decrease the biasing force of the spring on the tensioner arm 24 and therefore the tensioning force of the pulley 22 on the belt 14.

[0022] Examples of suitable tensioner structures may be found in PCT publication number, WO2010/094127A1 , and in US Provisional patent applications 61/382,892, 61 /381 ,929, and 61/391 ,266, all of which are hereby incorporated by reference. Examples of suitable actuators that can be used to drive the second end of the torsion spring may be found in US Patent Nos. 3,954,016, US 4,131 ,306, US 4,674,781 , US 4,850,466, US 4,885,954, US 4,893,704, US 5,338,076, US 5,634,676, US 5,862,903 and US 5,983,739, all of which are hereby incorporated by reference.

[0023] In reference to the tensioner embodiment shown in Figure 1 b, the graph shown in Figure 1c illustrates the relationship between the length of the belt 14, the force applied by the spring 26 on the tensioner arm 24 (shown at 90), and the resulting torque applied by the tensioner arm 24 on the belt 14 (shown at 92). The nominal belt length is shown at 96, and with plus and minus tolerances at 98a and 98b. In general throughout this disclosure, when it is stated that the tensioner arm is driven into the belt 14, is engaged with the belt 14, applies torque to the belt 1 or the like, it will be understood that it is actually acting on the belt 14 through the pulley. As the belt stretches, the arm 24 rotates progressively further as a result of the spring force acting on it. However, the spring force progressively decreases as the arm 24 rotates during belt stretch, since the spring 26 is progressively unwinding towards its rest position. Also, the rotation of the arm 24 changes the geometric (i.e. angular) relationship of the arm 24 to the applied hubload of the pulley 22 on the belt 14. As a result, even though the spring force may be decreasing linearly with rotation of the arm 24, the torque applied to the belt 14 varies non- linearly (sinusoidally in a typical configuration) with rotation of the arm 24. The result of these changes is shown in the graph in Figure 1 d, which illustrates the relationship between the belt tension generated by the tensioner, as a function of belt length. In theory it would be advantageous to be able to apply a certain target tension to the belt through the tensioner, which was constant regardless of belt length, and hence regardless of the angular position of the tensioner arm 24. This target tension is shown at 100 in Figure 1 d. However, when using a torsion spring such as spring 26, and a rotating tensioner arm 24, the curve shown at 102 is what results. A line indicating the nominal length of the belt 14 is shown at 96. The tensioner 20 is configured so that its response curve 102 to changes in belt length, even though non-linear, somewhat match up closely with the 'target' tension line 100.

[0024] In a typical tensioner of the prior art, the response curve applied by the tensioner is somewhat fixed in the sense that the tension applied by the tensioner on the belt 14 changes based on belt length only (ignoring dynamic issues relating to belt flutter and the like, which can temporarily affect the response of the tensioner). As a result, such tensioners are typically configured to apply a relatively high target tension to the belt because they must ensure that the belt has sufficient tension in it in the worst case scenarios (e.g. under harsh operating conditions where the belt is wet, where the load on the belt is high due to operation of all of the accessories driven by the belt, due to high rates of change of the engine rpm, and the like). However, in reality these worst-case scenarios are not the norm, and under most operating conditions, the belt tension that is applied by these prior art tensioners is needlessly high. This constant state of high tension in the belt unfortunately results in a needless consumption of power, overdesign of bearings and the like for pulleys that are engaged with the belt, potentially shortened operating life of the belt, increased emissions from the vehicle, and other disadvantages. [0025] The tensioner 20 is capable, however, of adjusting the tension in the belt 14 so that when the need arises for high tension, the tensioner 20 is capable of increasing the tensioning force applied to the belt 14, thereby increasing the tension in the belt 14. This can result, in some embodiments in a shift upwards of the tension response curve shown in Figure 1 d, from the curve 102, to the dashed line curve shown at 106. The curve 106 need not match the shape of the curve 102. Moreover the effective spring constant applicable to the tensioner 20 when in a 'high tension' mode as represented by the curve 106 is not critical. Thus, the shape of the curve 106 is not particularly critical. It is important mainly that the tension applied to the belt 14 is sufficiently high to prevent belt slip under the particular conditions that triggered the tensioner 20 to raise the tension, while ensuring that the tension applied to the belt 14 is not so high as to damage the belt 14 or the components engaged with the belt 14 (e.g. the pulleys driving and driven by the belt 14).

[0026] A primary purpose of providing sufficient tension in the belt 14 is to prevent belt slip, which is a condition in which the linear speed of the belt on a pulley is different than the linear speed of the periphery of the pulley itself (that is where the slipping occurs). This slipping is problematic for many reasons. One reason is that the slipping is inefficient, in the sense that the movement of the belt is not resulting in as much rotation as would be desired of the accessory driven by the pulley. Another reason is that the slipping can be audible in some circumstances, which can result in the impression on the part of the vehicle owner or passersby that there is a quality issue with the vehicle.

[0027] It is beneficial for the tensioner 20 to be able to determine if belt slip is taking place. To do this, the tensioner 20 may determine the linear speed of the belt 14, and the linear speed of one or more of the pulleys from accessories that are driven by the belt 14. The linear speed of the belt 14 may be determined by determining the linear speed of a pulley that is driven by the belt, but that relatively little rotational inertia or more broadly, resistance to rotation, associated with it. An example of such a pulley is the pulley 22 on the tensioner 20. Such a pulley typically has no belt slip associated with it and therefore provides a relatively accurate value for the speed of the belt 14 during operation of the belt tensioning system. To provide a value for the speed of rotation of the pulley 22 any suitable means may be used. For example, a sensor trigger, such as a magnet or a metal target, shown at 70, may be placed in the pulley as shown in Figure 2a, and a suitable sensor 72, such as a Hall effect sensor may be positioned in the tensioner arm 24. A plurality of the magnets or metal targets 70 may be spaced equally from each other about a circle at some radius on the pulley 22. Alternatively, as shown in Figure 2b, a plurality of sensor triggers, such as teeth, protrusions, or conversely, recesses, could be incorporated into the pulley 22 and a suitable sensor 72, such as a Hall effect sensor, could be mounted to a stationary element, such as a base mounting to sense the passage of the sensor triggers 70.

[0028] The pulley 22 may be machined from a suitable metal, or spun from metal sheet into a flanged configuration as shown. One or more teeth, recesses, protrusions, targets or other sensor triggers 70 may be machined into whichever edge of the pulley passes overtop of the sensor 72. As an example, the pulley may be machined from magnetic steel.

[0029] The sensor 72 is positioned to be able to detect the passing of the sensor triggers. For balance purposes, should the pulley be required to spin at very high RPM, it may be beneficial to have at least two sensor triggers 70 that are equally spaced apart on the pulley flange so as to balance the inertia of the targets when the pulley 22 spins at high speed.

[0030] As more sensor targets 70 are provided at equal spacings on the pulley edge the balance of the pulley 22 improves. Thus for applications where the pulley's RPM is high, it is more beneficial to configure the pulley 22 with an increased number of sensor triggers 70 to improve its dynamic balance. As the number of teeth (or more generally, sensor triggers 70) increases, the accuracy of the determined RPM increases in addition to the improvement in balance. As shown in Figure 2b the pulley 22 may have any suitable number of sensor triggers 70, such as, 10 sensor triggers.

[0031 ] The sensor 72 may include a processor which counts the passage the sensor triggers 70 overhead, and thereby determine the instantaneous RPM of the pulley 22. Once the RPM of the pulley 22 is known, the belt speed can be determined with high accuracy.

[0032] The pulley 22 may be used for determining the belt speed as described above, provided that there is sufficient belt wrap on the pulley 22. Additionally, in some circumstances there may not be sufficient room to install the sensor 72 or the triggers 70 on the tensioner arm 24 or pulley 22. If any of these conditions exist such that the belt speed cannot be determined using the tensioner 20, a separate idler pulley can be installed in a separate location in engagement with the belt 14 wherein a sufficient amount of belt wrap is provided.

[0033] An engine controller shown at 32 is provided in the vehicle and controls the operation of the engine 13 and the accessories. The engine controller 32 may be referred to as the vehicle control unit 32. The vehicle control unit 32 may receive sensor data from several sources, including signals from a speed sensor on the crankshaft 10 that indicates the speed of rotation of the crankshaft 10. The tensioning system, which includes the tensioner 20, may further include a tensioning system controller shown at 34. The tensioning system controller 34 controls the operation of the tensioner actuator 28. The tensioning system controller 34 additionally may receive signals from one or more other components indicative of the states of those components. For example, the tensioning system controller 34 may receive signals indicative of the crankshaft speed, the alternator voltage output, the alternator speed, the alternator temperature, the position of the tensioner biasing member 24, and the speed of the tensioner pulley 22 (as described above, using one or more sensor triggers 70 and sensor 72). The controller 34 can determine if any slip is occurring at any of the pulleys for the crankshaft 10, the alternator 16 and the compressor 18 based on any detected differences between their actual measured speed and the speed they should be rotating at based on their pulley diameter as compared to that of the tensioner pulley 22. If any slip is detected, or if the slip detected on any pulley exceeds an upper threshold value, such as about 3% to about 4%, the controller 34 may increase the actuator force so as to increase belt tension, so as to reduce or completely eliminate the slip. The increase in the actuator force may be by a discrete amount such as 25N after which the slip is rechecked to determine if it is below the upper threshold value. Alternatively, the increase in the actuator force may be continuous with continuous determination of the slip being made by the controller 34. In either case the controller 34 may stop increasing the tension once it determines that the slip is below the threshold value for all pulleys. Conversely, if the slip detected on all the pulleys is below a lower threshold value (which may be, for example, about 2%), then the controller may decrease the actuator force so as to reduce belt tension in order to provide improved fuel economy and reduce power losses. The decrease may be continuous, or it may be in discrete amounts.

[0034] The tensioning system controller 34 may also receive signals from the vehicle control unit 32 indicating what action the vehicle control unit 32 is about to take. For example, the vehicle control unit 32 may indicate to the tensioning system controller 34 that it is about to turn on the air conditioning compressor 18. As a result, the tensioning system controller 34 may be capable of using the incoming signals to determine that belt slip is about to occur, or that the load on the belt 14 is about to increase. In case of one of these determinations the tensioning system controller 34 may pre-emptively increase the actuator force on the tensioner pulley 22 so as to increase the belt tension prior to these events occurring. As a result, the belt slip is reduced (or even prevented).

[0035] In another scenario, where the vehicle is off, as the driver of the vehicle approaches the vehicle, the tensioning system controller 34 may sense the presence of the key fob, similar to the systems that sense the presence of the key fob for the purpose of unlocking the vehicle's doors in some luxury cars today. Instead of the controller 34 directly sensing the presence of the key fob, the vehicle may communicate to the controller 34 that some other system (such as the system responsible for unlocking the vehicle doors when sensing the presence of the key fob in vehicle so equipped). Alternatively, the tensioning system controller 34 may detect that the key fob was used to unlock the driver's door (by a user pressing an 'unlock' button on the key fob). As a result, the tensioning system controller 34 may determine that the vehicle's engine 13 may be started imminently. Engine startup introduces a relatively high load on the belt 14, and can cause belt slip in a belt that is under too little tension. In order to reduce the likelihood of belt slip, the tensioning system controller 34 may increase the actuator force on the tensioner pulley 22 prior to engine startup so as to prevent belt slip from occurring.

[0036] Instead of sensing the key fob, a less expensive option may be for the controller 34 to be programmed to increase the belt tension at certain times of the day on certain days. For example, the controller 34 may be equipped with a clock or may be programmed to determine the time and date from some other subsystem in the vehicle such as a GPS receiver, and may be programmed to detect patterns in terms of the times of day that the vehicle is driven. For example, the controller 34 may determine that, over the previous 30 days (or over any other suitable number of days) the vehicle has been started up on weekdays (i.e. Monday to Friday) at between 7:45am and 7:55am over 90% of the time and never earlier than 7:37am. The controller 34 may be further programmed to use this data to determine a time of day to increase the belt tension in anticipation that the engine will shortly be started. For example, the controller 34 may increase the belt tension at 7:43am every weekday, or, for example, at 7:35am every weekday. After increasing the tension in the belt 14, the controller 34 may enter, or return to, a sleep mode until it is awakened by some input, such as by the detection of the vehicle being started. In addition, time of day and date information can be used to predict the need for high tension in the belt 14 in other ways. For example, in embodiments where a humidity sensor or a moisture sensor are not present, the controller 34 may determine that, based on the time of day (e.g. early morning) and the date (e.g. early fall), there is an increased likelihood of belt slippage from dew on the belt 14. As a result, the controller 34 may be programmed to keep the belt 14 under high tension for a selected period of time after engine start up to allow the engine to warm up and the dew to evaporate. Conversely, in the mid- afternoon on a summer day, the controller 34 may be programmed to keep the belt 14 under high tension for a shorter selected period of time after engine start up due to an expectation that there is no dew on the belt 14.

[0037] Another example is where the controller 34 receives data from an ABS system and/or a traction control system to determine whether there is repeated pending wheel lockup or pending wheel slip during a particular use of the vehicle. This data, optionally in combination with temperature data and/or time and date data, can be used to infer that there is snow on the ground, or, depending on the sensed temperature, to infer that there may be slush or water on the ground. If snow, slush or water is inferred to be on the ground, then the controller 34 may determine that the belt tension should be increased due to the potential for the belt 14 to become wet from kickup or some other reason during driving, for example. If the temperature is below a selected level, the controller 34 may determine that the belt 14 is likely to be relatively stiff, and possibly loose. Also, at low temperature, grease or other lubricants used to facilitate the rotation of bearings and the like tends to 'stiffen' thereby increasing the resistance to turning of any rotating components. As a result, below a certain temperature, the controller 34 may therefore increase belt tension.

[0038] In another example, the tensioning system controller 34 may receive a signal that the humidity is high from a sensor such as a humidity sensor. This can be an indication that it is raining, or that the vehicle is in some sort of situation where the belt 14 is potentially wet. When the belt 14 is wet, it will be understood that there is an increased likelihood of belt slip at any given belt tension, as compared to a dry belt 14. When the tensioning system controller 34 determines that the humidity is high, the tensioning system controller 34 may automatically increase the actuator force on the pulley 34 so as to reduce the potential for belt slip. In a scenario where the tensioning system controller 34 has determined that a particular level belt tension is suitable for a particular situation, the tensioning system controller 34 may augment the belt tension to a higher level in a situation where it detects high humidity. An example of a suitable humidity sensor is an HIH series humidity sensor provided by Honeywell International Inc. Instead of, or in addition to, a humidity sensor, the tensioning system may include some other sensing means to detect a wet belt 14 or the potential for a wet belt 14, to trigger it to increase the belt tension. For example, a temperature sensor positioned to detect the temperature in the engine bay may be used, in conjunction with an ambient temperature sensor and a humidity sensor, to detect the potential for a wet belt in certain situations. A humidity sensor that is configured to determine relative humidity in the engine bay, would include a temperature sensor that would be positioned to detect the temperature in the engine bay. For example, when a vehicle is kept in a cool location such as in a garage or an underground parking lot, and is then started up and taken out on a hot day, there is the potential for water vapour in the air to condense on many components of the relatively cool vehicle including the accessory drive belt and the pulleys it engages, which can lead to belt slip or belt chirp. To handle this, the tensioning system controller 34 may be programmed to sense when the temperature in the engine bay is lower than the ambient temperature by more than a selected amount while the humidity is above a selected level, and if so, to increase the belt tension to a selected high tension setting. After the engine bay has warmed up (e.g. by operation of the engine) the controller 34 may determine that there is no longer the potential for belt slip or belt chirp and may accordingly reduce the belt tension. [0039] In order to determine whether it is raining, the controller 34 may receive signals from both the humidity sensor, and from the windshield wiper system. If the humidity is high and the windshield wipers are on, then the controller 34 may determine that it is raining. If the humidity is high but the windshield wipers are off, then the controller 34 may determine that it is not raining. If the humidity is not high but the windshield wipers are on, the controller may determine that it is not raining. As an example, the windshield wipers may have been put on for some other reason, such as to clean the windshield of dirt, snow or frost for example.

