Device for resistance detection
The device achieves non-contact resistance detection through electromagnetic coupling, addressing the challenges of galvanic connections by using electromagnetic induction for precise and cost-effective resistance measurement.
Patent Information
- Application Number
- PCT/EP2026/063920
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2025-07-21
- Filing Date
- 2026-05-13
- Publication Date
- 2026-11-19
AI Technical Summary
Existing resistance detection methods require a galvanic connection, which can be undesirable due to electromagnetic interference, construction constraints, or high costs, and alternative methods for non-contact resistance detection are needed.
A device utilizing electromagnetic coupling between circuits on insulating substrates, allowing for resistance detection without direct electrical contact, using electromagnetic induction or electrostatic induction to generate an induced signal in a galvanically isolated second circuit.
Enables precise resistance detection while protecting the control unit from electromagnetic interference and simplifying assembly, reducing costs, and facilitating disassembly.
Smart Images

Figure EP2026063920_19112026_PF_FP_ABST
Abstract
Description
Device for Resistance DetectionTechnical field
[0001] The invention relates to a device for resistance detection, and to a method for resistance detection.Background of the Invention
[0002] A variety of sensor systems are based on measurement of electric resistance. Among such sensors are pressure sensors, sensors for collision detection, sensors for occupancy detection, and wearable sensors for gait analysis. In all these cases, the sensor comprises a resistive element with an electrical resistance that depends on a parameter that is to be measured (e.g. pressure acting on the sensor), a status that is to be detected (e.g. a vehicle seat being occupied), or an event that is to be detected (e.g. a vehicle being involved in a collision). In some cases, a quantitative measurement of the electrical resistance is necessary while in other cases a qualitative measurement is sufficient, like detecting whether there is a significant increase or decrease in the electrical resistance. It is understood that the resistance of the sensor can be determined by applying a voltage of known magnitude and measuring the electric current. This, however, requires a galvanic connection between the voltage source and the sensor. Also, some kind of evaluation device, which can measure the current and interpret the result, needs to be connected to the sensor. In some cases, such an electrically conductive connection may be undesirable. One reason may be that the sensor could be subjected to external electromagnetic influences that could impair the function of the evaluation device. By using electronic signal transmission methods that ensure galvanic isolation, the detection of the electrical resistance of a sensor can be performed with improved precision. The detection is independent of possibly large voltage changes or large currents that can occur in the electric system containing the resistance sensor, or in its vicinity. Furthermore, galvanic isolation is a very effective method for breaking ground loops, and it can provide an important safety enhancement as it prevents accidental electric shocks, for example. Another reason could be that a conductive connection is difficult to realize due to construction constraints such as space limitations and complications in the assembly process. Finally, the typically employed means to realize conductiveconnections, for example, male-female connector pairs and cable connections can involve larger costs than the means of electronic signal transmission with galvanic isolation relying, for example, on inductive or capacitive coupling.Object of the invention
[0003] It is thus an object of the present invention to provide efficient means for resistance detection that are independent of a connection with a direct electric conduction path.
[0004] This problem is solved by a device according to claim 1.General Description of the Invention
[0005] The invention relates to a device for resistance detection. The device may also be referred to as a system for resistance detection. In this context, “resistance detection” may refer to a resistance measurement, i.e. a process in which the magnitude of an electric resistance is determined. However, “resistance detection” may also refer to obtaining a less accurate and / or more qualitative result, e.g., whether a resistance is within a certain interval or has a certain order of magnitude. It may also refer to determining whether a resistance has changed significantly. As will become apparent in the following, the device comprises hardware components, but some of its functions could be software-implemented.
[0006] The device comprises a first circuit disposed on a first substrate surface and comprising a first terminal and a first coupling device that is electrically connected to the first terminal and is adapted to generate an electromagnetic field extending in a coupling direction perpendicular to the first substrate surface in response to an electric input signal applied to the first terminal. The first substrate surface is a surface of a substrate that is also part of the device. Preferably, any of the substrates mentioned here and in the following are electrically insulating. The first circuit comprises a first terminal. The first terminal comprises at least one point or node that is adapted for an electrical connection, preferably it comprises two nodes. The first terminal is electrically connected to the first coupling device. Preferably, this refers to an electrically conducting connection, i.e., a connection with a direct electric contact. A thickness of the first circuit in general and the first coupling device in particular, measured perpendicular to the first substrate surface, may be very small,e.g., less than 0.05 mm. Such values are typical, e.g., for the conductor layer thickness in PCB technology or for the dried conductive ink layer thickness on printed electronics technology. The first coupling device is adapted to generate an electromagnetic field, wherein the term “electromagnetic field” refers to a field comprising electric and / or magnetic components. Due to the electric connection between the first terminal and the first coupling device, an electromagnetic field is generated by the first coupling device when an input signal is applied to the first terminal. The input signal is an electric signal corresponding to a voltage applied to the first terminal and a current flowing between the first terminal and the first coupling device. Preferably, the input signal is a time-dependent signal, although the voltage and / or the current may be constant for certain periods of time. The electromagnetic field generated by the first coupling device may extend from the coupling device in various directions. At least, it extends in a coupling direction that is perpendicular to the first substrate surface. To this respect, one could say that it extends above and / or below the first substrate surface, wherein the terms “above” and “below do not refer to any specific direction with respect to gravity. In some embodiments, an effective range of the electromagnetic field in the coupling direction can be limited, although it is preferably greater than the thickness of the first coupling device.
[0007] The device further comprises a second circuit disposed on a second substrate surface and comprising a second terminal for connection to a load resistance and a second coupling device electrically connected to the second terminal, like the first substrate surface, the second substrate surface is the surface of a substrate that is part of the device. Like the first terminal, the second terminal comprises at least one point or node that is adapted for an electrical connection, preferably it comprises two nodes. The second terminal is electrically connected to the second coupling device. Preferably, this refers to an electrically conducting connection, i.e. , a galvanic connection. A thickness of the second circuit in general and the second coupling device in particular, may be very small, e.g., less than 0.05 mm. the second terminal is either adapted for a connection to a load resistance or it is connected, e.g., permanently connected, to the load resistance. The load resistance can be any object having an electrical resistance. Specifically, this couldbe a resistance-based sensor cell of a pressure-sensitive sensor, a thermistor, a photoresistor, or the like.
[0008] At least in an operational state of the device, the second circuit is galvanically isolated from the first circuit and the electromagnetic coupling devices are positioned opposite each other in the coupling direction to form a coupling arrangement that is adapted to generate an induced signal in the second circuit in response to the input signal. The operational state is a state in which the device can be operated according to its intended use. In this state, the first circuit and the second circuit are galvanically isolated from each other. In other words, no direct current can flow between the two circuits. Specifically, no direct current can flow between the first coupling device and the second coupling device. However, coupling devices are positioned opposite each other in the coupling direction. Galvanic isolation also implies that the coupling devices are spaced from each other in the coupling direction. A substrate that forms at least one of the above-mentioned substrate surfaces may be interposed between the circuits, and specifically between the coupling devices. Together, the coupling devices form a coupling arrangement that is adapted to generate an induced signal in the second circuit in response to the input signal. More specifically, the input signal generates the above-mentioned electromagnetic field, which in turn influences the second coupling device, whereby the induced signal is generated. The induced signal may be caused by electromagnetic induction and / or electrostatic induction (i.e. , electrostatic influence). While reference is made here to an electromagnetic field generated by the first coupling device, it is understood that the induced signal also gives rise to an electromagnetic field and the resulting field effectively is caused by an interaction between the first coupling device and the second coupling device. Due to the induced signal in the second circuit, a voltage is applied to the second terminal and, if connected, to the load resistance. Correspondingly, a current may flow through the load resistance.
[0009] Furthermore, the device comprises a control unit adapted to apply the input signal to the first terminal and to detect the load resistance based on an input impedance at the first terminal. The control unit is a hardware-based but some of its functions may be software-implemented. The control unit is electrically connected or at least connectable to the first terminal. Furthermore, it can generate the inputsignal and apply it to the first terminal. Preferably, the control unit is adapted to control various characteristics of the input signal, e.g., its time evolution. Also, the control unit can detect the load resistance based on an impedance at the first terminal which is herein referred to as the “input impedance”. In general, the input impedance has a complex value. It is defined as the (complex) ratio between the voltage at the first terminal and the current flow from or into the first terminal. Therefore, the control unit is preferably adapted to measure the respective voltage and current. In this context, the voltage and current at the first terminal can be referred to as a “signal at the input terminal”. This may be different from the input signal, though. Specifically, the signal at the input terminal also comprises a component that is returned by the first and second circuit in response to the input signal. There are various ways how the control unit can apply the input signal and detect the input impedance. One option would be to apply a controlled voltage, e.g. having a fixed amplitude and frequency. In this case, the control unit would need to drive the current through the input terminal so that the controlled voltage is maintained. From the amplitude and phase of the current in relation to the voltage, the amplitude and phase of the input impedance can be calculated. Similarly, the control unit could apply a controlled current to the input terminal, in which case the amplitude and phase of the voltage could be used to calculate the amplitude and phase of the input impedance. Another option would be to use a directional coupler, which allows to separate signals based on their direction of propagation. In this case, the control unit could apply various types of input signal while the response signal or reflected signal resulting from this input signal could be measured separately from the input signal. Consequently, its amplitude and phase with respect to the input signal can be determined and from this, the input impedance can be i+r • calculated. Specifically, the input impedance can be calculated as Zin= Zc—1—in Vfwherein Zc= — is the characteristic impedance, defined by the voltage Vfand the 'fcurrent IJfof the input signal, and rin= —is the complex reflection coefficient,Vdefined by the voltage l^-of the input signal and the voltage Vrof the reflected signal.
[0010] The load resistance is detected based on the input impedance, which may imply that the control unit calculates the input impedance. However, in some embodiments, the input impedance may not be calculated explicitly, but thedetection of the load resistance may be based on one or several quantities (e.g. the above-mentioned voltage and current) from which the input impedance could be derived and which imply a certain input impedance. As mentioned above, “detecting” the load resistance may refer to determining the magnitude of the load resistance but could also refer to determining whether the load resistance is within a certain interval or whether the load resistance has undergone a (significant) change.
