GB2583336A - Method of determining a fault in an engine with EGR - Google Patents

Abstract

Disclosed is a method of diagnosing a fault in an internal combustion engine of a vehicle, the engine including at least one Exhaust Gas Recirculation (EGR) feedback line including an EGR control valve located in the line. The method comprises the steps of determining the actual pressure differential across the valve, determining the exhaust gas flow exiting the exhaust, determining a value of a computed reference pressure differential (EGRH dP*) based on the value of intake manifold air pressure, comparing the determined actual pressure across the valve with the reference pressure differential across the valve to determine a pressure differential error and diagnosing a fault based on the determined pressure differential error. The pressure differences and exhaust mass flow may be measured, estimated or modelled. The EGR may use a high-pressure line and/or a low-pressure line.

Description

METHOD OF DETERMINING A FAULT IN AN ENGINE WITH EGR

TECHNICAL FIELD

This application relates to EGR (Exhaust Gas Recirculation) systems in internal combustion engines and has application to systems with high pressure EGR feedback flow lines or low pressure EGR flow feedback lines as well as hybrid 10 EGR systems with both high and low pressure EGR.

BACKGROUND OF THE INVENTION

In internal combustion engines EGR (Exhaust Gas Recirculation) is mainly used to reduce the rate of formation of oxides of nitrogen thanks to effects like lower temperatures and lower oxygen concentration during combustion process. From the engine control prospective, an EGR (Exhaust Gas Recirculation) system is a closed loop control system that controls a percentage of EGR from the exhaust back to the inlet manifold using the EGR valve and throttle. The percentage of EGR is usually calculated indirectly considering a speed-density model, which accounts for both fresh air and EGR, and an AMF (Air Mass Flow) feedback.

In the US (United States) and Europe, OBD (On Board Diagnostic) legislation imposes OEMs (Original Equipment Manufacturer) to detect deviations of EGR 25 rates from nominal rates in terms of High, Low and Slow response EGR flow.

In case of hybrid EGR systems, correct pinpointing between high and low pressure EGR is one of many issues to face during the OBD function development, due to EGR control complexity. Accordingly high and low pressure EGR pinpointing is often bypassed, based on EGR demands or done by intrusive strategies which are not always accepted by OEMs due to their emissions impact.

In case of High Flow EGR, once the malfunction has been confirmed, a pinpointing is usually introduced to understand where the malfunction comes from.

For most of OEMs, due to control complexity, EGR malfunction pinpointing is often/; i) bypassed; ii) based on EGR demands; iii) done by intrusive strategies (e.g. EGR cut-off imposed throttle and/or intake shutter valve position, etc., which are not anymore accepted by OEMs; or iv) based on additional EGR temperature sensors (not always acceptable by OEMs) or more in general, required an additional diagnostic hardware It is an object of the invention to provide an improved method of EGR diagnostics.

SUMMARY OF THE INVENTION

In one aspect is provided a method of diagnosing a fault in a vehicle engine, said engine including at least one Exhaust Gas Recirculation (EGR) feedback line including a valve located therein, comprising: a) determining the actual pressure differential across said valve; b) determining the exhaust gas flow exiting the exhaust (EMF); c) determining a value of a computed reference pressure differential (EGRL dP*) from the value of intake manifold air pressure; d) comparing the results from a) and c) to determine a pressure differential error; e) diagnosing a fault based on the determined pressure differential error from step d).

In step c) said value may be determined from a pre-stored relationship between 25 values of EMT and values of computed reference pressure differential (EGRL dP*).

Said EGR feedback line may be a low pressure EGR feedback line.

In step a) the pressure differential may be determined from a pressure sensor across the inlet and outlet of the valve.

In step a) the pressure differential may be estimated or modelled.

In step a) the pressure differential may be estimated or modelled using the inputs of exhaust pressure and intake manifold pressure.

In step b) EIVIF may be estimated or modelled.

The method may include the further step of comparing the pressure differential error with a threshold value and where step e) comprises outputting an indication of error if the pressure differential error is higher than said threshold value.

Said threshold value is a calibrated threshold value based on operating conditions of the engine.

The method may include determining an Air Estimation Error and wherein step e) comprises diagnosing a fault based on the Air Estimation Error as well as the determined pressure differential error from d).

