WO2012087532A1

WO2012087532A1 - Charge pump systems with reduction in inefficiencies due to charge sharing between capacitances

Info

Publication number: WO2012087532A1
Application number: PCT/US2011/062890
Authority: WO
Inventors: Feng Pan; Jonathan H. Huynh
Original assignee: SanDisk Technologies LLC
Current assignee: SanDisk Technologies LLC
Priority date: 2010-12-20
Filing date: 2011-12-01
Publication date: 2012-06-28
Anticipated expiration: 2013-06-20
Also published as: US8294509B2; CN103270682A; US20120154023A1; US8421524B2; US20130038381A1; KR20140008324A; CN103270682B

Abstract

Improvements in the efficiency of two charge pump designs are presented. As a charge pump switches between modes, capacitances are charged. Due to charge sharing between capacitances, inefficiencies are introduced. Techniques for reducing these inefficiencies are presented for two different charge pump designs are presented. For a clock voltage doubler type of pump, a four phase clock scheme is introduced to pre-charge the output nodes of the pump's legs. For a pump design where a set of capacitances are connected in series to supply the output during the charging phase, one or more pre-charging phases are introduced after the reset phase, but before the charging phase. In this pre-charge phase, the bottom plate of a capacitor is set to the high voltage level prior to being connected to the top plate of the preceding capacitor in the series.

Description

CHARGE PUMP SYSTEMS WITH REDUCTION IN INEFFICIENCIES DUE TO CHARGE SHARING BETWEEN CAPACITANCES

FIELD OF THE INVENTION

[0001] This invention pertains generally to the field of charge pumps and more particularly to improve the efficiency of various charge pump designs by altering the timings for the charging of the capacitances.

BACKGROUND

[0002] Charge pumps use a switching process to provide a DC output voltage larger or lower than its DC input voltage. In general, a charge pump will have a capacitor coupled to switches between an input and an output. During one clock half cycle, the charging half cycle, the capacitor couples in parallel to the input so as to charge up to the input voltage. During a second clock cycle, the transfer half cycle, the charged capacitor couples in series with the input voltage so as to provide an output voltage twice the level of the input voltage. This process is illustrated in Figures la and lb. In Figure la, the capacitor 5 is arranged in parallel with the input voltage Vi to illustrate the charging half cycle. In Figure lb, the charged capacitor 5 is arranged in series with the input voltage to illustrate the transfer half cycle. As seen in Figure lb, the positive terminal of the charged capacitor 5 will thus be 2V_IN with respect to ground. By using several such stages, higher levels can be achieved.

[0003] Charge pumps are used in many contexts. For example, they are used as peripheral circuits on flash and other non-volatile memories to generate many of the needed operating voltages, such as programming or erase voltages, from a lower power supply voltage. A number of charge pump designs, such as conventional Dickson-type pumps, are know in the art. But given the common reliance upon charge pumps, there is an on going need for improvements in pump design, particularly with respect to trying to reduce the amount of layout area and the efficiency of pumps. SUMMARY OF THE INVENTION

[0004] According to a first set of aspects, a charge pump system includes a charge pump and clock generating circuitry. The charge pump includes a first stage having first and second legs. The first leg has: a first capacitor connected between a first internal node and a first clock signal; a first transistor connected between the first internal node and the first stage's input voltage; and a first switch, whereby a first output node is connected to ground when a third clock signal is asserted and the connected to the first internal node when the third clock signal is de-asserted. The second leg has: a second capacitor connected between a second internal node and a second clock signal; a second transistor connected between the second internal node and the first stage's input voltage; and a second switch, whereby a second output node is connected to ground when a fourth clock signal is asserted and connected to the second internal node when the fourth clock signal is de-asserted. The gate of the first transistor is connected to the second internal node and the gate of the second transistor is connected to the first internal node. The charge pump also has a first load capacitance connected between the first output node and ground and a second load capacitance connected between the second output node and ground. The clock generating circuitry provides the first, second, third and fourth clock signal. The first and second clock signal are non-overlapping such that when the first clock signal is high the second clock signal is low, and when the second clock signal is high the first clock signal is low, the third clock signal is de-asserted while the second clock signal is high and reasserted before the second clock signal goes high, and the fourth clock signal is de-asserted while the first clock signal is high and reasserted before the first clock signal goes high.

