Power device with current measurement

Abstract

Electronic device with: a circuit for a power device comprising the following: a main power transistor comprising a main drain terminal, a main source terminal, and a main gate terminal, a resistive load, and a sensor transistor with a source connected to the main source terminal, a gate connected to the main gate terminal, and a drain connected to a first terminal of the resistive load, a feedback circuit comprising at least one operational amplifier and at least one transconductor, and a circuit with at least one switch for connecting and disconnecting the circuit for the power device with the feedback circuit, wherein a first input of the at least one operational amplifier and a second connection of the resistive load are selectively connected to the main drain connection via the circuit, wherein a second input of the at least one operational amplifier and at least one output of the at least one transconductor are selectively connected to the drain of the sensor transistor via the circuit, and wherein the main power transistor has a first area or a first gate circumference, wherein the sensor transistor has a second area or a second gate circumference that is smaller than the first area or the first gate circumference, and wherein, during an ON state of the main power transistor, the output current of the feedback circuit's transconductor is proportional to the potential difference between the main drain terminal and the drain of the sensor transistor, resulting in a voltage across the resistive load of essentially zero; and wherein, during an OFF state of the main power transistor, the circuit is configured to disconnect the feedback circuit from the circuit for the power device, resulting in an increase in the drain-source voltage of the sensor transistor until the voltage is essentially the same as the drain-source voltage of the main power transistor.

Description

Area of ​​Disclosure

[0001] The disclosure relates to power semiconductor devices and current measurement in power semiconductor devices. In particular, but not exclusively, the disclosure relates to the use of an AlGaN / GaN transistor with a heterostructure and high electron mobility. Background to the disclosure

[0002] Gallium nitride (GaN) is a wide bandgap material suitable for power and RF semiconductor devices.

[0003] GaN technology enables transistors with high electron mobility and high saturation rate, both of which are of great importance in the field of power and RF electronics. The GaN material has additional advantages for power devices. The material's wide bandgap (E g = 3.39 eV) leads to a high critical electric field (E c=3.3MV / cm), which can lead to the development of components with a shorter drift range and thus a lower on-resistance when compared to a silicon-based component with the same breakdown voltage.

[0004] The use of an aluminum galium nitride (AlGaN) / GaN heterostructure also enables the formation of a two-dimensional electron gas (2DEG) at the hetero interface, in which the charge carriers have very high mobility values ​​(µ=2000cm). 2 / (Vs)). Furthermore, the piezopolarization charge present in the AlGaN / GaN heterostructure leads to a high electron density in the 2DEG layer (e.g., 1e 13 cm -2These properties enable the development of high electron mobility transistors (HEMTs) and Schottky barrier diodes with highly competitive performance parameters. Much of the research has focused on the development of power devices with AlGaN / GaN heterostructures. Several technologies exist that enable HEMT devices with normal on-state operation (Schottky gate-based technology) and normal off-state operation (insulated gate and p-GaN gate technology).

[0005] In power electronics, there is often a need to integrate protective and measuring circuits (e.g., current measurement, overvoltage protection, temperature measurement) into the main switch, which frequently operates in enrichment mode. Current measurement is one of the most important circuits that must be integrated into the power device.

[0006] Some current meters contain a measuring resistor that assists in detecting and / or measuring current. The current detection signal (V) CS The current across the measuring resistor is generally proportional to the current through the measuring resistor. Ideally, the current through the measuring resistor is a constant fraction of the current through the main current device over the entire intended current range and over the entire desired operating temperature range (e.g., -55°C to 150°C).

[0007] However, existing current meters present a conflict between the sensitivity of the device on the one hand and the linearity / accuracy of the current measurement under varying conditions on the other. This is because a high V CS The signal increases the device's sensitivity, while a low V CSThe signal improves the linearity and accuracy of the current sensor signal with respect to temperature and manufacturing process variations. From the perspective of linearity and accuracy, the signal V CS ideally close to zero volts, so that the current through the HEMT is a constant fraction of the current through the main power device over the entire current range.

[0008] The signal V CS Temperature can affect current measurement for various reasons. For example, the resistance of several components used in the current-sensing device, such as transistors and resistors, can vary with temperature. Furthermore, the precise current division ratio between, for example, a main HEMT transistor and a current-sensing HEMT transistor can also vary with temperature.

[0009] In contrast to silicon- or other semiconductor-based power devices, GaN HEMTs also suffer from a condition known as dynamic on-resistance (RDS(on)). on This condition is known as dynamic on-resistance. It can be described as a change in the on-resistance when a high voltage is applied to the main component. The dynamic on-resistance is important in relation to the power losses of the component during long-term operation, but it can also have a second-order effect in current sensing operations. For example, when a high voltage is applied, only the main HEMT and the current-sensing transistor of a GaN component are affected by the dynamic on-resistance. on affected, while the measuring load resistance R CS is practically unaffected. This further deteriorates the linearity of the current measurement across the entire current range and the accuracy of the current measurement as a function of temperature.

[0010] The applicant has therefore recognized that there is a need for a current measuring device with an improved compromise between sensitivity and measurement accuracy under various conditions.

[0011] US6433386 (B1) describes a read field-effect transistor (FET) capable of achieving many available read current ratios after fabrication, and a method for its fabrication. The read FET comprises a main cell array of metal-oxide-semiconductor field-effect transistor (MOSFET) cells connected in parallel and a main terminal connected to the sources of the main cells.

