An Online Phase Margin Monitor for Digitally Controlled Switched-Mode Power Supplies

An Online Phase Margin Monitor for Digitally Controlled Switched-Mode Power Supplies
An Online Phase Margin Monitor for Digitally
Controlled Switched-Mode Power Supplies
Jeff Morroni, Regan Zane and Dragan Maksimović
Colorado Power Electronics Center
University of Colorado at Boulder, USA
Email: {morroni, zane, maksimov}@colorado.edu
suited for one-time frequency response identification, e.g.
upon start up, or at other times when the system is in
steady state. In [6-11], frequency response information
and tuning of compensator parameters is performed based
on purposely induced limit-cycle oscillations in a
sequence of steps, assuming steady-state operation.
In this paper, inspired by the analog injection
technique [1], a method is proposed to measure the
crossover frequency and phase margin in a digitally
controlled power supply online, i.e., during normal
system operation. The proposed approach does not
require opening the feedback loop and is capable of
continuously updating the measured crossover frequency
and phase margin outputs in the presence of load
transients or other system disturbances. Applications of
the technique include fast design time verifications,
online dynamic performance monitoring of power
supplies in power distribution systems (such as servers or
spacecrafts [12-14]), as well as adaptive online tuning of
control parameters [15]. Section II details the proposed
approach. Section III presents experimental results.
Conclusions are given in Section IV.
Abstract — This paper presents a practical injection-based
method for continuous monitoring of the crossover
frequency and phase margin in digitally controlled
switched-mode power supplies (SMPS). The proposed
approach is based on Middlebrook’s loop-gain
measurement technique [1], adapted to digital controller
implementation. A digital square-wave signal is injected in
the loop, and the injection signal frequency is adjusted while
monitoring loop signals to obtain the system crossover
frequency and phase margin online, i.e., during normal
SMPS operation. The approach does not require open loop
or steady-state SMPS operation and is capable of
convergence in the presence of load transients or other
disturbances. Experimental results are presented for
various power stage configurations demonstrating close
matches between monitored and expected crossover
frequencies and phase margins.
I.
INTRODUCTION
Switching power converters are nonlinear systems
with dynamic responses that depend on the operating
point. Typically, based on averaged small-signal models,
switched-mode power supply (SMPS) feedback loops are
designed conservatively so that stability margins and
closed-loop regulation performance are maintained over
expected ranges of operating conditions and tolerances in
power stage parameters. At design time, it is a common
practice to measure the system loop gain using a network
analyzer to verify the system stability margins under
various conditions. Middlebrook’s injection technique [1]
has been widely adopted in practice as it allows loop gain
measurements without breaking the feedback loop. In
this way, designs can be verified offline to ensure desired
performance before system deployment.
With advances in digital control for high-frequency
DC-DC converters [2], it becomes possible to consider
alternative design approaches and techniques leading to
improved closed-loop dynamic responses or improved
robustness of SMPS. In particular, various methods have
been proposed to measure converter frequency responses
online [3-6] or to tune compensator parameters based on
online assessments of the frequency responses [6-10]. In
[3-5], a pseudo-random binary sequence perturbs the
converter duty cycle for the purpose of identifying the
open-loop control-to-output frequency response using
cross-correlation methods.
During duty-cycle
perturbation, the system is temporarily operated in openloop steady state. As a result, these approaches are best
978-1-4244-1668-4/08/$25.00 ©2008 IEEE
II.
STABILITY MARGIN MONITOR
Middlebrook’s analog loop gain measurement
technique is a well known and widely accepted approach
to measuring the loop frequency responses without
breaking the feedback loop [1, 16]. Figure 1 illustrates
this approach for the case of voltage injection Vz in series
with the loop. The measured gain Tv(s) is
0
Block 1
Z1(s)
+
+
_ G1(s)ve(s)
Σ
-
Block 2
Vˆz ( s)
vref(s)
Zs(s)
+
Vˆy ( s )
+
Vˆx ( s )
iˆ( s )
G2 ( s )Vˆx ( s )
Z2(s)
-
-Tv(s)
H(s)
Figure 1: Small-signal SMPS model illustrating analog loop gain
measurement technique using voltage injection without breaking the
feedback loop [1, 16].
859
Authorized licensed use limited to: UNIVERSITY OF COLORADO. Downloaded on November 12, 2008 at 13:03 from IEEE Xplore. Restrictions apply.
