The present document is directed at a method and a control unit for controlling the state of an (electro-) mechanical relay, notably for setting the control time instant for switching on or off the relay.

Mechanical relays are widely used e.g. in home appliances. When switching a relay at a random time instant relatively to the AC mains power sine wave, the relay contacts and/or its mechanical components may be subjected to degradation resulting from plasmatic arcs which form between the relay contacts during switching events of the relay. The phenomenon of plasmatic arcs may cause degradation of the relay mechanics by damaging the relay contact material, by damaging the relay return springs and/or by damaging other relay mechanical parts. Furthermore, extensive electromagnetic emission noise may be caused by the arc current between the contacts, wherein the noise may disturb operation of an appliance. In addition, a plasmatic arc may cause contact material to be evaporated and/or scattered within a relay, thereby affecting the overall reliability of the relay. Eventually, a damaged relay may affect the reliability of the home appliance which comprises the relay.

In order to avoid the phenomenon of plasmatic arcs, zero cross detection for detecting the zero crossing of the AC mains voltage sine wave may be used. The switching times of a relay may then be set based on the detected zero crossing time instants, such that the relay is switched on or off at time instants when the AC mains voltage sine wave exhibits a zero crossing.

A relay typically exhibits a control delay between the control time instant of a control signal for switching on or off the relay and the (dis)connect time instant at which the relay is actually switched on or off, thereby allowing or disabling the AC mains current to flow. In order to perform the switch-on or switch-off of the relay at a zero crossing time instant, the control time instant needs to be advanced with regards to the zero crossing time instant by the control delay. The present document is directed at the technical problem of estimating the control delay of a relay in an efficient and precise manner, notably in order to reduce the degradation of a relay during operation of the relay in conjunction with an AC power supply. This technical problem is solved by the subject-matter of each of the independent claims. Preferred examples are described in the dependent claims, in the following description and in the attached drawing.

According to an aspect, a method for controlling a relay is described. The relay comprises a movable anchor or armature and a coil. The coil is configured to exert a (magnetic) force onto the anchor subject to a coil current (notably a DC (direct current) current) at a control port, in order to connect contacts of the relay. The contacts of the relay may be configured to enable or to disable an electrical connection between load ports of the relay. The load ports may be coupled to an AC voltage (e.g. a 230V mains voltage at an AC frequency of 50Hz or an 110V mains voltage at an AC frequency of 60Hz).

During a switch-on event, the contacts of the relay may be connected, i.e. brought into contact with one another, subject to a coil current creating a magnetic field which attracts the movable anchor. The coil current may be generated by applying a control voltage to the control port of the relay. On the other hand, during a switch-off event, the contacts may be disconnected, i.e. separated from one another, subject to the coil current dropping to zero, such that the magnetic field is interrupted. The coil current may be interrupted by coupling the control port to ground.

The method may comprise applying, at a control time instant, a control signal to the control port for initiating a switch-off event or a switch-on event of the relay. The control time instant depends on a previously determined control delay of the relay. The control signal may be a step-up function increasing the control voltage from zero to a constant control voltage level (for initiating a switch-on event). Alternatively, the control signal may be a step-down function decreasing the control voltage from the constant control voltage level to zero (for initiating a switch-off event).

In particular, the method may comprise determining an upcoming or future zero crossing time instant of the AC voltage which is applied to the load ports of the relay. The zero crossing time instant may be the time instant at which the AC voltage changes polarity. The control time instant may be determined based on the previously determined control delay and based on the zero crossing time instant. In particular, the control time instant may be moved forward with respect to the (upcoming or future) zero crossing time instant by the control delay. By moving forward the control signal in dependence of the control delay of the relay, the disconnect time instant (at which the contacts of the relay disconnect) or the connect time instant (at which the contacts connect) may be moved closer to the zero crossing time instant (compared to an arbitrary selection of the control time instant). As a result of this, the generation of arcs between the contacts during a switch-off event or during a switch-on event may be reduced, thereby reducing degradation of the relay.

The method comprises determining a temporal course of the coil current subject to the control signal. As a result of the control signal for a switch-on event, the coil current may increase from substantially zero to a substantially constant on-current. The temporal course of the coil current until reaching the on-current may be determined. On the other hand, as a result of the control signal for a switch-off event, the coil current may decrease from the substantially constant on-current to substantially zero. The temporal course of the coil current until reaching substantially zero may be determined. The temporal course of the coil current may be sensed using a sensing or conversion resistor, which is configured to transform the coil current into a proportional coil voltage.