[0011] The inventive device therefore enables resistance detection without the need for a galvanic connection between the load resistance and the control unit. In some embodiments, this may help to protect the control unit from electromagnetic influences on the side of the load resistance. In other embodiments, the lack of a galvanic connection may facilitate disassembly of the device, e.g. in order to remove and replace the second circuit. Also, the resistance detection can be realized by relatively simple measurements and signal processing and / or calculations.
[0012] While reference is made to a first circuit, a first substrate, a second circuit, a second substrate etc., this includes the possibility that there may be more than one such element. In particular, there may be a plurality of second circuits, each comprising a second coupling device that may form a dedicated coupling arrangement with a first coupling device of the first circuit. The first circuit may comprise a dedicated first coupling device for each second coupling device. Alternatively, one first coupling device may be associated with a plurality of second coupling devices.
[0013] Although the structure of the input signal is generally not limited, it is preferred that the control unit is adapted to apply the input signal so that it comprises an alternating signal with at least one input frequency. More specifically, the input signal may be an alternating signal. Preferably, it is a sinusoidal signal that can be characterized by a single input frequency. However, in some embodiments, the input signal could comprise upper harmonics and / or a constant component (e.g., a direct-current component). In this case, the input frequency is the base frequency of the signal. However, the input signal could comprise a plurality of input frequencies, i.e., it could be any superposition of sinusoidal signals. Depending on the embodiment, a wide variety of input frequencies can be used. By way of example, the input frequency may be between 300 kHz and 1.5 GHz, or between 1 MHz and 500 MHz, but lower or higher frequencies are also conceivable. In someembodiments, the control unit may be configured to provide a superposition of sinusoidal signals with different input frequencies as an input signal, and to analyze the input impedance separately for each of the input frequencies used.
[0014] In one embodiment, the device comprises a first substrate with the first substrate surface and a second substrate with the second substrate surface. In this embodiment, each circuit is disposed on its own, dedicated substrate. However, this does not exclude the possibility that the two substrates may be permanently connected, e.g. bonded, to each other. Preferably, in operational state at least one substrate is interposed between the circuits in the coupling direction. I.e., in operational state, the first substrate and / or the second substrate is interposed between the first circuit and the second circuit. Accordingly, the interposed substrate provides or at least contributes to the galvanic isolation. However, it is also conceivable that the first substrate surface and the second substrate surface are disposed to face each other, in which case the galvanic isolation could be provided by an additional element.
[0015] In another embodiment, the device comprises a first substrate with the first substrate surface and the second substrate surface opposite the first substrate surface, whereby the first substrate is interposed between the circuits in the coupling direction. In other words, in this embodiment, both circuits are disposed on the same (first) substrate. However, since they are disposed on opposite surfaces, the first substrate is interposed between the first and the second circuit and provides galvanic isolation. This embodiment may also be advantageous in that the distance between the two circuits in the coupling direction is well-defined by the interposed substrate. This in turn leads to a stable coupling between the first and second coupling device through the electromagnetic field.
[0016] It is highly preferred that at least one substrate is made of flexible sheet material. Possible materials for a flexible foil include plastic, rubber, silicone or the like. The thickness of such a foil may be less than 1 mm, less than 0.1 mm or even less than 0.01 mm. However, apart from a foil, sheet materials like fabric or other textile materials, fibre-polymer composite materials or the like can be used. The circuit, including the coupling device that is disposed on the substrate surface of the respective substrate, can be applied to the substrate by flex circuit technologies. This may refer to, e.g., screen printing, inkjet printing, flex PCB or etched aluminumlaminates on polymer foil. In preferred embodiments, the first circuit and / or the second circuit may be referred to as a printed circuit. Likewise, the first coupling device and / or the second coupling device may be referred to as a printed device or printed structure.
[0017] As mentioned above, the first substrate and the second substrate may be connected to each other. Such a connection also helps to ensure proper positioning of the coupling devices relative to each other. This refers to a positioning parallel to the first and / or second substrate surface as well as to a positioning in the coupling direction. In one embodiment, the first substrate and the second substrate are detachably connected in operational state. A detachable connection could be realized by snap fasteners, hook-and-loop fasteners or the like. However, the substrates could also be non-detachably connected. This could specifically be achieved by techniques like gluing or ultrasonic welding. Specifically, the substrates could be connected by an insulating bonding layer. Such an insulating bonding layer could also extend over the first circuit and / or the second circuit, thus contributing to the galvanic isolation. Also, the bonding layer helps to guarantee a stable, defined distance between the coupling devices with respect to the coupling direction.
[0018] According to one preferred embodiment, the first coupling device and the second coupling device are coil devices adapted for magnetic coupling and the coupling arrangement comprises at least one coupling transformer. The term “coil device” is not to be construed in a limiting way. The respective device does not have to represent a spiral shape or helical shape. As mentioned above, the thickness of the coupling devices is preferably very low, wherefore a coil device may have a shape that is more two-dimensional than three-dimensional. One possible design of a coil device comprises a spiral shape with the innermost part of the spiral being connected to the outside by a conducting portion that traverses the windings of the spiral. Such a structure necessitates a multi-layer design in which the traversing portion is electrically insulated from the windings. Another embodiment corresponds to a pair of parallel conducting portions that form a spiral shape and are connected at the innermost part of the spiral. Such a structure removes the need for a traversing portion and a multi-layered design. In both cases, the windings of the spiral could have an arcuate, quasi-circular shape or a polygonal, e.g., rectangular shape. Another possibility would be a meandering shape with a plurality ofconducting portions that are alternatingly connected at opposite ends. Two coil devices together form a coupling transformer. Accordingly, the coupling device comprises at least one such coupling transformer, or preferably is a coupling transformer. It will be understood that the electromagnetic field is predominantly a magnetic field, although non-negligible electric field components may be present. Apart from the coil device, each of the circuits may comprise a capacitor connected parallel thereto. Such a capacitor may be used to adapt a resonance frequency of a circuit formed by the coil device and other parts of the circuit. In this context, it has to be considered that conducting portions connecting the coil device to the respective terminal also have non-negligible resistive, capacitive and / or inductive properties which need to be taken into account.
[0019] As mentioned above, the electromagnetic field may have an effective range that is relatively small. Specifically, in case of a magnetic coupling, it may be advantageous if the magnetic field does not extend too far from the first coupling device. Of course, it has to extend far enough to induce a sufficiently large voltage in the second coupling device. On the other hand, a large extension of the magnetic field could induce eddy currents in neighboring components, which would be detrimental to the operation of the device. One embodiment provides that at least one coil device comprises a plurality of conductor sections which extend in parallel to each other on a substrate surface, wherein preferably neighboring conductor sections are connected for opposite current flow. Such conductor sections could form a meandering structure or a double-spiral structure as described above. During operation, current flows through neighboring conductor sections in opposite directions, wherefore the magnetic fields generated by the individual sections cancel each other with increasing distance from the coil device. Accordingly, the effective range of the total magnetic field is limited.
[0020] If the inductance of the first coupling device and the second coupling device are LTI and LT2, and the mutual inductance of the transformer is Lmutuai, the coupling coefficient k is defined as, mutuaik = - LTI ’ L 2According to one embodiment, at least one coupling transformer has a coupling coefficient of at least 0.9 or at least 0.95. In other words, the coupling coefficientmay be equal to or at least close to 1. As will be explained below with respect to preferred embodiments, this may facilitate precise detection of the load resistance. For a coupling coefficient close to 1 , the input impedance as a function of the input frequency may have a single resonance, which may be referred to as the evenmode resonance. However, in other embodiments, the coupling coefficient may be less than 0.9, or even less than 0.7. In those cases, there may be two resonances, namely the even-mode resonance and an odd-mode resonance.
[0021] According to another embodiment, the first coupling device and the second coupling device each comprise at least one electrode adapted for capacitive coupling and the coupling arrangement comprises at least one coupling capacitor. It is understood that each coupling capacitor comprises at least one electrode of the first coupling device and at least one electrode of the second coupling device. In this embodiment, the electromagnetic field is predominantly electric, i.e., it acts predominantly by electrostatic induction or electrostatic influence, although non-negligible magnetic components may be present. Preferably, the coupling arrangement comprises two coupling capacitors. Each electrode may be constituted by a two-dimensional conductive structure on the respective substrate surface. The shape of the electrical could be rectangular, circular, or the like. To provide optimal coupling, it is preferred that the electrodes of one coupling capacitor have the same shape and size.
[0022] According to one embodiment, the control unit is adapted to detect the load resistance based on an absolute value of the input impedance. This may be advantageous in case of inductive coupling as well as in case of capacitive coupling. As mentioned above, the input impedance is generally a complex quantity, which can be described by a real part and an imaginary part or by an absolute value and a phase (which may also be referred to as the argument). As will be explained with respect to preferred embodiments, the absolute value of the input impedance may have a clear dependence on the load resistance. For example, for a certain input frequency, the absolute value of the input impedance may increase as a function of the load resistance. For instance, this may be true in case of inductive coupling if the input frequency is at least close to the even-mode resonance frequency. However, in some cases, the absolute value may also decrease as a function of the load resistance, e.g., in the case of capacitive coupling, when the frequency is atleast close to the frequency of the resonance feature with the smallest frequency. Evaluating the absolute value of the input impedance may not only enable a detection of a change in the load resistance but also to determine a specific value for the load resistance. For instance, the relation between the absolute value and the load resistance could be determined by a series of calibration measurements. The results of the calibration could then be stored in a lookup table of the control unit.
[0023] In another embodiment, the control unit is adapted to detect the load resistance based on a phase of the input impedance. This may also be advantageous in case of inductive coupling as well as in case of capacitive coupling. In some cases of inductive coupling, e.g., for an input frequency that is smaller (or larger, respectively) than the above-mentioned even-mode resonance frequency, the phase may increase (or decrease, respectively) as a function of the load resistance. In case of capacitive coupling at frequencies in a range below, respectively above the frequency of the resonance feature with the smallest frequency, e.g., the phase is a monotonically decreasing, respectively increasing function of the load resistance. Like the absolute value, the phase (or argument) may be used to detect a change in the load resistance and to determine a specific value for the load resistance.
[0024] Another embodiment provides that the control unit is adapted to detect the load resistance based on a derivative of the phase of the input impedance as a function of the input frequency. This embodiment implies that the control unit does not only apply a single input frequency but a plurality of input frequencies, e.g., one after another. Then, with knowledge of the phase as a function of the input frequency, the derivative can be calculated.