[0040] In yet another example, the tensioning system controller 34 may receive signals from the throttle sensor or from the vehicle control unit 32 that the driver is driving aggressively, and is repeatedly accelerating rapidly and/or is repeatedly letting off the throttle rapidly. In a less sophisticated embodiment, each time the driver accelerates (particularly if they accelerate hard), the tensioning system controller 34 will drive the actuator force up to increase the belt tension to prevent slip, and each time the driver suddenly releases the throttle, the tensioning system controller 34 will drive the actuator force down to decrease the belt tension since high tension may be considered not necessary. It will be understood that driving the actuator 28 consumes energy. Repeatedly driving the actuator force up and down can in some situations consume as much or more power than is saved by the overall reduction in average belt tension, and can shorten the life of the tensioner actuator 28 and other components. In a more sophisticated embodiment, after a sufficient number of repetitions of these actions, the tensioning system controller 34 may determine that this driving behaviour is likely to continue and as a result, the tensioning system controller 34 may simply leave the tensioning force at a high setting, so as to prevent hunting of the actuator 28 to reduce and increase the tensioning force in an attempt to achieve a constantly moving or oscillating target. Once the tensioning system controller 34 detects that the driver's driving behavior has changed and he/she is no longer driving as aggressively for a sufficient period of time, the tensioning system controller 34 may decide to permit the reduction of the belt tension. There are several ways of determining whether the driver of the vehicle is driving aggressively. One way is to receive signals relating to the engine rpm. If the controller 34 determines that the engine rpm cycles between a level that is above a first, 'high' level and a level that is below a second, 'low' level then the controller may determine that the vehicle is being driven aggressively. A parameter that may also be used by the controller 34 for this determination is the cycle time. In other words, if the engine rpm reaches into the 'high' range (i.e. above the 'high' level) and then falls into the 'low' range (i.e. below the 'low' level) sufficiently slowly, the controller 34 may determine that the engine is not 'cycling' and that this does not constitute sufficiently aggressive driving to warrant leaving the tensioning force in a high tension setting. However, if the engine rpm cycles between the 'high' range and the 'low' range in relatively short periods of time, then the controller 34 may determine that the engine is cycling and that this does constitute a situation where leaving the tensioner in a high tension setting is warranted. It will be noted that even if the controller 34 uses signals from the throttle sensor, the controller 34 can still be said to be basing the determination of whether the engine is cycling on engine rpm since the signals from the throttle sensor can have a rough correlation to engine rpm.

[0041] Another way of using the engine rpm is for the controller 34 to use the rate of change of the engine rpm to determine whether or not to leave the tensioner at a high tension setting. If the rate of change of the engine rpm repeatedly exceeds a particular 'high' level then the controller 34 may determine that the engine is cycling and may leave the tensioner at a high tension setting regardless of the actual engine rpm reached.

[0042] The controller 34 may be programmed to store events where it has determined that the vehicle is being driven aggressively in a database and may additionally store other information related to the event, such as, for example, the time of day, or some type of identification information that identifies which driver is driving the vehicle. For example, upon entry into the vehicle, the driver may have pushed one of the seat position memory buttons. This seat position button identifies the driver of the vehicle as Driver #2. The controller 34 may store the driver identification in memory. In the event that the controller 34 determines that a cycling event occurred (or in general that an aggressive driving event occurred), the controller 34 may store that in its memory along with the driver identification information. If the controller 34 determines that this particular driver has too many instances of aggressive driving, the controller 34 may make changes to the criteria that would trigger the controller 34 to put the tensioner in a high tension setting whenever it senses that that driver is driving the vehicle. Another way that the vehicle may identify the driver is by the key fob. In some luxury vehicles today, the vehicle's controller is capable of distinguishing one key fob from another. The vehicle's controller uses this information to adjust the seats, mirrors and the like to each driver's preferences based on which key fob is sensed. Thus the controller 34 may be able to determine which driver is driving the vehicle based on which key fob is being sensed by the vehicle's sensors.

[0043] Further with respect to reducing the tendency of the tensioner actuator 28 to hunt continuously, the tensioning system controller 34 may be programmed to may raise or drop the belt tension by discrete amounts as opposed to continuously adjusting it to achieve the optimal belt tension for dynamic (i.e. continuously changing) conditions. The tensioning system controller 34 may be programmed for this to occur always, or alternatively it may be programmed for this to occur only under certain conditions, while permitting hunting (i.e. continuous adjustment of the belt tension) under other conditions. An example of the use of a discrete change in belt tension is in a situation where the tensioning system controller 34 detects a constantly increasing required belt tension. In such a scenario the tensioning system controller 34 may decide to increase the belt tension to a selected high setting and to leave it there for a period of time, (unless it becomes necessary to increase it further), instead of continuously adjusting the belt tension to just match the necessary tension for each instant of time.

[0044] In another scenario, the tensioning system controller 34 may detect a situation where the engine speed is low (i.e. below a selected threshold). Resonance can occur in the belt 14 depending on certain factors, such as the engine speed, the load on the belt 14 and the belt tension. Resonance can cause the belt 14 to flutter and can in some situations cause damage to or failure of the belt 14. In particular, some belts that are proposed for use or are in use in vehicles today are relatively thinner than some belts previously used, which makes them even more prone to flutter. To inhibit this from occurring the tensioning system controller 34 may be programmed to detect situations where the engine speed and belt load would lead to an increased risk of resonance at a given belt tension, and to increase the belt tension so as to reduce the risk of belt flutter.

[0045] As shown in the examples described above, the tensioning system controller 34 may thus be capable of predicting situations in which belt slip may be imminent, or in which increased belt load may be imminent and may be able to control the belt tension before these events take place.

[0046] As described above, the tensioning system controller 34 can control the tensioning force by the pulley 22 on the belt 14 so as to reduce the belt tension where possible and to increase the belt tension when needed.

This provides an overall improvement in fuel economy for the vehicle, as compared to a tensioner that is not controllable. However, this is, in some ways, a passive approach to reducing belt tension to improve fuel economy (which may be referred to as fuel efficiency). In some embodiments, the tensioning system controller 34 may take an active approach to reducing belt tension to improve fuel economy.

[0047] For example, the tensioning system controller 34 may be operatively connected to one or more of the alternator 16, the air conditioning compressor 18 and any other accessories driven by the belt 14. It will be understood that each of these accessories represents a load on the belt 14, and each requires a certain amount of belt tension to operate without belt slip. By having two of these loads operate at the same time, as can commonly occur, the belt tension must be increased further in order to reduce the likelihood of slip. In an effort to reduce the maximum belt tension required to be applied by the tensioner, the tensioning system controller 34 may in some situations cause a reduction in the load associated with one of the accessories, when permitting another of the accessories to operate. For example, the tensioning system controller 34 may determine that the alternator 16 need not operate at its highest voltage (which increases the amount of magnetic drag is associated with the alternator 16) when the air conditioning compressor 18 is on. Thus, the tensioning system controller 34 may reduce the voltage of the alternator 16 at those times if possible, and may possibly shut off the alternator altogether. Similarly, the tensioning system controller 34 may reduce the refrigerant flow to the air conditioning compressor 18 (or may shut if off entirely) when it determines that the alternator 16 needs to run at high voltage. It will be noted that some alternator types (e.g. some modern high efficiency alternators) have a relatively high amount of drag associated with them in cold weather. This is due, at least in part, in some circumstances to the close tolerances that exist between the rotating and stationary components of such alternators. In cold weather, differential amounts of shrinkage of some components relative to each other can increase the amount of rotational friction that exists until the alternator warms up. Furthermore, when the vehicle sits idle (i.e. when the vehicle is off) for any period of time (even overnight) during cold weather, the vehicle's battery can lose charge relatively quickly, thereby causing the alternator 16 to operate at high load at vehicle startup as it attempts to charge the battery. As such, the tensioning system controller 34 may prevent the alternator 16 from operating during startup of the engine 13 in cold weather since engine startup already introduces a high belt load which implies a raised belt tension. Once the engine 13 has started up and the belt tension necessary to prevent slip is reduced, the tensioning system controller 34 may permit the alternator 6 to start up. In some embodiments, the controller 34 may prevent alternator 16 from operating until the temperature of the alternator 16 reaches above a selected temperature. A temperature sensor that is directly responsible for sensing the temperature of the alternator 16 may be provided for informing the controller 34 of the alternator temperature. If it is determined that the alternator 16 is needed to operate but is below a selected temperature, the controller 34 may put the tensioner at a high tension setting prior to start up of the alternator. The selected temperature below which the alternator 16 may be considered to be cold may be for example, less than or equal to about -20 degrees C. Optionally, this condition (the low alternator temperature) may override other logic used by the controller 34 to modify belt tension. In other words, the controller 34 may be programmed to generate a high belt tension whenever it detects that the alternator temperature is below the threshold value (assuming that alternator operation itself is needed), and to hold the high belt tension until the alternator temperature rises above a second threshold value, such as, for example, +20 degrees C.

[0048] With respect to the alternator 16, it is beneficial for the controller 34 to be able to estimate the torque needed to drive the alternator 16 in order to determine whether to change the belt tension. The alternator torque depends on several factors, including voltage, current and speed. These values can be mapped to estimate torque quickly and easily without significant computational requirements. The controller 34 can use this estimate to determine what belt tension is suitable.

[0049] It will be noted that the operative connection between the tensioning system controller 34 and the accessories may not be a direct one. For example, the tensioning system controller 34 may send instructions to the vehicle control unit 32 to stop or prevent operation of the alternator 16, and the vehicle control unit 32 may determine whether this is possible. For example, the vehicle control unit 32 may determine that it is not possible due to a critically low state of charge of the vehicle battery (not shown). In such an instance, the vehicle control unit 32 may send a signal back to the tensioning system controller 34 that it cannot stop, slow down or prevent operation of the alternator 16, in which case the tensioning system controller 34 may drive up the tensioning force accordingly. Overall, however, by providing some control over the accessories, the tensioning system controller 34 may be able to reduce the number of times the tensioning force needs to be driven up to high levels. The number of times that the tensioning force reaches high levels directly impacts the size of the shafts, bearings and brackets required for the various components associated with the accessories, so that they are sufficiently robust to resist deformation and fatigue. Reducing the number of times that the tensioning force reaches high levels can therefore result in the use of relatively smaller shafts, smaller and/or lighter-duty bearings, and lighter brackets. This reduction in weight of the components in turn results in an increase in fuel efficiency for the vehicle, and reduces rotational drag associated with the accessories which can result in a further increase in fuel efficiency for the vehicle.

[0050] Instead of sending instructions to the vehicle control unit 32, it is alternatively possible for the tensioning system controller 34 to directly send commands to one or more of the accessories. However, the vehicle control unit 32 may also be directly or indirectly operatively connected to the accessories, and it may issue overriding commands to the tensioning system controller 34 under certain conditions which prevent the tensioning system controller 34 from interfering with the vehicle control unit's operation of the accessories.

[0051] The tensioning system controller 34 may receive signals from one or more components that are indicative of the current belt tension so that the tensioning system controller 34 can determine if the current belt tension is suitable or not for the current set of conditions. The signals may be from a position sensor that indicates the position of the second end of the tensioner spring, which gives an indication of the tensioning force exerted by the pulley 22 on the belt 14. The position sensor could be a simple Hall-effect sensor which would send a signal proportionate to the distance between the sensor and the end of the spring. A suitable Hall-effect sensor could be a Honeywell SS 520 sensor, supplied by Honeywell International, whose headquarters are in Morristown, New Jersey, USA. Alternatively other more precise position sensing devices could be used, albeit at greater cost. Alternatively, the belt tensioning system may include other, more sophisticated devices, such as, for example, a strain gauge on one or more accessory shafts along with associated electronics for conditioning and signal amplification. Such devices are typically relatively high cost however. Such a device is described in US Patent No. 6,216,547 which is hereby incorporated by reference.

[0052] While the tensioning system controller 34 is shown as a separate device from the vehicle control unit 32, it is possible to provide the equivalent of the tensioning system controller 34 within the vehicle control unit 32 itself. In other words the tensioning system controller 34 could be a program module that resides in memory along with the program module that makes up the vehicle control unit 32. For the purposes of the claims provided below, however, the tensioning system controller 34 and the vehicle control unit 32 may nonetheless be considered to be separate elements regardless of whether they share hardware or even some software elements. The term 'controller' in reference to control system 34 is intended to be interpreted broadly so as to cover embodiments wherein there is a single control unit, and embodiments wherein there are multiple control units that control the operation of the tensioner 20. For the purposes of this disclosure the term 'controller' is to be considered synonymous with the term 'control system'.

[0053] In a particular embodiment shown in Figures 3a and 3b, the vehicle may be equipped with technology that shuts off the engine 13 temporarily in certain situations to reduce emissions and increase fuel efficiency. For example the vehicle control unit 32 may shut off the engine 13 hen the vehicle stops at a stoplight. When this occurs it may still be desirable to continue operation of one or more of the belt driven accessories, such as the air conditioning compressor, for example. To achieve this, the crankshaft 10 may be equipped with a crankshaft clutch 38 through which it connects to the crankshaft pulley 12. By disengaging the clutch 38 the belt 14 can now be driven without the engine 13 turning. The alternator 16 may in such an embodiment be an GU (motor/generator unit) which can operate as a motor, drawing power from an electrical source such as the vehicle battery (not shown). The MGU 16 can then drive the belt and the other belt-driven accessories. Such a system is described in WO2008/113186A1 , which is hereby incorporated by reference. It will be noted that in Figures 3a and 3b, only the components directly engaged with the belt 14 are shown, such as the crankshaft 10 and related components, the accessories and related components and the tensioner 20. Also, a water pump is shown at 55 in Figures 3a and 3b as one of the accessories. The difference between the embodiments shown in Figure 3a and the embodiment shown in Figure 3b is that the tensioner 20 in Figure 3a is a linear tensioner (as shown in Figure 1 a), and the tensioner 20 shown in Figure 3b is a rotary tensioner (as shown in Figure 1 b).

[0054] In such an embodiment, the tensioning system controller 34 may receive a signal from the vehicle control unit 32 indicating that the engine 13 is being shut off and may be programmed to reduce the belt tension to a low level so as to reduce the belt load, thereby increasing the amount of time the battery can support rotation of the MGU 16. As a further step to reducing the belt load the tensioning system controller 34 may also reduce the refrigerant flow to the air conditioning compressor 18 to a relatively low level so as to reduce the belt load associated with the compressor 18.

[0055] Alternatively, if flow reduction is not possible, the tensioning system controller 34 may attempt to mimic the reduction or the throttling of the flow of fluid by intermittently cycling the air conditioning compressor clutch on and off in a controlled fashion via a PWM pulse width modulation control strategy, or PPM, whereby the cycle time off and the cycle time on may be varied to optimize both the power savings potential as well as the cooling potential within the passenger compartment.