In addition, the method comprises detecting a disconnect time instant, at which the contacts of the relay disconnect, or a connect time instant, at which the contacts of the relay connect, based on a derivative of the temporal course of the coil current. The control signal may be for initiating a switch-off event. In this case, the disconnect time instant may be detected based on or using the first derivative of the temporal course of the coil current, notably based on a maximum of the first derivative of the temporal course of the coil current. On the other hand, the control signal may be for initiating a switch-on event. In this case, the connect time instant may be detected based on the second derivative of the temporal course of the coil current, notably based on a minimum of the second derivative of the temporal course of the coil current.

In addition, the method may comprise updating the control delay based on the detected disconnect time instant or based on the detected connect time instant. For updating the control delay, the time difference between the zero crossing time instant and the detected disconnect time instant or the detected connect time instant may be determined. The updated control delay may be increased or decreased based on the time difference (e.g. increased or decreased by a fraction of the time difference). The method may be repeated at subsequent cycles of the AC voltage, such that the determined control delay converges towards the actual control delay of the relay.

Using a derivative of the temporal course of the coil current during a switch-on or switch- off event, the actual disconnect time instants or connect time instants of a relay may be determined in a precise and reliable manner. By doing this, the generation of arcs during a switch-on or switch-off event may be reduced, thereby reducing degradation of a relay.

The method may comprise determining the control time instant as the zero crossing time instant minus the previously determined control delay and minus an offset. The offset may depend on a degree of reliability, notably a statistical scatter, of the control delay. By further offsetting the control time instant, the probability that an actual disconnect time instant or connect time instant occurs subsequent to the zero crossing time instant may be decreased (typically at the expense of increasing the probability that the actual disconnect time instant or connect time instant occurs prior to the zero crossing time instant). In view of the fact that, notably when switching off, an arc which is generated prior to the zero crossing time instant has a shorter duration than an arc which is generated subsequent to the zero crossing time instant, the use of an additional offset allows further reducing degradation of the relay (in particular, when making the offset dependent on the degree of reliability of the determined control delay).

The method may comprise detecting a peak of the first derivative of the temporal course of the coil current. The disconnect time instant may then be detected based on the peak, notably based on a time instant of a maximum of the peak, of the first derivative of the temporal course of the coil current. As a result of this, the disconnect time instant and by consequence the control delay may be determined in a particularly precise manner.

The method may comprise comparing the first derivative of the temporal course of the coil current with a peak detection threshold, in order to determine a (rectangular) envelope of the peak of the first derivative of the temporal course of the coil current. The disconnect time instant may then be determined in an efficient manner based on the envelope, notably based on a time instant of the center of the envelope. By determining the envelope of the peak, a binary signal may be generated, which allows the disconnect time instant to be determined in a particularly efficient manner (e.g. using a microprocessor).

The method may comprise determining a peak value of the peak of the first derivative of the temporal course of the coil current. Furthermore, the method may comprise determining information regarding the state of health of the relay, notably regarding the gap width of the gap between the contacts, based on the peak value. In particular, the information regarding the state of health of the relay may be determined based on experimental data which is indicative of the state of health of the relay, notably of the gap width, as a function of the peak value. Depending on the information regarding the state of health of the relay, it may be initiating that the relay is maintained or replaced. As a result of this, the reliability of an appliance comprising the relay may be increased.

According to a further aspect, a control unit for controlling a relay is described. The control unit may comprise analog and/or digital circuitry. The control unit is configured to apply, at a control time instant, a control signal to the control port for initiating a switch-off event or a switch-on event of the relay, wherein the control time instant depends on a previously determined control delay of the relay. Furthermore, the control unit may be configured to determine a temporal course of the coil current subject to the control signal, and to detect a disconnect time instant, at which the contacts of the relay disconnect, or a connect time instant, at which the contacts of the relay connect, based on a derivative of the temporal course of the coil current. In addition, the control unit may be configured to update the control delay based on the detected disconnect time instant or based on the detected connect time instant.

