U.S. patent application number 14/276834 was filed with the patent office on 2015-11-19 for digital-to-analog converter circuit for use in a power converter.
This patent application is currently assigned to Power Integrations, Inc.. The applicant listed for this patent is Power Integrations, Inc.. Invention is credited to Mingming Mao, Tiziano Pastore, Ricardo Luis Janezic Pregitzer, Peter Vaughan.
Application Number | 20150333764 14/276834 |
Document ID | / |
Family ID | 54539367 |
Filed Date | 2015-11-19 |
United States Patent
Application |
20150333764 |
Kind Code |
A1 |
Pastore; Tiziano ; et
al. |
November 19, 2015 |
DIGITAL-TO-ANALOG CONVERTER CIRCUIT FOR USE IN A POWER
CONVERTER
Abstract
A digital-to-analog converter circuit includes an input to
receive a digital input signal having multiple bits. A modulation
circuit is coupled to respond to less significant bits of the
digital input signal by outputting a modulation signal that
alternates between a logic low level and a logic high level. A
digital-to-analog circuit is configured to convert more significant
bits of the digital input signal to a first analog level. The
digital-to-analog circuit is configured to alternate an analog
output between the first analog level corresponding to a value of
the more significant bits and a second analog level corresponding
to one of adjacent values of the more significant bits in response
to the modulation signal.
Inventors: |
Pastore; Tiziano; (Los
Gatos, CA) ; Pregitzer; Ricardo Luis Janezic;
(Campbell, CA) ; Mao; Mingming; (Saratoga, CA)
; Vaughan; Peter; (Los Gatos, CA) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Power Integrations, Inc. |
San Jose |
CA |
US |
|
|
Assignee: |
Power Integrations, Inc.
San Jose
CA
|
Family ID: |
54539367 |
Appl. No.: |
14/276834 |
Filed: |
May 13, 2014 |
Current U.S.
Class: |
363/21.12 ;
341/145 |
Current CPC
Class: |
H03M 1/742 20130101;
H03M 1/68 20130101; H02M 3/33515 20130101; H03M 1/825 20130101;
H03M 1/745 20130101; H02M 3/33507 20130101 |
International
Class: |
H03M 1/68 20060101
H03M001/68; H03M 1/74 20060101 H03M001/74; H02M 3/335 20060101
H02M003/335 |
Claims
1. A digital-to-analog converter circuit, comprising: an input to
receive a digital input signal having multiple bits; a modulation
circuit coupled to respond to less significant bits of the digital
input signal by outputting a modulation signal that alternates
between a logic low level and a logic high level; a
digital-to-analog circuit configured to convert more significant
bits of the digital input signal to a first analog level, wherein
the digital-to-analog circuit is configured to alternate an analog
output between the first analog level corresponding to a value of
the more significant bits and a second analog level corresponding
to one of adjacent values of the more significant bits in response
to the modulation signal.
2. The digital-to-analog converter circuit of claim 1, wherein the
modulation signal is periodic with a period that is proportional to
a number of possible values that can be represented by the less
significant bits of the digital input signal.
3. The digital-to-analog converter circuit of claim 2, wherein a
duty cycle of the modulation signal is proportional to a value of
the less significant bits of the digital input signal.
4. The digital-to-analog converter circuit of claim 1, wherein the
modulation circuit includes: a counter coupled to receive a clock
signal and count in response to the clock signal, wherein the
counter is configured to reset a counter output to an initial value
when the counter counts a threshold number of times; and a
comparator coupled to receive the less significant bits of the
digital input signal and the counter output, wherein the comparator
is configured to output the modulation signal in response to a
comparison of the counter output and the value of the less
significant bits of the digital input signal.
5. The digital-to-analog converter circuit of claim 4, wherein the
threshold number is equal to a number of possible values that can
be represented by the less significant bits of the digital input
signal.
6. The digital-to-analog converter circuit of claim 4, wherein the
period of the modulation signal is equal to the period of the clock
signal multiplied by a number of possible values that can be
represented by the less significant bits of the digital input
signal.
7. The digital-to-analog converter circuit of claim 1, wherein the
digital-to-analog circuit is a binary-weighted digital-to-analog
converter.
8. The digital-to-analog converter circuit of claim 1, wherein the
digital-to-analog circuit includes a plurality of bit-to-analog
circuitries switchable in response to the more significant bits of
the digital input signal, each of the plurality of bit-to-analog
circuitries including a switch responsive to one of the more
significant bits of the digital input signal and a current
source.
9. The digital-to-analog converter circuit of claim 8 further
including a modulation source switchable in response to the
modulation signal, the modulation source being configured to output
a current that alternates between zero and a level to which a least
significant bit of the more significant bits of the digital input
signal is converted.
10. The digital-to-analog converter circuit of claim 9 further
including a summing block coupled to sum the analog output of the
digital-to-analog circuit and the current output by the modulation
source.
11. A control circuit for use in a power converter having a switch,
the control circuit comprising: a digital-to-analog converter
circuit, coupled to receive a digital input signal having multiple
bits, the digital-to-analog converter circuit including: a
modulation circuit coupled to respond to less significant bits of
the digital input signal by outputting a modulation signal that
alternates between a logic low level and a logic high level; a
digital-to-analog circuit configured to convert more significant
bits of the digital input signal to a first analog level, wherein
the digital-to-analog circuit is configured to alternate an analog
output between the first analog level corresponding to a value of
the more significant bits and a second analog level corresponding
to one of adjacent values of the more significant bits in response
to the modulation period signal; and a drive signal generator
coupled to receive the analog output and output a drive signal to
drive the switching of the switch, wherein the drive signal
generator is configured to adjust the drive signal to set at least
one of an on-time and a switching frequency of the switch in
response to the analog output.
12. The control circuit of claim 11, wherein the modulation signal
is periodic with a period that is proportional to a number of
possible values that can be represented by the less significant
bits of the digital input signal.
13. The control circuit of claim 12, wherein a duty cycle of the
modulation signal is proportional to the value of the less
significant bits of the digital input signal.
14. The control circuit of claim 11, wherein the modulation circuit
includes: a counter coupled to receive a clock signal and count in
response to the clock signal, wherein the counter is configured to
reset a counter output to an initial value when the counter counts
a threshold number of times; and a comparator coupled to receive
the counter output and the less significant bits of the digital
input signal, wherein the comparator is configured to output the
modulation signal in response to a comparison of the counter output
and the value of the less significant bits of the digital input
signal.
15. The control circuit of claim 11, wherein the digital-to-analog
circuit includes a plurality of bit-to-analog circuitries
switchable in response to the more significant bits of the digital
input signal, wherein a level to which each of the more significant
bits is converted is twice a level to which an adjacent lower bit
of the more significant bits is converted.
16. The control circuit of claim 15, wherein each of the plurality
of bit-to-analog circuitries includes a switch responsive to one of
the more significant bits of the digital input signal and a current
source.
17. The control circuit of claim 16, wherein each current source of
the plurality of bit-to-analog circuitries is configured to output
a current that is twice a current output by a current source of the
bit-to-analog circuitry that is responsive to an adjacent lower bit
of the more significant bits.
18. The control circuit of claim 15, wherein the digital-to-analog
circuit further includes a modulation source switchable in response
to the modulation signal, the modulation source being configured to
output a current that alternates between zero and a level to which
a least significant bit of the more significant bits of the digital
input signal is converted.
19. The control circuit of claim 18, wherein the digital-to-analog
converter circuit includes a summing block coupled to sum a current
output by the modulation source and the first analog level.
20. A control circuit for a power supply having a power switch, the
control circuit comprising: a modulation circuit coupled to receive
a digital signal and a clock signal, and to output a modulation
signal, wherein the modulation circuit is configured to set the
modulation signal to a first level corresponding to a value of the
more significant bits of the digital signal for a first number of
periods of the clock signal and to a second level corresponding to
an adjacent value of the more significant bits of the digital
signal for a second number of periods of the clock signal, the
first number and the second number being representative of a value
of less significant bits of the digital signal; and a drive signal
generator coupled to output a drive signal to drive switching of
the power switch, wherein the drive signal generator is configured
to adjust the drive signal to set at least one of an on-time and a
switching frequency of the switch in accordance with a setting of
the modulation signal.
21. The control circuit of claim 20 further including a
digital-to-analog circuit coupled to convert the modulation signal
to an analog output, the drive signal generator being configured to
adjust the drive signal to set at least one of an on-time and a
switching frequency of the power switch in response to the analog
output.
Description
REFERENCE TO RELATED APPLICATION
[0001] This application is related to U.S. patent application Ser.
No. ______ of Pastore et al., filed May 13, 2014, entitled
"MULTI-PACE CONTROLLER," and assigned to the Assignee of the
present application.
BACKGROUND INFORMATION
[0002] 1. Field of the Disclosure
[0003] The present invention relates generally to power supplies,
and more specifically, the invention relates to switch mode power
supplies.
[0004] 2. Background
[0005] Many electronic devices, such as cell phones, laptops, etc.,
are powered by a source of direct current (dc) power. Conventional
wall outlets generally deliver a high voltage alternating current
(ac) power that needs to be transformed to dc power in order to be
used as a power source by most consumer electronic devices. Switch
mode power converters, also referred to as switch mode power
supplies, are commonly used due to their high efficiency, small
size, and low weight to convert the high voltage ac power to a
regulated dc power. In one example, switch mode power converters
are used to provide regulated power to light emitting diode (LED)
devices.
[0006] One important consideration for a switch mode power
converter is the shape and the phase of the input current drawn
from the power source relative to the ac input voltage. The shape
of the ac input voltage is typically sinusoidal but because a
switching power converter presents itself as a non-linear load, the
shape of the input current drawn from the power source may become
distorted (non-sinusoidal) and/or out of phase with ac input
voltage. This results in increased power loss in the power
distribution systems.
[0007] Correction of the input current waveform to reduce shape
and/or phase mismatch with respect to input voltage is referred to
as power factor correction (PFC). The power factor may be defined
as the ratio of the average power over a cycle to the product of
the root mean square (rms) voltage and the rms current. That is,
the power factor may represent the ratio of the amount of usable
power to the amount of total power delivered to the load. As such,
the power factor may have a value between zero and one, with unity
power factor being the optimal. If the input current is sinusoidal
and perfectly in-phase with the input voltage, the power factor of
the power supply is one, and none of the energy delivered to the
load is returned to the power source. However, as the switch mode
power supply distorts the wave shape of the input current and/or
introduces a phase shift with respect to the input voltage, the
power factor decreases. Several regulatory agencies have set tight
standards that typically stipulate for greater power factors and/or
lower harmonic content of the input current.
[0008] One example application where switch mode power supplies may
be required to perform PFC is power conversion systems that are
used in light emitting diode (LED) lighting. Since the brightness
of light provided by LED lamps is a function of the current through
LEDs, the power supply used in such a system may also regulate the
current provided to LEDs at the output of the power supply. In
other words, the power supply may provide both output current
regulation and PFC.
[0009] Output current regulation is typically achieved by a power
supply controller by sensing the current provided to the LEDs. A
feedback signal is used to represent a current through the LEDs.
The power supply controller controls the transfer of energy from an
input to an output of the power supply in response to the feedback
signal. Switch mode power supplies typically respond very quickly
to fluctuations in the feedback signal by adjusting the energy
transfer to regulate the LED current at a desired level. However,
making rapid changes to the energy transfer can compromise the PFC
performance and cause the input current to be non-sinusoidal and/or
out of phase with the input voltage, resulting in a reduced power
factor.
