GB2440317A

GB2440317A - Functionality testing of secondary air injection systems

Info

Publication number: GB2440317A
Application number: GB0614834A
Authority: GB
Inventors: Sam Price
Original assignee: Ford Global Technologies LLC
Current assignee: Ford Global Technologies LLC
Priority date: 2006-07-27
Filing date: 2006-07-27
Publication date: 2008-01-30
Anticipated expiration: 2026-07-27
Also published as: GB0614834D0; GB2440317B

Abstract

A method and system for detecting faults in a secondary air injection system 10 of an internal combustion engine 11. The system has an air pump 16 for supplying air to the engine exhaust manifold 13 and, a programmable unit 21 for controlling operation of the pump and valve and a pump current draw sensor 31 for sampling electrical current at predetermined time periods spaced over at least on operational cycle of the system. The measured pump current is analysed by the control unit 21 to identify failures in the system. Current draw is measured in three phases of the operational cycle of the system; a first phase with the pump on and the valve closed, a second phase with the pump on and the valve on and a third phase with the pump off and the valve closed.

Description

Functionality TestinQ of Secondary Air Iniection Systems The present

invention relates to secondary air injection systems which are used on internal combustion engines to provide filtered air flow into the exhaust manifold Secondary air supply systems are used to provide a supply of air into exhaust system manifolds during the cold start up period. The hot un-burnt fuel particles leaving the combustion chambers mix with the injected air and heat up due to an exothermic reaction.

This leads to the combustion of un-burnt hydrocarbons and carbon monoxide that are present during the warm-up period. This combustion process increases the temperature of the exhaust gases and provides increased heating for the catalytic converter as the gases pass through. The increased heat generated in the catalytic converter reduces the time delay before the catalyst in the converter reaches its efficient operating temperature and therefore reduces exhaust gas pollutants Federal on board diagnostics legislation requires the detection of any fault with the secondary air injection (SAl) system that might lead to the vehicle emissions exceeding specified levels. One method of diagnosing faults in the SAl system is disclosed in US 6,497,094 which utilises readings of an oxygen sensor located upstream of a catalytic converter. The SAl system typically comprises an air pump with a filtered air connection to a control valve which is operated by an engine management system. The control valve regulates the air flow into the exhaust manifolds and prevents exhaust gases from blowing back into the air pump US 6,983,590 discloses a fault diagnosis method which utilises a pressure sensor mounted in the conduit between the air pump and the control valve and is used to take pressure measurements during several different operating phases in order to determine the condition of the SAl system. This "pressure sensor" based diagnostic strategy is utilised for fault diagnosis in some prior art vehicles and has the disadvantage that the diagnostic method requires the use of a separate pressure sensor to determine the correct operation of the air pump.

The present invention provides a method of detecting faults in a secondary air injection system without a dedicated pressure sensor.

According to the present invention there is provided a method of detecting faults in a secondary air injection (SAl) system having an air pump for supplying air to an exhaust manifold of an internal combustion engine and having a control valve controlling the air supply into the manifold and a programmable control unit for controlling operation of the pump and valve, wherein the method comprises the steps of: sampling electrical current draw by the air pump at predetermined time periods spaced over at least one operational cycle of the SAl system and analysing the measured pump current to identify failures in the SAl system.

Preferably the SAl system operation cycle is divided into three phase, a first phase with the pump on and control valve open, a second phase with the pump on and control valve closed, and a third phase with the pump off and control valve closed and the current draw is measured in each phase. In each phase, the pump current draw may be sampled, and preferably averaged, after a delay period at the start of each respective phase and preferably the pump current draw is corrected for atmospheric pressure and power supply voltage to provide a modified pump current draw.

For phase one, the modified pump current draw is compared with predetermined maximum and minimum values to determine some failure modes, and the modified phase one current draw is subtracted from the modified phase two current draw to provide a resultant which is compared with high and low threshold values to determine further failure modes. For phase 3 of the cycle the pump current draw may be compared to a predetermined maximum limit to determine a particular failure mode.