[0012] US10818786 (B1) describes a III nitride semiconductor-based heterojunction power device comprising a first heterojunction transistor formed on a substrate and a second heterojunction transistor formed on a substrate, wherein the second heterojunction transistor is used in sensing and protection functions of the first power heterojunction transistor.

[0013] US2022 / 0208761 (A1) describes a heterojunction power device based on a III nitride semiconductor, comprising: a first heterojunction transistor formed on a substrate, and a second heterojunction transistor formed on a substrate, the second heterojunction transistor being used in sensing and protection functions of the first power heterojunction transistor. The device also includes a monolithically integrated current sensing transistor having an substantially identical structure to the first heterojunction transistor, wherein the current sensing transistor is scaled to a smaller area or a shorter gate width compared to the first heterojunction transistor.

[0014] US11081578 (B1) describes a depletion-mode heterojunction device based on III nitride semiconductors comprising a substrate, a III nitride semiconductor region formed above the substrate, and at least two highly doped semiconductor regions spaced apart from each other.

[0015] US11217687 (B1) and US11404565 (B1) describe an integrated auxiliary gate terminal and pulldown network to achieve a normally off (E-mode) GaN transistor with a threshold voltage greater than 2 V, low gate leakage current and improved switching performance.

[0016] US6304108B1 describes a reference-corrected ratiometric current sensing circuit for sensing a current flowing through a load and a power-controlling pass-through device comprising a sensing device, a sensing resistor and a variable reference current source to provide a varying reference current.

[0017] US 2019 / 0 238 124 A1 describes a semiconductor chip with a correction circuit unit for correcting a digitally detected voltage from an analog-to-digital converter based on a detection result from the temperature sensor of a driver area and a detected result from the temperature sensor of a peripheral circuit area.

[0018] US 2020 / 0 395 905 A1 describes a multi-stage and forward-compensated complementary current field-effect transistor amplifier with a plurality of complementary pairs of novel current field-effect transistors connected in series to form the multi-stage amplifier. Summary

[0019] This disclosure generally relates to semiconductor structures and devices formed from wide-bandgap semiconductors, and in particular to Group III nitrides such as GaN and Al x GaN 1-x Structures, and heterostructures, that use two-dimensional electron gas layers (2DEG). Power devices with heterojunctions are high-voltage HEMTs.

[0020] According to a first aspect of the disclosure, an electronic device is provided comprising: a power device circuit comprising: (i) a main power transistor comprising a main drain terminal, a main source terminal, and a main gate terminal; (ii) a resistive sampling load; and (iii) a sampling transistor comprising a source terminal connected to the main source terminal, a gate terminal connected to the main gate terminal, and a drain terminal connected to a first terminal of the resistive sampling load. The electronic device further comprises: a feedback circuit comprising at least one operational amplifier and at least one transconductor; and a circuit comprising at least one switch for connecting and disconnecting the power device circuit from the feedback circuit.A first input of the at least one operational amplifier and a second terminal of the resistive load are selectively connected to the main drain terminal via the circuit. A second input of the at least one operational amplifier and at least one output of the at least one transconductor are selectively connected to the drain of the sensor transistor via the circuit. The main power transistor has a first area or gate circumference, and the sensor transistor has a second area or gate circumference that is smaller than the first area or gate circumference. During an ON state of the main power transistor, the output current of the feedback circuit's transconductor is proportional to the potential difference between the main drain terminal and the drain of the sensor transistor, resulting in a voltage of essentially zero across the resistive load.

[0021] If, during an OFF state of the main power transistor, the circuit is configured to disconnect the feedback circuit from the power device circuit, this results in an increase in the drain-source voltage of the sensor transistor until the voltage is essentially the same as the drain-source voltage of the main power transistor

[0022] Optionally, the output current of the feedback circuit is directly proportional to a current flowing through the main power transistor.

[0023] Optionally, the operational amplifier is configured to amplify a potential difference between the main drain terminal and the drain of the sensor transistor, wherein an output of the operational amplifier is connected to an input of the transconductor, and wherein an output current of the transconductor is proportional to the output of the operational amplifier.

[0024] Optionally, the feedback circuit is configured to supply the output current of the transconductor to the drain terminal of the sensor transistor to create a feedback loop configured to keep the potential difference between the main drain terminal and the drain of the sensor transistor at a constant value.

[0025] Optionally, the operational amplifier is configured to operate in a low-power mode using a conditioned gate driver signal, where the gate driver signal is inverted, time-shifted, or level-shifted.

[0026] Optionally, at least one transconductor is a p-channel transistor.

[0027] Optionally, the circuit is configured to: (i) connect the feedback circuit to the power supply circuit a predetermined time after the power supply circuit is turned on, and (ii) disconnect the feedback circuit from the power supply circuit a predetermined time before the power supply circuit is turned off.

[0028] Optionally, the circuit includes a first, a second, and a third switch configured to connect or disconnect the following: the main drain terminal to the first input of the operational amplifier of the feedback circuit, the drain of the sensor transistor to the second input of the operational amplifier, and an output of the transconductor to the drain of the sensor transistor.

