Embodiments of this application relate to the electronic circuit technologies, and in particular, to a clock signal generator, an on-chip clock system, and a chip.
With development of electronic technologies such as an artificial intelligence technology, a communications technology, and a semiconductor technology, functions of an on-chip system become increasingly complex. To ensure high reliability and a high running rate of the on-chip system, a clock signal generator generally needs to be disposed inside the on-chip system, so as to provide an accurate and stable clock signal for the on-chip system.
In a related technology, a flip-flop and a comparator are usually disposed in a clock signal generator. A signal that is output by an output end of the comparator is provided to an input end of the flip-flop, so that the flip-flop generates a clock pulse at an output end based on the signal received by the input end. Generally, there is a delay between the input end and the output end of the comparator, and the delay usually changes based on a static operating point, a temperature drift, and the like of a circuit. As a result, a clock frequency output by a clock output circuit changes, and stability of a frequency of a clock signal output by the clock signal generator is reduced. Therefore, how to enable the clock signal generator to output a clock signal with a stable frequency becomes a problem.
According to a clock signal generator, an on-chip clock system, and a chip provided in this application, input ends of a flip-flop are respectively coupled to a first electrode of a first transistor and a first electrode of a second transistor, so that an output frequency of the clock signal generator is related only to turn-on/off times of the first transistor and the second transistor. In this way, the clock signal generator can generate a clock signal with a stable frequency.
The following technical solutions are used in this application.
According to a first aspect, an embodiment of this application provides a clock signal generator. The clock signal generator includes a first transistor, a second transistor, a flip-flop, and a power supply end. A first electrode of the first transistor and a first electrode of the second transistor are coupled to the power supply end, and a second electrode of the first transistor and a second electrode of the second transistor are coupled to the common ground. A first input end of the flip-flop is coupled to the first electrode of the first transistor, and a second input end of the flip-flop is coupled to the first electrode of the second transistor.
The first input end and the second input end of the flip-flop are respectively coupled to the first electrode of the first transistor and the first electrode of the second transistor, so that a high-level signal or a low-level signal received by the first input end and the second input end of the flip-flop is related only to turn-on times or turn-off times of the first transistor and the second transistor, and a period or a frequency of a clock pulse signal at an output end of the flip-flop is related only to the turn-on times and the turn-off times of the first transistor and the second transistor. In this way, a clock frequency change caused by a change of a delay of a signal received by the flip-flop with a temperature can be avoided, so that a frequency of a clock signal output by the clock signal generator can be more stable.
In a possible implementation, the clock signal generator further includes a control signal generation circuit. Output ends of the control signal generation circuit are respectively coupled to a control electrode of the first transistor and a control electrode of the second transistor. The control signal generation circuit periodically provides a control signal to the control electrode of the first transistor and the control electrode of the second transistor, so that the first transistor and the second transistor are alternately turned on and turned off based on the control signal.
In a possible implementation, the control signal output by the output end of the control signal generation circuit is a voltage signal. In this case, the control signal generation circuit may include two output ends, where one output end is coupled to the control electrode of the first transistor, and the other output end is coupled to the control electrode of the second transistor. The control signal generation circuit may apply a generated voltage signal between the control electrode and the first electrode of the first transistor and between the control electrode and the second electrode of the second transistor, so as to alternately control turn-on and turn-off of the first transistor and the second transistor.
In a possible implementation, the clock signal generator further includes a first capacitor and a second capacitor. A first electrode of the first capacitor is coupled to the control electrode of the first transistor. A second electrode of the first capacitor is coupled to the common ground. The first transistor is periodically turned on or turned off based on charging/discharging of the first capacitor. A first electrode of the second capacitor is coupled to the control electrode of the second transistor. A second electrode of the second capacitor is coupled to the common ground. The second transistor is periodically turned on or turned off based on charging/discharging of the second capacitor.
