SYSTEM FOR CHECKING A CORRECT MOUNTING OF A SENSOR

Abstract

The present disclosure relates to a system for checking a correct mounting of a plurality of sensors, in particular of sensors mounted in an engine system, comprising a controller configured to communicate with the plurality of sensors, wherein the controller is configured to control a switching on or off of the power supply to at least a sub-group of the sensors and to check whether a sensor whose power supply was switched on is communicating, wherein the system is configured such that at least two sensors are connected to a common power supply port and are therefore powered on or off at the same time, wherein the at least two sensors comprise at least two different types of sensors.

Claims

1. A system for checking a correct mounting of a plurality of sensors, in particular of sensors mounted in an engine system, comprising a controller configured to communicate with the plurality of sensors, wherein the controller is configured to control a switching on or off of a power supply to at least a sub-group of the sensors and to check whether a sensor whose power supply was switched on is communicating, wherein the system is configured such that at least two sensors are connected to a common power supply port and are therefore powered on or off at the same time, wherein the at least two sensors comprise at least two different types of sensors.

2. The system of claim 1, wherein the controller comprises a check-routine that consecutively checks all the sensors having a power supply port controlled by the controller by switching on the respective power supply port, checking whether the sensors connected to the power supply port are communicating, and switching off the power supply port if further sensors connected to another power supply port remain to be checked.

3. The system of claim 1, wherein the plurality of sensors and preferably the at least two sensors connected to the common power supply port comprise at least one sensor that communicates cyclically and at least one sensor that communicates by interrogation, wherein preferably, the controller is configured to check the sensors that communicate cyclically and the sensors that communicate by interrogation in alternation.

4. The system of claim 3, wherein the plurality of sensors and preferably the at least two sensors connected to the common power supply port comprise at least one sensor that communicates cyclically and at least two sensors that communicate by interrogation, wherein the controller is configured to check by interrogation whether a first sensor communicating by interrogation is communicating, to wait for a cyclic communication of the sensor that communicates cyclically and only then check by interrogation whether a second sensor communicating by interrogation is communicating.

5. The system of claim 3, wherein at most one sensor that communicates cyclically is connected to any power supply port.

6. The system of claim 1, wherein the plurality of sensors and preferably the at least two sensors connected to the common power supply port comprise at least one NOx sensor and at least one NH3 sensor.

7. The system of claim 1, wherein at least one of the at least two sensors connected to the common power supply port has a first pin select state and wherein at least one of the at least two sensors has a second pin select state.

8. The system of claim 1, wherein all the sensors connected to the common power supply port have a different ID.

9. A system for checking a correct mounting of a plurality of sensors, in particular of sensors mounted in an engine system, the system comprising a controller configured to communicate with the plurality of sensors, wherein the controller is configured to control a switching on or off of a power supply to at least a sub-group of the sensors and to check whether a sensor whose power supply was switched on is communicating, wherein at least one sensor and preferably at least two sensors have a power supply port that is switched on and off together with the controller.

10. The system of claim 9, wherein a check-routine of the controller stops and issues a warning once a mounting error of one of the sensors having the power supply port that is switched on and off together the controller is detected.

11. The system of claim 1, wherein a check-routine of the controller consecutively checks all the sensors having a power supply port controlled by the controller and issues a common warning message containing all detected mounting errors of theses sensors after finishing the check-routine.

12. The system of claim 1, wherein the plurality of sensors communicate with the controller over a bus system, wherein the bus system preferably is a CAN-bus system.

13. The system of claim 1, wherein the plurality of sensors are sensors arranged on an exhaust gas aftertreatment system of the engine system.

14. An engine comprising the plurality of sensors and the system of claim 1.

15. A method for checking a correct mounting of a plurality of sensors, the sensors preferably mounted in an engine system, the method comprising controlling a switching on or off of a power supply to at least a sub-group of the sensors and checking whether a sensor whose power supply was switched on is communicating; wherein at least two sensors are connected to a common power supply port and are therefore powered on or off at the same time, wherein the at least two sensors comprise at least two different types of sensors; and/or at least one sensor and preferably at least two sensors have a power supply port that is switched on together with a power supply of a controller and are therefore powered on together with the controller.

