CN106885588B - Sensor Arrangement with Thermal EMF Compensation - Google Patents

Abstract

The present invention relates to the sensor arrangements compensated with thermo-electromotive force.The sensor arrangement for providing signal in response to the temperature difference between hall effect device output contactor and reference point includes the first contact head, the proximity positioned at Hall effect region；Second contact head is located near reference point；First conductor element, including the first and second end sections, first end part is thermally coupled to the first contact head and second end part is thermally coupled to the second contact head；Second conductor element, including the third and fourth end section, third end section are thermally coupled to the first contact head；Third conductor element, including the 5th and the 6th end section, 5th end section is thermally coupled to the second contact head, wherein described first and third end section electronically couple, second and the 5th end section electronically couple, at least two conductor elements in first, second, and third conductor element have substantially different Seebeck coefficient, and in the 4th and the 6th end section tapped off signal.

Description

Sensor Arrangement with Thermal EMF Compensation

related application

This application is a continuation-in-part of Application Serial No. 13/920,777, which was filed June 18, 2013 and is incorporated herein by reference in its entirety.

technical field

The present disclosure relates generally to sensors and, more particularly, to sensors, systems, and methods that compensate for thermal electromotive force (EMF) effects.

Background technique

Sensors can be affected by many different internal and external characteristics that can make the sensor output signal less accurate. One of these properties is the thermo-electromotive force (thermo-EMF), which is related to the effect that temperature can have on the movement of charges in a material. For example, a temperature gradient in a material can affect the flow of charge in a material by pushing charges in a particular direction, much like an applied electric field. This can be amplified in the presence of electric or magnetic fields or concentration gradients. Thermal EMF can also induce temperature-dependent charges in two basic cases: first, non-uniform temperature (ie, temperature gradient) in a homogeneous material; or, uniform temperature in a non-uniform material. For example, the second situation can occur at device contactors, where said voltage is referred to as thermal contact voltage. Both situations are undesirable with respect to sensor operation and output signal accuracy.

There are many different ways in which temperature can affect charge, only some of which are related to thermal EMF. For example, magnetic sensitivity and resistivity changes due to temperature in Hall effect devices are generally not related to any thermal EMF effects, and thus may not be addressed or compensated for by the embodiments discussed herein. However, particularly when the sensor is operated according to a spinning current or voltage scheme, the sensor output signal can be affected by thermal EMF. In one example, the sensor system includes a Hall plate that operates in sequential phases of operation. Different terminals of the Hall plate are tapped in each operating phase as power supply and output terminals, so that the current direction or the spatial distribution of the current is different for each phase. The rotational output signal can be obtained by combining the signals from the individual operating phases. In the absence of an applied magnetic field, the Hall plate (in fact, usually a magnetic field sensor) can experience an offset error resulting in the output signal. Offset errors in each operating phase can be largely eliminated in the rotation scheme due to the combination of individual operating phase signals, so that little or no residual offset remains in the combined output signal in the output signal.

Unfortunately, residual offset errors often exist, so some rotation scheme sensor systems provide residual offset compensation. Referring to Figure 1, such systems typically include a temperature sensor placed close to the Hall plate, since the offset correction is typically not invariant over temperature. Thus, the system is able to sense temperature, determine a compensation signal based on the temperature, and take this compensation signal into account in the rotation output signal. Therefore, this conventional scheme combines the stage temperature signals only by averaging, which can be considered equivalent to implicit low-pass filtering of slow temperature sensors. However, the challenge is to determine the compensation signal. Since the residual offset of the rotary Hall scheme is random, it depends on the actual individual device and the temperature of this device, and it can vary during the operational lifetime of the device. Thus, even if individual device calibrations can be efficiently and effectively performed during end-of-line testing, variations over the lifetime of the device can degrade the accuracy of the calibration and lead to thermal EMF-related residual offset errors.

Traditional solutions assume that thermal EMF effects are eliminated by using polarity inversion (ie only the polarity of the power supply changes) in sequential operating phases and using rotating voltage rather than current techniques. However, this may not be the case, since in practice the temperature distribution can vary when the polarity of the supply voltage is reversed.

Contents of the invention

According to one embodiment of the invention, the invention provides a sensor device configured to provide a signal in response to a temperature difference between an output contact of a Hall effect device and a reference point, said device comprising: a Hall effect region ; a first contact located near an outer surface of the Hall effect region; a second contact located near a reference point; a first conductor element comprising first and second end portions, the first end portion thermally coupled to the first contact Head and the second end portion is thermally coupled to the second contact head; the second conductor element, including the third and fourth end portion, the third end portion is thermally coupled to the first contact head; the third conductor element, including the fifth and the first Six end portions, the fifth end portion is thermally coupled to the second contact, wherein the first and third end portions are electrically coupled, the second and fifth end portions are electrically coupled, the first, second and third conductors At least two of the conductor elements have substantially different Seebeck coefficients, and the signal is tapped at the fourth and sixth end portions.

According to another embodiment of the present invention, the present invention provides a device configured to provide a signal in response to a temperature difference between a contactor of a Hall effect device and a reference point, the device comprising: a Hall effect region; a contact located near the outer surface of the Hall effect region and comprising at least one first contact; a first conductor element surrounding a substantial portion of the contact and comprising a plurality of second electrically coupled contacts surrounding a substantial portion of the contact; a contactor; a reference point electrically coupled to the first conductor element, wherein the thermal coupling between the reference point and the contact head is weaker than the thermal coupling between the second electrically coupled contactor and the contact head; and a second a conductor element comprising first and second end portions, the first end portion being electrically coupled to the plurality of second electrically coupled contacts, wherein the first and second conductor elements have substantially different Seebeck coefficients, and the reference point and the second end section are configured as tapped signals.

Description of drawings

A more complete understanding of the present disclosure may be obtained by considering the following detailed description of various embodiments of the disclosure when taken in conjunction with the accompanying drawings in which:

Figure 1 is a block diagram of the sensor system.

Figure 2A is a diagram of a Hall plate, according to an embodiment.

2B is a diagram of a vertical Hall sensor device, according to an embodiment.

2C is a diagram of a Hall plate, according to an embodiment.

2D is a perspective view of the potential distribution in the Hall plate during the operating phase, according to an embodiment.

Figure 2E is a coupling arrangement of Hall plates in a first phase of operation according to an embodiment.

Figure 2F is a coupling arrangement of Hall plates in a second phase of operation according to an embodiment.

Figure 2G is a coupling arrangement of Hall plates in a third phase of operation according to an embodiment.

Figure 2H is a coupling arrangement of Hall plates in a fourth phase of operation according to an embodiment.

Figure 2I is a block diagram of a sensor system, according to an embodiment.

Fig. 3 is a side sectional view of an arrangement of first and second sensor devices according to an embodiment.

Figure 4A is a diagram of a sensor coupling arrangement, according to an embodiment.

Figure 4B is a diagram of a sensor coupling arrangement, according to an embodiment.

5A is a block diagram of a sensor system, according to an embodiment.

Figure 5B is a block diagram of a sensor system according to an embodiment.

Figure 6A is a diagram of another coupling arrangement according to an embodiment.

6B is a block diagram of another sensor system according to an embodiment.

Figure 6C is a block diagram of another sensor system according to an embodiment.

Figure 7A is a diagram of a coupling arrangement of Hall plates, according to an embodiment.

Figure 7B is a diagram of a coupling arrangement of Hall plates, according to an embodiment.

Figure 7C is a diagram of a coupling arrangement of a Hall plate system, according to an embodiment.

7D is a diagram of a coupling arrangement of Hall plates in two phases of operation of a spinning current scheme according to an embodiment.

Figure 7E is a diagram of a coupling arrangement of Hall plates in two phases of operation of a spinning current scheme according to an embodiment.

8A is a depiction of temperature distribution in a Hall plate, according to an embodiment.

8B is a depiction of the transient temperature behavior of a Hall plate in a spinning current scheme, according to an embodiment.

Fig. 9 is a side sectional view of an arrangement of a sensor device according to an embodiment.

Figure 10 is a diagram of a sensor arrangement according to an embodiment.

Figure 11 is a diagram of another sensor arrangement according to an embodiment.

Figure 12 is a diagram of another sensor arrangement according to an embodiment.

Figure 13 is a diagram of another sensor arrangement according to an embodiment.

14A and 14B are circuit diagrams of sensor arrangements according to embodiments.

While the present disclosure may have various modifications and alternative forms, details thereof have been shown in the drawings as examples and will be described in detail. It should be understood, however, that the intention is not to limit the disclosure to the particular embodiments described. On the contrary, the intention is to cover all modifications, equivalents, and alternatives falling within the spirit and scope of the present disclosure as defined by the appended claims.

Detailed ways

Embodiments relate to sensor systems and methods capable of compensating for thermal EMF effects that can cause residual offset and other errors in the sensor system. In one embodiment, the sensor system comprises at least one temperature or temperature gradient sensor arranged close to a first sensor element configured to sense a physical quantity, such as a magnetic field , temperature, pressure, force, mechanical stress, or some other physical quantity. For example, in embodiments where the sensor system includes a Hall Effect magnetic field sensing system, the first sensor element includes, for example, a Hall plate, but in other embodiments, other types of magnetic field sensors (and more generally, sensors) can is used as the first sensor. In another embodiment, multiple temperature sensors can be used, where each temperature sensor is arranged close to a different sensor contact or element. In examples where the Hall plate operates according to a rotational operating scheme, the at least one temperature sensor can be configured to sense temperature in each phase of operation, and the individual sensed temperatures can be combined and used to provide a temperature correlation compensation signal.

For any particular material, thermal EMF (a property that the embodiments aim to compensate for) can be quantified by the Seebeck coefficient. For n-doped silicon with a concentration of about 10^16/cm ³ (which, in some embodiments, is typical for the active regions of the Hall effect devices discussed herein), at room temperature, The Seebeck coefficient is about −1,200 μV/°C. N-doped polysilicon with a resistivity of about 7 milliohm*cm has a Seebeck coefficient of about 200 μV/°C, and n-doped polysilicon with a resistivity of about 0.8 milliohm*cm has a Seebeck coefficient of about 80 μV/°C The Seebeck coefficient of . Aluminum, which is often used for metal interconnect lines in integrated circuit technology, has a negligible Seebeck coefficient of only about −0.5 μV/°C. These Seebeck coefficients are representative examples of materials that can be suitable for use in embodiments, but this list is not exhaustive nor limiting as to materials that could or could be used, as understood by those skilled in the art. Additionally, the Seebeck coefficient is not critical, and embodiments relate to reducing or eliminating underlying thermal EMF.

Embodiments relate to the case where an output signal is tapped at a terminal which can withstand small additive voltages caused by thermal EMF. In most cases a signal is tapped between two terminals which are at different temperatures, and thus the tapped signal has a small additive part proportional to the temperature difference between said terminals. In other cases, the signal can be tapped between a terminal and a reference terminal, such as, for example, one terminal at ground potential. Here, again, the terminal and the reference terminal can have different temperatures which can cause a small superimposed additive thermo-emf signal. In other cases, signals can be tapped at the same contactor at different times (eg, in different phases of operation), with multiple sampled values then combined to provide an overall signal. If the temperature at the contactor changes with time, a small additive thermal emf can be superimposed on the signal. It should be remembered that different temperatures at the terminals do not affect the isothermal signal (ie the signal if the terminals are at the same temperature), so the isothermal signal is not multiplied by some factor that depends on the temperature difference. Instead, the signal tapped between terminals at different temperatures is equal to the isothermal signal plus a small additive thermo-EMF contribution proportional to the temperature difference between the terminals. Embodiments aim to address this thermal EMF contribution.

Referring to FIG. 2A , a simplified diagram of a Hall plate 202 is depicted. Certain depictions in Figure 2 are not drawn to scale. The diagram of FIG. 2 and other diagrams that may be used herein are for example and descriptive purposes, and are generally supplemented by this discussion, other figures, and throughout this application. Some diagrams can include block diagrams, where the blocks can represent physical objects, functions, concepts and/or combinations thereof. Furthermore, although some embodiments discussed herein refer to magnetic field sensors including Hall devices and vertical Hall devices, in other embodiments other sensors can be used, and the use of Hall plate 202 in this embodiment and Depiction should not be considered limiting.

Hall plate 202 includes three contacts 204a, 204b and 204c. In an embodiment, the Hall plate 202 can be operated in a rotational scheme wherein the contacts 204a, 204b, 204c can be coupled differently in different phases of operation as power and signal contacts. As previously mentioned, in a spinning current or voltage scheme, the different contacts of the Hall plate 202 are coupled in each sequential phase of operation so that the current direction or spatial distribution of the current in the Hall plate 202 is specific to each phase of operation. rather different. In embodiments, the number of contactors and the number of operating phases can vary.