[0025] It should be noted that the absolute value, the phase and the derivative of the phase may not only be used individually but also in combination with each other to detect the load resistance.
[0026] The behavior of the input signal and the induced signal depends on various parameters. One parameter is the wavelength of the input signal in comparison to the dimensions of the first and second circuit. If the wavelength is long, i.e., the input frequency is low, signal propagation effects can be neglected. This may not be the case if the wavelength or, more precisely, a quarter of the wavelength is comparableto or smaller than the total length of the signal lines in the first circuit and the second circuit. Simulations based on a transmission-line model show that the absolute value, the phase and / or the derivative of the phase have a distinctive dependence on the load resistance for certain input frequencies that may be referred to as “characteristic frequencies”. For these characteristic frequencies, there is a distinct relation between the phase velocity and the total length of the signal lines. According to one embodiment, the control unit is adapted to apply the input signal with an input frequency that differs by less than 10%, preferably less than 5%, from a characteristic frequency fn=(n+1)Pp, wherein vPis the phase velocity of the signal, / is a total length of signal lines in the first and second circuit, and n is a nonnegative integer. This embodiment may be particularly advantageous if the load resistance is detected based on the absolute value of the input impedance. In this case, the absolute value at the characteristic frequencies may show significant differences for different load resistances.
[0027] In another embodiment, the control unit is adapted to apply a signal with a frequency between zero and a lowest characteristic frequency f0= —1 , or between two characteristic frequencies fn=(n+1)Pp, wherein the frequency preferably differs from the closest characteristic frequency by at least 5%, preferably at least 10 %. This embodiment may be particularly advantageous if the load resistance is detected based on the phase of the input impedance. In this case, the phase at the characteristic frequencies may only show minor differences for different load resistances. However, at frequencies that differ sufficiently from the characteristic frequencies, there may be considerable differences for different load resistances.
[0028] The invention also provides a method for resistance detection, using:a first circuit disposed on a first substrate surface and comprising a first terminal and a first coupling device that is electrically connected to the first terminal and is adapted to generate an electromagnetic field extending in a coupling direction perpendicular to the first substrate surface in response to an electric input signal applied to the first terminal, anda second circuit disposed on a second substrate surface and comprising a second terminal for connection to a load resistance and a second coupling device electrically connected to the second terminal, the second circuit beinggalvanically isolated from the first circuit, the electromagnetic coupling devices being positioned opposite each other in the coupling direction to form a coupling arrangement.
[0029] The method comprises at least the following steps:applying the input signal to the first terminal so that the coupling arrangement generates an induced signal in the second circuit in response to the input signal; anddetecting the load resistance based on an input impedance at the first terminal.
[0030] All these terms have been explained above with reference to the inventive system and therefore will not be explained again. It will be understood that the above-mentioned method steps can be performed by a control unit that can be connected to the first terminal. Preferred embodiments of the inventive method correspond to those of the inventive device.Brief Description of the Drawings
[0031] Further details and advantages of the present invention will be apparent from the following detailed description of not limiting embodiments with reference to the attached drawing, wherein:Fig.1 a top view of a first embodiment of an inventive device for resistance detection;Fig.2 is a top view of elements of the device of fig.1 ;Fig.3 is a cross-sectional view of the device of fig.1 ;Fig.4 is a diagram of an equivalent circuit for the device of fig.1 ;Fig.5 is a top view of elements of a second embodiment of an inventive device; Fig.6 is a top view of elements of a third embodiment of an inventive device; Fig.7 is a top view of elements of a fourth embodiment of an inventive device; Fig.8 is a diagram showing an absolute value of an input impedance as a function of frequency, for different values of a load resistance;Fig.9 is a diagram showing an argument of the input impedance as a function of frequency, for different values of the load resistance;Fig.10 is a diagram showing the absolute value of the input impedance at an evenmode resonance as a function of load resistance, for different values of a coupling coefficient;Fig.11 is another diagram showing the absolute value of the input impedance at the even-mode resonance as a function of load resistance, for different values of a coupling coefficient;Fig.12 is another diagram showing the absolute value of the input impedance as a function of frequency, for different values of the load resistance;Fig.13 is another diagram showing the argument of the input impedance as a function of frequency, for different values of the load resistance;Fig.14 is another diagram showing the absolute value of the input impedance as a function of load resistance, for different values of a coupling coefficient; Fig.15 is a top view of elements of a fifth embodiment of an inventive device; Fig.16 is a diagram of an equivalent circuit for the device of fig.15;Fig.17 is another diagram showing the absolute value of the input impedance as a function of frequency, for different values of the load resistance;Fig.18 is another diagram showing the argument of the input impedance as a function of frequency, for different values of the load resistance;Fig.19 is another diagram showing the absolute value of the input impedance as a function of frequency, for different values of the load resistance;Fig.20 is another diagram showing the argument of the input impedance as a function of frequency, for different values of the load resistance;Fig.21 is a diagram showing the absolute value of the input impedance as a function of load resistance, for different frequencies;Fig.22 is a cross-sectional view of a sixth embodiment of an inventive device; and Fig.23 is a top view of a seventh embodiment of an inventive device.Description of Preferred Embodiments
[0032] Fig.1 is a schematic top view of a first embodiment of an inventive device for resistance detection. The device comprises a first substrate 10 and a secondsubstrate 20. A first circuit 11 is disposed on a first substrate surface 10.1 , while a second circuit 21 is disposed on a second substrate surface 20.1. Both substrates 10, 20 are disposed on top of each other with respect to a coupling direction D that is perpendicular to the first substrate surface 10.1. The substrates 10, 20 may consist of polymer foil or sheet materials. Other flat materials including fabrics, other textile materials or fibre-polymer composite materials and laminated stacks of materials in a foil or sheet form can also be used. During assembly, the two substrates 10, 20 are bonded one to the other by the localized application of a liquid adhesive or using a laminated adhesive layer in one or several bonding areas 18, 28. These bonding areas 18, 28 may be disposed in the vicinity of coupling devices 13, 23 of the respective circuit 11, 21. However, other methods for attaching mechanically the first substrate 10 to the second substrate 20 using the bonding areas 18, 28 can also be used. These include point welding, ultrasound soldering, and riveting or clinching techniques. In some applications, the interconnection may not be intended to be permanent, e.g., because one of the two substrates 10, 20 is occasionally or frequently exchanged, or has to be removed temporarily for cleaning purposes. In these cases, a detachable connection may be used, e.g. using snap-button-fasteners, hook-and-loop fasteners or the like.
[0033] As can be seen in fig.2, which shows the two substrates 10, 20 separately, and in the sectional view of fig.3, each circuit 11, 21 comprises a coupling device 13, 23. In this embodiment, both coupling devices 13, 23 are coil devices 14, 24 which are configured to form a coupling arrangement 30 for inductive coupling, in this case a transformer 31. Both coil devices 14, 24 are realized with a conductive trace or line in the shape of a round Archimedean spiral. Standard printed electronics techniques may applied to form the coil devices 14, 24, and connection lines 12, 22 of the respective circuit 11, 21. These conductive structures may be deposited using printing techniques such as screen printing or inkjet printing with conductive inks, or they may be formed by lithography and etching processes applied to a planar layer of conductive material deposited on the substrate 10, 20. To electrically contact the inner side of the spiral-shaped coil device 14, 24 to the respective connection line 12, 22, a patch 17, 27 of electrically insulating dielectric layer is applied over a part of the printed electronics coil structure in a second processing step. The patch 17, 27 is then printed over in a third processing step witha conductive ink line to realize the electrical contact with the inner side of the spiral coil structure. Alternatively, the electric contact to the inner side of the spiral coil structure could be realized using through contacting with conductive vias in multilayer printed electronics realizations. Additionally, positioning marks (not shown in Figs.1 and 2) can be applied to both substrates 10, 20 to facilitate precise positioning during the assembly process. A precise positioning of the two substrates 10, 20 is preferred as it ensures a good and reproducible inductive coupling, wherefore the transformer 31 is guaranteed to have a near optimal and reproducible mutual inductance. An optimum mutual inductance is also facilitated by the coil structures having close to identical size and geometry.
[0034] As shown in fig.1 , the first coupling device is connected to a control unit 50 via a first terminal 16. The control unit 50 is adapted to apply an input signal, e.g., a sinusoidal voltage signal, to the first terminal 16. Also, the control unit 50 is adapted to determine an input impedance Zin at the first terminal 16, which is the ratio of the voltage Vin at the terminal 16 and the current hnflowing out through the terminal 16. This is also shown in the equivalent circuit diagram of fig.4, which will be explained in more detail further below. A load resistance Ri_, which may be a resistive element 40 like a sensor cell or the like, is connected to the second coupling device 23 via a second terminal 26. In fig.4, the load resistance RL is shown as part of a load impedance Zi_, although the capacitive and inductive properties of the resistive element 40 may be negligible. Due to the input signal, the first coupling device 13 generates a magnetic field which induces a voltage in the second coupling device 23. This leads to an induced signal in the second circuit 21 and a current flow through the load resistance RL. Therefore, the load impedance ZL also depends on the load resistance RL. In the following, the side of the device 1 with the first circuit 11 will also be referred to as the “primary side” while the side with the second circuit 21 will be referred to as the “secondary side”.
[0035] Fig.5 is a schematic top view of a part of an inventive device 1 according to a second embodiment. This embodiment differs from the first one in the shape and size of the coil devices 14, 24. In this case, both coil devices 14, 24 have rectangular shapes with a stepwise and uniform change in the length and the width with each turn, instead of the Archimedean spiral. The number of loops of the coil devices 14, 24 forming the transformer 31 in Figs.1, 2 and 5 is arbitrarily chosen to be eight.However, this is just an example and the number of loops as well as the shape and geometric parameters can be chosen to meet the requirements of the application, e.g., in view of size constraints and the required ranges for the mutual inductance and the self-inductance of the coil devices 14, 24.