[0056] In such an embodiment, the MGU 16 may also be used to start the engine 13 (referred to sometimes as a BAS (Belt Alternator Start) system, or a Belt-driven Starter Generator). To carry this out, the crankshaft clutch is engaged so that the crankshaft pulley 12 and the crankshaft 0 rotate together, and the MGU 16 drives the belt 14 in order to drive the crankshaft 10, thereby replacing a traditional starter motor. However, it will be noted that the crankshaft pulley 12 is positioned on a first side (shown at 40) of the MGU pulley 54. When the MGU pulley 54 drives the belt 14, the first side of the MGU pulley 54 is the slack side. Thus the belt tension on the first side 40 is relatively lower than the belt tension on the second side (shown at 44) of the MGU pulley 54. The lower belt tension introduces some potential for slip to occur between the belt 14 and the crankshaft pulley 12. It is possible to add a tensioner for the belt span between the MGU pulley 54 and the crankshaft pulley 12, however this is an expensive solution. In order to reduce the risk of slip when only one tensioner is provided (i.e. tensioner 20) the tensioning system controller 34 may use the tensioner 20, which, as can be seen is positioned between the crankshaft pulley 12 and the air conditioning compressor pulley 60, to increase the belt tension to a very high level so as to drive up the tension in the belt span (or belt spans) between the crankshaft pulley 12 and the MGU pulley 54 so as to reduce the likelihood of slip during starting of the engine 13 using the MGU 16. To assist in distributing the tension applied by the tensioner 20 throughout the belt 14 more evenly, one or more of the accessories may be shut off by the tensioning system controller 34 when the MGU 16 is used to start the engine 13. Once the engine 13 has been started, the MGU 16 may revert back to an alternator, the belt tension may be reduced and whatever accessories were shut off may be started up again. Increasing the belt tension has been described as being applicable during startup of the engine 13 by the MGU 16, while the tension has been described as being reduced by the controller 34 when the MGU 16 is driving the accessories. However, in a case where an accessory that is on the slack side of the MGU pulley 54 (e.g. the water pump 55) is determined to be necessary to operate without slip, the tension in the belt 14 may be increased by the controller 34 to facilitate it.

[0057] While the MGU 16 has been described as driving the belt 14 to drive the other accessories and/or to start the engine 13, it is alternatively possible for alternator 16 to be an alternator only, and for a separate electric motor to be provided for driving the accessories and/or to start the engine 13 via the belt 14.

[0058] Many of the scenarios described above relate to the controller 34 detecting a situation where increased belt tension may be required to prevent slippage from occurring, it may be possible for the controller 34 to detect upcoming situations where low belt tension may be permitted. For example, after detecting that the belt 14 potentially had dew on it and increasing the belt tension accordingly to prevent slip, the controller 34 may set a fixed period of time for the increased belt tension to occur, and, in the absence of other reasons to keep the belt tension up, the controller 34 may be programmed to reduce the belt tension automatically at the end of the fixed period on the assumption that the belt 14 is dry at that point.

[0059] Some systems of the prior art may adjust belt tension based on a determination that the current conditions are conducive to belt slip. However, this inherently means that there is a period of time during which the conditions exist and where belt slippage could occur prior to the system increasing belt tension. By detecting the potential for such situations before they occur (i.e. precognitively), the controller 34 is able to increase the belt tension before the need for increased belt tension arises, thereby avoiding a situation where slippage has already occurred before the tensioning system has had a chance to react.

[0060] Many parameters have been described herein as being useful for use by the controller 34 to determine the appropriate tension setting for the tensioner. All of these parameters may be combined into a map that may be stored in the controller memory. The controller 34 would then determine all the necessary inputs based on signals from sensors, from the vehicle ECU, and/or from other sources, and may then use those inputs with the map to determine a tension to apply to the belt 14, and/or a tensioner mode for the controller 34. An example of a tensioner mode would be a situation where the controller 34 has determined that the tensioner is hunting and then holds the tensioner at a particular tension setting for a set period of time to prevent it from hunting. Another example of a tensioner mode would be a situation such as during startup of the vehicle, whereby the tensioner may simply leave the tensioner at a high tension setting for a selected period of time, regardless of what signals are coming from the various sources.

[0061] Reference is made to Figures 4a-4c and 5a-5c, which show a tensioner 200 with an example of a tension adjustment system in accordance with another embodiment of the present invention.

[0062] The tensioner 200 includes a hub 221 that mounts to the engine 13, a tensioner arm 224 that is rotatably supported by an arm bushing 225 on the hub 221 for rotation about a hub axis between a free arm stop position and a load stop position, and which holds a tensioner pulley 222 for rotation about a pulley axis Ap, a tensioner biasing member 226 (which may be referred to as a spring 226 or a torsion spring 226, but which could be any suitable type of biasing member) for biasing the tensioner arm 224 towards the free arm stop position, and a tensioner actuator 228 which is operatively connected to the tensioner pulley 222 and tensioner arm 224 to change the tensioning force applied by the tensioner arm 224 on the belt 14 through the pulley 222. [0063] In the embodiment shown in Figures 4a-4c and 5a-5c, the tensioner actuator 228 is made up of a motor 230, in this instance with an integral gearbox, a worm 232, a sector gear 234 driven by the worm 232, and a spindle 236. The worm 232 is supported on bushings 238 in a bracket 240 that is part of the hub 221 , and is driven by the motor 230, which is controlled by the controller 34. The worm 232 causes rotation of the sector gear 234, which is connected to the spindle 236 via projections 242. The spindle 236 includes a driver slot 244 in it through which a spring tang 246 formed at the first end 248 of the spring 226. The second end of the spring 226 is shown at 250 and engages the tensioner arm 224. Rotation of the spindle 236 therefore causes rotation of the first end 248 of the spring 226. In a low tension situation, the tensioner 200 may engage the belt 14 with the arm 224 positioned as shown in Figure 5a. Rotation of the first end 248 in a first direction 252 causes the biasing force of the spring 226 on the tensioner arm 224 to increase progressively, thereby increasing the tensioning force of the tensioner arm 224, or more specifically the pulley 222, on the belt 14. Increasing the tensioning force causes rotation of the arm 224 in the clockwise direction in the views shown in Figure 5a-5c, until the tension in the belt 14 increases sufficiently that an equilibrium is reached between the tensioner arm 224 and the belt 14. Figures 5b and 5c show the tensioner 200 when the tensioning force has been increased from the low tension setting in Figure 5a, to midlevel tension setting (Figure 5b), and to a high tension setting (Figure 5c) respectively. It will be noted that the flights of the worm 232 are preferably configured to prevent the worm from being backdriven by the sector gear 234 so that when there is high tension in the belt 14, the biasing member 326 cannot relieve itself by backdriving the worm 232. If the rotation is in a second direction shown by arrow 254 in Figure 4a, the tensioning force applied to the belt 14 is reduced progressively.

[0064] The tensioner 200 is provided with two sensors shown at 260 and 262 respectively. The sensor 260 is provided for measuring the speed of the pulley 222 and may be similar to the sensor 72 shown in Figure 2a. The sensor 262 is provided for measuring the position of the tensioner arm 224. The sensor 262 may be any suitable type of sensor. For example it may be an angular position sensor, which is stationary and which can detect changes in the angular position of a circular (or, more precisely, disk-shaped) magnet 264 which is connected for rotation with the tensioner arm 224 and is aligned with the hub axis Ah (Figure 4b). The magnet 264 has a north pole that makes up 180 degrees of the magnet (i.e. a hemi-disk), and a south pole that makes up the other 180 degrees (i.e. the other hemi-disk). The position sensor 262 may be any suitable type of sensor that can detect the change in the position of the north and south poles of the magnet 264 as the arm 224 pivots about the hub axis Ah. An example of a suitable sensor is a model 2SA-10 Sentron sensor provided by Sentron AG, of Zug, Switzerland. A suitable position sensing system (made up of a sensor like sensor 262 and a circular magnet like magnet 264) is shown and described in US Patent 7,188,021 , which is incorporated herein by reference in its entirety. Instead of providing two sensors 260 and 262 in the tensioner 200, it would be possible to omit one of the sensors 260 or 262, or to provide the tensioner 200 with neither of the sensors 260 or 262.

[0065] Another sensor (not shown) may be provided to determine the position of the first end 248 of the biasing member 226.

[0066] The information from the sensors may be sent to the controller 34 which may be programmed to use the pulley speed information from sensor 260 to determine the speed of the belt 14, which the controller 34 can use when determining if there is any belt slip at one or more of the accessory pulleys or crankshaft pulley engaged with the belt 14. The information from sensor 262 may be used by the controller 34 to determine the precise position of the tensioner arm 24 which the controller 34 can use to determine the tensioning force being applied to the belt 14 and therefore the tension in the belt 14. The controller 34 can use this information as feedback to assist it in controlling the actuator 222 in order to apply a selected amount of tension in the belt 14 using any of the algorithms described herein.

[0067] Reference is made to Figures 6 and 7a-7c, which describes a tensioner 300 in accordance with another embodiment of the present invention. The tensioner 300 may have a hub 321 , a tensioner arm 324 that has a pulley 322, a biasing member 326 that drives the arm 324 towards the free arm stop position, and an actuator 328 that drives a spindle 336 that drives one end of the biasing member 326 to control the biasing force, and therefore the tensioning force, and therefore the belt tension. All of these components may be similar to their counterparts, 221 , 222, 224, 226, 228 and 236 shown in Figures 4a-4c, except that the tensioner arm 324 has a projection 370 that extends downwards into a slot 372 on the spindle, shown at 336. Referring to Figures 7a-7c, Figure 7a shows the tensioner 300 at a low tension setting. It will be noted that all of the tensioner arm 324, except for the projection 370 has been omitted from Figures 7a-7c so as to show the projection 370 and the slot 372 more clearly. As can be seen in Figure 7a, at this setting, the projection 370 is roughly in the middle of the slot 372, spaced from both ends, shown at 374 and 376 of the slot When the controller 34 determines that the belt tension should be raised, the motor shown at 330 drives the worm 332, which drives the sector gear 334, which drives the spindle 336, which drives the first end 348 of the biasing member 326 thereby increasing the biasing force of the biasing member 326 on the tensioner arm 324. This causes the tensioner arm 324 to rotate as it presses with more tensioning force into the belt 14. Figure 7b shows the tensioner arm 324 at a midlevel tension setting. It will be noted that in Figure 7b, the projection remains spaced from both ends 374 and 376 of the slot 372. As the controller 34 continues to drive the first end 348 of the biasing member 326, the first end 348 of the biasing member 326 will rotate angularly towards the second end shown at 350. If there is too much angular rotation of the first end 348 towards the second end 350, it may be possible that the biasing member 326 could become damaged. To prevent this, as the first end 348 rotates towards the second end 350, the first end 374 of the slot 372 on the spindle 336 rotates towards the projection 370. If the first end 348 of the biasing member 326 rotates by a selected angular amount towards the second end 350, the first end 374 of the slot 372 engages the projection 370. At this point, if the actuator 328 continues to drive the rotation of the first end 348 of the biasing member 326, the projection 370 (and therefore the tensioner arm 324) will be driven to rotate by the first end 374 of the slot 372 so that the relative angle between the first and second ends 348 and 350 of the biasing member 326 cannot decrease any further. Because the relative angle between the first end 348 and second end 350 of the biasing member 326 no longer decreases with further rotation of the sector gear 334, the biasing force applied by the biasing member 326 to the tensioner arm 324 no longer increases. However, because the tensioner arm 324 continues to be driven further and further into the belt 14 with further rotation of the sector gear 334 (by the motor 330), the tension in the belt 14 continues to increase. Figure 7c shows the tensioner arm 324 at a high tension setting where the projection 370 engages the first end 374 of the slot 372.

[0068] It will be noted that the projection 370 could instead be on the spindle 336 and the slot 372 could instead be on the tensioner arm 324. It will be further noted that there need not be a slot at all for engaging the projection 370. For example, a first projection could be provided on the spindle 336 that extends radially outwards and upwards. A second projection could be provided on the tensioner arm 324 and could extend radially outwards, or optionally radially outwards and downwards. A first engagement surface on the first projection would engage a second engagement surface on the second projection if the relative angle between the first and second ends of the biasing member decreases below a selected angle, thereby preventing any further decrease in the relative angle.

[0069] The tensioner 300 may includes sensors similar to sensors 260 and 262 in the embodiment shown in Figures 4a-4c. [0070] Reference is made to figure 8a, 8b, 9a and 9b, which shows a tensioner 400 in accordance with another embodiment of the present invention. The tensioner 400 may have a hub 421 , a tensioner arm 424 that has a pulley 422, a biasing member 426 (Figure 8b) that drives the arm 424 towards the free arm stop position, and an actuator 428 that may all be similar to the components 221 , 224, 222, 226 and 228 shown in Figure 4a-4c, except that the actuator 428 does not drive the first end shown at 448 of the biasing member 426, and instead drives a spindle 436 that drives a bumper arm 475 that includes a second biasing member 476 and a cover 477, into a projection 478 on the tensioner arm 424 to drive the tensioner arm 424 (more specifically the pulley 422 on the tensioner arm 424) into the belt 14, thereby controlling the tensioning force, and therefore the belt tension. Figure 9a shows the tensioner in an unboosted state. This may be considered a low tension setting, and the first biasing member 426 may be selected to have a spring constant that is relatively low for this purpose and may be configured to exert a relatively low biasing force on the tensioner arm 424 for this purpose. In this state the bumper arm 475 is unengaged with the projection 478 on the tensioner arm 424 and so the tensioner 424 operates as a typical tensioner according to the response curve 102 shown in Figure 1 c. It will be noted that in this state, the bumper arm 475 is sufficiently out of the way that the tensioner arm 424 can pivot throughout a desired range of motion without engagement between it and the bumper arm 475.

[0071] When it is desired to boost the tension in the belt 14, the motor shown at 430 drives worm 432, which drives sector gear 434, which drives spindle 436 to bring the bumper arm 475 into engagement with the projection 478 on the tensioner arm 424 (shown in Figure 9b), so as to urge the tensioner arm 424 into the belt 14 to increase the tension in the belt 14. Once engagement between the bumper arm 475 and the tensioner arm 424 occurs, further rotation of the motor 430 results in a progressively greater amount of compression of the second biasing member 476, which increases the amount of force it applies to the tensioner arm 424. This in turn increases the tensioning force applied by the tensioner arm 424 on the belt 14 which in turn increases the belt tension. Because the motor 430 is only engaged with the tensioner arm 424 through the second biasing member 476, the tensioner arm 424 can still pivot to accommodate dynamic fluctuations in belt tension even when the tension is being boosted, or more broadly, even when the tensioner arm 424 is engaged by the bumper arm 475.

[0072] Referring to Figure 9b, a moment arm M1 exists between the line of action through the pulley axis Ap of the pulley 422 and the pivot axis Ah of the tensioner arm 424. A moment arm M2 exists between the line of action of the bumper arm 475 on the projection 478 and the pivot axis Ah. It is optionally possible to configure the tensioner 400 so that the moment arm M2 is greater than the moment arm M1 so as to facilitate driving the pulley 422 into the belt 14 against the force exerted by the belt 14 on the pulley 422.

[0073] The tensioner 400 may include one or more sensors. A sensor shown at 460 may be similar to sensor 260 and is used for determining the speed of the pulley 422. A sensor shown at 462 may be similar to sensor 262 and is used for determining the position of the tensioner arm 424. A sensor shown at 465 is provided for determining the force applied by the second biasing member 476 on the tensioner arm 424, which is used in addition to the information from the sensor 462 which provides tensioner arm position information and therefore information regarding the force applied on the arm 424 by the first biasing member 426, so as to determine the tensioning force being applied to the belt 14. This can be used as feedback for the controller 34 while driving the motor 430 to boost the belt tension, in order to determine at what position to stop the motor 430. The sensor 465 may be, for example, a button load cell mounted to the cover 477. Instead of the sensor 465, the bumper arm 475 may contain a displacement measurement sensor 466, which can be used to measure the length of the bumper arm 475, thereby indicating the amount of compression there is in the second biasing member 476. This can be used by the controller 34 to determine the force exerted by the second biasing member 476 on the projection 478, which can be used to determine the overall tensioning force applied by the tensioner 400 on the belt 14 and thereby determine the belt tension. The displacement measurement sensor 466 may be configured to signal the controller only when in extended and retracted positions, or may be configured to signal to the controller 34 when it reaches any of three or more positions.