Although FIG. 2A depicts a single Hall plate 202 , in other embodiments, the sensor can include multiple Hall plates 202 that are similarly and differentially coupled to each other during different phases of operation. FIG. 2B depicts a cross-section of Hall device 212, where the top of device 212 is equal to the top of the semiconductor substrate—in contrast, FIG. 2A is a plan view of the major surface of the semiconductor substrate. 2B depicts one example of a vertical Hall device 212, which, as understood by those skilled in the art, includes three contacts 214a, 214b, and 214c and is capable of operating in a manner similar to Hall plate 202, except with The vertical Hall device 212 is sensitive to different magnetic field components than the Hall plate 202 . FIG. 2C is also a plan view and depicts an exemplary octagonal Hall plate 202 including four contacts or contact diffusions 204a, 204b, 204c, 204d. As previously mentioned, the size, shape, configuration and number of contacts can vary in embodiments, among other characteristics, and the Hall plate 202 will generally be used herein to refer to Embodiments or other Hall plates depicted, the vertical Hall device 212 of FIG. 2B , or other specific characteristics of the Hall plates, which can vary for each embodiment, without limitation. Hall plate 202 includes active region 226 and a boundary or isolation of active region 226 at perimeter 227 . FIG. 2D depicts an exemplary potential distribution within the Hall plate 202 in operation. The potential distribution within the Hall plate 202 results in a temperature distribution or a spatial distribution of temperatures within the Hall plate 202 . The effect of thermal EMF can affect the potential distribution within the Hall plate 202 and thus also the spatial distribution of the temperature within the Hall plate 202 .

In one embodiment, at least one temperature sensor can be arranged proximate to the Hall plate 202 . For example, the Hall plate 202 can be disposed on a semiconductor substrate, and the temperature sensor can be disposed physically close to one or more of the contacts 204a, 204b, 204c (typically, close to the Hall plate 202) or close to the Hall plate Another relevant part of 202. In one embodiment, a single temperature sensor is used and is configured to sense the Hall plate 202 associated with the different power dissipation that can occur in the Hall plate 202 during different operating phases of the rotation scheme. in the temperature. From these measurements, the system 100 is able to estimate thermal EMF changes or fluctuations for each phase. In another embodiment, the temperature sensor includes a plurality of temperature sensor elements, such as a temperature sensor element associated with each contactor of the Hall plate 202, a temperature sensor element located between each adjacent contactor of the Hall plate 202. A sensor element and/or a temperature sensor element comprising a plurality of wires or terminals coupled to or with the Hall plate 202 such that the spatial distribution of temperature in the Hall plate 202 can be sensed. However, in essence the at least one temperature sensor is intended to sense and measure the thermal EMF mismatch between different contacts in the Hall plate 202 (eg two contacts used in differential measurements).

In contrast to differential outputs, temperature sensors comprising a single output voltage or current can also be used in embodiments. In these embodiments, the total thermal EMF contributes to the output signal at the output contactor. An example could be the vertical Hall device 212 of Figure 2B, where current flows between two contactors and the output signal is the potential measured at the third contactor. Thus, regardless of whether the particular sensor implemented operates according to a single phase or according to a multi-phase spinning current or voltage scheme, the thermal EMF at each output or signal terminal in each phase of operation is related to the compensation effort. In other embodiments, the vertical Hall device 212 can also be used in a rotation scheme.

In an embodiment, the temperature sensor can be sampled during the sampling of the Hall plate 202 so that the instantaneous temperature can be obtained. In one example, the operating phase of the sensor device including the Hall plate 202 is at least about 1 microsecond (μs) long, such as about 2 μs, and requires a temperature that can keep pace and sample the signal at appropriate time intervals. sensor. Thus, in an embodiment, the temperature sensor is a temperature sensor with low thermal mass and thermal retention. In an embodiment, the temperature sensor can also have a high bandwidth. For example, if the Hall sensor 202 is operating at 500 kHz with a spinning current scheme, a suitable bandwidth can be about 1 MHz. In an embodiment, it is also beneficial that the temperature sensor has a high spatial resolution in order to detect the spatial gradient of the temperature in the Hall plate 202 . In an embodiment, the temperature sensor has a spatial resolution of at least about 10 micrometers (μm), such as about 1 μm. In addition, however, in embodiments, the temperature sensor can be relatively simple to avoid adding complexity and expense to the overall sensor system, given this and the underlying goal of thermal EMF compensation can be an additional feature in various sensor systems and implementations. The temperature sensor should also not introduce or increase parasitic effects such as high temperature leakage current or stray capacitance, especially those parasitic effects that can affect Hall sensors. Thus, temperature sensors that require too much power, very complex signal processing, large areas, or other features that can increase the cost of the temperature sensor or the demands on resources within the sensor system may not be suitable in most embodiments. suitable, but still capable of applicability in some specialized applications.

Regardless of whether the temperature sensor includes a single temperature sensor element or multiple temperature sensor elements, and whether the temperature sensor is implemented according to the embodiment of FIG. 3 , the embodiment of FIG. 4 , or according to some other embodiment discussed herein below, the temperature The sensor is capable of operating in every phase of operation of the rotation scheme of the Hall plate 202 or another sensor device. In other embodiments, the temperature sensor need not operate in every phase. Through the same or different circuitry (on-chip or off-chip), the various individual phase temperature signals from the temperature sensors can be combined, as can the individual phase signals from the Hall plate 202 . A temperature signal and a rotational output signal can be determined, wherein the temperature signal is used to determine a temperature dependent offset correction signal which can be combined with the rotational output signal to provide an overall system output signal.

For example, referring to FIG. 21 , a spatial gradient temperature sensor 230 can be thermally disposed adjacent to another sensor element 232 on a substrate 234, which in various embodiments can comprise a rotary Hall sensor or some other sensor element 232. other sensors. Sensor 230 is configured to sense a spatial gradient of temperature within sensor element 232 during at least a portion of at least one phase of operation. For example, in one embodiment, sensor 232 is capable of operating in multiple phases of operation according to a rotation scheme. Circuitry 236 , which in an embodiment can be disposed on or external to substrate 234 , is coupled to sensors 230 and 232 and can be configured to sense in combination with at least one phase of operation in each of at least one phase of operation. and combining signals from sensor 230 related to the sensed temperature spatial gradient in sensor 232 sampled during the same portion of the at least one operating phase. Those signals can then be combined to obtain an overall output signal indicative of a physical quantity. Using the combined signal from the sensor 230, an offset correction can be determined and used in combination to obtain an overall output signal, thereby correcting the overall output signal for thermal EMF-related offsets.

Turning to a specific exemplary embodiment of a temperature and/or spatial gradient temperature sensor, in one exemplary implementation, the sensor includes a pn junction. A pn junction is often used to measure temperature on a semiconductor substrate, and as such, can be suitable in various embodiments. The pn junction also has the advantage that it is relatively easy to arrange close to a contactor such as the Hall plate 202 . One exemplary embodiment of a temperature sensor 104 including a pn junction is depicted in FIG. 3 . The temperature sensor 104 is arranged close to the Hall plate 202 (in particular, the contactor 204a ), separated by a distance d on the substrate 302 with optional galvanic isolation. In other embodiments, the temperature sensor 104 and the Hall plate 202 can be electrically coupled, but this will affect the performance of one or both of the Hall plate 202 and the sensor 104, and thus may not be true in all implementations. suitable.

The temperature sensor 104 includes an n-tub 304 and a p-tub 306 , but these can be reversed in other embodiments and biased by a constant current source 308 . The voltage on the current source 308 is a strong function of temperature and, therefore, can be used as a temperature sensor. In various embodiments, additional components not specifically depicted, including signal conditioning circuit components such as precision amplifiers and analog-to-digital converters (ADCs), can be included.

An alternative implementation of the temperature sensor depicted in FIG. 3 is depicted in FIG. 4 . Here, temperature sensing is integrated with Hall or other sensors or types of devices to sense temperature and/or spatial gradients of temperature in Hall plate 202 by providing additional contacts, terminals, wires and/or other elements. The temperature sensing of the embodiment of FIG. 4 can be simpler than the temperature sensing of the embodiment of FIG. 3 and can determine the temperature or temperature gradient at the contacts of the Hall plate 202 or at the The temperature or temperature gradient within the contacts of the Hall plate 202 .

In FIG. 4A, two terminals 204a and 204b of Hall plate 202 are shown (eg, as depicted in FIG. some other configuration). In embodiments, terminals 204a and 204b can comprise wires, wires, or some other suitable configuration. Typically, a terminal or lead can be a device, circuit or circuit part that has a resistance approximately ten times smaller than the signal output resistance of the sensor device at the corresponding contactor at which the terminal or lead taps off. resistance, but this can vary in embodiments. In embodiments, it can be beneficial to have wires or terminals with low resistance and low parasitics. Terminals 204a and 204b and terminals 410a and 410b are generally at the same temperature T'. Although not depicted in FIG. 4A , terminals 204a and 204b can be coupled (eg, via a tungsten plug or stack of tungsten plugs) to a pair of transistors in a signal amplifier or other circuit that is also arranged isothermally.

The circuit and components of FIG. 4A can be arranged relative to one or more contacts of a Hall plate or other device, and in one embodiment, the Hall plate can include a low-doped n-region, or in other embodiments, The Hall plate can include some other configuration or composition. Each of the terminals 204 a and 204 b is coupled through a plug 402 to an interconnect line 404 a and 404 b , respectively, which can be arranged between intermetal oxide layers in or on the substrate 400 . In an embodiment, the plug 402 can include a tungsten-filled hole that has been etched or otherwise formed in the intermetallic oxide layer of the substrate 400, but in other embodiments, other tungsten-filled holes can be used. Materials, configurations and methods of formation. Interconnects 404a and 404b can comprise wires, wires, or other suitable structures, and couple the sensor's contacts 204a and 204b to contact diffusions 406a and 406b, respectively, via various additional plugs 403a, 403b, as described in the embodiment The various further plugs 403a, 403b can be the same or different from each other and/or the same or different from the plug 402 . In one embodiment, contact diffusions 406a and 406b include shallow n+ S/D contact diffusions formed in regions 408a and 408b. Each of contact diffusions 406a and 406b is coupled to terminals 410a and 410b via plugs 403a and 403b, respectively, and each contact diffusion can include an output contactor of the sensor for providing a signal related to a physical quantity sensed by the sensor. Or the reference point that will be used to sense the temperature difference between two points.

In an embodiment, the terminals 410a and 410b can comprise wires, metal wires, or other suitable materials or structures. In general, elements 204a, 204b, 404a, 404b, 410a, and 410b depicted in FIG. 4A can each be considered to be wires, terminals, interconnects, or some other structure (including in different are represented and are exemplary only), although different terminology may be used herein for purposes of illustration and description (eg, interconnect lines versus terminals and/or wires). In an embodiment it can be more interesting: especially when considered relative to each other and for where (ie between which terminals and/or wires and/or between contactors) to tap the signals or The Seebeck coefficients and/or materials that those elements 204a, 204b, 404a, 404b, 410a, and 410b comprise when otherwise considered in various embodiments.

For example, in an embodiment, interconnect lines 404a and 404b comprise different materials and/or have different Seebeck coefficients than terminals 410a, 410b. In one embodiment, interconnect lines 404a and 404b comprise a semiconductor material, such as polysilicon, silicon, germanium, single crystal semiconductor material, or polycrystalline semiconductor material, but in other embodiments other materials can be used to communicate with terminals 410a, The material of interconnect lines 404a and 404b has a different contact voltage than the material of interconnect 410b. In another embodiment, the materials can be chosen to maximize the difference in Seebeck coefficients between them. Thus, there can be differences in thermal EMF between materials that can be sensed and used to measure thermal EMF at terminals 410a and 410b and at terminals 204a and 204b. Throughout this document, the example in which interconnects 404a and 404b comprise polysilicon and terminals 410a and 410b comprise metal will generally be used, but this example should not be limiting for other embodiments. Thus, a potential can be tapped at the ends of terminals 410a and 410b, but also at terminals 204a and 204b, so that terminals 204a and 204b can be used to measure a differential signal, which can be compared with Compared to the differential signal at 410b, the two differential signals should differ from each other by the thermal contact voltage between the polysilicon of interconnect lines 404a and 404b and the metal of interconnect lines 410a and 410b.