Said Air Estimation Error may be compared with an AEE calibrated threshold here step e) comprises flagging a fault if the both AEE is more than said AEE calibrated threshold and said pressure differential error is higher than said threshold value.

Said AEE may be determined from comparing a value of estimated manifold intake flow and a flow estimate provided from a Speed Density model.

The speed density model may provided a flow estimate based on the Manifold Air Pressure and the Manifold Air Temperate and volumetric efficiency of the engine.

Said engine system includes a further EGR feedback line and said method include determining a further pressure differential error therefor and wherein in step e) comprises diagnosing a fault based on the determined pressure differential error form d and said further pressure differential error.

Said further EGR feedback line is a high pressure feedback line Step e) may comprise diagnosing a fault based on the determined pressure differential error from d) , said further pressure differential error. and said air estimation error.

BRIEF DESCRIPTION OF THE DRAWINGS

The present invention is now described by way of example with reference to the accompanying drawings in which: Figure 1 shows a chart of desired EGR flow 100 over time, and reference numeral 101 indicates the actual EGR flow when there is a problem such as a stuck open valve Figure 2 shows a flow circuits of a EGR hybrid system in a Diesel Engine with both high and low pressure EGR feedback loops (lines) Figure 3 depicts the AEE calculation; Figure 4 shows an example of the methodology according to one example; Figure 5 shows the high pressure EGR dP model is based on MAP sensor and shows a the calibration curve (map/chart); Figure 6 shows a refined example of methodology; Figure 7 illustrates another example of the invention; Figure 8 shows the low pressure EGR dP model is based on EMT input and shows a calibration curve (map/chart); Figure 9 shows a refined example of methodology; Figure 10 shows a yet further refined example of methodology; Figure 11 shows statistical results.

DESCRIPTION OF THE PREFERRED EMBODIMENTS

This invention provides an algorithm to detect a fault in an EGR system in terms of e.g. EGR high flow and an innovative method to correct pinpoint the fault in e.g. hybrid EGR systems.

Many possible EGR system issues cause a positive deviation from nominal EGR rates. These include a defective EGR valve which can be stuck open, opening too far or not properly closing. Also excessive vacuum signal or electrical signal to the EGR valve may cause deviation as well as malfunction of the EGR vacuum solenoid. An air leak in the pipe downstream from the compressor or Intake/exhaust manifold would cause similar problems as well as a defective throttle and/or VGT vanes.

Figure 1 shows a chart of desired EGR flow 100 over time, and reference numeral 101 indicates the actual EGR flow when there is a problem such as a stuck open valve. In steady state the actual EGR flow should be within upper and lower limits shown by the dotted lines.

The effect of the above failures can be critical for vehicle drivability, emissions and DPF (Diesel Particulate Filter) soot accumulation.

Figure 2 shows a flow circuits of a EGR hybrid system in a Diesel Engine with both high and low pressure EGR feedback loops (lines). In the example Air Intake is to an air filter 1 which then passes through an air meter 2 to the intake throttle. 3 This cold air intake is then compressed by a compressor 4 of a turbo system before entering the intake manifold 5 and hence cylinder of the engine for combustion with injected fuel. Between the compressor may be an intercooler 6 and an intake shutter valve 7. After combustion, exhaust air exits via an exhaust manifold 9 to drive a turbo 8 before entering the exhaust system 10 which comprises components as known in the art which may be a LNT (lean Nitrogen Trap) 11, a Diesel particulate filter (DPF) 12. (The DPF may alternatively be of the catalyzed-type, referred to as CDPF)" and an SCR system which includes the SCR itself 13 and urea injector 14 upstream thereof The exhaust gases then exit via a muffler 15.

As can be seen there is a high pressure EGR line 16 where exhaust gases directly from the exhaust manifold (i.e. not via any components such as the LNT DPF) is fed back into the intake manifold under the control of EGRH (EGR High) valve 17.

Between the DPF and the SCR system (including urea injector) is also provided a low pressure EGR return line 18 where exhaust gases form the DPF is via a Low Pressure (LP) EGR cooler 18 back to the air intake portion upstream of the compressor under the control of a EGRL (EGR Low) valve 19 in the low pressure EGR line.

