The present invention relates generally to voltage regulators and in particular, to switching-type regulators including but not limited to multi-phase switching type buck, boost, buck-boost regulators and synchronous versions of the same.
Embodiments of the invention are illustrated by way of example, and not by way of limitation, in the figures of the accompanying drawings in which like reference numerals refer to similar elements.
Various embodiments disclosed herein describe dynamically adjustable, multi-phase voltage regulators (VRs), which can provide for enhanced efficiency response characteristics.
Multiphase switching regulators such as so-called buck converters are commonly used to deliver a regulated DC voltage to components including, but not limited to, microprocessors, chipsets, memory cards and graphics accelerators. Switching type regulators generally use at least one or more switch and/or rectifier elements and an inductor to implement a “phase leg” to deliver switchably controlled current from an applied voltage source to a load through a regulated voltage node. When the load requires more current, the one or more switches are controlled to supply the load with current from the applied voltage for longer durations within a switching period. Conversely, the current is applied for smaller durations in the period when less current is required by the load.
Switching losses in the switch elements tend to dominate the overall power losses at light loads, while conduction losses tend to dominate power losses for larger loads. As a result, power conversion efficiency is typically at its highest for relatively intermediate load conditions, and tends to roll off at both light and higher load conditions under conventional operation.
With multi-phase regulators, several phase legs are commonly coupled to the regulated voltage node to deliver to it their current. The conduction time of the individual phase legs are staggered, typically uniformly, over a switching interval to minimize output voltage/current ripple, and to reduce the size of output capacitance needed to filter switching noise. This technique is commonly referred to as multi-phase power conversion. It is common for multiphase converters to be optimized for relatively high-load current conditions in order to reduce conduction type power losses. Unfortunately, electronic components, e.g., on computing platforms, can spend much time in a low-current “idle” mode, which can result in inefficient regulator operation and undesired loss of power. Such losses are magnified, for example, with battery operated platforms. Accordingly, new approaches may be desired.
The controller controls the phase legs so that the load current is distributed substantially evenly amongst all active phase legs (or phases). Depending on load current demand, a number of the phases are enabled so that each phase operates at a suitably efficient level over a given load current range. In some embodiments, the controller monitors the overall average load current and enables a selected number of the phases so that the average current per phase is maintained within a suitable window for desired efficiency. As the overall load current demand goes up, phases may be activated and conversely, as load current goes down, they may be dropped out. In some embodiments, the phases may be dropped, one at a time, as the load becomes lighter with the lightest load operation being supported by a single phase.
In the depicted embodiment, the driver section 312 comprises N drivers (Di); the current feedback section 332 comprises N current sensors (I-Sensei), and the output filter section 322 comprises N inductors (Li) and a capacitor C. The drivers (Di) are coupled to the inductors (Li) through the current sensors (I-Sensei), forming N phase legs 304i. The phase legs 304i are coupled, as shown, to a common output voltage node (VR) to feed current into regulated load 350.
As used herein, the term “phase leg” refers to a driver (i.e., one or more switching and/'or rectifier elements) coupled to one or more inductors and/or capacitors and to a regulated voltage node to controllably couple an applied supply (e.g., Vin) to a regulated load via the one or more inductors and/or capacitors. The phase leg may implement a phase of a multi-phase converter for any desired converter scheme including but not limited to a buck, synchronous buck, boost, buck-boost, or fly back converter. For example, in a synchronous buck type converter, a driver (Di) in a phase leg may comprise complementary PMOS and NMOS transistors coupled with their drains commonly connected to an inductor to switchably couple it to an applied high-side supply, couple it to a low-side supply, or tri-state it. On the other hand, the phase legs could be configured to implement a standard buck type driver, for example, with a switch such as a transistor coupled to an inductor having a diode coupled from the low supply referenced to the switch-inductor node to provide the inductor with current when the switch is open. Countless other driver and inductor/capacitor implementations will be appreciated by one of ordinary skill and are within the scope of the present invention.
