Control of the dwell time of ignition coils in spark-ignition internal combustion engine

Abstract

A method for controlling an ignition coil for a spark plug of an internal combustion engine, wherein the ignition coil is associated with a respective engine cylinder in which combustion is to occur, wherein the ignition coil is driven by a spark demand pulse. The method comprises determining a spark demand pulse by determining an end time point of the spark demand pulse, determining a spark demand pulse length based on a historic battery voltage profile of the associated battery voltage, determining a start time point of the spark demand pulse based on the determined end time point and the determined spark demand pulse length, driving the ignition coil with the determined spark demand pulse. A benefit of the invention is that a spark demand pulse is generated that is more appropriate for the prevailing battery voltage at the time the demand pulse is generated. Therefore, the process provides a more reliable approach for controlling the generation of spark at a spark plug as controlled by an ignition coil under conditions where the battery voltage may be compromised, for example for an aged battery and / or during cold engine conditions.

Description

[0001] PH24078-1

[0002] IMPROVEMENTS RELATING TO CONTROL OF IGNITION

[0003] COILS IN SPARK-IGNITION INTERNAL COMBUSTION ENGINE

[0004] 5 Technical Field

[0005] This disclosure relate to improvements in control of ignition coils for spark-ignition internal combustion engines (ICEs). Such ICEs may be powered by a variety of fuels that are liquid or gaseous at room temperature, such as gasoline or hydrogen.

[0006] 10

[0007] Background

[0008] Fuel injectors are integral components of a modern internal combustion engine, responsible for delivering fuel into the engine cylinders in a controlled and efficient manner. Fuel injectors work closely with ignition coils to ensure proper engine operation.

[0009] An ignition coil is responsible for converting the vehicle battery voltage (typically 12V in an ICE vehicle) into the high voltage needed to create a spark at the spark plug. The spark generated by the spark plug ignites the air-fuel mixture in the cylinder.

[0010] Ignition coils generally consist of two windings or coils: a primary coil and a secondary coil. Energizing the primary coil generates a magnetic field. When the current to the primary coil is interrupted, the magnetic field collapses which induces a high voltage in the secondary coil which is then delivered to the spark plug thereby generating a spark.

[0011] 25

[0012] The dwell time (or charge time) refers to the duration for which the primary coil is energized. During the dwell time, the primary coil stores energy by virtue of the magnetic field. The longer the dwell time, the more energy the coil can store, resulting in a more powerful spark. However, excessively long dwell times can cause premature again of the ignition coils. Proper control of

[0013] 30 the dwell time is crucial to ensure that the ignition system performs efficiently and the engine operates smoothly.

[0014] Low battery voltage and low temperatures can have an adverse effect on the performance of ignition coils. In the case of low battery voltage, typically caused by an aging or undercharged battery, the voltage available to charge the ignition coil during the dwell time may be insufficient, and this is particularly the case during engine cranking when the battery is being heavily loaded by the starter motor. This can lead to an incomplete charging process, where the ignition coil is unable to generate the necessary high voltage to produce a spark. This can cause a misfire, leading to engine performance issues, poor fuel efficiency, and higher

[0015] 40 emissions. Low temperatures can exacerbate these problems, as they reduce the ability of the battery to supply sufficient voltage. At colder temperatures, the viscosity of the engine oil increases, requiring more energy for starting and additional strain on the electrical system. As a consequence, the available voltage for the ignition system may be lower than required, resulting in a reduced dwell time and potentially weak or inconsistent spark production.

[0016] It is with these issues in mind that the embodiments of the invention have been devised.

[0017] Summary of the Invention

[0018] Against this background, according to an aspect of the invention, there is provided a method for controlling an ignition coil for a spark plug of an internal combustion engine, wherein the ignition coil is associated with a respective engine cylinder in which combustion is to occur, wherein the ignition coil is driven by a spark demand pulse. The method comprises determining a spark demand pulse by determining an end time point of the spark demand pulse, determining a spark demand pulse length based on a historic battery voltage profile of the associated battery voltage, determining a start time point of the spark demand pulse based on the determined end time point and the determined spark demand pulse length, driving the ignition coil with the determined spark demand pulse.

