DE102011001610B4 - Method for controlling magnetic coils (solenoids) - Google Patents

Abstract

Method for driving magnetic coil devices, comprising the following steps:
synchronous pulse width modulation (PWM) of an electrical current from a reference current source with a specific reference current (I _r ) through at least one current path addressed by components (T _r1 , T _{r2, ...,} T _rN1 ) from a number N _{1 of} selectable parallel current paths wherein k ₁ addressed current paths are opened synchronously for a first selectable period of time (T ₁ ) and are closed synchronously for a second selectable period of time (T ₂ ), and current paths not addressed by the N ₁ current paths N ₁ -k ₁ for the both periods (T ₁ + T ₂ ) are closed;
a. synchronous pulse width modulation of a solenoid coil current (I _s) by channeling it through at least one addressable from a number of N ₂ components (T _s1, T _{s2, ...,} T _SN2) parallel current paths, where k ₂ addressed paths at the same time for the first selectable period of time (T ₁ ) are opened synchronously and closed synchronously for the second selectable period of time (T ₂ ), and from the N ₂ Current paths not addressed N ₂ -k ₂ current paths are closed during both time periods (T ₁ + T ₂ );
b. Each path k _{1 of} the reference current is associated with at least one path k _{2 of} the coil current
c. Using the current mirroring method to multiply the reference current (I _R ) to obtain a defined coil current (I _S ), where each addressed path of k ₁ of N ₁ paths for the reference current is a multiply wider path of k ₂ of N ₂ paths for the coil current is assigned, so that these associated path pairs have the same ratio (Q) of the path widths for a given common length;
d. Measuring all voltage drops (e _ri ) along the N ₁ current paths for the reference current (I _r ) and thus along the modulating components (T _r1 , T _{r2, ...,} T _rN1 ) of the current path and forming a first average value (R) thereof and measuring all the voltage _drops (e _sj ) along the N ₂ current paths for the magnet coil and thus along the modulating components (T _s1 , T _{s2, ...,} T _sN2 ) for the solenoid _current and forming a second average value (S) therefrom,
e. Detecting a first sample (D) of the first average value (R) within at least one selected time slot (6) and detecting at least two samples (A _t1 , A _t2 , ..., A _tk ) of the second average value (S) within at least two selected time slots (3,4,5), wherein all samples are taken cyclically, but at least in two periods of the pulse width signal;
f. Use of the samples (A _t1 , A _t2 , ..., A _tk ) of the coil current as a characterizing vector for the electrical properties of the coil as a function of position and velocity of the magnet coil core and use of these samples (A _t1 , A _t2 , ... , A _tk ) for pulse width control or for the detection of faulty behavior (F / V) of the magnetic coil;
G. Use of an average value (A _m) of at least two of the samples obtained (A _t1, A _t2, ..., A _k) of the coil current paths and using the first sample value (D) of the reference current paths to a characterizing voltage ratio (E) between the paths for to determine the coil current (I _s ) on the one hand and the paths for the reference current (I _r ) on the other hand.

Description

The proposed invention relates to the technical field of electrical engineering, in particular to special features of semiconductor circuits for the control of solenoids (magnetic coils). Shown is a method to control the movements of the moving parts (armature) and to track their current position within the magnetic coils. In this invention, current mirroring and sampling of voltage drops at different times on parallel connected transistors are used to conduct the solenoid current. These transistors are used for coil current control by pulse width modulation (PWM). The invention allows improved accuracy and speed for electromechanically controlled systems. It makes it possible to detect misbehavior such as by blocking the armature (the moving part of the solenoid) within the solenoid coil.

Magnetic coil drivers are used in many applications, such as dispensing fluids, valves and magnetic switches. To control such solenoid coils, the use of PWM signals is more advantageous over that of DC. The PWM signal in this invention can be easily generated by a central processing unit (CPU). The current control is facilitated. The PWM duty cycle determines the value of the current through the solenoid coil and thus causes the movement of the armature within the coil. The maximum current through the solenoid is required at the beginning of the relative movement of the armature relative to the same, in particular during the transition to the active position. Due to the movement of the armature, the inductance of the solenoid changes, also due to the new relative position of the armature relative to the coil.

