Damper monitoring

Abstract

A method of damper monitoring comprises receiving a signal sent by a sensor configured to monitor the condition of a damper 10 and is sent while the damper is opening or closing, The method further in

Description

FIELD OF THE INVENTION The present invention relates to damper monitoring, and particularly to smoke control damper monitoring. BACKGROUND A damper can be used in a duct system to control the flow of air. Dampers can open and close accordingly to vary the volume of flow of air through the damper. Typically, a damper has either a single blade or a plurality of damper blades being moveable by an actuator between an open position for allowing airflow to pass through a passage / duct and a closed position for preventing airflow and / or smoke to pass through the passage / duct. Damper failure, particularly smoke control damper failure, can be dangerous because in the event a fire the damper is unable to close and therefore is unable to control the spread of the fire effectively. Failure is typically defined as the damper being unable to reach a predetermined defined position (fully open, fully closed, balanced position) within the time limit defined by the relevant European standard at the point of demand. (This is typically 30 seconds from receipt of the instruction to do so from the smoke control system, or upon the thermally actuated actuator (ETR) reaching a temperature of 72°C). Various types of damper monitoring are known for determining that the damper has failed, so that remedial action can be taken to fix the damper. For example, WO2014128480A2 discloses monitoring of the current of the operating signal to a damper, and when, for example, the current is zero when a 1A current is expected to be drawn for an operating damper, the system recognizes that that damper is not operating as required and sounds an alarm. EP3235546A1 discloses attaching a sensor to the damper blade, and a system which derives the position of the blade from the signal output by the sensor and also determines whether and in what time the maximum and minimum blade angles are reached. The system compares the data with a threshold value to determine whether or not the damper has failed its functional check. The above two documents disclose ways of damper monitoring that determine when the damper has failed so that remedial action can be taken. However, there is room for improvement in the way the damper is monitored. The present invention aims to overcome or mitigate at least one problem with the known damper monitoring systems. SUMMARY A first aspect of the present invention provides a method of damper monitoring, comprising: receiving a signal output by a sensor configured to monitor the condition of a damper, the signal being output while the damper is opening or closing, comparing a value of, or value derived from, the signal output by the sensor with a predetermined predicted failure threshold value, based on the comparison, predicting failure of the damper before the failure has occurred; and outputting an alert indicating that failure of the damper is predicted. Unlike WO2014128480A2 and EP3235546A1, the present invention allows failure to be predicted while the damper is still able to move but before failure has occurred. WO2014128480A2 detects failure by a zero current draw to the damper when a 1A current is expected. This is a likely indication of power failure to the damper. On the other hand, in the present invention the damper is still able to move but its failure to close within a required time is predicted before the failure occurs. This allows remedial action to be taken before the damper actually fails, and thus failure (or rather, predicted failure) can be determined at an earlier, or at least different, stage. EP3235546A1 can only detect failure once the failure has occurred. The present invention, on the other hand, allows failure to be predicted before it has occurred, and thus remedial action can be taken at an earlier stage. An example of a value derived from the signal is a trend value or an average value. In the below, unless the context indicates otherwise, “opening” may equally refer to closing or balancing of the damper blade. Preferably, failure of the damper is predicted when the value equals or exceeds, or equals or becomes less than, the predetermined predicted failure threshold value. Preferably, failure of the damper is predicted when the value equals or exceeds, or equals or becomes less than, the predetermined predicted failure threshold value at least two times. Preferably, failure of the damper is predicted when the value equals or exceeds, or equals or becomes less than, the predetermined predicted failure threshold value at least two times within a predetermined period of time, for example within 2 seconds. Preferably, the method comprises receiving data indicative of opening and / or closing cycles that the damper has made, wherein failure of the damper is predicted when the value equals or exceeds, or equals or becomes less than, the predetermined predicted failure threshold value at least two times within the same opening and / or closing cycle of the damper. Preferably, the method comprises receiving data indicative of