Along the same lines, it should be appreciated that the inductors could be implemented with any type of inductor (or transformer technique for that matter) to provide a desired amount of inductance, which might vary depending on design concerns and operating environment. For example, they could be made with magnetic cores or implemented with so-called “air-cores”. They could be arranged to be magnetically coupled to one another, or alternatively, some or all could be in a destructive coupling arrangement. In some cases, they might even be implemented with simple traces or the like depending on such factors as switching frequencies and design concerns.
(The term “PMOS transistor” refers to a P-type metal oxide semiconductor field effect transistor. Likewise, “NMOS transistor” refers to an N-type metal oxide semiconductor field effect transistor. It should be appreciated that whenever the terms: “transistor”, “MOS transistor”, “NMOS transistor”, or “PMOS transistor” are used, unless otherwise expressly indicated or dictated by the nature of their use, they are being used in an exemplary manner. They encompass the different varieties of MOS devices including devices with different VTs and oxide thicknesses to mention just a few. Moreover, unless specifically referred to as MOS or the like, the term transistor can include other suitable transistor types, e.g., junction-field-effect transistors, bipolar-junction transistors, and various types of three dimensional transistors, known today or not yet developed.)
In the depicted embodiment, the current feedback section 332 comprises current sensors 334 (I-Sensei) coupled in series between associated drivers (Di) and inductors (Li) to provide a signal (I)indicative of current in a phase leg. They may be implemented with any suitable device or combination of devices to monitor current in a leg without substantially impeding its current. In the depicted embodiment, each sensor is disposed between a driver and an inductor, but this is not required. For example, they could be inserted anywhere within a phase leg to indicate its current. In fact, a separate sensor element may not even have to be inserted in the phase leg current path. For example, in some embodiments, a transistor may be coupled across a transistor in a phase leg's driver (e.g., P-type transistor coupled to the applied supply) in a current mirror configuration, to indicate the current in the phase leg. The additional transistor forming the mirror may be coupled in series, e.g., with a resistor, to provide a voltage signal that is indicative of the current in the leg. In the depicted embodiment, the phase leg current signals (Ii) are voltage signals indicative of the current in their respective phase legs.
The current feedback section 332 also comprises circuitry (not shown) to sum the power leg currents and provide an overall average load current signal (IAVG) for the load 340. (For example, a capacitor could be used to integrate the sense current signals from active (enabled) phase legs 304i.) The average load current (IAVG), as well as the individual sense current signals (I1:IN) are provided back to the PWM 302.
With the depicted embodiment, the current feedback section 312 also has circuitry to generate an active voltage positioning signal (VAVP) to be provided to the voltage feedback section 342 for offset adjustment of the regulated output (VR) to account for load-line impedance.
The voltage feedback section 342 receives a sensed voltage (VSense) from the regulated output (VR), as well as the VAVP signal, to generate a proportional-integral-derivative (PID) signal to be used by the PWM 302 to track the regulated output voltage (VR) against a desired reference voltage (not shown). Additionally, in some embodiments, the voltage feedback section 342 may provide suitable feedback compensation (e.g., with a broadband feedback compensator circuit) to provide for stable system response for phase additions and removals (discussed more below).
PWM 302 generates control (e.g., drive) signals (P1 to PN) to be applied to the phase legs to controllably couple them to or decouple them from the associated applied voltage (Vin). The PWM 302 also generates phase enable (EN2 to ENN) signals to selectably enable or disable phase legs 3042 to 304N, respectively, depending on how much load current (as indicated by IAVG) is being drawn.
In operation, the PWM generates drive signals (P1 to PN) to controllably couple associated inductors to the applied voltage (Vin). Based on its applied drive signal (Pi), a driver (Di), in proportion to the duty cycle of its applied drive signal, controls the average amount of current provided to the load through its inductor (Li) and thus can be used to regulate the load voltage (VR). The drive signals (Pi) are thus said to be “pulse width modulated” for regulating the output load voltage (VR).
