DE102018204717B3 - Method and device for tank leak diagnosis in a motor vehicle - Google Patents

Abstract

The invention relates to a method and a device for tank leak diagnosis in a motor vehicle. In this method, the fuel tank is first evacuated and then shut off. When the fuel tank is shut off, a total tank pressure gradient is determined. From this total pressure gradient, the proportion of the increase in pressure caused by a tank leak and the proportion of the pressure increase caused by outgassing in the fuel tank are determined on the basis of the total pressure gradient. From the proportion of pressure increase caused by a tank leak, the diameter of the tank leak is determined.

Description

The invention relates to a method and a device for tank leak diagnosis in a motor vehicle.

According to country-specific legislation, it is necessary to check the tank ventilation system including fuel tank and activated carbon filter of a motor vehicle driven by a motor vehicle for leaks.

One known principle of such tank leak diagnosis is to evacuate and then shut off the fuel tank and calculate a leak diameter based on a tank pressure gradient that is established when the fuel tank is shut off.

In practice, it has been shown that misdiagnosis can occur in such a tank leak diagnosis.

In the US 5,355,864 A there is described a fuel evaporation processing system for an internal combustion engine in which a first control valve is disposed above an evaporative fuel supply passage extending from a fuel tank to a canister, with a second control valve disposed in a purge passage leading from the canister to the intake system of the internal combustion engine leads. There is a third control valve provided at an air inlet opening of the canister. A controller generates actuating command signals to the first to third control valves to close or open the same. The controller responds to an output from an in-tank pressure sensor that detects the pressure in the fuel tank and the operation command signals to detect an abnormality in the operation of the predetermined operation of the first to third control valves.

The DE 10 2006 006 842 A1 shows a leak detection system for a vehicle fuel tank. The system includes a sensor that responds to pressure in the fuel tank and generates a pressure signal based thereon. A control module begins a natural vacuum test with the engine off and determines a pressure change based on the pressure signal. The control module interrupts this test if the pressure change exceeds a pressure change threshold. The control module resumes the test if the pressure change is below the pressure change threshold.

From the US 5,575,265 A For example, a method of diagnosing leakage in a system for recovering fuel vapor vaporized from a fuel tank of the internal combustion engine is known. There are provided venting means for venting fuel vapor into the atmosphere. The method comprises the steps of detecting the leakage using means to close the system, depressurizing, pressurizing in the closed system, detecting a pressure or change in the closed system, and comparing an operating condition to a non-operating condition of the vent. The comparing step comprises obtaining pressure change rates or equivalent amounts of pressure change rates in the operation state of the deaerator and in the non-operation state of the deaerator, and calculating a difference by a ratio of the one of the pressure change rates and the equivalent amounts of the pressure change rates.

The object of the invention is to provide a method and a device for tank leak diagnosis in a motor vehicle, in which such misdiagnosis are avoided.

This object is achieved by a method having the features specified in claim 1. Advantageous embodiments and further developments of the invention are specified in the dependent claims 2-18. The claim 19 has a device for tank leak diagnosis the subject.

The advantages of the invention are in particular that a tank leak diagnosis according to the invention provides more accurate results than known tank leak diagnoses. In particular, faulty tank leak diagnoses can be avoided, which are due for example to the respective ambient conditions which have an effect on the outgassing behavior of the fuel.

• 1 eine Skizze einer Vorrichtung zur Tankleckdiagnose bei einem Kraftfahrzeug,
• 2 ein Diagramm, in welchem der Differenzdruck zwischen Umgebung und Kraftstofftank über der Zeit aufgetragen ist,
• 3 ein Diagramm, in welchem der Druckgradient über der Zeit aufgetragen ist,
• 4 ein Diagramm, in welchem Aktuatorstellungen über der Zeit aufgetragen sind,
• 5 ein Diagramm, in welchem Druckgradienten über der Zeit beim Vorliegen eines Lecks aufgetragen sind und
• 6 ein Diagramm, in welchem Druckgradienten über der Zeit beim Vorliegen eines leckfreien Kraftstofftanks aufgetragen sind.

Further advantageous features of the invention will become apparent from the following exemplary explanation with reference to FIGS. It shows

• 1 a sketch of a device for tank leak diagnosis in a motor vehicle,
• 2 a diagram in which the differential pressure between environment and fuel tank over time is plotted
• 3 a diagram in which the pressure gradient over time is plotted
• 4 a diagram in which actuator positions are plotted over time,
• 5 a graph in which pressure gradients are plotted over time in the presence of a leak, and
• 6 a graph in which pressure gradients over time in the presence of a leak-free fuel tank are plotted.