opening, closing and / or balancing cycles that the damper has made, wherein failure of the damper is predicted when the value equals or exceeds, or equals or becomes less than, the predetermined predicted failure threshold value at least two times within a predetermined number of opening, closing and / or balancing cycles of the damper. Preferably, the method comprises: deriving a trend value from values of the signal, comparing the trend value with a predetermined predicted failure threshold value, and based on the comparison, predicting failure of the damper before the failure has occurred. For example, the trend or gradient value is greater than a trend or gradient threshold value, so failure is predicted. Preferably, the method comprises: deriving an average value from values of the signal, comparing the average value with a predetermined predicted failure threshold value, and based on the comparison, predicting failure of the damper before the failure has occurred. For example, the average value is greater than an average threshold value, so failure is predicted. To illustrate the point, an upper threshold current value of, for example, 5.8A / 10s of data could be set as a threshold value. Therefore, when the average current value over 10s reaches or exceeds 5.8A, failure is predicted. Preferably, the method comprises: deriving a frequency value from values of the signal, comparing the frequency value with a predetermined predicted failure threshold value, and based on the comparison, predicting failure of the damper before the failure has occurred. Preferably, the method comprises: deriving a frequency trend value from values of the signal, comparing the frequency trend value with a predetermined predicted failure threshold value, and based on the comparison, predicting failure of the damper before the failure has occurred. Preferably, the predetermined predicted failure threshold value is a first threshold value, the method further comprising: comparing a value of, or value derived from, the signal output by the sensor with a second predetermined predicted failure threshold value which is different from the first threshold value, based on the comparison, predicting failure of the damper before the failure has occurred; and outputting an alert indicating that failure of the damper is predicted. The second threshold value may be set in the same way as, and may have any of the preferable features of, the first threshold value mentioned above. The signal may be indicative of rotation of the blade. The signal may be indicative of vibration of the blade. The signal may be indicative of current drawn by an actuator for a blade of the damper. For example, should the blade be damaged and its mass reduced, the characteristic frequency or cyclic signature would also change this would also be detectable with a change in drive current. The same is true for if the blade were to become heavier due to moisture absorption. The predicted failure threshold value may be a non-zero value of current. The method may include deriving the opening or closing time from a value of the signal over at least two opening or closing cycles, and deriving a predicted number of cycles before failure occurs, and outputting a predicted cycles before failure signal. The number of cycles may be displayed on a display unit. For example, the method may determine that, at the current rate of deterioration, the damper will fail in 100 cycles. A second aspect of the present invention provides a damper comprising a damper monitoring unit configured to perform the method of the first aspect, wherein the damper comprises the sensor. Alternatively, the damper monitoring unit and / or the sensor may be provided separately or remotely from the damper. Therefore, a damper monitoring system may be provided comprising the damper, damper monitoring unit and sensor. Preferably, the damper comprises: a damper blade, and an actuator for the damper blade, wherein the sensor is configured to monitor current drawn by the actuator. Preferably, the damper comprises: a damper blade, wherein the sensor is rigidly coupled to the damper blade, and is configured to monitor movement of the damper blade. A third aspect of the present invention provides a server comprising: a damper monitoring unit configured to perform the method of the first aspect; and a communications interface configured to receive the signal. A fourth aspect of the present invention provides a method of configuring a damper monitoring unit, comprising: receiving a signal output by a sensor configured to monitor the condition of a damper, acquiring a value of, or value derived from, the signal output by the sensor at a time that failure of the damper occurs, and setting a predicted failure threshold value to be a value which is less than or more than the value of, or value derived from, the signal at the time that failure of the damper occurs. The method of configuring the damper monitoring unit may be carried out before installation, or during commissioning of the damper. The preferable features of the first to third aspects are equally applicable to the fourth aspect. The controller may be configured to derive the opening or closing time from a value of the signal over