Preferably the internal combustion engine is also provided with a air /f uel ratio sensor, sometimes called an oxygen sensor or lambda sensor, located in the exhaust manifold to measure the concentration of Oxygen (02) in the exhaust gases in the manifold, said sensor being connected to the control unit, and the oxygen concentration monitored through the SAl system cycle, the air/fuel ratio being evaluated along with current draw values in phase 1 and phase 2 to further differentiate particular failure modes Preferably the SAl system is further provided with a pump voltage sensor and a control valve voltage sensor which are connected to the control unit for monitoring the electrical continuity through the pump and valve respectively In said method, the modified phase one current draw may be subtracted from the modified phase two current draw and the resultant utilised with the phase 3 diagnostic steps to determine pump tolerance.

The pump current draw may also be utilised by the control unit for calculating the mass air flow of the pump.

According to a further aspect of the present invention, there is provided a secondary air injection (SAl) system having an electrically powered air pump for supplying air to an exhaust manifold of an internal combustion engine, a control valve controlling the air supply into the manifold, and a programmable control unit for controlling operation of the pump and valve, and a sensor for sensing electrical current drawn by the air pump, the control unit being programmed to sample the pump drawn current at predetermined time periods spaced over at least one operational cycle of the system and analysing the sampled values of pump drawn current to identify failures in the SAl system.

The control unit is programmed to operate the SAl system cycle in three distinct phases, a first phase with the pump on and control valve open, a second phase with the pump on and control valve closed, and a third phase with the pump off and control valve closed and the current draw is measured in each phase, and the control unit may be programmed to sample the pump current draw in each phase after a delay period at the start of each respective phase.

The pump current draw sensor may comprise a shunt resistor located in the electrical feed to the pump, the current being derived by measuring voltage across the resistor.

The system may further comprise an atmospheric pressure sensor and a voltage sensor for measuring the voltage of electrical power supplied to the pump, the control unit being programmed to modify the pump current draw for atmospheric pressure and power supply voltage to provide a modified pump current draw. The control unit may be programmed to compare the modified phase one pump current draw with predetermined maximum and minimum values to determine some failure modes, and further programmed to subtract the modified phase one current draw from the modified phase two current draw and compare the resultant with predetermined high and low threshold values to determine other failure modes. The control unit may be programmed to compare the pump current draw in phase 3 with a predetermined maximum limit to determine a particular failure mode.

Preferably the control unit is connected to a air/fuel ratio sensor provided on the internal combustion engine in the exhaust manifold to measure the concentration of 02 in the exhaust gases in the manifold, and the control unit is programmed to monitor the 02 through the SAl system cycle, and to utilise the air/fuel ratio in combination with current draw values in phase 1 and phase 2 to further differentiate particular failure modes The SAl system may be further provided with at least one of a pump voltage sensor and a control valve voltage sensor which are connected to the control unit for monitoring the electrical continuity through a respective one of the pump and valve.

The control unit may be further programmed to subtract the modified phase one current draw is from the modified phase two current draw and utilise the resultant with the phase 3 diagnostic steps to determine pump tolerances.

Furthermore, the control unit is programmed to utilise the pump current draw values for calculating the mass air flow of the pump.

The invention will now be described by way of example with reference to the accompanying drawings of which: -Fig 1 is a schematic drawing of an engine and diagnostic system according to the present invention, Fig 2 is a schematic drawing of the diagnostic system of Fig.1 showing the secondary air injection system and its control in greater detail, Fig. 3 is a graph of pump electrical current draw versus time for different phases of operation of the SAl system Fig 4 is the graph of Fig. 3 showing the current draw sampling strategy, Fig.5 is a logic flow chart tor the fault diagnosis relating to phase 1 of the current draw sampling strategy, Fig. 5a shows a modification of the diagnosis in phase 1, Fig 5b shows a second modification to the diagnosis in phase 1, Fig 6 is a logic flow chart for the fault diagnosis relating to phase 2 of the current draw sampling strategy, Fig 6ashows a modification of the diagnosis in phase 2, Fig 6bshows a second modification to the diagnosis in phase 2, Fig 7 is a logic flow chart for the fault diagnosis relating to phase 3 of the current draw sampling strategy, Fig 7a shows modifications to the diagnosis relating to phase 3, Fig 8 is a logic flow chart showing the steps for checking electrical continuity for the secondary air pump, Fig 9 is a logic flow chart showing the steps for checking electrical continuity for the control valve, Fig 10 is a logic flow chart showing steps for the separation of faults in phase 2, Fig 11 is a logic flow chart showing steps for the separation of further faults in Phase 2,and Fig 12 is a logic chart for the calculation of the pump mass air flow into the exhaust manifold.