The control unit may comprise a conversion or sensing resistor configured to convert the coil current into a coil voltage. Furthermore, the control unit may comprise derivation circuitry which is configured to determine a first output signal indicative of a derivative of the course of the coil voltage, based on the course of the coil voltage. In addition, the control unit may comprise a microprocessor configured to sense the first output signal or a signal derived therefrom. The microprocessor may be further configured to detect the disconnect time instant or the connect time instant based on the sensed first output signal or the signal derived therefrom, and to update the control delay. Hence, the control unit may be implemented in an efficient manner using analog and digital circuitry.

The control unit may be configured to control a plurality of relays. The derivation circuitry may comprise an input portion and an output portion (wherein the output portion provides the first output signal). The control unit may comprise a plurality of conversion resistors for the corresponding plurality of relays. Furthermore, the control unit may comprise a plurality of input portions for the corresponding plurality of relays, and a single output portion for the plurality of relays. Hence, the output portion of the derivation circuitry may be used jointly for a plurality of different relays.

The control unit may be configured to detect the disconnect time instant or the connect time instant of the plurality of relays at a corresponding plurality of sequential time intervals, notably at a corresponding plurality of half cycles of an AC voltage applied to load ports of the plurality of relays. Hence, a plurality of relays of an appliance may be controlled and/or supervised in an efficient manner.

The control unit may comprise a comparator which is configured to compare the first output signal with a peak detection threshold to determine a second output signal which is indicative of a rectangular envelope around a peak of the first output signal. The microprocessor may be configured to detect the disconnect time instant or the connect time instant based on the second output signal, thereby further increasing the efficiency of the control unit.

According to a further aspect, a method for determining information regarding a state of health of a relay is described. The relay comprises a movable anchor and a coil which is configured to exert a force onto the anchor subject to a coil current at a control port, in order to connect contacts of the relay. The method comprises applying a control signal to the control port for initiating a switch-off event or a switch-on event of the relay. Furthermore, the method comprises determining a temporal course of the coil current subject to the control signal. In addition, the method comprises detecting a peak of a derivative of the temporal course of the coil current. The method further comprises determining a peak value of the peak of the derivative of the temporal course of the coil current, and determining information regarding the state of health of the relay, notably regarding a gap width of a gap between the contacts of the relay, based on the peak value.

It should be noted that the methods and systems including its preferred embodiments as outlined in the present document may be used stand-alone or in combination with the other methods and systems disclosed in this document. In addition, the features outlined in the context of a system are also applicable to a corresponding method. Furthermore, all aspects of the methods and systems outlined in the present document may be arbitrarily combined. In particular, the features of the claims may be combined with one another in an arbitrary manner.

The invention is explained below in an exemplary manner with reference to the accompanying drawing, wherein

Fig. 1 shows an example electromechanical relay;

Fig. 2a shows the temporal course or curve of the coil current during a switch-on event;

Fig. 2b shows the temporal course or curve of the coil current during a switch-off event;

Fig. 3a illustrates the connect time instant of a relay;

Fig. 3b illustrates the disconnect time instant of a relay;

Fig. 4 shows an example circuit for determining the control delay of a relay;

Fig. 5 shows an example circuit for determining the control delay of a plurality of relays;

Fig. 6 shows an example functional relationship between the peak coil current and the contact gap width of a relay;

Fig. 7 illustrates an example offset to be used for determining the control time instant for initiating a switch-on event or a switch-off event of a relay;

Fig. 8a shows a flow chart of an example method for controlling a switch-on event or a switch-off event of a relay; and

Fig. 8b shows a flow chart of an example method for determining the state of health of a relay.

As indicated above, the present document is directed at reducing the degradation of an electromechanical relay, which is used for switching on or off an AC power supply (e.g. an AC mains supply at 230V with an AC frequency of 50Hz). In this context Fig. 1 shows an example relay 1. The relay 1 comprises a coil 2 which is configured to generate a magnetic field, subject to a coil current l(t) which is applied to the control port 4. The coil 2 may comprise a coil core 8, which is configured to direct the magnetic field towards an armature or anchor 3 of the relay 1. The armature or anchor 3 is configured to be attracted by the magnetic field generated by the coil 2. The armature 3 is typically moved away from the coil 2 and the coil core 8 using a spring 10. The magnetic field which is generated by the coil 2 may cause a sufficiently high attracting force onto the armature 3, such that spring force for the spring 10 is overcome and such that the armature 3 is pulled towards the coil core 8 to switch on the relay 1.