[0005] In other aspects, a method of operating a charge pump is presented. The charge pump includes a first stage comprising: a first leg having a first capacitor with a first plate connected to a first internal node, a first transistor connected between the first internal node and the first stage's input voltage, and a first output node; a second leg having a second capacitor with a second plate connected to a second internal node, a second transistor connected between the second internal node and the second stage's input voltage, and a second output node, wherein the gate of the first transistor is connected to the second internal node and the gate of the second transistor is connected to the first internal node; a first load capacitance connected between the first output node and ground; and a second load capacitance connected between the second output node and ground. The method comprises: generating and supplying a first clock signal at a second plate of the first capacitor; generating and supplying a second clock signal at a second plate of the second capacitor, wherein the first and second clock signal are non-overlapping such that when the first clock signal is high the second clock signal is low, and when the second clock signal is high the first clock signal is low; generating a third clock, wherein the third clock signal is de-asserted while the second clock signal is high and reasserted before the second clock signal goes high; connecting the first output node to ground when the third clock signal is asserted and to the first internal node when the third clock signal is de-asserted; generating a fourth clock signal, wherein the fourth clock signal is de-asserted while the first clock signal is high and reasserted before the first clock signal goes high; and connecting the second output node to ground when the fourth clock signal is asserted and to the second internal node when the fourth clock signal is de-asserted.

[0006] Further aspects include a method of operating a charge pump including a plurality of capacitors, each having a first plate and a second plate. The method includes alternatingly operating the charge pump in a reset phase and a charging phase. The reset phase includes connecting the first plate of each of the capacitors to a low voltage value and connecting the second plate of each of the capacitors to a high voltage value. The charging phase includes connecting the capacitors in series, where the first plate of each capacitor except the first in the series is connected to the second plate of the preceding capacitor in the series, the first plated of the first capacitor in the series is connected to the high voltage value, and the second of the last plate in the series is connected to an output node of the charge pump. The method further includes a pre-charge phase after each reset phase and before the subsequent charging phase, where the pre-charge phase comprises connecting the second plate of one of the capacitors other than the last in the series to the first plate of next capacitor in the series and connecting first plate of said one of the capacitors to the high voltage value.

[0007] According to another set of aspects, a charge pump system includes a plurality of capacitors, each having a first plate and a second plate, switching circuitry, and control circuitry. By use of the switching circuitry, the first plate of each of the capacitors can be connected to a low voltage value and the second plate of each of the capacitors can be connected to a high voltage value, and the capacitors can be connected in series, with the first plate of each capacitor except the first in the series is connected to the second plate of the preceding capacitor in the series, the first plate of the first capacitor in the series is connected to the high voltage value, and the second of the last plate in the series is connected to an output node of the charge pump. The control circuitry is connected to the switching circuitry to alternatingly operate the charge pump in a reset phase and a charging phase, where the reset phase includes connecting the first plate of each of the capacitors to a low voltage value and connecting the second plate of each of the capacitors to a high voltage value, and the charging phase includes connecting the capacitors in series, and wherein the control circuitry further connects the switching circuitry in a pre-charge phase after each reset phase and before the subsequent charging phase, where the pre-charge phase comprises connecting the second plate of one of the capacitors other than the last in the series to the first plate of next capacitor in the series and connecting first plate of said one of the capacitors to the high voltage value.

[0008] Various aspects, advantages, features and embodiments of the present invention are included in the following description of exemplary examples thereof, which description should be taken in conjunction with the accompanying drawings. All patents, patent applications, articles, other publications, documents and things referenced herein are hereby incorporated herein by this reference in their entirety for all purposes. To the extent of any inconsistency or conflict in the definition or use of terms between any of the incorporated publications, documents or things and the present application, those of the present application shall prevail.

BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The various aspects and features of the present invention may be better understood by examining the following figures, in which:

Figure 1A is a simplified circuit diagram of the charging half cycle in a generic charge pump.

Figure IB is a simplified circuit diagram of the transfer half cycle in a generic charge pump.