With reference to Figure 1, there is shown schematically a motor vehicle internal combustion engine 11 connected to an exhaust manifold 13 which in turn is connected to a catalytic converter 14 for the treatment of exhaust gases G. Fuel is fed to the engine via fuel injectors 20 the operation of which is controlled by an engine control unit 21. The engine speed is controlled by a throttle 12 which is connected to an air intake 15.

A secondary air injection system (SAl) 10 comprises an air pump 16 having an air inlet 17 and an outlet conduit 18 which is connected to the exhaust manifold 13 via a control valve 19.

The operation of the engine 11 is controlled by the ECU 21 which contains engine maps and is programmable to control the pump 16, the valve 19 and many other engine components and operations as is well known. The ECU 21 is connected to a plurality of engine condition sensors which may include an engine speed sensor 22, an atmospheric pressure sensor 23, a throttle position sensor 24, and an air mass flow sensor 25 for sensing the air flow through the throttle 1 2 Other sensors for other vehicle parameters may also be connected to the ECU 21, for example a oxygen sensor ( sometimes called a lambda sensor) 28 may be provided on the catalytic converter 14, and a battery output voltage sensor (see Fig 2) may be provided.

The operation of the air pump 16 and the SAl control valve 19 are controlled by the ECU 21 and the operating electrical current Ip drawn by the pump 16 is measured by a current sensor 31 which is also connected to the ECU 21.

Now with reference to Fig.2, there is shown schematically the SAl system and its controls. The same reference numbers will be used for the same components previously described with reference to Fig. 1.The vehicle power supply, shown as battery 33, is connected in parallel to the secondary air pump 16 and the control valve 19 via switches 34 and 35 respectively. The switches 34, 35 may be solenoid operated switches which are controlled via the programmed ECU 21.

Electrical current Ip drawn by the air pump 16 is measured using the sensor 31, preferably deriving the current Ip by measuring the voltage across a shunt resistor, in the electrical feed line 36 to the air pump 16. The voltage Vp at the pump 16 and the voltage Vv at the control valve 19 are measured through voltage sensors 37 and 38 respectively and the battery output voltage Vs is measured via voltage sensor 29. The pump sensors 31 & 37, and the other voltage sensors 29 & 38 are all connected to the ECU 21.

With reference now to Fig. 3, there is shown an example of pump current draw during SAl operation. In order to derive a diagnostics strategy the pump current draw Ip is measured during three phases of operation of the SAl system as follows: -Phase 1 with the pump 16 in operation and the control valve 19 open.

-Phase 2 with the pump in operation and the valve 19 closed, and -Phase 3 with pump off and the valve 19 closed The current draw is sampled during the Phases 1 to 3 according to a diagnostic strategy as is shown in Fig. 4. The ECU 21 is pre-programmed with an Exhaust Back Pressure (EBP) model 41 and a diagnostic strategy. The diagnostic strategy will also be explained with reference to the logic flow chart shown in Figures 5-1 1.

With reference to Figure 5, and relating to Phase 1, the strategy begins at step 100 and at the beginning of phase 1, in step 101 the pump 16 is switched on and the control valve 19 is opened simultaneously. Following a delay in step 102 to allow the system to stabilise, and over a calibratable time period determined in step 103, the pump current Ip is sampled and averaged in step 104. The averaged value is then corrected against battery voltage Vs and atmospheric pressure in step 105.

Over the same time period, the output from the Exhaust Back Pressure (EBP) model 41 is recorded and averaged in step 106 (with the presumption that the model is accurate and has already been corrected for atmospheric pressure and other noise factors) and corrected pump current value from step 105 is modified with respect to the averaged EBP value to give a final phase 1' current value in step 106.

Step 107 determines if the phase 1 pump current 106 is below a pre-set maximum value, if the phase 1 pump current 106 is above the maximum value the pump may be jammed and this is flagged at step 108. If the phase 1 pump current 106 is below the maximum value it is passed to step 109 to determine if the phase 1 current value 106 is above a pre-set minimum value, If the current value 106 is below the minimum value, then the pump 16 may be inoperable and this fault is flagged at step 110. If the reply is yes to step 107 and step 109 then the strategy goes to phase 2 at step 111.

Referring now to Fig. 6 relating to Phase 2 which commences at step 200 after phase 1 is completed and the SAl system is no longer required to provide air into the exhaust gas stream, at step 201. At the beginning of phase 2, the ECU 21 in step 202 leaves the pump 16 running and closes the control valve 19. Following a delay determined in step 203 to allow the system to stabilise, the pump current Ip is sampled and averaged in step 205 over a time determined in step 204. The averaged value in step 205 is then corrected against battery voltage Vs and atmospheric pressure to give a final phase 2' current value in step 206 which can be used for fault diagnosis.

Fault analysis is initiated at step 208, in which the final phase 1' current value 106 is subtracted from the final phase 2' value 206. If the result is above a high threshold" at determined in step 209 then defined faults are flagged, for example pipe disconnection between the control valve 19 and the exhaust manifold 13 in step 210 If the result in step 209 is below the high threshold then the value is passed to step 211 to determine if the value is above a calibrated "low threshold". If the result is below the low threshold then the result is passed to step 212 to determine if the final phase 2' result is also below a valve stuck open' calibratable threshold, if so a fault is flagged at step 214 for disconnection between air pump 16 and control valve 19, or control valve stuck open. If the phase 2 result in step 212 is above the "valve stuck open "threshold then faults are flagged in step 213 for flow deterioration, control valve 19 stuck closed, and blockages in pipe 18.

If the value in step 211 is above the "low threshold "then in step 215 all conditions not in failure mode are passed on to phase 3 at step 300. -10-

With reference now to Fig 7, one it is determined in step 301 that phase 2 has been ended, then the pump is turned off by the ECU 21 in step 302 and phase 3' pump current Ip should be at, or very close to zero since pump 16 should no longer be operating and the control valve 19 remains closed. After a timed delay at step 303, the current is sampled at step 305 for a time period calibrated by step 304 If it is determined at step 306 that current is above a predetermined maximum value (max2 limit), I e. that there is still a current being drawn then a fault, for example pump stuck on" is flagged in step 307. If the phase 3 current is less than the predetermined maximum value then step 308 indicates that phase 3 is operating correctly Throughout the SAl operation the engine control unit 21 may additionally check for electrical continuity of the pump 16 through a series of diagnostic step (400 -410) and of the control valve 19 through a series of diagnostic steps (500 -510), by monitoring voltage flow through each of them (Vp and Vv).

Referring to Fig. 8 the ECU 21 initiates the continuity check for the pump 16 at step 400 and the "pump on" command is affected by the ECU 21 at step 401 If the pump 16 is not turned on, the pump voltage Vp of the pump 16 is checked via voltage sensor 37 at step 406. If there is no voltage, then step 409 checks that the measurement was confirmed throughout a specified time period, if yes then step 410 confirms pump continuity. If the "No voltage" response at step 406 was made and the time duration determined in step 409 has not been exceeded, the test steps loop back to step 401. If a voltage Vp is detected at step 406 and the time duration determined in step 407 has not been exceeded, the diagnostic step loop back to step 401. If a voltage Vp is detected at step 406 after the expiry of the time duration determined in step 407 then a "Short circuit "to ground is flagged up at step 408.

If the "pump on" command is effected at step 401, then if a voltage Vp is detected at step 402 and confirmed throughout the time duration determined in step 405 the pump 16 -11 -continuity is confirmed at step 410. If no voltage is detected at step 402 for the time period less than the duration determined at step 403 then the diagnostics steps are looped back to step 401. If no voltage is detected and the time duration in step 403 has been exceeded then a short circuit to ground is flagged at step 404 Referring to Fig 9, the ECU 21 initiates the continuity check for the control valve 19 at step 500 and the "valve open" command is effected by the ECU 21 at step 501. If the Valve open" command is NOT effected the valve 19 is checked for voltage Vv via voltage sensor 38 at step 506. If there is no voltage at step 506, then step 509 checks if the measurement was confirmed throughout a specified time period, if yes then step 510 confirms electrical continuity It the" No voltage" response at step 506 was made and the time duration determined in step 509 has not been exceeded, the test steps loop back to step 501. If a voltage Vv is detected at step 506 and the time duration determined in step 507 has not been exceeded, the diagnostic steps loop back to step 501. If a voltage Vv is detected at step 506 after the expiry of the time duration determined in step 507 then a "Short circuit" to battery is flagged up at step 508.

If the "valve open" command is effected at step 501, then if a voltage Vv is detected at step 502 and confirmed throughout the time period determined in step 505 the electrical continuity through valve 19 is confirmed at step 510. If "No voltage" is detected at step 502 for less than the time duration determined at step 503 then the diagnostics steps are looped back to step 501 If "No voltage " is detected and the time duration determined in step 503 has been exceeded, then a short circuit to ground is flagged at step 504.

The diagnostic system programmed into the ECU 21 can detect the failure modes listed below. The diagnostic step that will detect each failure mode is listed below: a) Air pump: * stuckoff-stepilO. -12-

* stuck on -step 307.

* pump jammed -step 108.

* flow deterioration (less air pumped) -step 213, * pump relay short to battery/ground -electrical continuity check steps 404 /408 b) check valve flow: * stuck closed -step 213.

* stuckoperi-step 214 * flow deterioration (restriction in valve) -step 213.

* pump relay short to battery/ground -electrical continuity -steps 504/508.

C) Pipe between pump and valve: * disconnection -step 214.

* blockage -step 213.

d) Pipes from valve to manifold(s): * disconnection -step 210 * blockage -step 213.

If desired, the individual faults may be separated out from each other by the introduction of additional diagnostic steps. Whilst the SAl is operating, the closed loop -13-fuelling can be monitored in parallel using the oxygen sensors 28 to determine the exhaust air/fuel ratio for each exhaust manifold. The degree of separation that is possible depends of whether the engine has a single exhaust manifold or a multi manifold layout The diagnostic steps that improve the separation of each failure mode are described below: In order to improve fault judgement separation in particular for a single exhaust manifold layout, modifications must be made to the basic flow charts shown in Figs 5 & 6 Referring to steps 104 and 205 in Figs 5 & 6 and figs 5a & 6a, the exhaust air/fuel ratio (Oxygen concentration) measured by sensor 28 must be sampled by the ECU 21 and averaged along with the other signals, forming new steps 104a and 205a shown in Figs 5a & 6a This provides for the separation of the various faults within step 213 and step 214 of Fig. 6.

The basic diagnosis cannot functionality distinguish between the following faults within step 213: -Flow deterioration (reduced pump air flow) -Flow deterioration (restriction in control valve) -Control valve stuck closed -Blockage in pipe between pumps & control valve -Blockage in pipe between control valve & manifold Referring now to Figs 6 and 10, if step 211 produces a "No" response and if the final phase 2 value deduced in step 206 is greater than the pre-set valve open threshold in step 212, then the new step 217 will determine if the phase 1 air/fuel ratio measured at 104a is greater than a threshold value If the ratio value is less than the threshold value, -14-then the "control valve stuck closed "failure mode is flagged at step 213a. If the ratio value is greater than the threshold value then the remaining flow deterioration and blocked pipe faults are detected at step 21 3b The diagnostics of the basis flow chart in Fig. 6 cannot distinguish between the following faults within step 214 of the basic flow diagram: -Control valve stuck open -Disconnection in pipe between pump & control valve If the final phase 2 value deduced in step 206 is less than the pre-set valve open threshold in step 212, then the new step 218 will determine if the phase 2 air/fuel ratio measured at step 205a is greater than a threshold value If the ratio value is greater than a second threshold value then the "control valve stuck open" failure mode is flagged at step 214a. If the phase 2 ratio value in step 218 is less than the threshold value then the failure mode for a disconnected pipe 18 between air pump 16 and control valve 19 is flagged up at step 214b In order to improve fault judgement separation in particular for a multi exhaust manifold layouts, modifications must be made to the basic flow charts shown in Figs 5 and 6. Referring to steps 104 and 205 Figs 5 and Fig 6, and Figs. 5b & 6b, the exhaust air/fuel ratio measured by the respective sensors 28 must be sampled for each manifold together with the other signals, forming new steps 104b and 205b. This provides for additional separation of the various faults within step 213 and step 214 of Fig. 6 For simplicity, a dual manifold layout is described below but this functionality can be scaled for any number of manifolds The averaged exhaust air/fuel ratio for manifold 2 (1 04b) is subtracted from the averaged exhaust air/fuel ratio for manifold 1 (104b) in step 113 to give the difference (or offset) in the air/fuel ratios from one manifold to the other. -15-

The basic diagnosis flow chart is shown in Fig 11 which is substantially similar to Fig such that only the differences will be described. Referring now to Figs 6 and 11, if step 211 produces a "No" response and if the final phase 2 value deduced in step 206 is greater than the pre-set valve open threshold in step 212, then the new step 217 will determine if the phase 1 air/fuel ratio measured at 104b is greater than a threshold value for either manifold If the ratio values are less than the threshold values, then the "control valve stuck closed "failure mode is flagged at step 213c. If the ratio value is greater than the threshold values, then the off-set value determined in step 113 is compared to threshold values for each manifold and determined in steps 219 and 220 If the offset value is less than the respective threshold values then a blocked pipe is flagged at steps 213d and 21 3e. The remaining flow deterioration and blocked pipe faults are detected at step 21 3f.

The difference between the phase 1' result in step 106 and the phase 2' result in step 206 may also be used to determine the tolerance of the pump 16 fitted to the vehicle.

This tolerance value could be used as a compensation factor for the failure thresholds This would increase the accuracy of the diagnostic judgements as, without knowing the pump tolerance, the failure thresholds would have to be calibrated based on the results of the "worst case" pump tolerance. Therefore, the thresholds would not be optimised for the full range of pump tolerances that could be fitted to the vehicle. This modification could also provide a compensation factor for the SAl pump mass air flow calculations which interact with the base engine air flow and fuelling control strategies. By adding this compensation factor the accuracy of the SAP mass air flow calculation would be improved as the current strategies have no reference to the tolerance (air flow characteristics) of the pump.

One method of calculating the pump tolerance and applying it to the basic flow diagram is shown below. Note that this can be also be integrated into the closed loop fuelling' modification as described previously -16-With reference now to Fig. 7a, the pump tolerance can be calculated from the diagnostic steps used in relation to faze 3. A pump tolerance value' is calculated after the step 308 and before the end of the proposed diagnostic flow in step 309, since the results from all the faults judgements are required before the pump tolerance value' can be calculated. The pump tolerance value' is only calculated if step 310 determines if no faults have been detected during the diagnostic test. If any faults are detected then the pump tolerance value' is not updated and the value that was calculated on the last passed' diagnostic test is stored in step 311 for the next time the test is run. If NO faults are determined at step 310 then in step 312 the phase 1' result (106) is subtracted from the phase 2' result (206) to give a phase offset' result. The phase offset' result at step 312 can be filtered and applied to previous pump tolerance' results in step 313.

The "calculated pump tolerances" experienced over a number ofdiagnostic tests can be used as an indication of the pump tolerance. A suitable diagnostic method provides that the phase offset' value of step 312 is subtracted from the pump tolerance' value calculated from the previous diagnostic test multiplied by a calibratable coefficient. This result is then added to the phase offset' value (312) to give a new pump tolerance' value (313) that would be used in the next diagnostic test.

The pump tolerance' value (313) may be used in the calculation of all of the failure thresholds described earlier. For each calibratable failure threshold in steps 107, 109, 209, 211. 212 and 306, the threshold value described in the basic proposed method could be replaced using a 2-Dimensional table whose x-axis is pump tolerance' ranged between the minimum and maximum pump tolerance specifications and whose y-axis would be the range of failure threshold values set for each of the various tolerances of pump. The failure threshold value used for the diagnostic test would be interpolated from the table based on the pump tolerance' value at that time -17-It may be beneficial to use low pass filters during the sampling and averaging of the signals in phases 1 and 2 (steps 104 and 205). The low pass filters would reduce the influence of any transient effects from the engine and secondary air system and electrical noise (interference) from the engine control system signals.

The measurement of the pump operating current Ip during "phase 1' may also be used to calculate the SAl pump mass air flow (MAF) which can be utilised by the engine air flow and fuelling control. With reference now to Fig. 12, there is shown a diagnostic flow chart for the calculation Qf the MAF of pump 16 into the exhaust manifold 13 The ECU requests at step 600 that the SAl system comes into operation for phase 1 at step 601 and the pump current (Ip) is measured and converted to a volumetric air flow in step 602 using pre-determined pump characteristics. The pump volumetric flow is corrected against battery voltage (Vs) in step 603. The result from step 603 is then modified in step 604 to take into account the thermodynamic effects of filling the pipes 18 that connect the pump 16 to the manifold with a pressurised gas. The result from step 604 is then corrected in step 605 for any flow restrictions (or changes in volumetric efficiency) though the control valve This will give the raw volumetric pump flow at the junction with the exhaust manifold 13 (with no exhaust back pressure present.) The volumetric flow from step 605 is then corrected for exhaust back pressure in step 606 in order to provide a total volumetric air flow into the exhaust manifold 13. The total volumetric air flow into the exhaust manifold 13 is then converted to total mass air flow in steps 607 -609 by using the characteristic gas law: m = PV/RT Where m = mass air flow, P = total air pressure (i.e atmospheric pressure + exhaust back pressure), V = volumetric air flow, R = characteristic gas constant for air, T = air temperature (Kelvin).

The result given in step 610 is the total SAl mass air flow into the exhaust manifold.

This can then be applied to base engine air flow and fuelling control strategies.

This proposed SAl system diagnostic proposal allows the pump current to be used to determine the operating condition of the system and has the advantage of removing the requirement for a pressure sensor and also improves the separation of the various failure modes from one another as compared with present commercial systems. With the additional proposed modifications further improvements in fault separation and judgement are achieved. -19-

Claims

CLAIMS

1. A method of detecting faults in a secondary air injection (SAl) system having an air pump for supplying air to an exhaust manifold of an internal combustion engine and having a control valve controlling the air supply into the manifold and a programmable control unit for controlling operation of the pump and valve, wherein the method comprises the steps of: sampling electrical current draw by the air pump at predetermined time periods spaced over at least one operational cycle of the SAl system and analysing the measured pump current to identify failures in the SAl system.

2. A method as claimed in Claim 1 wherein the SAl system operation cycle is divided into three phase, a first phase with the pump on and control valve open, a second phase with the pump on and control valve closed, and a third phase with the pump off and control valve closed and the current draw is measured in each phase.

3. A method as claimed in Claim 2, wherein in each phase the pump current draw is sampled after a delay period at the start of each respective phase.

4. A method as claimed in Claim 3, wherein the value for pump current draw is averaged over predetermined time duration.

5. A method as claimed in any one of claims 2 to 4 wherein the pump current draw is corrected for atmospheric pressure and power supply voltage to provide a modified pump current draw.

6. A method as claimed in Claim 5, when applied to phase one, wherein the modified pump current draw is compared with predetermined maximum and minimum values to determine failure modes -20 - 7. A method as claimed in Claim 6, wherein the modified phase one current draw is subtracted from the modified phase two current draw and the resultant compared with high and low threshold values to determine failure modes.

