Charge pump and phase-locked loop

Abstract

The invention discloses a charge pump and a phase-locked loop, and belongs to the field of chip design. The charge pump provided by the invention comprises a main circuit and a reference circuit, the main circuit comprises a first switch tube, a first current source, a second current source and a second switch tube which are connected in sequence; wherein the potential of a first node between the first current source and the second current source is equal to that of a second node on the reference circuit; the charge pump further comprises a first target switch tube and a second target switch tube. One end of the first current source is connected with the first node, and the other end of the first current source is connected with the second node through the first target switch tube, so that the potentials at the two ends of the first current source are equal when the charging path of the main circuit is disconnected; and one end of the second current source is connected with the first node, and the other end of the second current source is connected with the second node through the second target switch tube, so that the potentials at the two ends of the second current source are equal when the discharge path of the main circuit is disconnected. The charge pump is used for reducing the leakage current of the charge pump.

Description

Technical Field

[0001] The present application belongs to the field of chip design, and specifically relates to a charge pump and a phase-locked loop. Background Art

[0002] During the operation of the chip, the phase-locked loop inside the chip is used to provide high-quality clocks to different modules of the chip. Figure 1 As shown in FIG. 1 , a phase-locked loop generally includes a phase frequency detector, a charge pump, a loop filter, a frequency divider, and a voltage-controlled oscillator. The phase frequency detector detects the phase difference between the reference clock and the divided clock output by the frequency divider, and outputs a voltage signal. The charge pump converts the voltage signal into a current signal, which is integrated by the loop filter and then output to the voltage-controlled oscillator to change the output frequency, thereby achieving loop locking.

[0003] In the related art, in a phase-locked loop, a charge pump includes multiple metal oxide semiconductor field effect transistors (MOSFET). As the accuracy requirements of the phase-locked loop become higher and higher, the size requirements of the MOSFET become smaller and smaller. Under advanced process technology, the MOSFET switch in the charge pump has a short channel effect, the gate oxide layer becomes thinner and thinner, the channel length is correspondingly shortened, the gate's control ability over the channel becomes weaker, and the MOSFET switch cannot be completely turned off in the off state, resulting in serious leakage problems in the charge pump. Summary of the invention

[0004] The embodiments of the present application provide a charge pump and a phase-locked loop, which can solve the problem of charge pump leakage due to the short channel effect of the switch in the related art.

[0005] In a first aspect, an embodiment of the present application provides a charge pump, including: a main circuit and a reference circuit;

[0006] The main circuit comprises a first switch tube, a first current source, a second current source and a second switch tube connected in sequence; wherein the potential of a first node between the first current source and the second current source is equal to the potential of a second node on the reference circuit;

[0007] The charge pump further includes a first target switch tube and a second target switch tube;

[0008] One end of the first current source is connected to the first node, and the other end of the first current source is connected to the second node through the first target switch tube, so that when the charging path of the main circuit is disconnected, the potentials at both ends of the first current source are equal;

[0009] One end of the second current source is connected to the first node, and the other end of the second current source is connected to the second node through the second target switch tube, so that when the discharge path of the main circuit is disconnected, the potentials at both ends of the second current source are equal.

[0010] In a second aspect, an embodiment of the present application provides a phase-locked loop, comprising a phase frequency detector, a loop filter, a voltage-controlled oscillator, a frequency divider, and a charge pump as described in the first aspect; wherein the phase frequency detector is coupled to the charge pump, the charge pump is coupled to the loop filter and the voltage-controlled oscillator, the voltage-controlled oscillator is coupled to the frequency divider, and the frequency divider is coupled to the phase frequency detector.

[0011] In an embodiment of the present application, a charge pump includes a main circuit and a reference circuit; the main circuit includes a first switch tube, a first current source, a second current source and a second switch tube connected in sequence; wherein the potential of a first node between the first current source and the second current source is equal to the potential of a second node on the reference circuit; the charge pump also includes a first target switch tube and a second target switch tube; one end of the first current source is connected to the first node, and the other end of the first current source is connected to the second node through the first target switch tube, so that when the charging path of the main circuit is disconnected, the potentials at both ends of the first current source are equal; one end of the second current source is connected to the first node, and the other end of the second current source is connected to the second node through the second target switch tube, so that when the discharge path of the main circuit is disconnected, the potentials at both ends of the second current source are equal. In this way, a second node with the same potential as the first node is determined in the reference circuit of the charge pump, and one end of the first current source is connected to the second node through the first target switch tube. When the charging path of the main circuit is disconnected, the potentials of the two ends of the first current source can be equal, thereby avoiding the influence of the leakage current generated by the first switch tube being unable to completely shut down due to the short channel effect on the output of the first node through the first current source. When the discharge path of the main circuit is disconnected, one end of the second current source is connected to the second node through the second target switch tube. When the discharge path of the main circuit is disconnected, the potentials of the two ends of the second current source can be equal, thereby avoiding the influence of the leakage current generated by the second switch tube being unable to completely shut down due to the short channel effect on the output of the first node through the second current source. In this way, the problem of charge pump leakage caused by the short channel effect of the switch is suppressed to a certain extent. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 A schematic structural diagram of a phase-locked loop provided in an embodiment of the present application;

[0013] Figure 2 is a schematic structural diagram of a charge pump based on a source switch structure in the related art;

[0014] Figure 3 A schematic structural diagram of a charge pump provided in an embodiment of the present application;

[0015] Figure 4 A schematic structural diagram of another charge pump provided in an embodiment of the present application;

[0016] Figure 5 A schematic structural diagram of another charge pump provided in an embodiment of the present application;

[0017] Figure 6 A schematic structural diagram of another charge pump provided in an embodiment of the present application;

[0018] Figure 7 A schematic structural diagram of another charge pump provided in an embodiment of the present application.