The armature 3 is coupled to a contact 5, wherein the contact 5 of the armature 3 touches the contact 5 of the base of the relay 1 , when the armature 3 is pulled towards the coil core 8. As a result of this, the load ports 6 of the relay 1 are electrically coupled with one another, thereby allowing a current to flow between the load ports 6 and notably between the contacts 5.

By interrupting the coil current l(t), the magnetic field is interrupted and the armature 3 is pulled away from the coil core 8 due to the spring force caused by the spring 10. Eventually, the contacts 5 are disconnected from one another, thereby switching off the relay 1. In a relaxed or inactive state, the contacts 5 form a contact gap S having a certain gap width.

In the course of a switch-on event, the armature 3 is pulled towards the coil core 8, as soon as a coil current l(t) is applied to the control port 4. The coil current l(t) may be applied at a control time instant. At a subsequent connect time instant, the contacts 5 start touching each other, thereby switching on the relay 1 and thereby allowing a load current to flow via the load ports 6. At the connect time instant, the armature 3 typically does not yet touch the coil core 8. The time instant, at which the armature 3 touches the coil core 8 may be referred to at the close time instant, as the relay 1 is now fully closed. Once the relay 1 is fully closed, the (flexible) connecting plate between the armature 3 and the contact 5 of the armature 3 forms a flat spring 11 in a flexed condition, wherein the force of the flat spring 1 1 pulls the armature 3 away from the coil core 8. Hence, the magnetic field need to overcome the forces of the return spring 10 of the armature 3 and of the flat spring 11 of the contact 5. In the course of a switch-off event, the control port 4 is disconnected from a control voltage (at a control time instant). As a result of this, the coil current l(t) decreases gradually such that the magnetic force decreases causing the flat spring 1 1 to return to a relaxed state and subsequently causing the return spring 10 to pull the contact 5 of the armature 3 away from the contact 5 of the base of the relay 1. The time instant, at which the contacts 5 are disconnected may be referred to as the disconnect time instant.

As outlined in the introductory section, a switch-on event or a switch-off event may lead to plasmatic arcs between the contacts 5, if the contacts 5 are connected or disconnected at a time instant, when the voltage across the load ports 6 is relatively high (e.g. 30V or higher). This may lead to a degradation of the contacts 5 and of the mechanical components of the relay 1. In order to avoid such degradations, it is preferable to place the (dis)connect time instant of a switch-on event or a switch-off event at a zero crossing of the AC voltage which is applied to the load ports 6. In view of this, the control delay between the control time instant, at which the control voltage is applied to or disconnected from the control port 4, and the (dis)connect time instant, at which the contact gap S is closed or opened, needs to be determined in a precise manner.

Fig. 2a shows an example curve or course 201 of the coil current 200 during a switch-on event. It can be seen that the coil current 200 increases gradually, but exhibits an intermediate drop. Fig. 2b shows an example curve or course 202 of the coil current 200 during a switch-off event. It can be seen that the coil current 200 decreases gradually, but exhibits an intermediate peak.

Figs. 3a and 3b show the example curves or courses 201 , 202 of the coil current 200 during a switch-on event and a switch-off event, respectively. Furthermore, Figs. 3a and 3b illustrate different time instants for different incidents during a switch-on event and a switch-off event, respectively. In particular, Fig. 3a illustrates a time instant 311 at which the anchor or armature 3 starts to move. Furthermore, Fig. 3a shows the connect time instant 312 at which the contacts 5 of a relay 1 start touching each other. In addition, Fig. 3a shows a time instant 313 at which the anchor 3 and the coil core 8 start touching each other. Fig. 3a also shows a time instant 314 at which mechanical vibrations of the anchor 3 occur (subsequent to closing of the relay 1 ). The connect time instant 312 may e.g. be detected within a test environment by applying a (relatively small) DC test voltage to the load ports 6 and by detecting the time instant when the load ports 6 are short circuited.

Fig. 3a also shows the first derivative 301 of the coil current curve 201 , as well as the second derivative 302 of the coil current curve 201. It can be seen from Fig. 3a that the minimum of the second derivative 302 of the coil current curve 201 during a switch-on event coincides with the connect time instant 312. Hence, the control delay 340 for a switch-on event may be determined in a precise manner by

• determining the second derivative 312 of the coil current curve 201 during a switch-on event; and

• determining the time instant of the minimum value of the second derivative 312 of the coil current curve 201.