[0036] Fig.6 shows a third embodiment of an inventive device 1. In this example, both the first coupling device 13 and the second coupling device 23 comprise a single-loop coil with a meandering or toothed shape similar to a comb, in which the separation distance between two neighbouring conductor section with opposite directions of the current flow is small. Because it is a single-loop coil, a dielectric layer patch 17, 27 and a conductive line printed over it are not needed, unlike the first two embodiments. This simplifies the manufacturing procedure. A second advantage of this type of single loop coil is that the magnetic field decreases in the coupling direction D within a short distance from the substrate surface 10.1, 20.1. This is important, e.g., when only a small and flat volume is available for the device 1 and neighbouring elements (not shown) are made of electrically conductive material or containing conductive components, e.g., printed circuit boards, or metal housing parts. In these cases, namely, the magnetic field could generate eddy currents of considerable magnitude in the conductive structures, which would significantly reduce the magnetic flux that can be generated in the transformer and thereby also limit the effectively attainable self-inductance and mutual inductance of the coil devices 14, 24. Furthermore, the excitation of eddy currents in neighbouring conductive structures leads to unnecessary losses of electrical power, which can be largely avoided by using a coil structure as shown in fig.6, while still ensuring sufficiently large values for the mutual inductance and self-inductance.
[0037] Similar single loop coil geometries of the type shown in fig.6 can be devised by first elongating a single loop until it becomes very oblong and then folding the oblong loop shape to fit different paths, for example, a meandering path, or a spiralshaped path that fills the surface area of the printed electronics transformer. Fig.7 shows a further example of an interconnection using inductive coupling through a printed transformer 31 with a single loop coil geometry, where the elongated loop is wound along a spiral-shaped path. This geometry also helps to reduce the effective range of the magnetic field. For the single loop coil geometries of the type shown in figs. 6 and 7, the tolerance for the positioning of the first and the second substrate10, 20 upon each other during the assembly is on the order of a fraction of the distance separating two neighbouring printed conductive ink lines forming the loop of the coil, in order to ensure near optimal values for the mutual inductance. For printed transformers 31 based on coil devices 14, 24 as shown in figs.1, 2 and 4, the positioning tolerance during the assembly process can be considerably larger, typically on the order of a fraction of the diameter or width of the smallest loop of the coil.
[0038] With typical values for the width of the conductive trace lines and the separation distance between two neighbouring conductive traces on the order of 1 mm, the inductance that can be realized with printed coils with multiple windings using a surface area of 10 cm2are on the order of 1 |iH. With single loop coil geometries of the type shown in figs. 6 and 7 and using the same surface area of 10 cm2on the printed electronics sheet inductance values of the coil structure on the order of 100 nH can be achieved. As for other conventionally known transformer types, the printed transformers of the types described above can also contain multiple printed coil structures in the first and / or the second circuit 11, 21, and printed coil structures with a middle tap or with multiple taps are also conceivable.
[0039] Getting back to the equivalent circuit diagram in fig.4, a central component of the diagram is the transformer 31 that is realized by the overlapping coil devices 14, 24 when the two substrates 10, 20 are superposed upon each other during assembly. The secondary side of the transformer 31 consists of the second coil device 24 printed on the second substrate 20 that has an inductance LT2 and a serial resistance RT2. AS a simplification of the circuit model, the capacitance arising from the juxtaposition of the printed traces forming a loop or the loops of the coil structured is considered as a single capacitor connected in parallel with the coil and contributes to the parallel capacitance C2. The load impedance ZL or load resistance RL is connected to the secondary side of the transformer 31 via printed electronics lines (like connecting lines 22) with a serial resistance RS2 and a serial inductance LS2. The capacitance arising from these printed electronics lines are also assumed to be included as a further contribution to the parallel capacitance C2.
[0040] The load impedance ZL consists mainly of the load resistance RL considered to obey Ohm’s law, but can include possibly a further serial inductance LL and a further capacitance CL connected in parallel that can be thought of to arise fromfurther printed circuit lines and components that are possibly needed to complete the second circuit 21. The resistance RL may be the resistance of a part of a printed electronics circuit and / or of a component of the circuit established on the second substrate, for example, a sensor component like a thermistor, or a pressure sensor, or simply a switch. In these cases, the value of the resistance RL can thus change, and the main function of the device 1 is to determine the resistance RL and / or to detect a change in RL.
[0041] The first coil device 14 is characterized in the equivalent circuit diagram by an inductance LTI and a serial resistance RTI . The capacitance arising from the juxtaposition of the printed traces forming the coil structure on the primary side of the transformer are taken into account as a contribution to a capacitance Ci connected in parallel to the inductor LTI. The capacitance Ci includes a further contribution arising from the circuit lines printed on the first substrate (like the connection lines 12). These printed circuit lines connecting the primary side of the transformer are characterized by a serial resistance Rsiand a serial inductance Lsi. Additional auxiliary capacitances Cxi, CX2 can be connected in parallel with the capacitances Ci and C2. These capacitances Cxi, CX2 represent auxiliary capacitors that can be added to the circuit if needed for purposes of facilitating the determination of the resistance RL and the detection of a change of the resistance RL, which will be explained further below. The auxiliary capacitors can be added to the circuit, for example, in the form of SMD components or realized by printed electronics techniques on the substrates.
[0042] A good coupling between the printed coil structures on the primary and secondary side of the transformer 31 can be realized yielding large values of the mutual inductance Lmutuai and values for the coupling coefficient k = Lmutual / LTI ■ LT2 )1 / 2close to unity, when overlap mismatching between the two coil structures is avoided, and also conductive elements are not mounted too close to the printed transformer’s coil structures such that the excitation of Eddy currents in them, which reduce the effective magnetic field flux, is minimized.
[0043] The values for the serial resistance Rsiand RS2, and similarly RTI and RT2, for the serial inductance Lsiand l_S2, as well as for the capacitance contributions to Ci and C2 stemming from the printed circuit lines depend on the application and can vary largely depending on the details of the choices made for the materials like theconductive ink and substrate material and also on details of the chosen layout of the printed circuit, for example, the width and length of the printed circuit lines and the distances between line pairs. Typical values for the serial resistance Rsiand RS2 as well as RTI and RT2 are on the order of 10 Q / m of printed circuit line. For the serial inductance Lsiand l_S2, values on the order of a few times 100 nH / m of printed line pair can be assumed. The line capacitance of the parallel line pairs contributes on the order of a few times 10 pF / m to the parallel capacitance Ci and C2.
[0044] The resistance RL may represent the output of a sensor or a switch, of which the device 1 allows to detect the state and changes of that state, and can have values from a broad possible range from, e.g., 1 Q or less to 100 kQ or more. Depending on the application, a configuration of the device 1 can be arranged to either allow for the detection of the value of the resistance RL and continuous changes of that value or to detect in which state the resistive element 40 is out of two or more different states defined by different resistance value intervals for the resistance RL. For example, values of RL smaller than 1 kQ could result in the detection of an “on” state, while values larger than 1 kQ represents an “off” state.
[0045] As mentioned above, the control unit 50 is configured to determine the impedance Zin = Vin / lin at at least one input frequency f, at several discretely defined input frequencies, or also in a range of frequency values over which the input frequency f is swept. Both an amplitude and a phase of Zin as a function of frequency contain information on the load impedance ZL and on all the other elements shown in the diagram of fig.4. Depending on the requirements of the application and the layout of device, it may be possible that the only the information on the amplitude, or only the information on the phase, is sufficient to determine the value of the load resistance RL.
[0046] Looking at the equivalent circuit diagram in fig.4, it will be understood that the primary coil structure of the transformer 31 with the inductance LTI forms an LC circuit or tank circuit with the two capacitors Ci and Cxi. The serial resistance RTI of the coil functions as a damping resistor inside this circuit. Similarly, the coil structure of inductance LT2 on the secondary side of the printed transformer 31 forms a tank circuit with the capacitances C2 and CX2, wherein the serial coil resistance RT2 represents a damping term. The load impedance ZL is also connected to the secondary side of the transformer 31 via the serial resistance Rs2 and the serialinductance Ls2 forming a further parallel circuit with the capacitances C2 and CX2 and the inductance LT2. The effect of the load resistance Ri_, which represents the main contribution to Zi_, is that of an additional damping of resonator circuit oscillations. The tank circuits on the primary and the secondary side of the transformer 31 thus form a system of two coupled resonator circuits, in which the strength of the coupling between the two resonators is determined by the coupling coefficient k of the transformer 31.
[0047] The input impedance Zin as a function of the input frequency f and of the load impedance Zi_, or more precisely of the load resistance Ri_, constitutes a complex function arising from the rather intricate behaviour of the system of coupled resonators. How the system of coupled resonators reacts to an electronic excitation at the first terminal 16 depends sensitively on the coupling coefficient k as well as on the different parameters determining the equivalent circuit model and in particular on the value of the load resistance RL.
[0048] In some embodiments, the best sensitivity for the determination of the RL and changes in RL can be obtained when the device 1 is configured such that its circuit model is close to being symmetric in the sense that the resonance frequencies of the above-mentioned two resonator circuits are approximately equal or lie relatively close to each other within a relatively narrow frequency range. For realizations of the inventive device 1 that can be described by the equivalent circuit model shown in Fig.4, the resonance frequencies fi, f2 for the tank circuit on the primary side and on the secondary side of the transformer are approximately givenby 1 / (2TT(LC)1 / 2), where L is essentially given by LTI on the primary side and by LT2 on the secondary side, and where C is in essence Ci+Cxion the primary side and C2+CX2 on the secondary side. A parallel load capacitance CL that is possibly contained in the load impedance ZL can also be added to C2+CX2 to be more precise. By using a complete equivalent circuit model for a specific realization of the inventive device 1 , the precise values for the resonance frequencies fi, f2 can be determined without approximation from circuit simulations or by solving the set of equations underlying the equivalent circuit model. While the capacitances Ci and C2 are mainly fixed by the layout of the printed electronics circuit, the auxiliary capacitances Cxiand CX2 can be freely chosen to shift the resonance frequencies fi and f2 such thatthey are approximately equal and lie in a frequency range that may be preferred due to other requirements of the application.
[0049] Although realizations of the invention are possible in which the inductances LTI and LT2 of the first coil device 14 and of the second coil device 24 have values that differ from each other, for many cases of possible applications of the inventive device 1 the values for LTI and LT2 are preferentially chosen to be equal or approximately the same. This allows the largest possible mutual inductance Lmutuai of the transformer 31 to be achieved within the constraints of available surface area for the printed coil structures, and thereby a better sensitivity for the determination of RL is obtained.