[0019] Description of reference numerals:

[0020] NM0-N channel MOSFET; NM1-N channel MOSFET; NM2-N channel MOSFET; NM3-N channel MOSFET; PM0-P channel MOSFET; PM1-P channel MOSFET; PM2-P channel MOSFET; UP_N-switching signal generated by the phase frequency detector; UP_P-switching signal generated by the phase frequency detector; DN_N-switching signal generated by the phase frequency detector; DN_P-switching signal generated by the phase frequency detector; vref-reference voltage; pb-bias voltage; Iref-reference current; 100-charge pump; 110-main circuit; 110A-first Switching tube; 110B-first current source; 110C-second current source; 110D-second switching tube; Iout-first node; 120-reference circuit; 120A-third switching tube; 120B-first transistor; 120C-second transistor; 120D-fourth switching tube; Iout_dum-second node; 130-first target switching tube; 140-second target switching tube; 150-third target switching tube; 160-fourth target switching tube; 170A-first bias tube; 170B-second bias tube; 170C-third bias tube; 170D-fourth bias tube; 170E-fifth bias tube; C-capacitor. DETAILED DESCRIPTION

[0021] The following will be combined with the drawings in the embodiments of the present application to clearly describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments in the present application belong to the scope of protection of this application.

[0022] The terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described here, and the objects distinguished by "first", "second", etc. are generally of one type, and the number of objects is not limited. For example, the first object can be one or more. In addition, "and / or" in the specification and claims represents at least one of the connected objects, and the character " / " generally indicates that the objects associated with each other are in an "or" relationship.

[0023] like Figure 1 As shown in Figure 1, the phase-locked loop is a negative feedback loop. The phase-locked loop generally includes a frequency detector, a charge pump, a loop filter, a frequency divider, and a voltage-controlled oscillator. The frequency detector detects the reference clock f ref The divided clock f div The phase difference between them is used to output voltage signals (such as UP signal and DN signal). The charge pump converts the voltage signal into a current signal, which is then integrated by the loop filter and output to the voltage-controlled oscillator to change the output frequency f vco , thus achieving loop locking.

[0024] In a phase-locked loop, in order to increase the switching speed of the charge pump and reduce the channel charge injection caused by the switch, the relevant technology generally uses a charge pump with a source switch structure. Figure 2 As shown, Figure 2 A charge pump based on a source switch structure is shown, including a main circuit composed of four MOS tubes (ie PM2, PM1, NM2 and NM3) and a bias circuit composed of three MOS tubes (ie NM0, NM1 and PM0).

[0025] Among them, PM0, PM1 and PM2 are P-channel MOSFET devices, and NM0, NM1, NM2 and NM3 are N-channel MOSFET devices. In the charge pump, PM1 and NM2 can be used as current sources, PM1 is controlled by switch PM2, NM2 is controlled by switch NM3, and the bias circuit can be used to provide stable current for PM1 and NM2. The UP_N signal and the DN_P signal are switching signals generated by the frequency detector and phase detector. The Iout node between PM1 and NM2 is the output of the charge pump, and the capacitor C is part of the loop filter.

[0026] like Figure 2As shown, under ideal conditions, when UP_N is low, PM2 is turned on, and the power supply charges the capacitor C through the path from PM2 to PM1; when UP_N is high, PM2 is turned off, and PM2 can be equivalent to a very large resistor, and the power supply cannot generate a charging current from PM2 to PM1 to charge C; when DN_P is high, NM3 is turned on, and the capacitor C will discharge through the path of NM2 and NM3; when DN_P is low, NM3 is turned off and can also be equivalent to a very large resistor, and the capacitor C cannot discharge through the path of NM2 and NM3.

[0027] but, Figure 2 The typical charge pump structure shown is difficult to meet the needs of high-precision phase-locked loops. As the accuracy requirements of phase-locked loops become higher and higher, the size requirements of MOSFETs are getting smaller and smaller. Under advanced process technology, the MOSFET switch in the charge pump has a short channel effect, the gate oxide layer becomes thinner and thinner, the channel length is correspondingly shortened, the gate's control ability over the channel becomes weaker, and the MOSFET cannot be completely turned off in the off state, resulting in serious leakage problems in the charge pump.

[0028] For example, in Figure 2 In the backward process, when UP_N is high and DN_P is low, PM2 and NM3 can be close to being disconnected and no leakage current will be generated. However, in advanced processes, such as FinFET (Finned Field Effect Transistor) below 22nm, when UP_N is high and DN_P is low, the leakage current caused by the incomplete shutdown of PM2 and NM3 has a non-negligible impact on the circuit, resulting in serious leakage problems in the charge pump.