In other words, the signal measured between terminals 410a and 410b can include some unknown thermal EMF contribution caused at least in part by the different temperatures T1, T2 of contacts 406a, 406b, respectively, and in addition to that from polysilicon interconnect line 404a In addition to the contribution from and 404b, the signal measured between terminals 204a and 204b can include the same unknown thermal EMF contribution. There is a strong correlation between the additional polysilicon contribution and the unknown thermal EMF contribution since they are proportional to the temperature difference T1 − T2 of the contactors 406a and 406b. Therefore, the following formula can be used:

V(C1′)−V(C2′)=F[B]+Off′ and V(C1″)−V(C2″)=F[B]+Off″

where F[B] is a function of the magnetic field, where F[0]=0, and Off″=Off′+k(T1−T2), where k is the plug of material of terminals 410a and 410b and interconnect lines 404a and 404b Differences in Beck's coefficients. In one phase of operation, C1' represents terminal 410a, C2' represents terminal 410b, C1" represents terminal 204a, and C2" represents terminal 204b, but the coupling arrangement will vary from one phase of operation to the next Changes in the operating phase. Specifically refer to Figure 4:

V(204a)−V(204b)=k(404a)*(T1−T′)+k(408)*(T2−T1)+V(408ab)+k(404b)*(T′−T2)

V(410a)−V(410b)=k(410a)*(T1−T′)+k(408)*(T2−T1)+V(408ab)+k(410b)*(T′−T2)

→V(204a)−V(204b)−(V(410a)−V(410b))==(k(404a)−k(410a))*(T1−T′)+(k(404b)−k (410b))*(T′−T2)

It is assumed here that the conductive regions 408a and 408b have an effective Seebeck coefficient k(408), and that at isothermal conditions (T1=T2), a voltage V(408ab) (e.g., the output voltage of a Hall effect device) exists at, for example, a contact diffusion Between regions 408a and 408b tapped at 406a and 406b. This assumes that the difference in effective Seebeck coefficient between wire or interconnect 404a and wire or terminal 410a is the same as the difference in effective Seebeck coefficient between interconnect 404b and wire or terminal 410b. In an embodiment, this can be addressed by interconnects 404a and 404b comprising the same material and wires or terminals 410a and 410b comprising the same material, as discussed herein above. Note that it is the "effective" Seebeck coefficient of interest, since the individual Seebeck coefficients of a part or material can be the same. The effective Seebeck coefficients of other parts or parts of the sensor system are generally irrelevant as long as the parts or parts are substantially at a uniform temperature. For example, in FIG. 4A , the effective Seebeck coefficients for plugs 402 comprising tungsten are irrelevant because there is no spatial temperature gradient within each plug 402 . Thus, there is no difference if terminal 410a is coupled to contact diffusion 406a directly or via a single tungsten plug or via a stack of two or more tungsten plugs, as long as each tungsten plug does not have a significant temperature difference within its body. Thus, it would also be possible in FIG. 4A to have a first tungsten plug between contactor 406a and wire 404a and a stack of two further tungsten plugs between contactor 406a and terminal 410a, thereby There is no direct contact between the two other tungsten plugs of the stack and the wire 404a. In practice, a large number of tungsten plugs may be used between contact diffusion 406a and wire 404a or wire 410a, all electrically connected in parallel. Since the contact diffusions have a finite size, the space for tungsten plugs is limited there, and thus one generally tends to stack layers, each with a large number of plugs connected in parallel, as exemplified for a single plug per layer in Figure 4A shown. therefore,

V(204a)−V(204b)−(V(410a)−V(410b))=(k(404a)−k(410a))*(T1−T2)

This enables measurement of the temperature difference T1−T2 if the Seebeck coefficients as described above are different (eg, between wire or interconnect 404a and wire or terminal 410a and between interconnect 404b and wire or terminal 410b).

According to Off′=Off+Off _therm , the offset of the first sensor (eg, Hall effect device) includes the original offset Off and the thermoelectric contribution Off _therm . The thermoelectric contribution Off _therm has a strong correlation with the temperature difference T1−T2.

While this is valid for a single phase of operation, in the spinning scheme the Hall plate 202 operates in several phases and the total signal is calculated as the sum of all individual phases:

S'=Σ(V(C1')−V(C2')=Σ(F[B]+Off')=Σ(F[B]+Off+Off _therm ) measured at terminals 410a and 410b

as well as

S″=Σ(V(C1″)−V(C2″)=Σ(F[B]+Off″)=Σ(F[B]+Off+k(T1−T2) measured at terminals 204a and 204b +Off _therm ), where the index varies according to the phase of operation. The difference can then be calculated by the sensor system:

S″−S′=Σk(T1−T2).

Due to the strong correlation, this difference can be used to estimate the residual offset ΣOff _therm . One method used to obtain this estimate is to multiply by a predefined factor x:

ΣOff _therm ≅xΣk(T1−T2)

This factor x can depend on the technology and geometry of the Hall sensor and on the temperature, operating frequency, electrical bias and any mechanical stress acting on the sensor. In an embodiment, x can be determined based on characteristics of the sensor device in a laboratory or manufacturing level testing. Once ΣOff _therm is determined, ΣOff _therm can be subtracted from S' to obtain a signal without offset due to thermal EMF. Due to the rotating Hall scheme, the purely resistive offset (i.e., the offset that can be described by an asymmetric Wheatstone bridge circuit in the equivalent circuit diagram) disappears, ΣOff=0, so that S′−ΣOff _therm =S′− x(S″−S′)==(1+x)S′−xS″ without any offset.

Thus, a simple method for measuring the temperature difference between contact diffusions 406a and 406b can be as follows, where T1 is the temperature at contact diffusion 406a and T2 is the temperature at contact diffusion 406b:

T1−T2=(1/k)*(V(204a)−V(204b)−(V(410a)−V(410b)))

where k is the difference in Seebeck coefficients of metal lines (eg, terminals 410a and 410b ) and polysilicon lines (eg, interconnects 404a and 404b ). In an embodiment, this can be accomplished efficiently because V(410a)−V(410b) and V(204a)−V(204b) are already measured by the sensor's signal conditioning circuitry. Therefore, in an embodiment, no dedicated hardware such as preamplifiers and ADCs are required to measure T1−T2. In another embodiment, T1−T' can be obtained by determining V(410a)−V(204a), and T2−T′ can be obtained by V(410b)−V(204b), but the sensor does not need A dedicated preamplifier may be necessary.

Accordingly, the sensor system can be depicted as system 500 in FIG. 5A , system 500 including Hall plate 202 . As described elsewhere herein, the Hall plate 202 of FIG. 5A or any other figure (e.g., FIG. 6, FIG. 2A or 2C), have some other configuration not explicitly depicted, or include a vertical Hall device (eg, Hall device 212 of FIG. 2B or have some other configuration not explicitly depicted here). Tapping the output signal at contact diffusions 406a and 406b by elements forming pairs of two different materials with different Seebeck coefficients (eg, terminals 204a, 204b, 410a, 410b and interconnect lines 404a and 404b) as discussed above, A first set of phase signals (eg comprising, eg, metal terminals 410a and 410b) and a second set of phase signals (eg, comprising eg interconnection lines 404a and 404b of semiconductor material and tapped via terminals 204a and 204b) are obtained. The signals from all stages are then combined at circuit blocks 506 and 508 and the difference of the combined signals is determined at block 510 . In an embodiment, blocks 506 and 508 can be combined and time multiplexed, with metal or polysilicon phase signals sampled at any particular phase and then stored in memory until combined at block 510 . The output of block 510 is a measurement of temperature asymmetry in the sensor and/or temperature fluctuations of the contactors (eg, contactors 204a, 204b) during various operating phases. It may be equal to S″−S′=Σk(T1−T2). In an embodiment, this should be related to the remaining offset.

This difference (output of block 510 ) is then used at block 512 to estimate the remaining offset. Its output can be equal to xΣk(T1−T2). This estimated residual offset is then subtracted at block 514 from the rotated output signal from block 508 to obtain a residual offset with a significant reduction or removal, e.g. the overall output signal.

In fact, there are several different approaches for implementing a system 500, such as system 100 of FIG. 1, which is a conceptual or generalized description of a system and its operation according to an embodiment. In one embodiment, a first amplifier can be used for a first set of (eg, metal) phase signals and a second amplifier can be used for a second set of (eg, polysilicon) phase signals. In another embodiment, the first and second sets of phase signals can be multiplexed such that in a first rotation scheme, the first set of phase signals are amplified and processed, and in a second rotation scheme, the second set of phase The signal is amplified by the same amplifier(s). This second embodiment can be implemented more economically, but can be limited by the bandwidth associated with the change of the magnetic field between the first and second rotation schemes. However, at low bandwidths, this embodiment can be more accurate because any amplifier errors are canceled out when the two rotated output signals are combined. Another embodiment of system 500 is depicted in FIG. 5B , where the combined second set of phase signals (ie, the output of block 506 ) is also used to determine the overall output signal at block 512 .

Returning to FIG. 4 , FIG. 4B depicts one half of the system of FIG. 4A with the following coupling modifications: Instead of connecting the two wires 404 a , 410 a to the contact diffusion 406 a of the first sensor device 202 directly or via a tungsten plug 403 a, the contact diffusion There can also be a short wire of length d1 between 406a and the point where the two wires are connected together. If d1 is much smaller than d2, the temperature T11 can be much closer to the temperature T1 than to the temperature T'. Thus the differential input pair of evaluation circuit 499 (eg, a preamplifier in an embodiment) measures a signal proportional to the temperature difference T'−T11, which is close to the temperature difference T'−T1. T′−T1=x′*(T′−T11) holds. The same can be achieved for the second part of the system marked with "b" in Figure 4A such that T'−T2=x'*(T'−T22) holds, whereby T22 is the temperature at which wires 410b and 404b are shorted . If this wiring is used in conjunction with FIG. 5A, the output of block 512 will become xΣk(T1−T2)=xx′Σk(T11−T22). This shows that for offset compensation, a change in the temperature difference T1−T2 to T11−T22 means only a small modification of the factor x to xx′. If the distance d1 is not small relative to d2, this will lead to a significant deterioration in the measurement of temperature differences, because (i) these differences decrease, and (ii) the thermal influence of other parts of the system on T11, T22 rises, so that Finally, they are no longer primarily determined by the contact temperatures T1, T2.

In another embodiment, and referring to FIG. 6A , the voltage is measured between the terminals coupled to the first contactor, and then the voltage is measured between the terminals coupled to the second contactor. In other words, and referring to Figure 6A:

V(204a)−V(410a)=k(404a)*(T1−T′)+k(410a)*(T′−T1)

V(204b)−V(410b)=k(404b)*(T2−T)+k(410b)*(T′−T2)

If the Seebeck coefficients of interconnects 404a and 404b are the same and the Seebeck coefficients of wires or terminals 410a and 410b are the same, then:

V(204a)−V(410a)−(V(204b)−V(410b))=k(404)*(T1−T2)+k(410)*(T2−T1)=(k(404)− k(410))*(T1−T2)

In FIG. 6A, regions 408a and 408b are part of different heads (such as two contacts), each in each of the first and second heads of the Hall device, but in other embodiments In , regions 408a and 408b can be located in different devices or can include a portion of something other than a head within one or different devices. In an embodiment, circuit 401 couples two contacts 405a and 405b of regions 408a and 408b. Then:

(204a)−V(204b)=k(404a)*(T1−T′)+k(408a)*(T1′−T1)+V(1′1)+V(2′1′)+V( 22′)+k(408b)*(T2−T2′)+k(404b)*(T′−T2)

V(410a)−V(410b)=k(410a)*(T1−T′)+k(408a)*(T1′−T1)+V(1′1)+V(2′1′)+V (22′)+k(408b)*(T2−T2′)+k(410b)*(T′−T2)

→V(204a)−V(204b)−(V(410a)−V(410b))==(k(404a)−k(410a))*(T1−T′)+(k(404b)-k (410b))*(T′−T2)

Thus, the voltage drop across circuit 401 is represented by V(2'1'), and the voltage drop between 405a, 406a is represented by V(1'1), and the voltage drop between 406b, 405b is represented by V(22 ')express. For k(404a)−k(410a)=k(404b)−k(410b), we again obtain V(204a)−V(204b)−(V(410a)− V(410b), so that the temperature difference can be measured by the signal difference at terminals 410a, 410b and 204a, 204b.

Referring also to FIG. 6B, another embodiment of the system 500 includes two multiplication factors K1 and K2 that can be selected to achieve the same calculations as those illustrated in and discussed with reference to FIG. 5B calculation. Then, in a simplified depiction, and with reference to FIG. 6C , block 508 can perform in a single computational circuit block 508 one or more methods depicted at least conceptually as separate in FIGS. 5A , 5B, or 6B.

Returning briefly to FIG. 2 and the exemplary Hall device depicted therein, FIGS. 2E , 2F, 2G and 2H depict four stages of operation of a Hall plate 202 with four contacts 1 , 2 , 3 and 4 . The Hall plate 202 has 90 degree symmetry and is depicted as a simple square with contacts 1-4 arranged at the four corners, but this can vary in embodiments. In a rotating voltage scheme with four phases 1-4, the supply voltage Vs is coupled to a contactor 1-4 with the same phase number, and the contactor opposite this is coupled to ground potential. The two remaining contactors are shorted so that the output signal is equal to the current flowing between them. For example, in FIG. 2E , Phase 1 is depicted in which contactor 1 is coupled to Vs and contactor 3 is coupled to ground potential. The output signal is measured between contactors 2 and 4. The output current can thus be:

I24,1=F1[B]+Off1+k(T2,1−T4,1)+Off1,therm

Rotating one contactor clockwise in the coupling arrangement of Figure 2E in the next three stages provides the following output signals:

I31,2=F2[B]+Off2+k(T3,2−T1,2)+Off2,therm

I42,3=F3[B]+Off3+k(T4,3−T2,3)+Off3,therm

I13,4=F4[B]+Off4+k(T1,4−T3,4)+Off4,therm

The first element of each equation (eg, F2[B]) represents the magnetic field dependence, which is assumed to be different in each phase but need not be different in each phase. For example, Off1 represents a resistive offset term of the first phase which can depend on the applied potential and is fully defined by an equivalent circuit diagram, eg in the form of an asymmetric Wheatstone bridge circuit. For example, the term k(T2,1−T4,1) represents the thermal EMF caused by thermally coupled contacts, which can include aluminum or polysilicon interconnects (see eg FIG. 4 ). The last term Off1,therm represents the thermal EMF that occurs inside the active region of the Hall plate 202 due to non-uniform temperature and/or non-uniform doping gradient.