[0074] When it is desired to position the tensioner 400 in a low tension setting, the motor 430 may be rotated in the opposite direction as when boosting tension. When the controller 34 detects that the sensor 465 no longer senses any engagement between the bumper arm 475 and the projection 478, the controller 34 may be programmed to rotate the motor 430 by a selected number of revolutions to bring the bumper arm 475 sufficiently out of the way to accommodate the pivoting of the tensioner arm 424 when operating at the low tension setting.

[0075] It will be noted that, in the embodiment shown in Figures 8a, 8b, 9a and 9b, the element that holds the first end of the biasing member 426 is fixedly mounted to the hub 421 so that the first end of the biasing member 426 remains stationary and is not rotated with the spindle 436.

[0076] Reference is made to Figure 10, which shows a tensioner 500 in accordance with another embodiment of the present invention, which includes a hub 521 , a pulley 522, a tensioner arm 524, a biasing member 526 and an actuator 528, which is controlled by controller 34. The actuator 528 includes, among other things, a motor 530, a worm 532, a sector 534 that is connected for rotation with a spindle 536. The tensioner 500 may be similar to the tensioner 400 except that in the tensioner 500, the element that holds the first end of the biasing member shown at 526 is part of the spindle (shown at 536). Figure 1 1a shows the tensioner 500 at a low tension setting. In this state, the bumper arm shown at 575, which may be similar to bumper arm 475, is out of the way of the projection shown at 578 on the tensioner arm 524 so that the tensioner arm 524 can pivot throughout some range of motion as a result of engagement with the belt 14 without contacting the bumper arm 575. Figure 1 1b shows the tensioner 500 at a high tension setting. To boost the tension in the belt 14, the bumper arm 575 is driven by the motor shown at 530 toward the projection 578. This drives the first end 548 of the biasing member 526 however, which increases the torsion in the biasing member 526, and therefore drives the tensioner arm 524 to rotate further into the belt 14, increasing the belt tension. At some point, the spindle 536 is rotated sufficiently relative to the tensioner arm 524 that the bumper arm 575 engages the projection 578. At that point, further driving of the spindle 536 drives the bumper arm 575 further into engagement with the projection 578 which increases the biasing force applied by the biasing member 526, and which also increases the biasing force applied by the second biasing member shown at 576 which is part of the bumper arm 575. This assists in preventing damage to the first biasing member 526 resulting from overrotation of the first end 548, because the presence of the bumper arm 575 provides assistance so that a selected belt tension can be reached without all of the force having to be generated through the first biasing member 526. The remote drive arrangement shown for the tensioner 500 may be applied to any of the tensioners shown and described herein where a motor is utilized and would benefit from being positioned away from the engine block.

[0077] The tensioner 500 may includes sensors similar to the sensors 460, 462 and 465 shown in Figures 8a and 8b and a displacement measurement sensor similar to displacement measurement sensor 466.

[0078] Reference is made to Figure 12, which shows a tensioner 600 in accordance with another embodiment of the present invention. In a situation where the temperature in the immediate environment of the motor is high, the operating efficiency of the motor decreases, sometimes significantly. This is due to several factors, including the fact that the resistance of the copper wire in the motor increases with temperature. Thus at 60 degrees Celsius or at 80 degrees Celsius, the wire has a higher resistance than it does at 25 degrees Celsius. As a result, the torque capability of the motor is reduced at high temperature. To compensate for this the motor may have to be upsized significantly in order to be assured of being able to handle a given torque requirement. To assist in keeping the cost of the motor shown at 630 low, the tensioner 600 may be similar to the tensioner 200 except that the tensioner 600 includes a remote drive arrangement in that the motor 630 is operatively connected to the worm 632 through a flexible shaft 680. The flexible shaft 680 permits mounting the motor 630 away from the engine block, which means it may be possible to keep the motor 630 at a lower temperature than one that is mounted to or proximate to the engine block. The motor 40 may be positioned in a particularly cool part of the engine bay, such as near the grill or at some other cool air inlet. This would permit a less expensive motor to be used since it would not need to be oversized to compensate for a reduction in efficiency resulting from operating temperature.

[0079] Also, in any of the embodiments wherein a motor drives either the tang of the biasing member or drives a bumper arm, a locking mechanism (not shown) can be brought into provided to lock the spindle in a particular position. The locking mechanism may include a pawl or the like that engages an aperture or a detent or the like in the spindle. The spindle may include only a single aperture or detent, and is thus only held when it reaches a certain position, or it may have a plurality of plurality of apertures or detents about its circumference so that it can be held in a plurality of different positions. An actuator, such as a solenoid, would be provided to withdraw the pawl from the aperture/detent to permit the spindle to move to a different position or to return to a home position. Instead of this, the locking mechanism could include a cam that engages the tensioner arm, a wrap spring or any other suitable means.

[0080] If a gearbox is provided (as shown at 682) to reduce the speed and increase the effective torque provided by the motor 630, it is preferable that the gearbox be provided at the outlet end of the flexible shaft 680 (i.e. between the flexible shaft 680 and the worm shown at 632), as opposed to providing the gearbox between the motor 630 and the flexible shaft 680. As a result, the flexible shaft 680 is not relied upon to transmit high torque, which can result in its premature wear and/or windup in the flexible shaft 680. The remote mounting of the motor 630 permits the motor 630 to be positioned in a particularly cool part of the engine bay, such as near the grill or at some other cool air inlet.

[0081 ] Reference is made to Figure 13a, which shows a tensioner 700 in accordance with another embodiment of the present invention. The tensioner 700 includes a hub 721 , a tensioner arm 724 that is pivotable about a hub axis Ah, a pulley 722 mounted on the tensioner arm 724 for rotation about a pulley axis Ap, a biasing member 726 engaged between the hub 721 and the tensioner arm 724 to bias the tensioner arm 724 towards the free arm stop position, and a tensioner actuator 728. The tensioner arm 724 shown in Figure 13a includes a central portion 724a and an extendable portion 724b that is extendable from the central portion 724a by the actuator 728. A motor 730 that is mounted on the central portion 724a and drives a lead screw 790, which in turn causes linear movement of a traveler 792 mounted at one end of the tensioner arm 724. This linear movement causes extension or retraction of the tensioner arm 724 depending on the direction of rotation of the motor 730. Referring to Figures 13c and 13d, changing the length of the tensioner arm 724 (shown at L) changes the moment arm between the pulley 722 and the hub axis Ah, and also changes the hubload angle. Changing these properties changes the tensioning force applied by the pulley 722 on the belt 14, which depends at least in part on the moment arm M2 of the pulley 722 as compared to the moment arm M1 of the biasing member 726. Thus, when the moment arm of the biasing member 726 is constant as shown in Figure 13a, reducing the length L of the tensioner arm 724 increases the tension in the belt 14 because it increases the force of the pulley 722 acting on the belt 14. Similarly, extension of the tensioner arm 724 decreases the moment arm of the tensioner arm 724 on the belt 14, which decreases the tension in the belt 14. The length of the tensioner arm 724 may be determined using any suitable type of sensor, such as a Hall effect sensor that counts revolutions of the leadscrew 790. In order to stabilize the arm 724 during use, the extendable portion 724b may be generally U-shaped (albeit with a very shallow U in the embodiment shown in Figure 13b). To further stabilize the arm 724, a clamping mechanism (not shown) may be provided which removes any play that exists between the extendable portion 724b and the central portion 724a. It will be noted that any of the tensioners shown and described may include a damping member as is known in the art to dampen oscillations of the tensioner arm. The damping force from the damping member is preferably proportional to the torque applied.

[0082] Reference is made to Figures 14a and 14b, which shows a tensioner 800 in accordance with another embodiment of the present invention. The tensioner 800 may be similar to the tensioner 200 except that the tensioner 800 includes an actuator 828 that includes a motor 830 that turns a lead screw 890 which has a traveler 892 thereon. The traveler 892 engages projections 894 on the spindle, shown at 836. The spindle 836 engages the first end 848 of the biasing member 826 so as to control the biasing force of the biasing member 826 on the tensioner arm shown at 824. The tensioner 800 further includes a hub 821 that mounts to the engine. A pulley 822 is mounted for rotation on the tensioner arm 824.

USE OF HYDRAULIC ACTUATOR

[0083] Referring to Figures 15-43, a tensioner 910 is shown for an endless drive member 91 1 along with several components engaged with the belt 91 1 , including the MGU 16 (or alternatively an alternator 16) along with its pulley 54, the a/c compressor 18 along with its pulley 60, the crankshaft pulley 12 and an additional driven pulley shown at 915, which could be from an accessory such as the water pump 55 (Figure 3a). The pulley 915 includes a hydraulic actuator 913 which includes a cylinder 912. The cylinder 912 includes a first housing 914 and a first piston 916. The cylinder 912 at least partially defines a fluid chamber 918. The first piston 916 defines a first end of the fluid chamber 918. In the embodiment shown, the hydraulic actuator 913 further includes a second housing 920 (which may be referred to as a reservoir) is provided that is connected to the first housing 914 via a fluid passageway 922 (Figure 18) having a selected resistance to fluid flow. The second housing 920 defines part of the fluid chamber 918. A second piston 924 is movable in the second housing 920 and defines a second end of the fluid chamber 918. The fluid in the fluid chamber 918 (during use) is preferably a substantially incompressible fluid such as hydraulic oil. A first piston biasing member 926 is positioned to bias the first piston 916 towards the extended position (shown in several figures including, for example, Figure 18). A second piston biasing member 928 such as a spring is optionally provided, to assist in urging the second piston 924 into engagement with the fluid in the fluid chamber 918, even though this could be accomplished with a gas in the second housing 920 instead of the spring. A second piston position control device 930 is provided to control the position of the second piston 924. The control device 930 may include a threaded rod 932 that engages a threaded portion 934 at the end of the second housing 920. The threaded rod 932 may be rotated in one direction to drive it in towards the second piston 924 or in another direction to drive it away from the second piston 924. The threaded rod 932 may be used to set a limit position for the second piston 924. This limit position may be chosen so that the second piston 924 can be pushed by the fluid in the fluid chamber 918 by some selected amount. Alternatively, the threaded rod 932 could be brought right up into abutment with the second piston 924 so as to prevent the fluid from pushing the second piston 924 at all. Alternatively, the threaded rod 932 could be rotated to drive the second piston 924 forward by some selected amount, thereby forcing the first piston 916 towards the extended position. [0084] The first piston 916 is pivotally mounted at an external end to a base structure shown at 917a via a first cylinder pivot stud 919, about a cylinder pivot axis ACYL. The external end of the housing 914 is pivotally connected to a lever arm 936, via a second cylinder pivot stud 921 , about a lever arm engagement axis ALAE. The first and second pivot studs 919 and 921 are simply elements about which the ends of the housing 914 and piston 916 can pivot. They may alternatively be referred to as pivot posts 919 and 921.

[0085] The lever arm 936 is pivotally connected via a pivot stud 923 to a base structure 917b about a lever arm pivot axis ALAP. The base structure 917a and the base structure 917b may both be part of the same component, such as the engine block. Alternatively they may be portions of different components. A pulley 938 is rotatably mounted to the lever arm 936. The pulley is engaged with the endless drive member 911 to tension the endless drive member 91 1. In an alternative embodiment, it is possible for the pulley 938 to be directly mounted to the external end of the housing 914 and to omit the lever arm 936. In such an embodiment, the movement of the housing 914 would be constrained in some way, such as by guides that force it to travel linearly.

[0086] The threaded rod 932 may be driven to rotate by a motor shown at 940 optionally via a flexible shaft shown at 942. The motor 940 may be fixedly mounted (e.g. to a base structure that is spaced from the engine block) and the flexible shaft 942 may permit operative connection of the motor 940 to the threaded rod 932 even though the threaded rod 932 moves during operation of the tensioner 910. By mounting the motor remotely from the engine block it may be possible to keep the motor 940 at a lower temperature than one that is mounted to or proximate to the engine block. This would permit a less expensive motor to be used. Alternatively as shown in Figure 14, the motor 940 may be mounted to the second housing 920, in which case a shaft that is flexible is not necessary. [0087] If a gearbox is provided to reduce the speed and increase the effective torque provided by the motor, it is preferable that the gearbox be provided at the outlet end of the flexible shaft (i.e. between the flexible shaft 942 and the threaded rod 932), as opposed to providing the gearbox between the motor and the flexible shaft 942. As a result, the flexible shaft is not relied upon to transmit high torque, which can result in its premature wear and/or windup in the flexible shaft.

[0088] Where the motor 940 is positioned remotely from the second housing 920, the motor 940 may be positioned in a particularly cool part of the engine bay, such as near the grill or at some other cool air inlet.

[0089] The operation of the motor 940 may be controlled by any suitable control system. A generic control system is shown at 943 in Figure 15. The control system 943 may be made up of one or more individual components (e.g. one or more control units). The control system 943 may include the main ECU of the vehicle, or it may be separate from the main ECU of the vehicle. In an embodiment, the control system 943 may be a separate controller that communicates with the ECU of the vehicle via a wiring harness or via a CAN bus. A number of suitable types of relationship may exist between the ECU and any other controllers that together form the control system 943. For example, the ECU may sometimes or may all the time be a master to the other controllers in the control system 943.

[0090] The control system 943 may set the position of the threaded rod 932 using the motor 940 based on any suitable set of parameters. The control system 943 may, for example, drive the threaded rod 932 inwards to move the second piston 924 so as to move the first piston 916 to provide a selected high tension in the belt 91 1 in some situations where the control system 943 determines that the belt 911 is determined to be at risk of slippage or if slippage has been detected. An example of such a situation is upon startup of the vehicle, particularly on a cold day. Another example of such a situation is upon startup during a BAS (Belt Alternator Starter) belt starting sequence, where high belt tension is required only during the initial motor / alternator belt starting sequence, to spin the engine crankshaft by the belt. Another example of such a situation is during moments where there is high engine acceleration or deceleration.

[0091 ] The control system 943 may also drive the threaded rod 932 outwards to a position where it permits the second piston 924 to travel freely in the second housing 920 where, for example, a lower tension is acceptable for the belt 91 1 without risk of belt slippage. In some situations, the control system 943 may drive the belt inwards so as to simply abut the second piston 924 so as to prevent the second piston from permitting compression of the cylinder 912. This can be used during periods of belt flutter or belt resonance, during periods of high belt drive rotation torsional acceleration (engine torsionals), or for example, during situations where a system disturbance (e.g. belt flutter) is anticipated based on inputs from one or more sensors, in order to prevent tension loss / reduction due to the tensioner being forced back out of the belt during such periods of instability.

[0092] The control of the motor 940 by the control system 943 may be open-loop or may be closed loop whereby the control system 943 receives inputs from one or more sensors that indicate whether the position of the threaded rod 932 is as intended. Additionally the control system 943 may receive inputs from one or more other sensors which it uses to determine where to position the threaded rod 932. Such sensors may include, for example, temperature sensor, a rain sensor, a sensor related to belt slip, and the like. Suitable algorithms for the control system 943 are described in copending application 61/486,189, the contents of which are incorporated herein by reference.

[0093] The L-shaped configuration between the cylinder 912 and the second housing 920 facilitates rapid tuning using take apart components which can be switched out as required. However, an in-line configuration may alternatively be provided. Furthermore, a monotube configuration may be provided in which the entire fluid chamber and both the first and second pistons are contained in a single cylinder housing. In a monotube configuration, the connector at the external end of the housing could be shifted to the side so as to make room for the threaded rod to enter into the housing through the end of the housing.

[0094] In an alternative embodiment, the piston 916 and housing 914 could be reversed, whereby the piston 916 is connected to the lever arm 936 and the housing 914 is pivotally connected to the engine block.

[0095] In one aspect, the novel concept here is to block the return of the second piston to induce hydraulic lockout (piston lockout) thereby preventing, on demand, any retraction of the first piston 916 (i.e. compression of the cylinder 912) in response to a specific FEAD belt drive system requirement.

