DE102012214717A1 - Method for controlling power or voltage of load e.g. electric heater of diesel engine to monitor exhaust gas composition, involves changing duty factor of pulse width modulation such that actual voltage corresponds to reference voltage - Google Patents

Abstract

The method involves forming period of pulse width modulation (PWM) from a switching on region, a switching off region and two edge regions. Potential gradient over an electrical load e.g. electric heater (14), is determined such that actual effective voltage over the electrical load is computed from the potential gradient. The actual effective voltage is compared with preset effective reference voltage. An adjustable duty factor of the PWM is changed such that the actual effective voltage corresponds to the preset effective reference voltage. An independent claim is also included for a device for controlling power or voltage of an electrical load.

Description

State of the art

The invention relates to a method for power control or voltage control of an electrical load by means of a power output stage, wherein the power control or voltage control via a pulse width modulation (PWM) with an adjustable duty cycle and wherein a period of the pulse width modulation of a switch-on , a turn-off and two edge regions is formed.

The invention further relates to a device for power control or voltage control of an electrical consumer by means of a power output stage, wherein the power control via a pulse-width modulation (PWM) with an adjustable duty cycle and wherein a period of the pulse width modulation by a switch-on, a Ausschaltbereich and two edge regions is formed.

To set a power, an effective voltage or an effective current, it is known to drive electrical consumers via a pulse-width modulation (PWM). In this case, for example, a constant supply voltage of the electrical load in a predetermined duty cycle (TV) on and off. The ratio of the switch-on to switch-off time can be used to adjust the electrical power supplied to the load, the effective voltage and the effective current.

The switching process can be carried out via a correspondingly controlled power output stage. The duty cycle can be specified regulated or controlled.

The adjustment of the duty cycle takes place in controlled systems, assuming an ideal behavior of the electronic components used and thus the time course of the voltage supplied to the consumer. Deviations from this ideal behavior, for example due to a limited slope or delay times, have a direct effect on the power supplied to the load or the effective current and the effective voltage. The deviations can be justified in the, possibly by component tolerances different, response of the electronic components used. The edge steepness can additionally be limited by requirements in terms of electromagnetic compatibility (EMC).

Deviations from the ideal behavior have a particularly strong effect on small duty cycles of pulse-width modulation. Small duty cycles may arise, for example, in that electrical consumers are operated with different supply voltages. That's how it describes DE 10 2010 001 004 A1 a method for controlling actuators within a vehicle electrical system, which has different operating voltages or temporal electrical system voltage changes. It is provided that the actuator or actuators are controlled with different pulse width modulated drive signals, wherein the pulse width and the period of the drive signals are independently adjustable and adjusted depending on the currently applied vehicle electrical system voltage.

Exhaust gas sensors, for example, lambda sensors, such as are used today in modern internal combustion engines for monitoring and controlling the exhaust gas composition, often have an electric heater for setting a predetermined operating temperature of the exhaust gas probe. The heating power is set by a pulse-width modulation, which is designed for example to a 12V supply voltage. A possible use of the exhaust probe in a 24V system leads to very small duty cycles. Deviations from the ideal behavior of pulse-width modulation are compensated for in regulated operation from a certain cell temperature by a temperature controller. In controlled operation, for example, while a measure of the exhaust gas probe for operating a control is not available, the described deviations from the ideal behavior to strong temperature deviations with a correspondingly high risk can lead to failures due to a malfunction.

In order to nevertheless enable a controlled operation, fast, closely tolerated power output stages can be used for the control. However, this leads to an inadmissibly high EMC radiation.

Another possibility is to use a DC / DC converter instead of pulse width modulation. This leads to high costs. Additional disadvantages are a large space requirement and the high heat loss occurring during the process.

Also high costs and a high space requirement with simultaneous high hardware expenditure arise in the measurement and setting of a suitable average current value.

It is therefore an object of the invention to provide a method which an accurate and cost-effective power control of an electrical load via pulse-width modulation even at low duty cycle allows.

It is a further object of the invention to provide a corresponding device.

Disclosure of the invention

The object of the invention relating to the method is achieved by determining a voltage profile across the electrical load, calculating an actual effective voltage across the electrical load from the voltage profile, comparing the actual effective voltage with a predetermined effective setpoint voltage, and the duty cycle of the pulse-width modulation is changed such that the actual effective voltage of the predetermined effective target voltage corresponds.

