Belt retractor unit and vehicle having such belt retractor unit

Abstract

A belt retractor unit comprising a housing (12), a rotatably mounted belt reel (20), a blocking unit for blocking the belt reel (20), the blocking unit having an electrically controllable actuator unit (40), and a control device (50) for controlling the actuator unit (40), the control device having a sensor device (60) for measuring at least one horizontal acceleration value and a switching device (70) for controlling the actuator unit (40) in response to at least the at least one horizontal acceleration value, the belt retractor unit being adapted to detect a horizontal acceleration value (|a XY |) is within the first limit (a h 1) and the belt reel (20) is not blocked, and the horizontal acceleration value (|a XY |) is within the first limit (a h 1) and the belt reel (20) has an unblocked state in which the belt retractor (20) is not blocked, and the belt retractor (20) has a horizontal acceleration value (|a XY The magnitude of |) is the second limit (a h 2) A belt retractor unit that will not transition to the non-blocking state unless the load falls below this level.

Description

[Technical Field]

[0001] The present invention relates to a belt retractor unit as described in the preamble of claim 1. [Overview of the Initiative] [Problems that the invention aims to solve]

[0002] All modern passenger cars, as well as most trucks and buses, have seat belt systems. Such safety belt systems always have a belt retractor unit, which has a belt retractor comprising a housing and a belt reel rotatably mounted within the housing. A portion of the safety belt system strap is wound onto this belt reel, and the user can unwind the safety belt system strap from the belt reel against the force of a return spring acting between the belt reel and the housing. Further provided are blocking devices having an open state in which the belt spool is not blocked against the housing and a blocked state in which the belt spool is blocked against the housing. This blocking device typically has two independent sensors: a belt sensing sensor that senses the rotation of the belt reel and a vehicle sensing sensor that senses the vehicle position and / or vehicle acceleration (particularly negative vehicle acceleration, i.e., deceleration). Under normal driving conditions, i.e., when the strap is not stretched too quickly and the vehicle is not in an abnormal position or experiencing abnormal acceleration, the blocking device is in its unblocked state, and the user can stretch the strap, allowing for relatively free movement. However, if, for example, the strap is extended too quickly and / or the vehicle decelerates too quickly and / or the vehicle tilts too much from its normal orientation, the blocking device will enter its blocking state.

[0003] Currently, blocking devices are designed almost entirely mechanically, meaning that the complete blocking device (including sensors) is rigidly connected to the belt retractor housing. However, this rigid connection of the entire blocking device to the housing has drawbacks, particularly when the belt retractor unit is fixed to the vehicle seat, especially its backrest, as the position of the belt retractor, and therefore the position of the vehicle sensing sensors, can change relative to the vehicle.

[0004] For this reason, belt retractor units that are fully or partially electrically operated are known, and their blocking devices have a blocking unit connected to a housing having an electromagnet and a control device for controlling the electromagnet. This electromagnet is part of an electrically controllable actuator unit, and the state of the actuator unit (particularly no current or energized) determines the state of the blocking unit (the belt coil is rotatable or the belt coil is blocked). Typically, the blocking unit also has a return element (usually in the form of a spring) that opposes the electromagnet. This reset element may also be part of the actuator unit. For safety reasons (fail-safe), typically the no-current state is a blocked state (emergency state), and the energized state is an unlocked state (enabled state). Thus, a control device that drives an actuator unit having a power input connected to the vehicle's electrical system and a power output connected to the electromagnet has a passive switching state in which the power supply from the power input to the power output is interrupted, and an active switching state in which the power input is connected to the power output and therefore current flows through the electromagnet. This control device is generally electrically coupled to a blocking unit and can be positioned anywhere in the vehicle, particularly so as not to move with the backrest. The control device includes at least one sensor device and a switching device for controlling an actuator unit in accordance with data supplied by the sensor device. The control device may or may not form a structural unit.

[0005] The most important criterion for transitioning from an enabled state to an emergency state is that the magnitude of at least one horizontal acceleration value exceeds a limit. In many cases, it is not necessary to distinguish between acceleration in the longitudinal direction (X direction) and acceleration in the lateral direction (Y direction) of the vehicle, and only the magnitude of acceleration in the XY plane is considered.

[0006] A general belt retractor unit having such a blocking unit and such a control device is described, for example, in British Patent No. 2398824B. Such electrically operated belt retractor units have other advantages as they offer more possibilities for controlling the state (blocked / unblocked) of the belt retractor unit.