[0010] A switch mode power supply may use a controller to control
the switching (i.e., the turning on and turning off) of a power
switch to provide a desired output to a load. The controller may
regulate the output at a desired level in response to a feedback
signal representative of the output of the power supply. Some
controllers may use a digital control signal to adjust the
operating condition (e.g., on-time, switching frequency) of the
power switch in response to the feedback signal. Such a controller
may employ a digital-to-analog converter (DAC) to convert the
binary values of the control signal to corresponding discrete
levels of an analog signal that may be used to set the operating
condition of the power switch. For some types of DACs, such as
binary-weighted DACs, as the number of the bits of the control
signal is increased, the number of different operating conditions
to which the power switch can be set is increased. As a result, the
area on the silicon occupied by the DAC components such as current
sources, resistors, etc., may grow and make such an implementation
impractical.
BRIEF DESCRIPTION OF THE DRAWINGS
[0011] Non-limiting and non-exhaustive embodiments of the present
invention are described with reference to the following figures,
wherein like reference numerals refer to like parts throughout the
various views unless otherwise specified.
[0012] FIG. 1A shows a schematic diagram illustrating an example
switch mode power converter including a controller with a state
selector circuit and a driver circuit in accordance with the
teachings of the present invention.
[0013] FIG. 1B shows examples of relationships of switching
frequency with respect to a state signal of an example switch mode
power converter including a controller in accordance with the
teachings of the present invention.
[0014] FIG. 2A shows a circuit diagram illustrating one example of
the state selector circuit of the controller in FIG. 1A in
accordance with the teachings of the present invention.
[0015] FIG. 2B shows a circuit diagram illustrating another example
of the state selector circuit of the controller in FIG. 1A in
accordance with the teachings of the present invention.
[0016] FIG. 2C shows a circuit diagram illustrating yet another
example of the state selector circuit of the controller in FIG. 1A
in accordance with the teachings of the present invention.
[0017] FIG. 3A shows an example set of waveforms illustrating the
operation of the controller in FIG. 1A in accordance with the
teachings of the present invention.
[0018] FIG. 3B shows another example set of waveforms illustrating
the operation of the controller in FIG. 1A in accordance with the
teachings of the present invention.
[0019] FIG. 3C shows yet another example set of waveforms
illustrating the operation of the controller in FIG. 1A in
accordance with the teachings of the present invention.
[0020] FIG. 4A shows a flow diagram illustrating an example process
for adjusting the operating condition of a switch of a power supply
in accordance with the teachings of the present invention.
[0021] FIG. 4B shows a flow diagram illustrating one example of
detailed steps of one the process blocks of the process shown in
FIG. 4A in accordance with the teachings of the present
invention.
[0022] FIG. 4C shows a flow diagram illustrating one example of
detailed steps of another one of the process blocks of the process
shown in FIG. 4A in accordance with the teachings of the present
invention.
[0023] FIG. 5 shows a circuit diagram illustrating an example
driver circuit including one example of a modulated DAC in
accordance with the teachings of the present invention.
[0024] FIG. 6 shows a table illustrating different values of an
example digital signal received as an input by the driver circuit
in FIG. 5 and a timing diagram illustrating waveforms for various
signals that are associated with the example modulated DAC in
accordance with the teachings of the present invention.
[0025] FIG. 7 shows a collection of signal levels illustrating
example average values for the output of the modulated DAC of FIG.
5 in accordance with the teachings of the present invention.
[0026] FIG. 8 shows a circuit diagram illustrating another example
of a driver circuit in accordance with the teachings of the present
invention.
[0027] FIG. 9 shows a flow diagram illustrating an example process
for generating an analog signal having discrete levels in response
to a digital signal in one example of a power supply controller in
accordance with the teachings of the present invention.
[0028] Corresponding reference characters indicate corresponding
components throughout the several views of the drawings. Skilled
artisans will appreciate that elements in the figures are
illustrated for simplicity and clarity and have not necessarily
been drawn to scale. For example, the dimensions of some of the
elements in the figures may be exaggerated relative to other
elements to help to improve understanding of various embodiments of
the present invention. Also, common but well-understood elements
that are useful or necessary in a commercially feasible embodiment
are often not depicted in order to facilitate a less obstructed
view of these various embodiments of the present invention.
DETAILED DESCRIPTION
[0029] In the following description, numerous specific details are
set forth in order to provide a thorough understanding of the
present invention. It will be apparent, however, to one having
ordinary skill in the art that the specific detail need not be
employed to practice the present invention. In other instances,
well-known materials or methods have not been described in detail
in order to avoid obscuring the present invention.
[0030] Reference throughout this specification to "one embodiment",
"an embodiment", "one example" or "an example" means that a
particular feature, structure or characteristic described in
connection with the embodiment or example is included in at least
one embodiment of the present invention. Thus, appearances of the
phrases "in one embodiment", "in an embodiment", "one example" or
"an example" in various places throughout this specification are
not necessarily all referring to the same embodiment or example.
Furthermore, the particular features, structures or characteristics
may be combined in any suitable combinations and/or subcombinations
in one or more embodiments or examples. Particular features,
structures or characteristics may be included in an integrated
circuit, an electronic circuit, a combinational logic circuit, or
other suitable components that provide the described functionality.
In addition, it is appreciated that the figures provided herewith
are for explanation purposes to persons ordinarily skilled in the
art and that the drawings are not necessarily drawn to scale.
[0031] As will be discussed, an example power converter in
accordance with the teachings of the present invention includes a
controller with a state selector circuit that generates multiple
count signals in response to comparisons of a feedback signal of
the power converter with different threshold signals. In one
example, the state selector circuit generates two count signals,
first count signal and second count signal, and outputs a state
signal in response to these count signals. In the example, the
state selector circuit may be coupled to change the state signal at
a different rate depending on the values of first and second count
signals. In addition, the controller of the example power converter
includes a driver circuit that is coupled to generate a drive
signal in response to the state signal to drive the switching of a
power switch to control a transfer of energy from an input of the
power converter to an output of the power converter in accordance
with the teachings of the present invention.
[0032] To illustrate, FIG. 1A shows an example switch mode power
converter 100, also referred to as a switch mode power supply, with
a controller 150 that includes a state selector circuit 154 and a
driver circuit 156 in accordance with the teachings of the present
invention. In the illustrated example, power supply 100 receives an
input current T.sub.IN 113 and an input voltage V.sub.IN 102 to
output a dc output voltage V.sub.O 120 and a dc output current
I.sub.O 118 to a load 122. Input voltage V.sub.IN 102 may be
representative of an ac line voltage. Load 122 may include one or
more LEDs. In one example, input voltage V.sub.IN 102 is a
rectified and filtered ac voltage. As shown, input voltage V.sub.IN
102 is referenced to a ground terminal 104, which may also be
referred to as an input return terminal. Output voltage V.sub.O 120
is referenced to a ground terminal 105, which may also be referred
to as an output return terminal. In the example, input return
terminal 104 represents the lowest potential or the lowest voltage
against that all voltages on the input side of power supply 100 are
measured or defined, and output return terminal 105 represents the
lowest potential or the lowest voltage against that all voltages on
the output side of power supply 100 are measured or defined. In
some cases, input return terminal 104 and output return terminal
105 may correspond to the same voltage or potential. In some other
cases, input return terminal 104 and output return terminal 105 may
correspond to the different voltages or potentials.
[0033] As shown in the depicted example, power supply 100 may
further include an energy transfer element T1 124, a power switch
S1 170, a diode D1 114 and a capacitor C1 116. In the illustrated
example, energy transfer element T1 124 is a coupled inductor,
which is sometimes referred to as a transformer, with a primary
winding 110 and a secondary winding 112. In one example, primary
winding 110 has one end coupled to the input voltage V.sub.IN 102
and the opposite end coupled to power switch S1 170. Secondary
winding 112 has one end coupled to return terminal 104 and the
opposite end coupled to diode D1 114. Diode D1 114 is further
coupled to capacitor C1 116, which is coupled between the output of
power supply 100 and return terminal 104. While the polarities of
primary winding 110 and secondary winding 112, which are indicated
by the dots at one end of each winding (dotted ends have the same
polarity), show that power supply 100 is configured as a flyback
power supply, it should be appreciated that other power supply
topologies may also be used in accordance with the teachings of the
present invention. In the depicted example, power switch S1 170
represents the operation of a controlled semiconductor device such
as for example a metal oxide semiconductor field effect transistor
(MOSFET) or for example a bipolar junction transistor (BJT). As
shown, power switch S1 170 is coupled to energy transfer element T1
124 at primary winding 110 and to the input of power supply 100 at
return terminal 104.
[0034] In one example, controller 150 may be coupled to control the
switching of power switch S1 170 to control the energy transfer
from the input to the output of power supply 100, thereby
regulating an output quantity U.sub.O 153 (e.g., output voltage
V.sub.O 120, output current I.sub.O 118, or the combination of the
two) at a desired level. In addition, controller 150 may control
the switching of power switch S1 170 to provide input current
T.sub.IN 113 that is in phase with and proportional to input
voltage V.sub.IN 102. That is, controller 150 may control the
switching of power switch S1 170 to provide PFC. In the illustrated
example, controller 150 may provide a drive signal U.sub.DR 162 to
power switch S1 170 to control the switching (i.e., the turning ON
and turning OFF) of power switch S1 170. For example, in response
to drive signal U.sub.DR 162, power switch S1 170 may be switched
to a closed position, which is also referred to as being turned ON
or being in an ON state, and in turn, may conduct current that is
represented by a switch current I.sub.SW 126. Similarly, in
response to drive signal U.sub.DR 162, power switch S1 170 may be
switched to an open position, which is also referred to as being
turned OFF or being in an OFF state, in which power switch S1 170
may substantially prevent current conduction.
[0035] During the operation of power supply 100, when power switch
S1 170 is in the ON state, the voltage across primary winding 110
of energy transfer element T1 124 becomes substantially equal to
input voltage V.sub.IN 102 and causes current in primary winding
110 to increase linearly, which results in energy to be stored in
energy transfer element T1 124. When power switch S1 170 is in the
OFF state, the energy stored in energy transfer element T1 124
while power switch S1 170 was conducting begins to transfer to
output capacitor C1 116 and load 122. This energy transfer may
produce a pulsating current in diode D1 114, which may be filtered
by output capacitor C1 116 to produce a substantially constant
output voltage V.sub.OUT 120. In one example, the switching of
power switch S1 170 may produce a substantially constant output
current I.sub.O 118 to be provided to load 122.
[0036] As shown in the example, a clamp circuit 106 is coupled
across primary winding 110 of energy transfer element T1 124 and is
coupled to the input of power supply 100. In the example, clamp
circuit 106 operates to clamp turn-off spikes that result from
leakage inductance from primary winding 110 across the switching
device S1 170.
[0037] As further depicted in the example shown in FIG. 1A,
controller 150 may be coupled to sense switch current I.sub.SW 126
as a sensed switch signal U.sub.SSW 144. Any known technique to
sense current, such as for example receiving the voltage across a
resistor conducting the current, or for example receiving a scaled
current from a current transformer, or for example receiving the
voltage across the on-resistance of a MOSFET that conducts the
current, may be used to sense switch current I.sub.SW 126 and to
provide sensed switch signal U.sub.SSW 144 to controller 150 in
accordance with the teachings of the present invention.
[0038] In one example, controller 150 may be further coupled to
receive an input sense signal U.sub.INS 142 representative of input
voltage V.sub.IN 102 and an output sense signal U.sub.OS 132
representative of output quantity U.sub.O 153. In one example,
power supply 100 may include an input sense circuit 140 coupled to
sense input voltage V.sub.IN 102 and produce input sense signal
U.sub.INS 142 in response to input voltage V.sub.IN 102. Similarly,
power supply 100 may include an output sense circuit 130 coupled to
sense output quantity U.sub.O 153 and produce output sense signal
U.sub.OS 132 in response to output quantity U.sub.O 153.
[0039] In one example, controller 150 may be implemented as a
monolithic integrated circuit, with discrete electrical components,
or using a combination of discrete and integrated circuits. In
addition, controller 150 and power switch S1 170 may form a part of
an integrated circuit that is manufactured as either a hybrid or a
monolithic integrated circuit.