Fig. 3b illustrates various different time instants of different incidents during a switch-off event of a relay 1. In particular, Fig. 3b shows the control time instant 331 at which the control voltage is disconnected from the control port 4 of the relay 1 , in order to switch off the relay 1. Furthermore, Fig. 3b illustrates a time instant 332 at which the anchor release begins. In addition, Fig. 3b shows the disconnect time instant 333 at which the contacts 5 are effectively disconnected. Fig. 3b also shows the time instant 334 at which the anchor 3 reaches its relaxed position. In addition, Fig. 3b illustrates a bouncing incident 335 of the anchor 3 from the relaxed position.

Fig. 3b also shows the positive part (i.e. the positive values) of the first derivative 321 of the curve 202 of the coil current 200 during a switch-off event (the negative part of the first derivative 321 has been omitted; notably the negative part of the first derivative 321 has been cut off by the operational amplifier 401 of the circuit 400 of Fig. 4, which is connected in a single supply mode). It can be seen that the maximum of the first derivative 321 coincides with the disconnect time instant 333. Hence, the control delay 340 of a switch-off event may be determined in a precise manner by

• determining the first derivative 321 of the curve 202 of the coil current 200 during a switch-off event; and

• determining the time instant of the maximum value of the first derivative 321 (being the disconnect time instant 333). Determining a maximum or a minimum value may be difficult when using analog circuitry. In view of this, a comparator envelope 322 of the first derivative 321 of the curve 202 of the coil current 200 may be determined, using e.g. a comparator. The comparator envelope 332 exhibits a rectangular shape around the peak of the derivative 321. The disconnect time instant 333 may be determined in an efficient manner, by determining the duration of the rectangle of the comparator envelope 332 and by selecting the disconnect time instant 333 to be the time instant which cuts the duration of the rectangle of the comparator envelope 322 approximately in half.

As can be seen from Fig. 3b, the peak of the first derivative 321 may be (slightly) asymmetrical. As a result of this, the maximum value of the peak of the first derivative 321 may be (slightly) shifted with respect to the middle point of the comparator envelope 322. The asymmetry of the peak of the first derivative 321 and/or the required shift with respect to the middle point of the comparator envelope 322 for determining the disconnect time instant 333 may be determined experimentally prior to operation of a relay 1 and may be used subsequently during operation of the relay 1.

Fig. 4 shows an example (analog) circuit 400 for determining the disconnect time instant 333 and/or for determining the control delay 340. Fig. 4 shows the relay 1 which is controlled using the control signal 417. The coil current 200 through the coil 2 of the relay 1 may be converted into a corresponding coil voltage 410 using a sensing resistor R10. The coil voltage 410 may be differentiated using a differentiator or derivation circuit. The differentiator circuit comprises the C/R differentiator circuit 402 (comprising the capacitor C1 and the resistor R3). Hence, the first derivative 321 of the relay coil current vs. time (measured as a coil voltage 410 across R10) may be provided. The parts R1 , C2, RC2 and R2 form a low pass filter 403 in conjunction with the operational amplifier 401 and provide for amplification of the filtered signals 41 1 , 412. The output 413 of the operational amplifier 401 is indicative of the first derivative 321 of the coil current 200 (filtered and shaped by the operational amplifier 401 ).

The Schottky diode D1 (having a relatively low forward voltage) may be used to protect the input IN1 + of the operational amplifier 401 (when operated in single supply mode) against negative input voltages. The diode D10 has the function of a flyback diode. The operational amplifier circuit 401 shown in Fig. 4 comprises two separate operational amplifiers. The second operational amplifier is used as a comparator for producing a rectangular envelope 322 of the differentiator signal 413. In particular, the output signal 413 of the first operational amplifier is fed back towards the input IN2+ of the second operational amplifier (as signal 414). Furthermore, a reference signal 415 (referred to herein as the peak voltage threshold) is derived from the DC supply voltage VCC of the circuit 400 using the voltage divider comprising the resistors R5, R6. The output 416 of the second operational amplifier indicates a rectangular envelope 322 around the peak of the first derivative 321 of the curve 202 of the coil current 200.

