Electric circuit

The invention addresses low supply voltage operation in semiconductor circuits by using interconnected N-FET and P-FET transistors with transistor pairs to stabilize switching logic and reduce process variations, ensuring efficient operation in energy-constrained systems.

WO2026196257A4PCT designated stage Publication Date: 2026-11-12KALAJDŽIĆ, MARKO ALEKSANDAR
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
PCT/IB2026/054189
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2025-04-29
Filing Date
2026-04-28
Publication Date
2026-11-12

AI Technical Summary

Technical Problem

Existing semiconductor circuits face challenges in operating at low supply voltages due to process, voltage, and temperature variations, with process variations being particularly difficult to mitigate through design techniques.

Method used

The invention employs electrical circuits with specific configurations of AND, OR, and inverter gates constructed from interconnected N-FET and P-FET transistors, operating below the threshold voltage, and uses transistor pairs to dampen process variations, ensuring well-defined output levels and robust switching logic.

Benefits of technology

This design achieves stable and predictable switching logic at low supply voltages, reducing leakage current and minimizing the impact of process variations, enabling efficient operation in energy-constrained systems.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure IB2026054189_12112026_PF_FP_ABST
    Figure IB2026054189_12112026_PF_FP_ABST
Patent Text Reader

Abstract

The invention relates to an electric circuit for low supply voltages, comprising at least one AND gate, composed of at least two N-FET transistors, and / or at least one OR gate, composed of at least two P-FET transistors, and at least one inverter, composed of multiple transistors, wherein transistors of a Schmitt-trigger-like basic structure of the inverter are replaced by transistor pairs, wherein each transistor pair consists of a transistor of the basic structure and an additional transistor of a complementary conduction type, and wherein the electric circuit is operated at a supply voltage which is below the absolute value of the threshold voltage of the transistors.
Need to check novelty before this filing date? Find Prior Art

Description

Electrical circuit

[0001] The invention relates to an electrical circuit and the gate circuits constituting it. background

[0002] Semiconductor circuits have become indispensable in many areas of life.

[0003] Nevertheless, there is a growing need for semiconductor circuits that can or should only be operated with low supply voltages.

[0004] Examples include energy-constrained systems such as the Internet of Things, but also sensors, bioimplants, wearable devices, and especially devices powered by energy harvesting, where operation at low supply voltages is desirable. Perhaps the greatest challenge in energy-efficient circuit design is posed by so-called process, voltage, and temperature variations (PVT variations).

[0005] While voltage and temperature variations can be dampened by various mechanisms and techniques so that circuits remain within the target design space, process variations represent a fundamental limit that is difficult to change through design techniques.

[0006] Such techniques are essentially limited to enlarging all components or using process technologies that exhibit the lowest possible process fluctuations.

[0007] Previous approaches have focused on optimizing existing circuit topologies. However, this approach presents obstacles that are either impossible or extremely difficult to overcome. Technical task

[0008] It is therefore the object of the invention to provide an improvement that allows semiconductor circuits to be operated at low supply voltages.

[0009] The problem is solved by an electrical circuit for low supply voltages according to claim 1. Further advantageous embodiments are the subject of the dependent claims as well as the description and the figures.

[0010] The invention will now be described in more detail with reference to the figures.

[0011] It should be noted that different aspects are described, each of which can be used individually or in combination. That is to say, each aspect can be used with different embodiments of the invention, unless explicitly presented as a pure alternative.

[0012] Furthermore, for the sake of simplicity, reference will generally be made to only one entity at a time. Unless explicitly stated otherwise, the invention may also include several of the entities concerned. Therefore, the use of the words "a," "an," and "one" should only be understood as an indication that at least one entity is used in a simple embodiment.

[0013] Unless explicitly stated otherwise, the individual steps of a procedure described below can be arranged and / or combined in any order. Furthermore, the procedures can be combined with each other unless expressly indicated otherwise.

[0014] Insofar as standards, specifications, or the like are mentioned in this application, reference is always made to at least those standards, specifications, or the like applicable on the filing date. That is to say, if a standard / specification, etc., is updated or replaced by a successor, the standard / specification will be automatically superseded.

[0015] If the original is replaced, the invention is also applicable to it. Various embodiments are shown in the figures.

[0016] The invention relates to both the electrical circuits for forming a modular logic family and the gates constituting them, such as AND, OR or inverter.