The voltage profile is detected so high that it also describes the actual time course of the voltage in the edge regions of the pulses.

Due to the actual time profile of the voltage across the electrical, in particular ohmic consumer, the electrical power actually supplied to the load is determined in a known load resistance of the electrical load and described by the actual effective value of the voltage calculated therefrom. By comparing the actual effective voltage with the effective target voltage specified for achieving the desired power, a power deviation, as caused by a non-ideal behavior of the pulse-width modulation, can be detected and adjusted by a corresponding adjustment of the pulse duty factor. Modulation be compensated.

Even if the load resistance, for example, by an operation of the electrical load at different temperatures and a corresponding temperature coefficient of the electrical load, is not known, so leads here a direct comparison of the actual effective voltage with the predetermined effective target voltage and a corresponding adjustment of the duty ratio of Pulse-width modulation to a significantly improved control of the power input in the electrical load.

The adaptation of the duty cycle allows accurate power control or voltage control in electrical consumers, especially for very small duty cycles. Cost-effective power amplifiers can be used without any special demands on the component tolerances. Limited allowable edge slopes can be compensated and thus meet the requirements of the EMC. The method is inexpensive and can be implemented with a small amount of hardware and software while having a small footprint on a suitable electronic circuit. The method is applicable regardless of the type of amplifier used (discrete, integrated, high-side, low-side). Additional functions, for example to diagnose the power output stages used, can be implemented by means of a simple software extension.

The voltage profile can be determined at least over one period of the pulse-width modulation or via at least one pulse of the pulse-width modulation or over at least one edge of the pulse-width modulation pulse. As a result, depending on the required accuracy of the power control, the necessary measurement and calculation effort can be reduced. For example, for certain applications, a possible current flow in the off state can be neglected, so that no determination of the voltage curve is necessary here and only the switch-off time is included in the calculation of the actual effective voltage. Component scatters (slew rate rise, slew rate case, delays, etc.) have an effect in particular in the flank regions of the pulses and can be determined and corrected by a corresponding measurement of the voltage profile in the flank regions.

A simple and cost-effective determination of the voltage profile over the electrical load can take place in that the voltage profile is determined from voltage values determined with an analog-to-digital converter (ADC) at predetermined sampling times, and that the sampling times of the analog-to-digital converter match the expected voltage curve be adjusted.

The specification of the sampling times ensures a sufficiently high resolution in the determination of the voltage profile, in particular in the areas necessary for the determination of the effective actual voltage. It can therefore be provided that an increased sampling rate is provided in the edge regions of the pulse-width modulation pulse and that a reduced sampling rate is provided in the switch-off region and in the switch-on region of the pulse-width modulation. From a few measurements in the switch-off and in the switch-on can be usually concluded on the rest of these areas. As a result, the scope of the measurement points to be detected can be severely restricted without limitation with regard to the pulse duration, and the method can be implemented efficiently. Can the current flow in the turn-off range neglected If so, the sample rate in this range can be set to zero.

According to a preferred variant of the method it can be provided that the voltage profile is determined during each period or during a predetermined selection of periods of the pulse-width modulation. At typical PWM frequencies and drives, it is usually possible to deduce many pulses from a single pulse. Fluctuations of the control take place comparatively slowly, as a result of which the evaluation of a reduced number of pulses is sufficient. The frequency of the pulses to be evaluated depends on the required accuracy in the power control. Due to a reduced number of periods to be evaluated, the requirement for the computing power of a CPU used can be significantly reduced.

The determination of the effective actual voltage across the load can be carried out by squaring the voltage values across the electrical load at the respective sampling instants, by relating and sizing the squares of the voltage values to the current sampling rate and the period duration of the pulse width modulation, and that the actual effective voltage is determined from the root of the sum thus obtained.

Alternatively, it can be provided that a variable sampling rate is used, that the voltage values across the electrical load are squared at the respective sampling times, that the sum is formed from the products of the squares of the voltage values with the quotients of the time differences between every two sampling times and the period of the pulse-width modulation and that the actual effective voltage is determined from the root of the sum thus obtained. It is advantageous here that a sampling rate that is variable, for example, over a pulse of the pulse-width modulation is taken into account in the determination of the voltage values.