[0007] Based on this, the object of the present invention is to further improve a general belt retractor unit.

[0008] This problem is solved by a belt retractor unit having the features described in claim 1.

[0009] To protect occupants as much as possible, it is generally desirable that the seatbelt locking system activates early, especially before an actual accident occurs, for example, when the vehicle decelerates rapidly. For this purpose, the limit value of the amount of horizontal acceleration beyond which webbing locking begins should be relatively low. However, this results in the belt retractor unit switching to its blocked state (hereinafter also referred to as the emergency state) relatively frequently, but without any danger whatsoever, i.e., the belt reel block is consequently unnecessary and is subsequently removed again. This in itself is not a major problem and is accepted for the overall improvement of occupant safety.

[0010] However, when a preferred limit value (i.e., a low limit value) is selected for the safety of vehicle occupants, it has been found that, especially in poor road conditions, the belt blocking system may switch to its emergency state, where the belt reel is blocked against the housing, simply due to the swaying motion of the vehicle. This is not a problem in itself. However, especially in the cases mentioned above, it has been found that the accelerating force acting on the vehicle (and therefore in the direction of the sensor) frequently oscillates near the selected limit value in poor road conditions, and as a result, the belt retractor unit may frequently and often switch between the emergency and released states. On the one hand, this is associated with the generation of intrusive noise and leads to increased wear on the mechanical components involved. Naturally, both are undesirable, and the increased noise negatively impacts comfort. This is all the more true because a particular advantage of a typical seat belt retractor unit is that the housing of the retractor unit is located behind the seat and therefore close to the occupant's ears. Since the expected lifespan of the belt retractor unit should be designed to match the expected lifespan of the entire vehicle, increased wear is naturally already a disadvantage.

[0011] To solve the problems described above, the control unit of the actuator unit by the control device is controlled in a hysteresis manner, that is, with two mutually distinct limits on the amount of horizontal acceleration: a first limit at which the blocking unit changes from its released state to an emergency state (hereinafter referred to as the first emergency state), and a second limit at which the belt retractor unit returns from its first emergency state to the released state, the second limit being smaller in magnitude than the first limit. This prevents constant switching between forward and backward by keeping the belt retractor unit in the emergency state for longer periods in appropriate situations, especially under appropriate road conditions. Overall, this is far more convenient for the user than permanently switching back and forth between the emergency state and the enabled state. Naturally, this also reduces the number of switching cycles, which has a positive effect on the service life of the belt retractor unit.

[0012] The above can be further improved in that the belt retractor unit switches from its emergency state to the release state only when the magnitude of at least one horizontal acceleration value falls below a second limit value and remains below a third limit value for a predetermined period, which may be at least 500 ms. This third limit value is preferably between the second limit value and the first limit value, and may be identical to the first limit value in particular.

[0013] As already mentioned, blocking of the belt retractor unit must occur both when sufficient acceleration (usually negative acceleration, i.e., deceleration) occurs and when the vehicle tilts more than a predetermined value around its longitudinal axis or its transverse axis. In such tilts, the vehicle's Z-axis is always tilted relative to the vertical, so the magnitude of the tilt can be expressed as a single angle. A single classical mechanical sensor cannot distinguish between these two cases. The situation is different when using electronic sensors, where acceleration and tilt can be distinguished. Since it is clearly important that the belt retractor unit changes from the enabled state to the emergency state when both the acceleration limit and the tilt limit are exceeded, the sensor should measure both values. However, preferably, these values ​​are processed separately by the switching device, i.e., processed so that the tilt relative to the vertical axis is not considered when determining the first emergency state. This reduces the number of unnecessary transitions from the enabled state to the first emergency state, especially on rough roads, as it has been found that the effect of rough roads on acceleration in the XY plane is often relatively small, but the effect on vehicle tilt, especially around the X-axis, is often relatively large.

[0014] To compensate for not considering the inclination around the vertical axis when determining the first emergency condition, it is preferable that the belt retractor unit has a second emergency condition where the belt reel is also blocked relative to the housing and depends only on the inclination.