[0040] As further illustrated in the example depicted in FIG. 1A,
controller 150 may include a feedback signal generator 152, a state
selector circuit 154, and a driver circuit 156. In the illustrated
example, feedback signal generator 152 is coupled to receive input
sense signal U.sub.INS 142, output sense signal U.sub.OS 132, and
sensed switch signal U.sub.SSW 144 to produce a feedback signal
U.sub.FB 158 that is representative of output current I.sub.O 118
(i.e., current in load 122). In some cases, feedback signal
generator 152 may be configured to generate feedback signal
U.sub.FB 158 only in response to output sense signal U.sub.OS 132.
In some other cases, feedback signal generator 152 may be
configured to generate feedback signal U.sub.FB 158 in response to
input sense signal U.sub.INS 142, output sense signal U.sub.OS 132,
and sensed switch signal U.sub.SSW 144.
[0041] In the illustrated example, state selector circuit 154 is
coupled to receive feedback signal U.sub.FB 158. In response, state
selector circuit 154 outputs an N bit digital signal illustrated as
a state signal U.sub.ST 160. In some cases, state selector circuit
154 may also be coupled to receive input sense signal U.sub.INS
142. In operation, state selector circuit 154 gathers information
regarding certain properties of feedback signal U.sub.FB 158, which
may also be referred to as feedback information, at a sampling
frequency for a feedback period and adjusts state signal U.sub.ST
160 in response to the feedback information at the end of the
feedback period. In general, the feedback period is several times
greater than the period of a clock signal used to sample feedback
signal U.sub.FB 158. In other words, the feedback period may be
several times greater than a sampling period of feedback signal
U.sub.FB 158. For instance, in one example, the feedback period can
be half of the period of the ac line voltage (i.e., half line
cycle) and feedback signal U.sub.FB 158 can be sampled 512 times
during each feedback period. That is, the feedback period can be
512 times greater than the sampling period. In another example, the
feedback period can be equal to the period of the ac line voltage.
Additionally, the feedback period may be several times greater than
the switching period of power switch S1 170. That is, power switch
S1 170 may be switched between the ON state and the OFF state
several times (e.g., 1000 times) during the feedback period.
[0042] As shown in the example depicted in FIG. 1A, driver circuit
156 is coupled to receive state signal U.sub.ST 160 and output a
drive signal U.sub.DR 162 to drive the switching of power switch S1
170 such that an operating condition such as on-time and/or
switching frequency of power switch S1 170 is set according to an
operational state indicated by state signal U.sub.ST 160. In one
example, each one of 2.sup.N possible values of state signal
U.sub.ST 160 may represent a different operational state (i.e., a
different on-time and/or switching frequency) for power switch S1
170. As previously mentioned, in one example, state selector
circuit 154 does not adjust state signal U.sub.ST 160 until the end
of a feedback period. This also means that driver circuit 156 does
not adjust drive signal U.sub.DR 162 until the end of the feedback
period. In other words, the operational state (hence, the operating
condition of power switch S1 170) is maintained for the entire
feedback period in one example in accordance with the teachings of
the present invention.
[0043] FIG. 1B shows examples of relationships of switching
frequency of power switch S1 170 with respect to state signal
U.sub.ST 160 of FIG. 1A in accordance with the teachings of the
present invention. As shown in relationship 174 depicted in FIG.
1B, the switching frequency of power switch S1 170 varies with
state signal U.sub.ST 160. Specifically, state signal U.sub.ST 160
shown in FIG. 1A may be a 10 bit digital signal ranging from 0 to
1023, and the switching frequency of power switch S1 170 may
increase (e.g., from 50 kHz to 130 kHz) as state signal U.sub.ST
160 increases (e.g., from 0 to 1023). In another example, as shown
in relationship 176, the switching frequency of power switch S1 170
may increase (e.g., from 50 kHz to 130 kHz) as state signal
U.sub.ST 160 increases until reaching a certain value (e.g., until
state signal U.sub.ST 160 reaches 512) and may remain at a constant
switching frequency (e.g., 130 kHz) for greater values of state
signal U.sub.ST 160. It should be noted of course that FIG. 1B
provides examples for explanation purposes and that other similar
relationships may also exist between the on-time of power switch S1
170 and state signal U.sub.ST 160 in accordance with the teachings
of the present invention.
[0044] Referring back to the example controller 150 depicted in
FIG. 1A, the example state selector circuit 154 may adjust state
signal U.sub.ST 160 by an amount that is based on an operation mode
of state selector circuit 154. In the illustrated example, the
operation mode of state selector circuit 154 is determined
according to the feedback information at the end of a feedback
period. For instance, in response to the feedback information at
the end of a feedback period, state selector circuit 154 may be
operating in a coarse mode of operation. In this coarse mode, state
selector circuit 154 may update the sixth bit of state signal
U.sub.ST 160 in response to the feedback information. That is, in
the coarse mode, state selector circuit 154 may increase state
signal U.sub.ST 160 by 32 (i.e., 0000100000 binary) if the feedback
information indicates that state selector circuit 154 should
increase state signal U.sub.ST 160 and decrease state signal
U.sub.ST 160 by 32 (i.e., 0000100000 binary) if the feedback
information indicates that state selector circuit 154 should
decrease state signal U.sub.ST 160.
[0045] Similarly, in response to feedback information at the end of
the feedback period, state selector circuit 154 may be operating in
a fine mode of operation, and update the first bit, or least
significant bit, of state signal U.sub.ST 160. In other words, in
the fine mode, state selector circuit 154 may increase or decrease
state signal U.sub.ST 160 by 1 (i.e., 0000000001 binary) in
response to the feedback information. In one example, this means
that state selector circuit 154 may adjust state signal U.sub.ST
160 such that the rate of change in state signal U.sub.ST 160
(hence, the rate of change in the operating condition of power
switch S1 170) over multiple feedback periods in the fine mode of
operation is less than the rate of change in state signal U.sub.ST
160 over multiple feedback periods in the coarse mode of operation.
In this manner, state selector circuit 154 may vary the resolution
of changes made to state signal U.sub.ST 160 and thus, may vary the
rate of change in power delivery over multiple feedback periods to
load 122 in response to the feedback information in accordance with
the teachings of the present invention. For example, if the
feedback information indicates to state selector circuit 154 that
the rate of change (i.e., rate of increase or rate of decrease) in
power delivery over multiple feedback periods to load 122 should be
greater, state selector circuit 154 may operate in the coarse mode
and update state signal U.sub.ST 160, which therefore updates the
operating condition of power switch S1 170 by a greater amount in
accordance with the teachings of the present invention.
[0046] In the depicted example, the feedback information may
include a first information that may represent a difference between
a portion of a feedback period that feedback signal U.sub.FB 158 is
less than a threshold value, and a portion of the feedback period
that feedback signal U.sub.FB 158 is greater than the threshold
value. The feedback information may also include a second
information that may represent a difference between a portion of a
feedback period that feedback signal U.sub.FB 158 is less than a
lower limit and a portion of the feedback period that feedback
signal U.sub.FB 158 is greater than an upper limit. In one example,
the threshold value may represent a desired level of regulated
output current I.sub.O 118 at the output of power supply 100. The
lower limit may represent a level of output current I.sub.O 118
that is below the desired level (e.g., 10% below the desired level)
and the upper limit may represent a level of output current I.sub.O
118 that is above the desired level (e.g., 10% above the desired
level). In some applications, the first information may represent a
difference between an estimated average value of feedback signal
U.sub.FB 158 and the threshold value. In one example, state
selector circuit 154 may use the first information and/or the
second information to determine an operational state and hence, set
an operating condition of power switch S1 170 accordingly.
[0047] For example, if the portion of a feedback period that
feedback signal U.sub.FB 158 is less than the threshold value is
greater than the portion of the feedback period that feedback
signal U.sub.FB 158 is greater than the threshold value by a
certain amount, state selector circuit 154 may determine that the
power delivery to load 122 per unit time should be increased by a
greater amount to more quickly bring output current I.sub.O 118
closer to the desired level. In this case, state selector circuit
154 may operate in the coarse mode. However, once the portion of a
feedback period that feedback signal U.sub.FB 158 is less than the
threshold value approaches the portion of the feedback period that
feedback signal U.sub.FB 158 is greater than the threshold value
(e.g., the difference between the portion of a feedback period that
feedback signal U.sub.FB 158 is less than the threshold value and
the portion of the feedback period that feedback signal U.sub.FB
158 is greater than the threshold value falls within a certain
range), state selector circuit 154 may determine that changes to
the power delivery to load 122 should be made with finer
resolution. In this case, state selector circuit 154 may operate in
the fine mode. In this way, controller 150 can be configured to
respond more rapidly to larger transients and remain less
responsive to smaller disturbances at the input and/or the output
of power supply 100 in accordance with the teachings of the present
invention.
[0048] FIG. 2A shows a circuit diagram illustrating one example of
the state selector circuit 154 of the controller 150 in FIG. 1A
with increased detail in accordance with the teachings of the
present invention. It should be appreciated that similarly named
and numbered elements referenced below are coupled and function as
described above. As shown, state selector circuit 154 includes a
feedback signal processor 280 that is coupled to receive feedback
signal U.sub.FB 158, a sampling signal U.sub.SMP 272, and a
feedback period signal U.sub.PER 262. In response, feedback signal
processor 280 outputs a first count signal U.sub.CN1 222
corresponding to the first information and a second count signal
U.sub.CN2 232 corresponding to the second information.
[0049] As will be shown, in one example, during a feedback period
that is demarcated by feedback period signal U.sub.PER 262,
feedback signal processor 280 compares feedback signal U.sub.FB 158
with a threshold U.sub.TH 205, an upper limit U.sub.UP 201, and a
lower limit U.sub.LO 203 at a sampling frequency that is determined
by the frequency of sampling signal U.sub.SMP 272. Feedback signal
processor 280 updates first count signal U.sub.CN1 222 and second
count signal U.sub.CN2 232 based on the results of the comparisons
during the feedback period. In the illustrated example, threshold
U.sub.TH 205 corresponds to the threshold value (i.e., desired
value of output current I.sub.O 118), lower limit U.sub.LO 203
corresponds to the lower limit, and upper limit U.sub.UP 201
corresponds to the upper limit.
[0050] The example of FIG. 2A shows state selector 154 including a
feedback period signal generator 260 that generates feedback period
signal U.sub.PER 262 and a sampling clock generator 270 that
generates sampling signal U.sub.SMP 272. In one example, feedback
period signal generator 260 may output a pulse at set intervals
(i.e., a periodic pulse with a certain period) as feedback period
signal U.sub.PER 262. Each one of the intervals (i.e., the period
of feedback period signal U.sub.PER 262) demarcates a feedback
period. Stated differently, feedback period signal U.sub.PER 262
can indicate the beginning and the end of a feedback period. In one
example, the period of feedback period signal U.sub.PER 262 (i.e.,
the feedback period) may be equal to one half of the period the ac
line voltage, which may be several times (e.g., 512) greater than
the period of sampling signal U.sub.SMP 272 (i.e., the sampling
period). In other words, feedback signal processor 280 may update
first count signal U.sub.CN1 222 and second count signal U.sub.CN2
232 512 times during every feedback period. Additionally, feedback
signal processor 280 may set first count signal U.sub.CN1 222 to a
first initial value and second count signal U.sub.CN2 232 to a
second initial value at the beginning of each feedback period. In
the depicted example, the first initial value and the second
initial value may be the same and equal to zero. In some cases, the
first initial value may be different from the second initial
value.
[0051] As further illustrated in the example depicted in FIG. 2A,
feedback signal processor 280 includes a first counter Counter1
220, a second counter Counter2 230, a logic circuit 210, and
comparators 202, 204, and 206. First counter Counter1 220 is
coupled to receive sampling signal U.sub.SMP 272 at its CLK input,
feedback period signal U.sub.PER 262 at its RESET input, an output
207 of comparator 206 at its UP/DN input. First counter Counter1
220 is also coupled to output first count signal U.sub.CN1 222. In
operation, first counter Counter1 220 updates first count signal
U.sub.CN1 222 by counting up or down in response to output 207
during every sampling period. In one example, first counter
Counter1 220 counts up if output 207 is logic low and counts down
if output 207 is logic high. Comparator 206 is coupled to receive
feedback signal U.sub.FB 158 and set output 207 to logic high or
logic low in response to a comparison of feedback signal U.sub.FB
158 with threshold U.sub.TH 205. In one example, comparator 206 may
set output 207 to logic high if feedback signal U.sub.FB 158 is
greater than threshold U.sub.TH 205, and set output 207 to logic
low if feedback signal U.sub.FB 158 is less than threshold U.sub.TH
205. First counter Counter1 220 may increase first count signal
U.sub.CN1 222 by counting up when feedback signal U.sub.FB 158 is
less than threshold U.sub.TH 205 during a sampling period and
similarly, may decrease first count signal U.sub.CN1 222 by
counting down when feedback signal U.sub.FB 158 is greater than
threshold U.sub.TH 205 during a sampling period. In this way, first
counter Counter1 220 may output as first count signal U.sub.CN1 222
a signal that may be representative of the difference between a
portion of a feedback period that feedback signal U.sub.FB 158 is
less than threshold U.sub.TH 205 and a portion of the feedback
period that feedback signal U.sub.FB 158 is greater than threshold
U.sub.TH 205.
[0052] Continuing with the example depicted in FIG. 2A, second
counter Counter2 230 is coupled to receive sampling signal
U.sub.SMP 272 at its CLK input, feedback period signal U.sub.PER
262 at its RESET input, output 207 of comparator 206 at its UP/DN
input, an output 213 of logic circuit 210 at its EN input. Second
counter Counter2 230 is further coupled to output second count
signal U.sub.CN2 232. In operation, second counter Counter2 230
updates second count signal U.sub.CN2 232 by counting up or down in
response to output 207 during every sampling period if output 213
indicates that second counter 230 should be enabled, and maintains
second count signal U.sub.CN2 232 at the same value if output 213
indicates that second counter Counter2 230 should be disabled. In
one example, logic circuit 210 may be a two-input XOR gate coupled
to receive an output 211 of comparator 202 and an output 209 of
comparator 204 as inputs. As such, output 213 may be logic low
indicating that second counter Counter2 230 should be disabled when
output 211 and 209 are both logic low or are both logic high.
Output 213 may be logic high indicating that second counter
Counter2 230 should be enabled when only one of outputs 209 and 211
is logic high and the other one of outputs 209 and 211 is logic
low.
[0053] In the depicted example, comparators 202 and 204 are coupled
to receive feedback signal U.sub.FB 158 and set outputs 211 and 209
in response to the comparisons of feedback signal U.sub.FB 158 with
upper limit U.sub.UP 201 and with lower limit U.sub.LO 203,
respectively. Specifically, comparator 202 sets output 211 to logic
high if feedback signal U.sub.FB 158 is less than upper limit
U.sub.UP 201 and to logic low if feedback signal U.sub.FB 158 is
greater than upper limit U.sub.UP 201. Similarly, comparator 204
sets output 209 to logic high if feedback signal U.sub.FB 158 is
greater than lower limit U.sub.LO 203 and to logic low if feedback
signal U.sub.FB 158 is less than lower limit U.sub.LO 203. In other
words, second counter Counter2 230 is enabled to count up or down
in response to output 207 when feedback signal U.sub.FB 158 is
greater than upper limit U.sub.UP 201 or less than lower limit
U.sub.LO 203 during a sampling period. When feedback signal
U.sub.FB 158 is between upper limit U.sub.UP 201 and lower limit
U.sub.LO 203, however, second counter Counter2 230 is disabled and
keeps second count signal U.sub.CN2 232 constant.
[0054] In the example, when enabled, second counter Counter2 230
increases second count signal U.sub.CN2 232 by counting up if
output 207 is logic low and decreases second count signal U.sub.CN2
232 by counting down if output 207 is logic high. That is, second
counter Counter2 230 counts up when feedback signal U.sub.FB 158 is
less than lower limit U.sub.LO 203 during a sampling period. Second
counter Counter2 230 counts down when feedback signal U.sub.FB 158
is greater than upper limit U.sub.UP 201 during a sampling period.
In this way, second counter Counter2 230 may output as second count
signal U.sub.CN2 232 a signal that may be representative of the
difference between a portion of a feedback period that feedback
signal U.sub.FB 158 is less than lower limit U.sub.LO 203 and a
portion of the feedback period that feedback signal U.sub.FB 158 is
greater than upper limit U.sub.UP 201. In one example, both first
counter Counter1 220 and second counter Counter2 230 are configured
to have a maximum output count that is representative of a length
of time corresponding to a feedback period. The magnitudes of first
count signal U.sub.CN1 222 and second count signal U.sub.CN2 232
cannot exceed the maximum output count.
[0055] As further shown, state selector circuit 154 also includes a
decision circuit 240 and a state counter 250. Decision circuit 240
is coupled to receive first count signal U.sub.CN1 222 and second
count signal U.sub.CN2 232 from feedback signal processor 280 and
output a direction signal U.sub.DIR 244 and a mode signal U.sub.MD
242. In one example, direction signal U.sub.DIR 244 may be a
one-bit digital signal indicative of the direction of change in
state signal U.sub.ST 160 and mode signal U.sub.MD 242 may be a
two-bit digital signal indicative of an operation mode of state
selector circuit 154. In operation, decision circuit 240 may set
direction signal U.sub.DIR 244 to zero or one in response to first
count signal U.sub.CN1 222, and mode signal U.sub.MD 242 to one of
zero (i.e., 00 binary), one (i.e., 01 binary), and two (i.e., 10
binary) in response to both first count signal U.sub.CN1 222 and
second count signal U.sub.CN2 232. In the illustrated example,
decision circuit 240 may set direction signal U.sub.DIR 244 to one
indicating that state signal U.sub.ST 160 should be increased when
first count signal U.sub.CN1 222 is positive and set direction
signal U.sub.DIR 244 to zero indicating that state signal U.sub.ST
160 should be decreased when first count signal U.sub.CN1 222 is
negative.
[0056] Furthermore, decision circuit 240 may set mode signal
U.sub.MD 242 to zero, which may correspond to a coarse mode as an
operation mode if the magnitude of first count signal U.sub.CN1 222
is greater than a value X, the magnitude of second count signal
U.sub.CN2 232 is greater than a value Y, and both first count
signal U.sub.CN1 222 and second count signal U.sub.CN2 232 have the
same sign (i.e., both signals are either positive or negative).
Decision circuit 240 may set mode signal U.sub.MD 242 to one, which
may correspond to a medium mode as an operation mode if the
magnitude of first count signal U.sub.CN1 222 is greater than the
value X, the magnitude of second count signal U.sub.CN2 232 is
between the value Y and a value Z (less than the value Y), and both
first count signal U.sub.CN1 222 and second count signal U.sub.CN2
232 have the same sign. Decision circuit 240 may set mode signal
U.sub.MD 242 to two, which may correspond to a fine mode as an
operation mode if the magnitude of first count signal U.sub.CN1 222
is less than the value X, or the magnitude of second count signal
U.sub.CN2 232 is less than the value Z, or first count signal
U.sub.CN1 222 and second count signal U.sub.CN2 232 have different
signs. In one example, the value X represents a length of time that
corresponds to 5% of a feedback period (i.e., the value X is equal
to 5% of the maximum output count). In some cases, the value X may
also correspond to 5% of the maximum value of the difference
between an estimated average value of feedback signal U.sub.FB 158
and threshold U.sub.TH 205. The value Y represents a length of time
that corresponds to 20% of a feedback period (i.e., the value Y is
equal to 20% of the maximum output count) and the value Z
represents a length of time that corresponds to 10% of a feedback
period (i.e., the value Z is equal to 10% of the maximum output
count).
[0057] As shown in the depicted example, state counter 250 is
coupled to receive feedback period signal U.sub.PER 262 at its
UPDATE input, direction signal U.sub.DIR 244, mode signal U.sub.MD
242, first count signal U.sub.CN1 222, second count signal
U.sub.CN2 232. State counter is also coupled to output state signal
U.sub.ST 160. In operation, state counter 250 may update state
signal U.sub.ST 160 in response to the values of direction signal
U.sub.DIR 244 and mode signal U.sub.MD 242 at the time that a new
pulse in feedback period signal U.sub.PER 262 is received, which is
indicative of the end of a presently occurring feedback period
(i.e., the beginning of a new feedback period). For example, when
the value of mode signal U.sub.MD 242 at the end of a presently
occurring feedback period is zero, state selector circuit 154 is
set to operate in the coarse mode. When state selector circuit 154
is in the coarse mode, state counter 250 may update (increase or
decrease) the sixth bit of state signal U.sub.ST 160 (i.e.,
0000100000 binary) in response to the value of direction signal
U.sub.DIR 244 at the end of the presently occurring feedback
period. That is, when state selector circuit 154 is operating in
the coarse mode, state counter 250 may increase or decrease state
signal U.sub.ST 160 by 32 for the next feedback period based on the
value of direction signal U.sub.DIR 244 at the end of the presently
occurring feedback period. In one example, state counter 250
increases state signal U.sub.ST 160 if direction signal U.sub.DIR
244 is one and decreases state signal U.sub.ST 160 if direction
signal U.sub.DIR 244 is zero.
[0058] In another example, when the value of mode signal U.sub.MD
242 at the end of a presently occurring feedback period is one,
state selector circuit 154 is set to operate in the medium mode.
When state selector circuit 154 is in the medium mode, state
counter 250 may update the fourth bit of state signal U.sub.ST 160
(i.e., 0000001000 binary) in response to the value of direction
signal U.sub.DIR 244 at the end of the presently occurring feedback
period. That is, when state selector circuit 154 is operating in
the coarse mode, state counter 250 may increase or decrease state
signal U.sub.ST 160 by eight for the next feedback period based on
the value of direction signal U.sub.DIR 244 at the end of the
presently occurring feedback period. In one example, state counter
250 increases state signal U.sub.ST 160 if direction signal
U.sub.DIR 244 is one and decreases state signal U.sub.ST 160 if
direction signal U.sub.DIR 244 is zero.
[0059] In yet another example, when the value of mode signal
U.sub.MD 242 at the end of a presently occurring feedback period is
two, state selector circuit 154 is set to operate in the fine mode.
When state selector circuit 154 is in the fine mode, state counter
250 may update the first bit (least significant bit) of state
signal U.sub.ST 160 (i.e., 0000000001 binary). In the depicted
example, when state selector circuit 154 is operating in the fine
mode, state counter 250 is configured to update state signal
U.sub.ST 160 only if direction signal U.sub.DIR 244 and mode signal
U.sub.MD 242 maintain their values, and the final values of both
first count signal U.sub.CN1 222 and second count signal U.sub.CN2
232 at the end of a feedback period are above a minimum threshold
count (e.g., 3) for a certain number (e.g., 12) of consecutive
feedback periods. In one example, if this set of conditions is met
and direction signal U.sub.DIR 244 is one, state counter 250
increases state signal U.sub.ST 160 by one. If this set of
conditions is met but direction signal U.sub.DIR 244 is zero, state
counter 250 decreases state signal U.sub.ST 160 by one. Conversely,
if this set of conditions is not met such that direction signal
U.sub.DIR 244 changes and/or at least one of the final values of
first count signal U.sub.CN1 222 and second count signal U.sub.CN2
232 drops below the minimum threshold count, state counter 250
keeps state signal U.sub.ST 160 unchanged. For example, state
counter 250 may include a counter that only operates when state
selector circuit 154 is in the fine mode. The counter may be
configured to start counting from one when state selector circuit
154 enters the fine mode and count up at the end of each feedback
period if direction signal U.sub.DIR 244 maintains its value, and
the final values of both first count signal U.sub.CN1 222 and
second count signal U.sub.CN2 232 at the end of a feedback period
are above the minimum threshold count. If direction signal
U.sub.DIR 244 changes its value and/or at least one of the final
values of first count signal U.sub.CN1 222 and second count signal
U.sub.CN2 232 drops below the minimum threshold count, then the
counter is reset to one and state signal U.sub.ST 160 remains
unchanged. If direction signal U.sub.DIR 244 maintains its value,
and the final values of both first count signal U.sub.CN1 222 and
second count signal U.sub.CN2 232 remain above the minimum
threshold count for 12 consecutive feedback periods (i.e., if the
counter output reaches 12), state counter 250 updates state signal
U.sub.ST 160 based on the value of direction signal U.sub.DIR 244
and sets the counter output back to one. It is in this manner that
state selector circuit 154 can adjust the rate of change in the
operating condition of power switch S1 170 (hence, the rate of
change in power delivery to load 122) in response to the feedback
information in accordance with the teachings of the present
invention.
[0060] FIG. 2B shows a circuit diagram illustrating another example
of the state selector circuit 154 of the controller 150 in FIG. 1A
in accordance with the teachings of the present invention. It is
noted that state selector circuit 154 in FIG. 2B shares
similarities with state selector circuit 154 in FIG. 2A. It should
be appreciated that similarly named and numbered elements
referenced below are coupled and function as described above. One
difference between state selector circuit 154 in FIG. 2B and state
selector circuit 154 in FIG. 2A is that feedback period signal
generator 260 in FIG. 2B includes a one-shot circuit 264 coupled to
receive output 207 from comparator 206, and is therefore coupled to
generate feedback period signal U.sub.PER 262 in response to the
comparison of feedback signal U.sub.FB 158 with threshold U.sub.TH
205. One-shot circuit 264 is coupled to output a pulse in feedback
period signal U.sub.PER 262 when output 207 transitions from logic
high to logic low. In other words, feedback period signal generator
260 may output a pulse when feedback signal U.sub.FB 158 falls from
a level above threshold U.sub.TH 205 to a level below threshold
U.sub.TH 205. As such, each feedback period may correspond to a
length of time between consecutive instances of feedback signal
U.sub.FB 158 falling below threshold U.sub.TH 205. Additionally,
feedback period signal generator 260 may be configured to output a
pulse to terminate a feedback period if feedback signal U.sub.FB
158 does not fall below threshold U.sub.TH 205 within a certain
length of time (e.g., a timeout period) from the start of the
feedback period. For example, during startup, energy at the output
of power supply 100 may be at a level such that feedback signal
U.sub.FB 158 is below threshold U.sub.TH 205. In this case,
feedback period signal generator 260 may output a pulse every
timeout period to indicate that a presently occurring feedback
period has ended and a new feedback period has begun. That is, each
feedback period may be equal to the timeout period.
[0061] FIG. 2C shows a circuit diagram illustrating yet another
example of the state selector circuit 154 of the controller 150 in
FIG. 1A in accordance with the teachings of the present invention.
It is noted that selector circuit 154 in FIG. 2C shares
similarities with state selector circuit 154 in FIG. 2A. It should
be appreciated that similarly named and numbered elements
referenced below are coupled and function as described above. One
difference between state selector circuit 154 in FIG. 2C and state
selector circuit 154 in FIG. 2A is that feedback period signal
generator 260 in FIG. 2C includes a comparator 266 coupled to
receive input sense signal U.sub.INS 142 and a one-shot circuit 264
coupled to comparator 266 to output feedback period signal
U.sub.PER 262 in response to the output of comparator 266.
[0062] Specifically, in one example, comparator 266 may compare
input sense signal U.sub.INS 142 with a zero condition threshold
U.sub.ZC, which may be representative of a zero crossing threshold
for input voltage V.sub.IN 102. In response, comparator 266 may
output a logic high or logic low signal. One-shot circuit 264 may
be coupled to output a pulse in feedback period signal U.sub.PER
262 when the signal at the output of comparator 266 transitions
from logic high to logic low. In one example, comparator 266
outputs a logic high signal if input sense signal U.sub.INS 142 is
greater than zero condition threshold U.sub.ZC and a logic low
signal if input sense signal U.sub.INS 142 is less than or equal to
zero condition threshold U.sub.ZC. In other words, feedback period
signal generator 260 in FIG. 2C may output a pulse in feedback
period signal U.sub.PER 262 when input sense signal U.sub.INS 142
crosses zero condition threshold U.sub.ZC from a level that is
above the threshold to a level that is below the threshold, which
may also be referred to as a zero crossing event. As such, each
feedback period may correspond to a length of time between
consecutive zero crossing events.
[0063] FIG. 3A shows an example set of waveforms illustrating the
operation of the controller 150 in FIG. 1A in accordance with the
teachings of the present invention. In the illustrated example,
waveform 302 is a rectified periodic signal with a period T.sub.P
310 and may be representative of input voltage V.sub.IN 102.
Waveform 313 is a periodic signal with period T.sub.P 310 and may
be representative of input current I.sub.IN 113. As shown in the
example, waveform 313 is in phase with and proportional to waveform
302. Waveform 358 is an example waveform representative of feedback
signal U.sub.FB 158. In one example, waveform 358 may be
representative of output current I.sub.O 118 that is provided to
load 122 of power supply 100. As illustrated, waveform 358 is phase
shifted with respect to waveform 313. This could be due to output
capacitor C1 116 phase shifting output current I.sub.O 118 with
respect to input current I.sub.IN 113. It should be noted that
despite being phase shifted, waveform 358 may still be periodic
with a period T.sub.FB that is substantially equal to the period of
waveform 302 (i.e., period T.sub.P 310).
[0064] Further illustrated in FIG. 3A are waveform 362 and waveform
372. Waveform 362 may be representative of feedback period signal
U.sub.PER 262 of FIGS. 2A-2C, and waveform 372 may be
representative of sampling signal U.sub.SMP 272 of FIGS. 2A-2C. In
the illustrated example, waveform 362 includes pulses that are
generated in response to waveform 358 falling from a level above a
value 305 to a level below value 305. Value 305 may correspond to
threshold U.sub.TH 205 of FIGS. 2A-2C. Accordingly, in one example,
the length of time between consecutive pulses in waveform 362 may
correspond to a feedback period. For example, time points t.sub.n,
t.sub.n+1, and t.sub.n+2 may be the start and end points of
consecutive feedback periods, with time point t.sub.n indicating
the start of the n.sup.th feedback period, time point t.sub.n+1
indicating the end of the n.sup.th feedback period and the start of
the (n+1).sup.th feedback period, and time point t.sub.n+2
indicating the end of the (n+1).sup.th feedback period and the
start of the (n+2).sup.th feedback period.
[0065] In the illustrated example, waveform 322 and 332 are
representative of first count signal T.sub.CN1 222 and second count
signal U.sub.CN2 232 of FIGS. 2A-2C, respectively, and may be
updated every period of waveform 372 (i.e., every sampling period).
Specifically, waveform 322 may be incremented if waveform 358 is
less than value 305, and may be decremented if waveform 358 is
greater than or equal to waveform 305. Waveform 332 may be
incremented if waveform 358 is less than a value 303, which is
representative of lower limit U.sub.LO 203, may be decremented if
waveform 358 is greater than a value 301, which is representative
of upper limit U.sub.UP 201, and may be kept unchanged if waveform
358 is between value 301 and value 303.
[0066] FIG. 3A also illustrates a waveform 360, which may be
representative of state signal U.sub.ST 160. Final values of
waveforms 322 and 332 at the end of a feedback period may be used
to set the values of direction signal U.sub.DIR 244 and mode signal
U.sub.MD 242 and in turn, update waveform 360 to adjust the
operating condition of power switch S1 170 for the next feedback
period. For instance, waveform 360 may be updated by a different
amount depending on the set of conditions that is met by the final
values of waveforms 322 and 332 during a feedback period. In
addition, waveforms 322 and 332 may be set to an initial value such
as, for example, zero at the beginning of each feedback period
(e.g., at time points t.sub.n, t.sub.n+1, and t.sub.n+2). This
ensures that the final values of waveforms 322 and 332 at the end
of a feedback period reflect only the feedback information gathered
during that feedback period.
[0067] In the depicted example, at the end of the n.sup.th feedback
period, the final value of waveform 322 is positive and greater
than the value X, and final value of waveform 332 is positive and
greater than the value Y. Under these conditions, in one example,
direction signal U.sub.DIR 244 is set to one indicating that state
signal U.sub.ST 160 should be increased and mode signal U.sub.MD
242 is set to zero indicating that state selector circuit 154
should operate in the coarse mode. As a result, waveform 360 is
increased by 32 from a value K to a value (K+32) for (n+1).sup.th
feedback period. In the depicted example, the same set of
conditions is met by waveforms 322 and 332 at the end of
(n+1).sup.th feedback period and thus, waveform 360 is again
increased by 32 from value (K+32) to (K+64) for the next feedback
period starting at time point t.sub.n+2.
[0068] FIG. 3B shows another example set of waveforms illustrating
the operation of controller 150 of FIG. 1A in accordance with the
teachings of the present invention. One difference between waveform
358 in FIG. 3B and waveform 358 in FIG. 3A is that the final values
of waveforms 322 and 332 in waveform 358 in FIG. 3B at the end of
n.sup.th feedback period and (n+1).sup.th feedback period satisfy a
different set of conditions. Specifically, as shown in the example
depicted in FIG. 3B at the end of n.sup.th feedback period, the
final value of waveform 322 is positive and greater than the value
X, and the final value of waveform 332 is positive and between the
value Y and the value Z. In one example, the value Z is less than
the value Y. Under these conditions, direction signal U.sub.DIR 244
is set to one indicating that state signal U.sub.ST 160 should be
increased and mode signal U.sub.MD 242 is set to one indicating
that state selector circuit 154 should operate in the medium mode.
As a result, waveform 360 is increased by eight from value K to a
value (K+8) for (n+1).sup.th feedback period. In the depicted
example, the same set of conditions holds at the end of
(n+1).sup.th feedback period. Therefore, waveform 360 is again
increased by eight from value (K+8) to a value (K+16) for the next
feedback period starting at time point t.sub.n+2.
[0069] FIG. 3C shows yet another example set of waveforms
illustrating the operation of controller 150 of FIG. 1A in
accordance with the teachings of the present invention. One
difference between waveform 358 of FIG. 3C and waveform 358 of FIG.
3A and FIG. 3B is that the final values of waveforms 322 and 332 in
FIG. 3C at the end of the feedback periods from the n.sup.th
feedback period through (n+11).sup.th feedback period (not shown)
satisfy a different set of conditions. Specifically, at the end of
the n.sup.th feedback period, the final value of waveform 322 is
positive and less than the value X, and the final value of waveform
332 is positive and less than the value Z. Under these conditions,
direction signal U.sub.DIR 244 is set to one indicating that state
signal U.sub.ST 160 should be increased and mode signal U.sub.MD
242 is set to two indicating that state selector circuit 154 should
operate in the fine mode.
[0070] In one example, when operating in the fine mode, state
selector circuit 154 updates state signal U.sub.ST 160 only if the
values of direction signal U.sub.DIR 244 and mode signal U.sub.MD
242 remain the same, and both first count signal U.sub.CN1 222 and
second count signal U.sub.CN2 232 are above the minimum threshold
count for a certain number (e.g., 12) of consecutive feedback
periods. In the illustrated example, the final values of waveforms
322 and 332 remain positive and below the value X and the value Z,
respectively from n.sup.th feedback period through (n+11).sup.th
feedback period such that the values of direction signal U.sub.DIR
244 and mode signal U.sub.MD 242 remain at one and two,
respectively for 12 consecutive feedback periods. Additionally, the
values of first count signal U.sub.CN1 222 and second count signal
U.sub.CN2 232 at the end of each feedback period remain above the
minimum threshold count during this time. As a result, waveform 360
is increased by one from value K to a value (K+1) for (n+12).sup.th
feedback period starting at time point t.sub.n+12.
[0071] FIG. 4A shows a flow diagram illustrating an example process
400 for adjusting the operating condition of a switch of a power
supply in accordance with the teachings of the present invention.
It is noted that process 400 may be performed by a circuit similar
or identical to example state selector circuit 154 in FIG. 2A, FIG.
2B and in FIG. 2C in accordance with the teachings of the present
invention. In the depicted example, process 400 may begin at block
401. At block 403, a new feedback period may be started. In one
example, the feedback period may be started in response to a pulse
in an indicator signal (e.g., feedback period signal U.sub.PER
262). For example, feedback period signal generator 260 may output
pulses at set intervals as the indicator signal and each pulse may
indicate the start of a new feedback period. Alternatively,
feedback period signal generator 260 may output a pulse in response
to feedback signal U.sub.FB 158 falling below a threshold (e.g.,
threshold U.sub.TH 205). At block 405, a first count signal (e.g.,
first count signal U.sub.CN1 222) and a second count signal (e.g.,
second count signal U.sub.CN2 232) may be set to an initial value
(e.g., zero). In one example, the first count signal may be
representative of the difference between a portion of a feedback
period that feedback signal U.sub.FB 158 is less than the threshold
and a portion of the feedback period that feedback signal U.sub.FB
158 is greater than the threshold. The first count signal may be
generated by a counter (e.g., first counter Counter1 220). The
second count signal may be representative of the difference between
a portion of a feedback period that feedback signal U.sub.FB 158 is
less than a lower limit (e.g., lower limit U.sub.LO 203) and a
portion of the feedback period that feedback signal U.sub.FB 158 is
greater than an upper limit (e.g., upper limit U.sub.UP 201). The
second count signal may be generated by another counter (e.g.,
second counter Counter2 230). The first and the second count
signals may be set to the initial value when the counters are reset
in response to a pulse in the indicator signal.
[0072] At block 407, a new feedback sample (i.e., a new sample of
feedback signal U.sub.FB 158) may be obtained. In one example, each
feedback sample may be representative of the value of feedback
signal U.sub.FB 158 during a corresponding sampling period. Each
sampling period may be equal to the period of a clock signal (e.g.,
sampling signal U.sub.SMP 272) generated by a clock generator
(e.g., sampling clock generator). In one example, the period of the
clock signal may be several times (e.g., 512) smaller than the
feedback period. In other words, feedback signal U.sub.FB 158 may
be sampled several times (e.g., 512) during a feedback period.
[0073] At block 409, the feedback sample may be compared with the
threshold, the lower limit and the upper limit. Then, the first and
the second count signals may be updated based on these comparisons.
More particularly, the first count signal may be changed in
response to the comparison of the feedback sample with the
threshold and the second count signal may be changed in response to
the comparisons of the feedback sample with both the upper limit
and the lower limit.
[0074] FIG. 4B shows a flow diagram illustrating one example of
detailed steps that may occur in process block 409 in accordance
with the teachings of the present invention. For instance, examples
of these comparisons and the resulting changes in the first and the
second signals discussed on process block 409 of FIG. 4A are shown
in detail. Referring now to FIG. 4B, at block 412, the feedback
sample is compared with the threshold. If the sample is greater
than the threshold, process 409 proceeds to block 414 where the
first count signal is decremented. In one example, first counter
Counter1 220 may decrement first count signal U.sub.CN1 222 by
counting down. If the sample is not greater than the threshold,
process 409 proceeds block 416 where the first count signal is
incremented. In one example, first counter Counter1 220 increments
first count signal U.sub.CN1 222 by counting up.
[0075] At block 418, the feedback sample is compared with the upper
limit. If the sample is greater than the upper limit, process 409
proceeds to block 422 where the second count signal is decremented.
In one example, second counter Counter2 230 decrements second count
signal U.sub.CN2 232 by counting down. If the sample is not greater
than the upper limit, process 409 proceeds to block 420 where the
sample is compared with the lower limit. If the sample is less than
the lower limit, process 409 proceeds to block 424 where the second
count signal is incremented. In one example, second counter
Counter2 230 increments second count signal U.sub.CN2 232 by
counting up. If the sample is not less than the lower limit,
process 409 proceeds to block 426 where the second count signal is
kept unchanged.
[0076] Referring now back to FIG. 4A, at block 411, it may be
determined whether or not the feedback period has ended. In one
example, the feedback period may end in response to a new pulse in
the indicator signal. If the feedback period has not ended, process
400 returns to block 407. Otherwise, if the feedback period has
ended, process 400 proceeds to block 413.
[0077] At block 413, values of a direction signal (e.g., direction
signal U.sub.DIR 244) and a mode signal (e.g., mode signal U.sub.MD
242) are set in response to the final values of the first and the
second count signals at the end of the feedback period. In one
example, state selector circuit 154 may include a decision circuit
(e.g., decision circuit 240) coupled to receive the first and the
second count signals and set the values of direction signal and the
mode signal in response to the final values of the first and the
second count signals.
[0078] FIG. 4C shows a flow diagram illustrating one example of
detailed steps that may occur in process block 413 in accordance
with the teachings of the present invention. For instance, various
different conditions on the first and the second count signals, and
the corresponding values of the direction signal and the mode
signal are shown in detail. Referring now to FIG. 4C, at block 440,
it may be determined whether or not the magnitude of the first
count signal is greater than the value X, whether or not the
magnitude of the second count signal is greater than the value Y,
and whether or not both signals have the same sign. If these
conditions are met, process 413 proceeds to block 442. If not,
process 413 proceeds to block 444.
[0079] At block 442, the direction signal is set in response to the
sign of the first signal and the mode signal is set to a value that
indicates the coarse mode as the operation mode of state selector
circuit 154. For example, decision circuit 240 may set mode signal
U.sub.MD 242 to zero and direction signal U.sub.DIR 244 to zero if
the first signal is positive, and to one if the first signal is
negative.
[0080] At block 444, it may be determined whether or not the
magnitude of the first signal is greater than the value X, whether
or not the magnitude of the second signal is less than the value Y
but greater than the value Z, and whether or not both signals have
the same sign. If these conditions are met, process 413 proceeds to
block 446 where the direction signal is set in response to the sign
of the first signal and the mode signal is set to a value that
indicates the medium mode as the operation mode of state selector
circuit 154. If the conditions at block 444 are not met, process
413 proceeds to block 448 where the direction signal is set in
response to the sign of the first signal and the mode signal is set
to a value that indicates the fine mode as the operation mode of
state selector circuit 154.
[0081] Referring now back to FIG. 4A, at block 415, the operating
condition of a power switch (e.g., power switch S1 170) of a
switched-mode power supply may be adjusted in response to the
direction signal, to the mode signal, and to at least one of the
final values of the first count signal and the second count signal
at the end of a feedback period. For example, state selector
circuit 154 may change state signal U.sub.ST 160 based on the
values of direction signal U.sub.DIR 244 and mode signal U.sub.MD
242. This may change a drive signal (e.g., drive signal U.sub.DR
162) coupled to drive switching of the power switch, thereby
causing an operating condition of power switch S1 170 such as
switching frequency and/or on-time to change. As previously
mentioned, state selector circuit 154 may change state signal
U.sub.ST 160 by an amount that varies based on the operation mode
of state selector circuit 154. Therefore, state selector circuit
154 may vary the rate of change in the operating condition of power
switch S1 170 over multiple feedback periods. In addition, state
selector circuit 154 may keep state signal U.sub.ST 160 unchanged
based on direction signal U.sub.DIR 244 and at least one of the
final value of the first count signal and the final value of the
second count signal. For example, when state selector circuit 154
is operating in the fine mode, if direction signal U.sub.DIR 244
changes and/or at least one of the final value of the first count
signal and the final value of the second count signal drops below
the minimum threshold count during a certain number of consecutive
feedback periods, state selector circuit 154 may keep state signal
U.sub.ST 160 unchanged. In some cases, when state signal U.sub.ST
160 increases, switching frequency of power switch S1 170 may
increase resulting in more power to be delivered to load 122 and
conversely, when state signal U.sub.ST 160 decreases, switching
frequency of power switch S1 170 may decrease resulting in less
power to be delivered to load 122.
[0082] FIG. 5 is a schematic circuit diagram illustrating an
example driver circuit including one example of a modulated DAC
according to the teachings of the present invention. As shown,
driver circuit 156 can be coupled to receive a digital input signal
U.sub.IN 590 comprising (Q+P) bits, where Q and P are non-zero
integers. Driver circuit 156 can be coupled to output drive signal
U.sub.DR 162 in response to digital input signal U.sub.IN 590 to
control the switching of power switch S1 170. In one example,
digital input signal U.sub.IN 590 may correspond to state signal
U.sub.ST 160 of the controller in FIG. 1A. In this disclosure,
upper Q bits of digital input signal U.sub.IN 590 (i.e., bits
B.sub.1 through B.sub.Q) are referred to as "baseline bits" with
bit B.sub.Q being the most significant bit and bit B.sub.1 being
the least significant bit. Lower P bits of digital input signal
U.sub.IN 590 (i.e., bits M.sub.1 through M.sub.P) are referred to
as "modulation bits" with bit M.sub.P being the most significant
bit and bit M.sub.1 being the least significant bit.
[0083] Further illustrated in the example driver circuit is a
modulated DAC 540 coupled to receive digital input signal U.sub.IN
590 and a clock signal U.sub.CLK 512. In the illustrated example,
clock signal U.sub.CLK 512 is generated by a clock signal generator
510 that is included in driver circuit 156. In one example, clock
signal U.sub.CLK 512 is a periodic signal with a fixed period. In
response to digital input signal U.sub.IN 590 and clock signal
U.sub.CLK 512, modulated DAC 540 outputs an adjust signal U.sub.ADJ
580 to a drive signal generator 520. Modulated DAC 540 can include
a modulator 530 coupled to receive clock signal U.sub.CLK 512 and
the modulation bits. In response, modulator 530 generates a
modulation period signal UMP 538. In one example, modulation period
signal UMP 538 is a periodic signal that alternates between logic
high and logic low. The period of modulation period signal
U.sub.MDP 538 may be proportional to the number of possible values
(i.e., 2.sup.P) that can be represented by the modulation bits.
[0084] In the depicted example, modulator 530 includes a counter
532 and a comparator 535. Counter 532 has an UPDATE input coupled
to receive clock signal U.sub.CLK 512. Comparator 535 is coupled to
receive an output 534 of counter 532 and the modulation signal and
in turn, output modulation period signal U.sub.MDP 538. In one
example, counter 532 is configured to continually count up every
period of clock signal U.sub.CLK 512. Specifically, counter 532
increments output 534 by counting up every period of clock signal
U.sub.CLK 512 until output 534 reaches the maximum value of the
modulation bits (i.e., 2.sup.P-1). In response to reaching the
maximum value of the modulation bits, counter 532 sets output 534
back to zero and again counts up every period of clock signal
U.sub.CLK 512. In one example, the period of modulation period
signal U.sub.MDP 538 may be equal to 2.sup.P periods of clock
signal U.sub.CLK 512. In operation, comparator 535 can set
modulation period signal U.sub.MDP 538 to logic high or logic low
in response to comparing output 534 with the value of the
modulation bits. In one example, comparator 535 can set modulation
period signal U.sub.MDP 538 to logic high if output 534 is less
than the value of the modulation bits and may set modulation period
signal U.sub.MDP 538 to logic low if output 534 is greater than or
equal to the value of the modulation bits. In other words,
modulator 530 can adjust a portion of the period of modulation
period signal U.sub.MDP 538 that modulation period signal U.sub.MDP
538 is logic high and a portion of the period that modulation
period signal U.sub.MDP 538 is logic low in response to the value
of the modulation bits. That is, modulator 530 can adjust the duty
cycle of the modulation period signal U.sub.MDP 538 (i.e., the
ratio of a portion of the period of modulation period signal
U.sub.MDP 538 during which modulation period signal U.sub.MDP 538
is either logic high or logic low to the period of modulation
period signal U.sub.MDP 538) in response to the value of the
modulation bits. The duty cycle of modulation period signal
U.sub.MDP 538 times the period of modulation period signal
U.sub.MDP 538 may be referred to as a modulation time. As such, in
one example, the modulation time represents the portion of the
period of modulation period signal U.sub.MDP 538 during which
modulation period signal U.sub.MDP 538 is logic high.
[0085] In one example, the value of the modulation bits determines
the duty cycle of modulation period signal U.sub.MDP 538. For
example, if there are three modulation bits and the value of the
modulation bits is equal to three (i.e., 011 binary), the period of
modulation period signal U.sub.MDP 538 is eight (2.sup.3) periods
of clock signal U.sub.CLK 512 and the duty cycle of the modulation
period signal U.sub.MDP 538 is 3/8. Accordingly, modulation period
signal U.sub.MDP 538 may be set to logic high for three periods of
clock signal U.sub.CLK 512 and to logic low for five periods of
clock signal U.sub.CLK 512 during every period of modulation period
signal U.sub.MDP 538.
[0086] As further illustrated, modulated DAC 540 may include a DAC
577 comprising a group of switchable bit-to-analog circuitries 576
and a switchable modulation source 575. The number of switchable
bit-to-analog circuitries included in DAC 577 may be equal to the
number of baseline bits (i.e., Q). Group of switchable
bit-to-analog circuitries 576 and switchable modulation source 575
may be coupled between a voltage source VA and a summing block 586.
In the depicted example, group of switchable bit-to-analog
circuitries 576 is coupled to receive the baseline bits and provide
to summing block 586 a base signal U.sub.BASE 584, which is
responsive to the baseline bits. In one example, each one of
switchable bit-to-analog circuitries 564 to 570 includes a current
source and a switch. Each one of switchable bit-to-analog
circuitries 564 to 570 of the group can be switched in response to
one of the baseline bits. Therefore, base signal U.sub.BASE 584 may
be representative of the total sum of currents that is provided to
summing block 586 from the group. It should be noted that, in other
examples, each one of switchable bit-to-analog circuitries 564 to
570 may include other known circuit components such as a resistor,
a capacitor in place of a current source. In one example, summing
block 586 is a circuit node that is coupled to receive multiple
currents and output a signal representative of the sum of these
currents.
[0087] Furthermore, in the example modulated DAC 540, DAC 577 is a
binary weighted DAC. That is, each one of current sources 544 to
550 of the group of switchable bit-to-analog circuitries outputs a
current that has a magnitude that is weighted by power of two
relative to the current output by the current source of the
switchable bit-to-analog circuitry that is responsive to the
adjacent less-significant bit of the baseline bits. In other words,
switchable bit-to-analog circuitry 564 is responsive to the least
significant bit B.sub.1 and current source 544 outputs a current
with the lowest magnitude I.sub.B. Switchable bit-to-analog
circuitry 566 is responsive to bit B2 and current source 546
outputs a current that is twice as large as I.sub.B. The magnitude
of the current output by each one of the current sources of the
remaining switchable bit-to-analog circuitries successively doubles
such that current source 550 of switchable bit-to-analog circuitry
570 responsive to the most significant bit B.sub.Q outputs a
current that has magnitude 2.sup.QI.sub.B.
[0088] As further shown, switches 554 to 560 coupled to current
sources 544 to 550, respectively, can be switched in response to
one of the baseline bits. Specifically, each one of switches 554 to
560 can be switched in response to the bit of the baseline bits to
which the corresponding switchable bit-to-analog circuitry is
responsive. For example, switchable bit-to-analog circuitry 564
includes switch 554 that is coupled to current source 544. Since
switchable bit-to-analog circuitry 564 is responsive to the least
significant bit B.sub.1 of the baseline bits, switch 554 can be
switched in response to the least significant bit B.sub.1. In the
depicted example, a high value for one of the digits of the
baseline bits may close (i.e., enable) the respective switch to
couple the corresponding current source to summing block 586.
Conversely, a low value for one of the digits of the baseline bits
may open (i.e., disable) the respective switch to prevent current
from its respective current source from entering summing block 586.
Therefore, base signal U.sub.BASE 584 that enters summing block 586
is an analog signal representative of the value of the baseline
bits. In the example of FIG. 5, a low value for one of the baseline
bits is a high value for the complement of that one of the baseline
bits. As illustrated, a bar over the symbol for a bit of the
baseline bits represents the complement of the bit.
[0089] It should be noted that other examples of modulated DAC 540
may have multiple switches within each one of switchable
bit-to-analog circuitries to direct current from current sources
544 to 550 to other nodes for other reasons, such as for
calibration.
[0090] In the example DAC 577, switchable modulation source 575 is
coupled to output a modulated signal U.sub.MOD 582 in response to
modulation period signal U.sub.MDP 538. Switchable modulation
source 575 includes a current source 542 that outputs a current
that has magnitude equal to that of the current source of the
switchable bit-to-analog circuitry responsive to the least
significant bit B.sub.1 of the baseline bits. In other words,
current source 542 outputs a current that has a magnitude of
I.sub.B. Switchable modulation source 575 also includes a switch
552 coupled to current source 542. Switch 552 can be switched in
response to modulation period signal UMP 538. In one example, when
modulation period signal U.sub.MDP 538 is logic high, switch 552
closes (i.e., is enabled) to couple current source 542 to summing
block 586 and when modulation period signal UMP 538 is logic low,
switch 552 opens (i.e., is disabled) to prevent current from
current source 542 from entering summing block 586. That is, in
operation, switchable modulation source 575 may alternate modulated
signal U.sub.MOD 582 between I.sub.B and zero in response to
modulation period signal U.sub.MDP 538. In such cases, modulated
signal U.sub.MOD 582 is also a periodic signal with the same period
as that of modulation period signal U.sub.MDP 538 (i.e., 2.sup.P
periods of clock signal U.sub.CLK 512).
[0091] As depicted in the example in FIG. 5, adjust signal
U.sub.ADJ 580 output by modulated DAC 540 is the sum of currents
that are received by summing block 586, namely the sum of modulated
signal U.sub.MOD 582 and base signal U.sub.BASE 584. In one
example, this means that adjust signal U.sub.ADJ 580 is a periodic
signal with the same period as that of modulation period signal
U.sub.MDP 538 and modulated signal U.sub.MOD 582 (i.e., 2.sup.P
periods of clock signal U.sub.CLK 512). In operation, by
alternating modulated signal U.sub.MOD 582 between I.sub.B and zero
in response to modulation period signal U.sub.MDP 538, switchable
modulation source 575 can cause adjust signal U.sub.ADJ 580 to
alternate between base signal U.sub.BASE 584 and base signal
U.sub.BASE 584 plus I.sub.B during a period of adjust signal
U.sub.ADJ 580. Because the current source of the switchable
bit-to-analog circuitry that is responsive to the least significant
bit of the baseline bits outputs a current that has a magnitude of
I.sub.B, the difference in base signal U.sub.BASE 584 for two
adjacent values of the baseline bits is equal to I.sub.B.
Therefore, base signal U.sub.BASE 584 and base signal U.sub.BASE
584 plus I.sub.B may represent two adjacent discrete levels of
adjust signal U.sub.ADJ 580 with base signal U.sub.BASE 564
corresponding to the lower level, which may also be referred to as
the "base level." In this case, adjust signal U.sub.ADJ 580 is an
analog signal having discrete levels that are set in response to
the value of the baseline bits and the value of the modulation
bits.
[0092] As previously mentioned, modulated DAC 540 may alternate
adjust signal U.sub.ADJ 580 between two adjacent discrete levels in
response to modulation period signal U.sub.MDP 538. Specifically,
modulated DAC 540 may set a portion of the period of adjust signal
U.sub.ADJ 580 that adjust signal U.sub.ADJ 580 is the greater of
the adjacent levels in response to modulation period signal
U.sub.MDP 538. For example, modulated DAC 540 may set adjust signal
U.sub.ADJ 580 to the greater level when modulation period signal
U.sub.MDP 538 is logic high and set adjust signal U.sub.ADJ 580 to
the base level when modulation period signal U.sub.MDP 538 is logic
low. Modulated DAC 540 may thus set the portion of the period of
adjust signal U.sub.ADJ 580 during which adjust signal U.sub.ADJ
580 is the greater level to be equal to the modulation time.
[0093] In the depicted example, the average value of adjust signal
U.sub.ADJ 580 may over time become equal to one of several
additional levels that are equally spaced between the adjacent
levels of adjust signal U.sub.ADJ 580. More particularly, by
changing the modulation time between zero and (2.sup.P-1) periods
of clock signal U.sub.CLK 512 in response to the value of the
modulation bits, modulated DAC 540 may output an adjust signal
U.sub.ADJ 580 with an average value equal to one of (2.sup.P-1)
equally spaced levels between any pair of adjacent levels. In this
manner, modulated DAC 540 may generate a signal at its output with
the equivalent number of discrete levels and hence with the same
resolution that a conventional (P+Q) bit DAC can generate.
[0094] Modulated DAC 540 can thus reduce the area required for
implementing a DAC since modulated DAC 540 may need only Q circuit
components (e.g., current sources) to achieve the same resolution
as that of a (P+Q) bit DAC. Although the example modulated DAC of
FIG. 5 provides higher resolution by alternating the output between
adjacent levels, in other examples, an averaging circuit such as a
low pass filter, an integrator or the like can be used either in
the modulated DAC or outside of the modulated DAC to generate the
average value of adjust signal U.sub.ADJ 580.
[0095] In the illustrated modulated DAC 540, modulator 530 operates
in conjunction with a binary weighted DAC to alternate the output
between adjacent levels. In other cases, modulator 530 can operate
in conjunction with other types of DACs (e.g., thermometer coded
DAC, R-2R ladder DAC) that are appropriately modified. For example,
modulated DAC 540 may include a DAC that generates two levels for
the output in response to the baseline bits. One of the levels may
correspond to the value of the baseline bits and the other one of
the levels may correspond to one of the adjacent values of the
baseline bits. In this example, the DAC may alternate the output
between the two levels in response to modulation period signal
U.sub.MDP 538.
[0096] In some implementations, rather than generating a modulation
signal that alternates between a logic high level and a logic low
level, one or more signals that alternate between adjacent digital
levels can be received by a DAC. For example, a multiplexer can
output a digital signal that is set to a first level for a first
period of time and to an adjacent second level for a second period
of time within a period. The DAC can receive the digital signal at
the output of the multiplexer and convert the alternating digital
signal to an analog output as adjust signal U.sub.ADJ 580. The
first level may be determined, for example, based on the value of
the baseline bits or the value of the digital input signal received
by a modulated DAC. The first period of time and the second period
of time can be determined, for example, based on the value of the
modulation bits or in response to a separate signal that sets the
duty cycle of modulation period signal U.sub.MDP 538. In this case,
the multiplexer would be acting as a modulator and its output would
alternate between a higher digital level and a lower digital level.
In yet another implementation, rather than converting the digital
signal at the output of the multiplexer to an analog output, the
digital signal may be output as adjust signal U.sub.ADJ 580.
[0097] In the example driver circuit 156, drive signal generator
520 is coupled to receive adjust signal U.sub.ADJ 580 and in
response, output drive signal U.sub.DR 162. Drive signal U.sub.DR
162 may be a periodic signal that alternates between logic high and
logic low. In operation, drive signal generator 520 can set certain
properties of drive signal U.sub.DR 162 based on adjust signal
U.sub.ADJ 580. Examples of such properties of drive signal U.sub.DR
162 include period, ratio of logic high to logic low in a period,
etc. By setting the properties of drive signal U.sub.DR 162, drive
signal generator 520 can set the operating condition of power
switch S1 170. In one example, drive signal generator 520 may
include switches, capacitors, and comparators (not shown in FIG.
5). Drive signal generator 520 can use the charging and discharging
of those capacitors in response to adjust signal U.sub.ADJ 580 to
set one or more properties of drive signal U.sub.DR 162. In another
example, drive signal generator 520 can include a circuit (e.g., a
digital pulse width modulation circuit) to output a rectangular
signal as drive signal U.sub.DR 162. The circuit can be coupled to
receive an alternating digital signal (e.g., the digital signal at
the output of the multiplexer) and configured to set one or more
properties of drive signal U.sub.DR 162 in response to the
alternating digital signal.
[0098] FIG. 6 shows a table illustrating different values of an
example digital signal received as an input signal by driver
circuit 156 and a timing diagram illustrating waveforms of various
signals of modulated DAC 540. In the illustrated example, the table
includes nine rows from row A to row I where each row shows a
different value of the input signal that, in one example, may be
representative of input signal U.sub.IN 590. Specifically, the
input signal includes a total of 13 bits with upper ten bits set
aside as the baseline bits (i.e., Q equals ten) and lower three
bits set aside as the modulation bits (i.e., P equals three). As
shown, in each row of the table, the portion of the input signal
corresponding to the baseline bits is separated from the portion
corresponding to the modulation bits by a solid line. Further
illustrated in FIG. 6 is periodic waveform 612 with a period
T.sub.CLK 614 and that alternates between logic high H and logic
low L. In one example, waveform 612 is representative of clock
signal U.sub.CLK 512 generated by clock signal generator 510. FIG.
6 also includes a collection of example waveforms 620 for adjust
signal U.sub.ADJ 580. Each waveform of the collection corresponds
to the input signal in the respective adjacent row of the
table.
[0099] In the depicted example, since there are three bits used as
modulation bits, the value of the modulation bits can be one of
eight different values between zero and seven. Waveforms in
collection 620 are periodic with a period T.sub.ADJ 616 which is
equal to eight periods of waveform 612 (i.e., eight times period
Tux 614). As previously mentioned, the base level of adjust signal
U.sub.ADJ 580 is set in response to the value of baseline bits and
the duty cycle of modulation period signal U.sub.MDP 538 is set in
response to the value of the modulation bits. In the example, the
input signal in row A indicates 513 for the value of the baseline
bits and zero for the value of the modulation bits. In response to
receiving this input signal, modulated DAC 540 sets the base level
of adjust signal U.sub.ADJ 580 to 513I.sub.B and the duty cycle of
modulation period signal U.sub.MDP 538 to zero. In other words,
modulated DAC 540 sets modulation period signal U.sub.MDP 538 to
logic low for the entire period of modulation period signal
U.sub.MDP 538. Accordingly, the corresponding waveform of group 620
is constant and equal to the base level of 513I.sub.B. For rows B
through I, the value of the baseline bits indicated by the
respective input signal is the same and equal to 512 but the value
of the modulation bits decreases from seven in row B to zero in row
I. Thus, for rows B through I, the base level of adjust signal
U.sub.ADJ 580 is equal to 512I.sub.B, the greater of the adjacent
levels of adjust signal U.sub.ADJ 580 is equal to 513I.sub.B
(512I.sub.B+I.sub.B) and the modulation time is equal to the value
of the modulation bits times period Tux 614. In other words,
modulation period signal U.sub.MDP 538 is logic high for a length
of time that is equal to the value of the modulation bits times
period Tux 614. As a result, for rows B through I, each
corresponding waveform of group 620 is set to the 513I.sub.B for
the corresponding modulation time during each period T.sub.ADJ 616.
For example, in row B, the value of the modulation bits is seven,
the corresponding modulation time is seven times period T.sub.CLK
614 and the corresponding waveform of group 620 is set to
513I.sub.B for seven times period T.sub.CLK 614 during each period
T.sub.ADJ 616. Similarly, in row E, the value of the modulation
bits is four, the corresponding modulation time is four times
period T.sub.CLK 614 and the corresponding waveform is set to
513I.sub.B for four times period T.sub.CLK 614 during each period
T.sub.ADJ 616.
[0100] FIG. 7 shows a collection of signal levels illustrating
example average values for the output of the modulated DAC of FIG.
5. Specifically, each one of levels of collection 720 in FIG. 7
represents the average value of the corresponding waveform of
collection 620 that is responsive to the input signal in the
respective adjacent row of the table in FIG. 6. In other words,
FIG. 7 is an expanded view illustrating both the adjacent levels of
512I.sub.B and 513I.sub.B of adjust signal U.sub.ADJ 580 and the
additional seven (2.sup.3-1) equally spaced levels between these
adjacent levels. As shown, the levels of collection 720 that
correspond to the input signals in row A and I of the table in FIG.
6 are equal to 513I.sub.B and 512I.sub.B, respectively. As the
modulation time decreases from seven times period T.sub.CLK 614 in
row B to one times period T.sub.CLK 614 in row H, the corresponding
level of collection 720 (i.e., the average value of adjust signal
U.sub.ADJ 580) decreases by steps of (1/8)I.sub.B. For example, the
average value of the waveform of collection 620 corresponding to
row B is (512I.sub.B+(7/8)I.sub.B), the average value of the
waveform of collection 620 corresponding to row C is (512I.sub.B+(
6/8)I.sub.B) and so on. It should be noted that even though the
example waveforms and levels in FIG. 6 and FIG. 7 only illustrate
this behavior for one pair of adjacent levels of adjust signal
U.sub.ADJ 580 (i.e., 512I.sub.B and 513I.sub.B), the same behavior
can be seen for any pair of adjacent levels. Therefore, an adjust
signal U.sub.ADJ 580 generated by using ten current sources in this
manner may have the equivalent number of discrete levels and the
same resolution as that of a signal generated by a 13 bit DAC that
uses, for example, 13 current sources.
[0101] FIG. 8 is a schematic circuit diagram illustrating another
implementation of the driver circuit in FIG. 1A according to the
teachings of the present invention. Driver circuit 156 in FIG. 8
differs from driver circuit 156 in FIG. 5 in that counter 832 is
coupled to receive drive signal U.sub.DR 162 at its UPDATE input.
Consequently, the periods of modulation period signal U.sub.MDP
838, modulated signal U.sub.MOD 882 and adjust signal U.sub.ADJ 880
are all equal to 2.sup.P times the period of drive signal U.sub.DR
162. In the example driver circuit, since the period of drive
signal U.sub.DR 162 may vary in response to adjust signal U.sub.ADJ
880, the periods of modulation period signal U.sub.MDP 838,
modulated signal U.sub.MOD 882 and adjust signal U.sub.ADJ 880 may
also vary during the operation of controller 150.
[0102] FIG. 9 is a flow diagram illustrating an example process 900
according to the teachings of the present invention for generating
an analog signal having discrete levels in response to a digital
signal. In the depicted example, process 900 may begin at block
905. At block 910, an N bit digital signal is received as an input.
At block 920, the base level of an output signal is set in response
to upper Q bits of the input. More particularly, the base level may
be proportional to the value of a digital signal represented by
upper Q bits of the input. For example, Q may be chosen as ten and
the digital signal represented by upper Q bits of the input may be
equal to 512. In this case, the base level of the output signal may
be equal to 512 times a current that has a magnitude of
I.sub.B.
[0103] At block 930, a periodic signal is generated in response to
lower P bits of the input. In one example, the periodic signal has
a period proportional to the number of possible values that can be
represented by lower P bits of the input (i.e., 2.sup.P). For
example, when P is chosen as three, the period of the periodic
signal may be eight (2.sup.3) periods of a clock signal. The
periodic signal may alternate between logic low and logic high
during that period. At block 940, the duty cycle of the periodic
signal (i.e., the ratio of either a logic high or logic low portion
of the period of the periodic signal to the period of the periodic
signal) is set in response to the value of lower P bits of the
input. The duty cycle of the periodic signal times the period of
the periodic signal may represent a modulation time. Thus, in one
example, the portion of the period during which the periodic signal
is set to logic high may be equal to the modulation time and hence,
proportional to the value of lower P bits of the input. For
example, the value of lower P bits of the input may be equal to
five. In this case, the duty cycle of the periodic signal may be
5/8, the modulation time may be five periods of the clock signal
and the periodic signal may be logic high for five of eight periods
of the clock signal.
[0104] At block 950, the output signal may be set to an adjacent
level of the base level for the modulation time and to the base
level for the remainder of the period of the periodic signal. In
one example, the adjacent level corresponds to the greater of the
two closest levels of the base level that can be generated in
response to upper Q bits of the input. For example, the base level
of the output signal may increase or decrease in steps of value
I.sub.B in response to upper Q bits of the input. In this case,
when the base level is equal to 512I.sub.B, the adjacent level
becomes equal to 513I.sub.B. Consequently, the output signal may be
set to 513I.sub.B for the modulation time and to 512I.sub.B for the
remainder of the period of the periodic signal.
[0105] In this manner, the average value of the output signal
generated by process 900 may be equal to one of (2.sup.P-1)
additional levels that are equally spaced between any pair of
adjacent levels of the output signal. As such, the output signal
generated by using Q current sources according to process 900 can
achieve the same resolution as that of a signal generated by a
(Q+P) bit DAC that uses (Q+P) current sources.
[0106] In one example, the illustrated process may be performed by
a modulated DAC 540 or 840. Modulated DAC 540 or 840 may perform
process block 910 by receiving input signal U.sub.IN 590 or 890 as
an N bit digital signal. When group of switchable bit-to-analog
circuitries 576 or 876 performs process block 920, the base level
of an output signal (e.g., adjust signal U.sub.ADJ 580 or 880) is
set in response to upper Q bits of the digital signal. In one
example, group of switchable bit-to-analog circuitries 576 or 876
can set the base level of the output signal by outputting base
signal U.sub.BASE 584 or 884 to summing block 566 or 866.
Generating a periodic signal (e.g., modulation period signal
U.sub.MDP 538 or 838) in response to P lower bits of the digital
signal in process block 930 can be performed by modulator 530 or
830. Setting the duty cycle of the periodic signal (hence, the
modulation time) in response to the value of lower P bits of the
digital signal in process block 940 can also be performed by
modulator 530 or 830. Setting the output signal to an adjacent
level of the base level for the modulation time and to the base
level for the remainder of the period of the periodic signal in
process block 950 can be performed by switchable modulation source
575 or 875 and group of switchable bit-to-analog circuitries 576 or
876. In particular, switchable modulation source 575 or 875 can be
configured to couple a current source (e.g., current source 542 or
842) to summing block 566 or 866 to provide an additional signal
(e.g., modulated signal U.sub.MOD 582 or 882) in response to
modulation period signal UMP 538 or 838 such that the output signal
is set to the adjacent level of the base level for the modulation
time. Conversely, switchable modulation source 575 or 875 can be
configured to prevent the current source from coupling to summing
block 566 or 866 in response to modulation period signal U.sub.MDP
538 or 838 such that the output signal is set to the base level for
the remainder of the period of the periodic signal.
[0107] The above description of illustrated examples of the present
invention, including what is described in the Abstract, are not
intended to be exhaustive or to be limitation to the precise forms
disclosed. While specific embodiments of, and examples for, the
invention are described herein for illustrative purposes, various
equivalent modifications are possible without departing from the
broader spirit and scope of the present invention. Indeed, it is
appreciated that the specific example voltages, currents,
frequencies, power range values, times, etc., are provided for
explanation purposes and that other values may also be employed in
other embodiments and examples in accordance with the teachings of
the present invention.
* * * * *