A simple correction of the duty cycle can take place in that a factor is formed from the ratio between the actual effective voltage and the predetermined effective nominal voltage and that the duty cycle of the pulse width modulation is corrected by the factor or that from the deviation of the actual effective Voltage and the predetermined effective target voltage offset is formed and that the duty cycle of the pulse width modulation is corrected with the offset, each considered alone or in combination of the method. It can be provided that the correction is iterative over several consecutive pulses. Before a first correction, the factor is preferably set to a value of 1, the offset to a value of 0. The permissible range of the factor can be limited depending on the power output stage used.

According to a particularly preferred embodiment variant of the invention, provision may be made for determining a voltage value across the electrical load a reference voltage across the power output stage is determined in a turn-off and that the voltage value of the difference of the reference voltage and during a flank region or a switch-on above Power output stage specific measurement voltage is determined. In a series connection of the electrical load and the power output stage, in which the power output stage is grounded on one side, the reference voltage corresponds to the supply voltage. The difference between the reference voltage and the voltage currently present across the power output stage corresponds to the voltage drop across the electrical load in the respective measuring point. Changes in the supply voltage can thus be taken into account when determining the voltage profile across the electrical load. In the case of a single-ended power amplifier, the evaluation takes place in an analogous manner.

If the supply voltage exceeds the measuring range of an analog-to-digital converter used, it can be provided that the voltage across the power output stage is determined with the aid of a voltage divider to adapt to the measuring range of the analog-to-digital converter.

The object of the invention relating to the device is achieved in that the device contains means for determining a voltage profile across the electrical load, that the device contains a program sequence for calculating an actual effective voltage across the electrical load from the voltage curve, that the device is a comparison stage for comparing the actual effective voltage with a predetermined effective target voltage and that the device is adapted to change the duty cycle of the pulse-width modulation such that the actual effective voltage of the predetermined effective target voltage corresponds. The device enables the implementation of the method described.

An easily implemented power control or voltage control can be achieved by designing the electrical load as an ohmic load.

The method and the device can preferably be used for power control or voltage control of an electrical heater of an exhaust gas probe in the exhaust gas passage of an internal combustion engine.

The invention will be explained in more detail below with reference to an embodiment shown in FIGS. It shows:

1 a first electrical circuit for controlling the power of an electrical load,

2 a voltage curve output stage,

3 a voltage curve heater,

4 a controller with a voltage divider.

5 a second electrical circuit for power control of an electrical

consumer

1 shows a first electrical circuit for power control of an electrical load. The embodiment relates to the power control of an electric heater 14 a lambda probe, which is arranged in the exhaust passage of a diesel engine.

The electric heater 14 is with a voltage source 10 and via a control unit 12 with mass 17 connected. Line resistance of the leads to the electric heater 14 are in the illustration in a first lead resistance 13 summarized. The control unit 12 is executed as Engine Control Unit (ECU). Again, the line resistance of the leads are in a second supply resistance 11 summarized. The control unit 12 is via the second supply resistance 11 and over ground with the voltage source 10 connected. Between the electric heater 14 and the controller 12 is a first measuring connection 15 intended. To losses over the first supply resistance 13 Alternatively, an additional, not shown, measuring connection between the first supply line resistance can be compensated 13 and the heater 14 be provided.

The representation of the control unit 12 is limited to the components essential to the description of the invention. It contains a power output stage 16 in the form of a field effect transistor (FET) with a drain connection 16.2 , a gate connection 16.1 , a bulk connection 16.3 and a source port 16.4 , The bulk connection 16.3 and the source port 16.4 are with mass 17 , the drain connection 16.2 with the electric heater 14 connected.

To control the electric heater 14 supplied electrical power is applied to the gate terminal 16.1 a pulse width modulated signal (PWM signal) applied. During the turn-on of the PWM signal, the field effect transistor turns on and a current flows from the voltage source 10 through the electric heater 14 and the power output stage 16 to earth 17 , During the off times the current flow is interrupted. About the duty cycle of the PWM signal is in the electric heater 14 converted performance specified. According to the illustrated construction of the first electrical circuit is the power amplifier 16 as a lowside switch to ground 17 executed.