[0017] Exemplary electrical circuits according to the invention for low supply voltages comprise at least one AND gate according to the invention, constructed from at least two N-FET transistors, and / or one OR gate according to the invention, constructed from at least two P-FET transistors, as well as an inverter according to the invention, constructed from a plurality of transistors, wherein the transistors are connected to each other within a gate in a circuit according to the invention and wherein the supply voltage is below the absolute value of the threshold voltage of the transistors.

[0018] Exemplary schematic circuits of the gates according to the invention are now described with reference to Figures 1 to 3. Figure 1 describes an AND gate, Figure 2 an OR gate, and Figure 3 an inverter as the basic gate.

[0019] The AND gate – see Figure 1 – uses two N-FET transistors, T1 and T2, as an example. The gate G1 of the first N-FET transistor, T1, is connected to the source terminal S2 of the second N-FET transistor, T2. Furthermore, the gate G2 of the second N-FET transistor, T2, is connected to the source terminal S1 of the first N-FET transistor, T1. Likewise, the two drain terminals, D1 and D2, are connected to each other.

[0020] The connection between source terminal S1 and gate terminal G2 acts as the first input IN1, while the connection between source terminal S2 and gate terminal G1 acts as the second input IN2. The connection between drain terminals D1 and D2 acts as the output OUT. In other words, both IN1 and IN2 are connected to both N-FET transistors T1 and T2.

[0021] The OR gate – see Figure 2 – features two P-FET transistors, T1 and T2. The gate G1 of the first P-FET transistor, T1, is connected to the source terminal S2 of the second P-FET transistor, T2. Furthermore, the gate G2 of the second P-FET transistor, T2, is connected to the source terminal S1 of the first P-FET transistor, T1. Similarly, the two drain terminals, D1 and D2, are connected to each other.

[0022] The connection between source terminal S1 and gate terminal G2 acts as the first input IN1, while the connection between source terminal S2 and gate terminal G1 acts as the second input IN2. The connection between drain terminals D1 and D2 acts as the output OUT. In other words, both IN1 and IN2 are connected to both P-FET transistors T1 and T2.

[0023] The two-transistor AND and OR gates presented here are similar to pass-transistor logic (PTL). There is neither a pull-up path nor a pull-down path. Instead, the gates are created by connecting two N-FET or P-FET transistors in such a way that the input is applied to the gate of one and the source of the other. The drain terminals of both transistors are connected and function as the output. In this configuration, the transistors involved in the gate have no amplifying properties, since neither a ground nor a supply voltage is present. However, this also means that, due to the lack of a possible cross-current, virtually no leakage current flows.

[0024] Since the supply voltage is below the transistors' threshold voltage, well-defined output levels are established for all combinations of input signals in the AND and OR gates. Figure 10, based on a sample produced using a 22 nm FDSOI process, shows how the output levels commonly described as undefined in the two-transistor AND and OR gates presented here become well-defined potentials for functional signal propagation and maintain signal integrity. In other words, the AND and OR gates presented here ensure correct switching logic for all possible combinations of input signals.

[0025] The underlying principle can be illustrated using a minimal example of the AND and OR gates. Specifically, during the transition from input combination 11 (i.e., when a logic 1 is present at both input IN1 and input IN2) to input combination 00, the voltage level at the output OUT of the AND gate (see Figure 1) would typically be expected to settle at an indeterminate level, while the N-FET transistors T1 and T2 exhibit a steadily decreasing charge transport capacity due to the decreasing potential at their gate terminals G1 and G2. This expectation is particularly relevant when both input voltages decrease simultaneously.In this case, the AND gate can be reduced to an equivalent circuit as a minimal example, in which the input IN1 or IN2 is connected to the source or gate terminal of a single N-FET transistor, while the output OUT is at the drain terminal and in a real circuit is still connected to a capacitor, usually in the form of another gate connected to the output.

[0026] This equivalent circuit then represents an RC circuit with a time-varying resistance, implemented by the single N-FET transistor, which in the example above exhibits simultaneously decreasing source and gate potentials. If these potentials decrease simultaneously, the drain potential also decreases, following a time course similar to a logistic function, to the potential of the source terminal.

[0027] Similarly, a corresponding minimal example of the OR gate – see Figure 2 – can be constructed for the case where both input voltages change simultaneously from a logic 0 to a logic 1. Here, one would typically expect that with an increasing input voltage during the switching process, the charge transport capacity of the two P-FET transistors T1 and T2 in the OR gate decreases to such an extent that the output OUT exhibits an indeterminate voltage level at the end of the switching process.Since it is assumed here that both input signals IN1 and IN2 change simultaneously, an equivalent circuit for the OR gate can be constructed analogously to the AND gate, consisting of a single P-FET transistor, where the input IN1 or IN2 is connected to the source or gate terminal of this single P-FET transistor, whose drain terminal forms the output OUT, which, just like in the equivalent circuit of the AND gate, is connected to a capacitor.