[0019] The examples of the invention may also be considered to reside in an engine system comprising at least one piston movable within an associated engine cylinder, and a spark plug associated with the cylinder, the spark plug coupled to an ignition coil, wherein the ignition coil is driven by a spark demand pulse, the engine system having a control unit configured to carry out the steps as defined above.

[0020] A benefit of the invention is that a spark demand pulse is generated that is more appropriate for the prevailing battery voltage at the time the demand pulse is generated. Therefore, the process provides a more reliable approach for controlling the generation of spark at a spark plug as controlled by an ignition coil under conditions where the battery voltage may be compromised, for example for an aged battery and / or during cold engine conditions.

[0021] It is envisaged that the most benefit may be achieved by performing the methodology whilst the engine is in the cranking mode of operation. That is, when the engine is not producing power but is being driven by a starter motor in order to initiate self-sustaining operation. Therefore, references to “spark demand pulse” and “dwell period” may be understood to be those features occurring during a cranking mode of operation, for example a “cranking dwell period” or a “cranking spark demand pulse”. In this process, the historic battery voltage profile may reference vehicle battery voltage against time in respect of at least one previous combustion cycle of the associated cylinder, or another cylinder of the engine. Moreover, the historic battery voltage profile records vehicle battery voltage against time for a time window proximate to a top dead centre position (TDC) in respect of the combustion cycle of the associated cylinder, or another cylinder of the engine. By proximate to the TDC position of a cylinder, it is meant the time period represented by less than 100 degrees of crank angle, or less than 80 degrees of crank angle, or less than 50 degrees of crank angle. Preferably, the time window may extend from between 50 and 80 degrees of crank angle, and preferably between 60 and 70 degrees of crank angle to around the TDC position, and preferably just after the TDC position, for example between 0 and 30 degrees after TDC.

[0022] In one example, the spark demand pulse length is determined by calculating an ignition current value (lend) based on the historic battery voltage profile, and an associated current draw characteristic associated with the ignition coil, integrated over time.

[0023] The spark demand pulse length may also be determined with reference to one or more further engine ignition coil characteristics, such as its electrical resistance and electrical inductance.

[0024] The historic battery voltage profile that is used in the process may be a recorded set of voltage values during a specific time period in a previous cylinder event that occurred immediately prior to the cylinder event for which the spark pulse length is being determined. However, this is not essential, and the historic battery voltage profile may apply to a cylinder event that is a predetermined number of cylinder events prior to the current cylinder event. In this case, the historic battery voltage profile may be refreshed after a predetermined number of cylinder cycles.

[0025] In one example, the method may include applying a voltage compensation to the historic battery voltage profile to compensate for reduced vehicle battery voltage between cylinder combustion cycles. Therefore, if the measured voltage reference value, having been compared with the voltage value in the historic battery voltage provide at the equivalent time point in the cylinder event, is different, then a suitable compensation can be made to the historic battery voltage provide and any values that it provides. For example, a comparable voltage offset can be applied to all voltage values in the historic battery voltage profile. This feature is useful because it can factor in changes in battery voltages between the time at which calculations are being performed and the time when the historic battery voltage profile was recorded. The voltage compensation may be derived from a voltage measurement obtained at a reference position from a current cylinder combustion cycle. The end point of the spark demand pulse may be determined based on the receiving of an ignition timing signal. The ignition timing signal may be transmitted by a higher level engine control subsystem with the responsibility for determining the ignition timing of the engine.

[0026] In some examples, the method may include monitoring for a spark confirmation signal following termination of the spark demand pulse, and flagging a no-spark event if a spark confirmation is not received. This may provide further feedback information that a spark has been generated correctly. In the event that a no spark event is flagged, applying a further temporal increase to the spark demand pulse length for a subsequent combustion event.

[0027] The control unit may take the form of an engine control unit of the engine.

[0028] The examples of the invention may also be expressed as a control unit comprising a processor or processing means configured to perform the method as defined in any of the above-defined features, or a computer program product comprising instructions which, when the program product is executed by a computer, cause the computer to carry out the method as in the above-defined features, or a computer-readable storage medium or a data carrier comprising instructions, for example int eh form of a computer program product, which when executed by a computer, cause the computer to carry out the method of any of the above-defined features.