Description

BRIEF DESCRIPTION OF THE FIGURES

[0057] The figures show:

[0058] FIG. 1: a schematic drawing showing an embodiment of an inventive engine equipped with an inventive system,

[0059] FIG. 2: a diagram showing the checking of two sensors connected to separate power supply ports in case of a correct mounting;

[0060] FIG. 3: a diagram showing the checking of two sensors connected to separate power supply ports in case of an incorrect mounting in a case where the power supply port for the first sensor is not switched off after the check;

[0061] FIG. 4: a diagram showing the checking of two sensors connected to separate power supply ports in case of an incorrect mounting in a case where the power supply port for the first sensor is switched off after the check;

[0062] FIG. 5: a diagram showing the checking of three sensors connected to a common power supply port, wherein one of the sensors communicates cyclically;

[0063] FIG. 6: a diagram showing the checking routine for all the sensors of the embodiment shown in FIG. 1.

DETAILED DESCRIPTION

[0064] The general goal of the inventive strategy is to detect as soon as possible a potential inversion between sensors, in particular exhaust gas sensors (such as NOx and NH3 sensors) installed on an exhaust gas aftertreatment system of an engine, in order to ensure an optimized exhaust gas treatment.

[0065] FIG. 1 shows an embodiment of an inventive engine equipped with an inventive system. In particular, the exhaust gas aftertreatment system of the engine comprises an exhaust gas duct having several parallel paths, with a reactor, in particular an SCR catalyst, provided in each of the parallel paths. In the embodiment, four paths are provided, equipped with reactors 1 to 4. The present disclosure could also be applied to exhaust gas aftertreatment system with any other number of parallel paths.

[0066] A plurality of sensors is provided on the exhaust gas duct at predetermined positions in order to be able to control reductant injection for each of the reactors separately. The sensors communicate with a controller, in particular an ECU (engine control unit), by a serial bus such as a CAN bus.

[0067] In the embodiment shown, an NOx sensor is connected to a common exhaust gas duct of all the reactors upstream of the reactors. Further, for each reactor, an NOx sensor and an NH3 sensor is provided downstream of the respective reactor in the respective exhaust gas duct. The controller uses the sensor signals of the downstream sensors to separately control the exhaust gas aftertreatment in the respective reactor and in particular to control reductant injection into the respective exhaust gas stream.

[0068] Each sensor has a specific position at which it is to be mounted, corresponding to a position programmed into the software of the controller, such that measuring values provided by a sensor are correctly interpreted by the controller. For a correct mounting, each sensor therefore has to be mounted in the correct position on the exhaust gas aftertreatment system.

[0069] For example, in the system as shown in FIG. 1, the NOx sensor downstram 1 has to be fixed on the downstream side of SCR reactor 1 and it is the same logic for all other sensors.

[0070] The controller is coupled with electrical relays in order to switch the power supply of the sensors on and off and comprises a software routine that will check if the sensors are respectively positioned on the correct position, i.e. the correct SCR reactor, by checking whether a sensor that should be powered with a correct mounting is actually communicating.

[0071] As shown in FIG. 1, in the embodiment, the sensors NOx upstream and NH3 downstream 1 are connected to power supply port KL15, which is the controller activation signal. Details of this configuration will be described later on.

[0072] The sensors NOx downstream 1, NOx downstream 2 and NH3 downstream 2 are connected to a common power supply port that is switched on an off by relay 1.

[0073] The sensors NH3 downstream 3 and NOx downstream 3 are connected to a common power supply port that is switched on an off by relay 2.

[0074] The sensors NH3 downstream 4 and NOx downstream 4 are connected to a common power supply port that is switched on an off by relay 3.

[0075] The relays 1 to 3 are controlled by control outputs of the controller and are thereby switched on an off by the controller according to the check routine.

[0076] The system is adapted to be used with engines of different sizes. In particular, for a V12 engine, only two parallel exhaust gas ducts are present, and therefore only reactors 1 and 2 with the respective sensors. The check routine therefore ends after the check 5 indicated in FIG. 1.

[0077] For a V16 engine, three parallel exhaust has ducts are used, with reactors 1 to 3 with their respective sensors. The check routine therefore ends after the check 7 indicated in FIG. 1.

[0078] For a V20 engine, all four parallel exhaust has ducts are used, with reactors 1 to 4 with their respective sensors. The check routine therefore ends after the check 9 indicated in FIG. 1.

[0079] Before the specifics of the operation of the system shown in FIG. 1 are explained, we will first describe some fundamental properties of the system.

[0080] The disclosure is implemented to take into account a standard behavior of CAN buses such as CAN J1939 that as soon as a CAN device is powered, its corresponding information is sent to the controller and the controller can communicate with the CAN device. There is a CAN sensor particularity that is equally taken into account: [0081] Some devices send their own information by request: this kind of device needs a request from the controller in order to give information. For example, the NOx sensor might communicate in this way, i.e. by interrogation. [0082] Some devices send their own information by cycle; this kind of device sends its information every XXX ms as soon as the power supply is ON. For example, the NH3 sensor might communicate in this way, i.e. cyclically.