[0029] Based on this, in order to solve the problem of charge pump leakage due to the short channel effect of the switch in the related art, the embodiment of the present application provides a charge pump, which can suppress the leakage problem of the charge pump under advanced technology to a certain extent. The charge pump circuit structure provided by the embodiment of the present application is simple, and by adding a switch (such as the first target switch tube 130 and the second target switch tube 140 below), the voltages at both ends of the MOSFET tube that should be disconnected are consistent, so that there is no voltage difference at both ends of the MOSFET tube, which reduces the influence of the leakage current caused by the short channel effect of the switch on the charge pump to a certain extent.

[0030] The charge pump and phase-locked loop provided in the embodiments of the present application are described in detail below through specific embodiments and their application scenarios in conjunction with the accompanying drawings.

[0031] Figure 3 A schematic structural diagram of a charge pump provided in an embodiment of the present application.

[0032] like Figure 3 As shown, the charge pump 100 provided in the embodiment of the present application may include: a main circuit 110 and a reference circuit 120;

[0033] The main circuit 110 may include a first switch tube 110A, a first current source 110B, a second current source 110C and a second switch tube 110D connected in sequence;

[0034] The potential of the first node Iout between the first current source 110B and the second current source 110C is equal to the potential of the second node Iout_dum on the reference circuit 120;

[0035] The charge pump 100 further includes a first target switch tube 130 and a second target switch tube 140;

[0036] One end of the first current source 110B is connected to the first node Iout, and the other end of the first current source 110B is connected to the second node Iout_dum through the first target switch tube 130, so that when the charging path of the main circuit 110 is disconnected, the potentials at both ends of the first current source 110B are equal;

[0037] One end of the second current source 110C is connected to the first node Iout, and the other end of the second current source 110C is connected to the second node Iout_dum through the second target switch tube 140, so that when the discharge path of the main circuit 110 is disconnected, the potentials at both ends of the second current source 110C are equal.

[0038] In the embodiment of the present application, the reference circuit 120 can be a differential path of the main circuit 110, which is used to suppress non-ideal characteristics in the charge pump 100. For example, there are actually non-ideal factors such as clock feedthrough and charge sharing in the charge pump 100, which cause the net charge to flow into the capacitor C of the loop filter through the first node Iout, resulting in voltage ripples in the voltage-controlled oscillator input voltage, which are finally modulated by the voltage-controlled oscillator to generate reference spurious. The reference circuit 120 and the main circuit 110 form two differential paths, which can effectively suppress the reference spurious generated by non-ideal effects.

[0039] For example, Figure 3 As shown, in a specific embodiment, the reference circuit 120 may include a third switch tube 120A, a first transistor 120B, a second transistor 120C and a fourth switch tube 120D connected in sequence; the second node Iout_dum is located between the first transistor 120B and the second transistor 120C.

[0040] The third switch tube 120A corresponds to the first switch tube 110A, the first transistor 120B corresponds to the first current source 110B, the second transistor 120C corresponds to the second current source 110C, and the fourth switch tube 120D corresponds to the second switch tube 110D.

[0041] The working state of the third switch tube 120A is opposite to that of the first switch tube 110A, and the working state of the fourth switch tube 120D is opposite to that of the second switch tube 110D.

[0042] There is a second node Iout_dum on the reference circuit 120 , and the potential of the second node Iout_dum may be equal to the potential of the first node Iout on the main circuit 110 . The first node Iout is the output end of the charge pump, and the capacitor C is a part of the loop filter.

[0043] The first switch tube 110A and the first current source 110B can be understood as the charging path of the main circuit. In the charging path of the main circuit 110, the on / off state of the first current source 110B is controlled by the first switch tube 110A, and the on / off state of the first switch tube 110A is controlled by the UP_P signal.

[0044] The second current source 110C and the second switch tube 110D can be understood as the discharge path of the main circuit. In the discharge path of the main circuit 110, the on / off state of the second current source 110C is controlled by the second switch tube 110D, and the on / off state of the second switch tube 110D is controlled by the DN_P signal.

[0045] In the reference circuit 120, the on / off state of the first transistor 120B is controlled by the third switch tube 120A, and the on / off state of the third switch tube 120A is controlled by the UP_N signal. The on / off state of the second transistor 120C is controlled by the fourth switch tube 120D, and the on / off state of the fourth switch tube 120D is controlled by the DN_N signal.

[0046] Among them, the UP_N signal and the UP_P signal are switch signals with opposite phases, and the DN_P signal and the DN_N signal are switch signals with opposite phases. Both sets of signals are generated by a frequency detector and a phase detector.

[0047] It should be pointed out that the charge pump has four working states: Figure 3 The charging state shown (the charging refers to the charging of the first node Iout of the charge pump), as shown in Figure 4 The discharge state shown in FIG. 1 (the discharge refers to the discharge of the first node Iout of the charge pump), as shown in FIG. Figure 5 The fully open state shown and Figure 6 Shown in fully closed state.

[0048] The reference clock at the input of the phase-locked loop is f ref The phase of the feedback frequency division clock f div The charge pump is in phase Figure 3 In the charging state shown in FIG. 1 , the first switch tube 110A is turned on, the second switch tube 110D is turned off, and the charge pump outputs a charging current through a charging path (i.e., a path where the first switch tube 110A, the first current source 110B and the first node Iout are located) to charge the capacitor C of the loop filter in the phase-locked loop, resulting in an increase in the control voltage of the voltage-controlled oscillator and an increase in the frequency f output by the voltage-controlled oscillator. vco Increase.