Tv ( s ) =
− Vˆy ( s )
Z (s) ⎞ Z (s )
⎛
= T ( s )⎜1 + 1 ⎟ + 1
,
ˆ
Z
Vx ( s)
2 (s) ⎠ Z 2 ( s)
⎝
T (e
(1)
−V y
.
Vx
ϕ m = 180° + ∠T ⎛⎜ e
⎝
jωinjT
⎞⎟
.
⎠ ωinj =2πf c
fc = finj ,
(5)
V y = Vx .
(6)
Further, when (6) is satisfied, the phase margin can be
directly measured as:
ϕ = ϕ m = ∠V y ( f inj ) − ∠V x ( f inj ) .
Σ
_
Verr(t)
Switched-Mode
Power Converter
Vy
+
+ δ
-δ
Injection
Generator
δ
Stability Margin
Monitor
Vref(t)
+
ADC
-T
Vz
(7)
Based on (2)-(7), the crossover frequency and phase
margin can be monitored online in digitally controlled
systems without requiring any additional power stage
information. The monitoring can be performed
continuously during normal operation of the system at the
cost of a small output voltage perturbation imparted by
the injection Vz. However, the perturbation amplitude
seen at the converter output can be automatically
controlled by adjusting the signal injection amplitude δ,
as shown in Fig. 2.
A more detailed block diagram of the stability margin
monitor is shown in Fig. 3. Details regarding the design
Vout(t)
Σ
(4)
if
(2)
Vx
(3)
From (2) and (3), the crossover frequency fc is equal to
the injection source frequency,
From (2), the crossover frequency, fc, can be found as the
frequency where:
DPWM
) =1 ,
while the phase margin is obtained from
where T(s) is the actual loop gain. Clearly, T(s) ≈ Tv(s) as
long as ||Z1|| << ||Z2|| and ||T|| >> ||Z1/Z2||. In a SMPS with
analog voltage-mode PWM control, points where the
impedance conditions for the loop gain measurement
using voltage injection are well satisfied typically include
the converter output or the compensator output.
Figure 2 shows an implementation of the injectionbased loop gain measurement technique applied to a
digitally-controlled SMPS. The digital controller has the
standard architecture including a voltage A/D converter
(ADC), discrete-time compensator and digital pulsewidth modulator (DPWM). Similar to the analog voltage
injection approach, a small digital injection source Vz can
be added to a digital loop signal at a suitable point. For
example, injection can occur at the compensator input or
at the compensator output, as shown in Fig. 2. It should
be noted that a similar signal injection technique has been
proposed for the purpose of tuning a compensator gain to
achieve desired crossover frequency as part of an autotuning process [10].
Given the injection source Vz, and the fact that
||Z1/Z2|| = 0 in the digital part of the loop, the system loop
gain can be found as:
T=
jωinjT
PID Compensator
Gc(z)
Injection
Amplitude
Controller
finj
φ
Figure 2: Crossover frequency and phase margin monitor block diagram. The outputs of stability margin monitor are crossover frequency and
phase margin. The injection amplitude controller automatically adjusts the square-wave perturbation amplitude, δ, to result in minimum
(+/- 1 LSB) perturbation at the output voltage.
860
Authorized licensed use limited to: UNIVERSITY OF COLORADO. Downloaded on November 12, 2008 at 13:03 from IEEE Xplore. Restrictions apply.
To DPWM
Σ
Vx
From
Compensator
+
Band-pass
Filter
Σ
Vp_ref
_
Vp_err
δ
Injection
Generator
Vp
+
Peak
Detector
Vy
Vz +
Band-pass
Filter
Vy(finj)
Verr[n]
Injection
Generator
-δ
δ
∫
δ
Vx(finj)
finj
finj
∫
Peak Detector
+
||Vx(finj)||
Σ
Figure 4: Block diagram of the injection amplitude controller. The
injection amplitude, δ, is adjusted via feedback until the desired
output voltage perturbation magnitude is achieved.