The drive signals (P1 to PN) are skewed in time (phase shifted) so that switching noise, from each inductor is distributed in time. This reduces ripple and allows for a greater overall amount of current to be provided to the load than could otherwise be provided by a single phase leg. The PWM 302 monitors the average load current (IAVG) and compares it with one or more thresholds to cause an appropriate number of phases (phase legs) to be enabled for desired efficiency for the range of operational load currents. As the load current increases to a next higher “window”, the PWM engages an additional phase (phase leg). Conversely, if the load current goes down to a next lower window, it drops out (disengages) a phase. In this way, for the most part, it maintains the average current-per-phase within a desired efficiency range.
When a phase is added or dropped, the PWM 302 redistributes the remaining phases over the switching cycle. The feedback loop and the output capacitance should be selected so that the noise in the regulated voltage at the time of phase realignment remain within stable operation limits.
Rapid changes in load conditions may cause a spike or droop in the output regulated voltage. Accordingly, a voltage sense element (shown in the figure) to detect load transients may be employed to monitor such transients. The PWM 302 may be configured so that when such load transients are detected, the PWM reacts quickly to activate all phases to allow for rapid charging or discharging of the output capacitor in response to the transient event. For stable handling of load transient and corresponding phase adjustment events, it may be desirable (if not necessary) to have relatively low-bandwidth current sensing such as the averaging scheme discussed above with the generation of IAVG.
The control/DLL circuit 402 receives a reference signal 401 and generates from it four reference phase signals: CP1 to CP4. CP1 has a phase angle of 0°; CP2 has a phase angle of 120°; CP3 has a phase angle of 240°; and CP4 has a phase angle of 180°. In the depicted embodiment, the reference signal is a clock (e.g., pulse-train) signal used as a reference phase for the DLL circuitry to generate clocks with the desired phase angles. The differently-phased pulse clocks are then converted to triangular signals with corresponding phase angles to provide the CPi signals. However, other approaches could be used. For example, with other PWM approaches, pulse clocks could be used for the CP signals, or alternatively, if triangular CP signals are desired, a triangular reference wave could be used as the input reference and appropriate delay circuitry could be used to generate from it the differently-phased triangular signals. In some embodiments, reference phases and the differently phased signals with frequencies in excess of 10 MHz., e.g., 50 MHz., may be employed.
In the depicted embodiment, the DLL has delay stages with at least four tap points to provide the four control phase signals (CP1 to CP4) with their indicated phase relationships. these signals are then converted to triangular signals, with corresponding phase angles, and provided for the CPi signals. The control/DLL may also include circuitry (e.g., level shifting circuitry) to adjust an offset in each of the control phase signals in order to implement the pulse width modulation for control of the regulated voltage (VR).
Summing circuit 404 receives the PID signal and the sense current signal I1 from phase 1 and adds them together to generate a sum that is coupled to the negative input of comparator 412. The other input receives the CP1 control phase signal. The comparator output generates phase signal P1, which is a pulse train with a duty cycle that is controlled by the comparison between PID+I1 and CP1. The PID+I1 sum essentially serves as a moving comparator threshold against the CP1 triangular wave. When CP1 is below PID+I1, then P1 is Low. Conversely, when CP1 rises above PID+I1, then P1 goes High. Accordingly, the PID and I1 signals provide negative feedback because when they go up, less of the CP1 triangular will rise above them, resulting in a P1 with a smaller duty cycle. On the other hand, when they go down, more of each CP1 triangular wave will be above them, resulting in P1 having a larger duty cycle. (It should be appreciated that there are various ways to achieve the desired feedback, i.e., negative feedback with respect to PID and I1. For example, in an alternative embodiment, a DC reference could be applied at the positive input, and the CP1 signal could be applied to the summing circuit 404 to be added with PID and I1. In this way, the DC reference, rather than the triangular wave, could be adjusted in order to control the output voltage value. Alternatively, CP1 and PID could be combined, e.g., summed, together and applied to the negative input, and the DC reference could be summed with a negative version of I1, with this sum applied to the positive terminal. Various other ways and combinations could be implemented and are within the scope of the invention.)
Summer circuit 408 and comparator 416 function similarly except they act on CP3, PID, and I3 to generate phase signal P3. Summing circuit 406 and comparator 414 also work similarly, except they work on CP2 or CP4 (depending on the selection of switch 410), the PID signal and I2 to generate phase signal P2.