Figures 2A and 2B are a schematic illustrations of the operation of one variety of charge pump design of the voltage adder type. Figures 3A and 3B respectively show a schematic of voltage doubler type charge pump and a corresponding timing diagram.

Figures 4A and 4B illustrate the loss of efficiency due to charge sharing.

Figures 5A-C introduce a pre-charge phase into the process shown in Figures 4 A and 4B.

Figures 6A-D are a schematic illustration of the operation of the charge pump design of Figures 2A and 2B with pre-charging phases included.

Figure 7 is a schematic for implementing a 2-stage version for the arrangement of Figures 2 A and 2B.

Figure 8 is a schematic for implementing pre-charging in the charge pump of Figure 7.

Figures 9A and 9B are a schematic illustrations of the operation of one variety of charge pump design of the voltage adder type using a 4-phase clock.

DETAILED DESCRIPTION

[0010] Charge pumps find many applications in integrated circuit contexts where the system needs, in at least some phases of its operations, voltage values that extend beyond those available from the power supply. In particular, non-volatile memory devices often operate on fairly low supply voltage values, but require higher voltage levels for the writing and erasing of data. The techniques presented here can be advantageously used is such non-volatile memory devices, including the EEPROM based flash memory such as those described in US patent numbers 5,570,315, 5,903,495, 6,046,935 or the sort of "3D" structure presented in US patent number 7,696,812 and references found therein. Whatever their application, as charge pump are typically peripheral elements on a the circuit, there is a desire to make then as efficient, both in terms of power and area, as is practical. A number of different charge pump designs are known, having various relative advantages depending on how their application, but all of these can benefit from improved efficiency. Two specific designs are shown in Figures 2A and 2B and Figures 3A and 3B.

[0011] Figures 2 A and 2B are a schematic illustration of the operation of one variety of charge pump design of the voltage adder type. The pump includes multiple (here 3) stages, each with a capacitor, that alternately operations in a reset phase and a charging phase. In the reset or initialization phase, as shown in Figure 2A, each of capacitors CI 101, C2 103, and C3 105 has its bottom plate connected to ground (or more generally a lower voltage value) and its top plate connected to Vcc (or more generally a high voltage value). The output of this pump leg is taken from the node out and is not connected to the stage capacitors during reset. A load capacitor Ceq 107 is also connected between the out node and ground. During the charging or transfer phase, when the pump supplies a voltage to output node, the connections are switched as shown in Figure 2B so that the capacitors are connected in series, with the top plate of each connected to the bottom plate of the next in the series, except for the first capacitor CI 101 that has its bottom plate connected to Vcc and the last capacitor in the series (here C3 105) has its top plate connected to the output node. (The switching circuitry is not explicitly shown here to simplify the discussion, but see Figure 7 discussed below.) When operating, the pump alternates between the two phases, recharging the capacitors during the reset and then supplying the boosted charge in the charging phase.

[0012] Figures 3A and 3B respectively show a schematic of a clock voltage doubler type charge pump and a corresponding timing diagram. The pump has left and right legs each connected to receive from node A the stage's input voltage, here Vdd, and each having a corresponding capacitor 205 and 207. The bottom plate of capacitors 205 and 207 are respectively connected to receive the non-overlapping clock signals CLK1 and CLK2 and the top plates are respectively connected to the input voltage through a switch of NMOS transistor 201 and 203. The gates of these transistor are cross-coupled to the other leg at an intermediate node between the top plate of the capacitor of the other leg and the corresponding transistor of the other leg. The output nodes of the left and right legs are respectively taken at the nodes B and C. The node B is connectable to ground and the top plate of the capacitor 205 be a switch here composed of PMOS 213 and NMOS 215 whose gates are connected to the clock signal CLK1A, which is the inverse of CLK1. Similarly, the node C is connectable to ground and the top plate of the capacitor 207 be a switch here composed of PMOS 209 and NMOS 211 whose gates are connected to the clock signal CLK2A, which is the inverse of CLK2. Each output node is then connected to a load capacitor, 231 and 233 for the left and right output nodes, respectively. Here only a single pump stage is shown, but more generally there may be several such stages before, in which case the input voltage would come from the preceding stage or stages, after, in which case the output would supply the subsequent stages instead of being connected directly to the load capacitances, or both.