Peak Detector
_
Phase Detector
improved DC regulation, as explained in [17]. In
particular, the +/- 1 LSB periodic oscillation imposed by
Vz at the output voltage combined with the action of the
integrator in the PID compensator will work to position
the DC value of the output voltage in the center of the
zero error bin. The accuracy with which the output
voltage can be centered in the zero error bin then
becomes a function of the DPWM resolution rather then
the ADC resolution, which in general is finer to satisfy
conventional limit cycle criteria [18, 19].
||Vy(finj)||
φ
Figure 3: Blocks required to implement the stability margin monitor.
The injection frequency is adjusted, via feedback, until the filtered
peaks, ||Vx(finj)|| and ||Vy(finj)|| are equal. At this point, finj = fc and
φ=φm.
and implementation of each block in Fig. 3 are presented
in the following subsections.
B. Band-pass Filters and Peak Detectors
As described previously, in the proposed
implementation of Fig. 3, Vz is a 50% duty cycle squarewave injection with adjustable frequency determined by
the frequency command finj. However, (2)-(7) are based
on the assumption that Vz is a purely sinusoidal injection.
To account for the infinite odd harmonics introduced by
the square-wave, band-pass filters are used to remove all
unwanted frequency components of Vx and Vy. The
outputs of the band-pass filters, Vy(finj) and Vx(finj), then
contain only one frequency component, equal to the
injection frequency.
In Fig. 3, the band-pass filters, Gbp(z), are designed to
be high Q-factor filters with the pass-band of the filter
centered at finj. However, since finj changes in order to
satisfy (5), the filter pass-bands must also continuously
change. To understand how to realize adjustable bandpass digital filters, consider a general form 2nd order
digital band-pass filter
A. Injection Generator and Injection Amplitude
Controller
The injection generator block creates a 50% duty
cycle, square-wave perturbation with frequency
adjustable by the frequency command, finj. Practically,
this square-wave signal can be generated with a digital
counter, running off of a system clock having clock
frequency fclk, and a simple comparator. The frequency
resolution qfinj in finj depends on the ratio of the system
crossover frequency fc and the system clock frequency
fclk,
q finj =
f c2
.
f clk
(8)
In a typical system, the crossover frequency fc is a
fraction of the switching frequency fs, which, in turn is a
fraction of the system clock frequency. Hence, the
resolution qinj is typically a small fraction of fc.
To minimize the impact of the signal injection on the
output voltage ripple, it is desirable to control the
injection signal amplitude, δ, to obtain a minimum
detectable +/-1 least significant bit (LSB) output voltage
perturbation. However, in general the required injection
signal amplitude depends on finj. A block diagram of the
proposed feedback loop used to control δ to account for
changes in finj is shown in Fig. 4. The injection amplitude
controller takes as input the quantized output voltage
error Verr[n], which is then passed through a peak
detector. The peak output voltage error, Vp, is compared
to the desired LSB perturbation magnitude, Vp_ref. A
simple integral compensator then adjusts δ until the
desired output voltage perturbation is achieved.
A secondary benefit to purposely introducing a
periodic oscillation into a digital control loop involves
Gbp ( z ) = A
(z − 1)
2
z + Bz + C
.
(9)
Based on the discrete-time to continuous-time mapping
z=e
s
f sample
,
(10)
the pass-band center frequency fpb and the Q-factor of (9)
can be found as
f pb =
f sample
2π
⎛ B 2 − 4C
tan −1 ⎜
⎜
B
⎝
⎞
⎟,
⎟
⎠
861
Authorized licensed use limited to: UNIVERSITY OF COLORADO. Downloaded on November 12, 2008 at 13:03 from IEEE Xplore. Restrictions apply.
(11)
⎛ B 2 − 4C
tan −1 ⎜⎜
B
⎜
1
⎝
Q=
2
ln C
⎞
⎟
⎟⎟
⎠.
degrees in the detected phase,
f
q PM = 360° c .
(13)
f clk
Since fc is typically a fraction of the switching frequency,
which is a fraction of the system clock frequency, (13)
implies a phase detection resolution of several degrees.
The other main factor in the resolution/accuracy of
the phase detector is the sample rate, fsample, of the bandpass filters with respect to finj
(12)
Given fsample, fpb and Q, (11) and (12) can be used to solve
for the filter coefficients B and C. Then, by holding A, B
and C constant while varying fsample in proportion to finj,
the filter pass-band center frequency fpb shifts in
proportion to finj while Q stays constant.