[0083] The controller is in particular configured to check this communication behavior in order to differentiate what kind of sensors are connected or not.

[0084] In order to distinguish which sensor is plugged in which reactor, the present disclosure uses switching of the power supply to the sensors. In order to avoid the reception of information from all sensors (NOx and NH3) at the same time, and to be able to know the reception order, the software strategy of the controller will activate the electrical relays that switch the power supply ports on and off one after the other.

[0085] Therefore, this strategy handles the electrical command relays in a precise order. The strategy is thereby able to identify a sensor which is not well mounted.

[0086] However, the CAN J1939 behavior implies that a device is always present on the CAN network as long as the device is powered. So in case of sensor inversion, only one of the two sensor will be detected as bad if the power supply ports are switched on one after the other without switching them off again after the check. In other words, sequencing the power supply of the sensors through the relays is not sufficient and one or several sensor inversions might not be detected. This is exemplified by FIG. 2 and FIG. 3.

[0087] FIG. 2 shows the case where both sensor 2 and sensor 3 are correctly mounted and connected to power supply ports switched by relays 1 and 2, respectively.

[0088] As shown in FIG. 2, the controller is configured to switch relay 1 on and to check whether sensor 2 is communicating. Then, the controller switches on relay 2 and checks whether sensor 3 is communicating. Because the controller in each case receives a message from a sensor within the expected time frame, it concludes that the sensors are correctly mounted.

[0089] FIGS. 3 and 4 show the same basic situation, but with sensor 2 and 3 being inverted and therefore incorrectly mounted. Therefore, relay 1 will power sensor 3 instead of sensor 2 and relay 2 will power sensor 2 instead of sensor 3.

[0090] FIG. 3 shows the result of the check routine if the electrical relay that is powered first is not deactivated after the check on the respective sensor has been performed: [0091] 1. Activation of relay 1. [0092] Sensor 3 powered instead of sensor 2. [0093] Sensor 2 is not communicating and therefore seen by the controller as absent and thus badly positioned. [0094] 2. Activation of relay 2. [0095] Sensor 2 powered instead of sensor 3. [0096] However, sensor 3 is still powered by relay 1 from the previous phase. [0097] Therefore, sensor 3 is still communicating with the controller. [0098] Sensor 3 is seen by the controller as present and thus well positioned.

[0099] Therefore, on two badly positioned sensors, only one is seen to be wrong.

[0100] In order to avoid this situation, in a preferred embodiment, the controller deactivates the relays after the checking phase.

[0101] The resulting check routine (with deactivation of the electrical relay) is shown in FIG. 4: [0102] 1. Activation of relay 1. [0103] Sensor 3 powered instead of sensor 2. [0104] Sensor 2 seen absent by the controller and thus badly positioned. [0105] 2. Deactivation of relay 1. [0106] Sensor 3 not powered instead of sensor 2. [0107] 3. Activation of relay 2. [0108] Sensor 2 powered instead of sensor 3. [0109] Sensor 3 seen absent by the controller and thus incorrectly positioned. [0110] 4. Reactivation of relay 1 [0111] Sensor 3 powered instead of sensor 2. [0112] Sensor 2 powered instead of sensor 3.

[0113] On two wrongly positioned sensors, both are seen wrong by the controller. Therefore, for the relays controlled by the controller, the relay is switched off again before the next check is performed.

[0114] In the above examples, more than one sensor can be connected to each relay. In this case, additional functionality is provided by the present disclosure.

[0115] The following table shows an example of correspondence between calibration and physical placement of sensors in the embodiment of a system as shown in FIG. 1:

TABLE-US-00001 Sensor check number Physical sensor Calibration Comment Sensor NOx sensor NOX_UPSTREAM_ENGINE Alternate check 1 upstream NOx and Sensor NH3 sensor NH3_DOWNSTREAM_REACTOR_1 NH3 when check 2 downstream 1 both sensor Sensor NOx sensor NOX_DOWNSTREAM_REACTOR_1 type are check 3 downstream 1 onto the Sensor NH3 sensor NH3_DOWNSTREAM_REACTOR_2 application check 4 downstream 2 Sensor NOx sensor NOX_DOWNSTREAM_REACTOR_2 check 5 downstream 2 Sensor NH3 sensor NH3_DOWNSTREAM_REACTOR_3 check 6 downstream 3 Sensor NOx sensor NOX_DOWNSTREAM_REACTOR_3 check 7 downstream 3 Sensor NH3 sensor NH3_DOWNSTREAM_REACTOR_4 check 8 downstream 4 Sensor NOx sensor NOX_DOWNSTREAM_REACTOR_4 check 9 downstream 4

[0116] The customers are informed of the sensor positioning corresponding to the software calibration. The software will then check the correct mounting according to the software calibration.