The sum of the currents tapped (and k≈0) at the metal lines (eg, terminals 410a and 410b in FIG. 4 ) is:

And the sum of the currents tapped on the polysilicon lines (eg, interconnect lines 404a and 404b) is:

I _p =I _m +k(T _2,1 −T _4,1 +T _3,2 −T _1,2 +T _4,3 −T _2,3 +T _1,4 −T _3,4 )

Thus, referring to, for example, system 500 of FIG. 5A , the following formula can be determined at block 510:

I _p −I _m =k(T _2,1 −T _4,1 +T _3,2 −T _1,2 +T _4,3 −T _2,3 +T _1,4 −T _3,4 )

And it can be used as an input to block 512, which estimates the residual offset of the rotating voltage Hall plate due to the thermal voltage, since at and (T _2,1 −T _4,1 +T _3,2 −T _{1 ,2} +T _4,3 −T _2,3 +T _1,4 −T _3,4 ) is strongly correlated since the latter is the cause of the former. In this example the common mode potential of the output signal is floating, but in other embodiments could be tied to some predefined potential. These methods can also be applied to embodiments such as those disclosed in commonly owned U.S. Patent Application Serial Nos. 13/022,844 and 13/488,709 (which are incorporated herein by reference in their entirety): Hall contactors are used As a force sensing contactor, where the voltage or current at the force contactor is adjusted until the voltage or current respectively at the sensing contactor is at a certain predefined value. These contactors can be handled in the same manner as the various contactors discussed here, so that temperature sensors can be used to measure the temperature at each sensing contactor, or each sensing contactor is coupled to the Metal lines and polysilicon interconnect lines discussed with reference to FIG. 4 .

These and other embodiments can also include additional features, elements, functions and concepts. For example, a system as discussed herein can also include a heating element coupled to the contactor of the sensor to control the temperature of the contactor based on measurements performed by the temperature sensor, temperature gradient sensor, or temperature sensing circuit. Additionally or alternatively, in various embodiments, the ground reference can be adjusted in order to affect the non-linear current-voltage characteristics of the device and thus control the temperature of one or more contactors. Considering that the resistance of the Hall element generally increases when the reverse bias voltage of the Hall element's surrounding environment (e.g., the surrounding head, the substrate, or the shallow head as the top plate) increases, this effect can be exploited to control the Hall The power dissipation or the spatial distribution of the power dissipation of the contactors of the plate and thus the temperature distribution of the contactors of the Hall plate is controlled. In various embodiments, circumferential isolation structures or elements such as pn rings or trenches can also be used to implement such features. The control loop can be formed in such a way that the power dissipation or the spatial distribution of the power dissipation in the Hall plate is adjusted until the temperature difference signal of the temperature gradient sensor is minimized. Thereby, the adjustment can be frozen during a full rotation cycle, or it can be adjusted between operating phases within a full rotation cycle.

Another embodiment of Hall plate 202 is depicted in FIG. 7A, and another embodiment of Hall plate 202 includes four contactors C1, C2, C3 and C4 and four temperature sensors, described in this embodiment The four temperature sensors include diodes D1, D2, D3 and D4, but in other embodiments the four temperature sensors can include other devices or structures. Contactors C1-C4 include contact diffusions and are labeled in the same order as in Figures 2E-2H herein above. Generally, the same or similar reference numbers will be used throughout herein to refer to the same or similar elements, parts, structures or other features in the various drawings. In Figure 7, the top plate of the Hall effect device is optional and not depicted.

Terminals t1, t2, t3 and t4 are coupled to each contactor, ie t1 is coupled to C1, t2 is coupled to C2, t3 is coupled to C3, and t4 is coupled to C4. Diodes D1-D4 are coupled to each contactor, that is, D1 is coupled to C1, D2 is coupled to C2, D3 is coupled to C3, and D4 is coupled to C4; and temperature terminals tt1, tt2, tt3, and tt4 are coupled to each diode D1-D4, ie tt1 is coupled to D1, tt2 is coupled to D2, tt3 is coupled to D3, and tt4 is coupled to D4. In an embodiment, each diode D1-D4 is arranged in close thermal contact with its corresponding contact diffusion C1-C4.

As in other embodiments, a spinning current scheme can be implemented such that in the first operating phase of the spinning current cycle, current is supplied to Hall plate 202 at power contacts C1, C3 and the signal is tapped at signal terminals C2, C4. . More precisely, the supply current IsH is injected into the terminal t1 and flows into the contactor C1, while the second terminal t3 is tied to a reference potential VsL (such as ground potential or some other suitably chosen potential), And a first output voltage is measured at terminals t2 and t4. In one embodiment, currents IT2 and IT4 are drawn from terminals tt2 and tt4 such that:

|IT2|+|IT4|<|IsH|

In an embodiment, IsH is about 10 times to about 100 times IT2, and IT2 is equal to IT4. If current IT2 flows through diode D2, a voltage drop occurs across D2. The same situation exists for diode D4. Therefore, the voltage across tt4−tt2 is equal to the voltage across t4−t2 plus the difference between the voltages across temperature devices D2 and D4:

V(t4)−V(t2)=V(t4)−V(tt4)+V(tt4)−V(tt2)+V(tt2)−V(t2)=V(D4)+V(tt4)− V(tt2)−V(D2)

therefore:

V(tt4)−V(tt2)=V(t4)−V(t2)+V(D2)−V(D4),

Thus, voltage V(D2) is considered positive if the anode of diode D2 is positive with respect to the cathode, and the same applies to diode D4. According to one embodiment, the first temperature output voltage is then measured on terminals tt2 and tt4.

The temperature sensors depicted in FIG. 7A and described as diodes but as previously described can be varied in other embodiments are chosen such that the voltage across each temperature sensor is a strong function of temperature. In an embodiment, since diodes are known to respond to temperature changes by approximately −2 mV/°C., they can be suitable. However, in other embodiments, simple resistors can be used, such as those with a large temperature coefficient of resistance. Low-doped tips are common in integrated circuit technology and have a temperature coefficient of about 5000 ppm/°C.; then, for a voltage drop of about 1V across the resistor, 1V*5000 ppm/°C can be achieved. =5 mV/°C. Sensitivity of the temperature signal. However, a disadvantage of resistors as temperature devices is their electrical resistance, which increases the internal resistance of the Hall plate 202 and increases noise. In contrast, diodes have a much smaller internal resistance, which does not add much noise to the sensor signal. On the other hand, the resistance can also be realized in a layer above the silicon substrate and thus above the head of the Hall effect device. For example, polysilicon resistors R1 , R2 , R3 and R4 placed over respective contacts C1 - C4 can be used, and an example of such a configuration is depicted in FIG. 7B . In general, the temperature device can also be any bipolar circuit with a temperature-dependent voltage and, in an embodiment, a low internal resistance. In particular, this circuit can employ a feedback loop to reduce the resistance seen by the output signal of the Hall effect device.

FIG. 7C depicts a circuit diagram illustrating one way in which the spinning current Hall probe 202 is connected to preamplifiers A1 and A2 in a first phase of operation. The switches S1 , S2 , S3 and S4 are configured to connect any of the terminals t1 - t4 with any of the current source IsH, the reference voltage source VsL and/or the input of the amplifier A1 . Similarly, switches ST1-ST4 are configured to connect any of terminals tt1-tt4 with any input of amplifier A2. Current sources IT1-IT4 are configured to be switched on or off during any phase of operation, whereby the shading of IT1 and IT3 is intended to indicate that IT1 and IT3 may be in the first phase of operation (depicted in this figure and in FIG. 7A operating phase) is disconnected. Alternatively, all current sources IT1-IT4 may be switched on during all operating phases, which may minimize errors due to transient effects and/or self-heating. Note that the signs of the currents IT1-IT4 can be positive or negative, which means that these currents are drawn from the Hall device 202 (positive) or injected into the Hall device 202 (negative). This sign changes the common-mode potential at the input of the amplifiers A1, A2 and can be chosen appropriately. The amplifier A1 subtracts the two output signals at t2 and t4 during the first operating phase and provides the phase signal P1 at its output. Amplifier A2 subtracts the two temperature output signals at tt2 and tt4 during the first operating phase and provides at its output the phase temperature signal PT1. In one embodiment, A2 can be the same as Al if Al operates in a time-multiplexed manner.

The voltage across the temperature device (e.g., diodes D1-D4) can vary linearly with temperature (at least to a first order approximation):

V(D2)=VT20*(1+ST2*T2)

V(D4)=VT40*(1+ST4*T4)

Where T2 and T4 are the temperatures at contactors C2 and C4, ST2 and ST4 are the temperature sensitivities, and VT20 and VT40 are the voltages across D2 and C4 at zero temperatures T2 and T4. If the temperature devices are the same:

Then VT20=VT40 and ST2=ST4.

However, in most cases the temperature device has a mismatch:

VT20< >VT40 and ST2< >ST4

therefore:

V(D2)−V(D4)=VT20−VT40+VT20*ST2*T2−VT40*ST4*T4

Even if the temperature is the same, ie T2=T4, the voltage difference across the two temperature devices is usually not zero.

In embodiments where sources IsH and VsL are swapped, the system is able to handle these errors if the system performs a third phase of operation. Thus, in this third operating phase, the current source IsH is connected to terminal t3 and the reference voltage VsL is connected to t1. The temperature devices D1-D4 can still be connected in the same way as in operating phase 1. Then:

V′(D2)−V′(D4)=VT20−VT40+VT20*ST2*T2′−VT40*ST4*T4′

where an apostrophe or "'" indicates this stage of operation. Note that temperatures T2' and T4' differ from T2 and T4 because the Hall devices operate with different current directions, and due to small asymmetries and electrical nonlinearities, this can lead to slightly different temperatures (e.g., in implemented In the example about 0.01°C.). The system calculates the difference of the differential stage temperature signals as follows:

V(D2)−V(D4)−(V′(D2)−V′(D4))=VT20*ST2*(T2−T2′)−VT40*ST4*(T4−T4′)

And relate this difference to the thermoelectric error of the Hall output signal during the two phases of operation:

The first operation stage: V(t4)−V(t2)=S*B+k*(T4−T2)

The third operation stage: V′(t4)−V′(t2)=−S*B+k*(T4′−T2′)

Resistive offset terms are thus ignored since they are canceled over the full spinning current cycle. From the overall spinning current output signal, subtract the two signals from the first and third stages:

V(t4)−V(t2)−(V′(t4)−V′(t2))=2*S*B+k*(T4−T4′−T2+T2′).

If you do the same for the stage temperature signal, then:

V(tt4)−V(tt2)−(V′(tt4)−V′(tt2))=2*S*B+k*(T4−T4′−T2+T2′)+VT20*ST2*(T2 −T2′)−VT40*ST4*(T4−T4′)

Therefore, the phase signal has an error caused by the thermal EMF:

k*(T4−T4′−T2+T2′)

The stage temperature signal has an additional error caused by thermal EMF:

VT20*ST2*(T2−T2′)−VT40*ST4*(T4−T4′)

The system can compare the two (eg, by subtracting them). Therefore, the system is able to measure the following values:

VT20*ST2*(T2−T2′)−VT40*ST4*(T4−T4′).

By characterization in the laboratory, a typical relationship between k*(T4−T4′−T2+T2′) and VT20*ST2*(T2−T2′)−VT40*ST4*(T4−T4′) can be established. This relationship will be different for each device, for each production batch, but it should be stable over the lifetime of a particular device (i.e. as long as VT20*ST2 and VT40*ST4 are stable, which is in practice This is usually true where a stable pn junction or a stable resistor or other stabilizing device is used). This typical relationship can be used in the sensor system's algorithm to estimate the thermo-EMF error in the phase signal and eventually compensate for it (eg by subtracting the expected error from the phase signal).

For example, assume that there is no mismatch between the two temperature devices D2 and D4. Then, the additional error due to the thermal EMF in the stage temperature signal is

V(tt4)−V(tt2)−V′(tt4)+V′(tt2)−V(t4)+V(t2)+V′(t4)−V′(t2)=VT0*ST*(T2 −T2′-T4+T4′)

where we use VT0=VT20=VT40 and ST=ST2=ST4. If this is multiplied by an appropriate factor and added to the original phase signal in the rotation scheme, the thermo-EMF of the Hall effect device can be canceled:

Vcomp=V(t4)−V(t2)−V′(t4)+V′(t2)−x*{V(tt4)−V(tt2)−V′(tt4)+V′(tt2)−V (t4)+V(t2)+V′(t4)−V′(t2)}=2*S*B+k*(T4−T4′−T2+T2′)−x*VT0*ST*(T2 −T2′−T4+T4′)=2*S*B

for

k+x*VT0*ST=0

or

x=−k/(VT0*ST)

We refer to Vcomp as the thermal EMF compensation signal. This factor x can be determined empirically, but an approximation can be determined theoretically: if a diode is used as a temperature device, then

VT0*ST=−2mV/°C.

and k is eg the Seebeck coefficient of a low n-doped Hall region with k=−1500 µV/°C. This gives

x=−(−1.5 mV/°C.)/(−2 mV/°C.)=−0.75

Therefore, the thermal EMF compensation signal in the first operation stage is obtained by (1+x)*{V(t4)−V(t2)}−x*{V(tt4)−V(tt2)},

And in the third stage of operation, it is given by (1+x)*{V′(t4)−V′(t2)}−x*{V′(tt4)−V′(tt2)},

And in the overall rotation scheme, the two signals are subtracted.