Hysteretic comparators 418 and 420 are used to determine how many of the three phases (P1, P2, and P3) should be engaged. As with the other functional circuit blocks described herein, they may be implemented with any suitable circuitry including but not limited to analog circuit elements, digital logic elements, machine code, and the like. Both comparators receive at their positive inputs the average load current signal (IAVG), but comparator 418 receives at its negative input fixed reference signal (Vref3), while comparator 420 receives at its negative input fixed reference signal (Vref2). (The value of Vref3 will be greater than Vref2.) When IAVG is higher than Vref3 (and thereby Vref2 as well), the EN3 and EN2 signals will both be asserted to enable all three phases. On the other hand, when IAVG is higher than Vref2 but smaller than Vref3, then EN3 will de-assert and EN2 will assert so that phase 3 is dropped off and only phases 1 and 2 are active. Finally, when IAVG is less than both Vref2 (and thereby Vref3 as well), then both EN3 and EN2 will de-assert to disable both phase 3 and phase 2 so that only phase 1 remains active. Accordingly, Vref3 should coincide with the point when the average per-phase current becomes inefficiently low for three active phases, and VRef2 should be selected to coincide with the point when the average per-phase current is inefficiently low for two active phases.
The circuit of
When IAVG goes below Vref3, EN3 de-asserts, which causes phase 3 to be disabled. (Note that the phases, when dropped off, could be isolated from the load or disabled some other way in any suitable manner. For example, a transistor switch controlled by the pertinent enable signal could be used to isolate an associated switch power leg from the load.)
The de-assertion of EN3 also causes switch 410 to select CP4 instead of CP2. This results in the reference triangular at comparator 414 to now be at a relative phase angle of 180° instead of 120°. Accordingly, the two active phases, phase 1 and phase 2 are evenly re-distributed at 0° and 180°, respectively. In addition, note that the sense current for each phase is applied in a negative feedback manner (added to the PID signal) for its associated comparator. This results in reasonable load balancing, both for steady-state and transient conditions because if any switch power leg “comes up” too quickly, it will be inhibited by the duty cycle of its associated driver signal (Pi) decreasing. it should be appreciated that these circuit techniques may be scaled to larger numbers of phases. More switches may be required, but with a reference phase source having a sufficient number of tap points, the necessary phase combinations for the different phases (as other phases are removed or added) may reasonably be implemented for sufficiently even phase distribution.
With reference to
It should be noted that the depicted system could be implemented in different forms. That is, it could be implemented in a single chip module, a circuit board, or a chassis having multiple circuit boards. Similarly, it could constitute one or more complete computers or alternatively, it could constitute a component useful within a computing system.
The invention is not limited to the embodiments described, but can be practiced with modification and alteration within the spirit and scope of the appended claims. For example, it should be appreciated that the present invention is applicable for use with all types of semiconductor integrated circuit (“IC”) chips. Examples of these IC chips include but are not limited to processors, controllers, chip set components, programmable logic arrays (PLA), memory chips, network chips, and the like.
Moreover, it should be appreciated that example sizes/models/values/ranges may have been given, although the present invention is not limited to the same. As manufacturing techniques (e.g., photolithography) mature over time, it is expected that devices of smaller size could be manufactured. In addition, well known power/ground connections to IC chips and other components may or may not be shown within the FIGS. for simplicity of illustration and discussion, and so as not to obscure the invention. Further, arrangements may be shown in block diagram form in order to avoid obscuring the invention, and also in view of the fact that specifics with respect to implementation of such block diagram arrangements are highly dependent upon the platform within which the present invention is to be implemented, i.e., such specifics should be well within purview of one skilled in the art. Where specific details (e.g., circuits) are set forth in order to describe example embodiments of the invention, it should be apparent to one skilled in the art that the invention can be practiced without, or with variation of, these specific details. The description is thus to be regarded as illustrative instead of limiting.