[0028] This equivalent circuit also represents an RC circuit with time-varying resistance, in which the potential at the drain terminal adjusts to the potential at the source terminal, i.e., the potential of the input signals, over a period of time resembling a logistic function.

[0029] The time magnitude of this adjustment of the output voltage of the AND or OR gate to the respective input voltage corresponds to a typical and practical speed of gates designed for subthreshold operation.

[0030] Figure 10 shows, based on a pattern fabricated in a 22 nm FDSOI process, such a time course of the adaptation of the output potential to the input potential for the AND and OR gates at a supply voltage of 100 mV. The potential shown corresponds to the output node of the AND gate and simultaneously to the input node of the inverter according to the invention shown in Figure 6. This is a simulation of the same circuit as in Figures 12 and 13. The potential shown corresponds to the output node of the OR gate and simultaneously to the input node of the inverter according to the invention shown in Figure 7. This is a simulation of the same circuit as in Figures 14 and 15. NMOS and PMOS, respectively, denote the corresponding equivalent circuits, and NMOS Fit and PMOS Fit denote the corresponding fits of a mathematical model for the respective voltage waveform.

[0031] Furthermore, the use of only one transistor type within the AND gate or OR gate excludes changes in the switching logic caused by process-related corner variations, such as can occur in complementary CMOS circuits due to an imbalance between N-FET and P-FET properties.

[0032] The inverter gate – see Figure 3 – has, for example, 12 transistors T1…T12. Transistors T1 and T2 are P-FET transistors, while transistors T3 and T4 are N-FET transistors. The respective gate terminals G1…G4 are connected to each other on the input side IN. The drain terminal D1 of the first P-FET transistor T1 and the source terminal S2 of the second P-FET transistor T2 are connected, as are the source terminal S3 of the third N-FET transistor T3 and the drain terminal D4 of the fourth N-FET transistor T4. The drain terminal D1 of the first P-FET transistor T1 is also connected to the gate terminal G5 and the source terminal S5 of the fifth N-FET transistor T5. Similarly, the drain terminal D3 of the third N-FET transistor T3 is connected to the gate terminal G7 and the source terminal S7 of the seventh P-FET transistor T7.Likewise, the drain terminal D4 of the fourth N-FET transistor T4 is connected to both the gate terminal G8 and the source terminal S8 of the eighth P-FET transistor T8.

[0033] The drain terminal D2 of the second P-FET transistor T2 is connected to the drain terminal D3 of the third N-FET transistor T3 on the output side OUT. Similarly, the gate terminal G6 and the source terminal S6 of the sixth N-FET transistor T6, as well as the source terminal S7 and the gate terminal G7 of the seventh P-FET transistor T7, are also connected to the output side OUT.

[0034] Furthermore, the source terminal S2 of the second P-FET transistor T2 is connected to both the source terminal S9 and the gate terminal G9 of the ninth N-FET transistor T9, as well as to the source terminal S10 of the tenth P-FET transistor T10. The gate terminal G10 of the tenth P-FET transistor T10 is also connected to the output side OUT.

[0035] Furthermore, the source terminal S3 of the third N-FET transistor T3 is connected to both the source terminal S11 of the eleventh N-FET transistor T11 and the gate terminal G12 and the source terminal S12 of the twelfth P-FET transistor T12. The gate terminal G11 of the eleventh N-FET transistor T11 is also connected on the output side OUT.

[0036] Furthermore, both the drain terminal D12 of the twelfth P-FET transistor T12 and the drain terminal D10 of the tenth P-FET transistor T10 are connected to ground / minus / GND - indicated by the triangle.

[0037] Likewise, both the drain terminal D9 of the ninth N-FET transistor T9 and the drain terminal D11 of the eleventh N-FET transistor T11 are connected to supply voltage / Plus / VDD - marked by T-.