[0096] In another aspect, the novel concept here is to drive the small reservoir position in power mode, to extend (or retract) the strut to amplify the force by driving the first piston which may be larger than the second piston.

[0097] Different travel rates and load outputs can be achieved by manipulating the piston diameter ratios between the diameter of the first piston 916 and the diameter of the second piston 924.

[0098] In some embodiments the connection between the first housing 914 and the second housing 920 could be separated by a flexible hose - a stiff flex steel braided hose, for example, would probably be employed in such a case so as to minimize unwanted expansion of the hose under pressure, resulting in unwanted diametral elastic growth of the hose / tubing, which would ultimately result in adding an unwanted lag, or delayed response, to the final spring rate. Unwanted elasticity on the high pressure oil side is preferably eliminated in order to prevent force response lags, hence the use of non-elastic hoses where practical. [0099] In the exemplary embodiment presented in the illustrations, the majority of damping comes from the seals and damping rings installed within the first piston 916, as well as whatever damping comes from the flow of fluid.

[00100] Extra damping can be added to the tensioner by introducing such devices as a small orifice restriction and a blow off ball valve in the oil flow between the two housings, to induce restrictive orifice damping and steady blow off damping.

[00101] In HYDRAULIC LOCKOUT MODE, the rod 932 merely stops the backward motion of the floating reservoir piston, which ultimately hydraulically locks the larger piston 916 and prevents further insertion of the piston rod into the cylinder housing 914, effectively locking the tensioner arm 936.

[00102] In POWERED DRIVE MODE, the rod 932 is used to push the smaller floating piston 924, in order to force the larger tensioner cylinder piston 916 to move in response to the flow of oil from the reservoir 20. The technique, for example, may be used to push the pulley 938 harder into the belt under conditions where belt slip would otherwise be imminent, such as during a BAS start (in hybrid vehicles), or during wet or extremely cold conditions.

[00103] The rod 932 used to lock out or drive the reservoir piston 924 can be driven via a fine lead screw arrangement, as depicted in the figures, or it could be driven linearly (as opposed to the rotational actuation required by a lead screw arrangement) by a powerful linear solenoid - any actuator device which is capable of blocking movement of the smaller reservoir piston 924 or of driving the reservoir piston 924 (in powered drive mode described below) could be employed. The solenoid may actuate the rod 932 linearly via a flexible cable (e.g. a Bowden cable). This would permit the solenoid to be mounted remotely in a cooler spot in the engine bay, (e.g. near the grill) in similar manner to the remote mounting of the motor 940, described above. An example of a suitable solenoid with a flexible cable arrangement that may be used in some embodiments is provided by Trombetta, which is owned by Fulham & Company, and is located at Menomonee Falls, Wisconsin, USA.

[00104] It will be understood that, under typical conditions where temperature is not an issue, it is preferred to mount the motor 940 or solenoid in substantially direct engagement with the rod 932 (as opposed to a remote mounting with a flexible shaft or push/pull cable between them) so as to reduce the number of components between the motor 940 or solenoid and the rod 932, so as to increase the torque transfer efficiency between the motor 940 or solenoid and rod 932. However, in a situation where the temperature in the immediate environment of the load will be high and the motor is in that environment, the operating efficiency of the motor decreases, sometimes significantly. This is due to several factors, including the fact that the resistance of the copper wire in the motor increases with temperature. Thus at 60 degrees Celsius or at 80 degrees Celsius, the wire has a higher resistance than it does at 25 degrees Celsius. As a result, the torque capability of the motor is reduced at high temperature. To compensate for this the motor may have to be upsized significantly in order to be assured of being able to handle a given torque requirement.

[00105] The threaded rod 932 is in a hot environment due to its proximity to the engine. By mounting the actuator (e.g. the motor 940 or the solenoid) remotely from the rod 932, the actuator can be placed in a cooler environment where its operating efficiency is not degraded, or not degraded as much as it would be in the hotter environment. For example, the actuator may be placed near the grille of the vehicle where it is exposed to a strong airflow for cooling purposes.

[00106] Also a remote mounting of the actuator can be advantageous if there are space and size limitations proximate to where the rest of the tensioner 910 is mounted. [00107] Instead of using a solenoid, a vacuum actuator can be used in some embodiments to drive the rod 932 linearly. The vacuum actuator may draw power from the vacuum drawn from the air induction system. Vacuum drawn from the air induction system has been used in the past for other purposed such as to drive turbo waste gate motors and manifold induction duct actuator motors, for example.

[001 OB] In some embodiments a pneumatic actuator may be used to drive the rod 932 if a compressed air supply is available onboard the vehicle. Alternatively, a hydraulic actuator could be used, drawing power from a source of hydraulic fluid, such as the power steering system in vehicles that use a hydraulic power steering system, or, for example from the engine oil system. A compressed air supply, a hydraulic power supply, or a vacuum source are sometimes available on certain vehicles such as trucks and SUVs.

[00109] A vacuum actuator, a pneumatic actuator, or a hydraulic actuator may be switched via a relatively inexpensive solenoid powered valve, and could be controlled electronically, via either an analog or digital signal, from the control system 943.

[00110] An example of a suitable actuator may be a door lock actuator, similar to (but perhaps scaled up in power and resistance to high temperature) those described in US Patent Nos. US6067826, US5983739, US5634676 the contents of all of which are incorporated herein by reference. Such an actuator could be used successfully in some embodiments by incorporating a gearbox with worm drive and a high gear ratio. Backdriving of such an arrangement can be prevented by the worm drive (in embodiments wherein it is provided) and/or by providing some other structure such as one or more detent lock features, with a spring loaded lock and load feature as described in US Patent No. 5,983,739, the contents of which are hereby incorporated by reference.

[001 1] While a push-pull cable has been described as being usable in certain embodiments, other arrangements may be used, in addition to or instead of a push-pull cable. For example a linkage may be used such as the linkage shown in US Patent No. 5,634,676 the contents of which are hereby incorporated by reference, optionally with a push-pull cable. Using some linkages linear motion could be converted to rotary motion, or vice-versa. Such linkages may be used to alter the mechanical advantage by increasing force by reducing displacement, or by decreasing force by increasing displacement, depending on the embodiment.

[00112] The first piston biasing member 926 may be a coil spring, as shown in the figures, or it may be some other type of spring such as a closed cell foam (CCF) spring as shown in Figure 27. It could alternatively incorporate both a coil spring and a closed cell foam spring. An example of such a spring is supplied by BASF. An advantage to CCF springs is that they can collapse to only 20% of their original height (vs 40% for some steel coil springs). Another advantage to CCF springs is that it is easy to manufacture them with a desired constant spring rate or a spring rate that varies with the amount of compression of the CCF spring. This may be achieved by co-molding portions (e.g. layers) of the CCF spring, each having different properties. Properties that may be varied include density of the CCF, the cell size, the outer diameter and inner diameter of the CCF spring (in embodiments wherein they are hollow- cylindrical. Additionally you can easily tune their design to provide a selected amount of energy dissipation. The shorter collapsed length of the CCF spring could be used to provide a larger linear travel of the cylinder 912 for a given overall length of the cylinder 912. As another alternative, the first piston biasing member may include some other type of biasing member such as, for example, one or more Belleville washers, one or more wave washers, gas pressure, or a combination of two or more of these. All of these alternatives may also be used as the second piston biasing member 928 instead of or in addition to a coil spring.

[00113] In an example, the CCF spring may be used just as an end-of- travel jounce stop, to impart an elastically damped, high spring rate at the end of travel of a typical coil spring design. In another example, the CCF spring may be molded around metal coil springs or washer springs to impart new spring and new jounce stop properties to the overall spring assembly.

[00114] In an alternative embodiment shown in Figure 25, the second piston biasing member 928 may be positioned between the second piston 924 and a limit member 944 that is in the second housing 920. The threaded rod 932 engages the limit member 944. This structure is particularly advantageous when using the threaded rod 932 to set a limit for the amount of retraction that is possible for the first piston 916 in the first housing 914. With this structure, as the threaded rod 932 is advanced into the second housing 920, it advances the limit member 944 towards the second piston 924. This in turn causes the second piston biasing member 928 to compress, increasing its resistance to further compression. As a result, when the second piston 924 is moved to abut the limit member 944 because of an increase in tension in the belt 91 1 , the increased resistance offered by the biasing member 928 helps to reduce the abruptness of the engagement between the limit member 944 and the second piston 924.

[00115] Reference is made to Figure 26, which shows another embodiment of the present invention. In this embodiment, the tensioner includes a hub 946 that is mountable to a base structure such as the engine block (not shown). A tensioner arm 948 is pivotable about the hub 946. The pulley 938 is positioned at the end of the arm 948. A tensioning spring (not shown) and a damping element (not shown) are positioned about the hub 946, as they are in a typical tensioner. The cylinder 912 is mounted in such a way that one end (in this case the external end of the housing 914) is pivotaily mounted to a base structure, and the other end (in this case the external end of the piston 916) is pivotaily mounted to the tensioner arm 948. The second housing 920 is provided, with the second piston 924 therein. In this embodiment, the threaded rod 932 is engageable with the second piston 924 only through the limit member 944 and the second biasing member 928, however it is alternatively possible to omit the limit member 944 and to provide an arrangement similar to that shown in Figure 24, wherein the threaded rod 932 directly engageable with the second piston 924. The tensioner shown in Figure 26 may act substantially like a typical tensioner. However, the additional structure including the cylinder 912 and related components, permits the tensioner arm 948 to be locked hydraulically as described above (ie. by engagement of the threaded rod 932 or limit member 944 so as to prevent complete compression of the cylinder 912), and to be driven to a different (higher tension) position via the threaded rod 932 and a suitable actuator (eg. motor 940 and flexible shaft 942).

[00116] Reference is made to Figures 29a and 29b which show alternative structures for locking the tensioner 910 and for boosting the tension in the tensioner 910. A needle valve shown at 952 may be provided for controlling the flow of fluid between the first chamber portion shown at 918a and the second chamber portion shown at 918b. The needle valve 952 includes a flow control element 954 that is positionable in at least an open position wherein flow through the passageway 922 is permitted and a closed position (Figure 29b) wherein flow through the passageway 922 is prevented. The flow control element 954 may be positionable in only those two positions (by means, for example, of a two-position solenoid) or alternatively it may be positionable in three or more positions. In some embodiments it may be infinitely adjustable in position via proportional control. When the flow control element 954 is in the closed position, the tensioner 910 is locked hydraulically in position, in similar manner to the embodiment shown in Figures 15-28 in which the threaded rod 932 is adjusted to abut the second piston 924. When the flow control element 954 is in the open position it does not prevent the cylinder 912 from extending and retracting. When the flow control element is positionable in one or more partially open positions, an example of which is shown in Figure 29a, it can be used as an adjustable orifice to control the rate of flow of fluid between the chamber portions 918a and 918b, thereby controlling the rate of extension or retraction that is available to the cylinder 912 and acting as a damping member. While the threaded rod 932 is shown in Figures 29a and 29b it is not needed and the second housing 920 could simply have a closed end instead.

[00117] Also shown in Figures 29a and 29b is an alternative tension boost mechanism shown at 955. The tension boost mechanism includes a fluid reservoir 956 and a pressure source 957, such as an electric pump. The pressure source 957 connects to the chamber portion 918b (or alternatively to somewhere else in chamber 918), via conduit 958. When it is desired to boost the tension, the pump 957 operates to pump fluid into the chamber 918, forcing the second piston 914 backwards until the second biasing member 928 can no longer compress, and forcing the cylinder 912 to extend to drive the pulley 938 into the belt 91 1. When a desired tension is reached in the belt a valve (not shown ) in conduit 958 can be closed which would still permit the tensioner 910 to extend and retract, or alternatively, the locking valve 952 could be closed, however this would lock the tensioner 910 in position.

[00118] Reference is made to Figure 30, which shows the controller 34. As described above the controller receives input 960 that may include any one or more of the following:

ENGINE ECU PARAMETERS

Time information, which provides daily driver information such as time of vehicle use, days on / days off and the like;

Date information, which provides seasonal time of year approximations, and which may be available from SAT radio / GM OnStar™ / Ford Sync™ / ECU with battery backup;

Engine information, including RPM (based on crankshaft pulley speed) / engine on / engine start successful Transmission information including transmission gear position & shift status which provides rate of acceleration / deceleration information for belt slip prediction / transmission downshift & upshift information,

Cruise Control information: prolonged steady state speed would imply that use of lower belt tension is permissible / no rapid engine RPM transitions

VEHICLE ECU PARAMETERS

Windshield wiper status - Via CAN bus

Windshield rain sensor - Via CAN bus

Traction control system wheel slip status (acceleration) - wheel slippage indicates wet or slushy road conditions - predicts belt contamination / belt slip probability

ABS system wheel slip status (braking) - wheel slippage indicates wet or slushy road conditions - predicts belt contamination / belt slip danger probability Key fob Sensor in a simple embodiment it can be used to sense the approach of any driver and can adjust belt tension prior to initial cold engine start. In a more advanced embodiment it can sense approach of driver 1 / driver 2 - and can predicts driver driving style

Ignition Status: For a hybrid vehicle stop/start applications - if brakes are applied and if wheel speed = 0 and engine RPM = 0 would indicate to increase belt tension

Wheel Speed - Hybrid stop /start applications - if brakes are applied AND if wheel speed = 0 AND if engine RPM = 0 this is an indication to increase belt tension

Brakes - Hybrid stop /start applications - if brakes are applied AND if wheel speed = 0 AND engine RPM = 0 then this is an indication to increase belt tension Engine RPM - Hybrid stop /start applications - if brakes are applied AND if wheel speed = 0 AND if engine RPM = 0 then this is an indication increase belt tension

Headlight Status - If the high beams are on this may indicate fog or some other wet condition in which it would be beneficial to increase belt tension

Throttle position sensor, and the Brake position sensor: this includes the actual positions of the accelerator pedal and brake pedal, and the rate of change in the positions of the accelerator pedal and the brake pedal: these may be used to indicate whether the vehicle is coming to a stop, is leaving a stopped condition, or is in a panic stop, or is in a panic acceleration

Driver selectable settings, such as the driving mode (LUXURY, SPORT, ECO, NORMAL - used to control shift map for the transmission and in some cases the suspension settings). Seat settings, mirror settings and the like may also be used to indicate to the vehicle which driver is in it, which can be used to indicate whether or not high tension in the belt is likely to be needed.

FEAD COMPONENT INFORMATION

Battery voltage and battery outgoing current

Electric Fan on/off status - indicates that the alternator may need to be started to maintain the battery charge

Alternator regulator status (charge status : on / off)

Alternator speed via rotor rotation output

Alternator temperature

Alternator voltage

Alternator charge current Hydraulic power steering pressure switch: this indicates power steering status - on / off, if traveling on a highway no turns are anticipated and so the hydraulic power steering usage is expected to be low

Electric power steering motor status: indicates power steering status - on / off, if traveling on a highway no turns are anticipated and so the electric power steering usage is expected to be low

Air conditioning compressor clutch state : A/C compressor shaft on / A/C compressor shaft off

Waterpump SWP clutch state : waterpump impeller on / waterpump impeller off

Idler pulley speed ( for belt slip calculation where tensioner pulley speed sensor is not viable )

Tensioner pulley speed ( for belt slip calculation where tensioner pulley speed is provided)

Tensioner arm angle position (measures tensioner arm rotation angle using the tensioner arm position sensor (e.g. shown at 462)

Bumper arm information: This information could include: extended/retracted if using a two position sensor, or could include multiple positions between extended and retracted positions to provide more precise information regarding the force applied to the tensioner arm by the bumper arm. Optionally a continuous displacement sensor can be provided to indicate the amount of compression in the second biasing member more accurately. A load cell sensor could be provided which would provide direct information relating to the amount of force being applied by the bumper arm to the tensioner arm.