[0029] Further optional and advantageous features are referenced in the detailed description and the appended claims.

[0030] Brief Description of the Drawings

[0031] Examples of the invention will now be described with reference to the following figures:

[0032] Figure 1 is a schematic view of a simplified engine system in which the examples of the invention may be incorporated, where it should be noted the engine system shown provides a view of a single cylinder of which may be a multi-cylinder engine;

[0033] Figure 2 is a set of timing plots showing the conventional timing of a spark demand pulse or ‘dwell period’ (a), a coil current characteristic (b) associated with the spark demand pulse, and the spark generation (c) that is generated by the spark demand pulse,

[0034] Figure 3 is a flow chart depicting an example of a methodology for determining the duration and timing of a spark demand pulse in accordance with the invention, Figure 4 is a timing plot illustrating a measured battery voltage profile;

[0035] Figure 5 is a flowchart illustrating a subprocess of Figure 3;

[0036] Figure 6 is a set of timing plots showing a vehicle battery voltage profile (a) of a previous cylinder event that is stored in ECU memory, a spark demand pulse (b), a coil current characteristic (c) associated with the spark demand pulse, and a spark generation (c) that is generated by the spark demand pulse.

[0037] Detailed Description

[0038] In general, the examples of the invention provide strategies for controlling an ignition coil of a spark-ignition internal combustion engine. The strategies may provide benefits in terms of improving the reliability with which the ignition coil can energise a spark plug to generate a spark, particularly in conditions where battery voltage is relatively low, for example through an aged battery or reduced ambient temperatures.

[0039] Figure 1 depicts a simplified engine system 2 in schematic form which incorporates typical engine components that are relevant to the examples of the invention. It should be noted that a typical engine system would include many more components than is shown in Figure 1 as would be well understood by a person skilled in the art. However, a full discussion of a complete engine system is not required here so has been omitted for brevity.

[0040] The engine system 2 comprises a cylinder 4. A piston 6 is received slidably within the cylinder 4 in the usual manner. The piston 6 is connected to one end of a connecting rod 8 at a gudgeon pin 9, and the other end of the connecting rod 8 is connected to a rotatable crank 10 at a crank pin 12, which are located at a lower end of the cylinder 4, as shown in the drawing.

[0041] Although not shown, the cylinder 4 is provided in an engine block 14. Although a single cylinder 4 is shown in Figure 1 , it should be noted that this is for simplicity of description and a practical implementation would include a multi-cylinder engine.

[0042] A cylinder head 16 is provided at the upper end of the cylinder 4. Together, the cylinder head 16 and the piston 6 define a combustion chamber 18 in which combustion occurs when the piston 6 is at or near a top dead centre (TDC) position.

[0043] The cylinder head 16 provides an inlet duct 20 and an outlet duct 22. An inlet valve 24 is positioned to control the flow of a fuel air mixture into the combustion chamber 18 from the inlet duct 20. The flow of fuel / air mixture along the inlet duct 20 is controlled by a throttle valve 26. The throttle valve 26 is controllable through any conventional means as would be well understood by a skilled person. Likewise, the delivery of the air / fuel mixture into the inlet duct 20 is provided by a suitable fuel delivery system (not shown) that is not directly relevant to the invention and would be well understood by the skilled person. Therefore, a full discussion of such subsystems and components is not required here.

[0044] An outlet valve 28 is configured to control the flow of exhaust gas from the combustion chamber 18 into the outlet duct 22.

[0045] Movement of the inlet valve 24 is controlled by an inlet cam member 30, and movement of the outlet valve 28 is controlled by an outlet cam member 32. The cam members 30,32 act on the respective valves 24,28 in a manner that is known in the art, so a full discussion of the mechanism of operation will not be provided here.

[0046] A spark plug 34 is positioned within the cylinder head 16 such that a tip end of the spark plug 34 protrudes into the combustion chamber 18. In this way, the spark plug 34 is operable to ignite an air / fuel charge that is delivered to the combustion chamber 18.

