(19)
(11)EP 2 752 672 B1

(12)EUROPEAN PATENT SPECIFICATION

(45)Mention of the grant of the patent:
13.11.2019 Bulletin 2019/46

(21)Application number: 14000004.3

(22)Date of filing:  02.01.2014
(51)International Patent Classification (IPC): 
G01R 17/10(2006.01)
G01R 35/00(2006.01)
G01R 27/16(2006.01)
G01R 27/14(2006.01)
G01R 1/20(2006.01)

(54)

Method and system for measuring the resistance of a resistive structure

Verfahren und System zum Messen des Widerstandes einer resistiven Struktur

Procédé et système permettant de mesurer la résistance d'une structure résistive


(84)Designated Contracting States:
AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

(30)Priority: 04.01.2013 US 201313734581

(43)Date of publication of application:
09.07.2014 Bulletin 2014/28

(73)Proprietor: Linear Technology Corporation
Milpitas, CA 95035-7417 (US)

(72)Inventor:
  • Engl, Bernhard Helmut
    Colorado Sprins, CO 80920 (US)

(74)Representative: Müller-Boré & Partner Patentanwälte PartG mbB 
Friedenheimer Brücke 21
80639 München
80639 München (DE)


(56)References cited: : 
EP-A1- 1 607 753
DE-A1- 10 133 525
JP-A- S60 192 270
US-A- 5 804 979
WO-A1-2006/002446
DE-A1-102011 078 334
JP-A- S62 168 067
  
  • U. Tietze, Ch. Schenk: "Halbleiter-Schaltungstechnik", 1993, Springer Verlag, Berlin, XP002723495, ISBN: 3-540-56184-6 vol. 10, pages 371-373, * figures 13.12, 13.13 *
  
Note: Within nine months from the publication of the mention of the grant of the European patent, any person may give notice to the European Patent Office of opposition to the European patent granted. Notice of opposition shall be filed in a written reasoned statement. It shall not be deemed to have been filed until the opposition fee has been paid. (Art. 99(1) European Patent Convention).


Description

TECHNICAL FIELD



[0001] This disclosure generally relates to the electrical arts, and more particularly to calibration of resistive structures.

DESCRIPTION OF RELATED ART



[0002] In recent years there has been an effort to improve methods and circuits to measure the resistance of resistive structures. Information about the resistance of a resistive structure is useful in many applications, including accurate measurement of current flow. Discrete current sense resistors suitable for accurate output current measurement typically have several requirements. For example, such current sense resistors may require sufficiently low manufacturing tolerances, long term stability, a low temperature coefficient, and/or other limitations. As the specifications become more demanding the component cost and complexity of the system may increase.

[0003] Some known approaches accommodate less demanding specifications (and thus lower costs) by providing some form of measurement and/or calibration of a current sense resistor during system operation.

[0004] Various methods for calibrating resistive structures during uninterrupted system operation exist. For example, in a first category of methods, the temperature of the resistive structure is measured and then compensated for the drift over temperature by using the measured temperature and known temperature coefficients of the resistive material. For example, U.S. Patent No. 4,591,743 to Kung falls under this category.

[0005] A second category of methods, sometimes referred to as "DCR sensing" uses the parasitic resistance of an inductor of a switch-mode power converter. For example, U.S. Patent No. 6,469,481 to Tateishi falls under the second category.

[0006] A third category of methods periodically injects predetermined test or calibration currents at an end of the resistive structure and determines its resistance via Ohm's law. For example, U.S. Patent No. 7,536,577 to Desai et al., falls under this category.

[0007] A fourth category of methods infer the resistance of a current sense resistor at an output. For example, the resistance may be inferred from the switching parameters of a DC/DC converter and a measured input current to the DC/DC converter by a precision current sense resistor at an input. For example, this approach is described by G. Eirea and S.R. Sanders, High Precision Load Current Sensing Using On-Line Calibration of Trace Resistance, IEEE Transactions on Power Electronics, pages 907 to 914 (Vol. 23, Issue 2, March 2008), ("Eirea and Sanders.")

[0008] The foregoing prior art approaches have several disadvantages. For example, the first category of methods is unable to compensate for manufacturing tolerances by simply measuring temperature. Consequently, different initial resistance measurements and temperature coefficients of the resistance may result. Typically, a different initial resistance is accommodated by a one time calibration with a known test current at the end of the manufacturing process of the system or subsystem that includes the resistive structure.

[0009] Determining the temperature coefficient(s) of the resistive structure is much more costly. For example, a known test current is applied to the resistive structure at least at two different temperatures. The test equipment expenditure and the time for determining the temperature change for the stabilization, makes this individual determination of the temperature coefficient ("tempco") of each resistive structure impractical for most applications. Typically, it is necessary to compromise the temperature compensation by using batch-by-batch temperature coefficients that were obtained from several samples from a batch. Even worse, temperature coefficients may be used that were obtained by a single lab characterization of early prototypes. The foregoing approach regularly leads to poor performance of circuits over temperature.

[0010] - Another disadvantage of the first category of methods is the inability to detect and compensate for long-term drift of the resistance. Depending on parameters (e.g., such as the material of the resistive structure, its mechanical construction, and environmental influences), resistance may change by several percent for every 1000 hours of operation. For typical current sense resistors, this long-term drift effect is more pronounced at higher temperatures. A typical remedy (aside from specifying more costly resistors that have better stability) is regular recalibration, which typically increases the cost of ownership. Further, for this category, it implies periodic suspension of system operation.

[0011] The second category of methods relates to switch-mode power supplies. One disadvantage of this category is that the gain factor of the current measurement depends on the DCR of the inductor's coil winding, which (e.g., typically being made of copper), has a tempco in the order of 0.4% per Kelvin, which may be unacceptable for many applications. This category of methods typically requires matching a time constant of a discrete first resistor and capacitor pair (e.g., R1 X C1) to an L/DCR time constant of the inductor. This inductor may be a moving target due to the high DCR temperature coefficient and the high manufacturing tolerances (e.g., in the 10 percent range) of the inductor. Further, the inductance L is also dependent on core magnetization, which varies over inductor current. Although attempts to provide temperature compensation of the DCR shift by providing means to measure inductor winding temperature (e.g., such as seen in U.S. Patent No. 7,358,710 to Shiguo et al.,) have been made, they may not be accurate. For example, in such inductor DCR current sensing approaches, the difficulty in placing the temperature sensor (in this case a negative temperature coefficient (NTC)) in a suitable location such that the coil winding temperature is accurately tracked, is a source of a substantial loss of accuracy. Furthermore, none of the remedies of the second category helps against load life drift of the DCR resistance.

[0012] The third category of methods suffers from its inability to distinguish between the injected test and/or calibration currents and load current variations, which in typical systems occur in a random manner. For example, any change in load current alters the measured voltage differential over the current sense resistance to be calibrated. Accordingly, the accuracy of the calibration may be impaired if a change in load current occurs during the calibration. For example, the magnitude of the test current is typically designed to be much smaller than the maximum load current. In this regard, the outcome of the calibration may be wrong due to load current fluctuations that leak into the calibration channel.

[0013] The immediate remedy of switching OFF the load during the calibration may not be a practical solution for most cases since it interferes with the desired uninterrupted system operation. Even if switching OFF the load during calibration may be acceptable for some applications, such a calibration approach does not accurately track resistance changes while the system is in normal running condition. This is true even if a method to measure temperature of the resistive structure is provided. For example, there is a general difficulty in accurately measuring temperatures of distributed resistive structures (such as inductor windings and PCB traces) in a system environment (in contrast to a laboratory setup).

[0014] The fourth category of methods uses a precision current sense resistor on the input side of a switch-mode power supply. It will be understood that reliance on a precision current sense resistor instead of a regular (or no resistor) introduces additional cost. Accordingly, the fourth category merely avoids the voltage drop losses incurred by adding such a sense resistor on the low voltage output side of the switch-mode power supply. The current sense resistor itself is not able to be eliminated. Furthermore, the precision of the fourth category depends on the mode of operation of the switch-mode power supply, which may vary with load current. For example, the circuit discussed in High Precision Load Current Sensing Using On-Line Calibration of Trace Resistance does not achieve proper calibration at light loads and loses, calibration capability once the converter enters discontinuous conduction mode.

[0015] JP S62 168067 A relates to a constant current source connected between terminals where a constant current is supplied from the constant current source and conducted in a network. Voltages developed between the terminals are measured, after which the constant current source is disconnected and another constant current source is connected between the terminals. Voltages again develop between the terminals which are measured. Then, measured current values of the constant current sources and the measured voltages are used to calculate resistance values.

[0016] DE 101 33 525 A1 relates to a sensor with a bridge circuit, the diagonal voltage of which is used to determine a measurement value. A switch array is provided to change the bridge circuit between measurement and test modes. The test mode changes the arrangements of bridge resistors relative to each other. When the bridge is in test mode, the diagonal voltage is insensitive to changes in the measurement value.

[0017] "Halbleiter-Schaltungstechnik" by Schenk, C. and Tietze, U. relates to general concepts of voltage-driven transistor current sources.

[0018] DE 10 2011 078 334 A1 relates to a shunt resistor comprising two calibration terminals for injecting a calibration current.

[0019] In view of the foregoing, it would be desirable to provide an improved method and circuit to measure the resistance of a resistive structure while the system operates uninterrupted. It would also be desirable to measure the resistance of the resistive structure without having to measure component temperature. It would also be desirable to track and compensate for adverse effects (e.g., such as many manufacturing tolerances, temperature drift of the resistance, load life drift of resistance due to aging (poor long term stability, etc.,)) that may alter the resistance of the resistive structure.

MEANS FOR SOLVING THE PROBLEM



[0020] The object is achieved by the system as defined in claim 1 and by the method as defined in claim 15.

BRIEF DESCRIPTION OF DRAWINGS



[0021] The drawings are of illustrative embodiments. They do not illustrate all embodiments. Other embodiments may be used in addition or instead. Details that may be apparent or unnecessary may be omitted to save space or for more effective illustration. Some embodiments may be practiced with additional components or steps and/or without all of the components or steps that are illustrated. When the same numeral appears in different drawings, it refers to the same or like components or steps.

FIG. 1 illustrates a system having a resistive structure, consistent with an embodiment of the present invention.

FIG. 2 illustrates a more detailed view of a system having a resistive structure, consistent with an embodiment of the present invention.

FIG. 3 illustrates a system having a resistive structure wherein a calibration current source is reused as a tuning current source to provide a load current ripple, consistent with an embodiment of the present invention.

FIG. 4 illustrates a system that is capable of steering current in a resistive structure, consistent with an embodiment of the present invention.

FIG. 5 illustrates a system that steers current in a resistive structure by splitting a switch, consistent with an embodiment of the present invention.

FIG. 6 illustrates part of a system that measures the resistance of a resistive structure, consistent with an embodiment of the present invention.

FIG. 7 illustrates an exemplary resistive structure, consistent with an embodiment of the present invention.

FIG. 8 illustrates an exemplary resistive structure following a pseudo Kelvin connect approach, consistent with an embodiment of the present invention.

FIG. 9 illustrates a system having a tunable interpolator and a correlator circuit to guide the interpolator tuning process, consistent with an embodiment of the present invention.

FIG. 10 illustrates a system having a tunable interpolator and a correlator circuit based on polarity reversing switches driven by switching phases of a switch-mode power supply, consistent with an embodiment of the present invention.

FIG. 11 illustrates an exemplary resistive structure following another pseudo Kelvin connect circuit, consistent with an embodiment of the present invention.


DETAILED DESCRIPTION OF ILLUSTRATIVE EMBODIMENTS



[0022] In the following detailed description, numerous specific details are set forth by way of examples in order to provide a thorough understanding of the relevant teachings. However, it should be apparent that the present teachings may be practiced without such details. In other instances, well-known methods, procedures, components, and/or circuitry have been described at a relatively high-level, without detail, in order to avoid unnecessarily obscuring aspects of the present teachings.

[0023] The various examples discussed below relate to methods and circuits to measure the resistance of a resistive structure during uninterrupted system operation. Suitable resistive structures, among others, are discrete resistors or parasitic resistances associated with the metal layers of integrated circuits, printed circuit board foil, connection wiring, sheet metal, cable runs, or any other resistances, as long as they obey Ohm's law and have sufficiently low parasitic capacitances and inductances.

[0024] In one embodiment, the calibration process discussed herein is fully transparent to the system operation and typically does not interfere with it. The influence of a load current interference on the calibration result can be designed to be very low. In one aspect, calibration at low (or zero) load current may be performed without adversely affecting the calibration accuracy. In one aspect, the energy consumption of the calibration process discussed herein may be designed to be very low.

[0025] In one embodiment, the circuit and method described herein are able to perform calibration of a range of resistive materials in different shapes, forms, and sizes. The resistance accommodated can span many orders of magnitude. For example, resistance ranges of 100µΩ to 1Ω or more are supported. In one aspect, the resistive structure does not need to be a lumped element, such as a series connection of two resistors having three electrical nodes; rather, the concepts embodied herein also apply when the resistive structure is any distributed resistance (e.g., a mesh of resistors, such as from finite element analysis). In one embodiment, parasitic resistive structures in the form of printed circuit board traces or printed circuit board planes may be used as current sense resistors. In implementations where the inherent parasitic resistances (e.g., such as printed circuit board traces or wire runs) are used as precision current sense resistors after appropriate calibration, the additional voltage drop associated with adding a true current sense resistor can be avoided, thereby saving energy.

[0026] Reference now is made in detail to the examples illustrated in the accompanying drawings and discussed below. FIG. 1 illustrates a system 100 having a resistive structure 101, consistent with an embodiment of the present invention. In one embodiment, the resistive structure 101 has a total resistance Rx between a first node 130A and a second node 130B. At least one additional node (e.g., third node 130C) is provided that electrically and/or mechanically is located between the first node 130A and the second node 130B. For example, a first partial resistance Rx1 appears between the first node 130A and the third node 130C and a second partial resistance Rx2 appears between the third node 130C and the second node 130B. A total resistance Rx is provided by the sum of partial resistances Rx1 and Rx2. The magnitude of the resistances Rx, Rx1, and Rx2 may be unknown.

[0027] The resistive structure 101 may be part of a circuit. In one embodiment, the first node 130A is connected to a power supply 110, and the second node 130B is connected to a load 111. For example, the resistive structure 101 acts as a current sense ("shunt") resistor, that facilitates the measurement of a current flowing through resistive structure 101 by measuring the voltage across the first node 130A and the second node 130B and calculating the load current by Ohm's law, as provided by Equation 1 below.

when the magnitude of Rx is known.

[0028] In one embodiment, a calibration current circuit block 102 provides current (e.g., in the form of pulses) to resistive structure 101 (Rx). Accordingly, calibration current generating circuit block 102 is used to measure the resistance of resistive structure 101 (Rx). In one example, the calibration current generating circuit 102 is coupled to at least the third node 130C in at least one state of operation. "Coupled" in this context means that the calibration current need not be injected exactly into the third node; rather, any appropriate location in the vicinity of the third node may also be used to inject the calibration current. However, the closer to the third node the better the accuracy. A distinctive advantage provided by injecting a calibration current (ICAL) to the third node 130C of the resistive structure 101 is that reactive components (e.g., such as power supply decoupling capacitors) tied to nodes 130A and/or 130B (e.g., 110 and 111) do not impair or disturb the calibration current pulses (ICAL).

[0029] System 100 may include a control circuit 103 coupled to the calibration current generating circuit block 102. Control circuit 103 has at least two states of operation to change a calibration current (ICAL) or to guide the calibration current ICAL to either one of at least two of the three nodes of the resistive structure (i.e., 130A, 130B, and 130C) alternately. For example, instead of changing the calibration current (ICAL) between two states, (which may be "calibration current ON" and "calibration current OFF," two different magnitudes of the calibration current, or two different signs of the calibration current), the calibration current may be "ON" during the calibration but alternated between node 130C and node 130A, or be alternated between node 130C and node 130B. For example, the alternation may be accomplished by electronic switching means, such as a pair of MOSFETs (not shown in FIG.1) tied between the output of the calibration current source 102 and node 130C and nodes 130A/B of the resistive structure 101, respectively. It will be understood that adding such MOSFETs (or any other circuit elements) to the calibration current path still couples the calibration current generating circuit to the third node 130C of the resistive structure 101.

[0030] System 100 may include a signal conditioning circuit block 104 that is coupled to the first node 130A, the second node 130B, and the third node 130C to form a conditioned signal at its output 105. The signal at the output 105 provides the difference (or an amplified difference) between the electrical potential at the third node (130C) of the resistive structure 101 and the weighted average of the electrical potential at the first and second nodes (130A, 130B) of the resistive structure 101.

[0031] In one embodiment, system 100 includes a signal processing circuit block 106 that is coupled to the output 105 of the conditioning circuit block 104. The signal processing circuit block 106 uses the signal at the output 105 (provided by the conditioning circuit block 104) to determine the resistance Rx of the resistive structure 101. The signal processing block 106 is discussed in more detail later.

[0032] FIG. 2 is a more detailed representation of a system 200 having a resistive structure 201, consistent with an embodiment of the present invention. For example, the resistive structure 201 may be a piece of resistive material having at least a first connection node 230A, a second connection node 230B, and a third connection node 230C. Connection node 230C is between node 230A and node 230B. In one embodiment, connection node 230C is halfway between connection nodes 230A and 230B, in the center area, such that partial resistances Rx1 = Rac and Rx2 = Rcb are of substantially identical resistive value, as provided in Equation 2 below.



[0033] Accordingly, the desired response signal at node 230C is maximized, which decreases in amplitude the further node 230C is moved away from the center of the resistive structure 201. In one embodiment, node 230C (from which the signal is taken) is positioned away from the center (e.g., at a distance where the resistance is 1/3 of the total resistance). It should be noted that when node 230C is too close to one of the nodes 230A or 230B, then the response signal may degrade.

[0034] The calibration current generating circuit block 202 may comprise a regulation transistor M1. In one embodiment, the regulation transistor M1 is a double-diffused metal-oxide-semiconductor (DMOS) transistor. The drain of the regulation transistor M1 may be tied to the third connection node 230C. In one embodiment, the drain of regulation transistor M1 is coupled to the third connection node 230C by tying the drain to one or more locations in the vicinity of the third node of the resistive structure. In one embodiment, the calibration current generating circuit block 202 includes a reference resistor Rs having a first node coupled to the source of the regulation transistor M1 and a second node coupled to Vss (or ground). Reference resistor Rs generally does not see the load current ILOAD; rather, it sees the calibration current ICAL, which may be orders of magnitude smaller than the maximum load current. Further, the calibration current ICAL may comprise short pulses and have a small duty cycle. Accordingly, resistor Rs may be much smaller in size and of lower power rating (and hence, of lower cost) than a precision shunt resistor that is able to sustain the full load current. If an optional secondary calibration step (described later) is added, reference resistor Rs may also have a large tolerance, large temperature coefficient, and poor stability (and thus be of even lower cost), without substantially impairing the precision of the resistance calibration system 200 discussed herein.

[0035] Control circuit block 203 may comprise an op-amp 240 and a programmable reference voltage source VREF 242 coupled to the (e.g., DMOS) regulation transistor M1 and reference resistor Rs to form a current servo loop. For example, control circuit block 203 provides a calibration current as provided in Equation 3 below:



[0036] In one embodiment, at least two states of the control means are implemented by providing two states of the programmable reference voltage source VREF 242. For example, in a first state, zero voltage is provided and in a second state, a positive voltage is provided. The positive voltage may be approximately 1 V. In some embodiments, the reference voltage VREF 242 need not be too precise and it also may have a large temperature drift without substantially impairing the precision of the resistance calibration system 200 discussed herein, as long as the reference voltage is substantially stable for the brief periods of time during which the calibration current is switched "ON." For example, the reference voltage VREF 242 defines the gross magnitude of the calibration current pulse during a brief interval of time. If the reference voltage is off by some 10% to 30% from the designed typical value, the calibration 200 discussed herein still may work precisely as long as the circuitry does not exceed its operational margins and if the reference voltage is also used within the signal processing, which is described later.

[0037] In one embodiment, the signal conditioning circuit block 204 comprises an interpolator (IP) 220 that forms a weighted average voltage between the voltages at node 230A (Va) and at node 230B (Vb) as provided by Equation 4a below.



[0038] Accordingly, Equation 4a describes a proportional relationship. For example, when passive interpolators (e.g., based on resistors and/or capacitors) are used the proportionality factor typically is 0.5 and Equation 4a takes the form of Equation 4b below:



[0039] In another example, where active interpolators (e.g., based on operational amplifiers) are used, there may be an interpolator gain GI. In this regard, the proportionality factor may be other than 0.5, as provided by Equation 4c below:



[0040] In Equations 4a to 4c above, "a" is an interpolation parameter that may be constant or variable. It corresponds to the location of the third node 230C in relation to the first and second nodes 230A and 230B. An exemplary choice for the setting interpolation parameter "a" is discussed below. In many practical implementations, numerical values for interpolation parameter "a" may, for example, span a range of - 0.3 to +0.3, depending on the expected mechanical, thermal and electrical tolerances of the resistive structure 201. In some embodiments, numerical values for the interpolation parameter "a" may exceed this range. However, exceeding this range may reduce attainable accuracy of the calibration. For example, accuracy of the calibration for a given calibration current level is best when the signal Vavg has the largest amplitude, which occurs when interpolation parameter "a" is zero or close to zero. High accuracy can be accomplished by reducing the mechanical, thermal and electrical tolerances of the resistive structure to a reasonable level and by bringing the third node 230C as close to the electrical center equipotential line between the first and second nodes as reasonably possible (e.g., by proper exploitation of the given capabilities of the manufacturing process used to make the resistive structure). For basic geometries, such as a resistive structure having a rectangular shape, the preferred location to place the third node 230C (e.g., such that the interpolation parameter "a" is close to zero) is at the centroid of the shape between nodes 230A and 230B (e.g., at the intersection of the diagonals of the rectangular shape). For more complicated shapes, a finite element analysis may be conducted to find the optimum place for the third node.

[0041] Signal conditioning circuit block 204 also may include a difference amplifier (AMP) 222 that is configured to amplify the difference between a voltage Vc at the third node 230C and the weighted average voltage Vavg provided by interpolator 220 with a gain factor of GA, as provided by Equation 5a below.



[0042] If an active interpolator having an interpolator gain GI is used, then the input of the difference amplifier 222 coupled to the third node 230C of the resistive structure 201 has a gain as provided by Equation 5b below.



[0043] In one embodiment, this difference amplifier gain function of Equation 5a and the weighted average function of the interpolator 222 may be combined by configuring an operational amplifier according to known procedures to achieve a transfer function as provided by Equation 6 below.



[0044] For example, the calculation above involves a first input network coupling Va and Vb to a first input of an operational amplifier and a second input network coupling Vc to a second input of the operational amplifier. Further, a feedback network couples the output of the operational amplifier to the inverting input of the operational amplifier. The first and second input network and the feedback network may be resistive or capacitive.

[0045] In one embodiment, the interpolator 220 comprises a string of resistors having a center tap (or multiple selectable taps arranged around the center). The tap location (e.g., corresponding to interpolation parameter "a") is chosen to correspond to the location of the third node 230C in relation to the first and second nodes 230A and 230B. An exemplary procedure for the choosing interpolation parameter "a" is discussed below.

[0046] In one embodiment, the interpolator 222 may be combined with the amplifier 240 (e.g., configured as a switched capacitor circuit), where the interpolation parameter "a" is implemented as a ratio of input capacitors of the switched capacitor circuit, or is inherent in the switching sequence. An advantage of such a capacitive interpolator is that it does not draw DC currents from the resistive structure, so it may offer the same or better calibration accuracy as a high ohmic resistive interpolator does.

[0047] In various embodiments where a distributed resistive structure 201 is used (e.g., such as a PCB foil run), more than one node (e.g., C, C', C" of FIG. 8) near or at the center of the resistive structure 201 may be used to inject calibration currents. Further, more than one of these nodes may be used to provide a voltage Vc to an input node of the difference amplifier 222. In one embodiment, if more than one such node contributes to the voltage level at node Vc, the difference amplifier 222 may include multiple inputs to accommodate the multiple input voltages. For example, each voltage input to the amplifier 222 may have equal or individual gain factors. In one embodiment, finite element analysis of the distributed resistive structure 201 is used to determine optimum gain factors for each input of a multiple input difference amplifier. In the finite element model, the gain factors are easily determined from the relative magnitude of the electric potentials at the nodes providing the multiple input voltages when a known calibration current is injected at one or multiple other nodes in their vicinity.

[0048] The signal processing circuit block 206 may comprise a switched capacitor integrator and comparator (e.g., an analog to digital converter (ADC) based on delta-sigma modulator techniques). For example, the switched capacitor integrator and comparator are configured as a modulator of a delta sigma analog-to-digital converter. The modulator uses a reference voltage that tracks the reference voltage 242 that is used to regulate the calibration current ICAL. In one embodiment, the signal processing circuit block 206 adds or subtracts Vout and/or reference voltages from a stored integration result and checks for the sign of the integration result. A digital decimation filter may be added in hardware or implemented in software. In one embodiment, the reference voltage for the ADC is the same as the reference voltage 242 for the control circuit block 203. Put differently, the reference voltage for the ADC and reference voltage 242 both are derived from a common voltage source. In this case, the ADC result automatically corrects for any imprecision or temperature drift of the reference voltage source, as mentioned above.

[0049] With the foregoing overview of the system 200, it may be helpful now to consider a high-level discussion of an exemplary operation of the system 200. In a first phase of operation, reference voltage source VREF 242 is set to zero (i.e., calibration current Ical is OFF) and Vout at node 205 is sampled (e.g., by a delta sigma ADC within the signal processing block 206). This may be the first sample of a correlated double sampling (CDS) signal processing method.

[0050] In a second phase of operation, reference voltage source VREF 242 is set to a positive voltage (e.g., to 1.23V, which may be provided by a band-gap reference, but also can be from a reference that is much less precise than a band-gap reference). The calibration current drawn from the third node 230C is provided by Equation 7 below.



[0051] After a brief delay (i.e., when the calibration current is sufficiently stable), Vout at node 205 is sampled again. In one embodiment, this represents the second sample of the CDS signal processing method. It should be noted that other sampling, measurement, or digitizing approaches known in the art of signal processing may be used, as long as they achieve the capture of the Vout difference between the two phases of operation. In one embodiment, the first phase and second phase operations are periodic.

[0052] In one embodiment, an idle phase is included. For example, in the optional idle phase, reference voltage source VREF 242 is again set to zero volts. To save power during the optional idle phase, all the functional blocks discussed above (such as reference voltage source VREF 242, AMP 222, Interpolator 220, and the ADC) may be set to a power down state. The optional idle phase also helps to cool down the electronic components that generated the calibration current pulse. As noted above, the idle phase is optional since the same cool down effect can be achieved in one embodiment by making the first phase of operation significantly longer than the second phase of operation.

[0053] For example, the absolute value of the difference between the signal samples taken in the first phase and the second phase is proportional to the product of the total resistance Rx ICAL, even in the presence of offset voltages in the amplifiers or in the ADC. The proportionality factor itself may depend on the total system 200 gain. An advantage of the exemplary signal processing technique discussed herein is that parasitic thermocouple voltages at the three nodes 230A, 230B, and 230C do not have any influence on the measurement of Rx.

[0054] In one embodiment, if the reference voltage source VREF 242 is also used in the ADC, it is not included in the system gain equation for Rx / Rs. Accordingly, the reference voltage source VREF 242 need not be very precise.

[0055] Further, in one embodiment, if the interpolation parameter "a" discussed above is chosen such that the weights used in forming the weighted average of the voltages are proportional to the magnitude of the partial resistances Rx1 and Rx2, then the load current flowing though the resistive structure 201 (Rx) also has no effect on the measurement of the resistive structure 201 (Rx).

[0056] If the resistive structure is such that the interpolation parameter "a" does not need tuning (e.g., when there are no significant thermal gradients across the resistive structure or if its temperature coefficient of resistance is small), a trim circuit may be used to set the interpolation parameter "a" to an optimum value. For example, this optimum value may be determined once during the production of the system. For example, if the resistive structure is part of an integrated circuit, laser blown fuse links or some other known form of trimming circuitry may be used to choose one of several taps on a resistive interpolator to set the interpolation parameter "a" to an optimum value.

[0057] In one embodiment, automatic tuning of the interpolation parameter "a" of the interpolation (performed by interpolator 220) can be accomplished by varying interpolation parameter "a" until Vout is minimal in the first phase of operation, (e.g., when calibration current is zero).

[0058] Tuning circuitry for automatic tuning of the interpolation parameter "a" may take several forms. In a preferred embodiment, tuning is based on detecting a residual signal Vout (205) that is caused by a change in the magnitude of currents that enter the resistive structure 201 through the first node and exit at the second node (or vice versa), during operation of the system.

[0059] For example, the tuning circuitry or tuning method discussed herein may use the correlation between load current (e.g., observed as voltage Vab between nodes 230A and 230B) and Vout at node 205. The ideal tuning point is where the correlation is minimized. In one example, to achieve minimization of the correlation, the correlation circuit simply provides the sign of the correlation, and the tuning process changes the chosen tap on a resistive interpolator up or down, depending on the sign. After a brief period of operation, the tuning process jumps between the two most optimum taps. Interpolation parameter "a," which enters the gain equation for the calibration process as a factor (1-a2), follows from the chosen tap location according to Equation 5 above. This factor (e.g., close to 1.00 for small deviations from the ideal center tap case), may be ignored for systems calling for modest accuracy. However, in more accurate systems this factor may also be used by the signal processing block to correct for a change in calibration gain. Hence, the system can automatically track and correct for any deviations from the ideal center tap case.

[0060] FIG. 9 illustrates a correlation circuit 900, compatible with an embodiment of the present invention. In the example of FIG. 9, correlation circuit 900 comprises an analog four quadrant multiplier 950 coupled to an integrator 970. The differential voltage Vab across the input nodes of the amplifier 940 provides the first input signal and Vout provides the second input signal of the four quadrant multiplier 950. In one embodiment, differential voltage Vab is amplified by an amplifier 940 before providing it to the four quadrant multiplier 950. In one embodiment, a delta sigma modulator of an ADC is used in lieu of an analog integrator 970. To allow reuse of an integrator or ADC used in the calibration circuitry for the tuning process, a switch 960 may be used. For example, the switch 960 couples the integrator 970 to Vout in a "CAL" state while coupling the integrator 970 to the multiplier (950) in a "TUNE" state to provide a correlator. If the load current is unidirectional, a two quadrant multiplier may be used instead. The correlation also may be performed in the digital domain by correlating the PWM bit streams of two delta sigma modulators, one being provided the signal Vab, and the other one being provided by Vout at node 205. The signals Vab and/or Vout may be amplified before being sent to the multiplier.

[0061] In various embodiments, the correlation operation (i.e., the interpolation parameter "a" tuning process) may use the natural load current ripple caused by the inductor current ripple inherent to switch-mode power supplies, load current ripple that may be caused by changing the data processing (CPU) load, and the like. For example, current ripple of sufficient amplitude may not be available to perform the interpolator 220 tuning. In this regard, system software may provoke sufficient load current ripple by, for example, performing many computations from time to time, and less computations at other times, to modulate the load current. The system software also may intelligently control the tuning process to occur during such phases of operation where sufficient load current is present due to a varying computational load.

[0062] Load current ripple suitable for the correlation operation may also be generated by modulating the voltage feedback network of a switching or linear regulator power supply. For example, such load current ripple can be achieved by coupling a rectangular digital modulation signal to the feedback pin of a switching power supply via a resistor. For example, the resistance of the resistor is chosen such that a digital signal swing changes the output voltage of the switching power supply within the limits specified by the load. For example, for a load that can tolerate a 10% supply voltage variation, the modulation of the output voltage of the switching power supply can be set to 5%. In one embodiment, the period of the rectangular digital modulation signal is preferably is 10 to 100 times longer than the switching frequency. However, any other timing also may be used as long as the voltage control loop transient response to the modulation is detectable. For example, in the correlation operation, the voltage Vout of node 205 may be multiplied with a signal derived from the digital modulation signal. In one embodiment, a delayed version of the modulation signal is used, where the delay is chosen such that optimum sensitivity of the correlation process is achieved. Instead of using the modulation signal, the voltage Vout of node 205 may be also be multiplied by the voltage difference between the voltage at the first node 230A and the second node 230B (Vab) or an amplified version thereof in the correlation operation.

[0063] In one embodiment, the analog multiplier 950 is not used. Instead, the correlation operation uses the gate drive (switch pin) signal of a switching power supply. Reference now is made to FIG. 10, which illustrates a system having a tunable interpolator and a correlator circuit that is based on polarity reversing switches which are driven by switching phases of a switch-mode power supply, consistent with an embodiment of the present invention. The gate drive output(s) 1081, 1082 of a switching power supply controller 1080 may be amplified, shaped to a digital control signal, extended, or reduced in time duration, by a switch gate conditioning circuit 1090. The resulting digital control signals 1091 and 1092 may be used to set the polarity of the switch arrangement 1095. For example, the switch arrangement couples the voltage Vout 1005 or its polarity inverse 1006 from a difference amplifier 1022 to an integrator 1070 or analog to digital converter (ADC) (e.g., a delta sigma ADC). As shown in FIG. 10, the difference amplifier 1022 (e.g., corresponding to difference amplifier 222 of FIG. 2) may have a differential output, thereby allowing changing the polarity of the signals at nodes 1005 and 1006 before they are sent to the integrator 1070. Alternatively, in case of single ended difference amplifier, a polarity changing switch arrangement under control of the digital control signals 1091 and 1092 is used in the signal path before the difference amplifier 1022. However, additional gating for the desired timing window should then be added to its output. For example, the tunable interpolator 1020 corresponds to 220 in FIG. 2, and the resistive structure 1001 corresponds to 201 in FIG. 2. In one embodiment, the switching power supply 1000 includes a PFET 1083, an NFET 1084, an inductor 1085, and a capacitor 1086. The load 1111 is coupled to the output of the resistive structure 1001.

[0064] In one embodiment, where a switching power supply is not present or an existing regulator cannot be used for the correlation processes as described above, the sufficient load current ripple is provided by a tuning current source coupled to at least one of the nodes 230A and 230B. For example, the tuning current source is then pulsed to create the desired ripple. Alternatively, as shown in FIG. 3, a calibration current source may be reused as a tuning current source for a load current ripple by providing additional paths to couple the ripple current to node 330A or 330B (or both nodes 330A or 330B alternately), in addition to the basic coupling to node 330C. For example some DMOS transistors (e.g., M2, M3, M4) may be added in the output branch of the calibration current generating circuit (e.g., 202 in FIG. 2). Transistor M3 is switched ON and M2/M4 are switched OFF to inject a calibration current pulse into the third node 330C. To generate a tuning current pulse, M3 is switched OFF and either M2 or M4 are switched ON. In one embodiment, transistor M4 is omitted if ping-pong operation between nodes 330A and 330B (e.g., where transistor M4 and M2 are alternatingly turned ON) is not desired. In one example, generating such tuning current pulses depends on the particular load and power supply circuitry (not shown) that is coupled to nodes 330A and 330B.

[0065] FIG. 4 illustrates a system 400 that is capable of steering current in a resistive structure, consistent with an embodiment of the present invention. FIG. 4 illustrates a simplified configuration to more easily describe certain aspects of the invention. In one embodiment, several DMOS transistors (e.g., M1B and M1C) steer the calibration current pulses ICAL alternately to nodes 430B and 430C. The switching is provided by electronically controlled switches 450 and 452. Thus, the current pulses regulated by the operational amplifier 440 can be directed to either node 430C or node 430B, depending on which switch is closed. In one example, these pulses directed to node 430B of the resistive structure 401 may be used in lieu of natural load current ripple to tune the interpolation that is provided by the interpolator (within signal conditioning block 404). Alternatively, the drain of regulation transistor M1B may be tied to node 430A. In one embodiment, the drain of regulation transistor M1B is tied to the node 430A or 430B that has the smaller capacitance. Attaching the drain of regulation transistor M1B to the node with less capacitance forces the larger part of current pulse to flow though the resistive structure, thereby increasing the signal amplitude.

[0066] In various embodiments, the calibration current source may also be implemented by splitting a switch or pass transistor of a switch-mode or linear power supply. In this regard, FIG. 5 illustrates a system 500 that steers current in a resistive structure by splitting a switch, consistent with an embodiment of the present invention. For example, part of the load current ILOAD is periodically diverted from a first node (e.g., 530A) of the resistive structure 501 into a third node (e.g., 530C). This may be particularly beneficial if the resistive structure 501 and the switch or pass transistors (e.g., 5M2 and 5M1) are integrated on a switch-mode or linear-mode power supply chip. In one embodiment, the top metal layers of a semiconductor chip or bond wires may be used as the resistive structure 501.

[0067] It may be helpful now to consider a high-level description of an exemplary operation of the system 500. In one embodiment, in the first phase of operation, transistor 5M1 is OFF while transistor 5M2 carries all of the load current ILOAD. In the second phase of operation, 5M1 is turned ON thereby carrying part of the load current ILOAD. For example, at this stage, a part of ILOAD serves as the calibration current ICAL for the resistive structure 501. In one embodiment, the voltage drop across resistor 557 (Rs) is used to determine the calibration current ICAL or to regulate it to a desired value by including transistor 5M1 in the feedback loop. Since calibration current ICAL may be orders of magnitude less than the maximum load current ILOAD, both transistors 5M1 and resistor 557 may be configured to handle a much smaller current than transistor 5M2 and resistive structure 501. Accordingly, in this embodiment, the calibration current pulses do not contribute to the power consumption.

[0068] The foregoing embodiments can be augmented with an additional (e.g., secondary) calibration step, where the resistor 557 (Rs) of the calibration current branch ICAL is calibrated against an additional reference resistor or a reference current, as represented the secondary calibration current source 507 of FIG. 5. In one embodiment, secondary calibration current source 507 comprises a switchable (e.g., 569) reference current source 572. When switch 569 is ON, the reference current source 572 is coupled to resistor 557 (Rs) of the calibration current circuit block. For example, switch 569 is switched ON when the primary calibration system (for the resistive structure 501) is idle and does not produce a calibration current ICAL. During this phase of operation, the transistors 5M1 (or M1C and M1B of FIG. 4) are switched OFF, such that a secondary calibration current IREF flows through series resistor 557 (or reference resistor Rs of FIG. 4) undisturbed.

[0069] The magnitude of resistor 557 (Rs) is based on the circuit that measures the resistance, such as an ADC that measures the voltage across resistor 557 (not shown). For example, this secondary calibration step may be useful for implementations where the resistor 557 (or any resistor, such as Rs) associated with the calibration current generating circuit is unstable, unknown, subject to process tolerances, or has high temperature coefficients. In the example of FIG. 5, resistor 557 itself may be of the same metal layer as resistive structure 501. Accordingly, resistor 557 may have the same uncertainties - which are calibrated out in the secondary calibration step. However, since resistor 557 only needs to accommodate a small fraction of the maximum load current ILOAD, the resistor 557 can be made smaller in area and be higher ohmic than resistive structure 501. The same reasoning may apply to any resistors in the switchable reference current source 507 that generates the secondary calibration current Iref.

[0070] The reduction in size of the resistor 557 may have additional benefits. For example, if the magnitude of Iref is small enough, more stable and precise resistors (such as Silicon-Chrome thin-film resistors) may be used in the switchable reference current source 572 to stabilize Iref. Accordingly, including a secondary calibration capability (through reference block 507) may substantially relax the requirements for the material and construction of resistor 557. Such secondary calibration capability may also be added to any calibration current defining resistors Rs as discussed above, by adding a secondary calibration current source 507 to Rs, and providing a circuit to measure the voltage across reference resistor Rs that is caused by the secondary calibration current. The circuit within secondary calibration current source 507 may be similar to the example seen in block 202 of Fig. 2. It will be understood that any other circuit, means, and method known in the art to generate precision reference currents may also be used based on the concepts discussed herein.

[0071] Adding the secondary calibration step as discussed above further reduces the current and power levels used to generate the final precision reference current Iref on which the calibration of Rx of the resistive structure 101 is based in the end. This reduction of current and power levels can be several orders of magnitude per calibration step. This concept can be extended to a ternary (or higher) calibration step(s) until current and power levels are achieved for which a precision reference resistor can be provided at desired cost. The reference resistor also may be synthesized by a known technique that uses switched capacitor or charge balancing circuits that derive a reference current from a reference voltage source, a capacitor, and a time-base.

[0072] In some applications (e.g., where the typical broadband sinc spectrum of calibration current pulses is not desired), a resonant tank circuit may be used as a calibration current source. In this regard, FIG. 6 illustrates part of a system 600 that measures the resistance of a resistive structure 602, consistent with an embodiment of the present invention. The example of system 600 includes a resonant tank circuit block 608 comprising an inductor 6L1 in series with a capacitor 6C1. In one embodiment, the tank circuit 608 is excited by inductive or capacitive coupling. For example, the tank circuit 608 is excited by an excitation source 609 via capacitive coupling with capacitor 6C2. In one embodiment, the tank circuit 608 generates a pure sinusoidal calibration current of a predetermined frequency with no harmonics. In one embodiment, the tank circuit 608 is configured to a frequency band where it does not interfere with any sensitive signals in the system.

[0073] FIG. 7 illustrates an exemplary resistive structure 701, consistent with an embodiment of the present invention. For example, the resistive structure may be a distributed resistance provided by a printed circuit board (PCB) foil run 701. FIG. 7 shows an exemplary resistive structure 701 in which the electrical node 130C from FIG. 1 is essentially split into two nodes 730C and 730C' at (or near) the center of the resistive structure 701. The calibration current Ical is still injected at node 730C, but the signal conditioning circuit block 704 is now connected to node 730C'. This configuration yields a pseudo-Kelvin connection where the influence of the parasitic resistances in the Ical path are mitigated by splitting the third node into two nodes (e.g., 730C and 730C'). FIG. 7 illustrates the approximate equipotential lines 770 caused by Ical. It should be noted that when the calibration current Ical is flowing, the voltages at nodes 730C and 730C' are not the same. Thus, the connection may not be a true and ideal Kelvin connection. However, since the bulk of the resistive material between nodes 730C and 730C' still is part of the resistance from node 730A to node 730B, there is a linear relationship (or an additional gain factor to be accounted for) in the resistive structure 701 calibration process, which can be derived for a given geometry. For example, the linear relationship or the gain factor can be determined by direct measurement of a prototype or by finite element analysis software.

[0074] FIG. 8 illustrates another variation of the pseudo Kelvin connect approach, consistent with an embodiment of the present invention. For example, there is a two layer PCB board, having a bottom layer (striped) and a top layer (not striped). The drain of the transistor 8M1 is coupled to the bottom layer of the PCB foil runs. Preferably, the bottom layer PCB foil runs are of equal length and as symmetrical as possible, through via holes to physical nodes 830C' and 830C", where both receive part of the calibration current. Similarly, in one embodiment, nodes 830A, 830B, and 830C of the resistive structure 801 (in the top layer) are connected by via holes to the bottom layer traces running to the signal conditioning circuit block 804. In yet another embodiment that is able to handle very large currents, the resistive structure 801 may be a part of a plane within a multilayer PCB and the nodes 830A, 830B, 830C' and 830C" may be vias placed on the plane according to the teachings herein. As was previously discussed, the result of the calibration in this case may be scaled according to finite element analysis results to compensate for the gain factor associated with such distributed resistive structures.

[0075] FIG. 11 illustrates an exemplary resistive structure following another pseudo Kelvin connect circuit, consistent with an embodiment of the present invention. For example, the calibration current Ical from regulation transistor 11M1 is injected at a node 1130C of the resistive structure, which may be a printed circuit board trace or plane. The third node response signal is formed by averaging the voltages from adjacent nodes 1130C' and 1130C". In one example, the averaging is performed by two equally sized resistors or capacitors 1171, 1172 coupled to nodes 1130C' and 1130C", respectively. The common node 1104C of the resistors or capacitors used as coupling elements electrically corresponds to the third node (130C) of the idealized equivalent resistive structure comprising lumped elements as provided in the discussion of FIG. 1. It should be noted that other variations of how the calibration current generation circuit and the signal conditioning circuit 1104 may be coupled to a third node of an resistive structure by means of additional electronic components are possible without deviating from the scope and spirit of the embodiments discussed herein.

[0076] The components, steps, features, objects, benefits and advantages that have been discussed herein are merely illustrative. None of them, nor the discussions relating to them, are intended to limit the scope of protection in any way. Numerous other embodiments are also contemplated. These include embodiments that have fewer, additional, and/or different components, steps, features, objects, benefits and advantages. These also include embodiments in which the components and/or steps are arranged and/or ordered differently. For example, bipolar transistors (e.g., PNP or NPN) can be used instead of MOS transistors. A PNP may be used instead of NPN, and a PMOS may be used instead of NMOS. Accordingly, it is intended that the invention be limited only in terms of the appended claims.


Claims

1. A system (100; 200) for measuring electrical resistance, comprising:

a resistive structure (101; 201) having:

a first node (130A; 230A) at a first end of the resistive structure (101; 201);

a second node (130B; 230B) at a second end of the resistive structure (101; 201); and

a third node (130C; 230C) on the resistive structure (101; 201) between the first and second nodes (130A, 130B; 230A, 230B);

a calibration current generation circuit (102; 202) having an output coupled to the third node (130C; 230C) of the resistive structure (101; 201), and configured to sequentially couple at least two different calibration currents (ICAL) of different magnitude or sign into the third node (130C; 230C) of the resistive structure (101; 201);

a signal conditioning circuit (104; 204) comprising:

a first input coupled to the first node (130A; 230A) of the resistive structure (101; 201);

a second input coupled to the second node (130B; 230B) of the resistive structure (101; 201);

a third input coupled to the third node (130C; 230C) of the resistive structure (101; 201); and

an output (105; 205),

wherein the signal conditioning circuit (104; 204) is configured to provide at its output (105; 205) a signal proportional to a difference between a voltage at the third node (130C; 230C) and a weighted average of the voltages (Vavg) at the first and second nodes (130A, 130B; 230A, 230B) of the resistive structure (101; 201); and

a signal processing circuit (106; 206) coupled to the output (105; 205) of the signal conditioning circuit (104; 204), wherein the signal processing circuit (106; 206) is configured to calculate the resistance of the resistive structure (101; 201) from a difference between a first output voltage of the signal conditioning circuit (104; 204) when the calibration current (ICAL) is set to a first value, and a second output voltage of the signal conditioning circuit (104; 204) when the calibration current (ICAL) is set to a second value, the difference being proportional to a product of the resistance Rx of the resistive structure (101; 201) and the calibration current (ICAL).


 
2. The system (100; 200) of claim 1, wherein the calibration current generation circuit (102; 202) comprises a current servo loop having a regulation transistor (M1) that is configured to regulate the calibration current (ICAL) to a value determined by a reference voltage and a reference resistor (Rs).
 
3. The system (100; 200) of claim 2, wherein the regulation transistor (M1) of the current servo loop is configured to be switched between at least two current values; and/or
wherein the regulation transistor (M1) of the current servo loop is configured to be switched ON and OFF.
 
4. The system (100; 200) of claim 2 or 3, wherein the signal processing circuit comprises a switched capacitor integrator and a comparator, wherein:

the switched capacitor integrator and comparator may be configured as a modulator of a delta sigma analog-to-digital converter; and

the modulator may be configured to use a reference voltage that tracks the reference voltage that is used to regulate the calibration current (ICAL).


 
5. The system (100; 200) of any one of claims 1 to 4, wherein the signal conditioning circuit (104; 204) comprises:

an interpolator (220) having a first input coupled to the first node (130A; 230A) of the resistive structure (101; 201), a second input coupled to the second node (130B; 230B) of the resistive structure (101; 201), and an output; and

a difference amplifier (222) having a first input coupled to the output of the interpolator (220) and a second input coupled to the third node (130C; 230C) of the resistive structure (101; 201), wherein the difference amplifier (222) is configured to amplify a difference between the third node (130C; 230C) of the resistive structure (101; 201) and a voltage at the output of the interpolator (220).


 
6. The system (100; 200) of claim 5, wherein the interpolator (220) is configured to provide a voltage at its output based on:

a voltage Va at the first node (130A; 230A) of the resistive structure (101; 201);

a voltage Vb at the second node (130B; 230B) of the resistive structure (101; 201); and

an interpolation parameter "a," wherein,

the voltage at the output is equal or proportional to 0.5[(1-a)Va + (1+a)Vb].


 
7. The system (100; 200) of claim 6, further comprising a circuit for trimming the interpolation parameter "a" such that the voltage at the output of the interpolator (220) is substantially the same magnitude as the voltage at the third node (130C; 230C) of the resistive structure (101; 201) when the calibration current (ICAL) is zero,
the system (100; 200) optionally further comprising a circuit configured to tune the interpolation parameter "a" such that influence of load current flowing though the resistive structure (101; 201) on the signal at the output (105; 205) of the signal conditioning circuit is minimized.
 
8. The system (100; 200) of claim 6 or 7, further comprising a correlator circuit (900) having a first correlator input coupled to the output of the signal conditioning circuit,
wherein the correlator circuit (900) may comprise a multiplier (950) coupled to an integrator (970).
 
9. The system (100; 200) of claim 8, wherein:

the output of the signal conditioning circuit (900) is a differential output having positive and negative polarity output nodes; and

the correlator circuit (900) comprises a polarity interchanging switch (960) arrangement coupled to an integrator (970),

wherein the correlator circuit (900) may further include a second input, wherein the second input may be:

differential; and

coupled to the first and second nodes (930A; 930B) of the resistive structure (901).


 
10. The system (100; 200) of claim 8 or 9, further comprising at least one tuning current source coupled to at least the first or second node (930A, 930B) of the resistive structure (901), wherein the at least one tuning current source is configured to provide tuning current pulses for tuning the interpolator (920).
 
11. The system (100; 200) of claim 10, wherein the calibration current generation circuit (202) comprises a servo loop having:

a regulation transistor (M1) configured to regulate the calibration current (ICAL) to a value determined by a reference voltage (VREF) and a reference resistor (Rs);

a first current steering transistor (M3) coupled between the regulation transistor (M1) and the third node (330C) of the resistive structure; and

a second current steering transistor (M2/M4) coupled between the regulation transistor (M1) and one of (i) the first node (330A) or (ii) the second node (330B) of the resistive structure.


 
12. The system (100; 200) of claim 10 or 11, further comprising a plurality of switches (450, 452) having a common node coupled to a reference voltage and configured to render active one of (i) the first regulation transistor (M1B) or (ii) the second regulation transistor (M1C), wherein:

the calibration current generation circuit (02) comprises a servo loop having at least two regulation transistors (M1B, M1C) configured to regulate the calibration or tuning current to a value determined by a reference voltage and a reference resistor (Rs);

the first regulation transistor (M1C) is coupled to the third node (430C) of the resistive structure (401); and

the second regulation transistor (M1B) is coupled to one of the (i) first or (ii) second node (430A, 430B) of the resistive structure (401).


 
13. The system (100; 200) of any one of claims 5 to 12, further comprising:

an input voltage node (Vin);

a reference resistor (Rs);

at least two current steering transistors (5M1, 5M2), wherein a first current steering transistor (5M1) has a source coupled to the third node (530C) of the resistive structure (501) and a second current steering transistor (5M2) has a source coupled to the first node (530A) of the resistive structure (501);

a control circuit (503) having a separate output coupled to a gate of the at least two current steering transistors, respectively;

wherein:

the first node of the reference resistor (557) is coupled to the input voltage node (Vin);

the second node of the reference resistor (557) is coupled to the third node (530C) of the resistive structure (501) via the first current steering transistor (5M1);

the input node is coupled to the first node (530A) of the resistive structure (501) via the second current steering transistor (5M2); and

the control circuit (503) is configured to independently switch the current steering transistors ON and OFF; and

wherein the system (500) may further comprise a secondary calibration current source (507) coupled to the reference resistor.


 
14. The system (100; 200) of any one of claims 5 to 13, wherein the calibration current generation circuit (604) comprises a resonant tank circuit (608) that is coupled between the third node (630C) of the resistive structure (601) and to either the first or second node (630A, 630B) of the resistive structure (601).
 
15. A method of measuring resistance of a resistive structure (101; 201) having a first node (130A; 230A) at a first end of the resistive structure (101; 201), a second node (130B; 230B) at a second end of the resistive structure (101; 201), and a third node (130C; 230C) between the first node (130A; 230A) and the second node (130B; 230B) of the resistive structure (101; 201) and adapted to receive a calibration current (ICAL); a signal conditioning circuit (104; 204) comprising a first input coupled to the first node (130A; 230A) of the resistive structure (101; 201), a second input coupled to the second node (130B; 230B) of the resistive structure (101; 201), a third input coupled to the third node (130C; 230C) of the resistive structure (101; 201); and an output (105; 205), the method comprising:

determining a first calibration signal by measuring a voltage at the output (105; 205) when no calibration current is injected into the third node (130C; 230C) of the resistive structure (101; 201);

injecting a calibration current into the third node (130C; 230C) of the resistive structure (101; 201);

determining a second calibration signal by:

interpolating voltages at the first and second nodes (130A, 130B; 230A, 230B) of the resistive structure; and

subtracting the interpolated voltage from a voltage at the third node (130C; 230C) of the resistive structure (101; 201); and

determining an absolute value of a difference between the determined first calibration signal and the determined second calibration signal, wherein the absolute value is proportional to a product of the resistance Rx of the resistive structure (101; 201) and the calibration current (ICAL).


 
16. The method of claim 15, wherein the calibration current (ICAL) is injected into the third node (130C; 230C) of the resistive structure (101; 201) in a form of a pulse; and/orwherein the method further comprises performing the first calibration signal and the second calibration signal periodically; and/or
wherein the method further comprises measuring a current flowing between the first node (130A; 230A) and the second node (130B; 230B) of the resistive structure (101; 201).
 
17. The method of claim 15 or 16, wherein interpolating the voltage at the first node (130A; 230A) of the resistive structure (101; 201) and the voltage at the second node (130B; 230B) of the resistive structure (101; 201) includes determining a weighted average voltage (Vavg) between the voltages at the first node (130A; 230A) of the resistive structure (101; 201) and at the second node (130B; 230B) of the resistive structure (101; 201) based on:

a voltage Va at the first node (130A; 230A) of the resistive structure (101; 201);

a voltage Vb at the second node (130B; 230B) of the resistive structure (101; 201); and

an interpolation parameter "a," wherein,


 
18. The method of claim 17, further comprising:

amplifying a difference between the weighted average voltage (Vavg) and a voltage at the third node (130C; 230C) of the resistive structure (101; 201) by a factor GA; and

providing the amplified difference as an output voltage at the output (Vout),

wherein the output voltage at the output (Vout) may be based on:
a voltage Vc at the third node (130C; 230C) of the resistive structure (101), wherein,


 
19. The method of claim 17 or 18, further comprising tuning the signal conditioning circuit (104; 204) to compensate for a difference between a resistance between the first and third nodes (130A, 130C; 230A, 230C) of the resistive structure (101; 204) and a resistance between the second and third nodes (130B, 130C; 230B, 230C) of the resistive structure (101; 201).
 
20. The method of claim 19, wherein the tuning comprises varying the interpolation parameter "a" until the voltage at the output is minimized when no calibration current is injected into the third node (130C; 230C) of the resistive structure (101; 201).
 
21. The method of claim 20, wherein the tuning further comprises using a load current ripple as a signal source, wherein the load current ripple is caused by at least one of:

(i) a switch-mode power supply coupled to the first or third node (130A, 130C; 230A, 230C) of the resistive structure (101; 201); or

(ii) a load coupled to the first or third node (130A, 130C; 230A, 230C) of the resistive structure (101; 201); and

wherein the switch mode power supply may be modulated during the tuning process to generate a load current ripple.
 
22. The method of claim 20 or 21, wherein the tuning further comprises providing a current pulse to at least one of (i) the first node (130A; 230A) of the resistive structure (101; 201) or (ii) the second node (130B; 230B) of the resistive structure (101; 201) to create an artificial ripple current in the resistive structure (101; 201) to tune the signal conditioning circuit (104; 204).
 


Ansprüche

1. Ein System (100; 200) zum Messen des elektrischen Widerstands, das Folgendes umfasst:

eine resistive Struktur (101; 201), die Folgendes aufweist:

einen ersten Knoten (130A; 230A) an einem ersten Ende der resistiven Struktur (101; 201);

einen zweiten Knoten (130B; 230B) an einem zweiten Ende der resistiven Struktur (101; 201); und

einen dritten Knoten (130C; 230C) in der resistiven Struktur (101; 201) zwischen dem ersten und zweiten Knoten (130A, 130B; 230A, 230B);

eine Kalibrierstrom-Erzeugungsschaltung (102; 202) mit einem Ausgang, der mit dem dritten Knoten (130C; 230C) der resistiven Struktur (101; 201) gekoppelt ist, und konfiguriert ist, um mindestens zwei verschiedene Kalibrierströme (ICAL) unterschiedlicher Größe oder Vorzeichen in den dritten Knoten (130C; 230C) der resistiven Struktur (101; 201) sequentiell einzubinden;

eine Signalkonditionierungsschaltung (104; 204), die Folgendes umfasst:

einen ersten Eingang, der mit dem ersten Knoten (130A; 230A) der resistiven Struktur (101; 201) gekoppelt ist;

einen zweiten Eingang, der mit dem zweiten Knoten (130B; 230B) der resistiven Struktur (101; 201) gekoppelt ist;

einen dritten Eingang, der mit dem dritten Knoten (130C; 230C) der resistiven Struktur (101; 201) gekoppelt ist; und

einen Ausgang (105; 205),

wobei die Signalkonditionierungsschaltung (104; 204) konfiguriert ist, um an ihrem Ausgang (105; 205) ein Signal bereitzustellen, das proportional zu einer Differenz zwischen einer Spannung am dritten Knoten (130C; 230C) und einem gewichteten Mittelwert der Spannungen (Vavg) am ersten und zweiten Knoten (130A, 130B; 230A, 230B) der resistiven Struktur (101; 201); und

eine Signalverarbeitungsschaltung (106; 206), die mit dem Ausgang (105; 205) der Signalkonditionierungsschaltung (104; 204) gekoppelt ist, wobei die Signalverarbeitungsschaltung (106; 206) konfiguriert ist, um den Widerstand der resistiven Struktur (101; 201) aus einer Differenz zwischen einer ersten Ausgangsspannung der Signalkonditionierungsschaltung (104; 204), wenn der Kalibrierstrom (ICAL) auf einen ersten Wert eingestellt ist, und einer zweiten Ausgangsspannung der Signalkonditionierungsschaltung (104; 204), wenn der Kalibrierstrom (ICAL) auf einen zweiten Wert eingestellt ist, wobei die Differenz proportional zu einem Produkt aus dem Widerstand Rx der resistiven Struktur (101; 201) und dem Kalibrierstrom (ICAL) ist.


 
2. Das System (100; 200) nach Anspruch 1, wobei die Kalibrierstrom-Erzeugungsschaltung (102; 202) eine Strom-Servoschleife (current servo loop) mit einem Regeltransistor (M1) umfasst, der konfiguriert ist, um den Kalibrierstrom (ICAL) auf einen durch eine Referenzspannung und einen Referenzwiderstand (Rs) bestimmten Wert zu regeln.
 
3. Das System (100; 200} nach Anspruch 2, wobei der Regeltransistor (M1) der Strom-Servoschleife konfiguriert ist, um zwischen mindestens zwei Stromwerten geschaltet zu werden; und/oder
wobei der Regeltransistor (M1) der Strom-Servoschleife konfiguriert ist, um EIN- und AUS-geschaltet zu werden.
 
4. Das System (100; 200) nach Anspruch 2 oder 3, wobei die Signalverarbeitungsschaltung einen geschalteten Kondensator-Integrator und einen Komparator umfasst, wobei:

der geschaltete Kondensator-Integrator und der Komparator als Modulator eines Delta-Sigma-Analog-Digital-Wandlers konfiguriert werden können; und wobei

der Modulator konfiguriert werden kann, um eine Referenzspannung zu verwenden, die der Referenzspannung folgt, die zum Regeln des Kalibrierstroms (ICAL) verwendet wird.


 
5. Das System (100; 200) nach irgendeinem der Ansprüche von 1 bis 4, wobei die Signalkonditionierungsschaltung (104; 204) Folgendes umfasst:

einen Interpolator (220) mit einem ersten Eingang, der mit dem ersten Knoten (130A; 230A) der resistiven Struktur (101; 201) gekoppelt ist, einem zweiten Eingang, der mit dem zweiten Knoten (130B; 230B) der resistiven Struktur (101; 201) gekoppelt ist, und einem Ausgang; und

einen Differenzverstärker (222) mit einem ersten Eingang, der mit dem Ausgang des Interpolators (220) gekoppelt ist, und einem zweiten Eingang, der mit dem dritten Knoten (130C; 230C) der resistiven Struktur (101; 201) gekoppelt ist, wobei der Differenzverstärker (222) konfiguriert ist, um eine Differenz zwischen dem dritten Knoten (130C; 230C) der resistiven Struktur (101; 201) und einer Spannung am Ausgang des Interpolators (220) zu verstärken.


 
6. Das System (100; 200) nach Anspruch 5, wobei der Interpolator (220) konfiguriert ist, um eine Spannung an seinem Ausgang bereitzustellen, basierend auf:

einer Spannung Va am ersten Knoten (130A; 230A) der resistiven Struktur (101; 201);

einer Spannung Vb an dem zweiten Knoten (130B; 230B) der resistiven Struktur (101; 201); und

einem Interpolationsparameter "a", wobei,

die Spannung am Ausgang gleich oder proportional zu 0,5[(1-a)Va + (1+a)Vb] ist.


 
7. Das System (100; 200) nach Anspruch 6, ferner umfassend eine Schaltung zum Trimmen des Interpolationsparameters "a", sodass die Spannung am Ausgang des Interpolators (220) im Wesentlichen die gleiche Größe hat wie die Spannung am dritten Knoten (130C; 230C) der resistiven Struktur (101; 201), wenn der Kalibrierstrom (ICAL) null ist,
wobei das System (100; 200) optional ferner eine Schaltung umfasst, die konfiguriert ist, um den Interpolationsparameter "a" so abzustimmen (tune), dass der Einfluss des durch die resistive Struktur (101; 201) fließenden Laststroms auf das Signal am Ausgang (105; 205) der Signalkonditionierungsschaltung minimiert wird.
 
8. Das System (100; 200) nach Anspruch 6 oder 7, ferner umfassend eine Korrelatorschaltung (900) mit einem ersten Korrelator-Eingang, der mit dem Ausgang der Signalkonditionierungsschaltung gekoppelt ist,
wobei die Korrelationsschaltung (900) einen Multiplikator (950) umfassen kann, der mit einem Integrator (970) gekoppelt ist.
 
9. Das System (100; 200) nach Anspruch 8, wobei:

der Ausgang der Signalkonditionierungsschaltung (900) ein Differenzausgang ist, der Ausgangsknoten mit positiver und negativer Polarität aufweist; und wobei

die Korrelationsschaltung (900) eine mit einem Integrator (970) gekoppelte Polaritätswechsel-Schalteranordnung (960) umfasst,

wobei die Korrelatorschaltung (900) ferner einen zweiten Eingang beinhalten kann, wobei der zweite Eingang Folgendes sein kann:

differentiell; und

gekoppelt mit dem ersten und zweiten Knoten (930A; 930B) der resistiven Struktur (901).


 
10. Das System (100; 200) nach Anspruch 8 oder 9, das ferner mindestens eine Abstimmstromquelle umfasst, die mit mindestens dem ersten oder zweiten Knoten (930A, 930B) der resistiven Struktur (901) gekoppelt ist, wobei die mindestens eine Abstimmstromquelle konfiguriert ist, um Abstimmstromimpulse zur Abstimmung des Interpolators (920) bereitzustellen.
 
11. Das System (100; 200) nach Anspruch 10, wobei die Kalibrierstrom-Erzeugungsschaltung (202) eine Servoschleife umfasst, die Folgendes aufweist:

einen Regeltransistor (M1), der konfiguriert ist, um den Kalibrierstrom (ICAL) auf einen Wert zu regeln, der durch eine Referenzspannung (VREF) und einen Referenzwiderstand (Rs) bestimmt ist;

einen ersten Stromsteuerungstransistor (current steering transistor) (M3), der gekoppelt ist zwischen dem Regeltransistor (M1) und dem dritten Knoten (330C) der resistiven Struktur; und

einen zweiten Stromsteuerungstransistor (M2/M4), der zwischen dem Regeltransistor (M1) und einem von (i) dem ersten Knoten (330A) oder (ii) dem zweiten Knoten (330B) der resistiven Struktur gekoppelt ist.


 
12. Das System (100; 200) nach Anspruch 10 oder 11, das ferner eine Vielzahl von Schaltern (450, 452) umfasst, die einen gemeinsamen Knoten aufweisen, der mit einer Referenzspannung gekoppelt ist, und konfiguriert ist, um einen von (i) dem ersten Regeltransistor (M1B) oder (ii) dem zweiten Regeltransistor (M1C) aktiv zu machen, wobei:

die Kalibrierstrom-Erzeugungsschaltung (02) eine Servoschleife mit mindestens zwei Regeltransistoren (M1B, M1C) umfasst, die konfiguriert sind, um den Kalibrier- oder Abstimmstrom auf einen durch eine Referenzspannung und einen Referenzwiderstand (Rs) bestimmten Wert zu regeln;

der erste Regeltransistor (M1C) mit dem dritten Knoten (430C) der resistiven Struktur (401) gekoppelt ist; und wobei

der zweite Regeltransistor (M1B) mit einem der Folgenden: (i) dem ersten oder (ii) dem zweiten Knoten (430A, 430B) der resistiven Struktur (401) gekoppelt ist.


 
13. Das System (100; 200) nach irgendeinem der Ansprüche von 5 bis 12, das ferner Folgendes umfasst:

einen Eingangsspannungsknoten (Vin);

einen Referenzwiderstand (Rs);

mindestens zwei Stromsteuerungstransistoren (5M1, 5M2), wobei ein erster Stromsteuerungstransistor (5M1) eine Quelle aufweist, die mit dem dritten Knoten (530C) der resistiven Struktur (501) gekoppelt ist, und ein zweiter Stromsteuerungstransistor (5M2) eine Quelle aufweist, die mit dem ersten Knoten (530A) der resistiven Struktur (501) gekoppelt ist;

eine Steuerschaltung (503) mit einem separaten Ausgang, der jeweils mit einem Gate der mindestens zwei Stromsteuerungstransistoren gekoppelt ist;

wobei:

der erste Knoten des Referenzwiderstandes (557) mit dem Eingangsspannungsknoten (Vin) gekoppelt ist;

der zweite Knoten des Referenzwiderstandes (557) mit dem dritten Knoten (530C) der resistiven Struktur (501) über den ersten Stromsteuerungstransistor (5M1) gekoppelt ist;

der Eingangsknoten mit dem ersten Knoten (530A) der resistiven Struktur (501) über den zweiten Stromsteuerungstransistor (5M2) gekoppelt ist; und wobei

die Steuerungsschaltung (503) konfiguriert ist, um die Stromsteuerungstransistoren unabhängig voneinander EIN und AUS zu schalten; und

wobei das System (500) ferner eine sekundäre Kalibrierstromquelle (507) umfassen kann, die mit dem Referenzwiderstand gekoppelt ist.


 
14. Das System (100; 200) nach irgendeinem der Ansprüche von 5 bis 13, wobei die Kalibrierstrom-Erzeugungsschaltung (604) eine Resonanztankschaltung (resonant tank circuit) (608) umfasst, die zwischen dem dritten Knoten (630C) der resistiven Struktur (601) und entweder dem ersten oder zweiten Knoten (630A, 630B) der resistiven Struktur (601) gekoppelt ist.
 
15. Ein Verfahren zum Messen des Widerstands einer resistiven Struktur (101; 201) mit einem ersten Knoten (130A; 230A) an einem ersten Ende der resistiven Struktur (101; 201), einem zweiten Knoten (130B; 230B) an einem zweiten Ende der resistiven Struktur (101; 201) und einem dritten Knoten (130C; 230C) zwischen dem ersten Knoten (130A; 230A) und dem zweiten Knoten (130B; 230B) der resistiven Struktur (101; 201), der zum Empfangen eines Kalibrierstroms (ICAL) ausgelegt ist; eine Signalkonditionierungsschaltung (104; 204), die Folgendes umfasst: einen ersten Eingang, der mit dem ersten Knoten (130A; 230A) der resistiven Struktur (101; 201) gekoppelt ist, einen zweiten Eingang, der mit dem zweiten Knoten (130B; 230B) der resistiven Struktur (101; 201) gekoppelt ist, einen dritten Eingang, der mit dem dritten Knoten (130C; 230C) der resistiven Struktur (101; 201) gekoppelt ist, und einen Ausgang (105; 205), wobei das Verfahren Folgendes umfasst:

Bestimmen des ersten Kalibriersignals durch Messen einer Spannung am Ausgang (105; 205), wenn kein Kalibrierstrom in den dritten Knoten (130C; 230C) der resistiven Struktur (101; 201) eingespeist wird;

Einspeisen eines Kalibrierstroms in den dritten Knoten (130C; 230C) der resistiven Struktur (101; 201);

Bestimmen eines zweiten Kalibriersignals durch:

Interpolieren von Spannungen am ersten und zweiten Knoten (130A, 130B; 230A, 230B) der resistiven Struktur; und

Subtrahieren der interpolierten Spannung von einer Spannung am dritten Knoten (130C; 230C) der resistiven Struktur (101; 201); und

Bestimmen eines Absolutwertes einer Differenz zwischen dem bestimmten ersten Kalibriersignal und dem bestimmten zweiten Kalibriersignal, wobei der Absolutwert proportional zu einem Produkt aus dem Widerstand Rx der resistiven Struktur (101; 201) und dem Kalibrierstrom (ICAL) ist.


 
16. Das Verfahren nach Anspruch 15, wobei der Kalibrierstrom (ICAL) in den dritten Knoten (130C; 230C) der resistiven Struktur (101; 201) in Form eines Impulses eingespeist wird; und/oder wobei das Verfahren ferner das periodische Durchführen des ersten Kalibriersignals und des zweiten Kalibriersignals umfasst; und/oder
wobei das Verfahren ferner das Messen eines Stroms umfasst, der zwischen dem ersten Knoten (130A; 230A) und dem zweiten Knoten (130B; 230B) der resistiven Struktur (101; 201) fließt.
 
17. Das Verfahren nach Anspruch 15 oder 16, wobei das Interpolieren der Spannung am ersten Knoten (130A; 230A) der resistiven Struktur (101; 201) und der Spannung am zweiten Knoten (130B; 230B) der resistiven Struktur (101; 201) das Bestimmen einer gewichteten durchschnittliche Spannung (Vavg) zwischen den Spannungen am ersten Knoten (130A; 230A) der resistiven Struktur (101; 201) und am zweiten Knoten (130B; 230B) der resistiven Struktur (101; 201) beinhaltet, basierend auf:

einer Spannung Va am ersten Knoten (130A; 230A) der resistiven Struktur (101; 201);

einer Spannung Vb an dem zweiten Knoten (130B; 230B) der resistiven Struktur (101; 201); und

einem Interpolationsparameter "a", wobei


 
18. Das Verfahren nach Anspruch 17, das ferner Folgendes umfasst:

Verstärken einer Differenz zwischen der gewichteten durchschnittlichen Spannung (Vavg) und einer Spannung am dritten Knoten (130C; 230C) der resistiven Struktur (101; 201) um einen Faktor GA: und

Bereitstellen der verstärkten Differenz als Ausgangsspannung am Ausgang (Vout), wobei die Ausgangsspannung am Ausgang (Vout) beruhen kann auf:

einer Spannung Vc am dritten Knoten (130C; 230C) der resistiven Struktur (101),

wobei,


 
19. Das Verfahren nach Anspruch 17 oder 18,
ferner umfassend das Abstimmen (tuning) der SignalkonditionierungsSchaltung (104; 204), um eine Differenz zwischen einem Widerstand zwischen dem ersten und dritten Knoten (130A, 130C; 230A, 230C) der resistiven Struktur (101; 204) und einem Widerstand zwischen dem zweiten und dritten Knoten (130B, 130C; 230B, 230C) der resistiven Struktur (101; 201) zu kompensieren.
 
20. Das Verfahren nach Anspruch 19, wobei das Abstimmen das Variieren des Interpolationsparameters "a" umfasst, bis die Spannung am Ausgang minimiert ist, wenn kein Kalibrierstrom in den dritten Knoten (130C; 230C) der resistiven Struktur (101; 201) eingespeist wird.
 
21. Das Verfahren nach Anspruch 20, worin das Abstimmen ferner das Verwenden einer Laststromwelligkeit als Signalquelle umfasst, wobei die LaststromWelligkeit durch mindestens eine der Folgenden bewirkt wird:

(i) ein Schaltnetzteil, das mit dem ersten oder dritten Knoten (130A, 130C; 230A, 230C) der resistiven Struktur (101; 201) gekoppelt ist; oder

(ii) eine Last, die mit dem ersten oder dritten Knoten (130A, 130C; 230A, 230C) der resistiven Struktur (101; 201) gekoppelt ist; und

wobei das Schaltnetzteil während des Abstimmungsprozesses moduliert werden kann, um eine Laststromwelligkeit zu erzeugen.
 
22. Das Verfahren nach Anspruch 20 oder 21, wobei das Abstimmen ferner das Bereitstellen eines Stromimpulses für mindestens einen der Folgenden umfasst: (i) für den ersten Knoten (130A; 230A) der resistiven Struktur (101; 201) oder (ii) für den zweiten Knoten (130B; 230B) der resistiven Struktur (101; 201), zum Erzeugen eines künstlichen Welligkeitsstroms in der resistiven Struktur (101; 201), um die Signalkonditionierungsschaltung (104; 204) abzustimmen.
 


Revendications

1. Un système (100 ; 200) pour mesurer la résistance électrique, comprenant :
une structure résistive (101 ; 201) ayant :

un premier nœud (130A ; 230A) à une première extrémité de la structure résistive (101 ; 201) ;

un deuxième nœud (130B ; 230B) à une deuxième extrémité de la structure résistive (101 ; 201) ; et

un troisième nœud (130C ; 230C) sur la structure résistive (101 ; 201) entre le premier et le deuxième nœud (130A, 130B ; 230A, 230B) ;

un circuit de génération de courant d'étalonnage (102 ; 202) ayant une sortie couplée au troisième nœud (130C ; 230C) de la structure résistive (101 ; 201), et étant configuré pour coupler séquentiellement au moins deux courants d'étalonnage (ICAL) différents de grandeur ou de signe différents dans le troisième nœud (130C ; 230C) de la structure résistive (101 ; 201) ;

un circuit de conditionnement de signal (104 ; 204) comprenant :

une première entrée couplée au premier nœud (130A ; 230A) de la structure résistive (101 ; 201) ;

une deuxième entrée couplée au deuxième nœud (130B ; 230B) de la structure résistive (101 ; -201) ;

une troisième entrée couplée au troisième nœud (130C ; 230C) de la structure résistive (101 ; 201) ; et

une sortie (105 ; 205),

sachant que le circuit de conditionnement de signal (104 ; 204) est configuré pour fournir à sa sortie (105 ; 205) un signal proportionnel à une différence entre une tension au niveau du troisième nœud (130C ; 230C) et une moyenne pondérée des tensions (Vavg) au niveau du premier et deuxième nœud (130A, 130B ; 230A, 230B) de la structure résistive (101 ; 201) ; et

un circuit de traitement de signal (106 ; 206) couplé à la sortie (105 ; 205) du circuit de conditionnement de signal (104 ; 204), sachant que le circuit de traitement de signal (106 ; 206) est configuré pour calculer la résistance de la structure résistive (101 ; 201) à partir d'une différence entre une première tension de sortie du circuit de conditionnement de signal (104 ; 204) lorsque le courant d'étalonnage (ICAL) est réglé sur une première valeur et une deuxième tension de sortie du circuit de conditionnement de signal (104 ; 204) lorsque le courant d'étalonnage (ICAL) est réglé sur une deuxième valeur, la différence étant proportionnelle à un produit entre la résistance Rx de la structure résistive (101 ; 201) et le courant d'étalonnage (ICAL)


 
2. Le système (100 ; 200) d'après la revendication 1, sachant que le circuit de génération de courant d'étalonnage (102 ; 202) comprend une boucle d'asservissement de courant (current servo loop) ayant un transistor de régulation (M1) qui est configuré pour réguler le courant d'étalonnage (ICAL) à une valeur déterminée par une tension de référence et une résistance de référence (Rs).
 
3. Le système (100 ; 200) d'après la revendication 2, sachant que le transistor de régulation (M1) de la boucle d'asservissement de courant est configuré pour être commuté entre au moins deux valeurs de courant ; et/ou
sachant que le transistor de régulation (M1) de la boucle d'asservissement de courant est configuré pour être activé et désactivé.
 
4. Le système (100 ; 200) d'après la revendication 2 ou 3, sachant que le circuit de traitement de signal comprend un intégrateur à condensateur commuté et un comparateur, sachant que :

l'intégrateur à condensateur commuté et le comparateur peuvent être configurés comme modulateur d'un convertisseur analogique-numérique delta-sigma ; et que

le modulateur peut être configuré de manière à utiliser une tension de référence qui suit la tension de référence qui est utilisée pour réguler le courant d'étalonnage (ICAL).


 
5. Le système (100 ; 200) d'après l'une quelconque des revendications de 1 à 4, sachant que le circuit de conditionnement de signal (104 ; 204) comprend :

un interpolateur (220) ayant une première entrée couplée au premier nœud (130A ; 230A) de la structure résistive (101 ; 201), une deuxième entrée couplée au deuxième nœud (130B ; 230B) de la structure résistive (101 ; 201) et une sortie ; et

un amplificateur de différence (222) ayant une première entrée, couplée à la sortie de l'interpolateur (220), et une deuxième entrée couplée au troisième nœud (130C ; 230C) de la structure résistive (101 ; 201), sachant que l'amplificateur de différence (222) est configuré pour amplifier une différence entre le troisième nœud (130C ; 230C) de la structure résistive (101 ; 201) et une tension au niveau de la sortie de l'interpolateur (220).