Spindle position sensor - measures rotation of spindle - which provides spring tang wind up angle in embodiments where the first end of the biasing member is moved by the actuator. DEDICATED SENSOR INPUTS

Temperature sensor for under hood (engine bay) temperature, close to FEAD belt drive

Humidity sensor for under hood (engine bay) humidity, close to FEAD belt drive Temperature sensor - absolute outside ambient outdoor temperature

Humidity sensor - absolute outside ambient out door humidity

DEDICATED SENSOR AND CALCULATION PROCEDURES AND INPUTS

Pulley torsional vibration measurement can be as described in PCT publication WO2006045181 the contents of which are incorporated herein by reference. Belt elongation and belt stretch measurement can be as described in PCT publication WO2007143830 the contents of which are incorporated herein by reference.

Belt flutter / belt twist may be detected as a vibration or as an underhood noise by acoustic microphones, vibration sensors, or perhaps, by simple "touch" proximity sensors.

Input 960 may also include any other input described herein and may include other inputs also not described herein. The controller 34 uses these inputs to determine whether or not the change the tension in the tensioner. The controller 34 controls the motor 962 accordingly which may be the motor in any of the embodiments described herein.

USE OF LOAD / PRESSURE SENSORS

[00119] Reference is made to Figures 31 a-43, which illustrate embodiments with means for the determination of the tension in the belt 91 . In the tensioner shown in Figure 31a, for example, a pressure sensor 970 may be mounted to sense the fluid pressure in the fluid chamber 918 and to send signals to the controller 943 that indicate the sensed fluid pressure. The controller 943 can determine the tension in the belt 91 1 using this fluid pressure combined with other data indicative of the geometric relationship between the cylinder 912, the lever arm 936 and the hub load vector, shown at HLV. Examples of data that would be indicative of the aforementioned geometric relationship include the length of the cylinder 912, or the angular position of the lever arm 936. Even without the position data, however, the controller 943 could make a rough estimate of the belt tension by assuming that the cylinder 912 and the lever arm 936 are in a particular position, such as about halfway between the load stop and free arm positions.

[00120] The pressure sensor 970 may send signals to the controller 943 by either a hardwired connection, as shown in Figure 31a, or via a wireless connection as shown in Figure 31 b. A wireless connection may use any suitable technology as, for example Bluetooth or Zigbee technologies.

[00121] The pressure sensor 970 may be mounted to the second housing 920, as shown in Figure 31a. Alternatively, the pressure sensor 970 may be mounted to something else, such as the first housing 914. The pressure sensor 970 may be mounted so that it extends out from the second housing 920 (or from the first housing 914). Alternatively, the pressure sensor 970 may be mounted within the fluid chamber 918 (e.g. within the second housing 920 or within the first housing 914). While such an embodiment could be hardwired, it could alternatively be wireless, whereby it could be powered by way of a self- contained battery and could be programmed to go into a sleep mode when not in use so as to conserve energy so as to prolong the life of the battery, in similar fashion to tire pressure sensors installed on the wheels of some vehicles currently. The battery could have technology for harvesting energy from vibration or other sources in order to recharge itself. The use of a wireless transmitter from the pressure sensor 970 is preferable to a hardwired connection from the point of view of cost since the wired connection includes such costs as the electrical connectors, wiring harnesses and clips for them, and labour to install the wiring harnesses and clips.

[00122] The pressure sensor 970 may use any suitable technology for determining the pressure in the fluid chamber 918. For example, the pressure sensor 970 may incorporate a strain gauge, or piezoelectric technology.

[00123] In the embodiment shown, the second housing 920 is mounted substantially parallel to the first housing 914. This may be advantageous from a packaging perspective by being easier to fit into the sometimes tight space available about the engine in the engine bay of a vehicle. It will be noted, however, that the second housing 920 could be oriented in any other suitable orientation, such as the orientation shown in Figure 15 wherein it is generally perpendicular to the first housing 914.

[00124] Also, in the embodiment shown, the motor 940 used to drive the threaded rod 932 shown directly mounted to the second housing 920, in similar fashion to the embodiment shown in Figure 28. It is alternatively possible for the motor 940 to be mounted elsewhere, such as remotely from the second housing 920 in similar fashion to the embodiment shown in Figure 15. An electrical connector 945 is shown extending from the back of the motor 940 for connection to a power source within the vehicle. In a preferred embodiment, the electric connector 945 is connected to the controller 943 such that the controller 943 controls the operation of the motor 940.

[00125] The motor 940 shown in Figure 31a includes an associated gearbox 941. This gearbox 941 may, for simplicity's sake, be considered to be part of the motor 940.

[00126] Referring to Figure 32, a load cell 972 may be used to detect the force exerted by the cylinder 912 on the lever arm 936, which may be used along with other data by the controller 943 to determine the tension in the belt 91 1 , the other data being similar to that described above (i.e. data that is indicative of the geometric relationship between the cylinder 9 2, the lever arm 936 and the hub load vector HLV), or which may be used without other data to estimate the belt tension.

[00127] In the embodiment shown in Figure 32, the load cell 972 is placed on the first piston and is a compression load cell. The load cell 972 may have any suitable configuration for sensing the force exerted through the piston 916 and therefore the force exerted by the cylinder 912 on the level arm. For example, the load cell 972 may be a canister load cell, an S- or Z-beam load cell, a bending beam load cell or a shear beam load cell.

[00128] The load cell 972 in this example is a wired load cell that is connected via electrical connector 973 to the controller 943. The load cell 972 may alternatively be wireless and may incorporate a battery and a wireless transmitter for transmitting signals wirelessly to a receiver on the controller 943.

[00129] In the embodiment shown in Figure 33, the load cell 972 is positioned on the first housing 914 instead of being positioned on the first piston 916.

[00130] Referring to Figure 34, a hub load sensor 974 may be provided in any idler pulley that is driven by the belt 911 , such as, for example, the tensioner pulley 938 to determine the hub load incurred by that pulley. The hub load sensor 974 may have any suitable structure. For example, it may incorporate a compression-type strain gauge that acts between an inner race of a pulley bearing and a hub, as described in US Patent No. 6,484,593. The hub load sensor 974 may be used to detect the force of the belt 91 1 on the pulley 938 and so it constitutes a relatively direct way of determining the tension in the belt 91 1. It will be understood that the belt tension can be determined based on the sensed hub load and based on other data indicative of the geometric relationship between the belt spans 91 1a and 91 1 b on either side of the pulley 938 and the direction of the sensed hub load. The belt tension could alternatively be estimated using only the hub load, without other data indicative of the geometric relationship between the belt spans 91 1a and 91 1 b. [00131] As shown in Figure 34, the hub load sensor 974 may be connected to the controller 943 via a hardwired connection. It is alternatively possible, however for the hub load sensor 974 to be powered by a battery and to have a wireless transmitter so as to transmit hub load information wirelessly to the controller 943. The hub load sensor 974 may employ a strain gauge, as noted above, or piezoelectric technology, or any other suitable technology.

[00132] Referring to Figure 35, another way of determining the force of the belt 91 1 on the pulley 938 is by means of a bending load sensor 976 provided on a hub 978 of the pulley 938, which determines the bending load applied by the belt 91 1 on the hub 978 through the pulley 938. The bending load sensor 976 may be any suitable type of sensor, such as a bending beam load cell. The controller 943 can use the bending load sensed by the bending load sensor 976 either alone, or in combination with data indicative of the geometric relationship between the belt spans 91 1a and 91 1 b on either side of the pulley 938 and the direction of the sensed bending load, to determine the belt tension. A suitable bending beam load cell may be the model LBB200 or LBB300 provided by Futek Advanced Sensor Technology, Inc. of Irvine, California, USA.

[00133] In similar fashion to the embodiments described above, the bending load sensor 938 may be connected to the controller 943 via a hardwired connection, as shown, or may transmit bending load information to the controller 943 wirelessly.

[00134] Referring to Figure 36, another way of determining the belt tension is to use a bending load sensor 980 on either (or both) of the cylinder pivot studs 919 or 921 , which senses the force exerted by the cylinder 912 on either the base structure 917a or on the lever arm 936 (it is the same force in both cases). In the embodiment shown, the bending load sensor 980 is provided on the stud 921. The controller 943 can determine the tension in the belt 91 1 using the load sensed by the bending load sensor 980 either alone, or combined with other data indicative of the geometric relationship between the W

- 58 - cylinder 912, the lever arm 936 and the hub load vector, shown at HLV, in similar fashion to the belt tension determination made using the fluid pressure in the cylinder 912. The bending load sensor 980 may be similar to the bending load sensor 976 and may be a bending beam load cell.

5 [00135] Another possible location for a bending beam load cell aside from the locations described above would be on the stud shown at 991 that is used to mount the lever arm 936 to the base structure 917b.

[00136] In embodiments wherein a bending beam load cell is used, or wherein a pulley hub load sensor is used, it will be noted that, due to movement

10 of the various components of the tensioner during use, the orientation of the sensor may not be aligned always with the direction of the load acting on it. The controller 943 may attempt to compensate for this in some way so as to obtain a more accurate determination of the load itself. For example, it may compensate for this by obtaining data from one or more of the position sensors

15 described above. Alternatively it may compensate for it by estimating the position of the components based on some other data, such as the force itself, or accelerometer data from an accelerometer or based on some other information. Alternatively, the controller 943 may simply use the load data from the load sensor without compensating for it, on the assumption that it provides

20 sufficient accuracy in some embodiments.

[00137] In embodiments wherein a hub load sensor or a bending beam load cell is used, it may be used on any pulley (preferably an idler pulley) in the accessory drive. However, providing one of these devices on the tensioner pulley 938 or on the cylinder 912 permits it to be oriented relative to the 25 tensioner arm 936 in a selected way so that it is in a selected orientation as soon as the tensioner is installed on the engine. Optionally, the hydraulic actuator 913, the biasing member 926, the tensioner arm 936 and the pulley 938, could be provided on a single support bracket (not shown) that is itself mountable in a predetermined orientation to the engine. This would permit a bending beam load cell mounted in the stud 919 to be already oriented in a selected orientation when the tensioner is installed on the engine.

[00138] In embodiments wherein load-related sensors are provided (e.g. the pressure sensor, the load cells, hub-load sensors, strain gauges and the like in the embodiments shown in Figures 31a-36), the sensors can be used to provide feedback to the controller 943 so that closed loop control can be carried out when the controller 943 is driving the motor 940 in order to achieve a selected belt tension. Any closed loop control scheme could be used, such as PID control, fuzzy logic, etc. Alternatively, the controller 943 may control the motor 940 using open loop control. In such a case it would simply drive the motor a selected number of turns according to a look up table stored in memory, to bring the second piston 924 to a selected position.

USE OF ACCELEROMETER

[00139] Referring to Figure 37, an accelerometer shown at 982 may be provided on a moving element in the tensioner. The accelerometer 982 may be similar to those provided in some video game controllers and smartphones today and is capable of detecting movement of an object. The accelerometer 982 may be positioned on the lever arm 936 (shown in this embodiment to have a different shape than the lever arm 936 in other embodiments herein, in order to accommodate the accelerometer 982). The accelerometer 982 could alternatively be positioned on some other member, such as on one of the first or second housings 914 or 920, or possibly even on the first piston 916 (so as to detect the angular movement of the piston 916 about the pivot axis ACYL). The accelerometer 982 provides acceleration data to the controller 943, which can be used to detect both the amplitude and the frequency of the changes in acceleration of the lever arm 936 (or of whatever part that it is mounted to). Rapid and/or large changes in acceleration may indicate instability (e.g. resonance) in the tensioner, and may thus trigger the controller 943 to take some suitable action, such as to raise the pressure in the cylinder 912 by driving the threaded rod 932 into the second piston 924 with sufficient force to restore stability to the tensioner. In an embodiment, the threaded rod 932 may be driven just into engagement and the motor 940, preventing the movement of the piston 924 in one direction, thereby preventing the compression of the cylinder 912.

[00140] In an embodiment, the accelerometer 982 may include or may send signals to a math processor or an FFT processor that would determine the amount of movement (i.e. displacement) of the lever arm 936 and/or the frequency of oscillation of the lever arm 936 and/or the acceleration of the lever arm 936. Such a math or FFT processor could send signals indicative of its determinations to the controller 943. MEMS chip sensor technologies from companies such as Analog Devices, Honeywell or Sentron could be used in precalibrated sensors and FFT processors similar to those used in some consumer devices such as tablet computers and smartphones.

[00141] The accelerometer 982 could also be used in combination with limit switches or some other position sensing technology to assist in the determination of the positions of the components of the tensioner, (i.e. the cylinder 912, and the lever arm 936). For example if a limit switch is provided on the lever arm 936, and if the controller 943 adjusts the pressure in the cylinder 912 when the vehicle is started up so that the lever arm 936 hits the limit switch, the position of the lever arm 936 can be determined thereafter (until the vehicle is turned off), based on data from the accelerometer 982.

[00142] As with the embodiments above, the accelerometer 982 may connect to the controller 943 via a hardwired connection as shown in Figure 37, or via a wireless connection.

[00143] The accelerometer may be configured to measure acceleration in three axes. The orientation of the three axes may be as shown in Figure 37, whereby one of the axes (in the example shown, the x-axis) is generally parallel to the hub load vector when the tensioner arm 936 is in the nominal position when the tensioner 910 is in use on a new belt 911 and the engine is operating in a selected operating condition (e.g. idling). As a result of the match in the orientation of one of the axes of the accelerometer 982 and the hub load vector HLV, the displacement measured by the accelerometer 982 on that axis is approximately directly proportional to the displacement of the pulley 938. This is simpler than having to calculate an overall displacement based on the vector sums of displacements along a plurality of axes, which is what would occur in embodiments where none of the axes were parallel to the hub load vector. In general having a three axis accelerometer 982 (regardless of which way the axes are oriented) also permits the controller 943 to detect situations wherein there is any displacement in a direction away from the plane in which the tensioner arm moves (i.e. along the z-axis in the embodiment shown in Figure 37), or more than a selected amount of displacement along the axis that is in- plane but is generally perpendicular to the hub load vector when the tensioner arm 936 is in the nominal position when the tensioner 910 is in use on a new belt 91 and the engine is operating in a selected operating condition (e.g. idling), (i.e. the y-axis in the exemplary embodiment shown in Figure 37). Such a situation can occur when there is wear or damage to one or both of the tensioner arm 936 and the pivot stud 923.

[00144] It would be possible to use the three-axis accelerometer described above, but with the axes in any orientation (i.e. such that none of the axes are aligned with the hub load vector). It would still be possible to obtain all the information that was described above but it would be more processor intensive to obtain it since the controller 943 would determine the movement of the tensioner arm 936 via a vector sum of displacements along two or potentially three axes.

[00145] In an alternative embodiment, however, it would be possible to provide an accelerometer 982 with only one axis. Such an accelerometer would be less expensive than a three-axis accelerometer, but would still be capable of providing displacement information regarding the movement of the tensioner arm that would be the same as the information obtained from measuring the x-axis of the three-axis accelerometer with its axes oriented as shown in Figure 37.

USE OF POSITION SENSORS

[00146] Several technologies could be used to provide position data for the components of the tensioner to the controller 943. As will be understood, providing the position of any one of the first piston 916, the first and second housings 914/920, or the lever arm 936 is sufficient for the controller to determine the positions of the other of the components. This information can be used by the controller 943 in combination with the force or pressure information determined above to determine the belt tension. It is alternatively possible to use the position information for other purposes. For example the position information can be used on its own to estimate (roughly) the belt tension. Additionally or alternatively, the position information can be used to detect both the amplitude and frequency of movement of the lever arm 936 during operation of the engine, which the controller 943 can use to detect when the tensioner is unstable, or when there is belt flutter or the like taking place.