The 1 shows a sketch of a device for tank leak diagnosis in a motor vehicle. This device has an air filter 1 , a shut-off valve 2 , an activated carbon filter 3 , a differential pressure sensor 4 , a tank vent valve 5 , a ventilation valve 6 , an air filter 7 , a vent line 8th , a fresh air pipe 9 , a fresh air pipe 10 , a scavenging air line 11 , a scavenging air line 12 , a fuel tank 13 and a control unit 14 on.

The air filter 1 is about the fresh air line 9 with the shut-off valve 2 connected. The fresh air line 10 is between the shut-off valve 2 and the activated carbon filter 3 arranged. The activated carbon filter 3 is also on the scavenging air line 11 with the tank vent valve 5 in connection. The tank vent valve 5 is via the scavenging air line 12 with one in the 1 Combustion engine, not shown connected. Furthermore, the activated carbon filter is available 3 over the vent line 8th with the fuel tank 13 and with the vent valve 6 in connection. The ventilation valve 6 is to the air filter 7 connected. The differential pressure sensor 4 is in the fuel tank 13 arranged and is used to measure the pressure difference between the interior of the fuel tank 13 and its exterior space provided. An embodiment, not shown, is to position the differential pressure sensor at a different location of the device, for example in the region of the vent line 8th ,

A control unit 14 is designed to control the device for tank leak diagnosis. It evaluates sensor signals se including, among others, the differential pressure sensor 4 provided differential pressure signals include. Furthermore, the control unit generates 14 Control signals for the actuators of the device. By means of the control signal s1 becomes the shut-off valve 2 driven. The control signal s2 serves to control the ventilation valve 6 , The control signal s3 is for controlling the tank ventilation valve 5 intended.

In normal operation in the 1 The apparatus shown is the activated carbon filter 3 provided from the fuel tank 13 to adsorb exiting hydrocarbon gases. By means of the tank ventilation valve 5 takes place in normal operation in the 1 apparatus shown a regenerative purging of the activated carbon filter.

An advantageous embodiment of the invention is to further arrange a gas temperature sensor in the fuel tank.

In the context of carrying out a tank leak diagnosis, which is another mode of operation in the 1 The device operates according to one shown by the control unit 14 controlled tank diagnostic algorithm.

$${\dot{m}}_{HC}=k\frac{\left(p_{vap,HC}-p_{part,HC}\right)}{p_{Fueltank}}=k\frac{-\Delta p_{HC}}{p_{Fueltank}}$$

$$\begin{array}{l}pV=m{R}_{s}T\\ \dot{p}V=\dot{m}{R}_{s}T\end{array}$$

$$p_{fueltank}=p_{parc,HC}+p_{part,air}$$

$$\begin{array}{c}{\dot{p}}_{fueltank}=D\dot{T}{P}_{fueltank}\\ ={\dot{p}}_{part,HC}+{\dot{p}}_{part,air}\end{array}$$

$$D\dot{T}{P}_{fueltank}=\frac{T}{V}\ast \left(R_{air}\ast \sqrt{2{\rho }_{Air}}\alpha A\sqrt{-DTP}+R_{HG}\ast k\frac{-\Delta p_{HC}}{p_{fueltank}}\right)$$

The control unit does this 14 inter alia, use of the following equations: $${\dot{m}}_{air}=\sqrt{2{\rho }_{Air}}\alpha A\sqrt{p_{Fueltank}-p_{Ambient}}=\sqrt{2{\rho }_{ambient}}\alpha A\sqrt{-DTP}$$

$${\dot{m}}_{HC}=k\frac{\left(p_{vap.HC}-p_{part.HC}\right)}{p_{Fueltank}}=k\frac{-\Delta p_{HC}}{p_{Fueltank}}$$

$$\begin{array}{l}pV=m{R}_{s}T\\ \dot{p}V=\dot{m}{R}_{s}T\end{array}$$

$$p_{fueltank}=p_{parc.HC}+p_{part.air}$$

$$\begin{array}{c}{\dot{p}}_{fueltank}=D\dot{T}{P}_{fueltank}\\ ={\dot{p}}_{part.HC}+{\dot{p}}_{part.air}\end{array}$$

$$D\dot{T}{P}_{fueltank}=\frac{T}{V}*\left(R_{air}*\sqrt{2{\rho }_{Air}}\alpha A\sqrt{-DTP}+R_{HG}*k\frac{-\Delta p_{HC}}{p_{fueltank}}\right)$$

ṁ_air

der durch eine Leckage verursachte Luftmassenstrom [kg/s]

ṁ_HC:

Massenstrom des verdunstenden Kohlenwasserstoffs [kg/s]

ρ_Air:

Luftdichte [kg/m³]

α:

Durchflusskoeffizient [-]

A:

Leckfläche [m²]

p_fueltank:

Absolutdruck im Kraftstofftank [Pa]

p_ambient:

Umgebungsdruck [Pa]

DTP:

Differenzdruck (Kraftstofftank - Umgebung) [Pa]

k:

Ausgaskoeffizient [kg/s]

p_part,HC:

Partialdruck des Kohlenwasserstoffgases [Pa]

p_vap,HC:

Dampfdruck des Kohlenwasserstoffgases [Pa]

Δp_HC:

Differenz zwischen Partialdruck und Dampfdruck des Kohlenwasserstoffgases [Pa]

R_S:

Spezifische Gaskonstante [J/(kg K)]

V:

Gasvolumen im Kraftstofftank inkl. der Volumina des Aktivkohlefilters und der Leitungen zu den Ventilen [m³]

T:

Gastemperatur im Kraftstofftank [K]

t:

Zeit [s].

Here are:

ṁ _air

the air mass flow caused by a leak [kg / s]

_HC :

Mass flow of the evaporating hydrocarbon [kg / s]

ρ _Air :

Air density [kg / m ³ ]

α:

Flow coefficient [-]

A:

Leakage area [m ² ]

p _fueltank :

Absolute pressure in the fuel tank [Pa]

p _ambient :

Ambient pressure [Pa]

DTP:

Differential pressure (fuel tank environment) [Pa]

k:

Outlet gas coefficient [kg / s]

p _{part, HC} :

Partial pressure of the hydrocarbon gas [Pa]

p _{vap, HC} :

Vapor pressure of the hydrocarbon gas [Pa]

Δp _HC :

Difference between partial pressure and vapor pressure of the hydrocarbon gas [Pa]

R _S:

Specific gas constant [J / (kg K)]

V:

Gas volume in the fuel tank including the volumes of the activated carbon filter and the lines to the valves [m ³ ]

T:

Gas temperature in the fuel tank [K]

t:

Time [s].

The one from the control unit 14 Controlled tank diagnostic algorithm works in several phases.

In a first phase, an evacuation of the fuel tank takes place 13 , This evacuation is done by closing the shut-off valve 2 and adjusting a predetermined purge air mass flow through the tank vent valve 5 performed. The ventilation valve 6 is also closed during this phase. In this evacuation, both the partial pressure of the air and the partial pressure of the hydrocarbon gas in the fuel tank decreases 13 , As a result of the decrease in the hydrocarbon partial pressure below the vapor pressure, a mass flow from the liquid hydrocarbon phase into the gaseous phase sets in, as can be seen from equation (2) given above.

In a subsequent second phase, a determination of the pressure gradient takes place. This second phase begins as soon as a defined vacuum level is reached. At the beginning of this second phase, the tank vent valve 5 closed. The target pressure represents a compromise of the shortest possible evacuation time and sufficiently high leakage pressure gradient. The venting valve 6 can be opened momentarily to the connection between the vent valve 6 and the fuel tank 13 to fill with air. This results in a smaller effect on the hydrocarbon partial pressure during a subsequent third phase, which will be explained below. Subsequently, a shut-off of the fuel tank 13 , The positive pressure gradient occurring after this shut-off is additive in the presence of a leakage from an increase in the hydrocarbon partial pressure due to evaporation and an increase in the air partial pressure caused by the leakage. This is illustrated by equation (5) given above. The sum of both partial pressure gradients becomes the total pressure gradient at the differential pressure sensor 4 measured.

In a subsequent third phase, the negative pressure in the fuel tank 13 by opening the ventilation valve 6 raised to a pressure level that corresponds to the ambient pressure. The partial pressure of the hydrocarbon gas is not changed by this aeration. Therefore, according to the above equations (2) and (3), the partial pressure gradient of the hydrocarbon gas also remains approximately unchanged. If a short aeration phase is assumed, then approximately a constant hydrocarbon partial pressure gradient can be assumed during this phase. The leakage partial air pressure gradient can be neglected due to the balanced pressure level after the aeration phase. Is a representative cross section A and a flow coefficient α the between the ventilation valve 6 and the fuel tank 13 provided vent line known, then can be during this phase, the gas volume in the fuel tank 13 using equations (1) and (3) above, with a small error due to neglecting the evaporation and leakage mass flow. This error can be corrected after completion of the diagnosis if the evaporation and leakage pressure gradient has been quantified. Is the total volume of the fuel tank 13 known, then it also allows the current level of the fuel tank 13 determine.