[0050] In the case of a strong coupling of the two above-mentioned resonator circuits, i.e., when the coupling coefficient k at least close to, or even equal to, unity, the input impedance Zin as a function of frequency is largely dominated by a single resonance feature. This resonance feature corresponds to the even mode resonance of the system of coupled resonators, in which the currents on the primary and the secondary side of the transformer h and h oscillate essentially in phase, i.e., with a phase difference smaller than 90°. In other words, at the even-mode resonance, the direction of the current I2 oscillating on the secondary side of the transformer 31 with respect to the direction of the current h on the primary side is such that the voltage induced on the primary side by I2, namely, Lmutuai■ 2 i • f • I2, has the same sign as the self-induction voltage LT1■ 2 i • f •generated by the current h on the primary side of the transformer. When the resonance frequencies fi and f2 are about the same and the coupling coefficient k is close to unity, it can be inferred from the equivalent circuit model that the even-mode resonance occurs at a frequency fethat is smaller than fi or f2 by a factor of approximately 1 / V2. The even-mode resonance feature is the more pronounced the larger the load resistance RL is. The reason for this is that the load impedance ZL, in which the load resistance RL can be considered as the essential contribution, is connected in parallel via the serial resistance RS2 and inductance LS2 to the capacitance component of the resonator circuit C2+CX2 on the secondary side of the transformer 31 and thus acts as a bypass for the current. With larger values of the load resistance RL, the portion of the current I2 that is bypassed through RL is decreased and more energy peroscillation cycle can be stored in the capacitors of the tank circuit, which reduces the damping effect of RL.
[0051] By decreasing the load resistance RL to small values, the even-mode resonance feature in Zin becomes less prominent, and it actually vanishes from the spectrum of Zin when the magnitude of the impedance of the current bypass formed by the load impedance ZL, the serial resistance RS2 and inductance LS2 becomes smaller than the magnitude of the impedance of the capacitance of the tank circuit on the secondary side, which is approximately given by ((C2+ Cx2) ■ 2■ This is because the damping effect of RL hampers the resonant oscillations in the tank circuit on the secondary side of the transformer 31.
[0052] The even-mode resonance feature in the input impedance Zin is characterized by a peak in the amplitude and the changing of the phase from a value close to +90°, characteristic for an inductance dominated impedance, to a value close to -90°, characteristic of a capacitance dominated impedance, as the frequency is swept in positive direction over the range containing the even-mode resonance feature. Exemplary curves of the amplitude |Zjn| and the complex argument or phase arg(Zin) of the input impedance Zin are plotted versus the frequency for different values of the load resistance RL in figs. 8 and 9, respectively. These curves are obtained from equivalent circuit model calculations of Zin using an example of a configuration of the inventive device 1 with typical values for the circuit parameters, except for the values of the serial resistances Rsiand RS2 and the serial inductances Lsiand l_S2, which were assumed to be negligibly small for the sake of simplicity.
[0053] In figs.10 and 11, the peak value of |Zjn| for the even-mode resonance is plotted as a function of the load resistance RL as obtained from the equivalent circuit model calculations. The peak value of the amplitude |Zjn| at the even-mode resonance is a monotonically increasing function of the load resistance RL with a gradually decreasing slope as RL is increased to the larger values as well as when RL is decreased to the smaller values. As is shown in fig.11 , even in the range of small values of RL, for which the resonant oscillations in the tank circuit on the secondary side of the transformer 31 are prevented by the damping effect of RL and the even mode resonance feature vanishes from the spectrum of |Zjn|, the value of |Zin| at the even-mode resonance frequency feis a monotonically increasing functionof RL that can be used by the control unit 50 to sensitively determine the value of RL. For large values of RL, the equivalent circuit model calculations show that the peak value of |Zjn| at the even-mode resonance is gradually levelling off for RL values above 1 MQ.
[0054] As can be concluded from the curves in fig.9 for the different values of the load resistance RL, the phase of the input impedance Zin also represents a useful measure for the value of the load resistance RL. The phase itself is a monotonically increasing function of RL at frequency values that can be chosen in a broad range below the even-mode resonance frequency fe, and it is a monotonically decreasing function of RL when the frequency value is chosen in a broad range above the evenmode resonance frequency fe. At the even-mode resonance frequency fe, the phase of Zin appears to be close to zero invariant^ of the value of the load resistance RL. However, its derivative with respect to frequency, as can be seen from the slope of the curves in fig.9, provides a function of which the absolute value is monotonically increasing with the load resistance and can also serve as a sensitive measure for the value of RL.
[0055] The strong coupling of the system of two coupled resonators realized on the primary and secondary side of the printed transformer 31 cannot always be ensured, and there may also exist applications for which it can be of advantage when the layout is deliberately adapted to realize a specific value of the coupling coefficient k that is not close to unity. When the coupling strength between the two resonator circuits on both sides of the transformer 31 is decreased, the system of coupled resonators also shows, apart from the even-mode resonance, an odd-mode resonance.
[0056] For the odd resonant mode, the currents h and h on the primary side and on the secondary side, respectively, are oscillating essentially in phase opposition, i.e., with a phase difference that is larger than 90°. In opposition to the case of the even-mode resonance, in the case of the odd-mode resonance, therefore, the voltage induced on the primary side by I2, namely, (Lmutuai■ 2 i • ) • I2, has the opposite sign as the self-induction voltage (LT1■ 2 i • ) •generated by the current h on the primary side of the transformer 31. While the even-mode resonance frequency feis smaller than the resonance frequency fi or f2 of the individual uncoupled tank circuits, the frequency f0of the odd-mode resonance is larger thanfi or f2. In fact, the resonance frequencies of both the even and the odd mode depend on the value of the coupling coefficient k. From the equivalent circuit model with the simplification of negligibly small values for the serial resistances and inductances Rsi , Rs2, Lsi, and l_S2, it can be shown that feand f0lie away from fi or f2 by the factor (1 + fc)-1 / 2and (1 - fc)-1 / 2, respectively. As the coupling coefficient k approaches unity, the odd-mode frequency f0tends to infinity. When k is decreased to zero, feand f0approach the frequencies of the uncoupled tank circuits fi and f2, that are assumed to be closely the same, from the smaller frequency side, respectively from the side of the larger frequencies.
[0057] An exemplary spectrum of the amplitude |Zin| of the input impedance Zin is shown in fig.12 for a coupling coefficient k of 0.5. The curves plotted are obtained from calculations using the equivalent circuit model with the simplifying assumption of negligibly small values for the serial resistances Rsi , and RS2 and for the serial inductances Lsiand LS2. Fig.13 contains the corresponding curves for the complex argument or phase of Zin. Fig.12 clearly shows that the peak values of the amplitude |Zin| of the input impedance Zin at the even-mode resonance as well as at the oddmode resonance are both decreasing gradually as the load resistance RL is decreased from large values to a value close to 200 Q, at which both the even and the odd mode resonance feature appear to vanish from the spectrum of |Zjn| .
[0058] The value for RL close to 200 Q in this example calculation marks the resistance value corresponding to the magnitude of the impedance of the capacitance of the tank circuit on the secondary side of the transformer at the resonance frequency f2, which is approximately given by ((C2+ Cx2) ■ 2 ■ f2)-1= (^T2 / (G + Cx2))1 / 2- isalsoequal to the magnitude of the impedance of the inductance LT2- 2n- f2at the resonance frequency f2. This value for the amplitude of the impedance could also be called characteristic impedance of the circuit that connects the load resistance RL to the secondary side transformer 31 , since its role is similar to that of the characteristic impedance of a transmission line. When the load resistance RL is decreased below this value, more than half of the current h induced in the coil structure on the secondary side is bypassed through the circuit branch containing the load resistance RL that is in parallel to the capacitance C2+CX2 of the tank circuit, such that the oscillations of the resonant circuit are stronglydamped. This explains why the even and odd mode resonance features in Zin disappear when RL is smaller than this particular value.
[0059] As the load resistance RL is decreased below the characteristic impedance corresponding to the distinguished resistance value of 200 Q in this example, a peak arises in the spectrum of |Zjn| between the frequencies feand f0, which is called hereafter the “middle peak” feature. A very remarkable characteristic of this middle peak feature is that its peak value is monotonically increasing as the value of the load resistance RL is decreased down to zero. The middle peak feature can be understood as resulting from the reflection of the input signal at the load resistance RL, which acts gradually more like a short termination of a transmission line as RL is decreased from the value of the characteristic impedance to zero, and thus, the reflection coefficient for the induced signal is increased. The peak value of the middle peak feature in the |Zjn| spectrum is therefore a good measure for the load resistance RL in the range of smaller RL values below the characteristic impedance, where the even and the odd resonance features have both vanished from the spectrum of |Zjn| due to the strong damping effect on the oscillations of the tank circuit on the secondary side of the transformer 31. In fig.14 the value of the amplitude of the input impedance |Zjn| at the central frequency of the middle peak feature is plotted versus the load resistance RL for different values of the coupling coefficient k.
[0060] The appearance of the middle peak feature can also be understood from the equivalent circuit model. For small values of the load resistance RL below the characteristic impedance, the second circuit 21 on the secondary side of the transformer 31 can be viewed as consisting essentially of the inductance LT2 to which the small load resistance RL is connected in parallel, since most of the current h on the secondary side is flowing through the load resistance RL with only a small portion of I2 passing through the parallel circuit branch containing the capacitance C2+CX2 that has an impedance of larger magnitude. When RL is small, the effect of the secondary side of the transformer on the first coil device 14 is very similar to that of a well conducting structure in which eddy currents are induced. The current hon the secondary side is thus in phase opposition to the current h on the primary side and its magnitude is approaching k times h. From the equations of the equivalent circuit model with the simplifying assumption that the serial resistances Rsiand RS2and the serial inductances Lsiand LS2 are vanishingly small and that LTI and LT2 are approximately equal, it can be seen that the effect the secondary side circuit is that the voltage induced at the inductance LTI of the first coil device 14 becomes (LT1■ (1 - fc2) • 2 i ■ f) ■ I . From this it can be further deduced that the central frequency of this middle peak feature in the spectrum of |Zjn| is approximately given by (1 - k2)-1 / 2times the resonance frequency fi or the resonance frequency f2 of the uncoupled tank circuits that are assumed to be closely the same.