at least two opening or closing cycles, and derive a predicted number of cycles before failure occurs, and output a predicted cycles before failure signal. The number of cycles may be displayed on a display unit. For example, the controller may determine that, at the current rate of deterioration, the damper will fail in 100 cycles. According to another aspect of the present invention there is provided a non-transitory computer-readable storage medium comprising instructions which, when executed by at least one processor cause the at least one processor to perform any of the methods described herein. The instructions may be provided on one or more carriers. For example there may be one or more non-transient memories, e.g. a EEPROM (e.g. a flash memory) a disk, CD-or DVD-ROM, programmed memory such as read-only memory (e.g. for Firmware), one or more transient memories (e.g. RAM), and / or a data carrier(s) such as an optical or electrical signal carrier. The memory / memories may be integrated into a corresponding processing chip and / or separate to the chip. Code (and / or data) to implement embodiments of the present disclosure may comprise source, object or executable code in a conventional programming language (interpreted or compiled) such as C, or assembly code, code for setting up or controlling an ASIC (Application Specific Integrated Circuit) or FPGA (Field Programmable Gate Array), or code for a hardware description language. BRIEF DESCRIPTION OF THE DRAWINGS Preferred embodiments of the invention will now be described, purely by way of example, with reference to the drawings in which: Fig. 1 shows a schematic graph of current drawn by the damper actuator of the first embodiment of the invention plotted against time for a damper opening / closing cycle; Fig. 1 shows a schematic graph of current drawn by the damper actuator of the second embodiment of the invention plotted against time for a damper opening / closing cycle; Fig. 3 shows a damper in a closed position; and Fig. 4 illustrates a damper in an open position. DETAILED DESCRIPTION In the following description, failure is defined by the damper being unable to reach a predetermined defined position (fully open, fully closed, balanced position) within 30 seconds from receipt of the instruction to do so from the smoke control system, or 30 seconds from the thermally actuated actuator (ETR) reaching a temperature of 72°C. According to the first embodiment of the invention, a damper is provided which includes a damper monitoring unit. The damper monitoring unit includes a controller which receives an input signal and processes the input signal to monitor the damper’s condition. Based on the monitoring of the damper’s condition the controller can predict that a failure of the damper is likely to occur (i.e. that a failure is probable), and raise an alarm. The damper may be part of a system which includes an audible alarm unit or display unit for outputting an audible or visible alarm. The damper includes a damper blade, and an actuator for the damper blade. The input signal is supplied from a sensor configured to monitor current drawn by the actuator. In this embodiment, the damper includes the sensor, but this is not essential, and the sensor may be provided separately from the damper. Referring to Fig. 1, current drawn by the damper actuator is plotted against time for the damper opening / closing cycle. The upper dashed straight line (“Fault Upper Limit”) represents a threshold value which, if the current rises to reach it, indicates that the damper is unlikely to open / close within the required time (30 seconds in this embodiment). That is, failure would be predicted and a failure alert would be outputted. In this case, the threshold value is about 6.45A. Similarly, the lower dashed straight line (“Fault Lower Limit”) represents a threshold value which, if the current drops to reach it, indicates that the damper is unlikely to open / close within the required time. That is, failure would be predicted and a failure alert would be outputted. In this case, the threshold value is about 4.55A. The upper solid straight line (“Normal Upper Limit” or “Upper Tolerance”) represents a threshold value which, if the current rises to reach it, does not necessarily indicate failure is likely to occur. However, the controller would log the event and output a caution alert. In this case, the threshold value is about 6.2A. The Normal Upper Limit is lower than the Fault Upper Limit. The lower solid straight line (“Normal Lower Limit” or “Lower Tolerance”) works in a corresponding way to the Normal Upper Limit. In this case, the Normal Lower Limit is about 4.8A. The upper dotted straight line (“Linear (Trend Up)”) shows the gradient of the upper solid curve (“Trend Up”). The lower dotted straight line (“Linear (Normal)”) shows the gradient of the lower solid curve (“Normal”). Taking the lower solid curve (“Normal”) first, values of current drawn by the damper actuator are sampled every second at one-second intervals, and vary between Os and 30s over the opening / closing cycle. It can be seen that the “Normal” curve stays between the “Normal Upper Limit” and “Normal Lower Limit” and does not reach either of those lines, and neither