The first transistor and the second transistor are controlled through charging/discharging of the first capacitor and the second capacitor, so as to trigger level flipping of a first output end and a second output end of the flip-flop to generate a clock signal. A charging/discharging time of a capacitor is related to a material, a capacity, and a charging/discharging constant of the capacitor. Therefore, a clock frequency output by the clock signal generator can be controlled by controlling charging/discharging times of the first capacitor and the second capacitor. This avoids a change of a frequency of an output clock signal caused by a temperature drift or a change of a static operating point of another component such as an integrated operational amplifier in the clock signal generator, so that the frequency of the clock signal output by the clock signal generator can be more stable. In addition, in some application scenarios, when a relatively high clock frequency is required, a capacitor with a small capacity and a short charging/discharging time may be used. When a relatively low clock frequency is required, a capacitor with a large capacity and a long charging/discharging time may be used. In this way, a maximum frequency of a clock signal to be output by the clock signal generator can be effectively increased. This further improves performance of the clock signal generator. The control signal output by the output end of the control signal generation circuit is a current signal. The control signal generation circuit alternately and periodically provides the current signal to the first electrode of the first capacitor and the first electrode of the second capacitor to charge the first capacitor and the second capacitor.
In a possible implementation, the clock signal generator further includes a third transistor. A control electrode and a first electrode of the third transistor are coupled to the power supply end, and a second electrode of the third transistor is coupled to the common ground. The control signal generation circuit generates a current signal based on a voltage signal between the first electrode and the second electrode of the third transistor.
In a possible implementation, the clock signal generator further includes a current mirror circuit. The control signal generation circuit converts a voltage between the first electrode and the second electrode of the third transistor into a current, and provides the current to the current mirror circuit. The current mirror circuit performs mirror processing on the received current based on a preset proportion, and provides a processed current to the first electrode of the first capacitor and the first electrode of the second capacitor.
A requirement on the control signal generation circuit can be reduced by disposing the current mirror circuit. When a relatively large current needs to be input to the clock signal generator, the current mirror circuit may be used for implementation, so that stability of the control signal generation circuit can be improved. This further improves stability of the clock signal generated by the clock signal generator.
In a possible implementation, the clock signal generator further includes a first selector and a second selector. A control end of the first selector is coupled to the first output end of the flip-flop. A first input end of the first selector is coupled to the output end of the control signal generation circuit. A second input end of the first selector is coupled to the common ground. An output end of the first selector is coupled to the control electrode of the first transistor. A control end of the second selector is coupled to the second output end of the flip-flop. A first input end of the second selector is coupled to the output end of the control signal generation circuit. A second input end of the second selector is coupled to the common ground. An output end of the second selector is coupled to the control electrode of the second transistor. The first selector periodically gates the first input end and the second input end of the first selector under control of a clock signal that is output by the first output end of the flip-flop, so that the first capacitor is periodically charged/discharged. The second selector periodically gates the first input end and the second input end of the second selector under control of the clock signal that is output by the first output end of the flip-flop, so that the second capacitor is periodically charged/discharged.
In a possible implementation, the first input end of the flip-flop is coupled to the first electrode of the second transistor through an even quantity of phase inverters.
In a possible implementation, the second input end of the flip-flop is coupled to the first electrode of the third transistor through an even quantity of phase inverters.
A driving capability of a circuit can be improved by disposing the phase inverter. This increases flipping speeds of level signals at the first input end and the second input end of the flip-flop.
In a possible implementation, a quantity of first transistors, a quantity of second transistors, and a quantity of third transistors are in proportion. For example, the quantity of first transistors, the quantity of second transistors, and the quantity of third transistors are 1:1:1. A parameter of the first transistor, a parameter of the second transistor, and a parameter of the third transistor are the same. Specifically, the first transistor, the second transistor, and the third transistor may have a same physical parameter and a same operating parameter, that is, have same parameters such as a conduction voltage drop, an internal resistance, and a power. In this way, the temperature drift of the static operating point of the clock signal generator is reduced, so that the static operating point of the clock signal generator is more stable.
According to a second aspect, an embodiment of this application provides an on-chip clock system. The on-chip clock system includes the clock signal generator according to the first aspect.
According to a third aspect, an embodiment of this application provides a chip. The chip includes the on-chip clock system according to the second aspect.
To describe the technical solutions in embodiments of this application more clearly, the following briefly introduces the accompanying drawings for describing embodiments of this application. It is clear that the accompanying drawings in the following description show merely a part of embodiments of this application, and a person of ordinary skill in the art may still derive other drawings from these accompanying drawings without creative efforts.
The following clearly describes the technical solutions in embodiments of this application with reference to the accompanying drawings in embodiments of this application. It is clear that the described embodiments are a part but not all of embodiments of this application. All other embodiments obtained by a person of ordinary skill in the art based on embodiments of this application without creative efforts shall fall within the protection scope of this application.
The term “first”, “second”, or the like mentioned in this specification does not indicate any order, quantity, or importance, but is used only for distinguishing between different components. Likewise, the term “a/an”, “one”, or the like is not intended to indicate a quantity limitation either, but is intended to indicate at least one. The term “coupling”, “connection”, “link”, or the like is not limited to a physical or mechanical connection, but may include an electrical connection, whether directly or indirectly.
The term “unit” mentioned in this specification is usually a functional structure that is obtained through division based on logic, and the “unit” may be implemented only by hardware, or implemented by a combination of software and hardware.
In embodiments of this application, the term “and/or” describes an association relationship between associated objects and indicates that at least three relationships may exist. For example, A and/or B may indicate the following three cases: Only A exists, both A and B exist, and only B exists.
In embodiments of this application, the word “example”, “for example”, or the like is used to represent giving an example, an illustration, or a description. Any embodiment or design scheme described as “example” or “for example” in embodiments of this application should not be explained as being more or having more advantages than another embodiment or design scheme. Exactly, use of the word “example”, “for example”, or the like is intended to present a related concept in a specific manner.
In the descriptions of embodiments of this application, unless otherwise stated, “a plurality of” means two or more than two. For example, a plurality of processing units refer to two or more processing units. A plurality of systems refer to two or more systems.
In
In this embodiment, the transistor T1 and the transistor T2 each include a control electrode, a first electrode, and a second electrode. The first electrode of the transistor T1 and the first electrode of the transistor T2 are separately coupled to the power supply end Vdd. The power supply end Vdd is configured to receive electric energy provided externally. The second electrode of the transistor T1 and the second electrode of the transistor T2 are separately coupled to a common ground Gnd. The transistor T1 and the transistor T2 herein may be insulated gate field effect transistors, for example, may be PMOS transistors or NMOS transistors. This is not limited herein. When the transistor T1 and the transistor T2 are NMOS transistors, the control electrode may be a gate electrode, the first electrode may be a drain electrode, and the second electrode may be a source electrode. When the transistor T1 and the transistor T2 are PMOS transistors, the control electrode may be a gate electrode, the first electrode may be a source electrode, and the second electrode may be a drain electrode. The figure schematically shows a case in which the transistor T1 and the transistor T2 are NMOS transistors.
The SR flip-flop includes an input end S, an input end R, an output end QN, and an output end Q. The input end S of the SR flip-flop is connected to the first electrode of the transistor T1. The input end R of the SR flip-flop is connected to the first electrode of the transistor T2. The output end QN or the output end Q of the SR flip-flop is configured to output a clock signal.
As shown in
Specifically, as shown in
With reference to the circuit structure shown in
A working status of the transistor T1 is used as an example. A signal received by the input end S of the SR flip-flop is first described. When the transistor T1 is turned on, an electric potential at a node a is an internal resistance voltage drop of the transistor T1. Generally, the internal resistance voltage drop of the transistor T1 is very small and can be almost ignored. Herein, the electric potential at the node a may be considered as 0 V. In this case, the input end S of the SR flip-flop receives a low-level signal or “logic 0”. When the transistor T1 is turned off, the electric potential at the node a is an electric potential of the power supply end Vdd. In this case, the input end S of the SR flip-flop receives a high-level signal or “logic 1”. For a working status of the transistor T2 and a signal received by the input end R of the SR flip-flop, refer to descriptions of the transistor T1 and the signals received by the input end S of the SR flip-flop.