[0061] The curves of the amplitude |Zjn| of the input impedance at the even-mode resonance frequency feas a function of the load resistance RL for different values of the coupling coefficient k including the value 0.5 are also plotted in figs. 10 and 11. The curves show that also for values of the coupling coefficient k smaller than unity the value of |Zjn| at the even-mode resonance frequency feremains a useful measure for the load resistance RL over a large range for RL from 0 Q to a few MQ. The value of the amplitude |Zjn| at the odd-mode resonance frequency f0also shows a dependence on the load resistance RL, however, only for values of the coupling coefficient that are further away from unity, for example, 0.5, since the sensitivity to the value of RL decreases as k approaches unity, and it also decreases as k approaches zero. The sensitivity to the value of RL is generally weaker for the amplitude of the input impedance |Zjn| at the odd-mode resonance frequency f0than it is at the even-mode resonance frequency fe. The odd-mode resonance frequency f0as well as the central frequency of the middle peak feature in the |Zjn| spectrum show a strong dependence on the value of the coupling coefficient k, and can therefore be used as sensitive measures for the determination of the value of k.
[0062] In order to further improve the accuracy beyond the equivalent circuit model graphically represented in fig.4, a model extension representing the signal lines on the primary and secondary side using transmission-line theory may be appropriate, in particular when frequencies are considered for which the values of the total length of the signal lines reach the order of a quarter wavelength of the transmitted signals.
[0063] In another embodiment, which is shown in figs. 15 and 16, the coupling arrangement 30 is adapted for capacitive coupling and comprises two capacitors 32. Each capacitor 32 is constituted by one electrode 15 of the first circuit 11 and one electrode 25 of the second circuit 21. In the equivalent circuit diagram of fig.16, the coupling capacitors 32 are represented by capacitances Cd and CC2. During theassembly process, the two substrates 10, 20 shown in fig. 15 are superposed one upon the other with the print side of the substrates 10, 20 either both on the top side or both on the bottom side. When the assembly is done with the print side of the substrates 10, 20 on the same side (top side or bottom side, respectively), the substrate material of the first substrate 10 constitutes a dielectric layer between the two electrodes 15, 25 of the coupling capacitors 32. For a substrate thickness of 0.125 mm and a relative permittivity of 3 for the substrate material, a capacitance of 21 pF per 1 cm2can be realized in this way. In fig.15, the electrodes 15, 25 are rectangular in shape, but other shapes are possible. Preferably, the two electrode areas forming a capacitor 32 are exactly matching to each other and overlap entirely when the two coupling devices 13, 23 of the substrates are overlayed one upon the other during the assembly process. Since the surface area for the printed electrodes 15, 25 of the coupling capacitors 32 is typically large, on the order of 10 cm2, the tolerances in the positioning of one substrate 10, 20 with respect to the other during the assembly process are not required to be tight.
[0064] In fig. 15, bonding areas 18, 28 lie well outside of the area reserved for the electrodes 15, 25. However, it is also possible to apply a liquid adhesive or to use a laminated adhesive layer between the two substrates 10, 20 in the area of the electrodes 15, 25 for the bonding purpose, if the permittivity of the adhesive material is well-known and its dependence on temperature, relative humidity and aging is weak. In this case, the adhesive material contributes together with the substrate material to an effective permittivity which determines the capacitance of the coupling capacitor 32. As the capacitance is also dependent on the distance separating the two electrodes 15, 25 of the coupling capacitor 32, it can be preferable to use a thin adhesive layer between the two substrates 10, 20 that ensures a uniform and constant distance between the electrodes 15, 25 rather than a thin gap filled with air that could lead to changes and nonuniformity of the distance between the electrodes 15, 25 due to mechanical deformation in the application.
[0065] The coupling capacitors 32 could also be realized in a variant form, in which one of the coupling capacitor’s 32 electrodes 15, 25 is split into multiple electrodes. For example, the electrode 25 of the second circuit 21 could be divided into two electrodes with each occupying half the surface area of the electrode 15 of the first circuit 11. Then, each electrode 25 could be part of an individual second circuit 21.Using two of these coupling capacitors 32 for the coupling arrangement 30 a double connection is realized, in which the two different second circuits 21 are connected in parallel and can receive signals from the primary side and / or transmit signals to the primary side and with galvanic isolation between them being assured.
[0066] Fig. 16 is an equivalent circuit model for the case of a coupling arrangement 30 with galvanic isolation based on capacitive coupling. This is in many aspects similar to the equivalent circuit model shown in fig.4 for the case of inductive coupling. In the diagram of the equivalent circuit model for the capacitive coupling given in fig. 16, the same components C2, Rs2, LS2 and ZL on the side of the second substrate 20 and Ci , Rsi, and Lsion the side of the first substrate 20 are shown as in fig.4. The two coupling capacitors 32 represented in fig. 16 by the coupling capacitance Cd and CC2 in the circuit model replace the transformer components represented by LT2 and RT2 as well as LTI and RTI in fig. 4. In the case of the capacitive coupling, the capacitances Ci and C2 do not include the parallel capacitance contributions stemming from a coil structure as described above for the case of inductive coupling but represent only the capacitance of the printed connection lines 12, 22 connecting the coupling capacitors from the side of the first substrate 10, respectively from the side of the second substrate 20. Also, the auxiliary capacitances Cxi and CX2 are not needed in the capacitive coupling case.
[0067] For the embodiment using capacitive coupling a basic description of the operation and functioning of the interconnect can be given using an equivalent circuit model in which the signal lines on both sides of the interconnect are included as lumped elements, and for which a diagram is drawn in fig.16. From the equivalent circuit model the impedance Zin at the first terminal 16 can be obtained as a function of the frequency, either by circuit simulations or by analytically solving the system of equations representing the equivalent circuit model.
[0068] In the case of a coupling arrangement 30 based on inductive coupling the equivalent circuit model contains components with a relatively large inductance that play a dominant role, most notably the coil devices 14, 24. The equivalent circuit model for the case of the coupling arrangement 30 using capacitive coupling contains only the small serial line inductance contributions Lsiand LS2 from the pair of printed connection lines 12, 22 on the primary and on the secondary side of the device 1 that can be several times smaller than the inductance values of the coildevices 14, 24. Similarly, the line capacitance of the printed connection lines 12, 22 times their lengths form the main parallel capacitance components Ci and C2 in the equivalent circuit model in case of capacitive coupling. Auxiliary capacitors as in the inductive coupling case are typically not needed. The pair of coupling capacitors 32 of the coupling arrangement 30 are represented by the capacitances Cd and CC2 that form a serial connection of the conductor pair of the signal lines.
[0069] Fig.17 shows the amplitude |Zjn| of the input impedance as a function of the input frequency for different values chosen for the load resistance RL and for an example configuration of the device 1 with capacitive coupling according to figs. 15 and 16 using typical parameters. The corresponding curves for the complex argument or the phase of Zin are plotted in fig.18.
[0070] For this example of a configuration of the device 1 the capacitances Cd and Cc2 of the coupling capacitors 32 were chosen to be 0.133 nF. The printed connection lines 12, 22 are realized in the model in the form of two parallel conductors on a dielectric substrate 10, 20 using a width of 1 mm for the individual conductor, two times the half print width a of 0.5 mm, and a center-to-center separation s of the conductor pair of 5mm. The lengths h and I2 of the connection lines 12, 22 are set to 0.5 m and 0.75 m, respectively. In the calculations for the example case the values for the capacitance and the inductance per unit length of the signal line were approximated using known expressions for a parallel wire pair and assuming an effective relative permittivity of 1.5 and a relative electromagnetic permeability of one giving 18 pF / m and 0.92 pH / m, respectively. Furthermore, the resistance per unit length of an individual conductor line in a signal line pair was set to 30 Q / m in the example.
[0071] The curves for |Zjn| in fig.17 show a characteristic peak in the frequency range around 40 MHz for smaller values of the load resistance RL, more precisely for values smaller than a value close to 225 Q for the parameter choices made for the presented example calculation. This value corresponds to the characteristic impedance Zc= (Ls2 / C2)1 / 2of the printed signal lines 22 on the secondary side, which has the same value as on the primary side (Lsl / C])1 / 2, since the same configuration of the signal line was chosen. The peak in the |Zjn| curves for small values of RL is due to the resonance of the tank circuit formed by the inductance LS2 on the secondary side that is connected in parallel via the load resistor RL to the twocapacitances Ci and C2. The latter are in turn connected in parallel to each other by the coupling capacitors Cd and CC2, of which the amplitude of the impedance can be considered to be small with a value of about 30 Q at frequencies around 40 MHz. Indeed, the expression for the tank circuit’s resonance frequency (LS2(CI +C2)) / 2TT evaluates to about 40 MHz for the parameter values of the example configuration. The peak in the curves for |Zjn| disappears gradually as the value of RL is increased beyond values around Zc. This can be understood when considering the RC time required to charge the capacitor of the tank circuit via the resistance RL and the resistance RS2 of the connecting printed circuit lines. With increasing values of RL this RC time increases and eventually becomes larger than a period of oscillation of the resonant circuit, which gradually hampers and finally prevents the oscillations of the resonator.
[0072] As the value RL is increased to large values beyond Zc, the current through the serial inductance l_S2, the serial resistance RS2 and the load resistance RL on the secondary side becomes vanishingly small. For the large values of the resistance RL, therefore, the curves of |Zjn| show a characteristic minimum close to a frequency of 50 MHz. This minimum is caused by the resonance of the in-series LC circuit formed by the serial line inductance Lsion the primary side and the capacitance C1+C2 formed by the line capacitance contributions of the primary and secondary side connected in parallel by the coupling capacitors Cd and CC2. At the serial resonance frequency (Lsl• (Cx+ C2)) / (2TT), the complex impedance of the inductance LS1■ 2 i ■ f and of the capacitance 1 / ((CX+ C2) • 2ni ■ f) have equal magnitude and opposite sign, such that the input impedance Zin is approximately given by the serial line resistance Rsion the primary side. For frequencies larger than the serial resonance frequency, the input impedance Zin is dominated by the impedance contribution from the inductance Lsileading to a linear increase of |Zjn| with increasing frequency. For frequencies smaller than the serial resonance, Zin becomes a predominantly capacitive impedance resulting in the characteristic divergence of |Zjn| as the frequency approaches zero. For small values of the load resistance RL< ZC, the input impedance Zin is as well dominated by the impedance of the inductance Lsiat large frequencies according to the equivalent circuit model, and by a behaviour of an essentially capacitive impedance as the frequency tends to zero. Thus, the curves for all values of the load resistance RL exhibit a divergingbehaviour as the frequency approaches zero which also shows the effect of the galvanic isolation of the device 1 with capacitive coupling on the input impedance |Zin|.