does it reach the “Fault Upper Limit” nor “Fault Lower Limit” lines. In addition, as can be seen from the “Linear (Normal)” dotted line, the “Normal” curve does have a slight upward gradient, but the gradient is not sufficiently large to indicate that failure is likely. Therefore, controller would not output any alert to indicate failure is predicted for the “Normal” curve values (data). It should be noted that it is not possible to tell from Fig. 1 whether the damper does actually close within the required time (30s) because the position of the damper is not shown. However, testing is undertaken (during manufacture or commissioning of the system) during which the damper is opened and closed multiple times, and data is obtained giving a range over which the current varies during normal operation (i.e. operation in which the damper opens and closes within the required time). From that, threshold values at the boundaries of, or outside, the normal current range are determined which, if reached, indicate that failure is likely to occur. Referring now to the upper solid curve (“Trend Up”), the curve differs from the “Normal” curve in that the “Trend Up” curve reaches the “Normal Upper Limit” several times (on 14s, 18s, 22s, 26s and 28s-30s), and also reaches the “Fault Upper Limit” on 28s and 30s. In addition, the “Trend Up” curve has a larger upward gradient than the “Normal” curve. In this case, the controller is configured to have multiple predicted failure conditions, which is to say there are multiple different conditions in which the controller can determine that failure is likely, and then trigger an alert. These are described in detail below. Firstly, the controller is configured to output an alert if the current reaches the “Fault Upper Limit”. Therefore, in this case, when the current follows the “Trend Up” curve, the controller would output an alert after 28s. This is before the failure has occurred (i.e. before the damper has failed to close within 30s). The controller is also configured to output a caution alert if the current reaches the “Normal Upper Limit”. Therefore, in this case, an alert would first be output after 14s. In another embodiment, the controller may be configured to report a caution alert if the “Normal Upper Limit” is reached once, but a failure alert if reached multiple times (i.e. at least twice). Further, a gradient threshold value is set which, if the current gradient rises to reach it, means that the damper is unlikely to open / close within the required time. This threshold value is not shown on Fig. 1. The “Linear (Trend Up)” line has relatively steep gradient, and exceeds this gradient threshold value. Incidentally, in this case, the gradients are “global gradients” which are obtained from calculating the gradient using a relatively large number of data points (for example, data obtained over 15s, 20s or 25s, but could be obtained from data over 3s, 5s or 10s). The global gradient threshold value could be set, for example, to be a 1% variation between adjacent data points, in this case a 1% variation from one second to the next. It should be noted that, in that case, the 1% variation is between adjacent data points, but the threshold value is obtained using more than just data from two adjacent data points, in order to give a larger sample size. Alternatively, data from just two adjacent data points could be used, if desired. Therefore, in this case, when data over 15s is used to obtain the gradient, which is then found to reach or exceed a gradient threshold value, an alert that failure is predicted would be output after 15s. Additionally, or alternatively, average threshold values may be set. For example, in this case, an upper threshold value of 5.8A / 10s of data could be set as a threshold value. Therefore, when the average current value over 10s reaches or exceeds 5.8A, failure is predicted. In the present embodiment, the controller is configured to log the times that the various threshold values are reached, and also to log the time of other controller changes (for example, power shortages) and to cross-check whether they occurred at the same time, and, if so, output a caution alert. Fig. 2 shows a schematic graph of current drawn by the damper actuator of the second embodiment of the invention plotted against time for a damper opening / closing cycle. The second embodiment differs from the first embodiment in the following ways. The “Fault Upper Limit” and “Fault Lower Limit” threshold values are not used in the second embodiment. However, the controller is configured so that if two (or more) consecutive values of the current reach or exceed the “Upper Tolerance”, then the controller crosschecks whether they occurred at the same time as other system changes (for example, power shortages) and, if so, outputs a caution alert. Alternatively, instead of outputting caution alert, the controller may output a failure alert. The controller works with respect to the “Lower Tolerance” in a corresponding way. In the above embodiments, since the current will take some time to reach a steady state, a short time is allowed to elapse (to ignore initial inrush effects, for example) before the values of the current are taken, to allow the current to