The output signal 413 of the first operational amplifier (indicative of the first derivative 321 ) and/or the output signal 416 of the second operational amplifier (indicative of the rectangular envelope 322 of the peak of the first derivative 321 ) may be provided to an input of a microprocessor (not shown) for further analysis. The microprocessor may be configured to determine the disconnect time instant 333 based on at least one of the output signals 413, 416 (e.g. by determining the time instant which cuts the rectangular envelope 322 in half). The control delay 340 may then be determined as the duration between the control time instant 331 and the disconnect time instant 333.

Fig. 5 shows an extension of the circuit 400 for determining the control delays 340 of a plurality of different relays 1. As can be seen from Fig. 5 that most of the circuitry of circuit 400 may be used in common for the plurality of relays 1. Only the sensing resistor R10 for deriving the coil voltage 410 from the coil current 200, the flyback diode D10, the resistor R1 (which is part of a low pass filter 403) and the capacitor C1 (which is part of the differentiator circuit 402) need to be provided for each relay 1 (and may be referred to as the input portion of the deviation circuit). The capacitors C1 for the different relays 1 may be coupled at node 420 to the jointly used components of the circuit 400 (which comprise a jointly used output portion of the deviation circuit).

Hence, the circuit 400 of Fig. 4 may be used in a multi-relay environment by (time) multiplexing of the derivative signals of the different relays 1 (without using any particular multiplexer). The self-multiplexing can be done by splitting the C1 R3 differentiator 402 and the low pass filter 403 as shown in Fig. 5. As can be seen in Fig. 5 each relay 1 has its dedicated flyback diode D10, its dedicated coil current measuring resistor R10, resistor R1 and derivative capacitor C1. The remaining portion of the differentiator and comparator circuits 401 , 403 are common for all the relays 1. Such a self-multiplexing circuit provides an inexpensive and efficient way for providing an independent control of the switching-off time instants of multiple relays 1 with just one dedicated circuit 400 and relatively few passive parts added to each of the relays 1. The different relays 1 may exhibit different characteristics (e.g. with regards to electrical features and size of the relays 1 ).

The disconnect time instants 333 for the different relays 1 may be determined within different cycles or half cycles of the AC power supply, thereby performing a time multiplexed use of the joint portion of the circuit 400.

As indicated above, the output signals 413, 416 of circuit 400 each allow identifying the actual switch-off point (i.e. the disconnect time instant 333) of a relay 1 , e.g. by finding the middle time of the analogue pulse (i.e. the rectangular envelope) 322. Furthermore, the output signals 413, 416 may each be used for predictive monitoring of the mechanical health of a relay 1 , e.g. by monitoring the heights of the pulse or peak of the first derivative 321 (e.g. by monitoring the maximum value / voltage of the output signal 413 and/or by monitoring the pulse presence at the output signal 416). The maximum value of the first derivative 321 is related to the total accelerated mass of the mechanics and the flat spring deflection (i.e. the force) of the relay 1. In case of an obstacle between the relay contacts 305 (such as dirt, metal whiskers grown from the contact surface, etc.), which lowers the width of the contact working gap S, the peak voltage typically increases. On the other hand, if the width of the contact working gap S increases, e.g. by burning out or scattering of the contact material, the peak voltage may decrease.

Fig. 6 shows data 600 indicating the functional relationship between the width 601 of the gap S and the peak value 602 of the first derivative 321. It can be seen that the peak value 602 has an almost linear relationship with the gap width 601. Hence, the gap width 601 may be determined in a reliable manner based on at least one of the output signals 413, 416 of the circuit 400. The determined gap width 601 may then be used for preventive maintenance, e.g. for indicating on a user interface of an appliance that a relay 1 should be replaced (e.g. because the contacts 5 of the relay 1 are deteriorated). By doing this, the reliability of an appliance may be increased. The control of a relay 1 may be adapted in an iterative manner. A control unit 100 for controlling the relay 1 may be configured to store a value of the control delay 340 in a storage unit. Furthermore, the control unit 100 may be configured to determine the zero crossing time instant 705 of the AC power supply (see Fig. 7). If the relay 1 is to be switched on or off, then the control time instant 331 may be determined to be the zero crossing time instant 705 minus the stored control delay 340. As a result of this, the actual disconnection or connection of the contacts 305 of the relay 1 occurs at the zero crossing time instant 705, thereby avoiding the generation of arcs.