[0047] As is known, the piston 6 moves between a TDC position and a bottom dead centre (BDC) position within the cylinder 4 as is driven by expanding gases in the combustion chamber 18 and, in doing so, rotates the crank 10, that is connected to a crank shaft (not shown). The TDC position is the position in which the piston defines a minimum volume of the combustion chamber 18 (i.e. the piston’s closest position to the spark plug 34). Conversely, the BDC position is the position of the piston 6 when it defines a maximum volume of the combustion chamber 18 (i.e. the piston’s furthest position away from the spark plug 34).

[0048] A cycle of the piston such that the piston 6 travels from a BDC position, to the TDC position, and back to the BDC position is considered to be half of a full cycle in a 4-stroke ICE.

[0049] The spark plug 34 is coupled to an ignition unit 40. The ignition unit 40 incorporates the functionality of an ignition coil and associated control circuity required for operation. The ignition unit 40 is coupled to a vehicle battery 42 for the provision of a low voltage supply.

[0050] The ignition unit 40 is responsible for supplying the spark plug 10 with the high-voltage electrical charge necessary for spark generation. As is well known, the ignition unit 40 incorporates a transformer that steps up the voltage from a vehicle battery 42 the vehicle’s 12- volt battery to the tens of thousands of volts required to create a spark. Although the following components aren’t shown in Figure 1 , it will be understood that the ignition unit 40 consists of two windings: a primary winding, which is coupled to battery voltage via an associated control circuit, and a secondary winding which is connected to the spark plug 10. When current flows through the primary winding, a magnetic field is created. When the current is interrupted, the magnetic field collapses, inducing a high-voltage current in the secondary winding. This high- voltage pulse is communicated to the spark plug 10 to produce the spark at its tip, known as the spark gap.

[0051] The timing of the spark, referred to as spark timing, is crucial for the correct performance of the engine. The spark must occur at the correct moment in the cylinder cycle to ensure optimal combustion. In most engines, this timing is adjusted to occur just before the piston 6 reaches the top of its compression stroke (at TDC). This is known as spark advance.

[0052] At TDC, the piston 6 is at its highest point in the cylinder 4. Ideally, the spark should occur slightly before TDC, allowing time for the air-fuel mixture to ignite and begin expanding as the piston 6 starts its downward stroke. If the spark occurs too early (too much advance), the mixture may ignite before the piston 6 has reached the optimal compression, which can lead to engine knock or damage. Conversely, if the spark occurs too late (too much spark retard), the fuel may not burn efficiently, leading to a loss of power and increased emissions.

[0053] Modern engines typically use a timing control system (based on vehicle sensor data) to adjust the timing dynamically, depending on factors such as engine load, speed, and temperature. This ensures that the spark occurs at the most optimal point in each cycle, enhancing performance, fuel efficiency, and reducing harmful emissions.

[0054] In summary, the spark plug 34 and ignition unit 40 work together to initiate combustion in the cylinder 4, with the ignition unit 40 providing the necessary high-voltage current to the spark plug 10. The precise timing of the spark, closely tied to the position of the piston 5 in a particular cylinder cycle, is essential for engine efficiency and performance.

[0055] The ignition unit 40 is also coupled to a control unit 44. The control unit 44 in this example of the invention provides the controlling input to the ignition unit 40 in order to cause it to charge the spark plug 34 to generate a spark. The control unit 44 may be a dedicated control unit, or may represent functionality of the main engine control unit (ECU) of the vehicle. The control unit 44 is adapted to receive data input 46 to sense operational parameters of the engine to provide suitable control output signals to the ignition unit 40 as discussed above, for example providing a drive signal 48. The control unit 44 also receives a timing signal 47 from a crank angle sensor 49. In this way, the control unit 44 has a precise measurement of the position of the pistons within the respective cylinders so appropriate drive signals 48 can be determined. The control unit 44 includes a memory component 50. The memory component 50 stores data such as self-learnt control parameters and operating history data, as will be described in more detail later, which can be retrieved by the control unit 44 even after a power down cycle.