[0029] Optionally, the first, second and third switches consist of HEMTs in enrichment mode.

[0030] Optionally, the first, second, and third switches comprise T-type switches with two series-connected enhancement-mode transistors, wherein the respective gate terminals of the first, second, and third switches are connected to a switch enable signal and a high-impedance component, the first end of which is connected between the respective two enhancement-mode transistors and the second end of which is connected to ground.

[0031] Optionally, the feedback circuit includes a current mirror circuit configured to produce an output current identical to the output current of the transducer.

[0032] Optionally, the electronic device is configured to convert the output current of the current mirror circuit into an output voltage signal using a resistor.

[0033] Optionally, the electronic device includes a digitally programmable decoder configured to adjust the amplification of the current mirror circuit in response to one or more digital inputs to the digitally programmable decoder.

[0034] Optionally, the electronic component includes a broadband semiconductor chip and a silicon semiconductor chip, with the power circuitry and integrated circuitry located on the broadband semiconductor chip and the feedback circuitry on a silicon semiconductor chip.

[0035] Optionally, the wide bandgap semiconductor chip and the silicon semiconductor chip are integrated into a single system-in-package.

[0036] Optionally, the silicon semiconductor chip includes at least one of the following elements: a gate driver circuit, an overcurrent protection circuit, an overtemperature detection and protection circuit, an undervoltage detection or blocking circuit, a voltage regulator, a bandgap reference circuit, a level shifter, an ESD protection circuit, a startup circuit, a logic circuit, a memory and / or a slew rate control circuit.

[0037] Optionally, the wide bandgap semiconductor chip is a gallium nitride (GaN) chip.

[0038] Optionally, the electronic component includes a broadband semiconductor chip and a silicon semiconductor chip, with the power circuitry located on the broadband semiconductor chip and the switching circuitry and feedback circuitry on a silicon semiconductor chip.

[0039] Optionally, the main power transistor includes an active heterojunction transistor comprising a Miller clamping transistor, wherein the Miller clamping transistor is monolithically integrated with the driver circuit of the active heterojunction transistor.

[0040] In general, the electronic device of the present disclosure uses a feedback circuit to ensure that the measuring transistor is always operated at the same voltage level as the main power transistor. This ensures that all inferences about the current flowing through the main power transistor, which are drawn with the aid of the measuring transistor, are more accurate, since the measuring transistor (with which such inferences are drawn) is subject to essentially the same operating conditions (such as voltage, temperature, etc.) as the main power transistor due to the output of the feedback circuit.

[0041] More generally, the sensor transistor has a similar architecture to the main power transistor, meaning that the sensor transistor and the main power transistor have essentially identical structures (i.e., they are made of the same materials and layers) and operating principles. However, the sensor transistor, which serves to infer the current flowing through the main power transistor, is scaled to a different area or gate circumference than the main power transistor by a specific scaling factor X. The scaling factor X is said to be greater than 1, meaning that the sensor transistor is smaller than the main power transistor by that factor. A scaling factor of 2, for example, would mean that the area or gate circumference of the main heterojunction is twice that of the sensor transistor.Conversely, a scaling factor of less than 1 would mean that the sensor transistor is larger than the main power transistor.

[0042] The sensitivity S of an ammeter can be defined as the ratio between the output signal (voltage) V CS and the current through the main power unit I main be defined such that: S=VCSImain

[0043] For a current measuring device, "ideal" linearity means that the sensitivity of the current measurement is linear over the entire current range, i.e., that the derivative of the sensitivity with respect to the current (dS / dI) is equal to 1. main ) is a constant value, or close to it, across the entire current range of the device. The linearity of the signal can therefore be defined as a measure of how close this derivative is to a constant value across the current range.

[0044] With ideal linearity, the signal V csan accurate measure of the current through the main power unit 19 (see Fig. ) across the entire current range, since V CS increases linearly with the current through the main power unit 19. In reality, the signal V CS For many reasons, some of which are explained below, the current measurement is not perfectly linear. However, the closer the linearity is to a constant value, the better the accuracy of the current measurement solution with respect to the current.

[0045] A factor that affects the linearity of the current sensor signal V CS The temperature can influence the signal V. CS With respect to temperature, it can be defined such that the derivative of the sensitivity with respect to temperature (dS / dT) is approximately zero, i.e., an “ideal” accuracy with respect to temperature means that this derivative is zero over the entire temperature range and there are no fluctuations in the linearity of the signal V. CSwith regard to temperature (at least over an expected or desired operating temperature range).

[0046] Another factor affecting the linearity of the current sensor signal V CS One factor that can influence the (manufacturing) process variation is process variation. In this context, process variation refers to a general parameter that describes the variability in the structure of the current measuring device, e.g., differences between metal layers, 2DEG charges, doping profiles, trap distribution within different semiconductor layers or at interfaces, and gate leakage between devices on the same chip, on the same wafer, or from wafer to wafer or batch to batch. The accuracy of the signal V CSWith respect to process variations, it can be defined as follows: the derivative of the sensitivity with respect to process variations (dS / dPV) is close to zero, i.e., an "ideal" accuracy with respect to process variations means that this derivative is zero over the entire range of statistical process variations and there is no variation in the linearity of the signal V. CS with regard to process variations.