Brief Description of the Drawings

[0015] Here, the present invention will be described in more detail by way of preferred embodiments with reference to the drawings. The drawings are as follows. [Figure 1] FIG. 1 is a schematic view of a belt retractor unit, in which the belt retractor of the belt retractor unit is shown in a schematic side view, the control device is in an open switching state, and the belt retractor is in a locked state. [Figure 2] FIG. 2 shows a schematic top view of the belt retractor shown in FIG. 1 as viewed from direction R1. [Figure 3] FIG. 3 shows the control device shown in FIG. 1, the control device is in its closed switching state, and the belt retractor is in its unlocked state. [Figure 4a] FIG. 4a shows a side view of a vehicle and corresponding coordinates. [Figure 4b] FIG. 4b shows a top view of the vehicle in FIG. 4a and corresponding coordinates. [Figure 4c] FIG. 4c shows a top view of the vehicle in FIG. 4a as viewed from the rear and corresponding coordinates. [Figure 5] FIG. 5 shows the vehicle in FIG. 4a during a frontal collision. [Figure 6] FIG. 6 shows the vehicle in FIG. 4b during a side collision. [Figure 7] FIG. 7 shows the vehicle in the representation of FIG. 4a when tilted around the Y-axis. [Figure 8] FIG. 8 shows the vehicle in the representation of FIG. 4c when tilted around the X-axis. <​​​​​​​​​​​ [Figure 10] Figure 10 shows a flowchart of a first preferred control sequence for the logic unit. [Figure 11a] Figure 11a shows a schematic diagram of the mechanical analogue of the left branch of the control sequence shown in Figure 10. [Figure 11b] Figure 11b shows a schematic diagram of the mechanical analogue of the left branch of the control sequence shown in Figure 10. [Figure 12] Figure 12 shows a flowchart of the second preferred control sequence for the first logic subunit. [Modes for carrying out the invention]

[0016] The essential features of the belt retractor unit according to the present invention will be described first with reference to Figures 1 and 2. It should be noted here that the representation is very schematic and represents only the basic principles of the present invention. The belt retractor unit can be thought of as comprising a belt retractor 10 and a control device 50. Here, the control device 50 may, but does not have to be, directly connected to the housing of the belt retractor 10, and therefore the control device 50 is shown separated from the housing in Figures 1 and 3. Naturally, the control device and the belt retractor must be electrically connected to each other.

[0017] In the following, conduits are shown only schematically (they are not shown together with the output and return conductors). Here, power lines are shown as solid lines, and signal lines are shown as lines with a "dotted-dot" pattern.

[0018] The belt retractor 10 comprises, as is typical, a housing 10, a belt reel 20 rotatably supported within the housing and around which a portion of the strap 5 is wound, and a blocking unit for blocking the belt reel 20 within the housing 12. In the illustrated embodiment, the housing 12 has two housing plates 14a, 14b connected by connecting bolts 16, which should be understood as illustrative only. In principle, also shown herein, the blocking unit has a locking wheel 22 connected in a manner rotatably fixed to the belt reel 20. Further provided are locking claws 24a that, in the locked state (emergency state) (Figure 1), lock the locking wheel 22, and thus the belt spool 20, against the housing 12, but do not lock in the released state (Figure 3).

[0019] Importantly, the position of the claw 24a is controlled directly (as shown) or indirectly by an actuator assembly 40 having an electromagnet 42. In the illustrated embodiment, this effect is brought about by the fact that the actuator unit 40 has, in addition to the electromagnet 42, a plunger 44 driven by the electromagnet, which acts on a lever 24 that supports the claw 24a. When a sufficiently strong current flows through the electromagnet, the electromagnet pushes the plunger 44 outward. However, it should be noted that, as already stated, this configuration should be understood as illustrative only. Basically, the blocking unit has an electromagnet, and the actuator unit controls the blocking unit depending on the current flowing through the magnet. Typically, this also means that a spring, in this case a tension spring 30, or another elastic element is provided that uniquely defines the state of the blocking unit when the electromagnet 42 of the actuator unit 40 is current-free and therefore exerts no force on the plunger driven by it. This current-free state is the locked state, as also shown in the figure.

[0020] Such electrically controlled blocking units having electromagnets are known in the prior art. Therefore, the present invention also relates only to the control of the electromagnet, i.e., the control device 50.

[0021] As preparation for a detailed description of the present invention based on a description of the control device and its operation, some general descriptions are given first, and definitions are given with reference to Figures 4a to 8.

[0022] Figure 4a shows a schematic side view of the vehicle and its associated coordinates, namely the longitudinal X and vertical Z directions; Figure 4b shows a top view of the vehicle shown in Figure 4a, also showing its associated coordinates, namely the longitudinal X and lateral Y directions; and Figure 4c shows a rear view of the vehicles in Figures 4a and 4b, also showing its associated coordinates, namely the lateral Y and vertical Z directions.