Therefore, the thermal EMF compensation signal is a differential phase signal (e.g., V(t4)−V(t2)) and a differential phase temperature signal (e.g., V(tt4)−V(tt2)) (both have the same phase) linear combination of . In the above case, the stage temperature signal is weighted by a factor three times larger than the conventional stage signal: −x/(1+x)=−(−0.75)/(1−0.75)=3. Therefore, the main part of the signal comes from the stage temperature signal, and only a small part comes from the traditional stage signal, which is the difference with respect to the traditional scheme. Of course, the weighting of the differential stage temperature signal to the differential stage signal depends on the temperature sensitivity of the temperature sensor. As seen above, the linear combinations are phase independent: there is the same factor x for the first and third phases. However, assume that devices D2 and D4 are perfectly matched. If other devices D1 and D3 are used in the second and fourth phases of the rotary Hall cycle eg according to Fig. 2E-H and have different VT0 and ST parameters, this will also affect x. Therefore, usually in so-called quadrature operation phases, ie phases with different signal contacts (phases 1 and 2 are quadrature, whereas 1 and 3 are not) the linear combination for the thermoemf compensation signal is different. However, in non-orthogonal phases, such as those obtained by reversing the polarity of the supply voltage or supply current, the linear combination for the thermal EMF compensation signal is usually the same.

A fundamental aspect of various embodiments is that the thermal symmetry of the Hall effect device for power reversal can be optimized. If this is realized for all system effects (eg, geometric symmetry), there will always be some statistical asymmetry (eg, geometric asymmetry due to manufacturing tolerances), which should be present in the absence of Manufacturing tolerances cause temperature differences between contactors having the same temperature. These temperature differences can be measured by temperature difference sensors (i.e., sensors of the spatial gradient of temperature), and from these measurements, correction values are deduced online, which cancel the offset caused by the thermal emf, if these values are added to the Hall effect signal (rather than multiplying the Hall Effect signal). This feature of adding an offset correction value to an uncompensated signal in order to obtain a compensated signal can also be seen in the block diagrams in Figures 1, 2I, 5A, 5B and 6B. This distinguishes embodiments of offset compensation from conventional temperature compensation schemes that multiply the uncompensated Hall signal by some suitable temperature function (i.e., the temperature sensor signal) in order to obtain a temperature sensor with, for example, temperature independent behavior. compensation signal. While these conventional temperature compensation systems adjust system gain as a function of temperature, embodiments disclosed herein adjust system offset as a function of temperature gradient.

However, now assume that there is a mismatch between the two temperature devices D2 and D4. The mismatch can occur at VT20<>VT40 or ST2<>ST4, or both. Above we see that the subtraction of the signals of the first and third stages of operation eliminates the VT20 and VT40 terms, which is desired. Since the additional error caused by the thermal EMF in the stage temperature signal is VT20*ST2*(T2−T2′)−VT40*ST4*(T4−T4′), the system only has to deal with VT20*ST2< >VT40*ST4 mismatch between. Therefore, we can say that VT40*ST4=VT20*ST2*(1−MM), where MM is the mismatch between D2 and D4. The additional error due to the thermal EMF in the stage temperature signal is

VT20*ST2*(T2−T2′)−VT20*ST2*(T4−T4′)*(1−MM)=VT20*ST2*(T2−T2′−T4+T4′)+MM*VT20*ST2* (T4−T4′)

So instead of measuring T2−T2′−T4+T4′ in the presence of a mismatch in the temperature device, the system measures:

T2−T2′−T4+T4′+MM*(T4−T4′)

Therefore, as long as T4−T4′ is similar to T2−T2′−T4+T4′, the error is modest. However, when |T4−T4′|>>|T2−T2′−T4+T4′|, the error is large. In other words: the temperature fluctuations at the output contactors in the two operating phases with inverted supply voltage or current should not be large relative to the temperature difference fluctuations between the two output contactors in these operating phases. This means: when the power supply is reversed, the total power dissipation in the Hall effect device should change as little as possible. This also means: when the power supply is reversed, the power density in the vicinity of the output contactor should remain as constant as possible. Therefore, in an embodiment, the operating conditions should be such that the common mode potential at the output contact should remain the same or nearly the same when the power supply is reversed.

This is illustrated, for example, in Figures 7D and 7E. FIG. 7D depicts a Hall plate 202 with four contacts C1-C4 that is biased in an undesirable manner because, due to the electrical nonlinearity of the Hall effect device, when the supply voltage is reversed, the common Die potential (0.5*V(C2)+0.5*V(C4)) changes. Due to the electrical nonlinearity of the device, the common-mode output voltage is slightly less than half the supply voltage. Therefore, it is approximately 0.45*V(C1) in the first operating phase and 0.45*V'(C3) in the third operating phase. Assuming a perfectly symmetrical Hall effect device, V(C1)=V'(C3), ie if the supply currents are the same but with different polarities, then the supply voltages are the same. The common mode output voltage is also the same in both phases, and therefore, temperature T2 should be very similar to T2' (and T4 should also be similar to T4'). However, if the Hall effect device is slightly asymmetric such that contactor C1 is, for example, about 1% smaller than C3, this will cause V(C1) to be different from V'(C3), so the common mode in both phases The output voltage is also different. Additionally, the temperatures T2 and T2' (or T4 and T4') should be significantly different than before.

Figure 7E shows the same device biased in a more beneficial manner, where the common mode potential is controlled to be at the same level when the power supply is reversed. The operational amplifier OPA compares the reference voltage Vref with the sum V(C2)+V(C4). If the sum is the greater of the two, the output of the operational amplifier OPA rises, which pulls the gate of the NMOS up, which in turn sinks more current, which in turn pulls down V(C2) and V(C4) . Therefore, the common mode is controlled to the value Vref/2.

There are many other ways to hold the common-mode potential at a fixed value during rotating parts of the Hall scheme, and the examples discussed herein are not limiting. Many of these approaches are beneficial because they reduce matching requirements for temperature devices (eg, diodes D2 and D4). Typically, the goal is to keep the common-mode potential of the output constant when the power supply is reversed, but the differential potential V(C2)−V(C4) remains free to output the magnetic field signal. If the common-mode potential does not change, the power density (and therefore, the temperature distribution) should also not change. Note that the common-mode potential can have various effects on the Hall region: If the Hall region is isolated from the environment by a reverse-biased pn junction as usual, the common-mode potential determines the reverse bias and this The width of the depletion layer is determined, which defines the active width of the Hall region. The thinner the active Hall region, the higher its resistance will be. On the other hand, the common mode potential can affect the amount of free charge in the active Hall region, or at least part of the active Hall region (eg, through charge accumulation effects), and this also affects the resistance. Resistance also affects power dissipation and thus temperature distribution in the device.

In an embodiment, the system estimates T2−T4−T2'+T4' as shown above in the discussion herein for the simple case with a perfect match. Mismatches will result in reduced accuracy in such estimates, but the level of accuracy desired or needed can vary. In an embodiment, the temperature difference T2−T4 of the differential output of the Hall effect device can be estimated down to about 0.001°C. This provides a thermal emf voltage of approximately 1.5 μV for a Seebeck coefficient of 1500 μV/°C. A typical Hall effect device has a magnetic sensitivity of about 50 mV/T when operated on a 1V supply, so 1.5 μV corresponds to a 30 μT offset. This offset occurs in all phases of the rotating Hall probe, and it is random, whereby the offsets in the quadrature phases are presumably statistically independent, and the offsets in the non-orthogonal phases should also be substantially on independent. So if the rotation scheme has 4 stages of operation, the remaining offset should be about half or 1/sqrt(4). This gives a residual offset of about 15 μT, which roughly corresponds to observations performed by the inventors in the laboratory.

When a Hall Effect device operates according to a rotating Hall scheme, the temperature will also change with each new phase of operation. Furthermore, the system has some delay due to the thermal mass of the circuit arrangement, and this can cause the temperature in the nth operating phase to be also affected to some extent by the (n−1)th operating phase or generally any previous operating Effect of temperature during phase. Referring again to Figure 2E (Figure 2E shows the Hall effect device operating in the first phase), due to the electrical nonlinearity of the device, the contactor (C1) at the highest potential will also reach the highest temperature, while the contactor at ground C3 has the lowest temperature. If the second operating phase shown in Figure 2F immediately follows the first operating phase, the contactors C1 and C3 change roles: In phase 1, each of the contactors C1 and C3 is a power contactor, Whereas in Phase 2, each of the contacts C1 and C3 are now signal contacts. This gives the temperature difference between these two output contactors at the beginning of phase 2 when C1 is slightly hotter than C3. Therefore, it also provides the thermal emf voltage which causes the offset error. If the duration of phase 2 is significantly longer than the thermal time constant of the Hall effect device, then contactors C1 and C3 will end up at the same temperature at the end of phase 2, assuming perfect symmetry of the device. Thus, a transient temperature difference between the output contactors can occur such that it varies during a phase, and the duration of the operating phase can have a significant impact on these effects: if the rotating Hall scheme is executed very slowly, these transients The state effect can be neglected, however, if it is performed very quickly, the temperature during an operating phase can be mainly determined by the previous operating phase and only to a negligible extent by the current operating phase. Furthermore, the system measures the Hall signal and the temperature difference at the same time: at each point in time, when measuring the Hall signal, the thermal emf error is part of this Hall signal, and therefore, the system should also know the temperature difference.

Here, however, two types of systems can be distinguished: integrating systems and sampling systems. An integrating system such as, for example, a continuous-time sigma-delta analog-to-digital converter (CT-SD-ADC) integrates the Hall output signal during a certain time interval (eg, during an entire operating phase). In this case, the system is also able to simultaneously integrate the temperature difference between the output contactors. Alternatively, a successive approximation analog-to-digital converter (SAR-ADC) typically samples the Hall signal, which means it freezes the value using a sample-and-hold technique and then converts this static value. In this case, it is also possible to sample the temperature difference of the output contactor at the same time as the Hall signal is sampled.

Under static conditions, the temperature distribution in a 100 μm × 100 μm square silicon Hall plate with 5 μm thickness and typical nonlinearity operated at 3 V power supply in stage 4 of FIG. 2H is depicted in Figure 8A. It can be seen that contactor C4 is at the highest potential and also at the highest temperature (approximately 0.062°C above room temperature of 300 K).

FIG. 8B depicts the transient temperature behavior of several points in the Hall plate 202 of FIG. 8A during a spinning Hall cycle. The pulse-shaped curve represents the temperature of the four contacts, while the smooth curve is the temperature of the center of the Hall device 202 (eg, the active region 226 ). In this example, each phase of operation is approximately 10 μs long and initially 300 K exists before the device is powered up. Therefore, a fast thermal time constant of about 1-2 μs can be seen. There are also two different pulsed "teeth" in each phase of operation, the high potential power contactor and the low potential power contactor, and two equal "valleys". The equal "valley" is the output contactor. The temperature difference between the power contactor and the signal contactor is about 0.03°C, and the temperature difference between the two power contactors is less than about 0.001°C.

In an embodiment, the system is able to perform an automatic calibration if the initial mismatch between diodes D2-D4 is too large. Thus, in one embodiment, a heating element can be used which is designed in such a way that the same temperature is generated on D2 and D4. This heating element can be activated for a certain period of time and then measure the output of D2-D4 (i.e. once during end-of-line calibration, or during power-up of the sensor system, or repeatedly, e.g. every 100 ms ). If the output changes when the heating element is activated, this change is stored and subsequently subtracted from each measurement and used to correct subsequent measurements.

For example, if the heating element is off, the voltages across temperature diodes D2-D4 are V(D2), V(D4) and the temperatures at the diodes are T2, T4 respectively. If the heating element is on, the voltages are V"(D2), V"(D4), where the temperatures are T2+dT and T4+dT respectively. The system can measure these voltages and calculate (V(D2)−V(D4)−V″(D2)+V″(D4))/(V(D2) equal to 1+VT40*ST4/(VT20*ST2) −V″(D2)), which gives a mismatch in the temperature sensitivities of the two temperature sensors.

Alternatively, some parameter of one or both of diodes D2 or D4 can be adjusted to reduce the observed D2-D4 output variation to zero. For example, as discussed above, if D2 and D4 are diodes or resistors, the system can vary the current IT2-D2 until the temperature difference observed at D2-D4 does not depend on whether the heating element is on or off. This process adjusts the sensitivity of the D2 to match that of the D4. Although changes in IT2 also change the self-heating of the temperature sensor, this generally has no adverse effect on the system, as long as this self-heating is the same during the heater's on-time and off-time.

