TM
MP1580 – 2A, 380KHz STEP-DOWN CONVERTER
Compensation
In this case, the switching frequency is 380KHz,
so use a crossover frequency, fC, of 40KHz.
Lower crossover frequencies result in slower
response and worse transient load recovery.
Higher crossover frequencies can result in
instability.
The system stability is controlled through the
COMP pin. COMP is the output of the internal
transconductance error amplifier. A series
capacitor-resistor combination sets a pole-zero
combination to control the characteristics of the
control system.
Choosing the Compensation Components
The values of the compensation components
given in Table 4 yield a stable control loop for
the output voltage and capacitor given.
The DC loop gain is:
VFB
AVDC = RLOAD × GCS × AVEA
×
VOUT
Table 4—Compensation Values for Typical
Output Voltage/Capacitor Combinations
Where AVEA is the transconductance error
amplifier voltage gain, 400 V/V, GCS is the
current sense gain, (roughly the output current
divided by the voltage at COMP), 1.95 A/V and
RLOAD is the load resistance (VOUT / IOUT where
VOUT
2.5V 22µF Ceramic 7.5kꢀ 2.2nF None
3.3V 22µF Ceramic 10kꢀ 2nF None
15kꢀ 1.2nF None
33kꢀ 1nF None
200kꢀ 1nF 100pF
C2
R3
C3
C6
I
OUT is the output load current).
5V
22µF Ceramic
22µF Ceramic
The system has 2 poles of importance, one is
due to the compensation capacitor (C3), and
the other is due to the output capacitor (C2).
These are:
12V
560µF/6.3V
(30mꢀ ESR)
2.5V
3.3V
5V
560µF/6.3V
(30mꢀ ESR)
200kꢀ 1nF
250kꢀ 1nF
250kꢀ 1nF
82pF
56pF
27pF
GEA
fP1
=
2π× C3× AVEA
470µF/10V
(30mꢀ ESR)
Where P1 is the first pole and GEA is the error
amplifier transconductance (770µA/V).
220µF/25V
(30mꢀ ESR)
12V
and
To optimize the compensation components for
conditions not listed in Table 4, use the
following procedure:
1
2π × C2× RLOAD
fP2
=
Choose the compensation resistor to set the
desired crossover frequency. Determine the
value by the following equation:
The system has one zero of importance, due to
the compensation capacitor (C3) and the
compensation resistor (R3). The zero is:
2π × C2× fC VOUT
1
R3 =
×
fZ1
=
GEA × GCS
VFB
2π × C3×R3
If a large value capacitor (C2) with relatively
high equivalent-series-resistance (ESR) is
used, the zero due to the capacitance and ESR
of the output capacitor can be compensated by
a third pole set by R3 and C6. The pole is:
Putting in the known constants and setting the
crossover frequency to the desired 40KHz:
R3 ≈ 1.37 ×108 × C2× VOUT
Choose the compensation capacitor to set the
zero below ¼ of the crossover frequency.
Determine the value by the following equation:
1
fP3
=
2π × C6 × R3
0.22 × C2 × VOUT
The system crossover frequency (the frequency
where the loop gain drops to 1, or 0dB) is
important. A good rule of thumb is to set the
crossover frequency to approximately 1/10 of
the switching frequency.
C3 >
R3
MP1580 Rev. 3.0
12/5/2005
www.MonolithicPower.com
MPS Proprietary Information. Unauthorized Photocopy and Duplication Prohibited.
© 2005 MPS. All Rights Reserved.
7