The peak detectors take as inputs the filtered
waveforms, Vy(finj) and Vx(finj), and output ||Vx(finj)|| and
||Vy(finj)||, as shown in Fig. 3. The peak detectors give an
assessment of the magnitudes of each signal such that (6)
can be satisfied by closing an integral feedback loop
around the injection frequency.
fsample = γ finj.
(14)
where γ is an integer proportionality constant. Since
Vy(finj) and Vx(finj) are sampled at a rate proportional to finj,
their respective zero crossings could be shifted by as
much as one sample period, 1/fsample, from the actual zerocrossings. Therefore, the phase error satisfies
C. Integral Compensator
A slow integral compensator is used to process the
error between ||Vx(finj)|| and ||Vy(finj)||. Since the stability
margin monitor control loop is a sampled-system, its
bandwidth must be much slower than the injection
frequency (which upon convergence equals fc).
Therefore, a slow integral compensator is sufficient to
close the feedback loop in the stability margin monitor.
The output of the integral compensator is finj, the injection
frequency command, which is adjusted until there is no
error between ||Vx(finj)|| and ||Vy(finj)||, at which point (6)
is satisfied and fc = finj.
PM _ error ≤
360°
.
(15)
γ
As γ is increased, (15) approaches zero and the phase
margin resolution is dominated by (13). Note that (15)
gives the maximum phase error (i.e. the phase error is
guaranteed to be no larger than (15)).
III.
EXPERIMENTAL RESULTS
There are two experimental test-beds, shown in Fig. 6,
used to verify functionality of the proposed stability
margin monitor, a synchronous buck converter and a
boost converter which can be operated in continuous
conduction mode (CCM) or discontinuous conduction
mode (DCM).
4 μH
Point A
D. Phase Detector
The phase detector block diagram, used to monitor
φm, is shown in Fig. 5 and is similar to some approaches
used to detect phase in digital phase-locked loops [20].
The phase detector takes as input the filtered signals,
Vy(finj) and Vx(finj). These signals are passed through a
digital relay whose output is high when the input is above
zero and low when the input is below zero. The two relay
outputs are then XOR’d together to form an Enable pulse,
labeled in Fig. 5, which is high when the two inputs are
not equal. This enable pulse gives a direct relationship
between the phases of the two inputs signals. A counter
running at the system clock frequency, fclk, measures the
length of time Enable is high which is related to the phase
shift between Vy(finj) and Vx(finj).
There are two important sampling effects which
determine the resolution of the phase detector. First, the
system clock frequency determines the resolution in
+
12V
370 μF
+
_
5V
0.4
_
D
_
fc
Vref
ADC
Digital Controller
+
φm
(a) Synchronous buck converter
L
Point A
+
Vg
Vy(finj)
Relay
2 μF
+
_
Clk
30V
.025
_
Enable
XOR
Vx(finj)
Counter
φ
D
_
Digital Controller
Relay
fc
ADC
Vref
+
φm
(b) CCM or DCM boost converter
Figure 5: Block diagram of the phase detector. A system clock, Clk,
is used to measure the time shift between Vx(finj) and Vy(finj).
Figure 6: Experimental prototypes used for testing stability monitor (a)
Synchronous buck converter, (b) CCM or DCM boost converter.
862
Authorized licensed use limited to: UNIVERSITY OF COLORADO. Downloaded on November 12, 2008 at 13:03 from IEEE Xplore. Restrictions apply.
TABLE I
SUMMARY OF EXPERIMENTAL STABILITY MONITORING RESULTS COMPARING STABILITY MARGINS BASED ON MODEL [21],
MEASURED VIA PROPOSED MONITORING LOOP AND MEASURED VIA TRADITIONAL ANALOG LOOP GAIN MEASUREMENT [1]
Digital Compensator
Measured
fc
Average
Measured φm
Measured
fc
Measured
φm
(z − 0.762)(z − 0.91)
(z + 0.4)(z − 1)
9.22 kHz
55.0°
9.20 kHz
51.4°
9.58 kHz
53.56°
(z − 0.831)(z − 0.728 )
z (z − 1)
13.8 kHz
27°
13.89 kHz
28.4°
14.2 kHz
28.7°
(z − 1)
1.20 kHz
85.1°
1.23 kHz
80.7°
1.78 kHz
84.8°
(z − 0.8)
(z − 1)
6.87 kHz
65.6°
6.42 kHz
61.5°
6.68 kHz
67.6°
0.035
2. 0
z
The input injection magnitude, δ, is continuously
updated based on the injection amplitude controller
shown in Fig. 4 until the output voltage error Verr[n] is
minimum possible, +/- 1 LSB. In the experimental
prototypes, this amounts to +/- 0.4% output voltage
perturbation in the buck converter and +/- 1.6%
perturbation in the boost converter. The speed of the
input amplitude controller has been designed to be faster
then the monitoring control loop.