In a first time period, the transistor T1 is controlled to be turned on, and the transistor T2 is controlled to be turned off. In this case, the input end S of the SR flip-flop receives a low-level signal or “logic 0”, and the input end R of the SR flip-flop receives a high-level signal or “logic 1”. In this case, the Q end outputs the low-level signal or the “logic 0”. In a second time period, the transistor T1 is controlled to be turned off, and the transistor T2 is controlled to be turned on. In this case, the input end S of the SR flip-flop receives a high-level signal or “logic 1”, and the input end R of the SR flip-flop receives a low-level signal or “logic 0”. In this case, the Q end outputs the high-level signal or the “logic 1”. Therefore, based on a to-be-generated clock cycle, the transistor T1 and the transistor T2 are periodically controlled to be turned on or turned off, so that high-level and low-level signals, that is, clock pulse signals, can be alternatively output at the output end Q of the SR flip-flop.
It can be learned from
As shown in
In
In this implementation, the control signal generation circuit 20 may be a circuit whose output current and input voltage are in a proportional relationship. In the conventional technology, any circuit whose output current and input voltage can be in a proportional relationship may be used as the control signal generation circuit 20 shown in this application. This is not specifically limited herein. In specific implementation, the control signal generation circuit 20 may have a structure shown in
In some optional implementations, the clock signal generator 100 may further include a current mirror circuit 30.
In this optional implementation, a requirement on the control signal generation circuit 20 can be reduced by disposing the current mirror circuit 30. When a relatively large current needs to be input to the clock signal generation circuit 10, the current mirror circuit 30 may be used for implementation, so that stability of the control signal generation circuit 20 can be improved. This further improves stability of a clock signal generated by the clock signal generator 100.
With reference to
In
In the clock signal generator 100 shown in
Working principles of the selector M1 and the selector M2 are described by using the selector M1 as an example. When the control end k1 of the selector M1 receives “logic 0” or a low-level signal, the selector M1 gates the input end d1. In this case, an output end of a current mirror circuit 30 is connected to the control electrode of the transistor T1, the capacitor C1 is charged, and an electric potential at a node n1 gradually rises. When the electric potential at the node n1 rises to a specific value, so that a voltage between the control electrode and a second electrode of the transistor T1 is greater than a turn-on voltage, the transistor T1 is turned on. When the control end k1 of the selector M1 receives “logic 1” or a high-level signal, the selector M1 gates the input end d2. In this case, the control electrode of the transistor T1 is connected to the common ground Gnd, and the capacitor C1 is discharged. Under an action of the capacitor C1, the electric potential at the node n1 does not change abruptly. When the capacitor C1 is discharged to a specific value, and a voltage on the transistor T1 is less than the turn-on voltage, the transistor T1 is turned off. The working principle of the selector M2 is the same as that of the selector M1.
In
Specifically, a level signal at the output end QN of the SR flip-flop is the same as a level signal at an input end S of the SR flip-flop, and a level signal at the output end Q of the SR flip-flop is the same as a level signal at an input end R of the SR flip-flop.
It is assumed that at a current moment, the output end QN of the SR flip-flop outputs “logic 1” or a high-level signal, and the output end Q of the SR flip-flop outputs “logic 0” or a low-level signal. In a first time period, the capacitor C1 is charged, and the electric potential at the node n1 gradually rises. When the electric potential at the node n1 rises to a specific value, the transistor T1 is turned on, so that an electric potential at a node a1 is a low electric potential. The input end S of the SR flip-flop receives the low-level signal or the “logic 0”, and in this case, the output end QN of the SR flip-flop is triggered to flip. That is, the output end QN of the SR flip-flop outputs the “logic 0” or the low-level signal. The capacitor C2 is discharged, and an electric potential at a node n2 gradually decreases. When the electric potential at the node n2 decreases to a specific value, the transistor T2 is turned off, so that an electric potential at a node a2 is a high electric potential. The input end R of the SR flip-flop receives the high-level signal, and in this case, the output end Q of the SR flip-flop is triggered to flip. That is, the output end Q of the SR flip-flop outputs the “logic 1” or the high-level signal. Then, a second time period is entered.
