Datasheet

12
LTC1530
1530fa
Note that while the required R
DS(ON)
values suggest large
MOSFETs, the power dissipation numbers are only 1.39W
per device or less—large TO-220 packages and heat
sinks are not necessarily required in high efficiency appli-
cations. Siliconix Si4410DY or International Rectifier
IRF7413 (both in SO-8) or Siliconix SUD50N03 or Motorola
MTD20N03HDL (both in DPAK) are small footprint sur-
face mount devices with R
DS(ON)
values below 0.03Ω at 5V
of V
GS
that work well in LTC1530 circuits. With higher
output voltages, the R
DS(ON)
of Q1 may need to be signifi-
cantly lower than that for Q2. These conditions can often
be met by paralleling two MOSFETs for Q1 and using a
single device for Q2. Using a higher P
MAX
value in the
R
DS(ON)
calculations generally decreases the MOSFET
cost and the circuit efficiency and increases the MOSFET
heat sink requirements.
In most LTC1530 applications, R
DS(ON)
is used as the
current sensing element. MOSFET R
DS(ON)
has a positive
temperature coefficient. Therefore, the LTC1530 I
MAX
sink
current is designed with a positive 3300ppm/°C tempera-
ture coefficient. The positive tempco of I
MAX
provides first
order correction for current limit vs temperature. There-
fore, current limit does not have to be set to an increased
level at room temperature to guarantee a desired output
current at elevated temperatures.
Table 1 highlights a variety of power MOSFETs that are
suitable for use in LTC1530 applications.
Inductor Selection
The inductor is often the largest component in an LTC1530
design and must be chosen carefully. Choose the inductor
value and type based on output slew rate requirements
and expected peak current. The required output slew rate
primarily controls the inductor value. The maximum rate
of rise of inductor current is set by the inductor’s value, the
input-to-output voltage differential and the LTC1530’s
maximum duty cycle. In a typical 5V input, 2.8V output
application, the maximum rise time will be:
DC
VV
LL
MAX
IN OUT
−






=
185. A
sµ
either Q1 or Q2 with the power dissipation split up accord-
ing to the duty cycle:
DC Q
V
V
DC Q
V
V
VV
V
OUT
IN
OUT
IN
IN OUT
IN
()
()
1
21
=
=− =
−
()
The R
DS(ON)
required for a given conduction loss can now
be calculated by rearranging the relation P = I
2
R.
R
P
DC Q I
VP
V
I
R
P
DC Q I
VP
VV
I
DS ON Q
MAX Q
MAX
IN MAX Q
OUT
MAX
DS ON Q
MAX Q
MAX
IN MAX Q
IN OUT
MAX
()
()
()
()
()
()
()
()
1
1
2
1
2
2
2
2
2
2
1
2
=
[]
()
=
()
[]
()
()
=
[]
()
=
()
[]
−
()
()
P
MAX
should be calculated based primarily on required
efficiency or allowable thermal dissipation. A high efficiency
buck converter designed for the Pentium II with 5V input
and a 2.8V, 11.2A output might allow no more than 4%
efficiency loss at full load for each MOSFET. Assuming
roughly 90% efficiency at this current level, this gives a P
MAX
value of:
(2.8)(11.2A/0.9)(0.04) = 1.39W per FET
and a required R
DS(ON)
of:
R
VW
VA
R
VW
VV A
DS ON Q
DS ON Q
()
()
.
..
.
.
..
.
1
2
2
2
5139
2 8 11 2
0 020
5139
528112
0 025
=
()




=
=
()
−
()




=
Ω
Ω
APPLICATIO S I FOR ATIO
WUUU