[0013] Figure 3B shows an exemplary set of waveforms for the voltage double circuit of Figure 3A. The two clock signals CLK1 and CLK2 are non-overlapping so that when either one is high, the other is low. The switches for the output node connections are controller by the inverses of CLK1 and CLK2, CLK1A and CLK2A respectively. The outputs at nodes B and C are respectively V02 and V20. As shown, these outputs are (ideally) the clock signals doubled to 2Vdd.

[0014] The above discussion of Figures 3A and 3B and Figures 2A and 2B provided some of the basics of the operation of these two designs, but does not go into further detail on issues such as regulation, multi-stage (including variable stage) operation, or other design issues. More information on these and other issues on charge pumps that can be applied to the embodiments described here can be found, for example, in "Charge Pump Circuit Design" by Pan and Samaddar, McGraw-Hill, 2006, or "Charge Pumps: An Overview", Pylarinos and Rogers, Department of Electrical and Computer Engineering University of Toronto, available on the webpage "www.eecg.toronto.edu/~kphang/ecel371/chargepumps.pdf. Further information on various other charge pump aspects and designs can be found in US Pat. Nos. 5,436,587; 6,370,075; 6,556,465; 6,760,262; 6,922,096; 7,030,683; 7,554,311; 7,368,979; 7,795,952; and 7,135,910; US Patent Publication numbers 2009-0153230- Al; 2009-0153232-A1; 2009-0315616-A1; 2009-0322413-A1; and 2009-0058506- Al; and applications Nos. 11/295,906 filed on December 6, 2005; 11/303,387 filed on December 16, 2005; 11/845,939, filed August 28, 2007; 12/135,948 filed June 9, 2008; 12/506,998 filed on July 21, 2009; and 12/570,646 filed on September 30, 2009. Examples of a pump system with a variable number of branches can be found, for example, in US Pat. No. 5,781,473 and with a variable number of stages can be found, for example, in US patents numbers 6,370,075 and 6,486,728 and in US patent application number 12/634,385 filed on December 9, 2009.

[0015] For the charge pump stage described in the Background with respect to Figures 1A and IB, the output is twice the input voltage. Similarly, for a voltage adder type arrangement shown in Figures 2 A and 2B, each stage adds Vcc to the preceding stage. However, these are the ideal values and, in an actual application, efficiency will be lost due to charge sharing between capacitor plates. Considering Figures 2A and 2B further, in the reset phase, for each of the capacitors the top plates is at Vcc and bottom plate at ground. During the charging phase, the bottom plate of CI 101 will be at Vcc, but its top plate will have charge sharing with the bottom plate of C2 103. Similarly, the top plate of C2 103 will share charge with the bottom plate of C3 105 and the top plate of C3 105 will share charge with the load capacitor Ceq 107. Due to this charge sharing between plates, pump efficiency will be lost. This can be illustrated further with respect to Figures 4 A and 4B.

[0016] The example Figures 4 A and 4B shows three capacitors all having the same exemplary value of 2pF and the high voltage level is here taken as 5V. (Since this is a linear system, the superposition theorem can be applied to see only the effect of charge sharing while omitting the clocking effect.) In the initial, reset condition capacitors Cb 303 and Cc 305 are connected in series as shown, each with both of their plates at ground. The switch 307 is open and the top plate of Ca 301 is at 5V and the bottom plate is at ground. The charge on Ca 301 is Q(Ca)= 5V*2pF = lOpC. The other two capacitors have no charge, Q(Cb) = Q(Cc) = 0C, and node Na is at 5V and nodes Nb, Nc are both at 0V.

[0017] Figure 4B shows the final condition in the charging phase when the switch 307 is closed. Both of nodes Na and Nb will be at 3.33 V, while the node is Nc is only at 1.67V. Considering the case where Nc corresponds to the nodes of the pump of Figures 2A and 2B and Cc corresponds to the load capacitance Ceq 107, this illustrates how charge sharing between the capacitances can dilute the pumps efficiency.