In the second time period, the capacitor C1 is discharged, and the electric potential at the node n1 gradually decreases. When the electric potential at the node n1 decreases to a specific value, the transistor T1 is turned off, so that the electric potential at the node al is a high electric potential. The input end S of the SR flip-flop receives the high-level signal or the “logic 1”, and in this case, the output end QN of the SR flip-flop is triggered to flip. That is, the output end QN of the SR flip-flop outputs the “logic 1” or the high-level signal. The capacitor C2 is charged, and the electric potential at the node n2 gradually rises. When the electric potential at the node n2 rises to a specific value, the transistor T2 is turned on, so that the electric potential at the node a2 is a low electric potential. The input end R of the SR flip-flop receives the low-level signal, and in this case, the output end Q of the SR flip-flop is triggered to flip. That is, the output end Q of the SR flip-flop outputs the “logic 0” or the low-level signal.
In this way, through charging/discharging of the capacitor C1 and the capacitor C2, the output end Q and the output end QN of the SR flip-flop periodically output the high-level signal and the low-level signal. That is, the clock signals are formed at the output end Q and the output end QN separately. A clock frequency of the clock signal is a level signal flipping frequency of the output end Q or a level signal flipping frequency of the output end QN.
It can be learned from
A difference from the conventional technology shown in
In some implementations, the input end S of the SR flip-flop is coupled to a first electrode of the transistor T1 through an even quantity of phase inverters; and the input end R of the SR flip-flop is coupled to a first electrode of the transistor T2 through an even quantity of phase inverters.
In some implementations, parameters of the transistor T1, the transistor T2, and the transistor T3 are associated. Specifically, to provide a stable static operating point for the clock signal generator 100, the transistor T1 and the transistor T2 may have a same physical parameter and a same operating parameter, that is, have same parameters such as a conduction voltage drop, an internal resistance, and a power. There may be a proportional relationship between physical parameters and a proportional relationship between operating parameters of the transistor T1 and the transistor T3. In this way, a current flowing through the node a1 is the same as a current flowing through the node a2, and a current flowing through a node b and a current flowing through the node a1 have a proportional relationship. This reduces a temperature drift of the static operating point of the clock signal generator 100.
In the embodiments shown in
An embodiment of this application further provides an on-chip clock system. The on-chip clock system may include any clock signal generator described above and another structure such as a voltage source. The structure such as the voltage source is a conventional and common technology.
An embodiment of this application further provides a chip. The chip includes the foregoing on-chip clock system. The on-chip clock system including any clock signal generator described above may provide a stable clock signal for each module inside the chip, so that each module inside the chip works stably based on the clock signal.
Specifically, the chip may be an integrated circuit chip, and includes but is not limited to an artificial intelligence chip, a digital signal processing chip, an image processing chip, and the like. When the chip is configured to provide an image processing function and a data analysis and computing function for an application installed on a terminal device (for example, a mobile phone, a computer, or a wearable intelligent device), the electronic device may be a server device, and the chip may be disposed in the server device. When the chip is applied to the autonomous driving field to provide necessary vehicle-mounted computing, the electronic device may be a vehicle-mounted control device, and the AI chip may alternatively be disposed in the vehicle-mounted control device. Alternatively, the chip may be a 5G chip. In this case, the electronic device may be a base station hardware device that communicates with a terminal, and the AI chip may be disposed in the base station hardware device.
The foregoing describes embodiments of this application with reference to the accompanying drawings. However, this application is not limited to the foregoing specific implementations. The foregoing specific implementations are merely examples, but are not limitative. Inspired by this application, a person of ordinary skill in the art may further make many modifications without departing from the purposes of this application and the protection scope of the claims, and all the modifications shall fall within the protection scope of this application.
This application is a continuation of International Application No. PCT/CN2020/072413, filed on Jan. 16, 2020, the disclosure of which is hereby incorporated by reference in its entirety.
Number | Date | Country | |
---|---|---|---|
Parent | PCT/CN2020/072413 | Jan 2020 | US |
Child | 17865595 | US |