Number | Name | Date | Kind |
---|---|---|---|
6031886 | Nah et al. | Feb 2000 | A |
6278263 | Walters et al. | Aug 2001 | B1 |
6366069 | Nguyen et al. | Apr 2002 | B1 |
6433527 | Izadinia et al. | Aug 2002 | B1 |
6495995 | Groom et al. | Dec 2002 | B2 |
6600296 | Hazucha | Jul 2003 | B2 |
6670794 | Wang et al. | Dec 2003 | B1 |
6803752 | Chen | Oct 2004 | B1 |
6826028 | Schuellein | Nov 2004 | B2 |
6879138 | Dinh et al. | Apr 2005 | B2 |
6947273 | Bassett et al. | Sep 2005 | B2 |
7005835 | Brooks et al. | Feb 2006 | B2 |
7019502 | Walters et al. | Mar 2006 | B2 |
7030596 | Salerno et al. | Apr 2006 | B1 |
7061215 | Harris | Jun 2006 | B2 |
7071660 | Xu et al. | Jul 2006 | B2 |
7202643 | Miftakhutdinov | Apr 2007 | B2 |
7262628 | Southwell et al. | Aug 2007 | B2 |
7492134 | Tang et al. | Feb 2009 | B2 |
20050265269 | Saito et al. | Dec 2005 | A1 |
20060001408 | Southwell et al. | Jan 2006 | A1 |
20060280019 | Burton et al. | Dec 2006 | A1 |
20070013350 | Tang et al. | Jan 2007 | A1 |
20070096955 | New | May 2007 | A1 |
Number | Date | Country |
---|---|---|
1513247 | Mar 2005 | EP |
1513247 | Sep 2005 | EP |
11-127573 | May 1999 | JP |
2003-527062 | Sep 2003 | JP |
2005-86998 | Mar 2005 | JP |
2005-110498 | Apr 2005 | JP |
2006-180689 | Jul 2006 | JP |
2005076433 | Aug 2005 | WO |
WO-200576433 | Aug 2005 | WO |
Entry |
---|
“High-Frequency, Multiphase Controller”, Texas Instruments, SLUS578B—Oct. 2003—Revised May 2006; 30 pgs. |
“UCD9240 Digital Point of Load System Controller”, Texas Instruments, UCD9240 SLUS766—Apr. 2007; 21 pgs. |
Office Action Received for Japanese Patent Application No. 2008-087188, delivered on Aug. 3, 2010, 3 Pages of English Translation. |
Office Action Received for Japanese Patent Application No. 2008-087188, delivered on Aug. 16, 2011, 6 pages of Japanese office action including 3 pages of English Translation. |
Office Action Received for Chinese Patent Application No. 200810100384.9, issued on on Jan. 29, 2010, 9 Pages of Chinese office action including 3 pages of English Translation. |
Office Action Received for Germany Patent Application No. 10 2008 016 532.8 , mailed on on Mar. 24, 2009, 6 Pages of Germany office action including 3 pages of English Translation. |
Office Action Received for Japanese Patent Application No. 2008-087188, delivered on Feb. 8, 2011, 4 Pages of Japanese office action including 2 pages of English Translation. |
Third Office Action mailed May 16, 2012 for Chinese Patent Application No. 2008 10100384.9. |
Final Office Action mailed Feb. 8, 2011 for Japanese Patent Application No. 2008-087188, 4 Pages. |
Non-Final Office Action mailed Aug. 3, 2010 for Japanese Patent Application No. 2008-087188, 5 Pages. |
Non-Final Office Action mailed Mar. 24, 2009 for German Patent Application No. 10 2008 016 532.8-32, 6 Pages. |
Non-Final Office Action mailed Nov. 23, 2011 for Chinese Patent Application No. 2008 10100384.9, 17 Pages. |
Non-Final Office Action mailed Jan. 29, 2010 for Chinese Patent Application No. 2008 10100384.9, 17 Pages. |
Fourth Office Action mailed Dec. 10, 2012 for Chinese Patent Application No. 2008 10100384.9. |
Office Action received for Japanese Divisional Patent Application No. 2011-275607, mailed on Jul. 9, 2013, 2 pages of English Translation and 2 pages of Office Action. |
Number | Date | Country | |
---|---|---|---|
20080238390 A1 | Oct 2008 | US |