[0018] Similarly, the voltage doubler design of Figures 2A and 2B suffers from such a charge dilution at the capacitors. Ideally, the high value of both of V02 and V20 are twice the input value, 2Vdd; but due to parasitic loading and active load, the actual amplitude will be less.

[0019] To address this problem and improve pump efficiencies, the following sections consider the use of pre-charging and alterations to the clock signals, first for a pump having the arrangement shown in Figures 2A and 2B, and then for the voltage doubler design shown in Figures 3 A and 3B. Pre-charging for Stacked Capacitances

[0020] For low power supply operation, the stacking of capacitors in the sort of arrangement discussed above with respect to Figures 2A and 2B are used in pump clock drivers to generate high amplitude pump clocks to drive stages. As discussed with respect to Figures 4A and 4B, stacking capacitances is bad for power efficiency since charge sharing occurs and the internal node voltages will be lower than expected. This section addresses this issue and improves the efficiency of the pump through a combination of clock controls and selective pre-charging. The idea can be illustrated with respect to Figures 5A-C.

[0021] Figure 5 A is an initial condition in a reset phase, corresponds to Figure 4 A and is similarly labeled. Figure 5B introduces a pre-charge phase for Cb 303. In the pre-charge phase, the node Nb is charged to the high level of, here, 5V. At this point, both of nodes Na and Nb are at 5V and Nc is at 2.5V.

[0022] For the charging phase, the switch 307 is closed and the situation is similar to in Figure 4B, except that the voltage levels will be different due to the pre-charge phase. In this final condition, once the switch is closed the levels on nodes Na and Nb is 5V and at node Nc is 2.5V. Compared to Figures 4A and 4B, by pre-charging the node Nb with the supply level before it is connected to Na by the closing of switch 307 the node Nc is switched from 0V to 2.5V. This corresponding raises the voltage on Nc in the charging phase from 1.67V to 2.5V, improving its efficiency. This same principle can be applied to improve clock driver efficiencies for other pump designs as well.

[0023] Figures 6A-D illustrates how the pre-charging scheme can be applied to the pump design of Figures 2A and 2B. For the 3 stage embodiment shown there, an exemplary embodiment using two intermediate pre-charge phases, but in other embodiments less than all of the available pre-charge phases could be used. The high voltage level is again the on-chip high level of Vcc, but other values could be used. (For example, US patent number 7,586,362 uses regulation in an arrangement that could be used here as well.) Figures 6A and 6D correspond to Figures 2A and 2B and are similarly labeled, but the introduction of the intervening pre-charge stages improves the efficiency and results in a higher output voltage in the charging phase. [0024] After the reset phase in Figure 6A, Figure 6B shows "Phase 1", a first pre- charge phase. Capacitors CI 101 and C2 103 remain as in the reset phase, but C3 105 now has its top plate connected to the output node and its bottom plate is pre-charged to Vcc. In the next pre-charge phase of the sequence, or "Phase 2", as shown in Figure 6C, the top plate of the preceding capacitor in the series C2 103 has its top plate connected to the bottom plate of C3 105 and its bottom plate set to Vcc for pre- charging. In the charging phase (or "Phase 3") of Figure 6D, CI 101 and C2 103 are connected in series, with the bottom plate of CI 101 at Vcc, as in Figure 2B.

[0025] Figure 7 is a schematic for implementing the arrangement of Figures 2A and 2B, but for a 2-stage version to keep exposition more manageable. As shown, the capacitors 701 and 703 are arranged in series between Vcc at the bottom and the output node at the top, which is also connected to the load capacitance 705. When the clock signal CLK2 is low, the transistors 731, 733, and 735 are off and the series connections cut off. When CLKl is high (and CLK2 low), the transistors 711, 713, 723, and 725 are on and the reset phase conditions set. When the clock signals are reversed so that CLKl is low and CLK2 high, the transistors 711, 713, 723, and 725 are off, the transistors 731, 733, and 735 are on, and the charging phase occurs. The alternation of these provides the clock output. For this 2-stage example, considering the switches to be ideal, taking the internal capacitances at lOpF, Vcc as 2.5V, and a 500pF load capacitance to check performance, the output voltage is about 5.3V.

[0026] Figure 8 is a schematic of an implement of a pre-charge phase into the 2- stage design of Figure 7. A switch 741 has now been added, connected at a node between the two capacitances 701 and 703 and an added clock signal CLK4. Rather than control all of 731, 733 and 735 all by CLK2, these are now respectively controlled by CLK2, CLK2B, and CLK2A. During reset, CLKl is again high, and all of CLK4, CLK2A, CLK2B, and CLK2 are low. In the charging phase, CLKl is again low, and all of CLK4, CLK2A, CLK2B, and CLK2 are high. In between, however, to effect the pre-charging, after taking CLKl low, the other are raised sequentially in the order CLK4, CLK2A, CLK2B, CLK2. This allows the lower plate of 703 to pre- charge, followed by the passing of this charge to the output node, and then followed by the main clock driving the serial capacitance. Using the same values as in Figure 7, the introduction of the pre-charging allows the output to now be raised by about half a volt to about 5.8V.

Four Phase Clock Scheme for Clock Voltage Doubler

[0027] The clock voltage doubler type of pump, such as that shown in Figure 3A, can also seen improved efficiencies through charge sharing. This can be done by introducing a four-phase clocking scheme for clock doubler to improve power/area efficiency.

[0028] The exemplary clock voltage doubler of Figure 3 A will again be used for the discussion here, as shown in Figures 9A, but the clock signals CLKIA and CLK2A supplied the switches have been replaced by CLK1B and CLK2B. Although only one stage is stage is shown, as before there may be one or more preceding stages and, rather than the output nodes of the stages' legs being directly attached to the load capacitances 231, 233, there may also be one or more subsequent stages. Also, the switches by which these output nodes B and C are connectable to either ground or the top plate of the legs capacitance (respectively 205, 207) is again formed of a PMOS (213 or 209) and an NMOS (215 or 211) and the clock signals CLK1B, CLK2B are considered asserted when high. What is changed is, however, by the introduction of a four phase clock is the nature of the signals CLK1B and CLK2B and their relationship to CLK1 and CLKK2. This is illustrated with respect to Figure 9B.

[0029] As shown in the top two rows of Figure 9B, CLK1 and CLK2 are again non-overlapping clocks such that when either one is asserted (high), the other is low. Unlike CLKIA in Figure 3B, CLK1B in no longer just the inverter version of CLK1. Instead, the falling edge of CLK1B is now earlier than the rising edge of CLK1, the interval between these two events shown in Figure 9B as Tl . As CLK1B is then low while CLK2 is still high, this turns off NMOS 215, turns 213 on, and allows the node B to be pre-charged to Vdd through 201. This again reduces the amount of charge dilution, increasing the output. Similarly, the node C can be pre-charged by de- asserting CLK2B before CLK2 goes high.

[0030] Power consumption can also reduced by changing the timing for when

CLK1B and CLK2B are reasserted. If CLK1 has its falling edge earlier than the rising edge of CLK1B, as shown in the interval T2 of Figure 9B, this will allow a portion of the charge on node B (at the voltage V02) to be recycled through the transistor 213 back to the top plate of the capacitor 205 in reset phase of V02. Similarly, if CLK2 has its falling edge earlier than the rising edge of CLK2B, this will allow a portion of the charge on node C (at the voltage V20) to be recycled back to the top plate of the capacitor 207 in reset phase of V20.

[0031] The result of the pre-charging can be seen in the bottom two lines of Figure 9B. During the interval Tl, the level on node B of V02 begins to rise as it is pre-charged to Vdd. CLK 1 then goes high and V02 reaches it full value. Similarly, the voltage V20 on node C begins to pre-charge to Vdd once CLK2B goes low.

[0032] To give some exemplary values, taking the pump capacitors 205, 207 at 5pF, the load capacitances 231, 233 at lpF and Vdd at 2.5V, under the arrangement of Figures 2A and 2B the high value of V02 and V20 is 4.14V. For the four phase arrangement of Figures 9A and 9B using the same values, the high value of V02 and V20 is increased to 4.48V, illustrating the improved area efficiency and power efficiency. To obtain a similar increase by alter the pump capacitances would require a significant increase in their values.