[0049] The reference clock at the input of the phase-locked loop is f ref The phase lags behind the divided feedback clock f div The charge pump is in phase Figure 4 In the discharge state shown in FIG. 1 , the first switch tube 110A is turned off, the second switch tube 110D is turned on, and the charge pump outputs a discharge current through a discharge path (i.e., a path where the first node Iout, the second current source 110D, and the second switch tube 110C are located) to discharge the capacitor C of the loop filter in the phase-locked loop, resulting in a decrease in the control voltage of the voltage-controlled oscillator and a decrease in the frequency f output by the voltage-controlled oscillator. vco Reduce.

[0050] Among them, the frequency detector can accurately identify the phase difference between the feedback divided clock and the reference clock. After the negative feedback of the phase-locked loop itself, the phase difference is gradually reduced. When the phase-locked loop is finally locked, the phase difference is zero. When the phase-locked loop enters the locked state, ideally, no net charge flows into or out of the capacitor C of the loop filter through the first node. When the phase-locked loop is in the locked state, the charge pump has two states: fully open state and fully closed state.

[0051] like Figure 5 As shown, when the charge pump is in the fully-on state, the first switch tube 110A and the second switch tube 110D are both turned on, and the charging current and the discharging current match.

[0052] like Figure 6 As shown, when the charge pump is in the fully-off state, the first switch tube 110A and the second switch tube 110D are both disconnected, and there is no charging current or discharging current.

[0053] In practical applications, since the frequency detector and phase detector generate narrow pulses, when the phase-locked loop is in a locked state, the charge pump is in a fully closed state most of the time, and is only in a fully open state when a narrow pulse is input into the charge pump.

[0054] Under ideal conditions, these working states of the charge pump will not have a serious impact on the first node Iout. However, in practical applications, under advanced processes, MOSFET switches such as the first switch tube 110A and the second switch tube 110D have a short channel effect, and cannot be completely turned off in the cut-off state, resulting in leakage, causing the net charge to flow into the capacitor C of the loop filter through the first node Iout, or the capacitor C of the loop filter discharges the net charge to the ground through the first node Iout, resulting in voltage ripples in the voltage-controlled oscillator input voltage, thereby increasing the reference spurious. Therefore, it is necessary to suppress the short channel effect of the MOSFET, which cannot be completely turned off and causes leakage current to affect the output of the first node Iout.

[0055] Based on this, in an embodiment of the present application, a first target switch tube 130 and a second target switch tube 140 are set in a charge pump based on a differential circuit, and a second node with the same potential as the first node is determined in a reference circuit of the charge pump. One end of the first current source 110B is connected to the second node through the first target switch tube 130. When the charging path of the main circuit is disconnected, the potentials of both ends of the first current source 110B can be made equal, thereby avoiding the leakage current caused by the first switch tube 110A being unable to be completely turned off due to the short channel effect and affecting the output of the first node Iout through the first current source 110B.

[0056] Similarly, by connecting one end of the second current source 110C to the second node through the second target switch tube 140, the potentials at both ends of the second current source 110C can be made equal when the discharge path of the main circuit is disconnected, thereby avoiding the leakage current generated by the second switch tube 110D being unable to be completely turned off due to the short channel effect, which affects the output of the first node Iout through the second current source 110C.

[0057] According to the charge pump provided in the embodiment of the present application, it includes a main circuit and a reference circuit; the main circuit includes a first switch tube, a first current source, a second current source and a second switch tube connected in sequence; wherein the potential of the first node between the first current source and the second current source is equal to the potential of the second node on the reference circuit; the charge pump also includes a first target switch tube and a second target switch tube; one end of the first current source is connected to the first node, and the other end of the first current source is connected to the second node through the first target switch tube; one end of the second current source is connected to the first node, and the other end of the second current source is connected to the second node through the second target switch tube. In this way, a second node with the same potential as the first node is determined in the reference circuit of the charge pump, and one end of the first current source is connected to the second node through the first target switch tube. When the charging path of the main circuit is disconnected, the potentials of the two ends of the first current source can be equal, thereby avoiding the leakage current caused by the first switch tube being unable to be completely turned off due to the short channel effect, which affects the output of the first node through the first current source. When one end of the second current source is connected to the second node through the second target switch tube, the potentials of the two ends of the second current source can be equal when the discharge path of the main circuit is disconnected, thereby avoiding the leakage current caused by the second switch tube being unable to be completely turned off due to the short channel effect, which affects the output of the first node through the second current source. In this way, the problem of charge pump leakage caused by the short channel effect of the switch is suppressed to a certain extent.

[0058] In a specific embodiment, since the charge pump has the above-mentioned multiple working states, correspondingly, the first target switch tube and the first target switch tube also have different working states, so that the charge pump can suppress the leakage problem caused by the short channel effect of the switch in various working states.