[0023] Figure 5 shows a typical head-on collision. Here, the velocity changes in the X direction, i.e., the second derivative of the X coordinate with respect to time is not equal to 0, and d 2 X / dt 2 ≠0, and the magnitude of the acceleration in the X direction, and therefore in the XY plane, is greater than 0, |a XY |>0. Therefore, in the case of a lateral impact (Figure 6), the second derivative of the Y coordinate with respect to time is not equal to 0:d 2 Y / dt 2 ≠0, here too, the magnitude of acceleration in the XY plane is greater than 0:|a XY |>0. This means that one criterion for locking the strap is that the amount of acceleration in the XY plane exceeds a predetermined limit.

[0024] Figures 7 and 8 illustrate different scenarios that should result in the strap locking, i.e., the vehicle tipping over. Vehicle tilt always means that the vehicle's Z-coordinate (referred to here as Z') is tilted relative to the vertical Z. The lock criterion here should be that the tilt, which can be expressed by the angle ΔZ contained between Z and Z', exceeds a predetermined value. Since the dynamics (i.e., rate of change or acceleration) and direction are not important here, the amount of ΔZ can be set as the limit for locking.

[0025] In the preferred embodiment described below, all relevant orientation and acceleration data are measured by a common sensor device, but processed separately by a switching device downstream of this sensor device. First, a schematic circuit diagram is used for explanation. It should be emphasized that this schematic diagram represents only one possibility. The corresponding separation of information can also be carried out differently with respect to acoustic technology or software.

[0026] Figure 9a shows a schematic embodiment of the control device 50. This control device 50 may be designed as a control unit including all the relevant components. However, although it is possible to "distribute" the individual elements of the control device throughout the vehicle, this is generally not preferred.

[0027] As already mentioned, the control device 50 is used to control the flow of current from the onboard power supply to the electromagnet 42, that is, to enable or interrupt it. Therefore, it has a power input section 55 connected to the vehicle's electrical system and a power output section 56 connected to the solenoid, with a switching unit 80 positioned between them. When the switching unit is closed, current flows from the vehicle's electrical system through the electromagnet. When the switching unit 80 is open, no current flows through the switching unit 80. It should also be noted that in this position, a DC-DC converter acting as a constant current source may be provided to supply the solenoid with as much current as the solenoid requires when the switching unit is closed. This may also be designed using two stages to provide inrush current and lower holding current. However, this is not shown in detail below.

[0028] In addition to the switching unit 80 described above, the control device 50 includes a sensor device 60 and a logic unit 74. The logic unit 74 and the switching unit 80 together form a switching device 70.

[0029] In the illustrated preferred embodiment, the sensor device 60 (which may also be simply called a sensor) is a three-axis sensor that continuously measures orientation and acceleration in all three spatial directions and relays the measurement data to the switching device 70. As described above, the switching device consists of at least a logic unit 74 and a switching unit 80 functionally.

[0030] The data provided by the sensor device 60 is processed by the logic unit 74 in two separate operations. In one process, only acceleration in the XY plane is processed, and in the other process, tilt around the Z axis is processed. Thus, the logic unit 74 has, at least functionally, a first logic subunit 74a that processes acceleration evaluation in the XY plane and a second logic subunit 74b that processes tilt evaluation. In the illustrated embodiment, these two logic subunits 74a and 74b control two switches 80a and 80b connected in series, and these switches 80a and 80b form a switching unit 80. This embodiment of the switching unit is chosen specifically for illustrative purposes, but in practice, it can also be designed in this way.

[0031] As described above, the current-free state of the solenoid 42 corresponds to the locked state of the belt retractor 10 (Figure 1), and the energized state corresponds to the unlocked state.

[0032] As described above, the logic unit 74 has a first logic subunit 74a that evaluates the acceleration value in the XY plane and a second logic subunit 74b that evaluates the incline value ΔZ. Naturally, the unlocked state (solenoid energized state) exists only when both logic subunits 74a and 74b result in "unlocked". In terms of the circuit, this is implemented here such that the first logic subunit 74a activates the first switch 80a of the switching unit 80, and the second logic subunit 74b activates the second switch 80b of the switching unit 80. Since these two switches 80a and 80b are connected in series, in order to generate the unlocked state (enable state), both switches 80a and 80b must be in the flow-through position, i.e., closed (Figure 8a). In all other cases (Figures 9b to 9d), a locked state (emergency state) exists. Thus, this circuit forms a logic AND with respect to the unlocked state.

[0033] Next, referring to FIG. 10, the first algorithm according to the present invention executed by the logic unit 74 will be described. As described above, the two logic sub-units 74a and 74b operate in parallel, preferably independently of each other. In the flowchart of FIG. 10, the left tree shows the flowchart of the first logic sub-unit 74a, and the right tree shows the flowchart of the second logic sub-unit 74b. The measurement variable related to the first logic sub-unit 74a is the magnitude of the acceleration in the XY plane, that is, the horizontal acceleration |a XY |. The procedure is as follows.

[0034] After starting the vehicle, the first step may be to check whether the amount of acceleration |a XY | in the XY plane is less than the first limit value a h 1. Since the vehicle should be in this state, for example, if the acceleration value is greater than this limit, it indicates that either a technical defect or a very abnormal situation exists, and an alarm can be triggered. On the other hand, if the magnitude of the acceleration value is less than the first limit value a h 1, the first logic sub-unit 74a controls the first switch 80a so that the first switch 80a switches to the closed state corresponding to FIG. 9a.

[0035] The above review is continuously executed. As long as the amount of acceleration in the XY plane does not exceed the first limit a h 1, the first switch 80a remains closed. However, when the first limit a h 1 is exceeded, the first switch 80a is opened, and a first emergency state exists, which means that due to the acceleration in the XY plane, the electromagnet  42 has no current, and thus the belt coil 20 is locked with respect to the housing 12. The steps according to the present invention are as follows. After the switch is opened, the first logic sub-unit 74a, here, checks whether the amount of acceleration |a XY | in the XY plane is below the second limit value a h 2, and this second limit value a h 2 is the first limit value a hIt is less than 1. Otherwise, the first switch 80a remains closed, and therefore the first emergency state is maintained. Magnitude of acceleration in the XY plane |a XY | Moko's second limit a h Only when the value falls below 2, the first switch 80a is closed again, and the first emergency state no longer exists. Preferably, the second limit a h 2 is the first limit a h It is between 20% and 80% of 1.

[0036] Mechanical analogues are shown in Figures 11a and 11b. To move the ball from its minimum position, the radial direction R is used. r The force acting on the ball must exceed a first limit so that the ball can climb the steeper incline. Next, it moves onto a flatter incline and remains there, applying a lower holding force in the radial direction. Only when this falls below a second limit, which is smaller than the first limit, can the ball return to its starting position.

[0037] In parallel with the above, the second logic subunit 74b permanently checks the vehicle's tilt amount |ΔZ|. Here again, it is possible to check whether |ΔZ| exceeds the limit value G for the first time after starting the vehicle. If this is already the case when the vehicle starts up, an abnormal condition or technical defect must be assumed. On the other hand, if |ΔZ| is less than the limit value G, the second switch 80b is also closed, and the belt retractor unit as a whole is in its released state (Figure 9a).

[0038] A check is also continuously performed to see if |ΔZ| exceeds the limit G. As long as this limit is not exceeded, the second switch 80b remains closed. If this limit is exceeded, the second switch 80b opens, and the belt retractor unit is in its second emergency state, as defined herein. Depending on whether the first emergency state is also present (or not), either the switching state in Figure 9c or the switching state in Figure 9d exists. However, due to the AND circuit, the enable state (Figure 9a) exists only when neither emergency state is present.

[0039] When |ΔZ| falls below the limit G again, the second switch closes again. Hysteresis is not necessary here, but it is possible.

[0040] Figure 12 shows the second embodiment of the flowchart on the left, i.e., the operation of the first logic subunit 54a.

[0041] As can be seen from the figure, the functional sequence is initially the same as described above, but the second limit a h After falling below 2, another process follows before the first switch 80a is closed again. In practice, the timer is started and the magnitude of acceleration |a in the XY plane is measured for a selected time interval Δt, for example, 500ms or 1,000ms. XY | is the third limit a h Whether or not it exceeds 3 is checked continuously. This third limit a h 2 is the first limit a h It may be the same as 1, but the first limit a h 1 and the second limit a h It may be located between 2 and 3. Third limit a h As soon as the corresponding excess of 3 is detected, the time interval starts again. Within a continuous time interval Δt, the third limit a h Only if the excess of 3 does not occur, the first switch is closed again, and as a result, the belt retractor unit is released from its first emergency state.