Conclusion

[0033] Although the invention has been described with reference to particular embodiments, the description is only an example of the invention's application and should not be taken as a limitation. Consequently, various adaptations and combinations of features of the embodiments disclosed are within the scope of the invention as encompassed by the following claims.

Claims

IT IS CLAIMED:

1. A charge pump system comprising:

a charge pump, the charge pump including:

a first stage having first and second legs, the first leg having:

a first capacitor connected between a first internal node and a first clock signal;

a first transistor connected between the first internal node and the first stage's input voltage; and

a first switch, whereby a first output node is connected to ground when a third clock signal is asserted and the connected to the first internal node when the third clock signal is de-asserted; and the second leg having;

a second capacitor connected between a second internal node and a second clock signal;

a second transistor connected between the second internal node and the first stage's input voltage; and

a second switch, whereby a second output node is connected to ground when a fourth clock signal is asserted and connected to the second internal node when the fourth clock signal is de-asserted, wherein the gate of the first transistor is connected to the second internal node and the gate of the second transistor is connected to the first internal node;

a first load capacitance connected between the first output node and ground; and

a second load capacitance connected between the second output node and ground; and

clock generating circuitry that provides the first, second, third and fourth clock signal, wherein the first and second clock signal are non-overlapping such that when the first clock signal is high the second clock signal is low, and when the second clock signal is high the first clock signal is low,

wherein the third clock signal is de-asserted while the second clock signal is high and reasserted before the second clock signal goes high, and wherein the fourth clock signal is de-asserted while the first clock signal is high and reasserted before the first clock signal goes high.

2. The charge pump system of claim 1, wherein the third clock signal is reasserted after the first clock signal goes low and wherein the fourth clock signal is reasserted after the second clock signal goes low.

3. The charge pump system of claim 1, wherein the third clock signal is high when asserted and low when de-asserted and wherein the fourth clock signal is high when asserted and low when de-asserted.

4. The charge pump system of claim 1, further including one or more additional stages, each with first and second legs through with the first and second output nodes are respectively connected to the first and second load capacitances.

5. The charge pump system of claim 1, further including one or more additional stages from which the first stage's input voltage is supplied.

6. The charge pump system of claim 1, wherein the first switch includes: a PMOS transistor connected between first output node and the first internal node and having a gate connected to third clock signal; and

an NMOS transistor connected between the first output node and ground and having a gate connected to third clock signal; and

wherein the second switch includes:

a PMOS transistor connected between second output node and the second internal node and having a gate connected to fourth clock signal; and an NMOS transistor connected between the second output node and ground and having a gate connected to fourth clock signal.

7. A method of operating a charge pump, the charge pump including a first stage comprising: a first leg having a first capacitor with a first plate connected to a first internal node, a first transistor connected between the first internal node and the first stage's input voltage, and a first output node; a second leg having a second capacitor with a second plate connected to a second internal node, a second transistor connected between the second internal node and the second stage's input voltage, and a second output node, wherein the gate of the first transistor is connected to the second internal node and the gate of the second transistor is connected to the first internal node; a first load capacitance connected between the first output node and ground; and a second load capacitance connected between the second output node and ground, the method comprising:

generating and supplying a first clock signal at a second plate of the first capacitor;

generating and supplying a second clock signal at a second plate of the second capacitor, wherein the first and second clock signal are non-overlapping such that when the first clock signal is high the second clock signal is low, and when the second clock signal is high the first clock signal is low;

generating a third clock, wherein the third clock signal is de-asserted while the second clock signal is high and reasserted before the second clock signal goes high; connecting the first output node to ground when the third clock signal is asserted and to the first internal node when the third clock signal is de-asserted;

generating a fourth clock signal, wherein the fourth clock signal is de-asserted while the first clock signal is high and reasserted before the first clock signal goes high; and

connecting the second output node to ground when the fourth clock signal is asserted and to the second internal node when the fourth clock signal is de-asserted.

8. The method of claim 7, wherein the third clock signal is reasserted after the first clock signal goes low and wherein the fourth clock signal is reasserted after the second clock signal goes low.