[0038] The drain terminal D5 of the fifth N-FET transistor T5 and the drain terminal D6 of the sixth N-FET transistor T6 are connected to supply voltage / positive / VDD, whereas the drain terminal D7 of the seventh P-FET transistor T7 and the drain terminal D8 of the eighth P-FET transistor T8 are connected to ground / negative / GND. Similarly, the source terminal S4 of the fourth N-FET transistor T4 is connected to ground / negative / GND. The source terminal S1 of the first P-FET transistor T1 is connected to supply voltage / positive / VDD.

[0039] It should be noted that the number of transistors described above represents a functional description of the topology. A transistor provided in the circuit according to the invention can therefore be replaced by several sub-transistors without departing from the invention, provided that these, taken together, provide the electrical function of the transistor in question within the circuit according to the invention. For this purpose, the sub-transistors can, in particular, be connected in series and / or parallel. Likewise, functionally equivalent segmentations, divisions, or aggregations of transistor structures are included, as long as the circuit of the nodes according to the invention and the resulting switching function are retained.

[0040] The structure of the inverter according to the invention can be understood as a mapping rule, in which a standard CMOS inverter is used as a starting point, which on the one hand is extended by a Schmitt-trigger-like structure with doubling of the pull-up path and the pull-down path as well as the introduction of feedback transistors, and in which on the other hand each N-FET transistor and each P-FET transistor of the output structure is replaced by a specific complementary transistor pair.

[0041] Consequently, this design principle is not limited to the inverter according to the invention. Rather, the Schmitt-trigger-like extension, by replacing the individual transistors with transistor pairs according to the invention, can also be applied to other logic circuits that feature a Schmitt-trigger-like extension, such as NAND or NOR gates. The inverter plays a particularly important role here as a modular basic building block, thereby significantly advancing the technical effect of the invention in the circuit configuration according to the invention, since a multitude of elementary combinational and sequential logic circuits can be directly formed by connecting AND gates, OR gates, and inverters according to the invention. This allows for the provision of a logic family with a high degree of structural regularity.The use of these basic structures enables, in particular, highly regular layouts and highly symmetrical input nodes of the gates, thereby achieving improved matching of the components and thus improved predictability of the switching characteristics as well as increased uniformity of the gate behavior within an electrical circuit.

[0042] An inverter curve for the inverter under process and temperature fluctuations. FS 45°C, FS 25°C, SF 45°C, and SF 25°C is shown in Figure 11.

[0043] By connecting N-FET and P-FET transistors, as shown in Figures 4 and 5, i.e., by constructing the transistor pairs according to the invention, the switching behavior changes. Figure 4 assumes the case of an N-FET transistor, while Figure 5 shows the analogous case of a P-FET transistor. We will explain this below using the example of an N-FET transistor, although the corresponding effect is also observed with P-FET transistors.

[0044] If an N-FET transistor T1 is extended in the pull-down path by a P-FET transistor T2, the P-FET transistor T2 switches more strongly the closer the drain and source potentials of the N-FET transistor T1 are to ground (GND). The participation of the P-FET transistor T2 in the pull-down path is therefore dynamic, as the P-FET transistor T2 only develops the capacity to conduct current during the switching of the N-FET transistor T1. This implies that such process variations, which influence the charge transport of a transistor due to its type, are dampened in their effect on the circuit behavior by the additional charge transport provided by a complementary transistor.

[0045] In general, the construction of the transistor pairs according to the invention can be described as follows: An N-FET transistor T1 is extended by an additional P-FET transistor T2 by connecting the source terminal S2 and the gate terminal G2 to each other and to the source or drain terminal S1 or D1. The drain terminal D2 of the additional P-FET transistor is connected to ground / negative / GND or a feedback path. Since, by convention, the source terminal of the N-FET transistor is located on the side with the lower potential, whether the additional P-FET transistor is connected to the source or drain terminal of the N-FET transistor also depends on the position and function of the extended N-FET transistor in the electrical circuit.The connection of the additional P-FET transistor is particularly preferred to that source- or drain-side circuit node of the N-FET transistor which, with respect to the installation position of the formed transistor pair in the respective basic structure, is located on the output side. In the example of the inverter from Figure 3, the N-FET transistors T3 and T4 of the basic structure are accordingly extended at the drain terminal by an additional P-FET transistor, while the N-FET transistor T11 is connected to the corresponding additional P-FET transistor on the source terminal side, in order to further the technical effect of the invention.If, for example, the source terminal S12 and the gate terminal G12 of the additional P-FET transistor T12 were connected to the drain terminal D11 of the N-FET transistor T11 of the basic structure, the additional P-FET transistor T12 would not be connected to the output-side circuit node of the N-FET transistor T11 and would therefore not contribute to the electrical switching function as intended; rather, this could promote an undesired current path.