The circuit includes various sensors where parameters in the system can be measured and one or more of which are used in examples of EGR control and diagnostic, in both prior art and examples of the invention, as will be explained hereinafter. These include: a) Atmospheric pressure sensor 20. This may be optional and the atmospheric pressure may be assumed or derived from e.g. weather data in conjunction with a GPS system b) Air intake temperature sensor 21. T1 Again this may be optional and the atmospheric pressure may be assumed r derived from e.g. weather data in conjunction with a GPS system from a GPS c) Intercooler out temperature sensor/ intake temperature T22 = temperature of intake air entering cylinders /intake manifold d) Manifold (Intake) air pressure sensor 23 e) EGRH valve position (feedback) 24 for the high pressure EGR lone I) EGRL cooler out temperature 25= temp of gases entering EGRL valve g) ERGL valve position feedback 26 h) EGRL delta pressure 27 (pressure differential up and downstream of EGRL valve.

i) Air meter flow (28) Sensors used in aspects of the invention may include one or more of the following: i) Mass (Air) Flow Sensor (28); ii) Intake Manifold Pressure sensors (MAP) 23; iii) Intercooler out temperature sensor T22 or Intake manifold temperature sensor (not visible in figure 2).

A known system for analyzing EGR deviations is an EGR High Flow monitor is based on the intake port flow error, called Air Estimation Error (AEE), between a speed-density model and AMP (Air Mass Flow) feedback, during low EGR position demands and MAP (Manifold Absolute Pressure) changes.

Figure 3 depicts the AEE calculation: The Mass Airflow sensor 21 is used to determine the intake flow 28 (in the intake manifold) in block 30 to provide the (MAP based) intake flow at block 31. Because the MAF sensor (28) is measuring the flow where the air filter (1) is, while (MAF based) intake flow is a modeled flow after the intake throttle (close to intake manifold). Thus the value of intake flow can be determined based on MAF sensor signal as well as the flow coming from the low pressure EGR circuit. So 30 is the MAF = Air Meter Flow = 30, MAF-based Manifold (Intake Port) Flow = 31) The manifold pressure (MAP) i.e. he intake manifold pressure sensor or MAP from sensor 23, and manifold temperature (MAT which is the intake temperature sensor or intercooler out temp sensor (T22)) is input block 32 along with volumetric efficiency as an input; block 32 provides a SPD (speed density) based intake flow estimate. From these parameters. The flow estimate from 32 from the SPD 32 is then compared to the MAF intake flow results 31 at 33 to give an air estimation error 34, which is the degree to which the actual (high) EGR flow rate is higher or lower than the demand or expected value i.e. to determine whether there is too high or low EGRH flow.. This is then optionally compared with a calibrated threshold 35 at 36 to give an indication 37 of error, i.e. the indication 37 may be a binary output confirming error or not (i.e. flagging error) if the air estimation error is above the calibrated threshold.

So to recap the algorithm monitors the Intake Port Flow Error between the speed density model and air mass (MAF) flow feedback, during low EGR positions demands and NIAP changes. The air error is then compared with a negative calibrated threshold. Therefore when the air estimation error is negative the actual amount of exhaust gases, flowing back into the engine intake, is higher than the requested amount, leading to critical drivability conditions as long as fast soot accumulation.

The AEE monitor is not sufficient to pinpoint the EGR fault in case of e.g. hybrid EGR systems, therefore further monitoring methods have been developed in examples of the invention; which involve messuring or determmining further parameters.

Invention According to examples of the invention, diagnostic methods are used based on the high and/or low pressure EGR delta pressures errors, derived from the measured and modeled delta pressure on the EGR valves.; as will be explained hereinafter.

Thse further diagnostic methods can be used on their own to give an indication of EGR errors, or together, and each one can be used in conjunction with the AEE method described above to provide enhanced error estimation and the refinement of the AEE. In one example both the further diagnostic ethods as well as the AEE are used to give indications of the deviation (error) between actual EGR rates and the expected or demad rates.

Invention 1 High Pressure EGR diagnostic.

Example 1

In the following example it should be noted that the methodology can be used in 30 EGR systems whch do not include a low pressure EGR feedback as well as in hybrid systems with both low and high pressure EGR.

Figure 4 shows an example of the methodology according to one example. EGRH delta pressure (dP), that is the dP across the HP vlav ein the HP EGR line is determined, This can be determined in various ways such as from a pressure sensor between the inlet and outlet of the valve or measured//modelled estimated (e.g. using exhaust pressure sensor and intake manifold pressure sensor measuresas inputs). Alternatively it may be estimated or modelled (e.g. using intake manifold pressure measure sensor and estimation of exhaust pressure sensor).