Given the described experimental systems, Table I
summarizes the experimental performance of the stability
margin monitor with four different power stage
configurations. In Table I, finj and φm based on the
proposed monitoring approach closely match the
expected values for each case, based on the discrete-time
model of [21]. Further, the monitored phase margin
never deviates more than 11.25° from the predicted value,
thus satisfying the expected error given by (20). Note that
the phase margin results given in Table I are averaged
The nominal power stage parameters of the buck
converter are given in Fig. 6(a). The buck converter
output voltage ADC is a TI-THS1030 with an effective
output voltage LSB resolution of 20 mV. The nominal
switching frequency is 100 kHz.
Nominally, the boost converter power stage
parameters are as shown in Fig. 6(b) with Vg and L
depending on the mode of operation (CCM or DCM). In
CCM, Vg_CCM = 15V and LCCM = 100 μH. In DCM,
Vg_DCM = 10V and LDCM = 20 μH. The boost converter
ADC is an AD7822 with an effective output voltage
resolution of 512 mV. As with the buck converter, the
nominal switching frequency is 100 kHz.
In all power stages, the system clock frequency is
fclk = 50 MHz, which from (13) implies
q PM = 360°
fc
= 7.2 × 10 −6 ( f c ) .
50 MHz
(16)
The matched band-pass filters were implemented using
the following transfer function
Gbp ( z ) = 0.00195
z −1
z 2 + 1.989 z − 0.998
10
8
.
(17)
6
(a) Measured crossover frequency [kHz]
Where fsample = 32 finj, i.e., γ = 32. Based on (11) and (12),
the filter pass-band center frequency and Q-factor are
80°
60°
40°
f sample
20°
f pb =
Analog Injection
Results
Analytical
φm
2.0693
2.8203
Digital Stability
Margin Detection
Analytical
fc
Gc(z)
System 1
Buck
Converter
System 2
Buck
Converter
System 3
CCM Boost
Converter
System 4
DCM Boost
Converter
Analytical Stability
Margins
32
,
PM_error
(b) Measured phase margin
(18)
10
Q = 100 = 40dB .
(19)
8
6
Further, from (15) and (18), the maximum possible phase
error expected in hardware is
PM _ error ≤
360°
= 11.25° .
32
0.004
0.008
0.012
0.016
(c) Injection amplitude [DPWM LSB’s]
0.02
Time (s)
Figure 7: Experimentally observed dynamic performance of the
stability margin monitoring control loop. The dynamic response to a
change from Vg = 12V to Vg = 8V with the compensator of System 1:
(a) crossover frequency fc, (b) Phase margin φm with maximum
expected phase error based on (20), (c) Injection amplitude δ.
(20)
863
Authorized licensed use limited to: UNIVERSITY OF COLORADO. Downloaded on November 12, 2008 at 13:03 from IEEE Xplore. Restrictions apply.
monitor recognizes the bus voltage change and updates
the outputs accordingly. The high frequency noise seen
in the monitored phase margin is an artifact of the bandpass filter sample rate selection discussed previously.
Note however that the magnitude of the high frequency
noise is always less than the error predicted by (20), as
expected.
One advantage to the proposed solution to monitoring
stability margins is that the system does not require open
loop or steady-state operation. This allows the monitor to
run and converge despite power stage transients (load,
line, etc.). Figure 8 shows the load transient response of
System 1 of Table I (buck converter) and System 4 of
Table I (DCM boost converter) with and without the
stability margin monitor. First, Fig. 8(a) is the load
transient response, from 2.5A to 0A, of System 1 without
the phase margin monitor. As expected, the load
transient causes a deviation in output voltage, which
returns to steady state after some time due to the action of
the feedback loop. In Fig. 8(c), the same load transient is
imposed as in Fig. 8(a), but with the stability margin
over 100 samples. Table I also shows measured results
based on the standard analog injection technique,
obtained by introducing an analog voltage injection, Vz, at
the converter output (Point A in Fig. 6), with the digital
stability monitor disabled. The results from the standard
analog injection technique indicate close matches with
the measurements from the proposed digital technique
and the discrete-time model. Based on the cases tested in
Table I, worst case injection frequency and phase margin
resolution can be determined from (8) and (13). Based on
the highest crossover frequency for the tested systems,
the injection frequency resolution is always greater than 4
Hz while the phase margin resolution is always greater
then 0.1°.