[0046] A P-FET transistor T1 is extended by an additional N-FET transistor T2 by connecting the source terminal S2 and the gate terminal G2 to each other and to the source or drain terminal S1 or D1. The drain terminal D2 of the additional N-FET transistor is connected to a supply voltage / positive / VDD or a feedback path. Since, by convention, the source terminal of the P-FET transistor is on the side with the higher potential, the connection of the additional N-FET transistor to the source or drain terminal of the P-FET transistor depends on the position and function of the extended P-FET transistor in the circuit. The connection of the additional N-FET transistor is particularly preferred to that source- or drain-side circuit node of the P-FET transistor which, relative to the installation position of the resulting transistor pair in the respective basic structure, is located on the output side.In the example of the inverter shown in Figure 3, the P-FET transistors T1 and T2 of the basic structure are each extended at their drain terminals by an additional N-FET transistor, while the P-FET transistor T10 is connected to the corresponding additional N-FET transistor at its source terminal to further the technical effect of the invention. If, for example, the source terminal S9 and the gate terminal G9 of the additional N-FET transistor T9 were connected to the drain terminal D10 of the P-FET transistor T10 of the basic structure, the additional N-FET transistor T9 would not be connected to the output-side circuit node of the P-FET transistor T10 and would therefore not contribute to the electrical switching function as intended; rather, this could promote an undesired current path.

[0047] In a particularly preferred embodiment, the source terminal and the gate terminal of an additional transistor of complementary conductor type are connected together with that source- or drain-side circuit node of a corresponding transistor of the basic structure which, with respect to the installation position of the transistor pair in the respective electrical circuit, is located on the output side. The output-side location is defined here as that source- or drain-side circuit node of the corresponding transistor of the basic structure from which the current path relevant for the electrical switching function of the respective circuit extends to the output node.

[0048] It should also be noted here that the number of transistors described above represents a functional description of the topology. A transistor provided in the circuit according to the invention can therefore be replaced by several sub-transistors without departing from the invention, provided that these, taken together, provide the electrical function of the transistor in question within the circuit according to the invention. For this purpose, the sub-transistors can, in particular, be connected in series and / or parallel. Likewise, functionally equivalent segmentations, divisions, or aggregations of transistor structures are included, as long as the circuit of the nodes according to the invention and the resulting switching function are retained.

[0049] Such a division into sub-transistors can be achieved, for example, by extending the N-FET transistor T1 in the circuit shown in Figure 4 by adding two P-FET transistors T2 and T3, wherein the source and gate terminals S2 and G2 of the P-FET transistor T2 are connected to each other, to the gate terminal G3 of the P-FET transistor T3, and to the source or drain terminal of the N-FET transistor S1 or D1, wherein the drain terminal D2 of the P-FET transistor T2 is connected to the source terminal S3 of the P-FET transistor T3, and wherein the drain terminal D3 of the P-FET transistor T3 is connected to ground / negative / GND or a feedback path. The same applies to the circuit shown in Figure 5.

[0050] This means that in cases where N-FET or P-FET transistors have a comparatively increased or decreased capacity for charge transport due to manufacturing processes, variations in this capacity are at least partially compensated for by another transistor of a complementary type.

[0051] Therefore, variations in switching behavior in the entire pull-up path as well as in the entire pull-down path are dampened by the use of the described transistor pairs.

[0052] The basic elements according to the invention, namely AND gates, OR gates, and inverters, form a modular logic family from which not only NAND gates and NOR gates, but also further combinational and sequential logic circuits can be systematically assembled. The present invention is therefore not limited to the explicitly illustrated gates, but also includes circuit arrangements in which more advanced logic functions are formed by interconnecting the basic elements according to the invention.

[0053] It should be noted that the bulk connector connections shown in Figures 1 to 9 are merely examples and not limiting. The circuit topology is independent of the specific design of the bulk connectors. In particular, the possibility of applying body biasing or bulk biasing techniques remains unaffected. Such techniques can be used additionally without altering the functionality of the topology.

[0054] Unlike design techniques that influence the switching behavior of individual transistors, for example through body biasing, bulk biasing or other transistor-related optimization measures, the present invention takes a different approach.

[0055] For the inverter, the invention increases the number of transistors in a specific topological arrangement. The improvement is not primarily achieved at the level of controlling or parameterizing individual transistors, but rather through a topological design of the circuit itself. Such additional design techniques can be used as a supplement, but are not a prerequisite for the functionality of the circuit topology.