[0047] Therefore, a more temperature-independent method of current measurement is needed, which also provides better dynamic Rds. on exhibits tolerance. From this, it can be concluded that the voltage signal across the measured load should ideally be very small compared to the voltage drop between the source and drain terminals (or the resistance of the measured load should be small compared to the resistance of the HEMT sensor transistor) to improve the accuracy of the current measurement solution.

[0048] In general terms, the aim of the present disclosure is to provide a current measurement circuit capable of maintaining high linearity, accuracy, and high sensitivity over the desired current and temperature range, and exhibiting high immunity to changes in dynamic R on to offer. The present disclosure offers a solution by using a feedback circuit, as described below.

[0049] As mentioned previously, maintaining a largely identical drain-source bias for the main power transistor and the sensor transistor in both the on and off states can help improve the performance of the current sensing circuit.

[0050] As described above, in one example, the feedback circuit is configured to keep the drain potential of the sensor transistor largely at the same potential as the drain potential of the main power transistor. As described above, the drain terminal of the main power transistor can be connected as a first input to the feedback circuit, and the drain terminal of the sensor transistor can be connected as a second input to the feedback circuit. In one example, the feedback circuit can include an operational amplifier (op-amp) that receives the voltage signal inputs to the feedback circuit. In operation, the op-amp can be configured such that an increase in the input voltage (i.e., a difference between the drain potential of the main transistor and the drain potential of the sensor transistor) to the op-amp results in an increase in the output of the op-amp.The output of the operational amplifier can be connected to a transconductor, which may, for example, consist of at least one transistor. The output of the operational amplifier would thus be connected to the gate of the at least one transistor of the transconductor, which controls the output of the feedback circuit. The output of the feedback circuit can be (indirectly) connected to the sampling terminal of the sampling transistor, as described in the first aspect of the present invention, thereby creating a feedback loop configured such that the drain potential of the main power transistor and the sampling transistor is largely kept at the same potential. This is possible if the output current of the feedback circuit settles to an equilibrium output current of IDS / X, where IDS is the drain-to-source current through the main power transistor in one-state operation and X is the scale factor defined herein.

[0051] The output current IDS / X of the feedback circuit can be mirrored and provided as a current measurement output signal for the power device. The current signal can be converted into a voltage signal using a resistor. In some examples, the current mirror gain can be adjusted using a digital decoder.

[0052] It may be desirable for the drain terminals of the main power transistor and the sensor transistor to be disconnected as inputs to the feedback circuit in the off state. Likewise, the output of the feedback circuit can be disconnected from the drain terminal of the sensor transistor. This may be necessary to protect the operational amplifier, since the drain terminals of the main power transistor and the sensor transistor can reach a high voltage (e.g., >50 V) in the off state. This is the function of the circuit described in the first aspect of the present invention. The described switches must conduct during the on-state operating mode of the power transistor and block during the off-state operating mode of the power transistor. The signal controlling the switch can therefore be largely similar to the signal driving the power transistor.However, since it is important that the feedback circuit is always open when a high voltage is applied to the drain of the power transistor, the edges of the signal controlling the switch may be time-shifted relative to the edges of the signal controlling the power transistor. In other examples, the signal controlling the switch may be level-shifted relative to the gate driver signal of the power transistor.

[0053] In the off state, and with the feedback circuit disconnected from the drains of the power and measurement transistors, the measured load ensures that the drains of the power and measurement transistors are largely at the same potential. In this state, the off-state leakage is low, and therefore the voltage drop across the measured load is negligible. By maintaining a largely identical drain-source voltage across the main transistor and the sensor transistor during the off state, the dynamic Rds should be low. on -The rise of the main transistor and the sensor transistor should be largely identical.

[0054] In some examples, the power dissipation of an operational amplifier or feedback circuit can depend on the gate signal of the main power transistor. A conditioned version of the gate signal (e.g., inverted, time-shifted, level-shifted) can be used to enable or disable a low-power mode for the operational amplifier or feedback circuit.

[0055] The described design considerations aim to improve the linearity and accuracy of the current sensor signal with respect to temperature and manufacturing process variations, while simultaneously making the sensor immune to changes in the dynamic R on close.

[0056] The power device circuit in the above aspect of this disclosure can be based on a wide-bandgap semiconductor material. An example of such a material is gallium nitride (GaN), and in particular devices based on AlGaN / GaN heterojunction structures. Other examples of wide-bandgap semiconductor materials include silicon carbide, gallium arsenide, indium phosphide, and diamond. Other circuits described in the first aspect of this invention can also be based on a wide-bandgap material or on a different material, such that the overall circuit is implemented on two chips made of different semiconductor materials.

[0057] The electronic component described in the above aspect of this disclosure can be implemented on two different semiconductor materials, which can, however, be present on the same chip. An example is GaN-on-Si technology, where part of the circuit is implemented on GaN (e.g., the high-voltage components) and part of the circuit is implemented on silicon.

[0058] In some examples, the feedback circuit can be implemented on a co-packed silicon chip, rather than being implemented monolithically on the same GaN chip as the main and sensor transistors. Implementing the feedback circuit on silicon can offer several advantages, such as lower power consumption and higher speed, due to the availability of mature p-channel devices in silicon, which are at a much earlier stage of GaN technology development.