2 shows a voltage curve power amplifier 21 as it does during a pulse-width modulation pulse between the first measurement port 15 and mass 17 is present. The voltage curve power amplifier 21 is opposite to a voltage axis 20 and a timeline 22 applied.

3 shows a voltage curve heater 23 , The voltage curve heater 23 is opposite to in 2 introduced voltage axis 20 and the timeline 22 applied. One from the voltage source 10 provided supply voltage 24 is marked as a dotted line. The voltage curve heater 23 put that over the heater 14 It is formed from the difference of the supply voltage 24 and of in 2 shown voltage curve output stage 21 ,

Due to a non-ideal behavior of the driving components, for example due to limited edge steepness or delay times, the voltage curve heater deviates 23 from an ideal voltage waveform with rectangular pulses. Due to component variations, this deviation can be used for different output stages 16 be different. The slope is further limited by requirements of electromagnetic compatibility (EMC). Due to the internal resistance of the power output stage 16 when switched on, it is not the full supply voltage 24 over the electric heater 14 at.

Deviations from the ideal voltage curve lead directly to a deviation in the electric heater 14 converted electrical power from a predetermined setpoint. This has a particularly strong effect on small pulse width modulation pulse widths.

For the illustrated embodiment, it is provided that the lambda probe depending on present on-board voltage at supply voltages 24 of nominal system voltages of 12V or 24V. When operating at 24V, the duty cycle is ¼ of the value compared to 12V operation. Especially in the area of the so-called protective heating of the lambda probe with very small, controlled set performances, this results in very small duty cycles in the range of typically <1%. The switching edges of the heating signal here have a great influence on the actual in the electric heater 14 implemented performance. The steepness of the switching edges is limited by the components used, but also by the increased at 24V electromagnetic radiation. A deviation of the actually converted heating power from the specification leads to a high risk of field failures due to malfunction.

According to the invention, it is therefore provided, the actual voltage curve over the electrical load, in the present embodiment, the heater 14 to determine, to compare with a predetermined voltage waveform and adjust the duty cycle of the pulse width modulation in case of a deviation. For comparison, the actual voltage curve becomes a heater 23 determines an effective voltage, which is compared with a predetermined effective target voltage.

• 1. Bestimmung der Messspannung U_mess_i über der Leistungsendstufe 16 bei gesperrtem FET über eine vorgegebene Anzahl an Messpunkten.
• 2. Berechnung des Mittelwertes über die bestimmten Messwerte U_mess_i als Referenzspannung U_ref. Die Referenzspannung U_ref entspricht der Versorgungsspannung 24.
• 3. Messung des Spannungsverlaufs Endstufe 21 zu vorgegebenen Abtastzeitpunkten (U_mess_i). Anschließend für jeden Abtastzeitpunkt: – Berechnung des Spannungsabfalls U_Heizer_i über den Heizer 14 als Differenz zwischen der Referenzspannung und der jeweiligen Spannung U_mess: (U_Heizer_i = U_ref – U_mess_i). – Bildung des Quadrats von U_Heizer_i.
• 4. Bildung der Summe der Quadrate von U_Heizer_i über die Abtastzeitpunkte.
• 5. Bezug auf eine Periode der Puls-Weiten-Modulation: – Signal_1 = Summe ((U_Heizer_i)²/(Periodendauer PWM × f_sample)), wobei f_sample die Abtastrate darstellt.
• 6. Bestimmung der effektiven Spannung U_heff über den Heizer 14 als Wurzel aus dem Signal_1. Die effektive Spannung charakterisiert die aktuellen Eigenschaften des Spannungsverlaufs Heizer 23 inklusive von Flankentoleranzen, Verzögerungen, Strombegrenzungen und des Innenwiderstands der Leistungsendstufe 16.
• 7. Vergleich mit einer vorgegebenen effektiven Sollspannung U_hsoll durch Bildung eines Faktors F: F = U_heff/U_hsoll.
• 8. Korrektur des Tastverhältnisses der Puls-Weiten-Modulation auf Basis des Faktors F.