[0056] The control unit 44 may be operable to perform various engine monitoring and control objectives to manage the performance of the vehicle or power plant system into which it is installed. It should be appreciated that the control unit 44 may be any suitable control environment provided by the engine system 2. The control unit 44 may be the “engine ECU” of the engine system 2 or it may be another control unit which is configured to carry out other performance and monitoring tasks within the engine system 2 of the broader vehicle or power plant system. In particular, the control unit 44 may be a control environment provided specifically for the purposes of performing the method as discussed in this disclosure.

[0057] Irrespective of the functionality of the control unit 44, it will be appreciated that the control unit 44 has the necessary memory 50, processing environment 52 and communications interface 54 to be integrated into the engine system 2 and the broader system of an associated vehicle or power plant system.

[0058] Having described the general overview of the engine system 2, the discussion will now focus on a methodology that is implemented by the control unit 44 for controlling the ignition unit 40 and, thus, the spark generation of the spark plug 34.

[0059] Figure 2 illustrates a conventional spark demand pulse, SDP, which goes high for a predetermined time period, being around 3ms-6ms in circumstances where there is a good battery voltage available at moderate temperature conditions. The rising edge T1 of the spark demand pulse SDP represents the point at which battery voltage is applied to the ignition unit 40, and the falling edge T2 represents the point at which the applied voltage is terminated, thereby triggering a spark at the spark plug 34.

[0060] The duration of the spark demand pulse SPD is the temporal separation between T1 and T2. During this time period, oftentimes referred to as the ‘dwell period’, it can be seen that the coil current increases, and then terminates abruptly at T2, following which a spark is generated.

[0061] The timing of the spark generation process is determined in the usual way by the control unit 44 in a manner that would be familiar to a person skilled in the art, as the spark timing is linked to the operation of the engine based on engine speed, engine position (as determined by the crank angle sensor 49) torque demand, and other operational engine parameters, the detail of which is beyond the scope of this discussion. In ordinary circumstances when battery voltage is at nominal levels and at moderate temperatures, the energy that is accumulated in the ignition unit 40 over the dwell period is sufficient to generate a spark at the spark plug 34. It is customary to calculate the dwell period based on an average of the battery voltage for a given period in advance of the spark, typically greater than 100ms in advance of spark timing. This measure goes some way to compensate for a reduced battery voltage level since the calculations can be made to increase the dwell period if the battery voltage is deemed to be too low. However, it has been observed that this process can be fallible in some circumstances.

[0062] In accordance with an example of the invention, a control method 100 for the ignition unit 40, as shown in Figure 3, executes during engine operation in respect of each cylinder of the engine. The control method takes place during engine cranking, where the ignition unit 40 is fed with a voltage level which is determined by the battery 42 of the vehicle, and so is therefore more dependent on the voltage level than during normal operation during which the nominal 12V voltage level (running voltage usually between 13-14.5V) is held constant due to operation of the alternator.

[0063] The methodology discussed below is executed for each cylinder event in a succession of cylinder events during which a spark is generated by the spark plug 34 in order to trigger combustion within the cylinder as it approaches TDC position during a compression stroke. Therefore, in a four stroke engine running at approximately l OOOrpm, the method may have an execution interval of approximately 120ms, In a 4-cylinder 4-stroke engine, then the logic may run every 30ms (or equivalent crank angle timing in the rotational crank angle domain) as it needs to perform the same calculation for each cylinder. It should be appreciated that these values are merely exemplary. In practice, during cranking the crank speed will be lower than OOrpm so the execution interval may be increased compared to what has been mentioned here.

[0064] In general, the method 100 achieves the objective of determining the characteristics of a spark demand pulse. A spark demand pulse may be understood as being the voltage pulse that is applied to the ignition unit 40 (and, more specifically, the primary coil thereof), to charge the primary coil prior to rapid termination of the voltage pulse which triggers the high voltage supply to the spark plug 34 which generates a spark.

[0065] As a first step, the method determines whether a stored battery profile is available in memory 50. If there is no stored battery profile available, then the process continues to step 104 for the use of an alternative dwell period determination process. This may be a conventional process which would be well known to the skilled person. For example, an average battery voltage may be determined for a time prior in advance of the spark timing in order to determine the dwell period. Since this alternative process may be conventional, a full discussion will not be provided here.