[0073] As can be seen in fig. 17, the amplitude of the input impedance Zin exhibits distinctive differences for the different values of the load resistance RL. From measurements of |Zjn| at a frequency close to 40 MHz, for example, the value of the load resistance RL can be unambiguously and clearly inferred in the RL range from 1 Q to 10 kQ, since the curves of fig.17 show a monotonically decreasing dependence of |Zjn| on the value of RL. The curves for an RL of 10 kQ and 100 kQ, however, appear to be nearly indistinguishable. Also, e.g., for a frequency f chosen around 10 MHz or around 15 MHz, a functional mapping of |Zjn| on the value of RL that is monotonic can be recognized, which can be used to determine RL from measurements of |Zjn| . For the choices of the frequency f in that range, the amplitude |Zin| of Zin turns out to be a monotonically increasing function of RL in the interval between 1 Q and 10 kQ, and the derivative of |Zin| with respect to RL appears to be largest in the range from 10 Q to 1 kQ, thus enabling a sensitive determination of RL in that range.
[0074] Also, the complex argument or the phase of the input impedance Zin can be used to determine the value of the load resistance RL in the case of the device 1 based on capacitive coupling. The curves of the phase of Zin calculated for the considered example configuration from the equivalent circuit model that are shown in fig.18 exhibit the largest differences for the different values of RL in the frequency range around 30MHz with a width of about 10 MHz to 20 MHz. At a frequency f in that range, the functional mapping of the phase of Zin to the value RL of the load resistance is strictly monotonically decreasing over the entire interval from 1 Q to 100 kQ for RL. Thus measurements of the phase of Zin at an accordingly chosen frequency f enable the sensitive determination of RL in said range. The derivative of the phase of Zin with respect to the frequency can also serve as a sensitive measure for the value of RL. For example, when choosing the frequency f at which the derivative of the phase of Zin is determined to be in the vicinity of the tank circuit resonance frequency (LS2(CI + C2)) / (2TT) of about 40 MHz, the slope of thecurves in fig.18 clearly show that the derivative of the phase of Zin is a monotonically increasing function of RL.
[0075] From transmission line theory, formulae are known for the input impedance of a transmission line of a given length and with a defined load impedance at its termination as well as for including the resistive serial loss of the transmission line as approximation. With these formulae, a functional model for the device 1 using capacitive coupling for signal transmission that is based on transmission line theory can be established. The formulae from transmission line theory can directly provide an expression for the frequency-dependent impedance of the signal line on the secondary side that can be regarded as a transmission line with the load impedance ZL as the line termination impedance. Treating the coupling capacitors Cd and CC2 for simplicity as lumped elements that are connected in series with the input impedance of the signal line on the secondary side, an expression is obtained for the load impedance that can be considered to terminate the signal line on the primary side. Using this expression for the load impedance terminating the transmission line on the primary side and applying again the known formulae for the input impedance of a transmission line with a given load impedance at its termination the expression for the input impedance Zin according to a transmission line model is defined.
[0076] In fig. 19, the amplitude of the input impedance |Zjn| according to the transmission line model is plotted versus frequency for different values of the load resistance RL. The corresponding curves for the phase or the complex argument of Zin are shown in fig.20. In the transmission line model calculation of Zin for the curves shown in figs. 19 and 20, the parameters are chosen identical to the corresponding values used for the calculations based on the equivalent circuit model for the example configuration of the interconnect described above and for which the results are plotted in figs. 17 and 18. Comparing the curves in figs. 19 and 20 to the corresponding curves plotted in figs. 17 and 18, it can be observed that the curves of the two models show a qualitative similarity and are partially comparable quantitatively for frequencies up to about 70 MHz in the case of this example configuration of the device 1. For larger frequencies the main features present in the curves calculated with the transmission line model are entirely absent in the curves obtained by using the equivalent circuit model with the signal line representation aslumped components. This is not surprising as the equivalent circuit model with a representation of the signal lines using lumped components is expected to give a gradually more inaccurate and incomplete description as the frequency is increased into the range in which a quarter of a wavelength is on the order of the length of the signal lines or smaller. For the example configuration of the device 1 with capacitive coupling considered here an effective relative permittivity of 1.5 is assumed for the realization of the signal lines in the form of a pair of parallel conductors by the printed electronics techniques on a dielectric substrate. From the phase velocity vpfor the signal lines given by the speed of light in vacuum divided by the square root of the effective relative permittivity, the wavelength is obtained by dividing vpby the signal frequency. In the example configuration considered the total signal line length I on the primary and secondary side is 1.25 m, which corresponds to a quarter wavelength for a signal frequency of 49 MHz.
[0077] Apart from being a more accurate description for larger frequencies in particular, the transmission line model also allows for a better understanding of the main features that can be observed in the graphs of the amplitude and the complex argument of the input impedance Zin in figs. 19 and 20. Using the transmission line theory, the reflection of the signal at the load resistance RL is readily understood from the expression for the reflection coefficient. The reflection coefficient is defined as the relative difference of the line terminating load impedance RL and the characteristic impedance Zcof the signal line given by (RL-ZC) / RL+ZcThe reflection coefficient has two obvious and noteworthy properties. Its amplitude is zero for RL= Zc, it is larger the larger the absolute value of the difference between the load resistance RL and the characteristic line impedance Zcis, and it approaches unity for very large values of RL as well as when RL tends to zero. Second, the sign of the reflection coefficient is negative when the load resistance RL is smaller than the characteristic line impedance Zc, which means that the reflected signal obtains an additional phase shift of Pi, and positive for load resistance values larger than Zc. For the example configuration of the device 1 using capacitive coupling considered here the characteristic impedance Zcof the signal lines on the primary and the secondary side is 225 Q as given by the square root of the ratio of the line’s inductance per unit length to the line’s capacitance per unit length.
[0078] The curve of |Zjn| in fig.19 for the value of the load resistance RL of 225 Q is thus the curve for a matched load resistance terminating the signal line on the secondary side, that is to say, no signal reflection occurs at the load resistance RL. Apart from the divergence of the impedance contribution from the coupling capacitors Cd and CC2 as the frequency tends to zero, the |Zjn| curve for the value of 225 Q for RL therefore has a remarkably flat course approaching the 225 Q level as the frequency increases. The curve of the complex argument of Zin in fig.20 corresponding to the matched load resistance at the termination of the signal line on the secondary side follows a similarly flat course. For the values of the load resistance RL that are smaller than the characteristic line impedance Zc, or that are larger than Zc, at least a part of the signal is reflected at the load resistance RL. In the corresponding curves of |Zjn| in fig.19 peaks or local maxima, respectively dips or local minima, which occur at characteristic frequencies, can be observed.
[0079] If the phase of the reflected signal with respect to the input signal that is accumulated during the propagation to the load resistance RL and back to the input terminal 16 is an entire multiple of 2K, that is to say, an even multiple of , the total voltage amplitude at the input terminal resulting from the sum of the input signal and the reflected signal attains a local maximum, as both signals oscillate in phase, and thereby the amplitude of the input impedance |Zjn| also has a local maximum value. Conversely, if the accumulated phase difference of the reflected signal back at the input terminal 16 with respect to the input signal is K or K plus an entire multiple of 2K, which is to say, an odd multiple of , the reflected signal and the input signal are oscillating in phase opposition resulting in a local minimum of the total voltage amplitude at the input terminal 16, and consequently in a local minimum value for the amplitude of the input impedance |Zjn|.
[0080] In the example configuration considered for the device 1 based on capacitive coupling, therefore, peaks occur in the curves of |Zjn| for load resistance values RL smaller than the characteristic impedance Zc, when the total length of the signal lines on the primary and secondary side equals one quarter of a wavelength, three quarter of a wavelength, five quarter of a wavelength, and so on, which corresponds to the frequencies of approximately 49 MHz, 147 MHz, 246 MHz, and so on. At these frequencies the curves of |Zjn| for load resistance values larger than the characteristic impedance necessarily have a local minimum, as the reflectedsignal after it propagated back to the input terminal 16 is in phase opposition with the input signal. Similarly, the curves of |Zjn| for the load resistance values RL larger than the characteristic impedance Zchave thus a peak, when the total length of the signal line on the primary and the secondary side is an entire multiple of one half of the wavelength, which in the considered example configuration corresponds approximately to the frequency values of 98 MHz, 197 MHz, 295 MHz, and so on. At these frequencies the reflected signals for the cases in which the load resistance RL is smaller than the characteristic impedance Zcmust be in phase opposition with the input signal, due to the additional phase shift of % acquired in the reflection at the load resistor, and consequently the corresponding curves of |Zjn| show a local minimum. Depending on the requirements of the application of the device 1, the control unit 50 that is configured to determine the impedance Zin at the input terminal 16 can be set up for measurements at one or several frequencies or a range of frequencies, at which the sensitivity of Zin to the value of the load resistance RL is particularly large or just sufficient for the purpose targeted in the application. For the example configuration of a device 1 based on capacitive coupling considered for the curves plotted in figs 19 and 20, the sensitivity of the absolute value |Zin| of the impedance at the input terminal 16 to the value of the load resistance RL at five examples for the choice of the frequency f is shown in fig.21.
[0081] The curves of |Zjn| for the frequency f chosen at 1 MHz, 2 MHz and 3 MHz as examples show a rather flat course with a first plateau level on the order of 1 kQ for the small values of the load resistance RL between 0.1 Q to about 100 Q. As the load resistance RL is increased further, |Zjn| increases and reaches a maximum and then decreases again to approach a second plateau value on a considerably larger level at large values of RL than the first plateau value that it had at the small values of RL. This type of dependence of |Zjn| on RL can be useful to determine a state change or a switching of a system that is witnessed by a change in RL of an order of magnitude, or more. From the first plateau level of |Zjn| at small RL values to the second plateau value at large RL values lies a factor of at least 4 in the example curve for f = 3 MHz, about one order of magnitude in the curve for f = 2 MHz, and more than an order of magnitude for the curve at f = 1 MHz. It is also noteworthy that when a smaller value is chosen for the frequency f, the onset of the increase in |Zin| with increasing RL and the peak in |Zin| are shifted towards larger values of RL.