In a subsequent fourth phase, a determination of the hydrocarbon partial pressure gradient is carried out as follows: after closing the venting valve 6 Again, there is a positive pressure gradient. This corresponds approximately to the partial pressure gradient of the hydrocarbon gas, since the air-Partialdruckgradient can be neglected due to the balanced pressure levels. With the assumption of a constant hydrocarbon partial pressure gradient, the measured pressure gradient after the aeration phase can be used to determine the hydrocarbon fraction of the pressure gradient before the aeration phase. This makes it possible to clearly divide the pressure gradient determined in the second phase using equation (2) into the air partial pressure gradient and the hydrocarbon partial pressure gradient. With the equation (1) given above, it is possible to determine from this partial air pressure gradient a cross-section through which flows and, therefrom, a leak diameter.

A plausibility diagnosis of the differential pressure signal may be performed in the first, second and fourth phases. In particular, for a leak diagnosis while driving can also monitor the dynamics of the differential pressure signal during these phases. Thus, for example, with heavily sloshing fuel, the diagnosis can be stopped to avoid misdiagnosis. Furthermore, a start of the diagnosis can be prevented or at least delayed with greatly increased dynamics of the pressure difference signal.

Furthermore, a closed clamping vent valve 6 be recognized during the third phase by an evaluation of the differential pressure gradient. An open clamping vent valve 6 or large leaks, for example greater than 5 mm, may be recognized during the first phase by the fact that the target pressure during the evacuation can not be achieved within a predetermined period of time.

An associated calculation approach is explained in more detail below:

$$D\dot{T}{P}_4=\frac{T}{V}\ast \left(R_{air}\ast \sqrt{2{\rho }_{Air}}\alpha A\sqrt{-DT{P}_4}+R_{HC}\ast k\frac{-\Delta p_{HC}}{p_{fuelltank\mathrm{,4}}}\right)$$

It describes DTP the measured differential pressure. In general, the application of Equation (5) above at the end of the second phase and at the beginning of the fourth phase results in the following: $$D\dot{\mathrm{<figure-callout\; id="2"\; label="T\; ̇\; P"\; filenames="DE102018204717B3\_0021.png,DE102018204717B3\_0022.png"\; state="\{\{state\}\}">T\; </figure-callout>}}{P}_{}{}_2=\frac{T}{V}*\left(R_{air}*\sqrt{2{\rho }_{Air}}\alpha A\sqrt{-\mathrm{<figure-callout\; id="2"\; label="D\; T\; P"\; filenames="DE102018204717B3\_0021.png,DE102018204717B3\_0022.png"\; state="\{\{state\}\}">D\; </figure-callout>}T{P}_{}{}_2}+R_{HC}*k\frac{-\Delta p_{HC}}{p_{fuelltank2}}\right)$$

$$D\dot{\mathrm{<figure-callout\; id="4"\; label="T\; ̇\; P"\; filenames="DE102018204717B3\_0021.png,DE102018204717B3\_0022.png"\; state="\{\{state\}\}">T\; </figure-callout>}}{P}_{}{}_4=\frac{T}{V}*\left(R_{air}*\sqrt{2{\rho }_{Air}}\alpha A\sqrt{-\mathrm{<figure-callout\; id="4"\; label="D\; T\; P"\; filenames="DE102018204717B3\_0021.png,DE102018204717B3\_0022.png"\; state="\{\{state\}\}">D\; </figure-callout>}T{P}_{}{}_4}+R_{HC}*k\frac{-\Delta p_{HC}}{p_{fuelltank\mathrm{,\; 4}}}\right)$$

This describes the subscript digit 2 the circumstances at the end of the second phase and the lower number 4 the conditions at the beginning of the fourth phase.

With the assumption of a constant hydrocarbon partial pressure Δp _HC during the third phase results: $$A=\frac{D\dot{T}{P}_2-D\dot{\mathrm{<figure-callout\; id="4"\; label="T\; ̇\; P"\; filenames="DE102018204717B3\_0021.png,DE102018204717B3\_0022.png"\; state="\{\{state\}\}">T\; </figure-callout>}}{P}_{}{}_4*\frac{p_{fueltank\mathrm{,\; 4}}}{p_{fueltank2}}}{\sqrt{-DT{P}_2}-\sqrt{-\mathrm{<figure-callout\; id="4"\; label="D\; T\; P"\; filenames="DE102018204717B3\_0021.png,DE102018204717B3\_0022.png"\; state="\{\{state\}\}">D\; </figure-callout>}T{P}_{}{}_4}*\frac{p_{fueltank\mathrm{,\; 4}}}{p_{fuelta\mathrm{<figure-callout\; id="2"\; label="n\; k"\; filenames="DE102018204717B3\_0021.png,DE102018204717B3\_0022.png"\; state="\{\{state\}\}">n\; </figure-callout>}k2}}}*{\left(\frac{R_{air}T}{V}*\alpha *\sqrt{2{\rho }_{air}}\right)}^{-1}$$

vereinfacht sich obige Gleichung zu: $$A=\frac{D\dot{T}{P}_2-D\dot{T}{P}_4}{\sqrt{-DT{P}_2}}\ast {\left(\frac{R_{air}T}{V}\ast \alpha \ast \sqrt{2{\rho }_{air}}\right)}^{-1}$$

With the assumption of DTP ₄ = 0 Pa and $\frac{p_{fueltank\mathrm{,\; 4}}}{p_{fuelta\mathrm{<figure-callout\; id="2"\; label="n\; k"\; filenames="DE102018204717B3\_0021.png,DE102018204717B3\_0022.png"\; state="\{\{state\}\}">n\; </figure-callout>}k2}}\approx 1$

Simplifies the above equation to: $$A=\frac{D\dot{T}{P}_2-D\dot{T}{P}_4}{\sqrt{-\mathrm{<figure-callout\; id="2"\; label="D\; T\; P"\; filenames="DE102018204717B3\_0021.png,DE102018204717B3\_0022.png"\; state="\{\{state\}\}">D\; </figure-callout>}T{P}_{}{}_2}}*{\left(\frac{R_{air}T}{V}*\alpha *\sqrt{2{\rho }_{air}}\right)}^{-1}$$

The assumption of a constant hydrocarbon partial pressure gradient can lead to a non-negligible error with long-lasting aeration in the third phase. For this reason, the pressure gradient DṪP ₄ measured in the fourth phase can be corrected before being subtracted from the pressure gradient DṪP ₂ from the second phase. With the assumption of a constant temperature T and hence a constant hydrocarbon vapor pressure p _{vap, HC} (T), the following differential equation of the first order can be derived from equations (2) and (3) given above: $$\Delta {\dot{p}}_{HC}\left(t\right)={\dot{p}}_{part.HC}\left(t\right)-{\dot{p}}_{vap.HC}={\dot{p}}_{art.HC}\left(t\right)=\frac{R_{HC}T}{V}*k*\frac{\Delta p_{HC}\left(t\right)}{p_{fueltank}}$$

Assuming a constant absolute pressure P _fueltank , the following solution of the above differential equation is assumed: $$\Delta p_{HC}\left(t\right)=\Delta p_{HC.t=0}*e^{-\frac{RTk}{V{p}_{fueltank}}k}$$

With the assumption of DṪP ₄ = ṗ _{part, HC, 4} , the decrease of the hydrocarbon partial pressure gradient during the duration of the third phase can be described as follows: $$f_{cOr}=\frac{{\dot{p}}_{part.HC2}}{D\dot{\mathrm{<figure-callout\; id="4"\; label="T\; ̇\; P"\; filenames="DE102018204717B3\_0021.png,DE102018204717B3\_0022.png"\; state="\{\{state\}\}">T\; </figure-callout>}}{P}_{}{}_4}=\frac{{\dot{p}}_{part.HC2}}{{\dot{p}}_{part.HC\mathrm{,\; 4}}}=\frac{-\frac{R_{HC}T}{V{p}_{fueltank}}k\Delta p_{HC2}}{-\frac{R_{HC}T}{V{p}_{fueltank}}k\Delta {\mathrm{<figure-callout\; id="2"\; label="p\; H\; C"\; filenames="DE102018204717B3\_0021.png,DE102018204717B3\_0022.png"\; state="\{\{state\}\}">p\; </figure-callout>}}_{HC}*e^{-\frac{RTk}{V{p}_{fueltank}}\Delta t}}=e^{\frac{RTk}{V{p}_{fueltank}}\Delta t}$$

If the hydrocarbon partial pressure gradient is known, then using the equations (2) and (3) given above, the difference between the vapor pressure and the partial pressure can be determined Δp _HC approach the hydrocarbon gas at the end of the evacuation phase. Substituting this difference in relation to the absolute pressure difference during the first phase, then it can be concluded from the partial pressure of the hydrocarbon gas at the beginning of the first phase and from there to the vapor pressure of the fuel at the current gas temperature in the fuel tank. This temperature can be approximated by the ambient temperature or determined by a temperature model or by a corresponding sensor.

The 2 shows a diagram in which the differential pressure over time is plotted. This describes the curve K1 the course of the differential pressure sensor 4 measured differential pressure DTP over time in the four phases of tank leak diagnosis described above. It can be seen that in the course of the first phase, this differential pressure drops approximately linearly from the ambient pressure to a target pressure value which in the embodiment shown is about -18 hPa, then slightly increases in the course of the second phase to about -16 hPa, then rises steeply in the third phase to ambient pressure and finally in the fourth phase rises approximately constant. Furthermore, the curve describes K2 the course of the difference Δp _HC between the partial pressure and the vapor pressure of the hydrocarbon gas. It can be seen that this difference decreases approximately linearly during the first phase and then in the Continuously increases during the second, third and fourth phase. It can be seen from this graph that the hydrocarbon partial pressure only slightly changes between the second phase and the fourth phase, while the differential pressure increases significantly. The state of balanced pressure levels between the tank and the environment during the fourth phase thus leads to a negligible, due to leakage pressure gradient, the outgassing-related pressure gradient remains approximately identical to the second phase.

The 3 shows a diagram in which the pressure gradient over time is plotted. This describes the curve K3 the course of the gradient DTP of the differential pressure sensor 4 measured differential pressure over time in the four phases of the tank leak diagnosis described above. The curve K4 describes the course of the gradient of the partial pressure of the hydrocarbon gas. From the on the right side of the 3 shown enlarged detail is particularly apparent as divided in the embodiment shown, the measured differential pressure gradient on a conditional by a leak share and a conditional by an outgassing. During the fourth phase, the air partial pressure gradient caused by a leak is negligibly small, which is why it can be assumed approximately that the measured differential pressure gradient corresponds to the hydrocarbon partial pressure gradient caused by outgassing. By subtracting the hydrocarbon partial pressure gradient from the measured differential pressure gradient in the second phase, it is possible to deduce the air partial pressure gradient caused by a leak.

The 4 shows a diagram in which associated actuator positions are plotted over time. This describes the curve K5 the position of the tank ventilation valve 5 , the curve K6 the position of the ventilation valve 6 and the curve K7 the position of the shut-off valve 2 , From this diagram it can be seen that the shut-off valve 2 is closed during the entire diagnostic period. For pressure equalization during the third phase, the vent valve 6 opened to prevent a change in the hydrocarbon partial pressure in the fuel tank by purging the activated carbon filter. The tank vent valve 5 is used exclusively to evacuate the fuel tank. During the second and fourth phases all valves are closed.

The 5 and 6 illustrate the results of experiments.

The 5 shows a graph in which pressure gradients over time in the presence of a leak with a diameter of 0.9 mm diameter are plotted. This describes the curve K8 the course of the differential pressure sensor 4 measured differential pressure DTP over time. The curve K9 describes the course of the filtered gradient of the differential pressure sensor 4 measured differential pressure DTP over time. The curve K10 describes the course of the estimated gradient of the partial pressure of the hydrocarbon gas, based on the gradient of the differential pressure measured in the fourth phase (second 52,818 in the following table) DTP , It can be seen from this graph that at the end of the second phase (second 48,087) there is a significant difference between the estimated gradient of the hydrocarbon partial pressure and the measured gradient of the differential pressure DTP results. This indicates a pressure gradient due to leakage.

The test results obtained are shown numerically below: Time s 48.087 52,818 differential pressure hPa -14.253 -1.222 Filtered differential pressure gradient hPa / s 0.978 0,271 Estimated gradient of hydrocarbon partial pressure hPa / s 0,399 0,271

$${\dot{p}}_{part,air\mathrm{,2}}=D\dot{T}{P}_2-{\dot{p}}_{part,HC\mathrm{,2}}=\mathrm{0,579}\frac{hPa}{s}$$

This results in: $$D\dot{\mathrm{<figure-callout\; id="2"\; label="T\; ̇\; P"\; filenames="DE102018204717B3\_0021.png,DE102018204717B3\_0022.png"\; state="\{\{state\}\}">T\; </figure-callout>}}{P}_{}{}_2=0.978\frac{HPa}{s};D\dot{\mathrm{<figure-callout\; id="4"\; label="T\; ̇\; P"\; filenames="DE102018204717B3\_0021.png,DE102018204717B3\_0022.png"\; state="\{\{state\}\}">T\; </figure-callout>}}{P}_{}{}_4=\mathrm{0,271}\frac{HPa}{s};{\dot{p}}_{art.\mathrm{<figure-callout\; id="2"\; label="H\; C"\; filenames="DE102018204717B3\_0021.png,DE102018204717B3\_0022.png"\; state="\{\{state\}\}">H\; </figure-callout>}C2}=\mathrm{0,399}\frac{HPa}{s}$$

$${\dot{p}}_{part.a\mathrm{<figure-callout\; id="2"\; label="i\; r"\; filenames="DE102018204717B3\_0021.png,DE102018204717B3\_0022.png"\; state="\{\{state\}\}">i\; </figure-callout>}r2}=D\dot{T}{P}_2-{\dot{p}}_{part.\mathrm{<figure-callout\; id="2"\; label="H\; C"\; filenames="DE102018204717B3\_0021.png,DE102018204717B3\_0022.png"\; state="\{\{state\}\}">H\; </figure-callout>}C2}=0.579\frac{HPa}{s}$$

ermitteln, dass der Kraftstofftank ein Leck hat, dessen Durchmesser 0,9 mm beträgt. From this air fraction can be calculated using the above relationship $A=\frac{D\dot{T}{P}_2-f_{cOr}*D\dot{T}{P}_4}{\sqrt{-\mathrm{<figure-callout\; id="2"\; label="D\; T\; P"\; filenames="DE102018204717B3\_0021.png,DE102018204717B3\_0022.png"\; state="\{\{state\}\}">D\; </figure-callout>}T{P}_{}{}_2}}*{\left(\frac{R_{air}T}{V}*\alpha *\sqrt{2{\rho }_{air}}\right)}^{-1}$

Determine that the fuel tank has a leak whose diameter is 0.9 mm.

The 6 shows a diagram in which pressure gradients are plotted over time in the presence of a leak-free fuel tank. This describes the curve K11 the course of the differential pressure sensor 4 measured differential pressure DTP over time. The curve K12 describes the course of the filtered gradient of the differential pressure sensor 4 measured differential pressure DTP over time. The curve K13 describes the course of the estimated gradient of the partial pressure of the hydrocarbon gas, based on the gradient of the differential pressure measured in the fourth phase (second 123.49 in the following table) DTP , It can be seen from this graph that at the end of the second phase (second 115,479) there is only a small difference between the estimated gradient of the hydrocarbon partial pressure and the measured gradient of the differential pressure DTP results.

This indicates a very small pressure gradient due to leakage.

The test results obtained are shown numerically below: Time s 115.479 123.490 differential pressure hPa -14.578 0.081 Filtered differential pressure gradient hPa / s 0,347 0,179 Estimated gradient of hydrocarbon partial pressure hPa / s 0,325 0,180

$${\dot{p}}_{part,air\mathrm{,2}}=D\dot{T}{P}_2-{\dot{p}}_{part,HC\mathrm{,2}}=\mathrm{0,022}\frac{hPa}{s}$$

This results in: $$D\dot{\mathrm{<figure-callout\; id="2"\; label="T\; ̇\; P"\; filenames="DE102018204717B3\_0021.png,DE102018204717B3\_0022.png"\; state="\{\{state\}\}">T\; </figure-callout>}}{P}_{}{}_2=\mathrm{0,347}\frac{HPa}{s};D\dot{\mathrm{<figure-callout\; id="4"\; label="T\; ̇\; P"\; filenames="DE102018204717B3\_0021.png,DE102018204717B3\_0022.png"\; state="\{\{state\}\}">T\; </figure-callout>}}{P}_{}{}_4=\mathrm{0,179}\frac{HPa}{s};{\dot{p}}_{art.\mathrm{<figure-callout\; id="2"\; label="H\; C"\; filenames="DE102018204717B3\_0021.png,DE102018204717B3\_0022.png"\; state="\{\{state\}\}">H\; </figure-callout>}C2}=\mathrm{0,325}\frac{HPa}{s}$$

$${\dot{p}}_{part.a\mathrm{<figure-callout\; id="2"\; label="i\; r"\; filenames="DE102018204717B3\_0021.png,DE102018204717B3\_0022.png"\; state="\{\{state\}\}">i\; </figure-callout>}r2}=D\dot{T}{P}_2-{\dot{p}}_{part.\mathrm{<figure-callout\; id="2"\; label="H\; C"\; filenames="DE102018204717B3\_0021.png,DE102018204717B3\_0022.png"\; state="\{\{state\}\}">H\; </figure-callout>}C2}=\mathrm{0,022}\frac{HPa}{s}$$

ermitteln, dass der Kraftstofftank leckfrei ist.From this air fraction can be calculated using the equation given above $A=\frac{D\dot{T}{P}_2-f_{cOr}*D\dot{T}{P}_4}{\sqrt{-\mathrm{<figure-callout\; id="2"\; label="D\; T\; P"\; filenames="DE102018204717B3\_0021.png,DE102018204717B3\_0022.png"\; state="\{\{state\}\}">D\; </figure-callout>}T{P}_{}{}_2}}*{\left(\frac{R_{air}T}{V}*\alpha *\sqrt{2{\rho }_{air}}\right)}^{-1}$

Determine that the fuel tank is leak-free.

LIST OF REFERENCE NUMBERS

1

air filter

2

shut-off valve

3

Activated carbon filter

4

Differential Pressure Sensor

5

Tank ventilation valve

6

vent valve

7

air filter

8th

vent line

9

Fresh air line

10

Fresh air line

11

flushing air

12

flushing air

13

Fuel tank

14

control unit

se

sensor signals

s1

control signal

s2

control signal

s3

control signal

Claims (19)

Method for tank leak diagnosis in a motor vehicle, in which the fuel tank (13) is evacuated and then shut off, and with shut-off fuel tank (13) a Gesamttankdruckgradient is determined, wherein the proportion of the pressure increase caused by a tank leak and the proportion of an outgassing in Fuel tank (13) located fuel pressure increase is determined on Gesamtdruckgradienten and the diameter of the tank leak from the proportion of the pressure caused by the tank leak pressure increase is determined, characterized in that after the determination of the total pressure gradient directly connected to the fuel tank (13) ventilation valve (6 ) is opened, so that the total tank pressure in the fuel tank (13) is raised to the ambient pressure level. Method according to Claim 1 , characterized in that the evacuation of the fuel tank (13) by closing a shut-off valve (2) and setting a predetermined purge air mass flow through a tank vent valve (5) is made. Method according to Claim 2 , characterized in that the shut-off of the fuel tank (13) by closing the tank vent valve (5) is carried out when the measured by a pressure sensor total tank pressure in the fuel tank (13) has dropped to a predetermined vacuum level. Method according to one of the preceding claims, characterized in that after the increase of the total tank pressure to the ambient pressure level, the gas volume in the fuel tank (13) is determined. Method according to one of the preceding claims, characterized in that after the increase of the total tank pressure to the ambient pressure level of the current level of the fuel tank (13) is determined. Method according to one of the preceding claims, characterized in that after the increase of the total tank pressure to the ambient pressure level, the venting valve (6) is closed again. Method according to Claim 6 , characterized in that after closing the venting valve (6), a new determination of the total pressure gradient is made. Method according to Claim 7 , characterized in that a determination of the present before the opening of the venting valve (6) hydrocarbon content of the total pressure gradient is carried out by an evaluation of the determined after closing of the venting valve (6) Gesamtdruckgradienten. Method according to Claim 8 , characterized in that the proportion of the pressure rise caused by the tank leak from the measured total pressure gradient and the determined before the opening of the vent valve (6) present hydrocarbon content of the total pressure gradient is determined. Method according to Claim 9 , characterized in that the diameter of a present tank leak from the determined proportion of the pressure increase caused by the tank leak is determined. Method according to one of the preceding claims, characterized in that a plausibility check of a measured differential pressure signal is performed. Method according to Claim 11 , characterized in that in the context of the plausibility check the dynamics of the differential pressure signal is monitored while driving the motor vehicle. Method according to Claim 12 , characterized in that upon detection of an impermissibly high dynamics of the differential pressure signal, a start of the leak diagnosis is prevented or an already started leak diagnosis is canceled. Method according to one of the preceding claims, characterized in that a clamping of the closed venting valve (6) is detected by an evaluation of the gradient of the measured pressure difference signal. Method according to one of the preceding claims, characterized in that a clamping of the opened venting valve (6) and / or a larger tank leak is detected by an evaluation of the pressure curve during the evacuation of the fuel tank (13). Method according to Claim 15 , characterized in that the clamping of the opened vent valve (6) and / or a larger tank leak is detected when a predetermined target pressure is not reached within a predetermined time during the evacuation. Method according to one of the preceding claims, characterized in that the difference between the vapor pressure and the partial pressure of the hydrocarbon gas at the end of the evacuation phase is determined, this difference is set in relation to the absolute pressure difference (DTP) during the evacuation phase, the initial partial pressure of the hydrocarbon gas determine the evacuation phase, and to determine from this partial pressure the vapor pressure of the fuel at the present gas temperature in the fuel tank (13). Method according to Claim 17 , characterized in that the instantaneous gas temperature in the fuel tank (13) is approximated by the ambient temperature or determined using a temperature model or using a sensor. Device for tank leak diagnosis in a motor vehicle, characterized in that it comprises a control unit (14), which is designed such that it controls a method according to one of the preceding claims, and that it directly connected to the fuel tank (13) ventilation valve (6 ), which is controlled via the control unit (14).

Images