Note that there are typically different temperatures at D2 and D4 due to other heat sources present in the system or the operation of the Hall effect device. Thus, the system is only able to extract the difference in temperature difference between switching on and off of the heating element, so only superimposed temperature differences caused by symmetrical heating elements are relevant. In an embodiment, the heating element used for autocalibration should be completely symmetrical with respect to the two temperature sensors D2 and D4 (and for the quadrature phase, the same or different heating elements need to be completely symmetrical with respect to D1 and D3) such that it is at D2 The same temperature increase (in an embodiment, if possible, up to better than 1%, such as 0.1% or even better) occurs as at D4. There are several schemes that can be used to achieve this: for example, one heating element with respect to two temperature sensors D2 and D4 (or other diodes or temperature sensing devices as the case may be in any particular embodiment or configuration, where only For convenience the examples here relate to the description in the drawings) placed symmetrically, or each temperature sensor has its own dedicated heating element. In the second case, the heating elements should be perfectly matched, so now it can be seen that the problem of mismatch is only transferred from D2-D4 themselves to their corresponding heating elements. However, this may still be a viable option if the mismatch of the heating element is smaller than that of the temperature device. Even in this case, it can be beneficial to have a sufficiently large separation between the temperature sensor and the corresponding heating element, because every small change in layer thickness or other details can contribute to the temperature drift on the temperature sensor caused by the heater. have a big impact.

For example, assume that D2 and D4 are both diodes and have heating elements HT2 and HT4 comprising a resistive strip placed above them. The resistive strips and this particular placement are just one example of a suitable heating element and arrangement. Even if HT2 and HT4 are perfectly matched, the thickness of the intermetal dielectric or some other structure between HT2 and D2 may differ by eg 1% from the same structure between HT4 and D4. This is very likely since the vertical separation between HT2 and D2 is only about 10 μm or similar. Although HT2 produces the same heat and heat density as HT4, the temperatures at D2 and D4 will be different. However, the situation improves if the spacing between D2 and HT2 and also between D4 and HT4 is increased by a lateral distance of eg approximately 10 μm. Thermal coupling is then less tight, but more stable against production tolerances. On the other hand, the greater the distance between D2 and HT2, the greater the influence that surrounding areas and structures arranged there can have on the thermal coupling between them. Therefore, it can be beneficial that the layout of other circuitry around D2-HT2 and D4-HT4 is symmetrical in order to have exactly the same thermal coupling between D2-HT2 and D4-HT4. The same is true if only a single heating element is used for both D2 and D4; then, a mismatch between HT2 and HT4 is avoided. Of course, it is generally not practical to make the overall system layout symmetrical with respect to the heating elements and all the temperature sensors D1-D4. Thus, in an embodiment, it can be made symmetric over a certain distance, while at large distances the symmetry can be imperfect, as long as the coupling of the heat source and temperature sensor is still strong enough. Using the numbers given above, one can estimate whether some asymmetry is still acceptable: If the mismatch should be reduced to about 0.1% due to autocalibration, the thermal coupling between the heating element and D1-D4 must match at least Up to about 0.1%. Any asymmetries can be modeled in digital computer code (eg, finite element simulations), and thermal coupling can be studied.

If the main sensor has a contact diffusion of a first conductivity type (eg a Hall sensor has an n-doped contact diffusion), it may perhaps be possible to place a smaller diffusion tip of a second, opposite conductivity type within the contact diffusion of the first conductivity type. In the case of a Hall sensor, this would be a smaller p-head within an n-head. This gives a pn junction which can be used as a temperature device for this corresponding contactor. An example is depicted in FIG. 9, which is a version of FIG. 3 discussed above. This configuration saves space and provides a tighter thermal coupling between the contactor and temperature sensor. In an embodiment, the contactor 204a can be annular, completely surrounding the head 306 when viewed from the top (compared to the side sectional view of FIG. 9 ), so that current flowing into/out of the temperature device does not affect the potential distribution in the main sensor . Note that Figure 9 is a schematic diagram of a single contactor with an integrated pn temperature diode; in practice, Hall effect devices such as Hall plates or vertical Hall effect devices typically have three or more contactors , each contactor has such a pn junction. In general, all contactors used to tap signals in at least one phase of a rotary Hall scheme should have such a pn junction.

Fig. 10 is a diagram of a sensor arrangement 1000 according to an embodiment.

As an overview, the sensor arrangement 1000 overcomes the disadvantages of the previous embodiments in that the first conductor element 1020 consisting of a first material (e.g. polysilicon) having a large Seebeck coefficient and thus also a relatively high resistivity is only arranged Between the two contacts 1011, 1012 of the Hall device. There, both ends of the first conductor element 1020 are electrically coupled to a second conductor element 1030 and a third conductor element 1040, respectively, which are made of a second material (e.g., aluminum )constitute. The second material has a small Seebeck coefficient and therefore also has a lower resistivity than the first material, and as a result, the second conductor element 1030 and the third conductor element 1040 can be made relatively long without their internal resistance become impermissibly high. At the distal end of the second conductor element 1030 and the third conductor element 1040 a voltage is detected and from this a temperature difference between the two contact heads 1011 , 1012 is deduced.

The sensor arrangement 1000 comprises a Hall effect region shown by a dashed box in FIG. and signal line 1060.

The contact heads include a first contact head 1011 , a second contact head 1012 , a third contact head 1013 and a fourth contact head 1014 . The first contact 1011 is located near the outer surface of the Hall effect region. Optionally, the second contact 1012 may be located near the outer surface of the Hall effect region. Furthermore, optionally, the second contact 1012 may be located near the outer surface of the same Hall effect region as the first contact 1011 . The contact heads 1011, 1012, 1013, 1014 have a high doping concentration. Therefore, the contacts 1011, 1012, 1013, 1014 have low resistance, and their Seebeck coefficients are also relatively small. However, the Seebeck coefficients of the contacts 1011, 1012, 1013, 1014 and the Seebeck coefficient of the tungsten plug are irrelevant because these elements are so small that they have a uniform temperature. On the other hand, the Hall effect area is relatively large and there is a temperature gradient between the two sensing contacts. A Hall effect region may have a Seebeck coefficient of approximately 1.5 mV/degree.

The first conductor element 1020 is shaped as an elongated track and comprises two ends, a first end portion 1022 and a second end portion 1024 , between the first contact head 1011 and the second contact head 1012 . The first conductor element 1020 comprises, for example, unsilicided n- or p-doped polysilicon with a relatively high resistivity, or a shallow p-head over a Hall device, or metal with the second conductor element 1030 and the third conductor element 1040. compared to any material with a significantly different Seebeck coefficient. Polysilicon has a Seebeck coefficient of approximately 200 mV/Kelvin. The first end portion 1022 is thermally coupled to the first contact head 1011 . The second end portion 1024 is thermally coupled to the second contact head 1012 . The temperature difference between the first contact head 1011 and the second contact head 1012 is to be measured. The ends of the first conductor element 1020 are in contact with the usual aluminum of the wiring plane, so that the first and second thermocouples are present at the contact locations.

The second conductor element 1030 includes two ends—a third end portion 1032 and a fourth end portion 1034 . The third end portion 1032 is thermally coupled to the first contact head 1011 . The first end portion 1022 of the first conductor element 1020 and the third end portion 1032 of the second conductor element 1030 are electrically coupled in point form. In practice, they are often coupled by one or several tungsten plugs - if one uses several tungsten plugs, they are usually connected electrically in parallel, however, if the metal layer of the second conductor element is higher than metal 1, then according to It may also be necessary to use several tungsten plugs for series connection.

The third conductor element 1040 includes two ends—a fifth end portion 1042 and a sixth end portion 1044 . The fifth end portion 1042 is thermally coupled to the second contact head 1012 . The second end portion 1024 of the first conductor element 1020 and the fifth end portion 1042 of the third conductor element 1040 are electrically point-like coupled, or as described in the paragraph above.

The second conductor element 1030 and the third conductor element 1040 are composed of a material (eg, aluminum or copper) having a Seebeck coefficient of approximately 1 µV/degree, and have a lower resistivity than the first contactor element 1020 . The result is that the second conductor element 1030 and the third conductor element 1040 can have a relatively long length without their internal resistance becoming unacceptably high.

At least two of the first conductor element 1020, the second conductor element 1030 and the third conductor element 1040 have substantially different Seebeck coefficients, that is, preferably greater than 5 µV/°C, or preferably greater than 15 µV/°C, but the present disclosure is not limited in this respect. The difference in Seebeck coefficients may be any difference deemed suitable for the intended purpose.

The power line 1050 is electrically coupled to the third contact head 1013 and the fourth contact head 1014 . The power line 1050 is configured to supply current to the Hall device.

The signal line 1060 is electrically coupled to the first contact head 1011 and the second contact head 1012 . Signal line 1060 is configured to provide an output signal of the Hall device. When no magnetic field is present, it is expected that there will be no output signal from the output signal terminal 1062 of the signal line 1060 . However, there are thermocouples where the metal signal line 1060 touches the n-doped regions of the contacts 1011, 1012 with different Seebeck coefficients, resulting in a small thermal emf associated with the temperature difference between the first contact 1011 and the second contact 1012 (EMF) output signal. The first conductor element 1020 is electrically coupled to the second conductor element 1030 and the third conductor element 1040, but the conductor elements are not necessarily in electrical contact with the Hall effect region or any of the four contacts 1011, 1012, 1013, 1014 head coupling. Conversely, the power line 1050 and signal lines 1060, 1062 are in electrical contact with the Hall effect region.

The voltage at the distal ends of the second conductor element 1030 and the third conductor element 1040 may be tapped at the fourth end portion 1034 and the sixth end portion 1044 (i.e., the thermocouple signal terminals) and derived therefrom from the first contact 1011 and The temperature difference between the second contacts 1012. One of the two distal ends 1034, 1044 or one of the first, second and third conductor elements may optionally be coupled to a reference voltage Vref (not shown). The voltage at the thermocouple signal terminals (fourth end portion 1034 and sixth end portion 1044) is related to the metal of the third end portion 1032 and fifth end portion 1042 (that is, the metal is proportional to the temperature difference between the physical contacts between the polysilicon of the first conductor element 1020 .

During operation, no current flows through the first conductor element 1020 and the coupled second conductor element 1030 and third conductor element 1040 . The output voltage of the Hall device is measured at the output signal terminal 1062 . Additionally, the output voltage of the thermocouple was measured at the thermocouple terminals (fourth end portion 1034 and sixth end portion 1044). The two output voltages can then be combined and a correlation determined. If one of these voltages is large, the other voltage is usually large as well. A correlation can be determined and then the measured thermocouple voltage can be multiplied by some factor and subtracted from the Hall device output voltage. The result is the Hall device output voltage corrected for thermal EMF.

An advantage of the sensor arrangement 1000 over previous sensor arrangements is that the first conductor element 1020 is not as long and can be eg 50 µm. The first conductor element 1020 does not extend from the sensor arrangement 1000 to the remote amplifier, but instead only extends between the first contact head 1011 and the second contact head 1012 of the sensor arrangement 1000 . The result is that the internal resistance is not as great as in the case of amplifiers where the first conductor element 1020 extends perhaps 100-300 µm away, and therefore the temperature measurement is more accurate.

FIG. 11 is a diagram of another sensor arrangement 1100 according to an embodiment.

The sensor arrangement 1100 differs from the sensor arrangement 1000 of FIG. 10 in that the first conductor element is shaped as a circular plate 1120 with protrusions (first end portion 1122 and second end portion 1124 ) instead of being shaped into a thin Long track 1020. The first conductor element 1120 of this embodiment is wider and, therefore, is beneficial for reducing the internal resistance of the temperature measurement.

The sensor arrangement 1100 also differs from the sensor arrangement 1000 of FIG. 10 in that the signal wire of one of the thermocouples is combined with the signal wire of the Hall device output signal wire. More specifically, in addition to being electrically coupled to first end portion 1122 , third end portion 1032 is also electrically coupled to first contact head 1012 at contact 1164 . In this embodiment, the thermocouple is electrically coupled to the Hall device such that the common mode potential of the thermocouple is determined by the Hall device. No additional reference potential Vref is required and the signal wires are saved, ie one of the output signal wires with the output terminal 1162 is identical to one of the thermocouple wires with the fourth end portion 1034 .

Figure 12 is a diagram of another sensor arrangement 1200 according to an embodiment.

The sensor arrangement 1200 differs from the sensor arrangement 1100 of FIG. 11 in that instead of being shaped as a circular plate 1120 , the first conductor element 1220 is shaped as a square and smaller than the Hall effect area. In addition, the first conductor element 1220 has four protrusions instead of only two protrusions.

Sensor arrangement 1200 also differs from sensor arrangement 1100 of FIG. ) for thermal EMF compensation. As previously discussed, Hall devices typically operate in a spinning current mode with two phases of operation. In the first phase of operation, the supply current is sent through the diametrically opposite contact heads, and the output voltage is measured at the other two contact heads. In the second operating phase, there is switching, that is to say the first pair of contacts is used to tap the output voltage and the supply current is routed via the second pair of contacts. Since the output voltage is switched, it is desirable to know the temperature difference at the contacts used as output terminals. The contacts are swapped at each operating phase and therefore also the thermocouples. Four thermocouples are therefore required, wherein each thermocouple is assigned to one of the contacts 1011 , 1012 , 1013 , 1014 .

The first conductor element 1220 comprises four extensions between the first, second, fourth and third contacts 1011, 1012, 1014, 1013 - first, second, seventh and eighth end portions 1222, 1224, 1226, 1228. The first end portion 1022 is thermally coupled to the first contact head 1011 . The second end portion 1024 is thermally coupled to the second contact head 1012 . The seventh end portion 1226 is thermally coupled to the fourth contact head 1014 . The eighth end portion 1228 is thermally coupled to the third contact head 1013 . The temperature difference between the first contact head 1011 and the second contact head 1012 is to be measured during the first operation phase, and the temperature difference between the third contact head 1013 and the fourth contact head 1014 is to be measured during the second operation phase. The ends of the first conductor elements 1020 are in contact with the generally aluminum of the wiring plane, so that corresponding thermocouples are present at the contact points.

The fourth conductor element 1240 includes two ends—a ninth end portion 1242 and a tenth end portion 1244 . The seventh end portion 1226 of the first conductor element 1220 and the ninth end portion 1232 of the fourth conductor element 1240 are electrically coupled in a point-like manner, or via a tungsten plug as described above.

The fifth conductor element 1230 includes two ends—an eleventh end portion 1232 and a twelfth end portion 1234 . The eleventh end portion 1232 is thermally coupled to the third contact head 1013 . The eighth end portion 1228 of the first conductor element 1220 and the eleventh end portion 1244 of the fifth conductor element 1230 are electrically coupled in a point-like manner, or via a tungsten plug as described above.

Fourth conductor element 1240 and third fifth conductor element 1230 are similar to second conductor element 1230 and third conductor element 1240 in that they are composed of a material (e.g., aluminum or copper) that has a Seebeck coefficient of about 1 µV/degree And has a lower resistivity than the first contactor element 1220 . The result is that the fourth conductor element 1240 and the fifth conductor element 1230 can have a long length without their internal resistance becoming unacceptably high.

At least two of the first conductor element 1220, the fourth conductor element 1240, and the fifth conductor element 1230 have substantially different Seebeck coefficients, that is, preferably greater than 15 µV/°C, but the present disclosure is in This aspect is not limited. The difference in Seebeck coefficients may be any difference deemed suitable for the intended purpose.

During the first phase of operation, the voltage difference signal at the distal ends of the second conductor element 1030 and the third conductor element 1040 may be tapped at the fourth end portion 1034 and the sixth end portion 1044 (i.e., thermocouple signal terminals), And the temperature difference between the first contact head 1011 and the second contact head 1012 is deduced therefrom. The voltage at the thermocouple signal terminals (fourth end portion 1034 and sixth end portion 1044) is related to the metal of the third end portion 1032 and fifth end portion 1044 (that is, the metal is proportional to the temperature difference between the physical contacts between the polysilicon of the first conductor element 1020 .

During the second phase of operation, the voltage difference signal at the distal ends of the fourth conductor element 1240 and the fifth conductor element 1230 may be tapped at the tenth end portion 1234 and the twelfth end portion 1234 (i.e., thermocouple signal terminals). , and deduce therefrom the temperature difference between the third contact head 1013 and the fourth contact head 1014 . The voltage at the thermocouple signal terminals (the tenth end portion 1244 and the twelfth end portion 1234) is the same as the ninth end portion 1232 and the eleventh end portion 1232 (that is, the fourth conductor element 1240 and the fifth conductor element 1230 is proportional to the temperature difference between the metal and the physical contact between the polysilicon of the first conductor element 1220 .

During the first phase of operation, the sensor arrangement 1200 is supplied with current through the line 1260 and the output signal is tapped from the line 1160 . During the second phase of operation, this situation is reversed as the sensor arrangement 1200 is supplied with current through the line 1160 and the output signal is tapped from the line 1260 .

Figure 13 is a diagram of another sensor arrangement 1300 according to an embodiment.

The sensor arrangement 1300 differs from the sensor arrangements 1000, 1100 and 1200 of Figures 10-12, respectively, in that the thermocouples are linear or planar rather than in points. Thermocouples are intended to detect the average temperature across the contact, and therefore, it may be appropriate for the thermocouple to be linear or planar. More specifically, if a straight or curved path or area is connected directly to the first conductor element 1320 and to the second conductor element 1330 and the third conductor element 1340, then the first conductor element 1320 is connected to the second conductor element 1330 and the third conductor element 1340. A plurality of contact points between conductor elements 1340 may be arranged along the path or over the area. If the average temperature of the contact head 1310 is to be detected by a thermocouple, it is beneficial that the contact point between the first conductor element 1320 and the third conductor element largely surrounds the contact head.

Figure 13 illustrates the lower left corner of the Hall plate. Dashed lines are part of the Hall effect region. One contact head 1310 is shown, but there are identical contacts not shown. The contact head 1310 has a first contactor 1312 electrically coupled.

The first conductor element 1320 may be the same as the top plate of the Hall effect device. The top plate is a thin conductive plate covering the majority of the top surface of the Hall effect region. The main purpose of the top plate is to avoid the large electric field acting on the relatively low doped Hall effect region, since the large electric field can act as a force on the moving ions in the Hall effect region. If these ions move, they change the charge distribution in the Hall effect region, and this change in charge distribution changes the magnetic sensitivity and offset of the Hall effect device. The top plate is a metal or polysilicon plate that is electrically isolated from the Hall effect region by some interstitial dielectric layer and coupled to a reference potential, usually ground potential. Alternatively, a shallow p-head can be placed above the Hall-effect region as an n-head, whereby the p-head is coupled to a potential lower than the lowest potential in the Hall-effect device such that the gap between the top plate and the Hall-effect region The pn junctions are reverse biased, and therefore, no current flows between them.

This top plate may be the first conductive element 1320, however, said top plate should be composed of a material having a Seebeck coefficient significantly different from that of the interconnect layer. Furthermore, the top plate should be coupled to traces of the interconnection layer close to at least two signal/output contacts.

The top plate of this embodiment is essentially the same size as the Hall Effect region, except for the rectangular apertures 1324 used to access the corresponding contacts. The first conductor element 1320 has a ring of electrically coupled second contacts 1322 distributed over the first contact element 1320 to surround the contact head 1310 . This ring of the second contactor 1342 is in contact with the first conductor element 1320 . The ring of the second contactor 1342 measures the temperature of the contact head 1310 . By surrounding the contact head 1310 with a contactor, the average temperature on the contact head 1310 can be obtained.

The top plate as the first conductor element 1320 may comprise polysilicon and may be electrically isolated from the base Hall effect region by a dielectric layer. A shallow p-doped head, which is located in the surface of the n-doped Hall effect region, can alternatively be used as the cap. Such a head would benefit from having a higher Seebeck coefficient, which results in a larger signal at the same temperature gradient. The layout or geometry of these covers is similar to the embodiment of Figures 11-13.

If the first conductor element 1320 is isolated from the Hall effect region by some sort of gap electrically insulating layer, there can be electrical coupling between the first conductor element 1320 and the underlying Hall effect region. For this, the insulating layer should be opened, a tungsten plug inserted, and a small contact diffusion added above the Hall effect area to contact the tungsten plug. Preferably, this contactor is located in the center (in plan view) of the Hall effect region so that its potential does not increase much during the spinning current regime. This central fifth contact in the center of the Hall effect region does not introduce large asymmetries, and therefore, it does not add significant offset error to the Hall effect device. In this case, the common mode potential of the first conductor element 1320 is equal to the common mode potential of the Hall effect region. Alternatively, the first conductor element 1320 may be electrically coupled to an interconnection line and this interconnection line is routed to a reference potential. In this case, the contact can also be located near the center of the Hall effect region, but it can also be located anywhere on the first conductor element 1320 .

The second conductor element 1330 includes two ends—a first end portion 1332 and a second end portion 1334 . The first end portion 1332 is electrically coupled to the electrically coupled first contact 1312 . The second end portion 1334 is an output signal terminal.

The third conductor element 1340 includes two ends—a third end portion 1342 and a fourth end portion 1344 . The third end portion 1342 is electrically coupled to the ring of the second contactor 1342 that is electrically coupled. The fourth end portion 1344 is a thermocouple signal terminal which may be part of the Hall Effect device, and alternatively is a reference point external to the Hall Effect device.

In this embodiment, the Hall effect region coincides with the first conductive element 1320 . Contacts between the first conductor element 1320 and the second or third element 1330, 1340 are not located on the overhang, since these contacts do not require an overhang.

During operation, the voltage at the distal end of the second conductor element 1330 at the second end portion 1334 may be tapped in combination with the voltage at the corresponding terminal of the second contact (not shown) to obtain the contact 1310 and the second contact. The output voltage difference between the contacts, from which the temperature difference is derived. The voltage at the fourth end portion 1344 at the distal end of the third conductor element 1340 may be tapped off in combination with the voltage at the corresponding terminal of the second contact to obtain the output voltage of the Hall device. The two output voltages can then be combined and a correlation determined. If one of these voltages is large, the other voltage is usually large as well. A correlation can be determined and then the measured thermocouple voltage can be multiplied by some factor and subtracted from the Hall device output voltage. The result is the Hall device output voltage corrected for thermal EMF.

14A and 14B are circuit diagrams of a sensor arrangement 1400 according to an embodiment.

As an overview, this embodiment uses the Hall Effect devices themselves in a time-interleaved fashion, instead of a top plate above the Hall Effect devices, in order to correct for thermal EMF. In a first phase of operation, the Hall effect device is energized and the output signal is sampled. During this first phase of operation, an inhomogeneous temperature distribution is established between the Hall effect device output contacts C1 and C3. Then, in a second non-operating phase, the current flow through the Hall effect device is switched off so that the voltage distribution in the Hall effect device is no longer influenced by the current flow or the magnetic field. Instead, the voltage distribution is only influenced by the Seebeck effect, that is to say by the thermal voltage generated by the inhomogeneous temperature distribution. In other words, in this current-free state, a thermal voltage can be tapped at the Hall effect device output contacts C1 and C3 and a temperature difference derived therefrom. However, since the Hall effect device is not energized, there is no self-heating and the temperature profile established in the previous energized state now decays at a rate determined by the thermal time constant of the sensor arrangement 1400 . If the sensor arrangement 1400 is capable of detecting thermal voltages at the Hall effect device output contacts C1 and C3 quickly enough (e.g., within 1 μs or less), then these measurements can be compared to those in the energized first operating state. temperature inhomogeneity correlation. Then, the offset error due to temperature non-uniformity can be estimated and the Hall effect device output signal corrected for the offset error.

When the current is disconnected during the second non-operating phase, the voltage of the Hall effect device output contacts C1 , C3 is initially floating. An output contact C3 is coupled to the reference voltage through a first voltage source (1.15 V). Preferably, this voltage is the same as the voltage at the output contact C3 during the first operating phase, since at this output contact C3 the parasitic capacitance most rapidly undergoes charge inversion. In a current-free second non-operating phase, all other contacts C1, C2, C4 then experience a charge reversal of this voltage according to a time constant which approximately corresponds to the internal resistance of the contacts and the current in the network to ground. The product of the parasitic capacitances of the nodes (eg approximately 5kΩ • 200fF = 1ns). Capacitive transients are thus largely decayed after about 50 ns after supply current is disconnected, while thermal transients last up to 1 μs, or perhaps even 10 μs.

Figure 14A is a circuit diagram of a sensor arrangement 1400A during a first phase of operation.

The sensor arrangement 1400A comprises a Hall effect device, a preamplifier, switches S1, S2, S3, an NMOS current mirror, a PMOS current mirror, a feedback loop and a first voltage source.

The Hall effect device includes contacts C1-C4. The contacts C1-C4 are configured to supply power and tap the output voltage of the Hall effect device. The contacts C1 and C3 are the output terminals of the tap output signal. The preamplifier is configured to amplify the output signal. Contacts C2 and C4 are configured to supply current to the Hall effect device.

The PMOS current mirror is a current source, and current is supplied from this current source to the Hall effect device via switch SW1.

An optional feedback loop is configured to sink current to control the voltage at switch SW2. The feedback loop includes an operational transconductance amplifier (OTA) and a second voltage source (eg, 0.15 V). The OTA is configured to compare the voltages at its input terminals. At one input terminal is a reference voltage (a second voltage source of 0.15 V). If the voltage at switch SW2 is higher than the reference voltage (0.15 V), the OTA output is high, pulling up the gate of the NMOS current mirror, which in turn pulls down the voltage at switch SW2, thereby pulling down the voltage at the non-inverting input of the OTA. This negative feedback loop is thus configured to control the voltage at switch SW2 to the reference voltage (0.15 V). Switches SW1 and SW2 thus represent power supply terminals.

