MAX8513 MAXIM [Maxim Integrated Products], MAX8513 Datasheet - Page 23

• Manufacturer Part Number: MAX8513
• Description: Wide-Input, High-Frequency, Triple-Output Supplies with Voltage Monitor and Power-On Reset
• Manufacturer: MAXIM [Maxim Integrated Products]

When the output capacitance is comprised of parallel-
ing n number of identical capacitors whose values are
C
Thus the resulting f
capacitor.
The crossover frequency (f
when the closed-loop gain is equal to unity, should be
the smaller of 1/5th the switching frequency or 100kHz
(see the Switching-Frequency Setting section):
The loop-gain equation at the crossover frequency is:
where G
G
The loop compensation is affected by the choice of out-
put-filter capacitor used, due to the position of its ESR
zero frequency with respect to the desired closed-loop
crossover frequency. Ceramic capacitors are used for
higher switching frequencies (above 750kHz) because
of low capacitance and low ESR; therefore, the ESR
zero frequency is higher than the closed-loop crossover
frequency. While electrolytic capacitors (e.g., tantalum,
solid polymer, oscon, etc.) are needed for lower switch-
ing frequencies, because of high capacitance and ESR,
the ESR zero frequency is typically lower than the
closed-loop crossover frequency. Thus the compensa-
tion design procedure is separated into two cases:
The modulator gain at f
Since the crossover frequency is lower than the output
capacitors’ ESR zero frequency and higher than the LC
double-pole frequency, the error-amplifier gain must
have a +20dB/decade slope at f
slope of the error amplifier at crossover then adds to
Wide-Input, High-Frequency, Triple-Output Supplies
EACH
MOD(
Case 1: Ceramic Output Capacitor (operating at
f
with ESR of R
c)
EA(fc)
is the power modular gain at f
G
MOD fc
C
R
high switching frequencies, f
OUT
ESR
is the error-amplifier gain at f
G
f
( )
C
EA fc
______________________________________________________________________________________
ZESR
≤
=
=
( )
ESR_EACH
with Voltage Monitor and Power-On Reset
=
C
n
R
f
G
S
5
G
ESR EACH
MOD fc
is:
×
MOD DC
is the same as that of each
or
C
C
_
n
100
EACH
), which is the frequency
( )
(
, then:
kHz
C
)
= 1
. This +20dB/decade



f
and
PMOD
f
C
C
.



2
ZESR
C
, and
> f
C
)
the -40dB/decade slope of the LC double pole, and the
resultant compensated loop crosses over at the
desired -20dB/decade slope. The error amplifier has a
dominant pole at very low frequency (≈0Hz), and two
separate zeros at:
and poles at:
The error-amplifier equivalent circuit and its gain vs.
frequency plot are shown below in Figure 3.
In this case, f
ers’ closed-loop crossover frequency, f
error-amplifier gain has a +20dB/decade slope between
f
The gain of the error amplifier between f
Figure 3. Case 1: Error-Amplifier Compensation Circuit (Closed-
Loop and Error-Amplifier Gain Plot)
Z2
f
P
f
Z
2
and f
1
V
GAIN
OUT1
(dB)
=
=
G
P2
2
EA fZ fZ
2
π
π
. The error-amplifier gain at f
(
×
C11
×
Z2
R
1
R1
R
-
1
4
1
3
CLOSED-LOOP GAIN
and f
×
R2
×
2
G
C
)
C
EA fc
R4
11
5
=
f
P1
Z1
( )
and f
and f
G
REF
are selected to have the convert-
f
Z2
EA fc
=
( )
Z
P
f
C
G
2
3
EA
MOD fc
f
=
Z
f
C
=
2
f
P2
2
1
R3
C12
π
2
( )
=
×
π
f
P3
R
f G
×
C MOD fc
C
C
3
(
Z1
C5
R
, occur when the
is:
×
EA GAIN
1
f
Z



and f
+
1
2
C
C
1
R
FREQUENCY
COMP
5
5
( )
4
)
×
+
Z2
×
C
C
C
12
12
is:
11



23