[0056] While doubling the number of transistors (N → 2N) in combination with a specific interconnection of the resulting transistor pairs largely reduces the effect of corner variations at the expense of the quality of the logic levels, the additional use of Schmitt-trigger-like logic increases this logic level at the expense of the area due to the use of more than twice as many (N → 2N + 2) transistors.

[0057] This allows the realization of a circuit topology in which a Schmitt-trigger-like extension interacts with transistor pair logic.

[0058] The inverter according to the invention is not equivalent to a conventional Schmitt trigger. The Schmitt-trigger-like structure represents only one aspect. It also features a unique structural design in which transistors of an output structure are systematically replaced by complementary transistor pairs according to the invention. The technical effect of the inverter according to the invention does not result solely from the Schmitt-trigger-like extension as such, but only from its interaction with this replacement.

[0059] The use of two or more identical N-FET or P-FET transistors to form the AND or OR gates results, at the expense of amplifying properties of the switching logic, in a reduction of the number of transistors usually required, as well as robustness of the output levels against permutations of the input signals and the absence of such process variations as can occur in complementary CMOS circuits as a result of an imbalance between N-FET and P-FET properties.

[0060] Furthermore, the reduced variability of the output levels can be advantageous in applications where a uniform signal characteristic is desired.

[0061] Only the interaction of the first technique, namely the formation of symmetrical logic gates from identical transistors, and the second technique, namely on the one hand the Schmitt-trigger-like extension of complementary logic structures and on the other hand the replacement of the individual transistors by certain transistor pairs, enables a circuit design in which both the robustness against corner variations and the variability of the output levels under permutations of the input signals are improved in such a way that circuits according to claim 1 can provide a safe and stable switching logic in an otherwise inaccessible target design space at supply voltages below the threshold voltage.

[0062] According to one embodiment of the invention, the supply voltage is less than 150 mV. Obviously, the supply voltage depends on the material system and the achievable feature sizes. For example, the inventor has already produced samples based on a 22 nm FDSOI process with a supply voltage of 100 mV as an example of a voltage in the sub-threshold range.

[0063] Without limiting the generality of the invention, an inverter – see, for example, Figure 3 – can have at least 12 transistors. Particularly preferably, an inverter has exactly 12 transistors, especially in the circuit shown in Figure 3.

[0064] Furthermore, the size of a single AND gate can be chosen depending on the material system, the process technology used, and the achievable feature sizes. In an exemplary implementation based on a 22 nm FDSOI process, AND gates for two inputs with an area requirement of less than 3.2 µm² were realized.

[0065] According to one embodiment of the invention—illustrated by way of example in Figure 8—AND gates can also have a plurality of inputs, with each input being assigned two transistors, the transistors being of the same type. A generalization of the two-transistor AND gate can be made. Input IN1 is connected to the source terminal S1 of the first N-FET transistor T1 and the gate terminal GN of the Nth N-FET transistor TN. This can be continued until the last input INN. The last input INN is connected to the source terminal SN of the Nth N-FET transistor TN and the gate terminal G(N-1) of the (N-1)th N-FET transistor T(N-1). All drain terminals D1 … DN are connected together and form the output OUT.

[0066] Furthermore, the size of a single OR gate can be chosen depending on the material system, the process technology used, and the achievable feature sizes. In an exemplary implementation based on a 22 nm FDSOI process, OR gates for two inputs with an area requirement of less than 3.2 µm² were realized.

[0067] According to one embodiment of the invention – illustrated by way of example in Figure 9 – OR gates can also have a plurality of inputs, with each input being assigned two transistors, the transistors being of the same type. A generalization of the two-transistor OR gate can be made. Input IN1 is connected to the source terminal S1 of the P-FET transistor T1 and the gate terminal GN of the Nth P-FET transistor TN. This can be continued until the last input INN. The last input INN is connected to the source terminal SN of the P-FET transistor TN and the gate terminal G(N-1) of the (N-1)th P-FET transistor T(N-1). All drain terminals D1 … DN are connected together and form the output OUT.

[0068] As shown in Figures 6 and 7, NAND gates and NOR gates can be formed particularly advantageously by interconnecting the basic elements according to the invention, especially by connecting an AND gate or OR gate according to the invention with an inverter according to the invention. Such an interconnection is not merely to be understood as a subsequent combination of individual elements known per se, but rather is an expression of the modular logic family according to the invention. This family is structured in such a way that more complex logic functions can be systematically composed from the basic gates according to the invention.