[0066] If a stored battery profile is available, the process moves onto step 106. At step 106, the method calculates the required characteristics of a dwell period or ‘spark demand pulse’, in accordance with a novel methodology. Notably, the methodology is based in the determination of the available charging power that is available for charging the ignition unit 40 based on historical data of the battery voltage. Expressed in another way, therefore, the method may include the step of determining the available charging power, of an associated engine battery, for charging the ignition coil. By the term ‘historic’, it is meant that the battery voltage profile is used which relates to a previous cylinder event relating to the engine cylinder for which the spark demand pulse is being determined. Furthermore, it is acceptable to use a battery voltage profile for a previously-firing cylinder, which may not necessarily be the same cylinder for which the calculations are being performed.

[0067] An example of vehicle battery voltage profile 58 is shown in Figure 4. The vehicle battery voltage profile 58 is defined by a sampling window 60. The sampling window 60 shows the vehicle battery voltage that has been sampled at a high sampling rate for a predetermined time period during a previous cylinder event on the respective cylinder for which the dwell time is being calculated.

[0068] The sampling time of the sampling window 60 may be between 0.5 and 2ms, and it is envisaged that 1 ms would be particularly suitable.

[0069] The length of the sampling window 60 may be suitably selected in order to give a useful spread of voltage values. It is envisaged that a sampling window having a length extending from a short time after the TDC position of the cylinder to slightly after the BDC position of the cylinder would be appropriate. This represents a time length of potentially 180-230 degrees of crank angle, although it is envisaged that a crank angle of approximately 60 to 70 degrees before TDC should be sufficient.

[0070] It will be appreciated that the vehicle battery voltage profile 58 reduces from about 8V at a BDC position of the cylinder to approximately 6V at a TDC position, during cranking when the alternator isn’t running to support the battery voltage. It should be noted that these voltage values are just exemplary.

[0071] The vehicle battery voltage profile 58 is recorded during a previous cylinder event and stored in the memory 50 of the control unit 44. The stored battery voltage profile therefore represents a historic battery voltage profile because it relates to a previous cylinder event in relation to a cylinder event for which a spark demand pulse duration is being calculated by the methodology that will be discussed below. Thus, the stored vehicle battery profile 58 is used in determining the required characteristics of the spark demand pulse, as mentioned above at step 106.

[0072] An example sub-process is shown in Figure 5 which expands the process within step 106.

[0073] At step 202, the end time point of the dwell period is determined. The end time point is dependent on the spark timing derived from general vehicle operations, since the end time point marks the time at which a spark is required to be generated at the spark plug 34.

[0074] The steps that follow are based on an algorithmic approach to calculating the length of the spark demand pulse based on the energy that is available for the ignition unit 40 to generate a pulse from the vehicle battery which may vary based on its voltage level. Since the battery voltage is unknown, it is desired to obtain a more accurate prediction of battery voltage, thereby to derive a more accurate calculation of the current that will be supplied to the ignition unit 40 when the voltage is applied.

[0075] In this respect, since the ignition unit 40 can be compared to an LR circuit, having a specified resistance R and inductance L. Accordingly, at step 204, the electrical characteristics L and R for the ignition coil 40 are obtained. The values may be stored in memory component 50, for example in flash ROM of the control unit 44 in suitable calibration tables.

[0076] At step 206, the energy required in the ignition unit 40 at the end of a charge period may be determined. This can be a stored value in the memory component 50 in a suitable calibration table, since it is generally an understood parameter for the required energy of the ignition coil 40 to generate a spark successfully, although the stored value may suitably be adjusted by appropriate compensation strategies.

[0077] Once the energy required to charge the ignition unit 40 is known, an algorithmic approach to derive the duration of the spark demand pulse can be determined in the following manner.

[0078] The voltage across the ignition coil 40 can be expressed as:

[0079] Vs = VR+VL (1)

[0080] Equation (1) can be expressed in terms of the obtained electrical characteristics of the ignition coil (L and R) as:

[0081] Vs (t) = l(t)R + Ldl(t) / dt (2) So, the equation for charging of the ignition unit 40 can be expressed as: dl(t) / dt = Vs(t) / L - l(t)R / L (3)

[0082] The charging period of the ignition coil 40 results in energy being stored in the primary coil of the ignition unit 40 at the end of the charging period. The stored energy at the end of the charging period can be expressed as:

[0083] E = -2LI^d(4)

[0084] Therefore, it can be appreciated that the charging current at the end of the charging period determines the energy stored in the primary coil of the ignition unit 40. So, charging starts when current is at 0A and then increases to an end current level (lend) at the end of the dwell period (tend).