According to the in the 1 the circuit arrangement shown in the embodiment of the actual voltage waveform heater 23 from the difference between the supply voltage 24 and at the first measuring port 15 over the power output stage 16 measured voltage curve output stage 21 certainly. The following describes a possible procedure and an alternative procedure for evaluation:

• 1. Determination of the measuring voltage U_mess_i over the power output stage 16 with locked FET over a predetermined number of measuring points.
• 2. Calculation of the mean value over the determined measured values U_mess_i as reference voltage U_ref. The reference voltage U_ref corresponds to the supply voltage 24 ,
• 3. Measurement of the voltage curve of the power amplifier 21 at predetermined sampling times (U_mess_i). Then for each sampling time: - Calculation of the voltage drop U_Heizer_i via the heater 14 as the difference between the reference voltage and the respective voltage U_mess: (U_Heizer_i = U_ref - U_mess_i). - Formation of the square of U_Heizer_i.
• 4. Forming the sum of the squares of U_Heizer_i over the sampling times.
• 5. Referring to a period of pulse width modulation: - Signal_1 = Sum ((U_Heizer_i) ² / (Period PWM × f_sample)), where f_sample represents the sample rate.
• 6. Determination of the effective voltage U_heff across the heater 14 as the root of Signal_1. The effective voltage characterizes the current characteristics of the voltage curve heater 23 including edge tolerances, delays, current limits and the internal resistance of the power output stage 16 ,
• 7. Comparison with a given effective target voltage U_hsoll by forming a factor F: F = U_heff / U_hsoll.
• 8. Correction of the duty cycle of the pulse-width modulation on the basis of the factor F.

• 4a. Berechnung der quadrierten Werte von U_Heizer
• 5a. Berechnung des Summenwerts mit Berücksichtigung der Abtastung des ADC Signal_1 = Summe((U_heizer_i)²·Δt_ADC_i/Periodendauer PWM), wobei Δt_ADC_i die Dauer des jeweiligen Signalabschnitts beschreibt.

Alternatively, instead of items 4 and 5, the following procedure may be used. The alternative procedure shows the advantage that variable sampling rates over the pulse are included:

• 4a. Calculation of the squared values of U_Heizer
• 5a. Calculation of the sum value taking into account the sampling of the ADC Signal_1 = sum ((U_heizer_i) ² · Δt_ADC_i / period duration PWM), wherein Δt_ADC_i describes the duration of the respective signal section.

4 shows that in 1 introduced control unit 12 with an additional voltage divider over the power output stage 16 , The voltage divider is made up of a first resistor 30 and a second resistor 32 built up. A second measuring connection 31 is between the resistances 30 . 32 arranged.

The voltage curve power amplifier 21 and thus the voltage curve heater 23 can be measured preferably with an analog-to-digital converter (ADC), not shown. The voltage divider allows the adaptation of the measuring voltage to the measuring range of the analogue-to-digital converter. Such an analog-to-digital converter is in modern, as engine control unit (ECU) running ECUs 12 already planned. Through a simple extension of the controller 12 Around the voltage divider, this analog-to-digital converter can be used to determine the voltage waveform of the power amplifier 21 be used.

5 shows a second electrical circuit for power control of an electrical load. Here are the same, already too 1 introduced identifier used. In contrast to the first electrical circuit is the power amplifier 16 executed as a highside switch and over the first supply line resistance 13 with the voltage source 10 connected. The heater 14 is as an electrical consumer between the power output stage 16 and mass 17 arranged. At the first measuring connection 15 can the voltage curve heater 23 directly over the heater 14 to be determined to mass. Again, the use of a voltage divider is possible to adapt the measuring voltage to the measuring range of an analog-to-digital converter.

The method can be used with a small hardware outlay with little space requirement in the control unit 12 implement. The insert is not bound to exhaust probes, the method and the device can be used for many other electrical loads and regardless of the type of power output stage used 16 use. It can be used power amplifiers with higher tolerances and thereby costs can be saved. The voltage curve over the power output stage 16 can also be used to diagnose the power output stage 16 and the consumer.

To save computing power, it is advantageous to the sampling times to the expected voltage waveform amplifier 21 adapt. Can, for example, the power at locked power output stage 16 neglected, no or only a few measuring points are to be recorded here.

Component scattering is particularly relevant in the region of the edges of the pulses, so that high sampling rates make sense here.