reach a steady state. In the above embodiments, the first and last degrees of motion of the damper blade may be ignored to ensure repeatable data is used and ignoring the effects of end position peaks. The controller is configured to perform the method described above. The functionality of the controller described herein may be implemented in code (software) stored on a memory comprising one or more storage media, and arranged for execution on a processor comprising one or more processing units. The storage media may be integrated into and / or separate from the controller. The code is configured so as when fetched from the memory and executed on the processor to perform operations in line with embodiments discussed herein. Alternatively, it is not excluded that some or all of the functionality of the controller is implemented in dedicated hardware circuitry (e.g. ASIC(s), simple circuits, gates, logic, and / or configurable hardware circuitry like an FPGA). The controller is not limited to monitoring actuator current values or even current per se. Additionally, or alternatively, an accelerometer may be attached to the damper blade, or is at least located where a rigid coupling to the damper blade exists. One example scenario is the monitoring of the accelerometer during the process of opening / closing the damper. A “typical” accelerometer signature can be obtained from opening the damper. Any deviation outside this signature (with some tolerance based on statistical data and commissioning steps) would result in a fault or warning being raised. The controller may, like in the first embodiment, use “Fault Upper Limit” and “Fault Lower Limit” threshold values. These may be respectively greater than or less than the maximum or minimum acceleration values output by the accelerometer during the testing phase when the damper is opening / closing normally. Then, for example, if the controller determines at any point during the opening / closing cycle that the acceleration value reaches or exceeds the “Fault Upper Limit”, the controller outputs a failure alert. As another example, the “Fault Lower Limit” threshold value may be set in relation to the maximum acceleration value rather than the minimum acceleration value. Therefore, when the maximum acceleration value drops below its expected range of values, the controller outputs a failure alert. The maximum acceleration value may be determined, for example, from the rate of change of acceleration being zero in, for example, the first 15s of the cycle, or may be taken as the maximum acceleration value obtained within the first 15s of the cycle. As a further example, an image capturing device (which may be a camera) may be employed as the sensor to monitor the condition of the damper. The image capturing device (e.g. a camera) may be mounted in the duct in which the damper is installed, and may have the damper blade, or a part configured to move with the damper blade, in its field of view. The input signal may be supplied from the camera to a controller which may perform image processing on image data supplied from the camera to monitor the damper’s condition. The controller may determine the position of the damper blade from the input signal and may, for example, compare the position during the cycle with a “Fault Lower Limit” threshold value, and, if the position is the same as or lower than the threshold value, predict failure and output a failure alert. In this case, the blade would be moving too slowly to reach its required position within the required time. Referring to Fig. 3, there is illustrated a damper 10 which may include a damper monitoring unit of any of the embodiments referred to above. As discussed below, the damper 10 is shown in the closed position 10A. The damper 10 is placeable in a duct or surface such as a wall and controls airflow through the damper 10. The damper 10 has a damper housing 12 that forms a channel or passage 16 through which airflow can pass through the damper housing 12 and thus damper 10. The damper housing 12 generally forms the sides or walls of the passage 16 such that there is an inlet and outlet from the passage 16. As shown in Fig. 3, a four sided damper housing 12 is formed therefore providing a square or rectangular shape when viewed from the front. In the four sided damper 10 shown, the damper housing 12 has opposing sides formed between a top side 20 and a bottom side 18. Likewise, there are further opposing sides between a left side 26 and a right side 28 of the damper housing 12. Therefore, the periphery of the passage 16 is formed by the two sets of opposing sides. Airflow passes within these two sets of opposing sides through the passage 16 of the damper housing 12. Whilst a four sided damper has been described, numerous shapes for the damper 10 can be used depending on the requirements. For instance, a circular damper housing 12 can be used in some situations. Within the damper housing 12 are damper blades 14. The damper blades 14 are generally rectangular in shape when viewed from a plan view having a height and width and forming a damper blade surface. The damper blades 14 are arranged to extend across the width of the damper housing between the left side 26 and the right side 28 