[0059] The silicon chip can be housed in the same package as the GaN chip; that is, they are packaged together. Including a silicon chip in the package allows for the implementation of additional functions in a more mature technology. Examples of blocks that can be developed on silicon to complement the operation of the power transistor in GaN include gate drive, bandgap reference, overcurrent protection (OCP), temperature sensing and overtemperature protection (OTP), voltage regulators, slew rate control, UVLO, level shifters, ESD protection, logic circuits, start-up circuits, and so on.

[0060] It can still be advantageous to use certain additional components / circuits / circuit blocks in GaN. For example, high-voltage transistors in addition to the main power transistor, components / circuits where the parasitics in the connection to the main power transistor are important (e.g., Miller terminal), and / or components where adaptation to the process / temperature is important.

[0061] In some examples, the switches that connect / disconnect the feedback circuit may be made of GaN, and in others, of silicon. Designing these switches in GaN may be preferable because they need to block high voltage and can therefore employ a similar technology to the power transistor. Brief description of the drawings

[0062] The present disclosure can be better understood with reference to the accompanying drawings, which, however, are not to be understood as limiting the disclosure to the specific embodiments shown, but merely serve for explanation and understanding. In Fig. The diagram schematically shows an example of a sensor field-effect transistor (FET). Fig. schematically shows an example of a current measurement circuit with an external amplifier. Fig. schematically shows an example block diagram of a current measurement circuit according to one aspect of the present disclosure. Fig. Figure 1 schematically shows an exemplary current measurement circuit according to one aspect of the present disclosure. In Fig. This is an example of the impulse sequences according to one aspect of the present revelation. Fig. schematically shows a second example of a current measurement circuit with an external current output according to one aspect of the present disclosure. Fig. schematically shows a third example of a current measurement circuit with an external current output according to one aspect of the present disclosure. Fig. schematically shows a fourth example of a current measurement circuit according to one aspect of the present disclosure. Fig. schematically shows an example switch according to one aspect of the present disclosure. In Fig. A second example switch is shown schematically according to one aspect of the present disclosure. Fig. schematically shows a fifth example of a current measurement circuit according to one aspect of the present disclosure. Fig. shows a block diagram of an electronic device according to one aspect of the present disclosure. In Fig. is an example of a Miller bracket circuit according to one aspect of the present disclosure. Detailed description of the drawings

[0063] Fig. This shows an example of a sense FET. The sense FET includes a main FET Q. m and a sub-FET Q S The main FET and the sub-FET are connected in parallel, but there are two sources. The main source S m The main FET is grounded, while the sampling source S S of the sub-FET with an external sampling resistor R S(ext) is connected to prevent a voltage drop V s to detect the sampling resistance. An external connection for the R S(ext)Connection is provided via a pin. The main and sub-FETs consist of an array of identically constructed cells and therefore have identical current-voltage characteristics. However, the main FET can be equipped with many times (e.g., hundreds or thousands of times) the number of cells compared to the sub-FET. Consequently, a current I flows through the sub-FET. s , which represents a certain percentage of the electricity I m of the main FET, since the on-resistance of the MOSFET is inversely proportional to the number of cells in the FET, which means that the on-resistance of the sub-FET is much higher than that of the main FET.

[0064] Fig. Figure 1 shows an example of a monolithically integrated GaN chip 200 with a main power device or first power transistor 19 and a current-sense transistor 16. In this example, the current-sense transistor 16 is structurally identical to the main power device 19, but its surface area or number of fingers is reduced in a known ratio compared to the main power device, e.g., such that the current-sense transistor 16 is 100x or 1000x smaller than the main power device 19. The main power device 16 is a main high-voltage HEMT, and the current-sense transistor is a high-voltage sensing HEMT. The source of the current-sense transistor 16 is connected to a load or a current-sensing resistor 15. The resistor 15 can also be monolithically integrated with the main power device 19 and the current-sense transistor 16 or provided externally. In this arrangement, the sampling signal V CSproportional to the current through the scanning resistor 15. Under ideal conditions, the current through the measuring resistor 15 is a constant fraction of the current through the main current device 19 over the entire current range and over the entire desired temperature range (e.g. -55 C to 150 C).

[0065] The GaN chip 200 can V CS by increasing or decreasing the resistance R CS The sampling resistor 15 can be set. During linear one-state operation, the main HEMT device 19 and the HEMT sensor transistor 16 can be modeled as resistors controlled by the gate voltage potential (with the resistors Ron-main or Ron-sense ). At low voltage CS (which generally offers better linearity of the current measurement solution) is the ratio between R CS and Ron-sense Ideally small (e.g., less than 0.1). The reason for this is that with this model: VCS=RCSRCS+Ron−senseVds where V ds The voltage drop across the drain-source terminals of the current-sensing transistor 16 is due to R. CS compared to Ron-sense If the effect of the dynamic R is small, it will be less pronounced. on the current measuring transistor 16 was mitigated and the current measuring accuracy of the device was improved.