8. A method as claimed in Claim 6 or Claim 7, wherein in the pump current draw in phase 3 is compared to a predetermined maximum limit to determine a particular failure mode.

9. A method as claimed in any one of Claims 2 to 8, wherein the internal combustion engine is also provided with a air /fuel ratio sensor located in the exhaust manifold to measure the oxygen (02) concentration in the exhaust gases in the manifold, said sensor being connected to the control unit, and the 02 concentration is monitored through the SAl system cycle, and the air/fuel ratio may be applied to current draw values in phase 1 and phase 2 to further differentiate particular failure modes.

10. A method as claimed in any one of Claims 1 to 9 wherein the SAl system is further provided with a pump voltage sensor and a control valve voltage sensor which are connected to the control unit for monitoring the electrical continuity through the pump and valve respectively.

11. A method as claimed in Claim 5 wherein the modified phase one current draw is subtracted from the modified phase two current draw and the resultant is utilised with the phase 3 diagnostic steps to determine pump tolerance.

12. A method as claimed in any one of Claims 1 to 5, wherein the pump current draw is utilised by the control unit for calculating the mass air flow of the pump.

13. A secondary air injection (SAl) system having an electrically powered air pump for supplying air to an exhaust manifold of an internal combustion engine, a control valve controlling the air supply into the manifold, and a programmable control unit for -21 -controlling operation of the pump and valve, and a sensor for sensing electrical current drawn by the air pump, the control unit being programmed to sample the pump drawn current at predetermined time periods spaced over at least one operational cycle of the system and analysing the sampled values of pump drawn current to identify failures in the SAl system.

14. A system as claimed in Claim 13, the control unit is programmed to operate the SAl system cycle three distinct phases, a first phase with the pump on and control valve open, a second phase with the pump on and control valve closed, and a third phase with the pump off and control valve closed and the current draw is measured in each phase.

15. A system as claimed in Claim 14, wherein the control unit is programmed to sample the pump current draw in each phase after a delay period at the start of each respective phase 16. A system as claimed in any one of claims 13 to 15, wherein and further comprising an atmospheric pressure sensor and a voltage sensor for measuring the voltage of electrical power supplied to the pump, and the control unit is programmed to modify the pump current draw for atmospheric pressure and power supply voltage to provide a modified pump current draw.

17. A system as claimed in Claim 16, wherein the control unit is programmed to compare the modified phase one pump current draw with predetermined maximum and minimum values to determine failure modes 18. A system as claimed in Claim 17 wherein the control unit is programmed to subtract the modified phase one current draw from the modified phase two current draw and compare the resultant with predetermined high and low threshold values to determine failure modes.

-22 - 19 A system as claimed in any one of Claims 14 to 18, wherein the control unit in programmed to compare the pump current draw in phase 3 with a predetermined maximum limit to determine a particular failure mode.

20. A system as claimed in any one of Claims 14 to 19, the control unit is connected to a air/fuel ratio sensor provided on the internal combustion engine in the exhaust manifold to measure the 02 concentration in the exhaust gases in the manifold, and the control unit is programmed to monitor the air/fuel ratio through the SAl system cycle, and to utilise the air/fuel ratio in combination with current draw values in phase 1 and phase 2 to further differentiate particular failure modes.

21. A system as claimed in any one of Claims 14 to 20, wherein the SAl system is further provided with at least one of a pump voltage sensor and a control valve voltage sensor which are connected to the control unit for monitoring the electrical continuity through a respective one of the pump and valve.

22. A system as claimed in Claim 16, wherein the control unit is programmed to subtract the modified phase one current draw from the modified phase two current draw and utilise the resultant with the phase 3 diagnostic steps to determine pump tolerance.

23. A system as claimed in any one of Claims 14 to 22 wherein the control unit is programmed to utilise the pump current draw values for calculating the mass air tiow of the pump.

24. A system as claimed in any one of Claims 14 to 23, wherein the pump current draw sensor is provided by a shunt resistor in the electrical feed to the pump, the current being derived trom the voltage measured across the resistor.