[0042] The above-described method achieves the avoidance of numerous switching cycles despite the low threshold for belt reel locking. [Explanation of Symbols]

[0043] 10 Belt Retractor 12 Housing 14a, b Housing plate 16 connecting bolts 18 Holder for spring and actuator unit 20 Belt Reels 22 Lock Wheels 24 Lever 24a Nail 30 Tension spring 40 Actuator Units 42 Electromagnet 44 plungers 50 Control device 55 Power Input Section 56 Power output section 60 Sensor device 70 Switching devices 74 Logical Units 74a First Logical Subunit (Acceleration Evaluation in the XY Plane) 74b Second logical subunit (slope evaluation ΔZ) 80 Switching Units 80a First switch 80b Second switch

Claims

1. A belt retractor unit, Housing (12) and A belt reel (20) is rotatably mounted inside the housing (12), A blocking unit for blocking the belt reel (20) from the housing (12), wherein the blocking unit has an electrically operable actuator unit (40), A control device (50) for operating the actuator unit (40) comprises a sensor device (60) for continuously measuring at least one horizontal acceleration value, and a switching device (70) for operating the actuator unit (40) according to at least one horizontal acceleration value, The belt retractor unit has at least one horizontal acceleration value (|a XY The magnitude of |) is the first limit value (a h 1 A first emergency condition in which the belt reel (20) is blocked from the housing (12) when the horizontal acceleration value (|a) exceeds the value of at least one horizontal acceleration value (|a) XY The magnitude of |) is the first limit value (a h 1 ) is below a certain level, and the belt reel (20) is in an open state where it is not blocked by the housing (12), The belt retractor unit starts measurement by a timer based on that the magnitude of the at least one horizontal acceleration value (|a XY |) is lower than a second limit value (a h 1 ) which is smaller than the first limit value (a h 2 ), and shifts from the first emergency state to the release state only when the magnitude of the at least one horizontal acceleration value (|aXY|) does not exceed a third limit value (ah3) during a predetermined period (Δt) during the measurement by the timer. A belt retractor unit characterized by this.

2. A belt retractor unit, Housing (12) and A belt reel (20) is rotatably mounted inside the housing (12), A blocking unit for blocking the belt reel (20) from the housing (12), wherein the blocking unit has an electrically operable actuator unit (40), A control device (50) for operating the actuator unit (40) comprises a sensor device (60) for continuously measuring at least one horizontal acceleration value, and a switching device (70) for operating the actuator unit (40) according to at least one horizontal acceleration value, The belt retractor unit has a first emergency state in which the magnitude of at least one horizontal acceleration value (|a XY|) exceeds a first limit value (a h 1) and the belt reel (20) is blocked against the housing (12), and a release state in which the magnitude of at least one horizontal acceleration value (|a XY|) falls below the first limit value (a h 1) and the belt reel (20) is not blocked against the housing (12). The belt retractor unit will not transition from the first emergency state to the release state unless the magnitude of at least one horizontal acceleration value (|a XY|) falls below a second limit value (a h 2) which is smaller than the first limit value (a h 1). The belt retractor unit is characterized in that the switching device (70) has two switching elements connected in series with each other, each of the two switching elements is controlled by an individual logic subunit, and is configured to generate the open state only when both of the two switching elements are in a conductive state.

3. Starting from the first emergency state, the at least one horizontal acceleration value (|a XY The magnitude of |) again exceeds the third limit value (a) for at least 250 ms. h 3 The belt retractor unit according to claim 1 or 2, characterized in that it changes to the released state only when it does not exceed ).

4. The third limit value (a h 3 ) is the second limit value (a h 2 ) is greater than the first limit value (a h 1 The belt retractor unit according to claim 3, characterized in that it is as follows:

5. The belt retractor unit according to any one of claims 1 to 4, wherein the sensor device (60) further measures the inclination (ΔZ) of the sensor device (60) with respect to the vertical axis and transmits it to the switching device (70), however, this does not take the inclination (ΔZ) into consideration when determining the first emergency state.

6. The belt retractor unit according to claim 5, characterized in that the belt retractor unit has a second emergency state in which the inclination (ΔZ) exceeds the inclination limit (G) and the belt reel (20) is blocked against the housing (12).

7. The second emergency condition is the horizontal acceleration value (|a XY The belt retractor unit according to claim 6, characterized in that it does not depend on the size of the |.

8. The belt retractor unit according to any one of claims 1 to 7, characterized in that the sensor device (60) has a three-axis sensor.

Images