[0085] Form this, equation (3) can be rearranged as:

[0086] In equation (5), ^(t.) is the historic vehicle battery voltage profile of a previous cylinder event (as is shown in Figure 3), At is the sample interval (e.g. 1 ms sampling rate in the sampling window 60), R and L are known characteristics of the ignition unit 40 which are functions of temperature which can be looked up from calibration tables.

[0087] So, using equation (5) and starting from an initial value of current that corresponds to the required energy in the ignition unit 40, the equation can be iterated from lend to l=0 to determine the length of the spark demand pulse. This process is represented by step 208.

[0088] Once the duration of the spark demand pulse has been determined, the process moves to step 208, at which the start time point of the spark demand pulse can be set based on the end time point and the calculated duration. This process can be seen pictorially in Figure 6. The end time point is a defined point as set by the required spark timing. The charge duration is calculated based on the historic vehicle battery voltage profile, shown at the top of Figure 6, and as described in detail above. Since the battery voltage is relatively low, the charge duration has been extended compared to what a ‘healthy’ battery voltage would be. This is seen as the time period AT which extends the rising edge of the spark demand pulse or the ‘start time point’.

[0089] The resultant charge duration is therefore extended compared to the case where the battery voltage is at a higher value. The benefit of this is that the process compensates for lower battery voltage levels that would otherwise compromise the ability of the ignition unit 40 to charge to a sufficient level in order to generate a spark.

[0090] Once the duration of the spark demand pulse and the start time point have been determined, the subprocess ends and returns back to method 100 at step 108 (Fig 3) at which point the ignition unit 40 is charged with the determined values, thereby to generate a spark at the spark plug 34 at the required ignition timing.

[0091] It will be appreciated that various adaptations may be made to the specific implementations described above without departing from the inventive concept as defined by the claims.

[0092] In one possible adaptation, the historic battery volage provide that is used in the above process may be modified by the use of a reference voltage measurement that is taken at the beginning of the calculation period. Therefore, if the measured voltage reference value, having been compared with the voltage value in the historic battery voltage provide at the equivalent time point in the cylinder event, is different, then a suitable compensation can be made to the historic battery voltage provide and any values that it provides. For example, a comparable voltage offset can be applied to all voltage values in the historic battery voltage profile. This feature is useful because it can factor in changes in battery voltages between the time at which calculations are being performed and the time when the historic battery voltage profile was recorded.

[0093] In some examples, it is envisaged that the method may include monitoring for a spark confirmation signal following termination of the spark demand pulse. Some vehicle ignition systems may thus be provided with a built-in feedback path that confirms the generation of a spark. Such a system has the ability to flag a no-spark event if a spark confirmation is not received. Beneficially, in the event that a no spark event is flagged, the method may be adapted to apply a further temporal increase to the spark demand pulse length for a subsequent combustion event.

[0094] In the above described methodology, it should be appreciated that the historic battery voltage profile is used to provide a more accurate determination of the energy with which the battery is able to apply charging current to the ignition coil. The historic battery voltage profile may be used in other approaches to determine the dwell duration and, thus, an appropriate start time point for the spark demand pulse.

[0095] For example, the historic battery voltage profile could be used for the purpose of calculating the average voltage over a standard dwell period and then adjusting the ‘standard’ dwell period so that the calculated average voltage converges on a reference voltage that has been selected as the voltage that would apply at the end of the dwell period for a battery with an acceptable voltage level (i.e. not aged, or operating under a low temperature condition). In this way, therefore, the historic battery voltage profile is used to adjust a conventionally-derived dwell period to account for the reduced energy availability in the affected battery.

[0096] In another example, a suitable methodology that also uses a historic battery voltage profile applying to a cylinder event during cranking prior to a TDC position, and may comprise the following steps.