At typical PWM frequencies and controls, it can be concluded from the evaluation of a single or fewer pulses to many pulses, so that only a predetermined proportion of the pulses must be measured and evaluated. For a typical broadband lambda probe, the pulse width modulation frequency is about 100 Hz. Here it may be sufficient to evaluate only every 100th pulse, for example, which leads to a significant reduction of the measurement and calculation effort.

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• DE 102010001004 A1 [0006]

Claims (13)

Method for power control or voltage control of an electrical consumer with the aid of a power output stage ( 16 ), wherein the power control or voltage control via a pulse width modulation (PWM) with an adjustable duty cycle and wherein a period of the pulse width modulation of a power-on, a turn-off and two edge regions is formed, characterized in that a voltage waveform is determined over the electrical load, that from the voltage curve, an actual effective voltage is calculated across the electrical load, that the actual effective voltage is compared with a predetermined effective target voltage and that the duty cycle of the pulse-width modulation is changed so that the actual effective voltage of the predetermined effective target voltage corresponds. A method according to claim 1, characterized in that the voltage profile is determined at least over a period of the pulse-width modulation or at least one pulse of the pulse-width modulation or at least one edge of the pulse width-modulation pulse. A method according to claim 1 or 2, characterized in that the voltage profile is determined with an analog-to-digital converter (ADC) at predetermined sampling times voltage values and that the sampling times of the analog-to-digital converter are adapted to the expected voltage waveform. Method according to one of Claims 1 to 3, characterized in that an increased sampling rate is provided in the edge regions of the pulse-width modulation pulse and that a reduced sampling rate is provided in the switch-off region and in the switch-on region of the pulse-width modulation. Method according to one of claims 1 to 4, characterized in that the voltage profile is determined during each period or during a predetermined selection of periods of the pulse-width modulation. Method according to one of Claims 1 to 5, characterized in that the voltage values across the electrical load are squared at the respective sampling instants, that the squares of the voltage values are related to the current sampling rate and the period duration of the pulse width modulation and that the actual effective voltage is determined from the root of the sum thus obtained. Method according to one of claims 1 to 5, characterized in that a variable sampling rate is used, that the voltage values are squared over the electrical load at the respective sampling times, that the sum is formed from the products of the squares of the voltage values with the quotients of the Time differences between each two sampling times and the period of the pulse-width modulation and that the actual effective voltage is determined from the root of the sum thus obtained. Method according to one of claims 1 to 7, characterized in that a factor of the ratio between the actual effective voltage and the predetermined effective target voltage is formed and that the duty cycle of the pulse-width modulation is corrected by the factor or that of the deviation an offset is formed between the actual effective voltage and the predetermined effective target voltage, and that the duty cycle of the pulse width modulation is corrected with the offset, taken alone or in combination of the methods. Method according to one of claims 1 to 7, characterized in that for determining a voltage value above the electrical load, a reference voltage across the power output stage ( 16 ) is determined in a switch-off region and that the voltage values are determined from the difference between the reference voltage and a voltage during a flank region or a switch-on region ( 16 ) certain measurement voltage is determined. Method according to one of claims 1 to 8, characterized in that the voltage across the power output stage ( 16 ) is determined by means of a voltage divider to adapt to the measuring range of the analog-to-digital converter. Device for power control or voltage control of an electrical consumer with the aid of a power output stage ( 16 ), wherein the power control via a pulse width modulation (PWM) with an adjustable duty cycle and wherein a period of the pulse width modulation is formed by a power-on, a Ausschaltbereich and two edge regions, characterized in that the device comprises means for Determining a voltage profile across the electrical load includes that the device includes a program flow for calculating an actual effective voltage across the electrical load from the voltage waveform, that the device includes a comparison stage for comparing the actual effective voltage with a predetermined effective target voltage and that the device to is designed to change the duty cycle of the pulse-width modulation such that the actual effective voltage of the predetermined effective target voltage corresponds. Apparatus according to claim 10, characterized in that the electrical load is designed as a resistive load. Application of the method and the device according to one of claims 1 to 11 for power control or voltage control of an electric heater ( 14 ) an exhaust gas probe in the exhaust passage of an internal combustion engine.

Images