[0059] Among them, the charging path of the main circuit 110 may include a first switch tube 110A and a first current source 110B; the discharging path of the main circuit 110 may include a second current source 110C and a second switch tube 110D; when the charging path of the main circuit is disconnected, the first target switch tube 130 is in a conducting state to make the potentials at both ends of the first current source 110B equal; and / or, when the discharging path of the main circuit is disconnected, the second target switch tube 140 is in a conducting state to make the potentials at both ends of the second current source 110C equal. In this way, the leakage current generated by the first switch tube 110A due to the short channel effect and unable to be completely turned off and flowing through the first current source 110B can be avoided to affect the output of the first node Iout, and / or, the leakage current generated by the second switch tube 110D flowing through the second current source 110C can be avoided to affect the output of the first node Iout. The following examples are given according to different situations:

[0060] The first case: when the first switch tube 110A is in the on state, the charging path of the control main circuit 110 is turned on, and the second switch tube 110D is in the off state, the discharge path of the control main circuit is turned off, and the charge pump 100 is in the charging state;

[0061] When the charge pump 100 is in the charging state, the first target switch tube 130 is in the disconnected state, and the second target switch tube 140 is in the on state, so that the potentials at both ends of the second current source 110C are equal, so as to avoid the leakage current generated by the second switch tube 110D being unable to be completely turned off due to the short channel effect, and affecting the output of the first node Iout through the second current source 110C.

[0062] like Figure 3 As shown, the charge pump 100 is in a charging state. Ideally, when the UP_P signal is at a high level, the first switch tube 110A is turned on, and the power source charges the capacitor C through the charging path from the first switch tube 110A to the first current source 110B.

[0063] like Figure 3 As shown, in an ideal case, when the DN_P signal is at a low level, the second switch tube 110D is turned off. At this time, the second switch tube 110D can be equivalent to a very large resistor, and the capacitor C cannot be discharged through the discharge path of the second current source 110C and the second switch tube 110D.

[0064] However, in practical applications, under advanced processes, the second switch tube 110D has a short channel effect and cannot be completely turned off in the off state, resulting in leakage in the discharge path.

[0065] In the embodiment of the present application, the on / off state of the second switch tube 110D is controlled by the DN_P signal, and the on / off state of the second target switch tube 140 is controlled by the DN_N signal. The DN_P signal and the DN_N signal are switch signals with opposite phases to each other, and the working state of the second switch tube 110D is opposite to that of the second target switch tube 140.

[0066] When the second switch tube 110D is disconnected, making the second current source 110C in the off state, the second target switch tube 140 is turned on, making the other end of the second current source 110C connected to the second node. Because the potentials of the first node Iout and the second node Iout_dum are equal, the potentials of both ends of the second current source 110C are equal. In this case, even if the second switch tube 110D cannot be completely turned off due to the short channel effect when it is disconnected and leakage occurs, since the potentials of both ends of the second current source 110C are equal, the capacitor C will not generate a discharge path from the discharge path of the second current source 110C and the second switch tube 110D to the ground, thereby avoiding the leakage current from affecting the capacitor C.

[0067] In this way, when the second switch tube 110D in the charge pump cannot be completely turned off in the cut-off state due to the short channel effect, the second target switch tube 140 is set to level the potentials at both ends of the second current source 110C, making it difficult for the leakage current to affect the capacitor C through the second current source 110C, thereby avoiding the problem of charge pump leakage due to the short channel effect of the second switch tube 110D.

[0068] The second situation: when the first switch tube 110A is in the off state, the charging path of the control main circuit 110 is disconnected, and the second switch tube 110D is in the on state, the discharging path of the control main circuit 110 is connected, the charge pump 100 is in the discharging state;

[0069] When the charge pump 100 is in a discharging state, the first target switch tube 130 is in an on state, and the second target switch tube 140 is in an off state, so that the potentials at both ends of the first current source 110B are equal. At this time, the leakage current generated by the first switch tube 110A being unable to be completely turned off due to the short channel effect can be avoided from affecting the output of the first node Iout through the first current source 110B.

[0070] like Figure 4 As shown, the charge pump 100 is in a discharging state. Ideally, when the DN_P signal is at a high level, the second switch tube 110D is turned on, and the capacitor C is discharged through the second current source 110C and the discharge path of the second switch tube 110D.

[0071] like Figure 4 As shown, in an ideal case, when the UP_P signal is at a low level, the first switch tube 110A is turned off. At this time, the first switch tube 110A can be equivalent to a very large resistor, and the power supply cannot generate a charging current to charge C through the charging path from the first switch tube 110A to the first current source 110B.

[0072] However, in actual applications, under advanced processes, the first switch tube 110A has a short channel effect due to its small size, and cannot be completely turned off in the cut-off state, resulting in leakage in the charging path.

[0073] In the embodiment of the present application, the on / off state of the first switch tube 110A is controlled by the UP_P signal, and the on / off state of the first target switch tube 130 is controlled by the UP_N signal. Since the UP_N signal and the UP_P signal are mutually inverted switch signals, the working states of the first switch tube 110A and the first target switch tube 130 are opposite.

[0074] When the first switch tube 110A is disconnected, making the first current source 110B in an off state, the first target switch tube 130 is turned on, making the other end of the first current source 110B connected to the second node. Because the potentials of the first node Iout and the second node Iout_dum are equal, the potentials of both ends of the first current source 110B are equal. In this case, even if the first switch tube 110A cannot be completely shut down due to the short channel effect when it is disconnected and leakage occurs, since the potentials of both ends of the first current source 110B are equal, the power supply will not inject current into the capacitor C through the first switch tube 110A and the first current source 110B, thereby avoiding the leakage current from affecting the capacitor C.

[0075] In this way, when the first switch tube 110A in the charge pump cannot be completely turned off in the cut-off state due to the short channel effect, the first target switch tube 130 is set to level the potentials at both ends of the first current source 110B, making it difficult for the leakage current to affect the capacitor C through the first current source 110B, thereby avoiding the problem of charge pump leakage due to the short channel effect of the first switch tube 110A.