The value of MAP at 40 (MAP is the intake manifold pressure 23) is determined (e.g. measured by sensor or estimated/modelled by appropriate inputs from senosrs) and input into a look up table or map 41 which gives a "computed reference" value 42 EGRH dP* which is a computed pressure differential, as will be explained in more detail herenafter. This value can be considered the ideal or what the value of dP should be for that value of MAP. This is compared with the EGRH dP pressure 43 which is the pressure difference measured/estimated across the EGR HP valve as described above, to give an EGRH dP error 44 which is the degree to which the actual EGRH dP in the HP deviates from demand/expected.

This is optionally compared with a calibrated threshold 45 at 46 to give an indication 47 of error i.e... The indication 47 may be a binary output confirming error or not (i.e. flagging error if the air estimation error is above the calibrated). So in ther words concerning the high pressure EGR delta pressure model, a MAP-based model has been set With regards obtainign a value of the computed pressure differential i.e. the reference value EGRH dP* 42 for comparison, a calibration curve, map, or chart can be prepared. This is shown in more detail in figure 5. The high pressure EGR dP model is based on MAP sensor and it is evaluated considering a test with a normally functioning high pressure EGR valve.

In hybrid systems the high pressure EGR dP model is based on MAP sensor and the calibration curve (map/chart) shown in figure 5 and can be evaluated considering a test which covers various EGR engine operating points with a normally functioning high pressure EGR valve and a malfunctioning low pressure EGR valve, reproducing EGR high flow conditions i.e. the plot can in more complex examples consider a malfunctioning low pressure EGR valve and a functioning high pressure EGR valve), reproducing EGR high flow conditions. A curve can be provided taking to account all observations and used to determine EGRH dP* from MAP pressure.

To optimize model fitting, the calibration engineer can discard points that are not included in the enable conditions (points 48 in the figure).

Example 2

Figure 6 shows an enhanced example. This incorporates the method described above (the logic of figure 4 shown by the dotted box) with AEE determination methodology of figure 3. Essentially the parameter or indicaton of error 47 is fed into a fault management (logisl function block) 50 where a further input is the output 37 from an AEE system as described above; the figure shows a simplifed version of the AEE model of figure 3. Thus the output of the known AEE as well as the output 47 of the example are analusyed together to provide an enhanced diagnostics output indicative of error 50. In one aspect thee output 37 and 47 errors are compared in order to avoid false error indication; so the EGR delta pressure errors monitors works in conjucntion with AEE monitor in the Fault Manager subsystem to avoid false detections. So only if an error is flagged in both the AEE sytem at 37 and the sysem of figure 4 is an error flagged at 50. If there is no error flagged at either 37 or 47 then not error is flagged at 50.

Invention 2 Low Pressure EGR diagnostic Figure 7 illustrates an example of the invention. Here at 52 the pressure differential across the control valve in the low pressure EGR circuit EGL dP is determined e.g. from pressure sensor 27 to give a value of low pressure EGR delta pressure (differential) . Again alternatively, this can be determined in various ways such as from a pressure sensor between the inlet and outlet of the valve or measured//modelled estimated (e.g. using exhaust pressure sensor and intake pressure sensors measures as inputs). Alternatively it may be estimated or modelled (e.g. using intake manifold pressure measure sensor and estimation of exhaust pressure sensor).

The exhaust flow that exits the exhaust system e.g. exhause muffler flow EMF 53 is determined from sensors or from a model i.e. tThe exhaust muffler flow may be a modelled parameter. The inputs of this model can be intake manifold pressure (MAP), intake manifold temperature (MAT) and injected fuel demand..

This value of EMF is then used in a look up table or map 54 to provide a computed reference value 55 which is EGRL dP* which is a computed reference pressure differential, as will be explained in more detail herenafter. This can be regarded as the ideal presssure differential or that which should be across the EGR L valve is a normally operating system for that value of EMF.

This computed reference value of EGRL dP* is compared with the EGRL dP pressure 52 which is the pressure difference measured across the EGR LP vlave to give an EGRL dP error 56 which is a parameteter idicative of error or the degree to which the actual EGR flow in the LP deviates from demand/expected.

Thus 56 which is the degree to to which actual EGRL flow differs from expected demand vlaues, is optionally compared at 60 with a calibrated threshold 57 to give an indication of error 58. The indication 58 may be a binary output confirming error or not (i.e. flagging error) if the air estimation error is above the calibrated.

So to recap this example provides a low pressure EGR delta pressure model, a muffle /exhaust flow based model has been set), Figure 8 shows in more detail how the vlaue of EGRL dP* which is a computed pressure differential, is determined. The low pressure EGRL dP* model is based on Exhaust Mass Flow as an input and it is evaluated during tests which cover various EGR operating points t with a normally operating EGR LP valve e.g. reproducing EGR high flow conditions. A curve /map look up tables can be provided taking to account all observations and used to determine EGRL dP* from EMT.

So the calibration curve can be determined by running a normally functioning valve at various conditions and looking at measured/computed EN/IF and the pressure differential across the EGRL valve. In hybrid systems the calibration curve/map/mapping can be provided by running tests with a functioning EGRL valve and a malfunctioning EGRH valve. So where the system is a hybrid system the test to produce the calibration curve may be performed with a normally operating low pressure EGR valve and a malfunctioning high pressure EGR valve, reproducing EGR high flow conditions.

Figure 9 shows an enhanced example. This incorporates the AEE method described above with respect to figure 3and the logic described above with reference to figure 7 (the latter shown by the dotted box). Essentially the indicaton of error 58 is fed into a fault management (logical function block) 61_ where a further input is the output from an AEE system as described above; the figure shows a a simplifed version of the AEE model of figure 3. Thus the output of the known AEE 37 as well as the output 58 of the systtem/model of figure 7 of the example are analusyed together by 61 to provide an enhanced diagnostics output indicative of error 62. In one aspect the error indications output from 37 and 58 are compared in order to avoid false errors. So the EGR L delta pressure error system works in conjucntion with AEE monitor in the Fault Manager subsystem to avoid false detections. So only if an error is flagged in both the AEE system at 37 and the system of figure 7 at 58, is an error flagged at 62.

A further enhanced system is shown in fgure 10 which shows both methods described with reference to figure 4 and figure 7 used together with the AEE system. Here the fault manager (logic) 70 receives error diagnostics (flagging) outputs 37, 58 and 47 from the sub-sytems. These are analysed of determine an improved reliable errors dignostic 71. In one example only if an error is flagged in both the AEE sytem at 37 and one or both the outputs 47 and 48 indicate/flag an error, is an error flagged at 71.

Alternatively an error is flagged at 71 is both 47 and 58 flag an error regardless of the output at 37.

A solid and mature OBD system means that the models can distinguish between worst performing acceptable parts (WPA) and best performing unacceptable parts (BPU) and maximize their separation in order to minimize OBD system error occurrences.

EGR high flow monitor can be considered as a threshold monitor (malfunctions are detected before emissions exceed OBD emission thresholds with a safety factor) with: a) BPU given by drivability issues, fast DPF soot accumulation (generally x2 with respect to nominal DPF soot accumulation) or OBD Threshold Limits (OTLs); b) WPA given by nominal parts + limit hardware on MAP, MAF air flow meter = mass air flow) and atmospheric pressure sensor (sensor deviation are given by the OEM -the atmospheric pressure sensor is used to compute the worst part acceptable (WPA). In On board diagnostics (OBD) there is a distinction between WPA and best part unacceptable (BPU).) which have the most negative impact on AEE; c) AEE as first monitor parameter d) High and/or low pressure EGR delta pressure error as second monitor parameter Figures 11a and l lb depicts a statistical plot sof WPA (201)and BPU (200) Gaussian distributions with respect to the first monitor parameter of the air estimation error in the cases of the EGRH and EGRL respectively. It thus shows the statistical separation on EGR high flow monitor for high and low pressure EGR sub systems. In and off cycles are considered. This plots show validation of monitor robustness. On the left (200) of the plot are the histogram results (distribution) for the tests with working EGR valves and on the right, the results (distribution) 201 (the test with malfunctioning EGR valves (red distribution) and the line trace is the consequent normal distribution. If there is a large separation of the distributions, the monitor can robustly detect an EGR malfunction without false passes or false fails. This method is used in OBD to assess the monitor robustness, therefore it's not a part of the invention logic but of the invention validation.

The metric to evaluate monitoring robustness is given by DM/SS factor defined as: (Di ence of e Means / Sum of nd rd Deviations) Robust monitors have DM/SS > 3 (no overlap of distributions).

WPA/BPU statistical plots also help to set the threshold which distinguish a working system from a faulty system. In case of EGR high flow monitor, the usage of a second monitor parameter (The secondary monitor parameters are EGRH dp Error and EGRL dp Error; the primary monitor parameter is the Air Estimation Error.) for pinpointing requires a threshold determination logic; accordingly cloud-plotting all low pressure EGR BPUs with enable conditions (which are mainly low calibrated EGR position demands, calibrated MAP values) of high pressure EGR, helps to set high pressure delta pressure error threshold and vice versa.

Examples of the EGR high flow monitor offers many advantage compared to current EGR high flow strategies: i) Non-intrusive monitor, ii) In-Use-Performance-Ratio (1UPR) improvement compared to intrusive strategies; iii) no need of engine idle for leak detection; iv); innovative and independent pinpointing logic in case of hybrid EGR systems. In addition, having an independent pinpointing logic between low and high pressure EGR, solves the issue of correct pinpointing in case of fault AMIE, in fact, large negative deviations on AMIE generate a negative AEE but does not impact EGR delta pressure errors, avoiding false EGR detection v) No additional logic during DPF regeneration

Claims (1)

1. CLAIMS1. A method of diagnosing a fault in a vehicle engine, said engine including at least one Exhaust Gas Recirculation (EGR) feedback line including a valve located therein, comprising: a) determining the actual pressure differential across said valve; b) determining the exhaust gas flow exiting the exhaust (EMF); c) determining a value of a computed reference pressure differential (EGRL dP*) from the value of intake manifold air pressure; d) comparing the results from a) and c) to determine a pressure differential error; e) diagnosing a fault based on the determined pressure differential error from step d).;2. A method as claimed in claim 1 where in step c) said value is determined from a pre-stored relationship between values of EMF and values of computed reference pressure differential (EGRL dP*) 3. A method as claimed in claim 1 or 2 where said EGR feedback line is a low pressure EGR feedback line.4. A method as claimed in claim 1 to 3 wherein in step a) the pressure differential is determined from a pressure sensor across the inlet and outlet of the valve.5. A method as claimed in claim 1 to 4 wherein in step a) the pressure differential is estimated or modelled.6. A method as claimed in claim 5 wherein in step a) the pressure differential is estimated or modelled using the inputs of exhaust pressure and intake manifold pressure.7. A method as claimed in claims 1 to 6 wherein in step b) EMF is estimated or modelled.8. A method as claimed in claim I to 7 wherein including the further step of comparing the pressure differential error with a threshold value and where step e) comprises outputting an indication of error if the pressure differential error is higher than said threshold value.9. A method as claimed in claim 1 to 8 wherein said threshold value is a calibrated threshold value based on operating conditions of the engine.10. A method as claimed in claim 1 to 9 including determining an Air Estimation Error and wherein step e) comprises diagnosing a fault based on the Air Estimation Error as well as the determined pressure differential error form d).11. A method as claimed in claim 1 to 10 wherein said Air Estimation Error is compared with an AEE calibrated threshold here step e) comprises flagging a fault if the both AEE is more than said AEE calibrated threshold and said pressure differential error is higher than said threshold value.12. A method as claimed in claim 1 to 12 wherein said AEE is determined 20 form comparing a value of estimated manifold intake flow and a flow estimate provided from a Speed Density model.13. A method as claimed in claim 1 to 12 where the speed density model provided a flow estimate based on the Manifold Air Pressure and the Manifold Air Temperate and volumetric efficiency of the engine.14. A method as claimed in claim 1 to 13 where said engine system includes a further EGR feedback line and said method include determining a further pressure differential error therefor and wherein in step e) comprises diagnosing a fault based on the determined pressure differential error form d and said further pressure differential error.15. A method as claimed in claim 1 to 14 wherein said further EGR feedback line is a high pressure feedback line 16. A method as claimed in claim Ito 15 wherein step e) comprises diagnosing a fault based on the determined pressure differential error from d) , said further pressure differential error. and said air estimation error.