[0097] Firstly, determining the end time point of the dwell period is determined based on provided spark demand timing. Secondly, the voltage at the end time point of the dwell period can be predicted using the stored historic battery voltage profile, which can be characterised as Vend. The end voltage Vend can therefore be adjusted based on a measured voltage reference proximate to the start of the dwell time (e.g. between 60 and 70 degrees of crank angle prior to the dwell time) and compared with the voltage at the same time point in the historic battery voltage profile of an associated cylinder, which can be characterised as Vref. The two determined voltage levels Vref and Vend can thereafter be used to determine an average or aggregate, average or compound voltage level (Vavg) using a weighting factor, W, for example using the expression Vavg= Vref * W + Vend * (1-W). The weighting factor may be a calibratable constant that can be determined offline. The calculated voltage level Vavgmay thereafter be used in a standard dwell period mapping or calibration table which associated voltage with temperature, current and a dwell constant k. Such a dwell period map would be well understood by a skilled person.

Claims

Claims1 . A method for controlling an ignition coil for a spark plug of an internal combustion engine, wherein the ignition coil is associated with a respective engine cylinder in which combustion is to occur, wherein the ignition coil is driven by a spark demand pulse, the method comprising determining a spark demand pulse by: determining (202) an end time point of the spark demand pulse, determining (206) a spark demand pulse length based on a historic battery voltage profile of the associated battery voltage, determining (208) a start time point of the spark demand pulse based on the determined end time point and the determined spark demand pulse length, driving the ignition coil with the determined spark demand pulse.

2. The method of Claim 1 , wherein the historic battery voltage profile references vehicle battery voltage against time in respect of at least one previous combustion cycle of the associated cylinder, or another cylinder of the engine.

3. The method of Claim 2, wherein the historic battery voltage profile records vehicle battery voltage against time for a time window proximate to a top dead centre position (TDC) in respect of the combustion cycle of the associated cylinder, or another cylinder of the engine.

4. The method of Claims 2 or 3, wherein the spark demand pulse length is determined by calculating an ignition current value (lend) based on the historic battery voltage profile, and an associated current draw characteristic associated with the ignition coil, integrated overtime.

5. The method of any of Claims 2 to 4, wherein the available charging power is determined also with reference to one or more further engine ignition coil characteristics.

6. The method of any one of Claims 2 to 4, further comprising: applying a voltage compensation to the battery voltage profile to compensate for reduced vehicle battery voltage between cylinder combustion cycles.

7. The method of Claim 6, wherein the voltage compensation is derived from a voltage measurement obtained at a reference position from a current cylinder combustion cycle.

8. The method of any one of the preceding claims, wherein the end point of the spark demand pulse is determined based on the receiving of an ignition timing signal.

9. The method of any one of the preceding claims, wherein the time duration of the spark demand pulse length is the temporal separation between the start time point and the end time point.

10. The method of any one of the preceding claims, further comprising: monitoring for a spark confirmation signal following termination of the spark demand pulse, and flagging a no-spark event if a spark confirmation is not received.

11. The method of Claim 10, wherein, in the event that a no spark event is flagged, applying a further temporal increase to the spark demand pulse length for a subsequent combustion event.

12. The method of Claim 12, wherein the further temporal increase to the spark demand pulse length is achieved by advancing the start time point of the spark demand pulse for a subsequent combustion event.

13. An engine system comprising at least one piston movable within an associated engine cylinder, and a spark plug associated with the cylinder, the spark plug coupled to an ignition coil, wherein the ignition coil is driven by a spark demand pulse, the engine system having a control unit configured to carry out the steps of: determining (202) an end time point of the spark demand pulse, determining (206) a spark demand pulse length based on a historic battery voltage profile of the associated battery voltage, determining (208) a start time point of the spark demand pulse based on the determined end time point and the determined spark demand pulse length,driving the ignition coil with the determined spark demand pulse.

14. A control unit comprising a processor configured to perform the method of any one of Claims 1 to 12.

15. A computer program product comprising instructions which, when the program product is executed by a computer, cause the computer to carry out the method of any one of Claims 1 to 12.

Images