[0066] A solution for the compromise between sensitivity and linearity of the signal V CS consists of V CS to keep relatively low in order to improve the linearity and accuracy of the current measurement (e.g. with respect to temperature and process fluctuations), and to provide an external amplifier 50 connected to node V CS is connected to increase the sensitivity of the current measuring device, as in Fig. This is shown. In this case, the sensitivity of the current measuring device is determined by the ratio between the output signal V. o(that the product of V CS and the amplifier's gain of 50 corresponds to) and the current I main the main power device 19 provides the necessary current. However, providing an external amplifier 50 leads to a longer bill of materials and thus higher production costs. Furthermore, parasitic components (e.g., parasitic inductances, capacitances, and resistances) resulting from the additional connections between the GaN chip 200 and the external chip of the amplifier 50 can affect the current measurements, especially at high operating frequencies. Any attempt to monolithically integrate the amplifier 50 with the power switching device 19 within the GaN chip 200 is significantly hampered by the fact that the potential at the current sensing node V CS is very low (in most cases lower than the threshold voltage of an enrichment transistor) and the measuring load 15 is connected to ground.

[0067] On the other hand, as already mentioned, it is often advantageous to limit the maximum voltage across a load. If this sampling voltage is too high (e.g., greater than 0.3 V), the current through the current-sampling transistor can no longer scale with the current in the main transistor, thus compromising the linearity of the sampling signal V. CSis affected. Furthermore, as mentioned earlier, linearity and accuracy can be further affected across the temperature range and in the presence of other factors such as dynamic on-resistance (specific to GaN HEMTs) and manufacturing tolerances. It follows from the above that it is sometimes advantageous to keep the maximum voltage drop across the measured load as small as possible, e.g., less than 0.3 V, less than 0.2 V, or even less than 0.1 V. When the load and the transistor are connected in series, the sum of the voltages across the measured load and the measuring transistor (i.e., between the drain and source terminals of the measuring transistor) is equal to or approximately equal to the voltage drop between the drain and source terminals of the main transistor.Since the voltage drop across the source-drain terminals of the main transistor is often less than 2 V in normal one-state operation, and it is advantageous for the voltage drop across the sampling load to be on the order of 0.2 V or less, the resistance of the sampling load can be chosen to be less than 1 / 10 of the one-state resistance of the sampling load transistor. Typical values ​​for these resistors can be in the ohm range (e.g., 1–5 ohms). However, it goes without saying that all the above values ​​can vary considerably depending on the specific ratings and applications in which the device is used.

[0068] Consequently, for the current-measuring devices described in this disclosure, it is advantageous, but not required, that the voltage signal at the measured load be small compared to the voltage drop between the source and drain terminals of the measuring transistor (and / or that the resistance of the measured load be small compared to the resistance of the measuring transistor) to improve the accuracy of the current-measuring solution. This applies similarly when the measured load is placed between the drain terminals of the main power transistor and the measuring transistor, as in Fig. depicted.

[0069] Fig. Figure 1 shows a schematic diagram of a current sensor architecture according to the present disclosure, illustrating at a high level the advantages of a feedback circuit. The architecture comprises a power device circuit 10, an integrated circuit 30, and a feedback circuit 40. The power device circuit 10 includes a power transistor Q1, a sensing transistor Q2, and a resistive load R_S. The integrated circuit 30 connects and disconnects the power device circuit 10 from the feedback circuit 40 based on a Switch_Enable signal. The feedback circuit 40 includes an operational amplifier AMP1 (401) and a transconductor 402.

[0070] In the Fig. In the illustrated example, the feedback circuit 40 consists of an operational amplifier AMP1 and an additional p-channel transistor Q3 as a transconductor. The circuit 30 consists of three switches 1, 2, 3. In the steady state of the power transistor, it is ensured that the measuring transistor always "follows" the power transistor with respect to the voltages between the terminals and that the voltage drop across the resistive load is virtually zero.

[0071] In active current sensing mode, switches 1, 2, and 3 are closed when the gate signal is HIGH and the feedback circuit 40 is connected. To ensure that the feedback circuit 40 is protected, the signal to close switches 1, 2, and 3 can be delayed relative to the gate signal going HIGH. This ensures that sufficient time is allowed for the drain-source potential at the main power transistor to fall during turn-on before the feedback circuit 40 is connected. An example pulse train is shown in Fig. depicted.

[0072] When switched on, the VDS of the power transistor is also established across the measuring transistor due to the feedback circuit. Ideally, the current through the resistor (R_S) is zero, since VDS ideally equals VDSS.

[0073] In this way, the output of the feedback circuit contains a current signal that represents a known ratio to the current through the main power transistor. The measuring transistor is similar in construction to the main power transistor, but has a significantly smaller area or gate circumference, by a specific ratio m=X, which determines the ratio between the measured current in the measuring transistor and the actual current in the power transistor.

[0074] When the gate signal is LOW and the main power transistor is operating in the off state, switches 1, 2, and 3 open, and the feedback circuit is interrupted. This prevents a high voltage from occurring at the input of the feedback circuit when the transistor is on and off. To ensure that the feedback circuit is switched off before the drain potential of the main power transistor rises, the signal switches 1, 2, and 3 can precede the gate signal by a short time interval, as shown in Fig. depicted.

[0075] The voltage drop across the power transistor is established via the measuring transistor through the resistor R_S. The voltage drop across R_S is negligible because the leakage current through the sensor transistor in the off state is very low (e.g., <1 µA). Therefore, the dynamic R on -Changes in both the sensor transistor and the power HEMT were simulated.

