16
LT1579
APPLICATIONS INFORMATION
WUU
U
Five milestones are noted on the timing diagram. Time A
is where the primary input voltage drops enough to trip the
low-battery detector LB1. The trip threshold for LB1 is set
at set at 5.5V, slightly above the dropout voltage of the
primary input. At time B, the BACKUP flag goes low,
signaling the beginning of the transition from the primary
source to the secondary source. Between times B and C,
the input current makes a smooth transition from V
IN1
to
V
IN2
. By time C, the primary battery has dropped below the
point where it can deliver useful current to the output. The
primary input will still deliver a small amount of current to
the load, diminishing as the primary input voltage drops.
By time D, the secondary battery has dropped to a low
enough voltage to trip the second low-battery detector,
LB2. The trip threshold for LB2 is also set at 5.5V, slightly
above where the secondary input reaches dropout. At time
E, both inputs are low enough to cause the LT1579 to enter
dropout, with the DROPOUT flag signaling the impending
loss of output regulation. After time E, the output voltage
drops out of regulation.
Some interesting things can be noted on the timing
diagram. The amount of current available from a given
input is determined by the input/output voltage differen-
tial. As the differential voltage drops, the amount of
current drawn from the input also drops, which slows the
discharge of the battery. Dropout detection circuitry will
maintain the maximum current draw from the input for the
given input/output voltage differential. In the case shown,
this causes the current drawn from the primary input to
approach zero, though never actually dropping to zero.
Note that the primary begins to supply significant current
again when the output drops out of regulation. This occurs
because the input/output voltage differential of the pri-
mary input increases as the output voltage drops. The
LT1579 will automatically maximize the power drawn
from the inputs to maintain the highest possible output
voltage.
Thermal Considerations
The power handling capability of the LT1579 is limited by
the maximum rated junction temperature (125°C). Power
dissipated is made up of two components:
1. The output current from each input multiplied by the
respective input to output voltage differential:
(I
OUT
)(V
IN
– V
OUT
) and
2. Ground pin current from the associated inputs multi-
plied by the respective input voltage: (I
GND
)(V
IN
).
If the primary input is not in dropout, all significant power
dissipation is from the primary input. Conversely, if SS has
been asserted to minimize power draw from the primary,
all significant power dissipation will be from the second-
ary. When the primary input enters dropout, calculation of
power dissipation requires consideration of power dissi-
pation from both inputs. Worst-case power dissipation is
found using the worst-case input voltage from either input
and the worst-case load current.
Ground pin current is found by examining the Ground Pin
Current curves in the Typical Performance Characteris-
tics. Power dissipation will be equal to the sum of the two
components above for the input supplying power to the
load. Power dissipation from the other input is negligible.
The LT1579 has internal thermal limiting designed to
protect the device during overload conditions. For con-
tinuous normal load conditions, the maximum junction
temperature rating of 125°C must not be exceeded. It is
important to give careful consideration to all sources of
thermal resistance from junction to ambient. Additional
heat sources nearby must also be considered.
Heating sinking for the device is accomplished by using
the heat spreading capabilities of the PC board and its
copper traces. Copper board stiffeners and plated through-
holes can also be used to spread the heat. All ground pins
on the LT1579 are fused to the die paddle for improved
heat spreading capabilities.
The following tables list thermal resistances for each pack-
age. Measured values of thermal resistance for several
different board sizes and copper areas are listed for each
package. All measurements were taken in still air on 3/32”
FR-4 board with one ounce copper. All ground leads were
connected to the ground plane. All packages for the
LT1579 have all ground leads fused to the die attach
paddle to lower thermal resistance. Typical thermal