 
6. Le système (100 ; 200) d'après la revendication 5, sachant que l'interpolateur (220) est configuré pour fournir une tension à sa sortie basée sur :

une tension Va au premier nœud (130A ; 230A) de la structure résistive (101 ; 201);

une tension Vb au deuxième nœud (130B ; 230B) de la structure résistive (101 ; 201); et

un paramètre d'interpolation "a," sachant que,

la tension à la sortie est égale ou proportionnelle à 0,5[(1-a)Va + (1+a)Vb].


 
7. Le système (100 ; 200) d'après la revendication 6, comprenant en outre un circuit pour ajuster le paramètre d'interpolation "a" de manière que la tension à la sortie de l'interpolateur (220) soit essentiellement de la même grandeur que la tension au troisième nœud (130C ; 230C) de la structure résistive (101 ; 201) lorsque le courant de calibration (ICAL) est zéro,
le système (100 ; 200) comprenant facultativement en outre un circuit configuré pour accorder le paramètre d'interpolation "a" de manière que l'influence du courant de charge traversant la structure résistive (101 ; 201) sur le signal à la sortie (105 ; 205) du circuit de conditionnement du signal soit minimisée.
 
8. Le système (100 ; 200) d'après la revendication 6 ou 7, comprenant en outre un circuit corrélateur (900) ayant une première entrée de corrélateur couplée à la sortie du circuit de conditionnement de signal,
sachant que le circuit corrélateur (900) peut comprendre un multiplicateur (950) couplé à un intégrateur (970).
 
9. Le système (100 ; 200) d'après la revendication 8, sachant que :

la sortie du circuit de conditionnement de signal (900) est une sortie différentielle ayant des nœuds de sortie de polarité positive et négative ; et que

le circuit corrélateur (900) comprend un agencement de commutateur d'échange de polarité (960) couplé à un intégrateur (970),

sachant que le circuit corrélateur (900) peut en outre inclure une deuxième entrée, sachant que la deuxième entrée peut être :

différentielle; et

couplée aux nœuds premier et deuxième (930A ; 930B) de la structure résistive (901).


 
10. Le système (100 ; 200) d'après la revendication 8 ou 9, comprenant en outre au moins une source de courant d'accord couplée à au moins le premier ou le deuxième nœud (930A, 930B) de la structure résistive (901), sachant que l'au moins une source de courant d'accord est configurée pour fournir des impulsions de courant d'accord pour accorder l'interpolateur (920).
 
11. Le système (100 ; 200) d'après la revendication 10, sachant que le circuit de génération de courant d'étalonnage (202) comprend une boucle d'asservissement présentant :

un transistor de régulation (M1) configuré pour réguler le courant d'étalonnage (ICAL) à une valeur déterminée par une tension de référence (VREF) et une résistance de référence (Rs) ;

un premier transistor de commande de courant (current steering transistor) (M3) couplé entre le transistor de régulation (M1) et le troisième nœud (330C) de la structure résistive ; et

un deuxième transistor de commande de courant (M2/M4) couplé entre le transistor de régulation (M1) et l'un parmi (i) le premier nœud (330A) ou (ii) le deuxième nœud (330B) de la structure résistive.


 
12. Le système (100 ; 200) d'après la revendication 10 ou 11, comprenant en outre une pluralité de commutateurs (450, 452) ayant un nœud commun couplé à une tension de référence, et configuré pour rendre actif l'un parmi (i) le premier transistor de régulation (M1B) ou (ii) le deuxième transistor de régulation (M1C), sachant que :

le circuit de génération de courant d'étalonnage (02) comprend une boucle d'asservissement ayant au moins deux transistors de régulation (M1B, M1C) configurés pour réguler le courant d'étalonnage ou d'accord à une valeur déterminée par une tension de référence et une résistance de référence (Rs) ;

le premier transistor de régulation (M1C) est couplé au troisième nœud (430C) de la structure résistive (401) ; et que

le deuxième transistor de régulation (M1B) est couplé à l'un parmi (i) le premier ou (ii) le deuxième nœud (430A, 430B) de la structure résistive (401).


 
13. Le système (100 ; 200) d'après l'une quelconque des revendications de 5 à 12, comprenant en outre :

un nœud de tension d'entrée (Vin) ;

une résistance de référence (Rs) ;

au moins deux transistors de commande de courant (5M1, 5M2), sachant qu'un premier transistor de commande de courant (5M1) présente une source couplée au troisième nœud (530C) de la structure résistive (501) et qu'un deuxième transistor de commande de courant (5M2) présente une source couplée au premier nœud (530A) de la structure résistive (501) ;

un circuit de commande (503) ayant une sortie distincte couplée à une grille des au moins deux transistors de commande de courant, respectivement ;

sachant que :

le premier nœud de la résistance de référence (557) est couplé au nœud de tension d'entrée (Vin) ;

le deuxième nœud de la résistance de référence (557) est couplé au troisième nœud (530C) de la structure résistive (501) via le premier transistor de commande de courant (5M1) ;

le nœud d'entrée est couplé au premier nœud (530A) de la structure résistive (501) par l'intermédiaire du deuxième transistor de commande de courant (5M2) ; et que

le circuit de commande (503) est configuré pour activer et désactiver indépendamment les transistors de commande de courant ; et

sachant que le système (500) peut en outre comprendre une source de courant d'étalonnage secondaire (507) couplée à la résistance de référence.


 
14. Le système (100 ; 200) d'après l'une quelconque des revendications de 5 à 13, sachant que le circuit de génération de courant d'étalonnage (604) comprend un circuit LC (resonant tank circuit) (608) qui est couplé entre le troisième nœud (630C) de la structure résistive (601) et le premier ou le deuxième nœud (630A, 630B) de la structure résistive (601).
 
15. Un procédé de mesure de résistance d'une structure résistive (101 ; 201) présentant un premier nœud (130A ; 230A) à une première extrémité de la structure résistive (101 ; 201), un deuxième nœud (130B ; 230B) à une deuxième extrémité de la structure résistive (101 ; 201) et un troisième nœud (130C ; 230C) entre le premier nœud (130A ; 230A) et le deuxième nœud (130B ; 230B) de la structure résistive (101 ; 201) et adapté pour recevoir un courant d'étalonnage (ICAL) ; un circuit de conditionnement de signal (104 ; 204) comprenant une première entrée couplée au premier nœud (130A ; 230A) de la structure résistive (101 ; 201), une deuxième entrée couplée au deuxième nœud (130B ; 230B) de la structure résistive (101 ; 201), une troisième entrée couplée au troisième nœud (130C ; 230C) de la structure résistive (101 ; 201) et une sortie (105 ; 205), le procédé comprenant le fait de :

déterminer un premier signal d'étalonnage en mesurant une tension à la sortie (105 ; 205) lorsqu'aucun courant d'étalonnage n'est injecté dans le troisième nœud (130C ; 230C) de la structure résistive (101 ; 201) ;

injecter un courant d'étalonnage dans le troisième nœud (130C ; 230C) de la structure résistive (101 ; 201) ;

déterminer un deuxième signal d'étalonnage par le fait de :

interpoler des tensions aux premier et deuxième nœuds (130A, 130B ; 230A, 230B) de la structure résistive ; et de

soustraire la tension interpolée d'une tension au troisième nœud (130C ; 230C) de la structure résistive (101 ; 201) ; et de

déterminer une valeur absolue d'une différence entre le premier signal d'étalonnage déterminé et le deuxième signal d'étalonnage déterminé, sachant que la valeur absolue est proportionnelle à un produit de la résistance Rx de la structure résistive (101 ; 201) par le courant d'étalonnage (ICAL).


 
16. Le procédé d'après la revendication 15, sachant que le courant d'étalonnage (ICAL) est injecté dans le troisième nœud (130C ; 230C) de la structure résistive (101 ; 201) sous la forme d'une impulsion ; et/ou sachant que le procédé comprend en outre le fait de mettre en œuvre périodiquement le premier signal d'étalonnage et le deuxième signal d'étalonnage ; et/ou
sachant que le procédé comprend en outre le fait de mesurer un courant s'écoulant entre le premier nœud (130A ; 230A) et le deuxième nœud (130B ; 230B) de la structure résistive (101 ; 201).
 
17. Le procédé d'après la revendication 15 ou 16, sachant que le fait d'interpoler la tension au niveau du premier nœud (130A ; 230A) de la structure résistive (101 ; 201), et la tension au niveau du deuxième nœud (130B ; 230B) de la structure résistive (101 ; 201) comprend le fait de déterminer une tension moyenne pondérée (Vavg) entre les tensions au niveau du premier nœud (130A ; 230A) de la structure résistive (101 ; 201) et au niveau du deuxième nœud (130B ; 230B) de la structure résistive (101 ; 201) en se basant sur :

une tension Va au premier nœud (130A ; 230A) de la structure résistive (101 ; 201);

une tension Vb au deuxième nœud (130B ; 230B) de la structure résistive (101 ; 201) ; et

un paramètre d'interpolation "a", sachant que


 
18. Le procédé d'après la revendication 17, comprenant en outre le fait de :

amplifier une différence entre la tension moyenne pondérée (Vavg) et une tension au troisième nœud (130C ; 230C) de la structure résistive (101 ; 201) par un facteur GA : et

fournir la différence amplifiée en tant que tension de sortie à la sortie (Vout),

sachant que la tension de sortie à la sortie (Vout) peut être basée sur :

une tension Vc au troisième nœud (130C ; 230C) de la structure résistive (101),

sachant que,


 
19. Le procédé d'après la revendication 17 ou 18,
comprenant en outre le fait d'accorder le circuit de conditionnement de signal (104 ; 204) pour compenser une différence entre une résistance entre le premier et le troisième nœud (130A, 130C ; 230A, 230C) de la structure résistive (101 ; 204) et une résistance entre le deuxième et le troisième nœud (130B, 130C ; 230B, 230C) de la structure résistive (101 ; 201).
 
20. Le procédé d'après la revendication 19, sachant que le fait d'accorder comprend le fait de varier le paramètre d'interpolation "a" jusqu'à ce que la tension à la sortie soit minimisée lorsque aucun courant d'étalonnage n'est injecté dans le troisième nœud (130C ; 230C) de la structure résistive (101 ; 201).
 
21. Le procédé d'après la revendication 20, dans lequel le fait d'accorder comprend en outre le fait d'utiliser un courant de charge ondulant comme source de signal, sachant que l'ondulation du courant de charge est provoquée par au moins un des éléments suivants :

(i) une alimentation à découpage couplée au premier ou au troisième nœud (130A, 130C ; 230A, 230C) de la structure résistive (101 ; 201) ; ou

(ii) une charge couplée au premier ou au troisième nœud (130A, 130C ; 230A, 230C) de la structure résistive (101 ; 201) ; et

sachant que l'alimentation à découpage peut être modulée pendant le processus d'accord pour générer une ondulation du courant de charge.
 
22. Le procédé d'après la revendication 20 ou 21, sachant que le fait d'accorder comprend en outre le fait de fournir une impulsion de courant à au moins l'un parmi (i) le premier nœud (130A ; 230A) de la structure résistive (101 ; 201) ou (ii) le deuxième nœud (130B ; 230B) de la structure résistive (101 ; 201) pour créer un courant ondulé artificiel dans la structure résistive (101 ; 201) afin d'accorder le circuit (104 ; 204) de conditionnement de signal.
 




Drawing


























Cited references

REFERENCES CITED IN THE DESCRIPTION



This list of references cited by the applicant is for the reader's convenience only. It does not form part of the European patent document. Even though great care has been taken in compiling the references, errors or omissions cannot be excluded and the EPO disclaims all liability in this regard.

Patent documents cited in the description




Non-patent literature cited in the description