[0069] This is particularly advantageous for the implementation of extensive digital circuits at supply voltages below the threshold voltage, since in practical circuit arrangements the basic gates according to the invention are regularly connected in series.

[0070] In the presented topology, it makes no difference whether one designs a NAND gate directly, as this would lead to a corresponding structure. The same applies to NOR gates. Similarly, the same holds true for interconnections of multiple AND gates or multiple OR gates with at least one inverter.

[0071] An inverter curve for NAND gates for the voltage sweep of input combination 01 → 11 under process and temperature fluctuations. FS 45°C, FS 25°C, SF 45°C, and SF 25°C is shown in Figure 12.

[0072] An inverter curve for NAND for the voltage sweep of the input combination 00 → 11 under process and temperature fluctuations. FS 45°C, FS 25°C, SF 45°C, and SF 25°C is shown in Figure 13.

[0073] An inverter curve for NOR for the voltage sweep of the input combination 01 → 11 under process and temperature fluctuations. FS 45°C, FS 25°C, SF 45°C, and SF 25°C is shown in Figure 14.

[0074] An inverter curve for NOR for the voltage sweep of the input combination 00 → 11 under process and temperature fluctuations. FS 45°C, FS 25°C, SF 45°C, and SF 25°C is shown in Figure 15.

[0075] According to a further embodiment of the invention, the transistors used are selected from complementary field-effect transistors. The circuit topology according to the invention is not limited to a specific process technology, but can be implemented in field-effect transistor technologies that provide N-FET and P-FET transistors. That is, the invention allows for a technology-neutral design.

[0076] The present invention thus expands the design techniques.

[0077] According to the invention, transistor pairs can reduce process fluctuations affecting only the N-FET transistors by incorporating the corresponding P-FET transistors. Conversely, process fluctuations affecting only the P-FET transistors can be reduced by incorporating the corresponding N-FET transistors. Simultaneously, the specific interconnection of identical N-FET and P-FET transistors allows the design of AND and OR gates with the properties described above.

[0078] The present invention, however, is not limited to these techniques individually. Rather, it is the interaction of the Schmitt-trigger-like extension of complementary logic structures by replacing individual transistors with specific complementary transistor pairs, on the one hand, and the specific interconnection of similar N-FET or P-FET transistors to form AND or OR gates, on the other, that opens up the possibility of a highly regular and robust cell library suitable for logic at low supply voltages. This opens up a previously unattainable area in the design space.

[0079] The invention is further explained with the aid of figures. These show: Fig. 1

[0080] an exemplary schematic implementation of an AND gate according to embodiments of the invention, Fig. 2

[0081] an exemplary schematic realization of an OR gate according to embodiments of the invention, Fig. 3

[0082] an exemplary schematic realization of an inverter according to embodiments of the invention, Fig. 4

[0083] an exemplary schematic implementation of a complementary transistor pair, Fig. 5

[0084] an exemplary schematic implementation of a complementary transistor pair, Fig. 6

[0085] an exemplary schematic realization of a NAND gate according to embodiments of the invention, Fig. 7

[0086] an exemplary schematic realization of a NOR gate according to embodiments of the invention, Fig. 8

[0087] an exemplary schematic realization of a multiple AND gate according to embodiments of the invention, Fig. 9

[0088] an exemplary schematic realization of a multiple OR gate according to embodiments of the invention, Fig. 10

[0089] a time-dependent adjustment of the output voltage to the input voltage of exemplary AND and OR gates according to embodiments of the invention with equivalent circuit diagram and fitting of a mathematical model, Fig. 11

[0090] Exemplary voltage curves of exemplary implementations of inverters according to embodiments of the invention, Fig. 12

[0091] Exemplary voltage waveforms for asymmetrical input voltages at exemplary implementations of NAND gates according to embodiments of the invention, Fig. 13

[0092] Exemplary voltage waveforms for symmetrical input voltages at exemplary implementations of NAND gates according to embodiments of the invention, Fig. 14

[0093] Exemplary voltage waveforms for asymmetrical input voltages at exemplary implementations of NOR gates according to embodiments of the invention, Fig. 15

[0094] Exemplary voltage waveforms for symmetrical input voltages at exemplary implementations of NOR gates according to embodiments of the invention.