[00147] An exemplary way of determining the position of the lever arm 936 is to use an absolute rotary position sensor 984 and a dipole magnet 985 that is disk-shaped with one hemi-disk having one polarity and the other hemi- disk having the opposite polarity, as shown in Figures 38a and 38b. The sensor 984 (Figure 38b) may be positioned to sense relative motion between itself and the dipole magnet 985. The sensor 984 in this embodiment positioned on a mounting bracket 986 that is mounted to the tensioner arm 936 and therefore rotates with the arm 936 about axis ALAP. The dipole magnet 985 may be mounted to the mounting bolt shown at 989. In the embodiment shown in Figure 38b, the dipole magnet 985 has an extension 987 that is fixedly received in an aperture 988 in the bolt 989, however it will be understood that the magnet 985 may be mounted to the bolt 989 or to any other suitable stationary base structure in any suitable way. A suitable sensor and magnet for use in this embodiment are shown and described in US Patent 7,188,021. When the tensioner arm 936 rotates, the sensor 984 rotates. This rotation changes the orientation of the sensor 984 relative to the magnet, which is sensed by the sensor 984 and is communicated to the controller 943. The controller 943 can use this information in combination with the force or pressure information from one of the other sensors described above, to determine the belt tension in belt 91 1 . Alternatively, the controller 943 could use the sensor information to determine the position of the pulley 938, which is indicative of the amount of stretch present in the belt 91 1. The amount of stretch in the belt 911 can be used to estimate the belt tension in the belt if certain material properties of the belt 91 1 such as its elastic modulus, are known.

[00148] In the embodiment shown in Figure 38b, the sensor 984 rotates through the angular range of the tensioner arm 936, while the magnet 985 is stationary. It is alternatively possible for the sensor 984 to be mounted stationary (e.g. by way of an extension on the sensor 984 that fits in an aperture 988 in bolt 989) and for the magnet 985 to rotate with the tensioner arm. The cable leading to the sensor 984 (shown at 979 in Figures 38a and 38c) may be routed so that it clears the bracket 986 so as to not obstruct movement of the tensioner arm 936 (Figure 38a) relative to the cable 991. Other arrangements for the sensor 984 and magnet 985 may be provided that are similar to the arrangements of sensor and magnet shown in Figures 4a- 2.

[00149] Instead of determining the angular position of the lever arm 936 for use in determining the belt tension, it would be possible to determine the belt tension using the length of the cylinder 912. In an alternative embodiment shown in Figure 39, a linear position sensor 990 is provided on the cylinder 912 for determining its length. The sensor 990 may be any suitable type of sensor, such as, for example, a linear variable differential transformer (LVDT), such as one that is provided by RDP Electrosense of Pottstown, Pennsylvania, USA. A first portion 991 of the LVDT is connected to the first piston 916 via bracket 993 and the second portion 992 of the LVDT is connected to the first housing 914 via bracket 995. As the cylinder 912 changes length, the interaction of the first and second portions 991 and 992 changes the voltage output from the LVDT. The LVDT may be connected to the controller 943 so that the controller 943 detects the change in voltage. Determining the length of the cylinder 912 permits the controller 943 to determine the position of the pulley 938 which in turn permits the controller 943 to estimate the belt tension based on the pulley position and the elastic modulus of the belt 91 1 in addition to numerous other system properties, including, but not limited to: other properties of the belt 911 , of the engine, and of the accessories driven by the belt 91 1. Alternatively, it is possible to determine the belt tension using the length of the cylinder 912 (to establish the positions of the cylinder 912 and the lever arm 936) and the information from one of the force or pressure sensors described above. Another example of a linear sensor that may be suitable for the embodiment shown in Figure 39 is a magnetorestrictive Temposonics™ sensor provided by MTS Sensors Group of Cary, NC, USA. Another example of a linear sensor that could be employed is a capacitive linear sensor, also referred to as a capacitive linear encoder.

[00150] Instead of mounting the linear sensor 990 to the first piston 916 and the first housing 914, it would alternatively be possible to mount the linear sensor 990 elsewhere on the base structure 917b and the lever arm 936 to measure the movement of the lever arm 936. In such an embodiment, the linear sensor could be fixedly mounted at its first and second ends to first and second portions of a telescoping structure. The first and second portions of the telescoping structure could be pivotally mounted at their distal ends to the base structure 917b and to the lever arm 936. Such an embodiment is shown in Figure 40. It will be noted that in the embodiment shown in Figure 40, the orientation of the linear sensor is generally parallel to the hub load vector HLV at least in the tensioner arm position shown, which is the nominal position of the tensioner arm 936 when the tensioner 910 is in use on a new belt 91 1 and the engine is operating in a selected operating condition (e.g. idling). As a result of the match in the orientation of the axis of the sensor 990 and the hub load vector, the displacement measured by the linear position sensor 990 is approximately directly proportional to the displacement of the pulley 938.

[00151] With reference to Figure 41 , it is possible to use a different type of displacement measurement device then the sensors shown in Figures 39 and 40. For example, a string potentiometer shown at 997 may be used. A string shown at 998 is extendable from a potentiometer body 999 (but is biased to withdraw into the body 999) and connects to some moving part of the tensioner 910, such as the tensioner arm 936. The connection itself may be provided by any suitable structure at the end of the string 998, such as a simple loop, a carabiner, a magnet, a hook or any other suitable structure. The body 999 is connected to a base structure 917c, which may be any suitable base structure. Movement of the tensioner arm 936 causes the string potentiometer 997 to rotate in one direction of the other. The string potentiometer 997 sends signals to the controller 943 indicative of the position of the tensioner arm 936. Using pulleys at selected positions, the string 998 could be routed through a complex path and so the potentiometer body 999 could be positioned quite remotely from the tensioner arm 936.

[00152] From the point of view of ease of installation, it is preferable to have a tensioner with relatively few components that require individual mounting to their associated base structure. For that reason, embodiments wherein the motor 940 is mounted to the end of the second housing 920 and where the second housing 920 is fixedly connected to the first housing 914 may be advantageous in many applications. In some embodiments, wherein the temperature in proximity to the engine 13 is very high, and/or when the there are extreme levels of vibration that will be incurred by the tensioner (e.g. from torsional vibration from the crankshaft pulley 12, or from vibration or shaking of the engine 13 itself), it may be advantageous to position the motor 940 away from the rest of the tensioner 910 where it can be thermally isolated from the engine 13 and where it is physically isolated with respect to vibration from the rest of the tensioner 910. With reference to Figure 43, to provide this aforementioned thermal and physical isolation, the second housing 920 may be connected to the first housing 914 via an intermediate conduit shown at 1 100 that is flexible. This permits the second housing 920 to be mounted fixedly to a base structure 917d and to remain fixed in position there through the range of movement of the cylinder 912. The intermediate conduit 1 100 may be made from any suitable material that is at least somewhat flexible to permit relative movement between the first and second housings 914 and 920. The intermediate conduit 1 100 may have one or more inflexible segments and one or more flexible segments which provide sufficient flexibility to the intermediate conduit 1 10 that the aforementioned relative movement between the first and second housings 914 and 920 can take place. The inflexible segments may be made from a suitable metal or any other suitable material. The flexible segments may be made from hose or any other suitable material that can provide sufficient flexibility to permit movement of the first housing 914 throughout its range of movement between the load stop position and the free arm stop position. Whatever portions of the intermediate conduit 1 100 are flexible may be made from rubber or any other suitable polymeric material. Other examples of suitable material for the conduit 1100 would be a high pressure hydraulic thermoplastic hose.

[00153] The intermediate conduit 1 100 may be made from a material that flexes in length and/or in diameter when pressurized. In such situations, if steps to prevent such expansion were not taken, the conduit 1 100 would have a certain level of spring compliance, damping, and may cause a delay in the response of the system to pressurization, each time the hose is initially pressurized from a rest state. To reduce (and optionally to effectively eliminate) the amount of expansion that the conduit 1 100 may include an outermost layer that is a braided steel, braided carbon fibre, or even braided fabric. The amount of elastic compliance available in the conduit 1 100 can be selected. Computer modeling or any other suitable means can be used to assist in this process. The flexibility of the conduit 1100 can be selected (specifically, properties such as minimum bending radius, collapsibility, stiffness and other parameters) based on the specific details of the application (e.g. the routing of the conduit, the temperatures it will be exposed to, etc).

[00154] The ends of the conduit 1100 may connect to the first and second housings 914 and 920 by any suitable type of connection. For example, the connections may be threaded connections with suitable seals provided. A suitable locking material can be used (e.g. LOKTITE™ ) to prevent the conduit 1 100 from loosening or separating from the housings 914 and 920. Alternatively, the ends of the conduit 1 100 may be inserted into the housings 914 and 920 and may connect thereto via a threaded compression coupling that incorporates a conical collar that is threaded into position and compresses and locks the conduit 1 100 into place. Alternatively, the conduit 1 100 may be crimped or staked at their connections to the housings 914 and 920. In embodiments wherein the conduit 1 100 has metallic (e.g. brass) end fittings, the connections may be soldered, brazed or welded, e.g. using laser welding techniques. Such connections could also be used where segments of the conduit 1 100 join other segments of the conduit 1 100.

[00155] It would be preferable for any stresses that are associated with the flexing of the conduit 1100 to be removed from the connections between the ends of the conduit 1100 and the housings 914 and 920 (and from connections between different segments of the conduit 1 100 if there are any). To accomplish this, strain relief structures may be provided across any connection points to inhibit flexing from taking place at the connection points themselves. Such strain relief structures would ensure that substantially all of the flexing in the conduit takes place away from the connection points. A strain relief structure could be a simple rigid member (e.g. a metal bar or sleeve) that is connected at one end to one of the housings 914 or 920 away from the associated connection, and at the other end to the conduit 1 100 also away from the associated connection. The rigidity of the strain relief structures thus prevents flexing of the conduit 1 100 at the connection. Any other suitable strain relief structures for protecting the connections may alternatively be employed, such as a sleeve made from a non-rigid material such as rubber or some other polymer.

[00156] In order to guide which portions of the conduit 1 100 flex during operation of the tensioner 910 certain portions of the conduit 100 may be mounted fixedly to a base structure shown at 917e.

[00157] The mounting of the second housing 920 (and the motor 940) may be by simple mounting lugs as shown, without the need for elastomeric vibration isolation members or the like, particularly where the base structure being mounted to is some element such as a portion of the vehicle chassis that is isolated from the engine 13. If desired, however, isolation members, such as elastomeric members or springs can be provided in the mounting structure of the second housing 920 to help keep the motor 940 isolated from vibration and the like. The mounting structure used for the second housing (and the motor 940 if the motor 940 has any direct contact with the base structure 917d) can additionally or alternatively incorporate thermal isolation elements to inhibit the transfer of heat into the motor 940 through the base structure 917d.

[00158] Several benefits arise from this configuration. As a result of this configuration, the motor 940 can be directly mounted to the second housing 920 for driving the threaded rod 932 (not shown in this figure) without exposing the motor 940 to the movements and vibrations that may be incurred by the cylinder 912 during operation of the tensioner 910 and which could reduce the operating life of the motor 940. Additionally, because the motor 940 is physically separated from the engine 13, a thermal barrier shown at 1 102 can be provided and positioned between the motor 940 and the engine 13. This permits the motor 940 to be kept cooler than it might be if it were in closer proximity to the engine 13 as could be the case with embodiments in which it is mounted directly to the end of the second housing 920 and the second housing 920 is fixedly connected to (and proximate to) the first housing 912 as shown in Figures 31a-41. Keeping the motor 940 cooler can increase the operating life of the motor 940 and can reduce the potential for certain kinds of failure such as failure resulting from thermal shock or the like. Additionally or alternatively, keeping the motor 940 cooler permits the use of a less expensive motor 940. It will be understood that all the advantages described for the motor 940 are applicable to the gearbox 941 . The thermal barrier 1 102 is made from a thermally insulative material and thus inhibits the transmission of heat from the engine 13 to the motor 940. Suitable materials for the thermal barrier 1 102 include thin foil sheets which can reflect incoming radiated heat away from the motor 940. Other suitable materials include, for example, fibrous material with low thermal conductivity. Additionally or alternatively, the thermal barrier 1 102 may include a hard shell made from a suitable material having low thermal conductivity. The thermal barrier may be in the form of a wall, as shown, a blanket that fully or partially envelops the motor 940 and optionally the second housing 920, or a rigid enclosure that fully or partially envelops the motor 940 and optionally the second housing 920. In embodiments where a blanket or an enclosure is provided, cool air from some source, such as a duct from behind the radiator, shown at 1 104, could be provided to direct cool air to the motor 940.

[00159] To further assist in keeping the temperature of the motor 940 below a selected level, the motor 940 may be positioned behind the vehicle's radiator 1 104 and fan (shown at 1106), as shown in Figure 43. Air flow from the radiator 1104 and fan 1 106 could thus help to carry heat away from the motor 940.

[00160] A coating may be applied to any metal or polymeric structure to assist in inhibiting the conduction of heat from that structure towards the motor 940. An example of such a coating is provided by Techline Coatings of Midlothian, Texas, USA under the names CermaKrome™, PolyPhen™ and PowerKote™.

[00161] Another technology that could be employed to protect the motor 940 could be in the form of a thermoelectric member as the thermal barrier 1 102. A thermoelectric member, such as that which is provided by Tellurex Corporation of Michigan, USA, is powered electrically to transfer heat from one side (which may be referred to as the cool side), through an array of semiconductor pellets to the other side where the heat is released (which may be referred to as the hot side). The cool side of the thermoelectric member would be the side facing the motor 940, and the hot side is the side facing away from the motor 940. Optionally cooling fins could optionally be provided on the hot side. Optionally air flow can be provided through the fins to remove heat therefrom. Air flow to the thermoelectric member can be provided by any suitable means, such as by a fan or by positioning the thermal barrier 1 102 in the path of air entering the engine area of the vehicle while the vehicle is being driven. Instead of cooling the hot side by providing an air flow over it, it may be possible to cool the hot side by positioning some other member against the hot side so as to permit heat to be conducted from the hot side into the other member.

[00162] Some thermoelectric members are capable of generating electricity from the temperature gradient that exists across them in use. The thermoelectric member could in some embodiments be powered at least in part by the heat coming from the engine 13 itself so as to reduce the cost of its operation.

[00163] In some embodiments, it may be possible to apply thermoelectric members in the form of tiles or the like directly to the exterior of the motor 940 (e.g. the motor housing) so as to provide more direct heat transfer from the motor 940 into and through the thermoelectric member. As an alternative, it may be possible to form (e.g. by molding or the like), a single thermoelectric member that is shaped to snugly receive the motor 940 therein thereby providing good heat transfer from multiple faces of the exterior of the motor 940 into and through the thermoelectric member.

[00164] While the thermoelectric member has been described above as being used to cool the motor 940 in the embodiment shown in Figure 43 wherein the second housing 920 and the motor 940 are remote from the first housing 914, it is optionally possible for the thermoelectric member to be applied to the motor 940 in the form of tiles or in the form of a single, molded member that at least partially surrounds the motor 940 in embodiments wherein the motor 940 is mounted to the end of the second housing 920 and the second housing is fixedly connected to the first housing 914 as shown, for example in Figures 31a-41.

[00165] In an embodiment, the motor 940 itself may be used to determine the position of the pulley 938. For example, the motor 940 (or some element that is rotated by the motor 940, such as an element in the gearbox 941 , or the threaded rod 932) may incorporate a turn counter which may be a magnet and a sensor (i.e. an encoder) that can send a signal to the controller 943 indicative of how many rotations the motor 940 has undergone. If the controller 943 is capable of retaining in memory the true position of the motor 940 after the vehicle is turned off, then the controller 943 is capable of determining the absolute (or true) position of the motor 940 during use. Alternatively if the controller 943 is not capable of retaining in memory the true position of the motor 940 after the vehicle is turned off, then the controller 943 is capable of determining the incremental or relative motor position.