[0076] The example in Fig. shows how a current IDS / X can be obtained in the circuit. In another example, which is in Fig. As shown, the measured current IDS / X can be mirrored using transistors Q4 and / or Q5. In the case of current mirroring with transistor Q5, the measured current can be provided as an external current output. With current mirroring using transistor Q4, the output current can be converted into a current-measuring voltage signal using an external resistor. The external resistor can be selected to meet the specifications of the control unit used in the application with the electronic device.

[0077] In another example, additional flexibility can be achieved by using a 501 digital decoder, which can adjust the current mirror gain by changing the size of transistors Q4 and / or Q5. This is shown in Fig. The input of the digital decoder can be provided as external terminals g0, g1, so that the gain of the current mirror can be adjusted by signals outside the chip.

[0078] As described here, the feedback circuit 40, and consequently the operational amplifier AMP1, is switched off when the gate signal is low. Since the operational amplifier is inoperative during this time, it would be advantageous to limit its power dissipation. This can be achieved by supplying a low-power enable signal to the operational amplifier during the period when the gate signal is low. The low-power enable signal can therefore be an inverted gate driver signal, as shown in Fig. depicted.

[0079] Fig. Figure 1 shows an example for switches 30a, 30b, and 30c. The switch can be a simple enhancement transistor similar to the main power transistor (in construction, but not in size). In the case of switch 30a, for example, the drain terminal of the switch is connected to the drain terminal of the main power transistor.

[0080] Fig. shows a schematic representation of the switch SWITCH1, which controls the V DS to the IN input of the operational amplifier AMP1 according to the present disclosure. The other switches, SWITCH2 and SWITCH3, also have the same structure. These switches prevent a high voltage from being applied to the operational amplifier during the power circuit's shutdown and also allow feedback from the transconductor to the drain terminal of the sensor transistor.

[0081] Fig. Figure 1 shows another example of the present disclosure, in which the current sensing architecture comprises a power circuit 10 with a power transistor Q1 and a sensing transistor Q2 implemented on a wide-bandgap semiconductor chip 60, and a feedback circuit 40 implemented on a silicon chip 20a. Implementing the power circuit on a wide-bandgap semiconductor chip can enable effective high-voltage operation. The circuit 30 can also preferably be implemented on a wide-bandgap semiconductor chip 60, since the switches are required to block high voltage and can therefore employ a similar technology to the power transistor. Because the switches allow the selective application of signals to the input of the operational amplifier and protect it from high voltage, it can be advantageous to implement the operational amplifier on a silicon chip.

[0082] In one implementation of this example, the wide-bandgap semiconductor chip 60 can be made of gallium nitride (GaN). Other examples of wide-bandgap semiconductor chips include silicon carbide, gallium arsenide, indium phosphide, and diamond. Implementing the feedback circuit 40 on a silicon chip, rather than monolithically integrating the feedback circuit on the GaN chip, can offer advantages in terms of speed, power dissipation, design flexibility, yield, etc., since silicon technology is more mature than GaN technology, particularly with regard to the availability of p-channel devices.

[0083] In another example, the electronic device comprises a system in a package with a GaN semiconductor chip 60 and a silicon semiconductor chip 20, as in Fig. The GaN semiconductor chip 60 contains the power transistor and the sense transistor, while the silicon semiconductor chip accommodates a selection of control and protection circuits. These control and protection circuits include, but are not limited to, gate drivers, overcurrent protection, overtemperature detection and protection, operational amplifiers, voltage regulators, bandgap reference, slew rate control, UVLO, level shifters, ESD protection, logic circuits, startup circuits, and / or other control and protection circuits. A Miller clamp 104 may also be present on the GaN semiconductor chip 60.

[0084] Fig. shows an example of the in Fig.The Miller clamp shown is 104. The Miller clamp can be driven by a driver circuit (e.g., an inverter) located on the same GaN die or, alternatively, on the silicon die. In other examples, the Miller clamp can be a distributed Miller clamp, meaning it consists of a network of subtransistors connected to a distributed network of main power subtransistors. This can result in a Miller clamp subtransistor and its associated main power transistor being located closer together, so that the connections between them have smaller parasitic components. This can suppress oscillations and / or capacitive coupling effects in the device during operation.

[0085] It should be noted that the embodiments described above are only some examples and that the present disclosure can also be implemented in other embodiments, e.g., when all circuits are monolithically integrated in a GaN semiconductor package, when a combination of GaN and silicon semiconductor devices is present in a single package, or when one or more other types of semiconductor devices, such as silicon carbide, gallium arsenide, indium phosphide, or diamond, are present in a single package.

[0086] The person skilled in the art will understand that in the foregoing description and the attached claims, positional terms such as "above", "overlapping", "below", "sideways", etc., are used with reference to conceptual representations of a device, e.g., those showing standard cross-sectional perspectives and those depicted in the attached drawings. These terms are to be understood as referring to a device in an orientation such as that shown in the attached drawings.

[0087] Although the disclosure has been described in terms of preferred embodiments, as set out above, it is to be understood that these embodiments serve only for illustration and that the claims are not limited to these embodiments. The person skilled in the art may make modifications and alternatives to the disclosure which fall within the scope of the appended claims. Any feature disclosed or illustrated in the present description may be included in the disclosure, either alone or in a suitable combination with any other feature disclosed or illustrated herein.