Figure 7 shows the experimentally observed
dynamics of finj, φm and δ, captured in Chipscope (a
Xilinx embedded FPGA logic analyzer), under a power
stage line transient. In particular, Fig. 7(a), Fig. 7(b) and
Fig. 7(c) show a bus voltage change in the synchronous
buck converter power stage from 12V to 8V with the
compensator of System 1. Under this change, the
1 LSB
(c) System 1 (synchronous buck) with
monitoring loop
(a) System 1 (synchronous buck) without
monitoring loop
1 LSB
(b) System 4 (DCM boost) without
monitoring loop
(d) System 4 (DCM boost) with monitoring
loop
Figure 8: Output voltage and inductor current waveforms: (a) System 1 (synchronous buck) without crossover frequency and phase margin
monitoring during a 2.5A Æ 0A load transient, (b) System 4 (DCM boost) without the crossover frequency and phase margin monitor
during a 3A Æ 0.05A load transient (c) System 1 with crossover frequency and phase margin monitoring during a 2.5A Æ 0A load
transient, (d) System 4 with the crossover frequency and phase margin monitor during a 3A Æ 0.05A load transient
864
Authorized licensed use limited to: UNIVERSITY OF COLORADO. Downloaded on November 12, 2008 at 13:03 from IEEE Xplore. Restrictions apply.
monitor control loop running. Note the oscillation
imposed by Vz is only +/- 1 LSB due to the action of the
feedback loop controlling δ. Also, note that the frequency
of Vz is equal to the crossover frequency. Load transient
results are presented in Fig. 8(b) and Fig. 8(d) for System
4 of Table I, the DCM boost converter. Since the boost
converter is operating in DCM, the load transient (from
3A to 50mA) significantly affects the crossover
frequency, as indicated by the oscillation frequency
before and after the transient in Fig. 8(d). Based on the
discrete-time model [21], the expected crossover
frequency after the load transient is 4.1 kHz, which is
approximately the frequency of oscillation seen in Fig.
8(d).
In Fig. 8(c) and Fig. 8(d), notice the output voltage
perturbation combined with the action of the integrator in
the PID compensator centers the DC value of the output
voltage in the zero error bin with accuracy related to the
DPWM resolution rather than the ADC resolution, as
discussed previously. Since in general the DPWM
resolution is finer than the ADC resolution, this equates
to more precise DC regulation accuracy with the imposed
output voltage oscillation than without.
As a final note, the hardware required to implement
all of the above described blocks is summarized in Table
II. As indicated, to implement the entire stability margin
monitor requires a relatively modest gate count and no
additional memory.
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
TABLE II
REQUIRED LOGIC RESOURCES TO IMPLEMENT
DIGITAL STABILITY MONITOR
Function
Logic Gates
Injection/Clock Generator
1262
Band-pass Filters
2288
Peak Detectors
1220
Integral Compensator
1034
Amplitude Controller
1448
7252
Total
III.
[12]
[13]
[14]
[15]
CONCLUSIONS
This paper presents a practical method for
continuously monitoring the crossover frequency and
phase margin in digitally controlled switching power
converters. The proposed approach does not require open
loop operation and is capable of converging to correct
results in the presence of load transients or other
disturbances. Further, the stability margin monitoring
requires and ensures that only +/- 1 LSB output voltage
perturbation is caused by the monitor. Experimental
results are presented for four different system
configurations indicating close matches between
monitored and expected crossover frequencies and phase
margins.
Experimental results are also presented
showing the observed output voltage and inductor current
during a load transient, indicating that the control loop is
unaffected by disturbances.
[16]
[17]
[18]
[19]
[20]
[21]
REFERENCES
[1]
[2]
R.D.Middlebrook, “Measurement of Loop Gain in
Feedback Systems,” Int. J. Electronics, 1975, pp. 485-512.