[0117] The check routine comprises several phases, as shown in FIG. 6:

[0118] After activation of KL15, sensors 1 and 2 are checked. As KL15 is the ECU activation signal, it cannot be deactivated because otherwise the ECU is switched off. Without the ECU activation signal, the ECU is only power supplied but not processing.

[0119] So all the checks, also those following sensors checks 1 and 2, are done with KL15 activated but only the sensor checks 1 and 2 are linked to this signal. If a mis-positioning is detected at this stage, the check routine is interrupted and a warning message is issued, prompting the user to check sensors 1 and 2 and to correct the mis-positioning before the check-routine is started again or continued for the remaining sensors.

[0120] Therefore, if a positioning error is seen on the sensor checks 1 or 2, the strategy breaks which means that it is stopped and an error is risen. Then the customer has to correct the sensor 1 and 2 positions, and switch on the ECU again in order to continue the check for all other sensors. In other words, the strategy is only complete after a minimum of 2 ECU power on if a positioning error is present on sensor check 1 or 2. This implementation is used in order to cope with a lack of free pins on the controller.

[0121] After activation of relay 1, sensors 3, 4 and 5 are checked. Then, relay 1 is switched off before relay 2 is switched on and sensors 6 and 7 are checked. Then, relay 2 is switched off before relay 3 is switched on and sensors 8 and 9 are checked.

[0122] In the above method, because more than one sensor is connected to each relay, an inversion for sensors on the same relay cannot be detected by switching power on and off. However, some additional measures help to improve the checking functionality.

[0123] In particular, a sensor ID is given to a group of two sensors that can be differentiated by the SW by wiring. For the same ID, a pin select (adding an electrical mass) changes the source address of the sensor. This can for example be done for the NOx sensors.

TABLE-US-00002 ID Sensor without pin select Sensor with pin select X NOx upstream NOx downstream 1 Y NOx downstream 2 NOx downstream 3 Z NOx downstream 4 NOx downstream 5

[0124] Therefore, according to a predefined correct mounting that is checked by the controller, relays are connected onto the sensors according to the rule that two sensors cannot be connected to the same electrical relay if their ID is the same.

[0125] This will for example result in the following strategy for the NOx-Sensors:

TABLE-US-00003 Phase (represented Sensor check by relay) number Physical sensor Calibration 1 Sensor check NOx upstream NOX_UPSTREAM_ENGINE 1 Sensor check NOx NOX_DOWNSTREAM_REACTOR_3 2 downstream 3 2 Sensor check NOx NOX_DOWNSTREAM_REACTOR_2 3 downstream 2 Sensor check NOx NOX_DOWNSTREAM_REACTOR_1 4 downstream 1 3 Sensor check NOx NOX_DOWNSTREAM_REACTOR_4 5 downstream 4

[0126] In this way, the positioning error is physically limited, and the strategy helps to detect an inversion.

[0127] Further, the SW strategy can also detect an inversion between NH3 and NOx sensors due to the different communication behavior. Therefore, only one NH3 sensor is connected to each power supply port.

[0128] Further, as shown in FIG. 5, if two NOx-sensors are connected to a power supply port together with an NH3 sensor, the first NOx-sensor is checked, then the software routine waits for the next cyclic communication by the NH3 sensor and then the next NOx sensor is checked. Therefore, between two checks on the NOx sensors, the controller waits for a communication from the NH3 sensor in order to avoid that the check on the second NOx sensor is corrupted by the cyclic communication by the NH3 sensor. The NH3 sensor can be checked on any receipt of a cyclic message.

[0129] Further, on a more general level, NOx sensors and NH3 sensors can be checks in alternation. The reason is to avoid a NH3 sensor answer in place of another NH3 sensor as they send cyclic message. The alternating order makes sure that there is a NOx sensor between two NH3 sensors which sends its information after a request.

[0130] The check routine of the present disclosure is particularly robust and fast because it checks a predetermined configuration. In this way, the controller searches for what is expected in case of a correct mounting and if it does not receive the expected answer after a certain amount of time knows that there is a mounting error for this position, and can remove an unknown parameter from the equation.