Claims

Electrical circuit for low supply voltages, comprising at least one AND gate constructed from at least two N-FET transistors, and / or at least one OR gate constructed from at least two P-FET transistors, and at least one inverter constructed from a plurality of transistors, wherein transistors of a Schmitt-trigger-like basic structure of the inverter are replaced by transistor pairs, each transistor pair consisting of one transistor of the basic structure and an additional transistor of complementary line type, wherein at least one AND gate and / or at least one OR gate and at least one inverter are interconnected to form different logic switching functions within the electrical circuit, wherein the electrical circuit is operated at a supply voltage that is below the absolute value of the threshold voltage of the transistors.where one or more of the transistors can each be replaced by several sub-transistors, provided that the electrical function within the replaced circuit is maintained. Electrical circuit according to claim 1, characterized in that the inverter has at least 12 transistors. An electrical circuit according to one of the preceding claims, characterized in that transistors of a Schmitt-trigger-like basic structure of the inverter are replaced by transistor pairs, wherein each N-FET transistor of the basic structure is assigned an additional P-FET transistor, the drain terminal of which is connected to a ground / minus / GND or a feedback path and the source terminal and gate terminal are connected together to the same circuit node as the source terminal or the drain terminal of the N-FET transistor of the basic structure, wherein each P-FET transistor of the basic structure is assigned an additional N-FET transistor, the drain terminal of which is connected to a supply voltage / plus / VDD or a feedback path and the source terminal and gate terminal are connected together to the same circuit node as the source terminal or the drain terminal of the P-FET transistor of the basic structure. Electrical circuit according to one of the preceding claims, characterized in that the AND gate is constructed from two N-FET transistors, wherein the gate terminal of a first N-FET transistor is connected to the source terminal of a second N-FET transistor, the gate terminal of the second N-FET transistor is connected to the source terminal of the first N-FET transistor, and the drain terminals of the two N-FET transistors are connected to each other. Electrical circuit according to one of the preceding claims, characterized in that the OR gate is constructed from two P-FET transistors, wherein the gate terminal of a first P-FET transistor is connected to the source terminal of a second P-FET transistor, the gate terminal of the second P-FET transistor is connected to the source terminal of the first P-FET transistor, and the drain terminals of the two P-FET transistors are connected to each other. An electrical circuit according to one of the preceding claims, characterized in that the AND gate has N inputs and N N-FET transistors, wherein the first input is connected to the source terminal of the first N-FET transistor and the gate terminal of the N-th N-FET transistor, wherein each i-th input from the second to the N-th input is connected to the source terminal of the i-th N-FET transistor and the gate terminal of the preceding (i-1)-th N-FET transistor, wherein i is a natural number greater than 1 and at most equal to N, wherein the drain terminals of the N N-FET transistors are connected to each other, where N is a natural number greater than 2. An electrical circuit according to one of the preceding claims, characterized in that the OR gate has N inputs and N P-FET transistors, wherein the first input is connected to the source terminal of the first P-FET transistor and the gate terminal of the N-th P-FET transistor, wherein each i-th input from the second to the N-th input is connected to the source terminal of the i-th P-FET transistor and the gate terminal of the preceding (i-1)-th P-FET transistor, wherein i is a natural number greater than 1 and at most equal to N, wherein the drain terminals of the N P-FET transistors are connected to each other, where N is a natural number greater than 2. Electrical circuit according to one of the preceding claims, characterized in that the AND gate in a MOSFET technology for two inputs has an area requirement of less than 10 µm². Electrical circuit according to one of the preceding claims, characterized in that the OR gate in a MOSFET technology for two inputs has an area requirement of less than 10 µm². Electrical circuit according to one of the preceding claims, characterized in that the inverter in a MOSFET technology has an area requirement of less than 10 µm². Electrical circuit according to one of the preceding claims, characterized in that a NAND gate formed by interconnecting an AND gate according to claim 1, 4 or 6 with an inverter according to claim 1, 2 or 3, in a MOSFET technology has an area requirement of less than 20 µm². Electrical circuit according to one of the preceding claims, characterized in that a NOR gate formed by interconnecting an OR gate according to claim 1, 5 or 7 with an inverter according to claim 1, 2 or 3, in a MOSFET technology has an area requirement of less than 20 µm². Electrical circuit according to one of the preceding claims, characterized in that the electrical circuit is implemented in a field-effect transistor technology that provides N-FET transistors and P-FET transistors. Electrical circuit according to one of the preceding claims, characterized in that the electrical circuit is operated at a supply voltage of at most 150 mV.