Claims

[1] Electronic device with: a circuit for a power device comprising the following: a main power transistor comprising a main drain terminal, a main source terminal, and a main gate terminal, a resistive load, and a sensor transistor with a source connected to the main source terminal, a gate connected to the main gate terminal, and a drain connected to a first terminal of the resistive load, a feedback circuit comprising at least one operational amplifier and at least one transconductor, and a circuit with at least one switch for connecting and disconnecting the circuit for the power device with the feedback circuit, wherein a first input of the at least one operational amplifier and a second connection of the resistive load are selectively connected to the main drain connection via the circuit, wherein a second input of the at least one operational amplifier and at least one output of the at least one transconductor are selectively connected to the drain of the sensor transistor via the circuit, and wherein the main power transistor has a first area or a first gate circumference, wherein the sensor transistor has a second area or a second gate circumference that is smaller than the first area or the first gate circumference, and wherein, during an ON state of the main power transistor, the output current of the feedback circuit's transconductor is proportional to the potential difference between the main drain terminal and the drain of the sensor transistor, resulting in a voltage across the resistive load of essentially zero; and wherein, during an OFF state of the main power transistor, the circuit is configured to disconnect the feedback circuit from the circuit for the power device, resulting in an increase in the drain-source voltage of the sensor transistor until the voltage is essentially the same as the drain-source voltage of the main power transistor. [2] Electronic device according to claim 1, wherein an output current of the feedback circuit is directly proportional to a current flowing through the main power transistor. [3] Electronic device according to one of claims 1 to 2, wherein the operational amplifier is configured to amplify a potential difference between the main drain terminal and the drain of the sensor transistor, wherein an output of the operational amplifier is connected to an input of the transconductor, and where the output current of the transconductor is proportional to the output of the operational amplifier. [4] Electronic device according to claim 3, wherein the feedback circuit is configured to supply the output current of the transconductor to the drain terminal of the sensor transistor in order to generate a feedback loop configured to maintain the potential difference between the main drain terminal and the drain of the sensor transistor at a constant value. [5] Electronic device according to any one of claims 1 to 4, wherein the operational amplifier is configured to operate in a low power mode by a conditioned gate driver signal, wherein the gate driver signal is inverted, time-shifted or level-shifted. [6] Electronic device according to any one of claims 1 to 5, wherein the at least one transconductor is a p-channel transistor. [7] Electronic device according to any one of claims 1 to 6, wherein the circuit is configured to (i) connect the feedback circuit to the power device circuit a predetermined time after the power device circuit is switched on and (ii) disconnect the feedback circuit from the power device circuit a predetermined time before switching it off. [8] Electronic device according to any one of claims 1 to 7, wherein the circuit comprises a first, a second and a third switch configured to connect and disconnect: the main drain connection to the first input of the operational amplifier of the feedback circuit, connects the drain of the sensor transistor to the second input of the operational amplifier, and connect the output of the transconductor to the drain of the sensor transistor. [9] Electronic device according to claim 8, wherein the first, second and third switches comprise HEMTs in enrichment mode. [10] Electronic device according to claim 8, wherein the first, second and third switches comprise T-type switches with two enhancement mode transistors connected in series, wherein the respective gate terminals of the first, second and third switches are connected to a switch enable signal and a high-impedance component, the first end of which is connected between the respective two enhancement mode transistors and the second end of which is connected to ground. [11] Electronic device according to any one of claims 1 to 10, wherein the feedback circuit comprises a current mirror circuit configured to produce an output current identical to the output current of the transducer. [12] Electronic device according to claim 11, wherein the electronic device is configured to convert the output current of the current mirror circuit into an output voltage signal using a resistor. [13] Electronic device according to claim 11, comprising a digitally programmable decoder configured to set an amplification of the current mirror circuit in response to one or more digital inputs into the digitally programmable decoder. [14] Electronic device according to any one of claims 1 to 13, comprising a wide bandgap semiconductor chip and a silicon semiconductor chip, wherein the power device circuit and the circuitry are provided on the wide bandgap semiconductor chip and the feedback circuitry is provided on a silicon semiconductor chip. [15] Electronic device according to claim 14, wherein the wide bandgap semiconductor chip and the silicon semiconductor chip are contained in a single system-in-package. [16] Electronic device according to claim 14, wherein the silicon semiconductor chip comprises at least one of the following elements: a gate driver circuit; an overcurrent protection circuit; a circuit for detecting and protecting against over-temperatures; an undervoltage detection or an interlock circuit; a voltage regulator; a bandgap reference circuit; a level converter; an ESD protection circuit; a startup circuit; a logic circuit; a memory; and / or a circuit for controlling the rate of ascent. [17] Electronic device according to claim 14, wherein the wide bandgap semiconductor chip is a gallium nitride (GaN) chip. [18] Electronic device according to any one of claims 1 to 13, comprising a wide bandgap semiconductor chip and a silicon semiconductor chip, wherein the power device circuit is provided on the wide bandgap semiconductor chip and the circuit and feedback circuit are provided on a silicon semiconductor chip. [19] Electronic device according to any one of claims 1 to 18, wherein the main power transistor comprises an active heterojunction transistor comprising a Miller clamp transistor, and wherein the Miller clamp transistor is monolithically integrated with the driver circuit of the active heterojunction transistor.

Images