3.
If the DC1134 will operate above 10A, replace
Q1 with a suitable PowerPAK SO-8 package
MOSFET such as the Si7880ADP suitable for
up to 20A. The SO-8 pads on the front of the
board include the extra heat-sinking area this
package requires. Alternately, for low power
applications, Q1 may be replaced with a
MOSFET in a SSOT-6 package using pads on
the back of the board.
4.
The “TIMER SELECT” jumper chooses the
startup time limit. It may enable either the ad-
justable timer with external timer capacitor CT
or the built in internal timer with a fixed startup
time 100ms. CT is stuffed with a 0.68µF ca-
pacitor at the factory, which provides a 8ms
startup time. To customize the startup time, se-
lect CT = T
START
/12.3(ms/µF).
5.
The soft start capacitor, CSS, sets the current
ramp rate at which the inrush current will in-
crease when turning on. CSS is stuffed with a
68nF capacitor at the factory, which provides a
5mV/ms inrush sense resistor voltage ramp or
0.5A/ms for both rails.
6.
The LTC4222 does not require an RC network
on the GATE pin for compensation. However,
an RC network may be used to limit the rate at
which the GATE rises to provide an inrush cur-
rent below the internal current limit. Pads for
R6, R106, CG1 and CG2 on the back of the
board are available for this purpose. R6 and
CG1, as well as R106 and CG2 may be stuffed
with 100K resistor, and CG = C
LOAD
*
20uA/I
INRUSH
for this purpose.
7.
The ADIN1 and ADIN2 turrets may be used to
provide signals to take 10 bit measurements of
any voltage. Jumpers “ADIN1 SELECT” and
“‘ADIN2 SELECT” choose between measuring
the input rail voltage or measuring the voltage
on the ADIN1, ADIN2 turrets respectively. The
rail input voltages are measured through resis-
tor dividers. The +12V divider is assembled to
a 15.4V full scale range to match the range of
the Q1 SOURCE voltage measurement. The
+3.3V divider is assembled to a 4.3V full scale
range to match the range of the Q101
SOURCE voltage measurement. To change
the full scale of the input measurement, select
R11 or R111 as
R = V
RANGE
*12.4k/1.23-12.4k.
The signals at the ADIN1 and ADIN2 turrets
are also measured through a voltage dividers,
but the bottom resistor is not stuffed at the fac-
tory, defaulting to a 1.28V full scale voltage. To
select a larger full scale voltage, choose R17
and R117 as
R = 12.8/(V
RANGE
-1.28)k.
8.
If the system uses short connector pins to
sense board insertion, place “EN1” and “EN2”
jumpers
to position SHORT PIN1 and SHORT
PIN2, respectively. Using the SHORT PIN1
and SHORT PIN2 turrets, connect the afore-
mentioned short pins. Those are direct connec-
tions to the LTC4222's EN1 and EN2 pins; if
deleterious voltages are anticipated, add a se-
ries resistor and clamping. EN1 and EN2 are
good for -0.3 to +12V on their own. EN1 and
EN2 are also logic compatible with a 1.235V
threshold and 130mV hysteresis.
9.
The I
2
C address, assigned to the board, is se-
lected on the board by using jumpers “ADR0”,
“ADR1” and “ADR2” to pull the address pins
high, low, or allow them to float. An address ta-
ble is shown in the data sheet. The evaluation
software automatically scans and identifies the
I2C address, regardless of the setting.
10.
After any necessary component changes have
been made, connect suitable loads between
+12V/5A OUTPUT and GND and +3.3V/5A
OUTPUT and GND. This may be a passive re-
sistive load or an active electronic load box. If
long leads are present between the DC1134
and load, install 10µF or more at C1 and C101.
This should eliminate the chance of MOSFET
oscillation and large negative excursions at
+12V/5A OUTPUT.
11.
Connect power supplies capable of supplying
10A to the +12V INPUT and GND turrets and