such that a single blade can extend between these opposing sides. Therefore, the longest side (e.g. width) of a rectangular blade 14 extends between the left side 26 and the right side 28. The damper blades 14 are also arranged within the damper housing 12 between the top side 20 and the bottom side 18. In this configuration, multiple blades are arranged in parallel between the top side 20 and bottom side 18. Whilst the terms, top 20, bottom 18, left 26 and right 28 sides have been used for the walls of the damper housing 12, it is to be understood that the walls can be orientated in any manner not limited by the terms “top”, “bottom”, “left” and “right”. The damper blades 14 can be orientated between a fully closed position 10A where airflow through the passage 16 is prevented to a fully open position 10B (Fig. 4) where the airflow through the passage 16 is permitted to a maximum volume flow. Each damper blade 14 is pivotable on its own axis extending between the left 26 and right 28 sides of the damper housing 12. Therefore, each damper blade 14 can be moved to rotate about its axis, i.e. towards or away from the top 20 and bottom 18 sides of the damper housing 12. The passage 16 can be blocked to airflow when the damper blades 14 are rotated to an orientation that fills the passage 16 of the damper housing 12. Therefore, when the blade 14 is orientated substantially vertically, such that its surface is faced against the airflow through the passage 16, the damper 10 is closed, i.e. in the closed position 10A. Multiple blades 14 are positioned in the damper housing 12 and to close the damper 10 all of these are orientated substantially vertically such that airflow cannot pass through the passage 16. Therefore, when the blade is viewed from the front as a rectangle, the shortest side (e.g. height) of a rectangular blade 14 is positioned between the top side 20 and the bottom side 18. It will be appreciated that the complete prevention of airflow through the passage 16 may not occur in the closed position 10A due to sealing with surfaces, such as adjacent blades 14 and the sides of the damper housing 12. Referring to Fig. 4, the damper 10 in the open configuration 10B is shown. The damper blades 14 can be moved between a closed 10A and an open position 10B. In the open position 10B the damper blades 14 are arranged to permit flow through the passage 16. The rotation of the blades 14 through their axis to a position wherein the blade 14 is orientated substantially horizontally, such that its surface is faced in-line with the airflow through the passage 16 to permit airflow over the blade 14 (and in-between multiple blades 14), will result in the damper 10 being open, i.e. in an open position 10B. The open position 10B and closed position 10A describe a fully open and a fully closed configuration. Therefore, in the fully open position 10B, the blade 14 is orientated as flat and horizontal as is permissible to the flow of air through the passage 16. However, it will be appreciated that the blade 14 may not be completely horizontal in this configuration. There can be positions in which the damper blades 14 are held between fully open 10B and fully closed 10A where airflow is controlled between the two extremes. Whilst six damper blades 14 have been shown in Figs. 3 and 4, any number of blades 14 can be provided. For instance, it is possible to have a single blade 14 to control the opening and closing of the passage 16. The damper blades 14 can be all the same size or they can have different sizes. However, a set of damper blades 14 being all the same size except for one blade 14 allows for a mostly uniform arrangement where the one differently sized blade is used as a sizing blade which allows a damper 10 to be accurately sized for a particular duct or damper space without changing the sizes of all the blades 14. The blades 14 in the closed position 10A do not need to be sized so that they fill the passage 16 such that the blade edges touch blade edges of adjacent blades 14. Instead, the blades 14 can overlap in the closed position 10A such that a front surface of a blade 14 overlaps with a back surface of an adjacent blade 14. This overlapping configuration provides a larger surface area for forming a seal with a blade 14. Therefore, in the closed position 10A, the blades 14 do not need to be fully vertical to be in a fully closed position 10A, instead the blades 14 are substantially vertical. In some cases, seals are provided at the edges of the blades 14. Referring to Figs. 3 and 4, there is shown an actuator body 22 arranged on the side of the damper housing 12. In this case the actuator body 22 is arranged on the right side 28 of the damper body 12. The actuator body 22 forms part of the damper 10 and provides the actuation and control for moving the damper blades 14, such as between open and closed positions 10B, 10A. The actuator body 22 may include the damper monitoring unit and current sensor configured to monitor current drawn by the actuator. Preferred embodiments of the invention have been described purely by way of example, and various modifications, additions and / or omissions will present themselves to one skilled in the art, all of which form part of the invention.