[0082] With the choice for the frequency of 49 MHz, at which a peak occurs in the |Zin| curves in fig.19 for RL values smaller than Zc, |Zjn| shows very distinctive changes, when the load resistance RL is changed over each of the orders of magnitude in the range from 1 Q to 10 kQ. The sensitivity of |Zjn| at 49 MHz to changes in RL becomes however considerably weaker for RL values larger than 10 kQ. This type of dependence of the value of |Zjn| as a function of RL, as it is monotonically decreasing over a large range of RL, lends itself well for the precise determination of the value of RL over a large range of values. The curve of |Zjn| versus RL in fig.21 for the frequency choice of 98 MHz, at which the curves in fig.19 for RL values larger than Zchave a peak, is a monotonically increasing function of RL and appears to be even better suited for the precise determination of RL over a large range, since it shows a generally larger slope and over roughly six orders of magnitude in the load resistance RL.
[0083] Similar to the curves shown in fig. 18 for the results from the equivalent circuit model representing the signal lines with lumped circuit elements the curves for the complex argument of Zin drawn in fig. 20 obtained from the transmission line model also display distinctive courses of the curves for the different values of the load resistance RL. The curves for the complex argument of Zin for the different values for RL distinguish themselves mostly in the frequency ranges between the characteristic frequency values, at which the peaks and dips occur in the curves for |Zin| as described above and shown in fig. 19.
[0084] As can be seen in fig.20, at a frequency f chosen in the range between 10 MHz and 40 MHz, for example, the complex argument of Zin is clearly a monotonically decreasing function of RL, which allows the determination of the value of RL in the range from 1 Q to 100 kQ. As another example, the frequency f for the measurement of the phase of Zin could be chosen in the range between 50 MHz and 90 MHz, in which the phase of Zin represents a monotonically increasing function of RL in the range from 1 Q to 100 kQ. In both frequency range examples, the curves of the complex argument of Zin distinguish themselves most between different values of the load resistance RL in the range from 10 Q and 10 kQ facilitating a sensitive determination of RL from the phase measurement. On the other hand, the differences between the curves for RL = 1 Q and RL = 10 Q and between those for RL = 10 kQ and 100 kQ are somewhat smaller, indicating that the sensitivity for thedetermination of RL from the phase is reduced accordingly. This is a very similar if not the same observation that can be made for the curves for the complex argument of Zin shown in fig.18 that were calculated using the equivalent circuit model with a lumped elements representation of the signal lines on the primary and secondary side. Thus, the control device 50 can also be configured to infer the value of RL from the determination of the complex argument of Zin. It is also advantageous that the input frequency f at which the measurement of the phase is done, can be chosen quite freely from within several relatively broad frequency ranges and that the accuracy and the stability of the frequency required for the measurement of the phase of Zin are reduced compared to the measurement of the amplitude |Zjn| at the characteristic frequencies at which the spectrum of |Zjn| shows a peak or a dip feature.
[0085] The derivative of the phase of Zin with respect to frequency determined at a frequency f at or in the vicinity of one of the characteristic frequencies, at which the spectrum of |Zin| shows a peak or a dip, can also be used as a sensitive measure for the value of the load resistance RL, as can clearly be observed from the slope of the curves shown in fig.20 at the different characteristic frequencies. To determine the derivative of the phase of Zin with respect to frequency in the control unit 50, a frequency modulation technique or other methods known in the art can be employed.
[0086] Fig.22 shows a sectional view of a portion of another embodiment of an inventive device 1. This embodiment can be used for both types of coupling. For inductive coupling, the coupling arrangement 30 corresponds to the arrangement of the coil devices 14, 24 described above, while for capacitive coupling, it is constituted by the pairs of electrodes 15, 25. In this embodiment, the coupling arrangement 30 is formed between a first circuit 11 on a first substrate surface 10.1 or the top side and a second circuit 21 located on a second substrate surface 10.2 or bottom side of a double-sided printed first substrate 10. Thus, a coupling and signal transmission between the two surfaces 10.1, 10.2 can be established without a direct electric contact between the two circuits 11, 21 that would otherwise have to be realized using an electronic via technology that is adapted to the type of substrate used. Of course, the bonding areas 18, 28 are not needed in this embodiment.
[0087] It is also possible to include multiple coupling arrangements 30 with galvanic isolation with only one first substrate 10 but a plurality of second substrates 20. Accordingly, a non-conducting connection can be established between one first circuit 11 and a plurality of second circuits 21, each of which may be connected to one or several resistive elements 40. A corresponding embodiment of an inventive device is shown in fig. 23. Thus, several load resistances can be read out by control unit 50 on the first substrate 10.List of Reference Symbolsdevice10, 20 substrate10.1, 10.2, 20.1 Substrate surface 11, 21 circuit12, 22 connection line13, 23 coupling device14, 24 coil device15, 25 electrode16, 26 terminal17, 27 patch18, 28 bonding area30 coupling arrangement 31 coupling transformer 32 coupling capacitor 40 Resistive element 50 control unitD coupling direction
Claims
Claims1. A device (1 ) for resistance detection, comprising:- a first circuit (11) disposed on a first substrate surface (10.1 ) and comprising a first terminal (16) and a first coupling device (13) that is electrically connected to the first terminal (16) and is adapted to generate an electromagnetic field extending in a coupling direction (D) perpendicular to the first substrate surface (10.1 ) in response to an electric input signal applied to the first terminal (16),- a second circuit (21 ) disposed on a second substrate surface (10.2, 20.1 ) and comprising a second terminal (26) for connection to a load resistance (RL) and a second coupling device (23) electrically connected to the second terminal (26), wherein at least in an operational state of the device (1), the second circuit (21) is galvanically isolated from the first circuit (11) and the electromagnetic coupling devices (13, 23) are positioned opposite each other in the coupling direction (D) to form a coupling arrangement (30) that is adapted to generate an induced signal in the second circuit (21) in response to the input signal, and- a control unit (50) adapted to apply the input signal to the first terminal (16) and to detect the load resistance (RL) based on an input impedance (Zin) at the first terminal (16).
2. A device according to claim 1 , wherein the control unit (50) is adapted to apply the input signal so that it comprises an alternating signal with at least one input frequency.
3. A device according to any of the preceding claims, comprising a first substrate (10) with the first substrate surface (10.1) and a second substrate (20) with the second substrate surface (20.1 ), wherein preferably, in operational state at least one substrate (10, 20) is interposed between the circuits (11 , 21 ) in the coupling direction (D).
4. A device according to any of the preceding claims, comprising a first substrate (10) with the first substrate surface (10.1) and the second substrate surface (10.2) opposite the first substrate surface (10.1), whereby the first substrate (10) is interposed between the circuits (11 , 21 ) in the coupling direction (D).
5. A system according to any of the preceding claims, wherein at least one substrate (10, 20) is made of flexible sheet material.
6. A system according to any of the preceding claims, wherein in operational state the first substrate (10) and the second substrate (20) are detachably or non- detachably connected.
7. A system according to any of the preceding claims, wherein the first coupling device (13) and the second coupling device (23) are coil devices (14, 24) adapted for magnetic coupling and the coupling arrangement (30) comprises at least one coupling transformer (31).
8. A system according to any of the preceding claims, wherein at least one coil device (14, 24) comprises a plurality of conductor sections which extend parallel to each other on a substrate surface (10.1, 10.2, 20.1), wherein preferably neighboring conductor sections are connected for opposite current flow.
9. A system according to any of the preceding claims, wherein at least one coupling transformer (31) has a coupling coefficient of at least 0.9, or a coupling coefficient of less than 0.9.
10. A system according to any of the preceding claims, wherein the first coupling device (13) and the second coupling device (23) each comprise at least one electrode (15, 25) adapted for capacitive coupling and the coupling arrangement (30) comprises at least one coupling capacitor (32).
11. A system according to any of the preceding claims, wherein the control unit (50) is adapted to detect the load resistance (RL) based on an absolute value of the input impedance (Zin).
12. A system according to any of the preceding claims, wherein the control unit (50) is adapted to detect the load resistance (RL) based on a phase of the input impedance (Zin).
13. A system according to any of the preceding claims, wherein the control unit (50) is adapted to detect the load resistance (RL) based on a derivative of the phase of the input impedance (Zin) as a function of the input frequency.
14. A system according to any of the preceding claims, wherein the control unit (50) is adapted to apply a signal with a frequency that differs by less than 10%,preferably less than 5%, from a characteristic frequency fn= - — wherein vPis the phase velocity of the signal, / is a total length of signal lines in the first and second circuit (11, 21 ), and n is a nonnegative integer.
15. A system according to any of the preceding claims, wherein the control unit (50) is adapted to apply a signal with a frequency between zero and a lowest characteristic frequency f0= — , or between two characteristic frequencies fn=1p, wherein the frequency preferably differs from the closest characteristic frequency by at least 5%, preferably at least 10 %.
16. A method for resistance detection, using:- a first circuit (11) disposed on a first substrate surface (10.1 ) and comprising a first terminal (16) and a first coupling device (13) that is electrically connected to the first terminal (16) and is adapted to generate an electromagnetic field extending in a coupling direction (D) perpendicular to the first substrate surface (10.1 ) in response to an electric input signal applied to the first terminal (16), and- a second circuit (21 ) disposed on a second substrate surface (10.2, 20.1 ) and comprising a second terminal (26) for connection to a load resistance (RL) and a second coupling device (23) electrically connected to the second terminal (26), the second circuit (21) being galvanically isolated from the first circuit (11) and the electromagnetic coupling devices (13, 23) being positioned opposite each other in the coupling direction (D) to form a coupling arrangement (30),the method comprising:- applying the input signal to the first terminal (16) so that the coupling arrangement (30) generates an induced signal in the second circuit (21) in response to the input signal; and- detecting the load resistance (RL) based on an input impedance (Zin) at the first terminal (16).