The voltage at switch SW1 is not directly controlled; it is controlled by a feedback loop. When the first operating phase ends and the second non-operating phase begins, switches SW1 and SW2 are opened to interrupt current flow through the Hall effect device. The Hall effect device still needs to be at a certain voltage, and that voltage is the first voltage via switch SW3 (that is, 1.15 V in this case). When the current through the Hall effect device is turned off, the voltage of the Hall effect device at output contact C3 does not change because it is always coupled to the first voltage source of 1.15 V through switch SW3. The preamplifier needs this voltage because the preamplifier can only measure microvolts; if the voltage changes by 0.5 or 1 V, the preamplifier will go into full saturation and won't work.

The NMOS and PMOS current mirrors may alternatively be replaced by any other current mirror circuits such as those comprising bipolar transistors or cascodes. Furthermore, the first and second voltage sources are not limited to the specific voltage amounts mentioned. These voltage sources may be of any voltage amount suitable for the intended purpose.

In this embodiment, there are no dedicated thermocouples because the output contacts C1 and C3 serve as conventional Hall effect signal outputs during the first phase of operation and as thermocouples during the second phase of non-operation. In order to use the output contacts C1 and C3 as thermocouples, no current can flow through the Hall effect device. This is the reason for the two stages.

As depicted in Figure 14A, the first phase of operation is the conventional phase of operation, except that switch SW3 couples the first voltage source to the preamplifier. In order for this output line from contact C3 not to jump in voltage after the current through the Hall effect device is disconnected, the sensor arrangement 1400 pins this output to the first voltage source (1.15V).

Figure 14B is a circuit diagram of the sensor arrangement 1400B in a second non-operational phase.

In the second non-operational operation period immediately following the first operation period, the switches SW1 and SW2 are turned off, as shown in FIG. 14B . As a result, current no longer flows into or out of the Hall effect device. However, the voltage at one of the two output terminals of the Hall effect device (in this case output terminal C3) is still maintained at 1.15 V via the first voltage source because switch SW3 is still closed. The Hall effect device voltage is not free floating because switch SW3 remains closed and ties the Hall effect device output and the preamplifier input to the first voltage source (1.15 V). At this preamplifier input, there is no change between the first and second stages because it is coupled to the first voltage (1.15 V). Another preamplifier input is coupled to the output contact C1 which will now vary. In the first phase of operation, the difference between the output contacts C1 and C3 is caused by the magnetic field of the Hall effect device. But since current no longer flows through the Hall effect device during the second non-operating phase, the magnetic field no longer affects the output voltage, which now decreases. But the thermal EMF voltage is still there. The output contacts C1 and C3 are at slightly different temperatures (eg, about 20 mK apart), and each of these output contacts acts as a thermocouple. As a result, there is a voltage difference of perhaps 800µV between the inputs of the preamplifiers during the second non-operational phase.

The preamplifier must rapidly measure this thermal voltage in a time period shorter than the thermal time constant of the temperature difference decay. If appropriate, sample and hold elements can also be used to detect thermal voltages immediately after switches SW1 and SW2 are opened and provide the preamplifier with more time to process the voltage.

When switches SW1 and SW2 are open, a small glitch voltage appears across Hall effect output contacts C1 and C3. Interference voltages are unavoidable since all circuit nodes are loaded with unavoidable stray capacitance. In this regard, after opening switch SW1, the connection of the Hall effect device to switch SW1 drops from approximately 2.25 V to 1.15 V (ie, -1.1V). Simultaneously, the connection of the Hall effect device to switch SW2 rises from 0.15V to 1.15V (ie, +1V). This asymmetry (-1.1 V vs. +1.0 V) occurs because of the electrical nonlinearity of the Hall effect device, so that with the vanishing magnetic field, the voltage at the output contacts C1 and C3 is not exactly at the supply voltage the middle of the , but shifts a bit toward higher voltages. If the stray capacitances at the two connections are similar in magnitude (due to the symmetry of the Hall-effect device and similar conductor connections), then negative and positive charging currents that do not exactly cancel each other must be expected, and the difference in Switching of switches SW1 and SW2 results in short voltage pulses. For example, the input through the preamplifier is simply disconnected from the Hall effect device during switching, the circuit arrangement 1400 should mask this pulse.

There are many methods for defining the common-mode potential at the input of the preamplifier during the second non-operational phase. For example, in the first phase of operation, the auxiliary circuit may tap the potential at contact C3 or C1 or contacts C3 and C1 and charge the capacitor to V(C3) or V(C1) or (V(C3)+ V(C1))/2. During the second non-operational phase, the auxiliary circuit no longer charges this capacitor but instead couples it between ground and any contact C1 , C2 , C3 or C4 . The potential at contact C3 or C1 is then defined by this capacitor during the entire second non-operating phase.

An advantage of using the Hall effect device directly as a thermocouple is that the metal/semiconductor junction (the metal/semiconductor structure forming the thermocouple) is better thermally coupled to the output contact. The Hall effect device output contacts C1 and C3 are measured directly to obtain the true temperature of the contacts, rather than the temperature of a thermocouple close to the contacts, resulting in a more accurate temperature measurement. Furthermore, the n-doped contact of a Hall effect device is usually the element with the largest Seebeck coefficient. This is the worst case for a Hall effect device as it has the largest thermally induced error. During the second non-operating phase, the Hall effect device is used as a thermocouple, but it is a very sensitive thermocouple. If the top plate is used as a thermocouple, the Seebeck coefficient is 100 or 200 microvolts per Kelvin, while the Hall effect device itself has a Seebeck coefficient of about 1500 microvolts per Kelvin, which is 10 times the Seebeck coefficient of the former times. This increases the thermal offset signal compared to previous embodiments where the Hall effect region is not part of the thermocouple.

This embodiment is not only applicable to conventional Hall effect devices. For example, this embodiment is also applicable to mechanical stress sensor devices, or alternatively to vertical Hall effect devices. Hall effect devices can typically only measure the magnetic field component perpendicular to the chip surface, and perpendicular Hall effect devices can measure the in-plane magnetic field component parallel to the chip surface. The top plate is electrically isolated from each Hall-effect region, for example by a dielectric isolation layer or by a blocking pn junction, and can be coupled to a voltage through connecting lines. Such a top plate can be produced in a semiconductor process from a material having a certain Seebeck coefficient which has an absolute value greater than that of the aluminum or copper metallization. And such a top plate can be provided with a plurality of connecting wires which are in contact with the top plate in each case in the vicinity of the contacts of the Hall effect region, and thus the temperature at these contacts can be measured.

Accordingly, various embodiments of sensors, systems and methods for compensating for thermal EMF effects have been discussed herein with reference to several exemplary embodiments and depictions that are not limited to the general concept. For example, having discussed examples in relation to Hall effect sensors, other sensor types can be used, including other magnetic field sensors, mechanical stress sensors, and the like. In general, however, residual offset can be associated with temperature fluctuations at the sensor contactors, which in turn can be associated with thermal EMF, and based on the temperature sensed at one or more sensor contactors, by adding a correction term or Compensating the signal or by implementing a control loop, the residual offset can be reduced or eliminated.

Various embodiments of systems, devices, and methods have been described herein. These examples are given by way of example only, and are not intended to limit the scope of the present disclosure. Furthermore, it should be appreciated that the various features of the described embodiments can be combined in various ways to yield many additional embodiments. Furthermore, while various materials, dimensions, shapes, configurations and locations, etc., have been described for use with the disclosed embodiments, materials, dimensions, shapes, configurations and locations other than those disclosed may be utilized without exceeding the scope of the present disclosure. Materials, dimensions, shapes, configurations, locations, etc. other than location, etc.

Those of ordinary skill in the art will appreciate that the present disclosure may include fewer features than shown in any individual embodiment described above. The embodiments described herein are not intended to be an exhaustive representation of the ways in which the various features of this disclosure may be combined. Accordingly, the embodiments are not mutually exclusive combinations of features; rather, the disclosure can include combinations of different individual features selected from different individual embodiments, as understood by those of ordinary skill in the art. Furthermore, elements described with reference to one embodiment can be implemented in other embodiments even when not described in such an embodiment unless otherwise indicated. Although a dependent claim may refer to a specific combination in a claim with one or more other claims, other embodiments can also include a combination of the subject matter of said dependent claim with each other dependent claim or one or more Combinations of individual features with other dependent or independent claims. Unless stated that a particular combination is not intended, such combinations are proposed herein. Furthermore, it is intended that features of this claim be included in said independent claim even if a claim is not directly dependent on any other independent claim.

Any inclusion by reference of the above documents is limited so as not to include subject matter contrary to the express disclosure herein. Any inclusiveness by reference of the documents above is also limited such that no claims contained in said documents are hereby incorporated by reference. Any inclusiveness by reference of a document above is also limited such that any definitions provided in said document are not incorporated herein by reference unless expressly included herein.

For purposes of interpreting the claims of the present disclosure, it is expressly intended not to invoke the provisions of 35 U.S.C. Section 112(f) unless the specific terms "means for" or "for. .....A step of".

While the foregoing has been described in conjunction with exemplary embodiments, it should be understood that the term "exemplary" means an example rather than the best or best. Accordingly, this disclosure is intended to cover alternatives, modifications, and equivalents, which may be included within the scope of this disclosure.

While specific embodiments have been shown and described herein, it will be understood by those of ordinary skill in the art that various alternative and/or equivalent implementations may be shown and described instead without departing from the scope of the present disclosure. specific example of . This disclosure is intended to cover any adaptations or variations of the specific embodiments discussed herein.

Claims (17)

1. A sensor device configured to provide a signal in response to a temperature difference between an output contactor of a Hall effect device and a reference point, said device comprising: Hall effect area; a first contact located near an outer surface of the Hall effect region; a second contact head located near the reference point; a first conductor element comprising first and second end portions, the first end portion being thermally coupled to the first contact and the second end portion being thermally coupled to the second contact; a second conductor element comprising third and fourth end portions, the third end portion being thermally coupled to the first contact; a third conductor element comprising fifth and sixth end portions, the fifth end portion being thermally coupled to the second contact, wherein the first and third end portions are electrically coupled and the second and fifth end portions are electrically coupled, at least two of the first, second and third conductor elements have substantially different Seebeck coefficients , and tap the signal at the fourth and sixth end sections. 2. The sensor device of claim 1, wherein the substantially different Seebeck coefficients differ by more than 15 µV/°C. 3. The sensor device of claim 1, wherein the second contact is located near an outer surface of the Hall effect region. 4. The sensor device of claim 3, wherein the second contact is located near an outer surface of the same Hall effect region as the first contact. 5. The sensor device according to claim 1, wherein the first and third end portions are electrically point-coupled, and the second and fifth end portions are electrically point-coupled. 6. The sensor device of claim 1, wherein the third end portion is electrically coupled to the Hall effect region via at least one contact. 7. The sensor device of claim 1, wherein the first conductor element is shaped as an elongated track. 8. The sensor device of claim 1, wherein the first conductor element is shaped as a plate. 9. The sensor device of claim 1, further comprising: a third contact head located near the reference point; a fourth contact located near the outer surface of the Hall effect region; The first conductor element includes seventh and eighth end portions, the seventh end portion being thermally coupled to the fourth contact and the eighth end portion being thermally coupled to the third contact; a fourth conductor element comprising ninth and tenth end portions, the ninth end portion being thermally coupled to the fourth contact; a fifth conductor element comprising eleventh and twelfth end portions, the eleventh end portion being thermally coupled to the third contact head, Wherein the seventh and ninth end portions are electrically coupled, the eighth and eleventh end portions are electrically coupled, at least two of the first, fourth and fifth conductor elements have substantially different Seebeck coefficients, and tap the signal at the tenth and twelfth end sections. 10. The sensor device of claim 9, wherein the substantially different Seebeck coefficients of the at least two of the first, fourth and fifth conductor elements differ by more than 15 µV/°C. 11. The sensor device of claim 9, wherein the ninth end portion is electrically coupled to a third contact. 12. The sensor device of claim 9, wherein the third contact is located near an outer surface of the Hall effect region. 13. The sensor device of claim 12, wherein the third contact is located near the outer surface of the same Hall effect region as the fourth contact. 14. The sensor device of claim 9, wherein the first conductor element is shaped as a plate. 15. A sensor device configured to provide a signal in response to a temperature difference between a contactor of a Hall effect device and a reference point, the device comprising: Hall effect area; a contact head positioned adjacent an outer surface of the Hall effect region and including at least one first contactor; a first conductor element surrounding a substantial portion of the contact head and including a plurality of second electrically coupled contacts surrounding a substantial portion of the contact head; reference point, electrically coupled to the first conductor element, wherein the thermal coupling between the reference point and the contact head is weaker than the thermal coupling between the second electrically coupled contactor and the contact head; and a second conductor element comprising first and second end portions, the first end portion being electrically coupled to the plurality of second electrically coupled contacts, wherein the first and second conductor elements have substantially different Seebeck coefficients, and the reference point and the second end portion are configured to tap the signal. 16. The sensor device of claim 15, wherein the substantially different Seebeck coefficients differ by more than 15 µV/°C. 17. The sensor device of claim 15, wherein the plurality of second electrically coupled contacts are arranged according to a curvature.