9. The method of claim 7, wherein the third clock signal is high when asserted and low when de-asserted and wherein the fourth clock signal is high when asserted and low when de-asserted.

10. The method of claim 7, wherein the charge pump further includes one or more additional stages, each with first and second legs through with the first and second output nodes are respectively connected to the first and second load capacitances.

11. The method of claim 7, wherein the charge pump further includes one or more additional stages from which the first stage's input voltage is supplied.

12. The method of claim 7, wherein the first leg further includes:

a PMOS transistor connected between first output node and the first internal node and having a gate connected to receive the third clock signal; and

an NMOS transistor connected between the first output node and ground and having a gate connected to receive the third clock signal; and wherein the second leg further includes:

a PMOS transistor connected between second output node and the second internal node and having a gate connected to receive the fourth clock signal; and

an NMOS transistor connected between the second output node and ground and having a gate connected to receive the fourth clock signal.

13. A method of operating a charge pump including a plurality of capacitors, each having a first plate and a second plate, the method comprising:

alternatingly operating the charge pump in a reset phase and a charging phase, the reset phase including:

connecting the first plate of each of the capacitors to a low voltage value and connecting the second plate of each of the capacitors to a high voltage value, and

the charging phase including:

connecting the capacitors in series, where the first plate of each capacitor except the first in the series is connected to the second plate of the preceding capacitor in the series, the first plated of the first capacitor in the series is connected to the high voltage value, and the second of the last plate in the series is connected to an output node of the charge pump, and

wherein the method further includes, a pre-charge phase after each reset phase and before the subsequent charging phase, where the pre-charge phase comprises connecting the second plate of one of the capacitors other than the last in the series to the first plate of next capacitor in the series and connecting first plate of said one of the capacitors to the high voltage value.

14. The method of claim 13, wherein the low voltage value is ground.

15. The method of claim 13, w herein the high voltage value is the on-chip high voltage level.

16. The method of claim 13, wherein the charge pump includes more than two capacitors and the pre-charge phase has a plurality of sub-phases, where the second plate of a corresponding one of the capacitors other than the last in the series is connected to the first plate of next capacitor in the series and connecting first plate of said one of the capacitors to the high voltage value,

wherein the pre-charge sub-phases are performed sequentially in the reverse order in which the corresponding ones of the capacitors occur when serially connected.

17. The method of claims 13, wherein the charge pump further includes a load capacitor connected between the output node and ground.

18. A charge pump system, comprising:

a plurality of capacitors, each having a first plate and a second plate;

switching circuitry, whereby the first plate of each of the capacitors can be connected to a low voltage value and the second plate of each of the capacitors can be connected to a high voltage value, and whereby the capacitors can be connected in series, with the first plate of each capacitor except the first in the series is connected to the second plate of the preceding capacitor in the series, the first plated of the first capacitor in the series is connected to the high voltage value, and the second of the last plate in the series is connected to an output node of the charge pump; and

control circuitry connected to the switching circuitry to alternating operate the charge pump in a reset phase and a charging phase, where the reset phase includes connecting the first plate of each of the capacitors to a low voltage value and connecting the second plate of each of the capacitors to a high voltage value, and the charging phase includes connecting the capacitors in series, and wherein the control circuitry further connects the switching circuitry in a pre-charge phase after each reset phase and before the subsequent charging phase, where the pre-charge phase comprises connecting the second plate of one of the capacitors other than the last in the series to the first plate of next capacitor in the series and connecting first plate of said one of the capacitors to the high voltage value.

19. The charge pump system of claim 18, wherein the low voltage value is ground.

20. The charge pump system of claim 18, w herein the high voltage value is the on-chip high voltage level.

21. The charge pump system of claim 18, wherein the charge pump system includes more than two capacitors and the pre-charge phase has a plurality of sub- phases, where the second plate of a corresponding one of the capacitors other than the last in the series is connected to the first plate of next capacitor in the series and connecting first plate of said one of the capacitors to the high voltage value,

wherein the control circuitry performs the pre-charge sub-phases sequentially in the reverse order in which the corresponding ones of the capacitors occur when serially connected.

22. The charge pump system of claim 18, further including a load capacitor connected between the output node and ground.