The control delay 340 may be determined repeatedly, using the scheme and/or the circuit 400 outlined in the present document. For this purpose, subsequent to a switch-on event or a switch-off event, the time difference between the actual connect time instant 312 or the actual disconnect time instant 333 and the zero crossing time instant 705 may be determined. The time difference may be used to increase or decrease the stored value of the control delay 340, thereby updating the control delay 340. The updated control delay 340 may then be stored in the storage unit of the control unit 100. By doing this, a precise control of a relay 1 may be maintained over the lifetime of the relay 1. Furthermore, the repeated determination of the first derivative 321 of the curve 202 of the coil current 200 during a switch-off event may be used for monitoring the state of health of the relay 1.

The creation of an arc during a switch-on or a switch-off event may be avoided if the voltage value of the AC supply voltage 700 at the connect time instant 312 or the disconnect time instant 333 is at or below a pre-determ ined voltage threshold 704 (as illustrated in Fig. 7). The pre-determined voltage threshold 704 typically depends on the environmental conditions (notably the humidity) of the relay 1. Typically, voltage thresholds 704 are around 30V (or 20V in case of high humidity) for a switch-off event.

As can be seen from Fig. 7, the AC supply voltage 700 is only below the voltage threshold 704 for a relatively small fraction (e.g. 6%) of the total time of a half cycle. In case of an AC frequency of 50Hz, the total time of a half cycle is 10ms and the time interval which is available for safely disconnecting the contacts 305 of the relay 1 may be as low as 400 to 600ps. Hence, the control delay 340 needs to be determined with relatively high precision and/or with a relatively high degree of reliability. Fig. 7 shows an example statistical distribution 710 of the control delay 340 around the exact value of the control delay. The temporal width of the distribution is typically an indication of the degree of reliability. It can be seen from Fig. 7 that the control delay 340 which is determined using the schemes outlined in the present document, may deviate from the exact value of the control delay. If using the determined control delay 340 as an offset to the zero crossing time instant 705, it may occur that the resulting disconnect time instant 333 is prior or subsequent to the zero crossing time instant 705, such that the AC supply voltage 700 is higher than the voltage threshold 704 at the resulting disconnect time instant 333. This may lead to an arc between the contacts 5 of the relay 1.

If the actual value of the control delay is lower than the determined control delay 340, the resulting disconnect time instant 333 is prior to the zero crossing time instant 705, such that the AC supply voltage 700 is decreasing. By consequence, the arc between the contacts 5 is maintained only for a relatively short time interval. On the other hand, if the actual value of the control delay is higher than the determined control delay 340, the resulting disconnect time instant 333 is subsequent to the zero crossing time instant 705, such that the AC supply voltage 700 is increasing. By consequence, the arc between the contacts 5 is maintained for a relatively long time interval, until the following zero crossing time instant.

In the case which is illustrated in Fig. 7, the probability 71 1 that an arc is created on the decreasing slope of the AC supply voltage 700 (i.e. prior to the zero crossing time instant 705) is equal to the probability 712 that an arc is created on the increasing slope of the AC supply voltage 700 (i.e. subsequent to the zero crossing time instant 705). The impact of an arc on the deterioration of a relay 1 typically increases with increasing duration of the arc. Hence, it may be beneficial to increase the probability 71 1 in order to decrease the probability 712. This may be achieved by increasing the determined control delay 340 with an offset 703, such that the center of the distribution 710 shown in Fig. 7 is moved towards the decreasing slope of the AC supply voltage 700. The offset 703 may be determined such that the probability 712 is below a pre-determined probability threshold (close to zero). In particular, the offset 703 may be determined in dependence of the distribution 710 of the determined control delay 340. As illustrated in Fig. 7, applying an offset 703 to the control delay 340 corresponds to applying the offset 703 to the zero crossing time instant 705, thereby providing an offset zero crossing time instant 702. The determined control delay 340 may be subtracted from the offset zero crossing time instant 702 to determine the control time instant 331 for initiating a switch-off event.

Fig. 8a shows a flow chart of an example method 800 for controlling a relay 1 , wherein the relay 1 comprises a movable anchor or armature 3 and a coil 2. The coil 2 is configured to exert a (magnetic) force onto the anchor 3 subject to a coil current 200 at a control port 4 or the relay 1 , in order to connect contacts 5 of the relay 1. The contacts 5 may be used to couple or to disconnect load ports 6 of the relay 1. The load ports 6 may be coupled to an AC voltage 700 comprising a sequence of (sinusoidal) cycles at an AC frequency (e.g. 50Hz or 60Hz).