[00166] In any embodiment where the position of the tensioner pulley 938 is being used to determine the belt tension, a relationship needs to be determined between pulley position and the belt tension needs to be established during tensioner development. An example of this relationship is shown in the graph 1000 shown in Figure 42. The graph 1000 shows tensioner arm angular position vs. belt tension for a particular tensioner configuration. The tensioner arm angular position is shown in three different scales, all of which are in degrees but which have different zero-points. The vertical line 1002 at 37.5 degrees shows the nominal position of the tensioner arm when the engine 13, belt 91 1 and tensioner 910 are manufactured precisely to their specified dimensions and have precisely their specified properties. The vertical line 1004 at 44.8 degrees shows the equivalent position of the tensioner arm when the engine 13 is manufactured at the high end of its tolerance range and the belt 911 is manufactured at the low end of its tolerance range (i.e. long engine, short belt). The vertical line 1006 at 29.8 degrees shows the equivalent position of the tensioner arm when the engine 13 is manufactured at the low end of its tolerance range and the belt 911 is manufactured at the high end of its tolerance range (i.e. short engine, long belt). A curve shown at 1008 represents the relationship of the belt tensioner to the tensioner arm's angular position. Three curves 1008 are shown in Figure 42. The curve 1008b is the curve that is applicable when the tensioner biasing member 926 is manufactured precisely to its specifications of length and biasing force. The curve 1008a is the curve that is applicable when the tensioner biasing member 926 is manufactured to one end of its tolerance range in terms of dimensions and other properties such that it applies a higher biasing force than nominal when the tensioner arm 936 is at any given angle. The curve 1008c is applicable when a tensioner biasing member 926 is manufactured at the other end of its tolerance range in terms of dimensions and other properties such that it applies a lower biasing force than nominal when the tensioner arm 936 is at any given angle. Thus, if there is no calibration of the tensioner 910 to determine the particular curve that is applicable, then it can be seen that there is a range of belt tensions that could exist for a given tensioner arm position, due to tolerances in the various components during manufacture. In other words, if there is no calibration, then the belt tension would only be known with a potential error (such as, for example, +/- 15% as shown in the graph 1000) for a given position of the tensioner arm 936. If there is no calibration of the position sensing device used in the tensioner 910, the controller may assume that the tensioner arm 936 is at the nominal position when initially installed, and so it will be noted that the position of the tensioner arm 936 will be known with a potential error (such as, for example, about 7.5 degrees). CALIBRATION OF THE TENSIONER

[00167] In an embodiment, the tensioner 910 is capable of calibrating itself upon initial installation in the vehicle (optionally before there is a belt on the engine or before there is any tension in the belt). For example, the controller 943 may drive the motor 940 until the tensioner arm 936 reaches one of its end positions (i.e. its load stop position or its free arm stop position). The controller 943 may detect the tensioner 910 reaching the end position by any suitable means (e.g. by providing limit switches at the end positions, or by monitoring the current to the motor 940 so as to detect when the motor 940 has stalled). Upon determining that the tensioner 910 has reached an end position, the controller 943 may consider this to be a 'home' position. The controller 943 may then drive the motor 940 in the opposite direction until the tensioner 910 reaches the other end position. When moving the tensioner arm 936 to the load-stop position, a suitable tool (not shown) could be attached to the tool receiving feature shown at 1001 and the tool could be used to drive the tensioner arm 936 to its load stop position against the bias of the biasing member 926. The controller 943 may note the number of rotations of the motor 940 required to reach the other end position, and can thereby tell the position of the tensioner arm 936 at any time by tracking the number of rotations it has carried out in either direction. This is known as a two-point calibration. It is alternatively possible to calibrate the tensioner 910 using one or more other known positions (e.g. using additional limit switches at selected points along the path of motion of the tensioner arm 936) to further improve the accuracy of the calibration.

[00168] The tensioner 910 could alternatively be calibrated using a single point calibration, wherein the tensioner arm 936 is driven by the motor 940 to one end of its travel (e.g. the free-arm stop position). The controller 943 in such an instance would be able to determine the position of the tensioner arm 936 by tracking the number of motor turns that the motor 940 has undergone in each rotational direction, if the angular distance of the tensioner arm per turn of the motor 940 is known. While the turns of the motor 940 are described herein, it will be understood that the controller 943 could track the number of turns of any element that is connected directly or indirectly to the motor 940, such as the threaded rod 932, or some element in the gearbox 941 , as noted above. It will be noted that some motors are not capable of recalling their exact position after power-down and would require recalibration upon power-up. Other motors are capable of recalling their exact position even after power-down and subsequent power-up in which case recalibration would not be needed.

[00169] - With any of the aforementioned types of calibration it is possible to reduce the potential error in the determined position of the tensioner arm 936. If the tensioner 910 is calibrated in terms of position vs. belt tension during engine or vehicle manufacture, then it would also be possible to reduce the potential error in the determined belt tension based on the determined position.

[00170] When the controller 943 determines the belt tension it may do so either by calculation, or more preferably it would do so via a look up table based on the determined position and possibly based on other factors, such as, for example, temperature.

[00171] In an embodiment, the controller 943 may be programmed to determine belt tension (at least approximately) from the sensed motor current, since the motor current is, in at least some situations, proportional to the amount of force being exerted by the motor 940, which is proportional to the amount of force applied by the belt on the pulley 938, which is proportional to the belt tension. In this embodiment, belt tension could be determined without such elements as the pressure sensor, any load cells or strain gauges or the like. It will be understood that using the motor current may be relatively imprecise as compared to some of the other systems and methods described herein for determining belt tension. [00172] Stall detection on the motor 940 may be beneficial for other purposes than determining when the tensioner has reached an end position. For example, if the controller 943 determines that the motor 940 is stalled when the controller 943 determines that the tensioner has not yet reached an end position, the controller 943 may be programmed to stop the motor 940 to protect the motor 940 from damage and may indicate that there is a problem to the vehicle driver.

[00173] In the embodiments described herein, if the controller 943 has brought the tensioner to a high-tension setting, it could return the tensioner to a low-tension setting based on sensor input, or based on other criteria, such as the passage of a pre-determined amount of time.

[00174] In the embodiments wherein the vehicle is a hybrid vehicle and the tensioner 910 is used, the MGU 16 may be used in place of an alternator, or alternatively an alternator 16 and a separate motor (not shown) may be used. The MGU 16 will be discussed herein for simplicity, however, it will be understood that in any case where the MGU 16 is described, a motor and a separate alternator could be used.

[00175] The MGU 16 may have any one or more of several functions. One function, as described above, is to operate the MGU 16 as a motor and to use it to start the engine 13 (Figure 1 a) (which is a function referred to above as a Belt Alternator Start function). Another function would be to use the MGU as a motor to drive selected accessories (e.g. the a/c compressor shown at 18) when the engine 13 is off (e.g. when the vehicle is stopped at a stoplight), as described above. Yet another function may be for torque smoothing, wherein the MGU 16 is operated in such a way as to cancel out at least some torsional vibration that is exerted on the belt 91 1 by the crankshaft pulley 12. In cases where the device 16 is an MGU, it can be used to dynamically cycle the MGU pulley clockwise and counterclockwise in order to phase-match and to hence cancel out a belt drive resonance event, torsional pulley vibration, and / or belt flutter event. Alternatively if the device 16 is an alternator it can still be cycled on and off or between high and low loads to at least partially cancel out one of the aforementioned situations. Such use is described in US Patent No. 6,998,823, which is incorporated herein by reference. Yet another function for the MGU 16 is for regenerative braking (an alternator could also be used for this purpose). Yet another function is for launch assist, whereby the MGU 16 is operated as a motor and drives the belt 91 while the engine 13 also drives the belt 91 1. The MGU 16 thus at least reduces the effort that the engine 13 has to use to drive the accessories, and may itself drive the crankshaft pulley 12 through the belt 91 1 . In either case, the MGU 16 provides a boost in torque. In some of these situations, such as, for example, when the MGU 16 is being used as a (BAS) Belt Alternator Start device, it may be desirable for the tensioner to increase the belt tension to a high setting to reduce the likelihood of belt slip. The controller 943 may control the input voltage to the motor 940 directly from the on and off signals to the MGU 16 such that when the MGU 16 is operated as a motor, the controller 943 drives the tensioner 910 to the high tensioner setting and when the MGU 16 returns to operation as a generator, the controller 943 brings the tensioner 9 0 to the low tension setting.

[00176] Alternatively, when the MGU 16 is being operated as a motor, the controller 943 may be programmed to keep the belt tension relatively low to help reduce the parasitic losses incurred when it is used to drive any accessories. When it is to be used to start the engine 13, the MGU 16 can be used first to drive up the belt tension, so that when the MGU 16 is driven to start rotation of the crankshaft pulley 12, belt slip is unlikely to occur. It will be noted that when the engine 13 is stopped, the amount of resistance of the crankshaft pulley 12 is quite high, and as a result the torque required to drive the crankshaft pulley 12 can be relatively high. In order to describe the driving up of the belt tension, the belt spans are identified in Figure 15. The belt spans are shown at S1 to S5. It will be noted that, during normal operation of the engine 13 (Figure 1 ) the crankshaft pulley 12 is driving the belt 91 1 and is rotating clockwise in the view shown in Figure 15. The belt span S1 may be referred to as the belt span leaving the crankshaft pulley 12. The belt span S2 may be referred to as the belt span leaving the accessory pulley 915 (e.g. water pump pulley 915), or may be referred to as the belt span leading to the tensioner pulley 938. The belt span S3 may be referred to as the belt span leaving the tensioner pulley 938 or may be referred to as the belt span leading to the pulley 60 for the MGU 16 or for the alternator 16. The belt span S4 may be referred to as the belt span leaving the alternator 16 or may be referred to as the belt span leading to the a/c compressor pulley 60. The belt span S5 may be referred to as the belt span leading to the crankshaft pulley 12.

[00177] When the crankshaft is driving the belt 91 1 the spans S4 and S5 may be considered to be tight and the spans S1 , S2 and S3 may be considered relatively slack (compared to spans S4 and S5). However, when the MGU 16 is operated as a motor, belt spans S4 and S5 are slack and spans S1 , S2 and S3 are relatively tight. In order to drive up the belt tension, the MGU 16 may be driven briefly in a direction (counterclockwise in the view shown in Figure 15) opposite to its normal direction of rotation. This increases the tension in the belt spans S4 and S5 and drives slackness in the belt 91 1 into the spans S3 and S2. As soon as the belt spans S2 and S3 slacken the tensioner biasing member 926 drives the tensioner pulley 938 into the belt 91 1 to take up the slack. To ensure that the tensioner 910 holds the belt at the level of tension that it has at that moment, the motor 940 is commanded by the controller 943 to drive the threaded rod 932 up to the floating piston 924 so that it abuts it, thereby preventing the cylinder 912 from expanding and preventing the belt 91 1 from pulling the tensioner pulley 938 during normal (clockwise in the figure) rotation of the MGU 16. Once the rod 932 is abutted with the floating piston 924, the MGU 16 can be driven in the clockwise direction briefly to build slack into the belt spans S4 and S5. The MGU 16 can then be driven again in the counterclockwise direction to try to bring this slack from spans S4 and S5 into the spans S2 and S3, where again the slack will be taken up by movement of the tensioner pulley 938 under the bias of the biasing member 926. This movement of the pulley 938 will be accompanied by an extension of the cylinder 912 and consequent movement of the floating piston 924 away slightly from the end of the threaded rod 932. Again the threaded rod 932 will be driven back into abutment with the floating piston 924, locking the piston 924 and therefore preventing the belt 911 from pulling the pulley 938 in the load stop direction (i.e. clockwise in Figure 15). This reciprocation of the MGU 16 can continue for a selected number of cycles, or for a selected period of time or until a selected belt tension is reached. Once this is completed, the belt 91 1 is ready to be used to drive the crankshaft pulley 12 to start the engine 13. In a preferred embodiment, this cycling of the MGU can take place as soon as or shorter after the engine 13 is shut off as the vehicle is coming to a stop (e.g. at a stoplight or at a stop sign). As a result, the vehicle is ready to start up the engine 13 as soon as the driver indicates to the vehicle that he/she wants it. By readying the belt 91 1 to start up the engine 13 relatively quickly after shutting down the engine during a stop, can be valuable in the event that the driver needs to accelerate away quickly, such as in an emergency situation in which the driver needs to avoid a collision from another vehicle. In an alternative embodiment, the belt tension may be kept relatively low for a period while the vehicle is stopped to reduce the amount of energy that is expended by the MGU 16 to drive the accessories such as the a/c compressor 18 while the vehicle is stopped at a stoplight and the engine 13 is off. At some point, such as when the driver depresses the accelerator pedal to pull away from the stoplight, the MGU 16 can be cycled back and forth as indicated above so as to drive up the belt tension. The cycling of the MGU 16 back and forth as indicated above may be referred to as 'feathering'. The belt tension achieved during this process may be as high as 2000 N, or even higher for a brief moment when the MGU 16 is driving the crankshaft pulley 12.

[00178] When driving the belt tension up to high levels, such as when feathering the MGU to prepare the belt 91 1 for driving the crankshaft pulley 12, it may be preferable to use load sensing or pressure sensing devices instead of position sensing devices, since load sensing devices are a relatively more direct means of determining the belt tension and are, in that sense, a more accurate means of determining belt tension. In general when driving the tension in the belt up to high levels, it is beneficial for the controller 943 to be aware of the loads in the tensioner 910 particularly in embodiments where there is no spring compliance (such as the biasing member 928 shown in the embodiment in Figure 26, or such as the use of a conduit that is elastically expandable in the embodiment shown in Figure 43). Without spring compliance, very small linear movement of the threaded rod 932 into the second piston 924 could cause very large changes in belt tension and in the loads that are applied to pulleys and the like in the tensioner 910 and in the accessory drive system in general. As a result, when driving up the belt tension, the movement of the threaded rod 932 is preferably precisely controlled and is controlled with good feedback on the loads in the tensioner 910 in order to inhibit damage to tensioner and accessory drive components.

[00179] For greater certainty, it will be understood that in at least some embodiments, and even some embodiments wherein the vehicle is a hybrid vehicle, the device 16 may be an alternator instead of an MGU.

[00180] While the term 'belt' has been used in the description herein, it will be understood that the tensioning system of the present invention can apply to many different kinds of endless drive element, both synchronous (e.g. toothed) and asynchronous (e.g. non-toothed), including accessory drive belts, timing belts, timing chains and the like.

[00181] In the embodiments shown in Figures 15-43, it will be noted that the particular arrangement of the hydraulic actuator 913, the tensioner arm 936, the pulley 938 could be different for different applications, while still using some means (such as the motor 940 and the threaded rod 932) for selectively preventing or limiting the compression of the cylinder 912 by preventing movement in a selected direction of the floating piston 924, and/or while still using some means (such as the motor 940 and the threaded rod 932) for driving the cylinder 912 to extend thereby driving the pulley 938 into the belt 91 1 to increase the belt tension.

[00182] In the embodiments shown in Figures 1 -14b sensors and the like are described for detecting certain properties and conditions. Additionally, several control schemes are described based on several conditions and signals from the described sensors. For greater certainty, it will be noted that the control schemes and the sensors described in relation to the embodiments shown in Figures 1-14b are applicable to the embodiments shown in Figures 15-43.

[00183] For greater certainty, anywhere in this disclosure where a belt is described or shown, the belt may alternatively be any endless drive member, whether synchronous (e.g. toothed) or asynchronous (non-toothed).

[00184] While the description contained herein constitutes a plurality of embodiments of the present invention, it will be appreciated that the present invention is susceptible to further modification and change without departing from the fair meaning of the accompanying claims.