[0076] The third situation: when the first switch tube 110A is in the on state, controlling the charging path of the main circuit 110 to be turned on, and the second switch tube 110D is in the on state, controlling the discharging path of the main circuit 110 to be turned on, the charge pump 100 is in the fully open state.

[0077] like Figure 5 As shown, the charge pump 100 is in a fully on state. When the charge pump is in a fully on state, the charging path of the main circuit 110 is turned on, and the discharging path of the main circuit 110 is turned on, and the charging current output by the first current source 110B matches the discharging current output by the second current source 110C.

[0078] At this time, since the first switch tube 110A and the second switch tube 110D are in the on state, there is no leakage current due to the short channel effect in the off state. As for the leakage current caused by other non-ideal factors, it is much smaller than the charging current and the discharging current, and its influence can also be ignored.

[0079] The fourth situation: when the first switch tube 110A is in the off state, the charging path of the control main circuit 110 is disconnected, and the second switch tube 110D is in the off state, the discharging path of the control main circuit 110 is disconnected, the charge pump 100 is in the fully off state.

[0080] like Figure 6As shown, the charge pump 100 is in a fully open state. When the charge pump 100 is in a fully closed state, the first target switch tube 130 and the second target switch tube 140 are both in a conducting state, so that the potentials at both ends of the first current source 110B are equal and the potentials at both ends of the second current source 110C are equal, which can prevent the leakage current generated by the first switch tube 110A being unable to be completely turned off due to the short channel effect from affecting the output of the first node through the first current source 110B, and can also prevent the leakage current generated by the second switch tube 110D being unable to be completely turned off due to the short channel effect from affecting the output of the first node through the second current source 110C.

[0081] In addition, since the reference circuit 120 is a differential path of the main circuit 110, based on a similar concept to the first target switch tube 130 and the second target switch tube 140, the embodiment of the present application may also set a third target switch tube 150 and a fourth target switch tube 160 on the reference circuit 120 to eliminate reference strays caused by non-ideal factors.

[0082] In a specific embodiment, Figure 7 As shown, the reference circuit 120 may include a third switch tube 120A, a first transistor 120B, a second transistor 120C and a fourth switch tube 120D connected in sequence; the second node Iout_dum is located between the first transistor 120B and the second transistor 120C;

[0083] The charge pump 100 may further include a third target switch tube 150 and a fourth target switch tube 160;

[0084] One end of the first transistor 120B is connected to the second node Iout_dum, and the other end of the first transistor 120B is connected to the second node Iout_dum through the third target switch tube 150;

[0085] One end of the second transistor 120C is connected to the second node Iout_dum, and the other end of the second transistor 120C is connected to the second node Iout_dum through the fourth target switch tube 160 .

[0086] In this way, the present application can suppress non-ideal factors such as clock feedthrough, charge sharing, etc. that actually exist in the charge pump 100 through the third target switch tube 150 and the fourth target switch tube 160 .

[0087] Among them, the on / off state of the third switch tube 120A is controlled by the UP_N signal, and the on / off state of the third target switch tube 150 is controlled by the UP_P signal. The UP_N signal and the UP_P signal are mutually inverted switch signals, and the working states of the third switch tube 120A and the third target switch tube 150 are opposite.

[0088] Among them, the on / off state of the fourth switch tube 120D is controlled by the DN_N signal, and the on / off state of the fourth target switch tube 160 is controlled by the DN_P signal. The DN_N signal and the DN_P signal are switch signals with opposite phases to each other. The working states of the fourth switch tube 120D and the fourth target switch tube 160 are opposite.

[0089] For example, Figure 3 As shown, when the charge pump 100 is in the charging state, the third target switch tube 150 is in the on state to make the potentials at both ends of the first transistor 120B equal, thereby suppressing the leakage current through the first transistor 120B, and suppressing the reference spurious to a certain extent.

[0090] like Figure 4 As shown, when the charge pump 100 is in the discharging state, the fourth target switch tube 160 is in the on state, so that the potentials at both ends of the second transistor 120C are equal, thereby suppressing the leakage current through the second transistor 120C, and suppressing the reference spurious to a certain extent.

[0091] like Figure 5 As shown, when the charge pump 100 is in the fully open state, the third target switch tube 150 and the fourth target switch tube 160 are both in the on state, so that the potentials at both ends of the first transistor 120B are equal and the potentials at both ends of the second transistor 120C are equal, thereby suppressing the leakage current through the first transistor 120B or the second transistor 120C, and suppressing the reference stray to a certain extent.

[0092] In addition, it is mentioned above that the potential of the first node Iout is equal to the potential of the second node Iout_dum. The present application can adopt a variety of implementation methods to ensure that the potential of the first node Iout is equal to the potential of the second node Iout_dum.

[0093] In a specific embodiment, Figure 7 As shown, the charge pump 100 further includes a bias circuit, which may include a first bias tube 170A and a second bias tube 170B;

[0094] The drain and gate of the first bias tube 170A are coupled to the first node Iout, and the source of the first bias tube 170A is grounded;

[0095] The drain and gate of the second bias tube 170B are coupled to the second node Iout_dum, and the source of the second bias tube 170B is grounded;

[0096] The operating current I1 of the first bias tube 170A is equal to the operating current I2 of the second bias tube 170B, and the size of the first bias tube 170A is the same as the size of the second bias tube 170B, so that the potential of the first node Iout is equal to the potential of the second node Iout_dum.

[0097] It can be understood that, through the diode connection mode of the first bias tube 170A and the second bias tube 170B, it can be ensured that the potential of the first node Iout is equal to the potential of the second node Iout_dum.

[0098] The operating current I1 of the first bias tube 170A and the operating current I2 of the second bias tube 170B may be the same current provided inside the phase-locked loop or two equal but different currents, which is not specifically limited in the present application.

[0099] When the charge pump is in a charging state, a discharging state, a fully open state or a fully closed state, the operating current I1 of the first bias tube 170A and the operating current I2 of the second bias tube 170B may be equal, and this application does not impose any specific restrictions.

[0100] In this way, two bias tubes with the same size and the same working current are connected to the first node Iout and the second node Iout_dum in a diode connection manner, respectively, so that the potential of the first node Iout can be equal to the potential of the second node Iout_dum.

[0101] Alternatively, in another specific embodiment, the charge pump may further include a unit gain buffer (not shown), wherein a first input terminal of the unit gain buffer is coupled to the first node, and a second input terminal and an output terminal of the unit gain buffer are coupled to the second node, so that the potential of the first node is equal to the potential of the second node. The unit gain buffer may be an operational amplifier with an amplification factor of 1, so that the potential of the first node is equal to the potential of the second node.

[0102] Of course, the present application can also set the potential of the first node on the main circuit to be equal to the potential of the second node on the reference circuit in other ways, which will not be repeated here.

[0103] In addition, if Figure 7 As shown, in the charge pump 100, the bias circuit may further include a third bias tube 170C, a fourth bias tube 170D, and a fifth bias tube 170E. The third bias tube 170C may provide a bias voltage pb for the gate of the first current source 110B, and provide a bias voltage pb for the gate of the first transistor 120B. The fourth bias tube 170D and the fifth bias tube 170E may provide a reference voltage vref for the gate of the second current source 110C, and provide a reference voltage vref for the gate of the second transistor 120C.

[0104] In the embodiment of the present application, the first current source 110B, the first transistor 120B and the third bias tube 170C can be P-channel MOSFET devices; the first switch tube 110A, the first target switch tube 130, the third switch tube 120A and the third target switch tube 150 can be P-channel MOSFET devices or CMOS transmission gate devices, and the present application does not make specific restrictions.

[0105] In the embodiment of the present application, the second current source 110C, the second transistor 120C, the first bias tube 170A, the second bias tube 170B, the fourth bias tube 170D, and the fifth bias tube 170E can be N-channel MOSFET devices; the second switch tube 110D, the second target switch tube 140, the fourth switch tube 120D and the fourth target switch tube 160 can be N-channel MOSFET devices or CMOS transmission gate devices, and the present application does not make any specific restrictions.

[0106] It should be noted that the charge pump provided in the embodiment of the present application may be a charge pump with a source switch structure, thereby reducing the leakage current of the charge pump under advanced processes without affecting the switching speed of the charge pump.

[0107] For example, Figure 7 As shown, the first current source 110B may include a first P-channel MOSFET, and the second current source 110C includes a first N-channel MOSFET. The source of the first P-channel MOSFET is coupled to the first switch tube 110A, and the source of the first N-channel MOSFET is coupled to the second switch tube 110D; the drain of the first P-channel MOSFET is coupled to the drain of the first N-channel MOSFET.

[0108] In this way, the first switch tube 110A is coupled with the source of the first current source 110B, and the second switch tube 110D is coupled with the source of the second current source 110C, forming a structure based on a source switch. Compared with the charge pump based on a drain switch structure in the related art, the embodiment of the present application uses a charge pump with a source switch structure. The drain and gate of the current source (i.e., the first current source 110B and the second current source 110C) always maintain a normal bias, and the source impedance is low, so the switching speed is fast, and it can be applied to a high-bandwidth phase-locked loop. It can be seen that the charge pump provided in the embodiment of the present application uses a source switch structure, which has a low leakage characteristic while retaining the fast switching characteristics of the source switch structure.

[0109] Moreover, compared with the traditional source charge pump, since the power supply voltage is often smaller in advanced processes, this structure has only 4 layers of MOS devices from the power supply to the ground, which is suitable for a power supply voltage of 0.8 volts. In extreme cases, the power supply voltage can be reduced to 0.5 volts, which is suitable for advanced processes.

[0110] In addition, based on the same concept as the charge pump provided in any of the above embodiments, the present application embodiment also provides a phase-locked loop. Figure 1 As shown, the phase-locked loop provided in the embodiment of the present application may include: a phase frequency detector, a loop filter, a voltage-controlled oscillator, a frequency divider and a charge pump; wherein the phase frequency detector is coupled to the charge pump, the charge pump is coupled to the loop filter and the voltage-controlled oscillator, the voltage-controlled oscillator is coupled to the frequency divider, and the frequency divider is coupled to the phase frequency detector.

[0111] Among them, the charge pump in the phase-locked loop can be the charge pump provided by any of the above embodiments. Furthermore, the phase-locked loop provided by the embodiment of the present application can realize the function of the charge pump provided by any of the above embodiments and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0112] It should be noted that, in this article, the terms "comprise", "include" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise one..." do not exclude the presence of other identical elements in the process, method, article or device including the element. In addition, it should be pointed out that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved, for example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0113] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a computer software product, which is stored in a storage medium (such as ROM / RAM, a disk, or an optical disk), and includes a number of instructions for a terminal (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0114] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the present application and the scope of protection of the claims, all of which are within the protection of the present application.

Claims

1. A charge pump, applied to a phase-locked loop, characterized in that: The charge pump comprises: a main circuit and a reference circuit; The main circuit comprises a first switch tube, a first current source, a second current source and a second switch tube connected in sequence; wherein the potential of a first node between the first current source and the second current source is equal to the potential of a second node on the reference circuit; The charge pump further includes a first target switch tube and a second target switch tube; One end of the first current source is connected to the first node, and the other end of the first current source is connected to the second node through the first target switch tube, so that when the charging path of the main circuit is disconnected, the potentials at both ends of the first current source are equal; One end of the second current source is connected to the first node, and the other end of the second current source is connected to the second node through the second target switch tube, so that when the discharge path of the main circuit is disconnected, the potentials at both ends of the second current source are equal.

2. The charge pump according to claim 1, characterized in that The first node is the output end of the charge pump; The charging path of the main circuit includes the first switch tube and the first current source; the discharging path of the main circuit includes the second current source and the second switch tube; When the charging path of the main circuit is disconnected, the first target switch tube is in a conducting state so that the potentials at both ends of the first current source are equal; And / or, when the discharge path of the main circuit is disconnected, the second target switch tube is in an on state, so that the potentials at both ends of the second current source are equal.

3. The charge pump according to claim 2, characterized in that: When the first switch tube controls the charging path of the main circuit to be turned on, and the second switch tube controls the discharging path of the main circuit to be turned off, the charge pump is in a charging state; When the charge pump is in a charging state, the first target switch tube is in a disconnected state, and the second target switch tube is in a conductive state, so that the potentials at both ends of the second current source are equal.

4. The charge pump according to claim 2, characterized in that When the first switch tube controls the charging path of the main circuit to be disconnected, and the second switch tube controls the discharging path of the main circuit to be turned on, the charge pump is in a discharging state; When the charge pump is in a discharging state, the first target switch tube is in an on state, and the second target switch tube is in an off state, so that the potentials at both ends of the first current source are equal.

5. The charge pump according to claim 2, characterized in that: When the first switch tube controls the charging path of the main circuit to be disconnected, and the second switch tube controls the discharging path of the main circuit to be disconnected, the charge pump is in a fully-off state; When the charge pump is in a fully-off state, the first target switch tube and the second target switch tube are both in an on state, so that the potentials at both ends of the first current source are equal and the potentials at both ends of the second current source are equal.

6. The charge pump according to any one of claims 1 to 5, characterized in that: The charge pump further comprises a bias circuit, wherein the bias circuit comprises a first bias tube and a second bias tube; The drain and the gate of the first bias tube are both coupled to the first node, and the source of the first bias tube is grounded; The drain and the gate of the second bias tube are both coupled to the second node, and the source of the second bias tube is grounded; The operating current of the first bias tube is equal to the operating current of the second bias tube, and the size of the first bias tube is the same as the size of the second bias tube, so that the potential of the first node is equal to the potential of the second node.

7. The charge pump according to any one of claims 1 to 5, characterized in that: The charge pump further includes a unit gain buffer, a first input terminal of the unit gain buffer is coupled to the first node, and a second input terminal and an output terminal of the unit gain buffer are both coupled to the second node, so that the potential of the first node is equal to the potential of the second node.

8. The charge pump according to any one of claims 1 to 5, characterized in that: The first current source includes a first P-channel MOSFET, and the second current source includes a first N-channel MOSFET. The source of the first P-channel MOSFET is coupled to the first switch tube, and the source of the first N-channel MOSFET is coupled to the second switch tube; the drain of the first P-channel MOSFET is coupled to the drain of the first N-channel MOSFET.

9. The charge pump according to any one of claims 1 to 5, characterized in that: The reference circuit comprises a third switch tube, a first transistor, a second transistor and a fourth switch tube connected in sequence; the second node is located between the first transistor and the second transistor; The charge pump further includes a third target switch tube and a fourth target switch tube; One end of the first transistor is connected to the second node, and the other end of the first transistor is connected to the second node through the third target switch tube; One end of the second transistor is connected to the second node, and the other end of the second transistor is connected to the second node through the fourth target switch tube.

10. The charge pump according to claim 9, characterized in that When the charge pump is in a charging state, the third target switch tube is in a conducting state, so that the potentials at both ends of the first transistor are equal; When the charge pump is in a discharging state, the fourth target switch tube is in a conducting state, so that the potentials at both ends of the second transistor are equal; When the charge pump is in a fully-on state, the third target switch tube and the fourth target switch tube are both in a conducting state, so that the potentials at both ends of the first transistor are equal and the potentials at both ends of the second transistor are equal.

11. A phase-locked loop, characterized in that: The phase-locked loop includes: a phase frequency detector, a loop filter, a voltage-controlled oscillator, a frequency divider, and a charge pump as described in any one of claims 1 to 10; wherein the phase frequency detector is coupled to the charge pump, the charge pump is coupled to the loop filter and the voltage-controlled oscillator, the voltage-controlled oscillator is coupled to the frequency divider, and the frequency divider is coupled to the phase frequency detector.

Images