The method 800 comprises applying 801 , at a control time instant 331 , a control signal to the control port 4 for initiating a switch-off event or a switch-on event of the relay 1. The control time instant 331 typically depends on a previously determined control delay 340 of the relay 1. In particular, the control time instant 331 may occur prior to an upcoming zero crossing time instant 705 of the AC voltage 700, depending on the control delay 340, such that the actual disconnection or connection of the contacts 5 occurs as close as possible to the zero crossing time instant 705 (or to an offset zero crossing time instant 702).

In addition, the method 800 comprises determining 802 a temporal course 201 , 202 of the coil current 200 subject to the control signal. In particular, it may be determined how the coil current 200 evolves during the switch-off event or during the switch-on event (until the time instant at which the relay 1 is fully switched-off or fully switched-on).

Furthermore, the method 800 comprises detecting 803 a disconnect time instant 333, at which the contacts 5 of the relay 1 disconnect, or a connect time instant 312, at which the contacts 5 of the relay 1 connect, based on a derivative 321 , 302 of the temporal course 201 , 202 of the coil current 200. In particular, the disconnect time instant 333 may be determined based on the first derivative 321 of the temporal course 202 of the coil current 200. On the other hand, the connect time instant 312 may be determined based on the second derivative 302 of the temporal course 201 of the coil current 200. In addition, the method 800 may comprise updating 804 the control delay 340 based on the detected disconnect time instant 333 or based on the detected connect time instant 312. Fig. 8b shows a flow chart of an example method 810 for determining information regarding a state of health of a relay 1. The relay 1 comprises a movable anchor 3 (also referred to herein as an armature) and a coil 2. The coil 8 is configured to exert a force onto the anchor 3 subject to a coil current 200 at a control port 4 of the relay 1 , in order to connect contacts 5 of the relay 1.

The method 810 comprises applying 81 1 a control signal to the control port 4 for initiating a switch-off event or a switch-on event of the relay 1. In particular, a (DC) voltage may be applied to the control port 4, in order to initiate a switch-on event. The (DC) voltage may be disconnected from the control port 4 or the control port 4 may be short circuited, in order to initiate a switch-off event.

In addition, the method 810 comprises determining 812 a temporal course 202 of the coil current 200 subject to the control signal. During a switch-on event, the temporal course 201 may comprise or may be limited to the time interval during which the coil current 200 increases to a substantially constant on-current used for maintaining the relay 1 in the on- state. On the other hand, during a switch-off event, the temporal course 202 may comprise or may be limited to the time interval during which the coil current 200 drops to substantially zero.

Furthermore, the method 810 comprises detecting 813 a peak of a derivative 321 , 302 of the temporal course 201 , 202 of the coil current 200. In particular, a peak of the first derivative 321 of the temporal course 202 of the coil current 200 during a switch-off event may be detected. The peak may be detected using the circuit 400 described in the present document.

In addition, the method 810 comprises determining 814 a peak value 602 of the peak of the (first) derivative 321 , 302 of the temporal course 202 of the coil current 200. The method 810 further comprises determining 815 information regarding the state of health of the relay 1 , notably regarding a gap width 601 of a gap S between the contacts 5 of the relay 1 , based on the peak value 602. For this purpose, experimental data 600 may be used, which indicates the state of health of the relay 1 as a function of the peak value 602. The aspects which are outlined in the present document enable a reliable and efficient control of a relay 1. The control scheme allows reducing the deteriorations of a relay 1 , which are caused by switch-on or switch-off events of the relay 1. Furthermore, monitoring of the state of health of a relay 1 may be performed, thereby enabling preventive maintenance of the relay 1 and of an appliance (notably a home appliance, such as a dishwasher, a washing machine, an oven, etc). In addition, failures of a relay 1 may be detected.

It should be noted that the description and drawings merely illustrate the principles of the proposed methods and systems. Those skilled in the art will be able to implement various arrangements that, although not explicitly described or shown herein, embody the principles of the invention and are included within its spirit and scope. Furthermore, all examples and embodiment outlined in the present document are principally intended expressly to be only for explanatory purposes to help the reader in understanding the principles of the proposed methods and systems. Furthermore, all statements herein providing principles, aspects, and embodiments of the invention, as well as specific examples thereof, are intended to encompass equivalents thereof.