STATE OF THE ART

• US 7 740 225 B1 „Self Adjusting Solenoid Driver and Method“ schlägt ein Verfahren zum automatischen Anpassen des Treiberausganges vor, wobei die Versorgungsspannungsdrift kompensiert wird. Der Treiberschaltkreis erzeugt ein Ausgangssignal für die Magnetspule mit einer zeitabhängigen Komponente, welche eine Funktion der Versorgungsspannung ist. Zum Beispiel kann diese Komponente die Spitzenstromdauer sein. Folglich passt sich der Treiber automatisch der Versorgungsspannung an, und kann auch bei unterschiedlichen Versorgungen verwendet werden. Des Weiteren ist in diesem Patent ein Verfahren genannt, worin der Treiberschaltkreis ein Ausgangssignal als Funktion der Versorgungsspannung liefert.
• US 7 161 787 B2 „Low Power Solenoid Driver Circuit“, ein Magnetspulentreiber für geringe Leistungsaufnahme, verwendet ein Steuersignal von dessen Abhängigkeit die Magnetspule mit zwei unterschiedlichen Energieniveaus versorgt wird, ein höherer „Setz-Pegel“ und ein niedrigerer „Halte-Pegel“. Die Erzeugung dieser zwei Pegel wird durch das Steuersignal sichergestellt, sodass der „Setz-Pegel“ immer dann gilt, wenn der Spulenstrom aktiviert wird. Unerwünschte zeitliche Schwankungen des Signals werden dadurch minimiert.
• US 6 175 484 B1 „Energy Recovery Circuit Configuration for Solenoid Injector Driver Circuits“ beschreibt, dass die induzierte Energie in der Spule im Passivverhalten durch Verkoppeln mehrerer Magnetspulen miteinander, wieder genutzt wird. Dazu dienen Schalter für höhere Spannungen.
• US 4 729 056 A beschreibt einen „Solenoid Driver Control Circuit with Initial Boost Voltage“, also einen Magnetspulen-Ansteuerungsschaltkreis mit kurzen Ansprechzeiten und minimierter Verlustleistung und mit geringer Belastung für den Schaltkreis.
• US 5 703 748 A „Solenoid Driver Circuit and Method“ ermöglicht eine Verringerung der Geschwindigkeit des beweglichen Ankers. Patent US 6,061,224 zeigt einen „PWM Solenoid Driver“ also einen Treiber mit Pulsweitenmodulation. Der spart Energie, indem die Signalpulsbreite während der Übergangsphase zwischen zwei Endpositionen des Ankers der Spule reduziert wird.

Various proposals for solenoid control can be found in the literature:

• US 7 740 225 B1 "Self Adjusting Solenoid Driver and Method" proposes a method of automatically adjusting the driver output, compensating for the supply voltage drift. The driver circuit generates an output signal for the solenoid having a time dependent component which is a function of the supply voltage. For example, this component may be the peak current duration. As a result, the driver automatically adapts to the supply voltage and can also be used with different supplies. Furthermore, this patent mentions a method in which the driver circuit provides an output signal as a function of the supply voltage.
• US Pat. No. 7,161,787 B2 Low Power Solenoid Driver Circuit, a low power solenoid driver, uses a control signal whose dependency provides the solenoid with two different energy levels, a higher "set level" and a lower "hold level". The generation of these two levels is ensured by the control signal, so that the "set level" always applies when the coil current is activated. Unwanted temporal fluctuations of the signal are thereby minimized.
• US 6 175 484 B1 "Energy Recovery Circuit Configuration for Solenoid Injector Driver Circuits" describes that the induced energy in the coil is reused in the passive behavior by coupling several solenoids together. These are switches for higher voltages.
• US 4 729 056 A describes a "Solenoid Driver Control Circuit with Initial Boost Voltage", that is, a solenoid drive circuit with short response times and minimized power dissipation and with low stress on the circuit.
• US 5 703 748 A "Solenoid Driver Circuit and Method" allows a reduction in the speed of the moving armature. patent US 6,061,224 shows a "PWM Solenoid Driver" so a driver with pulse width modulation. This saves energy by reducing the signal pulse width during the transition phase between two end positions of the armature of the coil.

Compensation of drift, reduction of signal fluctuations and energy loss, as well as speed improvement are thus the different viewpoints in the known publications of the invention. These are more or less different additional requirements placed on the circuits.

In US 4 604 675 A "Fuel Injection Solenoid Driver Circuit" uses a solenoid driver for injection pumps, a storage capacitor for power generation and details to reduce thermal losses. Different ways to reduce the magnetic coil current have been described in the patent literature. If the coil armature has reached the target position, the coil current is reduced, eg in US 5 347 419 A "A Current Limiting Solenoid Drive", in US Pat. No. 6,469,885 B1 "A Power Saving Circuit for Solenoid Driver" and in US 4,890,188 A "A Solenoid Driver System".

It is important to have a precise knowledge of the position of the armature within the solenoid coil. This can save energy through power reduction. The position of the armature can be determined via the induction change. A known method for inductance determination is based on the control with a reference current over a specific period of time and a subsequent phase for determining the fall time of the current value. The differences in the recorded current consumption times serve to inform about the anchor position, since there is a dependence on the inductance. Is in US 5 053 911 A Solenoid Closure Detection disclosed.

US 5,784,245 A "Solenoid Driver and Method for Determining Solenoid" describes the use of two reference current pulses, one in the "off" state and the other in the "on" state of two reference current pulses of the solenoid. The duration of the current decay phases and the ratio of these time periods are calculated. Based on this ratio, the anchor position can be determined via calibration tables or calculation.

From the DE 195 29 433 A1 In addition, a method for monitoring a power amplifier module is known, with reference current sources with current mirror and also parallel current paths are used for synchronous current monitoring.

Fiction BACKGROUND

The aim of such inventions is to provide a method of accurately determining the current through a solenoid and detecting the exact position of the armature within it. For the coil current detection, for example, the voltage drop across a known ohmic resistance could be measured, then amplified and finally further processed.

In the invention disclosed herein, a more accurate method for determining the solenoid current is proposed. It combines the known methods of current mirroring and the detection of voltage drops on the one hand small transistors which are connected to a reference current source, and on the other hand to larger power transistors which are connected via the magnetic coil to the supply voltage. The measured voltage drops between the source (or collector) and drain (or emitter) terminals of the transistors are sampled at specific time slots. This results in a unique signature of the solenoid and this signature correlates to a particular armature position relative to the solenoid.

Furthermore, an error detection is possible. The signature of the solenoid reflects the shape of the current curve over time. This shape is roughly detected via the voltage drops across the driver transistors for the solenoids during the current "ON" phase of the PWM signal, which PWM signal is applied to the gate (or base) terminals and turns on the defined current. The signature depends on the one hand on the properties of the magnetic coil (the coil inductance, the coil resistance), and on the other hand on the anchor position, which changes these sizes.

Accurate measurement and analysis of this signature using preferably a microprocessor is the core object of the invention. The approximate detection of the current curves via sample and hold technique (sample and hold method). A value table of several samples over time is formed, which compares with stored values, e.g. from a calibration cycle.

If the measurement of a voltage drop using a known resistance used to determine the coil current, would result in addition to the power loss caused thereby undesirable heat generation. In addition, the integration of accurate resistances is not easy and requires trimming in production. Consequently, such a solution is unsuitable, especially in higher power applications.

OBJECT OF THE INVENTION

Based on the prior art, a new method has been sought to drive a magnetic coil with a movable armature, with higher speed and positioning accuracy with the lowest possible energy consumption to improve performance should contribute. Possible errors should be detected early to avoid possible damage or consequential damage in good time. The value of the current density should be kept as constant as possible, but also be adaptable to a large current amplitude range. The current position of the armature or moving part through the solenoid coil should be determined as accurately as possible. A simple control principle was sought, which allows both the PWM control with micro-process control but also allows closed-loop control. Changes in the supply, for example due to the battery condition or due to different applied supply voltages within a range, changes due to changing ambient temperatures should be considered as well as production-related fluctuations of the system properties. Of the Influence of all these parameters should be minimized.

THE INVENTION

The above objects of the invention have been achieved according to the claims set forth. For this purpose, the developed method for driving magnet coil devices shows the following elements:

Current from a reference current source is passed through at least one current path addressed by a number N ₁ components. The components serve for the simultaneous pulse width modulation of the current and are electrically connected in parallel. Each component of the subpaths selected by the driver is opened synchronously for a first selectable period of time and closed synchronously for a second selectable period of time. Unnecessary paths are closed the entire time. Simultaneously with the pulse width modulation of the reference current, the pulse width modulation of the solenoid coil current is carried out by passing this current through an equal or multiple number of addressable by further N ₂ components parallel current paths. Again, the addressed paths are opened synchronously at the same time for the first selectable period of time and closed synchronously for the second selectable period of time and unnecessary paths are closed the entire time.

Along all the parallel current paths for the reference current on the one hand and for the coil current on the other hand, the voltage drops across the components serving for the modulation are measured and averages are formed.

The average value for the reference current voltage drop is sampled and buffered within at least one selected time slot of the entire pulse period. The average value for the coil current drop is also sampled and latched within at least two selected time slots. All samples are taken cyclically, but at least in two periods of the pulse width signal.

The sample values for the coil current-related voltage drop average value at the modulating components serve as a characterizing vector for the electrical properties of the coil, as a function of the position and velocity of the magnet coil core. The values of this signature thus obtained, given by the sample vector, are used by a control device for pulse width control or for the detection of magnetic coil malfunction. Sample values for the coil current determining voltage drop are averaged and the ratio to that of the reference current path determined. Thus, reference current and coil current can be determined and more precisely controlled.

It is essential for the method that current mirroring is used for multiplying the reference current of the setting of a defined coil current. For this purpose, each addressed path for the reference current is assigned a multiply wider path for the coil current. These mutually associated path pairs each have the same ratio of the path widths to each other at a given common length.

Each reference current path is assigned at least one coil current path.

To form the samples, a known sample-and-hold technique is preferably used, wherein the voltage values of the PWM modulated paths are latched as averaged signals in at least one capacitive element by means of switching elements. In this case, control signals serve, for example, from a microcontroller to select the time slots within the PWM signal period.

For the method, it is favorable that at least two voltage patterns from two time windows are detected in succession and then evaluated in parallel.

An analog-to-digital conversion is advantageously used in order to be able to store the detected voltage patterns, which are present in analogy in the storage capacitors, in digital registers and to further process them as a result of digital processing.

Microprocessor processing is used to acquire the signature (a state-characterizing representation) by storing in registers related to each other, and to allow analysis of the acquired samples, for example, by software. In particular, the voltage relationship between the voltages on the modulating elements of the paths for the coil current and the voltage drop across the reference current path is also determined using stored weighted reference values determined either by calibration or calculation or from simulation results, preferably by digital calculation.

Via a microprocessor, the pulse width over the modulating elements is advantageously determined by control signals at their inputs. These modulating elements turn on the current paths for a certain period of time and the rest of the time. This will be the mean current value through the solenoid varies, so the current increase as well as the current drop regulated.

A selection of the involved paths is expediently selected via control signals. This selection thus serves to determine the total current density within the paths.

• 1 zeigt die zwei Hauptfunktionsblöcke nämlich das Leistungsmodul 11 und das Abtast-Modul 12. Diese werden normal mit einem Digitalblock für Evaluierungs- und Steuerungsaufgaben verbunden.
• 2 zeigt eine mögliche Realisierung des Leistungsmoduls 11 aus 1 mit Transistoren vom n-Typ um die zwei unterschiedlichen Ströme mit PWM-Signalen zu steuern.
• 3 zeigt einen möglichen Aufbau des Abtastmoduls 12 aus 1 bestehend aus den Blöcken zur Mittelwertbildung 31,32, die Abtasteinheit 33 und dem Block 34 für die Weiterverarbeitung der Abtastwerte.
• 4 zeigt eine mögliche Realisierung der Abtasteinheit 33 von 3 mit Operationsverstärker OPAMP in Spannungsfolger-Schaltung mit Eingangskondensator und Ausgangskondensatoren sowie den erforderlichen Schaltern.
• 5 ist ein Blockschaltbild für den möglichen Aufbau für den Block 34 für die Weiterverarbeitung der Abtastwerte aus 3 mit drei Blöcken: ein Block 51 zur Mittelung zweier abgetasteter Spannungsabfälle die repräsentativ für den Spulenstrom sind; ein zweiter Block 52 zur Durchführung der Signaturprüfung, um entweder eine gültige Signatur oder ein Fehlersignal zu erzeugen und ein dritter Block 53 zur Bestimmung des Verhältnisses zwischen den gemittelten, Magnetspulenstrom-basierten Spannungsabfall-Abtastwerten und dem Abtastwert aus dem Referenzstrompfad.
• 6 zeigt ein typisches Zeitdiagramm der Steuersignale 3,4,5,6 an den Eingängen der Abtastschalter, am Steueranschluss der Transistoren 1 und das Zeitdiagramm am Ausgang des gemittelten Spannungsabfalls 2a aufgrund des Spulenstromes und des Spannungsabfalls 2b aufgrund des gespiegelten Stromes zum ausgeschalteten Referenz-Strompfad.

The invention is explained in more detail by the following examples:

• 1 Namely, the two main functional blocks show the power module 11 and the sampling module 12. These are normally connected to a digital block for evaluation and control tasks.
• 2 shows a possible implementation of the power module 11 from 1 with n-type transistors to control the two different currents with PWM signals.
• 3 shows a possible construction of the scanning module 12 1 consisting of the averaging blocks 31, 32, the scanning unit 33 and the block 34 for the further processing of the samples.
• 4 shows a possible implementation of the sampling unit 33 of 3 with operational amplifier OPAMP in voltage follower circuit with input capacitor and output capacitors as well as the required switches.
• 5 is a block diagram of the possible structure for the block 34 for the further processing of the samples from 3 with three blocks: a block 51 for averaging two sampled voltage drops representative of the coil current; a second block 52 for performing the signature check to generate either a valid signature or an error signal and a third block 53 for determining the ratio between the averaged, solenoid current-based voltage drop samples and the sample from the reference current path.
• 6 shows a typical timing diagram of the control signals 3,4,5,6 at the inputs of the sampling switches, at the control terminal of the transistors 1 and the timing diagram at the output of the averaged voltage drop 2a due to the coil current and the voltage drop 2b due to the mirrored current to the reference current path turned off.

The proposed invention is essentially through the blocks 11 and 12 as in 1 shown solved. A pulse width modulation signal PWM controls the two currents I _r and I _s in the power module 11 , A current is the reference current I _r which originates from a defined current source. The other current is the solenoid current I _s . This path connects the positive supply voltage across the coil and the modulating elements to the negative power ground. The outputs of the power module 11 are the voltage signals e _ri , e _sj which represent an image of the current _therealong along the modulating elements. Some of these voltage signals belong to the reference current I _r and form the voltage drops e _ri along the modulating elements, such as switches for the reference current. The remaining voltage signals belong to the magnetic coil current I _s and form the voltage _drops e _sj along the modulating elements, eg switches for the coil current.

In 2 is a possible circuit of the power module 11 shown, the n-type N ₁ transistors T _r1 , T _{r2, ...,} T _rN1 , for example, NMOS transistors used to connect the reference _current source to the reference ground terminal; and it uses N ₂ transistors T _s1 , T _{s2, ...,} T _{sN2 of the} same type to connect the solenoid _current to the ground (power ground). The N ₁ (or N ₂ ) transistors are connected in parallel with each other at their drain and source terminals. The associated control gates are connected to the signals c _r1 , c _r2 ,... C _ri ,... C _rN1 for the reference _{current side} and to the signals c _s1 , c _s2 ,... C _si , _{sN2 is connected} for the power current side of the coil supply. It is important that each reference transistor T _r1 , T _r2,..., T _{rN1 has} a corresponding power transistor T _s1 , T _s2,..., T _sN2 for the coil current, which has a multiple Q in its dimensioning M _si = i * Q = M _ri * Q, where M _ri = i and i = 1,2,3, ...

M usually represents a certain ratio of width to length of the transistor channel, and the smallest transistor has the value M = 1. This allows the same signal at port s _ri and at port s _{si to mirror} the current of one path to the other path, the amplitude of the mirrored current being proportional to the ratio of transistor dimensions M _si / M _ri = Q.

The drain terminals of a portion of the transistors T _r1 , T _{r2, ...,} T _rN1 are connected to reference ground GND _r and the associated source terminals are connected to the reference _current source I _r . At each modulating transistor T _r1 , T _{r2, ...,} T _rN1 , a voltage drop e _ri between drain and source is detected by means of small switches sw _{r1 +} , sw _r1- , sw _{r2 +} , sw _r2- , ..., sw _{rN1 +} , sw _N1- . Tapped.

Similarly, the voltage _drops e _sj at the transistors T _s1 , T _{s2, ...,} T _sN2 via the switch sw _{s1 +} , sw _s1- , sw _{s2 +} , sw _s2- , ..., sw _{sN2 +} , sw _sN2 tapped. These transistors conduct the Coil current I _s . The transistors connect the coil to ground GND _p . The modulating signals c _r1 , c _r2 , ... c _ri , ... c _rN1 and c _s1 , c _s2 , ... c _si , ... c _sN2 determine which transistors are used and with the pulse width modulation signal on whose control input is acted upon.

These output signals e _ri , e _{sj of} the power _module 11 are used as input signals to the scanning module 12 guided. In this scanning module 12 Further steps of the invented method are performed. These are in 3 shown. First, the voltage drops e _ri , e _{sj of} the power module in blocks 31, 32 are averaged. In the block 31 the signal S is formed, which maps the time-dependent voltage signal between the drain and source terminals of the coil driving transistors and in the block 32 R is the time-dependent voltage signal between the drain and source terminals of the transistors for modulation of the reference current is formed. In the scanning unit 33 the two signals S, R are sampled in certain time windows. One possible scenario of these signals is in 6 shown using the pulses 3 . 4 . 5 and 6 , The scanning unit 33 in 3 shows the associated input _terminals c _t1 , c _t2 , ..., c _{tk of} the pulses for sampling the S signal and c _tr for the R signal. Advantageously, it is the signals 3,4,5 during PWM in "ON" status and pulses 6 while PWM is timing in the "OFF" state.

The result is the output signals of the scanning unit 33 , as in 3 shown, the signals A _t1 , A _t2 , ..., A _{tk of} the sampled amplitudes of the input signal S at k different times and the additional signal D representing the sampled reference signal R.

In 6 the sampling times are shown in relation to the PWM signal.

One possible implementation of the sampling unit is in 4 which shows an operational amplifier OPAMP in voltage follower circuit with a gain of 1. A capacitor C _s is charged to the value of the average voltage drops S or R at the defined times (during sampling windows 3, 4, 5 and 6). This value is also stored in the capacitors C ₁ , C ₂ ,..., C _k and C _D by means of the switches s _t1 , s _t2 ,..., S _tk and s _tr and the OPAMP during the same times.

This results in the analog values A _t2 , ..., A _tk and D in the capacitors, which then in the in the further processing unit 34 in Fig.3 are provided for further processes. This unit evaluates the validity of the amplitude vector, for example by comparison with reference values. Then either an error signal F or the signature value valid signal is sent to the control unit 35 which may be a central processing unit CPU.

The signal processing of the CPU includes the control of the pulse width modulation signals for the control inputs c _r1 , c _r2 , ... c _ri , ... c _rN1 and c _s1 , c _s2 , ... c _si , ... c _sN2 . A second signal of the processing unit 34 is the ratio E between a value of the sampled values A _t1 , A _t2 ,..., A _tk related to all or some coil current, of the voltage _drops in the form of an average value A _m and the signal D (E = A _m / D). The modules of the finishing unit 34 out 3 are in 5 shown.

1. 1) eine Referenzstromquelle für den Referenzstrom,
2. 2) mindestens zwei parallele Transistor-Kanäle für den Referenzstrom mit Steuer-Gate- oder Basisanschlüssen zur Deaktivierung oder Ansteuerung durch Pulsbreitenmodulation, wobei die Transistorkanäle unterschiedliches Breite zu Längeverhältnis aufweisen,
3. 3) zumindest ebenso viele Transistorkanäle zur Kontrolle des Magnetspulenstroms, wobei jeder Transistorkanal einem der parallelen Transistorkanäle für den Referenzstrom zugeordnet ist. Es können auch mehrere Magnetstrom-Transistorkanäle einem Referenzstrom-Transistorkanal zugeordnet sein. Die Zuordnung erfolgt dadurch, dass die Gate- oder Basisanschlüsse mit demselben PWM Ansteuer(oder Deaktivierungs-)signal verbunden sind. Das Breite zu Längeverhältnis der Magnetstromtransistorenkanäle ist ein konstantes Vielfaches des Breite zu Längenverhältnisses der Referenztransistorkanäle,
4. 4) Spannungsabgreifstrukturen an jedem Transistor für den Referenzstromanteil und an jedem Transistor für die Spulenstromanteile,
5. 5) Vorrichtungen zur Mittelung der Spannungswerte an den Transistoren für den Referenzstromanteil und der Spannungswerte an den Transistoren für die Spulenstromanteile,
6. 6) Sample- und Hold-Strukturen vorzugsweise mit Operationsverstärker, Speicherkondensatoren (C₁,C₂, .., C_k,C_D) und Analogschaltern zur Erfassung der Signatur, welche eine Anzahl zeitliche Abfolge von Abtastwerten während einer PWM-Periode darstellt,
7. 7) einen Zeitgeber und Auswahlblock. Dieser steuert die Gate- bzw. Basisanschlüsse der Transistoren jedes Strompfades für den Referenzstrom oder den Magnetspulenstrom derart, dass der zugehörige Transistorkanal hochohmig gesperrt ist oder über Pulsbreitensignale (PWM) geöffnet und geschlossen wird. Die Auswahl entscheidet die Höhe der Stromdichte erlaubt dadurch den gewünschten Mittelwert des Magnetspulenstromes mehr oder weniger rasch zu erreichen, und um die Zeitschlitze für die Erfassung der Samples mithilfe der Analogschalter über Steuerleitungen (c_t1,c_t2,...c_tk,c_tr) zu bestimmen,
8. 8) einen Evaluierungsblock. Dieser erlaubt es, aus den erfassten Abtastwerte (A_t1, A_t2,...,A_tk) des Spulenstromes - als Signatur für die aktuelle Position des Magnetspulen-Ankers bei einer gegebenen Geschwindigkeit - gegenüber gespeicherten Referenz-Signaturen,
9. 9) einen Rechenblock (53) für das Verhältnis (E) zwischen Abtastwerten (A_t1, A_t2,...,A_tk) der Spannungsabfälle (e_sj) bei zumindest einigen Transistoren für die Spulenstrompfade, wobei die Spannungsabfälle durch eine Vorrichtung (51) gemittelt werden, einerseits und mit zumindest einem Abtastwert (D) des Spannungsabfalls der Referenzstrompfade andererseits, und 10) einen Mikroprozessor.

A possible device for carrying out the method according to the invention has the following components:

1. 1) a reference current source for the reference current,
2. 2) at least two parallel transistor channels for the reference current with control gate or base terminals for deactivation or control by pulse width modulation, wherein the transistor channels have different width to length ratio,
3. 3) at least as many transistor channels for controlling the solenoid current, each transistor channel being associated with one of the parallel transistor channels for the reference current. It is also possible to assign a plurality of magnetic current transistor channels to a reference current transistor channel. The assignment is made by connecting the gate or base terminals to the same PWM drive (or disable) signal. The width to length ratio of the magnetic current transistor channels is a constant multiple of the width to length ratio of the reference transistor channels,
4. 4) voltage tap patterns on each transistor for the reference current component and on each transistor for the coil current components,
5. 5) means for averaging the voltage values at the transistors for the reference current component and the voltage values at the transistors for the coil current components,
6. 6) sample and hold structures, preferably with operational amplifiers, storage capacitors (C ₁ , C ₂ , .., C _k , C _D ) and analog registers for detecting the signature, which represents a number of time series of samples during a PWM period,
7. 7) a timer and selection block. This controls the gate or base terminals of the transistors of each current path for the reference current or the magnetic coil current such that the associated transistor channel is high-impedance locked or opened and closed via pulse width signals (PWM). The choice decides the amount of current density allowed to reach the desired mean value of the solenoid current more or less quickly, and the time slots for the detection of the samples using the analog switches via control lines (c _t1 , c _t2 , ... c _tk , c _tr ) to determine
8. 8) an evaluation block. This allows, from the acquired samples (A _t1 , A _t2 , ..., A _tk ) of the coil current - as a signature for the current position of the solenoid coil at a given speed - against stored reference signatures,
9. 9) a computational block ( 53 ) for the ratio (E) between samples (A _t1 , A _t2 , ..., A _tk ) of the voltage _drops (e _sj ) in at least some transistors for the coil current paths, the voltage drops being caused by a device ( 51 ), on the one hand and with at least one sample (D) of the voltage drop of the reference current paths on the other hand, and 10) a microprocessor.

Claims (6)

A method of driving magnetic coil devices, comprising the steps of: synchronous pulse width modulation (PWM) of an electrical current flowing from a reference current source having a particular reference current (I _r ) through at least one of components (T _r1 , T _{r2, ...,} T _rN1 ) addressed current path of a number N ₁ selectable parallel current paths, k ₁ addressed current paths for a first selectable period (T ₁ ) are opened synchronously and closed synchronously for a second selectable period of time (T ₂ ), and from the N ₁ current paths N ₁ -k ₁ unaddressed current paths for the two time periods (T ₁ + T ₂ ) are closed; a. Synchronous pulse width modulation of a magnetic coil current (I _s ) by channeling it through at least one of a number of N ₂ via components (T _s1 , T _{s2, ...,} T _sN2 ) addressable parallel current paths, k ₂ addressed paths at the same time for first selectable period of time (T ₁₎ are synchronously opened and closed synchronously (T ₂₎ for the second selectable period of time, and from the N ₂ flow paths unaddressed N ₂ -k ₂ flow paths during two time periods (T ₁ + T ₂₎ are closed ; b. Each path k _{1 of} the reference current is assigned at least one path k _{2 of} the coil current c. Using the current mirroring method to multiply the reference current (I _R ) to obtain a defined coil current (I _S ), where each addressed path of k ₁ of N ₁ paths for the reference current is a multiply wider path of k ₂ of N ₂ paths for the coil current is assigned, so that these associated path pairs have the same ratio (Q) of the path widths for a given common length; d. Measuring all voltage drops (e _ri ) along the N ₁ current paths for the reference current (I _r ) and thus along the modulating components (T _r1 , T _{r2, ...,} T _rN1 ) of the current path and forming a first average value (R) thereof and measuring all the voltage _drops (e _sj ) along the N ₂ current paths for the magnet coil and thus along the modulating components (T _s1 , T _{s2, ...,} T _sN2 ) for the solenoid _current and forming a second average value (S) therefrom, e. Detecting a first sample (D) of the first average value (R) within at least one selected time slot (6) and detecting at least two samples (A _t1 , A _t2 , ..., A _tk ) of the second average value (S) within at least two selected time slots (3,4,5), wherein all samples are taken cyclically, but at least in two periods of the pulse width signal; f. Use of the samples (A _t1 , A _t2 , ..., A _tk ) of the coil current as a characterizing vector for the electrical properties of the coil as a function of position and velocity of the magnet coil core and use of these samples (A _t1 , A _t2 , ... , A _tk ) for pulse width control or for the detection of faulty behavior (F / V) of the magnetic coil; G. Using an average value (A _m ) of at least two of the sampled values (A _t1 , A _t2 ,..., A _tk ) of the coil current paths and using the first sample value (D) of the reference current paths by a characterizing voltage ratio (E) between the paths for to determine the coil current (I _s ) on the one hand and the paths for the reference current (I _r ) on the other hand. Method according to Claim 1 , characterized in that a sample-and-hold technique is used to detect voltage drops (e _ri , e _sj ) of the pulse width modulated paths as averaged signals (R, S) in at least one capacitive element (C _s ; C ₁ , C ₂ , ..., C _k , C _D ) are buffered via switching elements (s _t1 , s _t2 , ..., s _tk and s _tr ) and their control signals (c _t1 , c _t2 , ..., c _tk , and c _tr ) for selected time slots. Method according to Claim 1 or 2 , characterized in that at least two voltage drops from two time slots are detected in succession and then evaluated in parallel. Method according to one of Claims 2 or 3 , characterized in that analog-to-digital conversion is used in order to store the detected voltage drops in digital registers and further processing as a result of digital. Method according to one of Claims 1 to 4, characterized in that microprocessor processing is used for signature formation and analysis of the acquired samples (A _t1 , A _t2 , ..., A _tk ) and by forming the voltage ratio between the voltage _drops (e _sj ) of the paths for the coil current and the voltage drops (e _ri ) of the reference current paths also using stored weighted reference values, which are determined either by calibration or calculation or from simulation results. Method according to Claim 5 , characterized in that a microprocessor for generating the control signals (c _r1 , c _r2 , ... c _ri , ... c _rN1 , c _s1 , c _s2 , ... c _si , ... c _sN2 ) is used ,

Images