Claims

1. A method of damper monitoring, comprising:receiving a signal output by a sensor configured to monitor the condition of a damper, the signal being output while the damper is opening or closing,comparing a value of, or value derived from, the signal output by the sensor with a predetermined predicted failure threshold value,based on the comparison, predicting failure of the damper before the failure has occurred; andoutputting an alert indicating that failure of the damper is predicted.

2. A method according to claim 1, wherein failure of the damper is predicted when the value equals or exceeds, or equals or becomes less than, the predetermined predicted failure threshold value.

3. A method according to claim 2, wherein failure of the damper is predicted when the value equals or exceeds, or equals or becomes less than, the predetermined predicted failure threshold value at least two times.

4. A method according to claim 3, wherein failure of the damper is predicted when the value equals or exceeds, or equals or becomes less than, the predetermined predicted failure threshold value at least two times within a predetermined period of time.

5. A method according to claim 3, comprising receiving data indicative of opening and / or closing cycles that the damper has made,wherein failure of the damper is predicted when the value equals or exceeds, or equals or becomes less than, the predetermined predicted failure threshold value at least two times within the same opening and / or closing cycle of the damper.

6. A method according to claim 3, comprising receiving data indicative of opening, closing and / or balancing cycles that the damper has made,wherein failure of the damper is predicted when the value equals or exceeds, or equals or becomes less than, the predetermined predicted failure threshold value at least two times within a predetermined number of opening, closing and / or balancing cycles of the damper.

7. A method according to claim 1, comprising:deriving a trend value from values of the signal,comparing the trend value with a predetermined predicted failure threshold value, andbased on the comparison, predicting failure of the damper before the failure has occurred.

8. A method according to claim 1, comprising:deriving an average value from values of the signal,comparing the average value with a predetermined predicted failure threshold value, andbased on the comparison, predicting failure of the damper before the failure has occurred.

9. A method according to claim 1, comprising:deriving a frequency value from values of the signal,comparing the frequency value with a predetermined predicted failure threshold value, andbased on the comparison, predicting failure of the damper before the failure has occurred.

10. A method according to claim 9, comprising:deriving a frequency trend value from values of the signal,comparing the frequency trend value with a predetermined predicted failure threshold value, andbased on the comparison, predicting failure of the damper before the failure has occurred.

11. A method according to any preceding claim, wherein the predetermined predicted failure threshold value is a first threshold value, the method further comprising:comparing a value of, or value derived from, the signal output by the sensor with a second predetermined predicted failure threshold value which is different from the first threshold value,based on the comparison, predicting failure of the damper before the failure has occurred; andoutputting an alert indicating that failure of the damper is predicted.

12. A method according to any preceding claim, wherein the signal is indicative of rotation of the blade.

13. A method according to any preceding claim, wherein the signal is indicative of vibration of the blade.

14. A method according to any one of claims 1 to 10, wherein the signal is indicative of current drawn by an actuator for a blade of the damper.

15. A method according to claim 14, wherein the predicted failure threshold value is a non-zero value of current.

16. A damper comprising:a damper monitoring unit configured to perform the method of any preceding claim, wherein the damper comprises the sensor.

17. A damper according to claim 16, comprising: a damper blade, and an actuator for the damper blade, wherein the sensor is configured to monitor current drawn by the actuator.

18. A damper according to claim 16, comprising: a damper blade, wherein the sensor is rigidly coupled to the damper blade, and is configured to monitor movement of the damper blade.

19. A server comprising:a damper monitoring unit configured to perform the method of any preceding claim; anda communications interface configured to receive the signal.

20. A method of configuring a damper monitoring unit, comprising:receiving a signal output by a sensor configured to monitor the condition of a damper,acquiring a value of, or value derived from, the signal output by the sensor at a time that failure of the damper occurs, and5 setting a predicted failure threshold value to be a value which is less than or morethan the value of, or value derived from, the signal at the time that failure of the damper occurs.

21. A non-transitory computer-readable storage medium comprising instructions10 which, when executed by at least one processor cause the at least one processor to perform the method of any of claims 1 to 15, or 20.

Images