D. Maksimovic, R. Zane and R. Erickson, “Impact of
digital control in power electronics,” in Proc. IEEE
International Symposium on Power Semiconductor
Devices & ICs, May 2004, pp. 13-22.
B. Johansson and M. Lenells, “Possibilities of obtaining
small-signal models of DC-to-DC power converters by
means
of
system
identification,”
in
Proc.
Telecommunications Energy Conference, 2000, pp. 65-75.
B.Miao, R. Zane and D. Maksimovic, “Practical on-line
identification of power converter dynamic responses,” in
Proc. IEEE Applied Power Electronics Conference,
March 2005, pp. 57-62.
J. Morroni, A. Dolgov, M. Shirazi, R. Zane and D.
Maksimovic, “Online Health Monitoring in Digitally
Controlled Power Converters,” in Proc. IEEE Power
Electronics Specialist Conference, June 2007, pp. 112118.
A. Leva, “PID autotuning algorithm based on relay
feedback,” in Proc. IEEE Conf. on Decision and Control,
Dec. 2003, pp. 66-75.
M. Shirazi, L. Corradini, R. Zane, P. Mattavelli and D.
Maksimovic, “Autotuning techniques for digitally
controlled point-of-load converters with wide range of
capacitive loads,” in Proc. IEEE Applied Power
Electronics Conference, Feb. 2007, pp. 14-20.
I. Kaya and D.P. Atherton, “Exact parameter estimation
from relay autotuning under static load disturbances,” in
Proc. IEEE American Control Conference, 2001, pp.
3274-3279.
W. Stefanutti, P. Mattavelli, S. Saggini and M. Ghioni,
“Autotuning of Digitally Controlled Buck Converters
based on Relay Feedback”, IEEE Trans. Power Electron.,
Vol. 22, pp. 199-207, Jan. 2007.
L. Corradini, P. Mattavelli and D. Maksimovic, “Robust
Relay-Feedback Based Autotuning for DC-DC
Converters,” in Proc. IEEE Power Electronics Specialist
Conference, June 2007, pp. 2196-2202.
Z. Zhao and A. Prodic, “Limit-Cycle Oscillations Based
Auto-Tuning System for Digitally Controlled DC-DC
Power Supplies,” IEEE Trans. Power Electron, pp. 22112222, Nov. 2007.
R. Button, “Intelligent Systems for Power Management
and Distribution,” NASA 2002.
B.Bailey, “Power System Protection – Power Monitoring
– Data Logging – Remote Interrogation System,” ICPS
1989.
S. Javidi, E. Gholdston, P. Stroh, “Space Station Freedom
Power Management and Distribution Design Status,”
ECEC 1989.
J. Morroni, R. Zane, D. Maksimovic, “Adaptive Tuning of
Digitally Controlled Switched-Mode Power Supplies
based on Desired Phase Margin,” in Proc. IEEE Power
Electronics Specialist Conference, June 2008.
R.W. Erickson and D. Maksimovic, “Fundamentals of
Power Electronics,” 2nd Edition, Boston: Kluwer
Academic Publishers, 2001.
Z. Zhao, A. Prodic, “Non-zero Error Method for
Improving Output Voltage Regulation of Low-Resolution
Digital Controllers for SMPS,” in Proc. IEEE Applied
Power Electronics Conference, March 2008, pp. 11061110.
A. Peterchev, S. Sanders, “Quantization resolution and
limit cycling in digitally controlled PWM converters,
“IEEE Trans. Power Electron., vol. 18 pp. 301-308, Jan.
2003.
H. Peng, A. Prodic, E. Alarcon and D. Maksimovic,
“Modeling of quantization effects in digitally controlled
DC-DC converters,” IEEE Trans. Power Electron., vol.
22, pp. 208-215, Jan. 2007.
W. Lindsey and C. Chie, “A Survey of Digital PhaseLocked Loops,” Proceedings of IEEE, pp. 410-431, April
1981.
D. Maksimovic and R. Zane, “Small-signal discrete-time
modeling of digitally controlled DC-DC converters,”
IEEE Trans. Power Electronics, Letters, pp. 2552-2556,
Nov. 2007.
865
Authorized licensed use limited to: UNIVERSITY OF COLORADO. Downloaded on November 12, 2008 at 13:03 from IEEE Xplore. Restrictions apply.
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement