EP0576448B1 - Process and device for tank ventilation - Google Patents

Abstract

A process for alternating execution of phases with and without tank ventilation during the operation of an internal combustion engine (10) with a tank ventilator (21, 24-26) is characterized in that the ratio of the time-intervals with and without tank ventilation is chosen so that it is dependent on operating data of the engine or of the tank ventilator. Preferably the quantity measured is a measure of the quantity of fuel to be regenerated during tank ventilation, and the above-mentioned ratio is increased with respect to an initial ratio in favour of the tank ventilation time-interval if the value of the measured quantity exceeds an upper limit (Dp-SMW; FTEA-SWU). This process makes it possible to equip the corresponding device with an adsorption filter (24) and a tank ventilation valve (25) for smaller throughputs than hitherto, without the risk of fuel vapours escaping into the atmosphere. If large quantities of fuel vapor are produced, the tank ventilation time-interval is increased with respect to the basic adaptation time-interval. The smaller adsorption filter is therefore still adequately regenerated despite the smaller cross-section of the tank ventilation valve.

Description

The invention relates to a method and a device for alternating execution of phases with and without tank ventilation when operating an internal combustion engine with tank ventilation system.

State of the art

EP-A-0 208 069 describes a method according to which phases with and without tank ventilation, namely tank ventilation phases and basic adaptation phases, alternate in a fixed pattern. 5 minutes are given for the tank ventilation period and 1 minute for the basic adaptation period. In practice, the first period is somewhat shorter and the second is somewhat longer.

According to EP 0 208 069, the time periods mentioned can also be changed depending on the engine speed in order to be able to carry out predetermined time cycles even when the accelerator pedal is operated frequently.

The duration of the tank ventilation period, together with parameters of the tank system and the associated internal combustion engine, determine the size of the adsorption filter in which fuel vapor from the tank is adsorbed, and these sizes also determine the diameter of the tank ventilation valve, with the aid of which the adsorption filter is flushed with air becomes. The size of the adsorption filter and the cross section of the tank ventilation valve must be such that Even with the greatest possible amount of fuel vapor, essentially all fuel vapor can be adsorbed during the basic adaptation periods and desorbed again during the tank ventilation periods.

The general problem in technology is to operate devices using such methods and to design them in such a way that the components are used as sensibly as possible. This problem also applied accordingly to methods and devices for carrying out phases with and without tank ventilation when operating an internal combustion engine with a tank ventilation system.

Presentation of the invention

The method according to the invention is defined in claim 1.

The device according to the invention is defined in claim 9.

Since this method and this device no longer use a predefined time grid for the phases mentioned, they can use the components involved more flexibly than was previously possible.

According to a development of the method, tank ventilation is carried out continuously at full load without lambda control with the tank ventilation valve fully open. This is based on the knowledge that at full load without lambda control in the phases without tank ventilation, no basic adaptation can be carried out, so that it makes more sense to use the entire time for tank ventilation. Due to the fact that the valve is kept open rather than keyed, it is little stressed.

According to another development, if a diagnostic method for the functionality of the tank ventilation system is started during a tank ventilation phase, which requires a temporary closing of the tank ventilation valve, a basic adaptation phase is started immediately with the closing of the valve and the next tank ventilation phase becomes at least partial compensation for the previous one that has been canceled Extended phase. As a result, the diagnostic time is used sensibly in parallel for adaptation.

It is particularly advantageous to use all the variants mentioned together.

The method with a variable ratio of the time periods mentioned makes it possible to design the adsorption filter and the tank ventilation valve for the throughput of an average amount of fuel from the tank ventilation instead of a maximum amount. These parts, which are thus smaller than previously, are nevertheless capable of satisfactorily satisfying even very high amounts of fuel vapor, as they occur from time to time to vent, because in this case the tank ventilation period is extended at the expense of the basic adaptation period. Shortening the basic adaptation period z. B. up to 1 minute and extending the distance between two such periods to z. B. 15 minutes (duration of the extended tank ventilation period) leads to disadvantages only in exceptional cases, e.g. B. with very fast uphill driving a relatively steep road. Here it could actually be necessary that the factor taking the air density into account in the basic adaptation in the period mentioned by z. B. should change 5% or more. Since he cannot do this because of the locked basic adaptation, the required change in the fuel injection times must be absorbed by the manipulated value of the lambda control, which is, however, in principle possible without any problems, since the typical stroke of lambda controls is approximately 15%. Difficulties arise briefly with transient processes when changing between greatly different operating states, since then only the relatively sluggish control must adapt to the new operating state without optimized support by a well-adapted pilot control. So there is apparently the disadvantage that in the case of extremely high fuel vapor accumulation in the tank ventilation and thus greatly extended tank ventilation periods, while at the same time being very fast and steep uphill, short-term increases in harmful gas emissions can occur during transient processes. However, all of these conditions will very rarely be met together. However, there is always the advantage that a smaller adsorption filter and a smaller tank ventilation valve can be used. This leads to permanent fuel savings, albeit an extremely small one, due to the reduced weight of these parts and thus also to lower pollutant gas emissions. Furthermore, the energy required to manufacture and operate the parts downsized. The mentioned, rarely occurring disadvantage is therefore offset by far-reaching advantages.

Theoretically, the amount of fuel vapor generated in the tank ventilation would be most accurately determined by a flow meter between the tank and the adsorption filter. However, such a flow meter would be extremely expensive and complex if it were to work accurately. It is easier to determine the pressure difference between the tank pressure and the ambient pressure. This requires a pressure difference sensor on the tank, but it is advisable to attach it in many ways in modern tank ventilation systems, which is therefore often available anyway for other reasons. The greater the pressure difference measured by this sensor, the stronger the fuel gas in the tank. The ratio of tank ventilation to basic adaptation time can accordingly be made dependent on this pressure difference. Another very advantageous possibility is to make the ratio mentioned dependent on the tank ventilation adaptation factor itself. This is because it is a direct measure of the amount of fuel vapor currently generated during tank ventilation. However, this value is not updated during the basic adaptation period.

If there is very little fuel vapor in the tank ventilation, it is advantageous to extend the basic adaptation periods at the expense of the tank ventilation periods. In the latter time periods, the tank ventilation valve is actuated in a clocked manner, while in the basic adaptation time periods it is closed without current. It therefore contributes significantly to increasing the service life of the tank ventilation valve if it is only activated when this is actually required for tank ventilation. Another control type of low load mentioned above is to keep the valve open all the time, which is without a load at full load Lambda control is possible.

The answer to the question of how much the tank ventilation period has to be extended in order to prevent the adsorption filter from becoming oversaturated depends not only on how much fuel vapor is supplied to the filter from the tank, but also on how good the filter is in each case Operating state can be rinsed. At idle and with low loads, the pressure at the outlet of the tank ventilation system is so low that the amount of purge gas must be limited by partially closing the tank ventilation valve (corresponding duty cycle). At very high load, especially at full load, however, the flushing effect is sometimes small even when the tank ventilation valve is fully open. It is therefore advantageous to increase the tank ventilation period not only with an increasing amount of fuel vapor supplied to the adsorption filter, but also with an increasing load, that is to say a decreasing purge effect.

drawing

• Fig. 1: block diagram of an internal combustion engine with tank ventilation system and lambda control and function groups for tank ventilation adaptation and basic adaptation;
• 2: Flow diagram for explaining a procedure for increasing the tank ventilation period at the expense of the basic adaptation period on the basis of a differential pressure signal:
• FIG. 3: flow chart corresponding to that of FIG. 2, but for additionally reducing the ratio between the tank ventilation and basic adaptation time period, the change in the ratio taking place on the basis of the tank ventilation adaptation factor;
• Fig. 4: Flow chart for explaining a method for alternately executing basic adaptation and tank ventilation adaptation.
• 5: Flow chart for explaining a method for exclusively performing tank ventilation at full load; and
• 6: Flow chart for explaining a method for starting basic adaptation immediately with the closing of the tank ventilation valve during a tank ventilation phase for diagnostic purposes.

Description of exemplary embodiments

1 shows an internal combustion engine 10 with an intake pipe 11, in which a throttle valve 12 and an injection valve 13 are arranged, and with an exhaust pipe 14, in which an oxygen sensor 15 is attached. The injection times with which the injection valve 13 is operated are determined by adapted pilot control with lambda control. For this purpose, injection times are read out from an injection time map 16 as a function of speed n and load L and are linked to adaptation variables and a control factor FR. The control factor FR is provided by a lambda controller 17, which forms this factor on the basis of a control algorithm based on a control deviation, which corresponds to the difference between a lambda setpoint value read from a setpoint map 18 and the actual lambda value supplied by the lambda probe 15 . The control factor FR, that is to say the manipulated value of the lambda control, is the basis for adapted values as they are formed by a basic adaptation device 19 and a tank ventilation adaptation device 20. The basic adaptation device 19 calculates various correction variables in any known manner. In Fig. 1 illustrates three unspecified quantities for the basic adaptation. In this case, the first additive leakage air error can adapt, the second multiplicative air density changes can be compensated, and the third in turn can adapt additive change-in and fall-time changes of the injection valve 13. The tank ventilation adaption device 20 provides a multiplicative factor FTEA for the tank ventilation, which has the value one during inactive tank ventilation, but in the case of active tank ventilation an adapted value greater or less one, depending on whether the tank ventilation contains a leaner or richer mixture in the Intake pipe leads when it is provided in the mixture formation without tank ventilation adaptation.

As just mentioned, fuel can be supplied to the internal combustion engine 10 in two ways, namely either via the injection valve 13 or via a ventilation line 21 of a tank ventilation system. The fuel injector 13 receives its fuel from a tank 23 via a fuel pump 22. This tank 23 is vented via an adsorption filter 24, a tank vent valve 25 and the vent line 21. As long as the tank ventilation valve 25 is closed, fuel vapor escaping from the tank 23 collects in the adsorption filter 24. During this time, basic adaptation takes place. The tank ventilation adaptation device 20 receives the value one as an input value, with the result that no adaptation is carried out. It outputs the value one as the tank ventilation factor FTEA.

As soon as the tank ventilation valve 25 is opened, the negative pressure prevailing in the ventilation line 21 reaches into the adsorption filter 24, whereupon this purge air is sucked in through a ventilation line 26. This desorbs fuel held in the adsorption filter and guides it the suction pipe 11 to. In this phase the tank ventilation is adapted. For this purpose, the tank ventilation adaptation device 20 receives the output signal FR from the lambda controller, and it outputs the tank ventilation adaptation factor FTEA. The basic adaptation device 19 receives the value one as an input value in this tank ventilation period. As a result, the basic adaptation sizes remain unchanged and continue to be output according to their last status.

The tank vent valve 25 is not necessarily fully opened in the tank venting periods. Rather, it is usually controlled with a certain duty cycle, which is read out from a duty cycle map 27 as a function of speed n and load L. The pulse duty factors are dimensioned such that a maximum amount of air can pass through the tank ventilation valve 25. At idle, this amount is limited relatively sharply, while at full load the tank ventilation valve is opened completely. When the adsorption filter 24 is completely regenerated, the duty cycle TVH read out from the duty cycle map 27 remains unchanged. Otherwise, it is reduced with the aid of a limit control 28 depending on the value of the tank ventilation factor FTEA. The limit value control outputs a factor FTVH that takes a maximum of one. The richer the mixture from the vent line 21 into the intake manifold 11, the more the duty cycle TVH read out from the duty cycle map 27 is reduced with the help of the factor FTVH mentioned.

The switching between basic adaptation GA and tank ventilation adaptation TEA takes place with the aid of a sequence control 29.

The arrangement described so far completely agrees with an embodiment of a conventional arrangement match. The difference lies in the specific design of the sequence controller 29. In known methods and devices for alternately executing basic adaptation GA and tank ventilation adaptation TEA, the sequence controller bases fixed values on the basic adaptation period and the tank ventilation period, typically 1.5 minutes and 4 minutes . In the invention, however, the sequence control 29 varies the ratio of the tank ventilation to the basic adaptation period depending on the amount of fuel that arises during the tank ventilation.

A direct measure of the amount of fuel vapor generated in the tank ventilation is the value of the tank ventilation adaptation factor FTEA. If this value indicates a very rich tank ventilation mixture, the tank ventilation period is extended and the basic adaptation period is shortened. In the opposite case, the time periods mentioned are changed in reverse. However, it should be noted that when choosing the size FTEA as a measure of the amount of fuel generated in the tank ventilation, the basic adaptation period must not be chosen too long, since the size FTEA is not updated during this time and it is therefore unknown whether there is a lot or has little fuel accumulated in the adsorption filter 24.

Very large basic adaptation periods can, however, be selected if the pressure difference between the internal pressure of the tank 23 and the atmospheric pressure is used as a measure of the amount of fuel to be regenerated. For this purpose, a differential pressure sensor 30 is connected to the tank. Its signal is fed to the sequence controller 29. The differential pressure is an immediate indication of whether much or little fuel has evaporated and should be regenerated accordingly. The differential pressure was initially very low and therefore a long basic adaptation period was chosen, but is If an increase in the differential pressure is observed during this period, the basic adaptation can be stopped and tank ventilation can be carried out.

2 how the basic adaptation period T_GA and the tank ventilation period T_TEA can be selected depending on values of the differential pressure Dp. In a step s2.1, it is first examined whether Dp is less than a lower threshold value Dp_SWU. If this is the case, an extended basic adaptation period of 10 minutes and a customary tank ventilation period of 4 minutes are set in a step s2.2. Otherwise, it is queried in a step s2.3 whether Dp is smaller than an average threshold value Dp_SWM. If this is the case, conventional time periods are selected, as entered in step s2.4 in FIG. 2. Otherwise, it is queried in a step s2.5 whether the differential pressure Dp is below a high threshold value Dp_SWH. If this is the case, the basic adaptation period is shortened to 1 minute in a step s2.6, and the tank ventilation period is extended to 6 minutes. Otherwise, that is to say with a very high differential pressure, the tank ventilation period is extended even further in a step s2.7, namely to 15 minutes. However, the basic adaptation period remains at 1 minute. In the exemplary embodiment, this is the shortest period of time within which the basic adaptation can still be carried out expediently.

FIG. 3 illustrates a similar procedure if, instead of the differential pressure Dp, the tank ventilation adaptation factor FTEA is used as a measure for the amount of fuel to be regenerated in the tank ventilation. The differences are that in the latter case the basic adaptation period must not be extended for a reason mentioned above and that the factor mentioned increases with increasing The amount of fuel becomes smaller, while the differential pressure increases in this case. This leads to changed queries.

In a step s3.1 it is examined whether the value of FTEA is less than a lower threshold FTEA_SWU. If this is the case, the basic adaptation period is shortened to the minimum value of 1 minute in a step s3.2, and the tank ventilation period is extended to 10 minutes. Otherwise, it is queried in a step s3.3 whether the value of FTEA is below a high threshold FTEA_SWH. If this is the case, the usual time periods are set in step s3.4, which represent the initial ratio of tank ventilation to basic adaptation time period. Otherwise, the tank venting period is shortened to 3 minutes in a step s3.5, while the basic adaptation period is increased slightly to 2 minutes. A longer extension is not justifiable because the value FTEA is not updated during the basic adaptation phases and it is therefore unclear whether the amount of fuel to be regenerated has changed.

It is particularly advantageous to combine the procedures explained with reference to FIGS. 2 and 3, namely in that the ratio of tank ventilation to basic adaptation time period is actually set with the help of the exact value FTEA, but with extended basic adaptation time periods The differential pressure Dp is used to check whether the basic adaptation should be terminated due to the increasing amount of evaporated fuel.

4 illustrates how the change from basic adaptation and tank ventilation phases can be controlled. In a step s4.1, after two marks A and B have been run through (see also FIG. 5), basic adaptation is first started. In a subsequent step s4.2, a query is made as to whether Basic adaptation is currently running. Since this is the case after the start of the method, it is checked whether the basic adaptation period T_GA has already expired (step s4.3). The information on the current time period T_GA is supplied by a block bl. Shortly after the start of the method, this period of time has not yet expired, whereupon step s4.3 is followed by a step s4.8, in which a query is made as to whether the method should be ended. This is not yet the case, whereupon the processes are repeated from step s4.2. If it is determined after some time in step s4.3 that the current value of the basic adaptation time period T_GA has been reached, the basic adaptation GA is ended in a step s4.5 and the tank ventilation adaptation TEA is started. It is then checked (step s4.6) whether the current tank ventilation period T_TEA has already expired. The value of this time period is made available from a block b2. If the time has not yet expired, after passing through two marks C and D (see also FIG. 6), steps s4.8, s4.2 and s4.6 are repeated until the time period T_TEA has expired. Then the tank ventilation adaptation is ended and the basic adaptation is started again (step s4.7). After step s4.8 of querying the end of the method, the sequence described from step s4.2 may follow again.

The current values of T_GA and T_TEA, as they are read from blocks b1 and b2, are determined according to one of the methods explained with reference to FIGS. 2 and 3. For the time period T_TEA, it is indicated in brackets in block b2 that this variable can additionally be selected depending on the load. This takes into account the fact that at high loads on the adsorption filter 24 there is only a slight pressure drop between the vent line 21 and the vent line 26, so that the filter is only slightly regenerated. It is now assumed that the differential pressure sensor 29 is a constant one Differential pressure is measured. The amount of fuel vapor generated at this average differential pressure can be regenerated better at medium loads than at high ones. It is therefore advantageous to choose the ratio of the tank ventilation to the basic adaptation period not only as a function of the differential pressure Dp, but also as a function of the speed n and load L. The load condition is of lesser importance, however, if the ratio mentioned is based on the tank ventilation Adaptation factor FTEA is set. If, under higher loads, regeneration is initially insufficient, this leads to a reduction in the FTEA factor, which automatically results in an increase in the tank ventilation period.

It should be noted that there are many strategies for basic adaptation and tank ventilation adaptation. However, what is important in the method and device described above is completely independent of the respective adaptation method. The only decisive factor is that the time periods for the adaptations, however they are carried out, depend on the value of a variable which is a measure of the amount of fuel to be regenerated in the tank ventilation, and which may also depend on the load state of the adaptation-provided internal combustion engine.

FIG. 5 illustrates an exemplary embodiment of how it can be used independently or between marks A and B in the course of FIG. 4. It is examined whether full load is present (step s5.1). If this is the case, tank ventilation is carried out (step s5.2) and step s5.1 is repeated until it is found there that the queried condition is no longer fulfilled. This procedure is based on the knowledge that at full load in engines with lambda control, this is generally switched off, which is why no basic adaptation can be carried out, so it is not worthwhile. to interrupt the tank ventilation, which does not work too effectively at full load anyway.

FIG. 6 illustrates an exemplary embodiment of how it can be used independently or between marks C and D in the course of FIG. 4. It is examined (step s6.3) whether a tank system diagnosis should be carried out with the tank ventilation valve closed. Such a method is described in a parallel application. According to this, the tank ventilation valve is closed after a negative pressure builds up on the adsorption filter in order to use the time behavior of the resulting reduction in the negative pressure to re-establish the functionality of the system. The closing of the valve and the diagnosis are the subject of a step s6.2 in FIG. 6. When the valve is closed, the tank ventilation phase is ended, a basic adaptation phase is started and an enlargement factor for the next tank ventilation period is output (step s6.3). . The advantage of this measure has already been stated above. The magnification factor has the value two in the exemplary embodiment. When used together with the acquisition of a measurement variable in accordance with the sequence of FIG. 3, it makes sense to limit the maximum tank ventilation time period, as obtained by multiplication by the enlargement factor, for the reasons explained in connection with FIG. 3.

Claims (9)

1. Method for alternately carrying out phases with and without fuel tank venting during the operation of an internal combustion engine (10) with a fuel tank venting system (21, 24-26) and lambda controller (17), characterized in that the ratio of the time periods with and without fuel tank venting is selected as a function of operational data of the fuel tank venting system,

   - a variable (Dp;FTEA) being measured which is a measure of the quantity of fuel to be regenerated during the fuel tank venting,

   - and the ratio of the time periods with and without fuel tank venting being increased in favour of the time period with venting in relation to an original ratio if the quantity of fuel exceeds, in terms of the measurement variable, an upper limit (Dp_SMW; FTEA_SWU).
2. Method according to Claim 1, characterized in that the aforesaid ratio is reduced in relation to the original ratio if the quantity of fuel drops, in terms of the measurement variable, below a lower limit (DP_SUW; FTEA SWH).
3. Method according to one of Claims 1 or 2, characterized in that, when the aforesaid ratio is increased, the time period without venting (T_GA) is reduced only as far as a prescribed minimum value, and further increasing is effected by prolonging the fuel tank-venting time period (T_TEA).
4. Method according to one of Claims 1 to 3, characterized in that the pressure difference (Dp) between the fuel tank pressure and ambient pressure is used as measured variable.
5. Method according to one of Claims 1 to 4, characterized in that the fuel tank-venting adaptation factor (FTEA) is used as measured variable.
6. Method according to one of Claims 1 to 5, characterized in that, in the case of a large load, the fuel tank-venting time periods (T_TEA) are prolonged to a greater extent than in the case of a relatively small load.
7. Method according to one of the preceding Claims, characterized in that in the case of a full load without lambda control, fuel tank venting is carried out continuously with the fuel tank venting valve (25) opened entirely.
8. Method according to one of the preceding Claims, characterized in that if a fuel tank venting valve (25) is closed for diagnostic purposes during a fuel tank venting phase, a basic adaptation phase for the lambda control is started immediately and the next fuel tank venting phase is prolonged.
9. Device (29) for alternately carrying out phases with and without fuel tank venting during the operation of an internal combustion engine (10) with a fuel tank venting system (21, 24-26), characterized in that the said device is designed in such a way that it selects the ratio of the time periods with and without fuel tank venting as a function of operational data of the fuel tank venting system.

Images