Design Note 59
Issue 1 December 2000
Load Switch
Extended battery life is becoming more
important in today’s applications.
Load Switch
C o n s u m e r s a r e d e m a n d i ng mo r e
complex features resulting in greater
power demands in products such as
laptop computers, mobile phones, etc.
Load switching is an effective technique
for disconnecting power to electronic
subsystems, that are not required,
extending battery life.
It is important that the load switch has
minimal losses in order to maximise
efficiency and extend battery life for the
end user. Size is also critical, with the
consumer requiring smaller, lighter
portable equipment.
Load Switch
Load Switch
Using Zetex High Density MOSFETs or
SuperSOT Transistors for load
switching, high efficiency and reduced
space can be achieved. This is because
their low on-resistance or saturation
voltage enable high current density in
small packages.
Load Switch
Theory of Operation.
Figure 1. shows a typical portable
s ys t em ar c hi t ec t ure. The var ious
sections of the system are selected
when required, via the load switch. The
power management circuitry controls
the load switch and thus power to the
relevant subsystem.
Figure 1.
Typical Portable System Architecture.
DN 59 - 1
Design Note 59
Issue 1 December 2000
Figure 2a.
Typical MOSFET Load Switch.
Figure 2b.
Typical Bipolar Load Switch.
Figures 2a and 2b are typical load switch
schematics for MOSFET and bipolar
technologies. The capacitors, C1, may
not be required in some applications
(see inrush current section).
The load switch consists of a P-Channel
pass element, Q1 (PNP transistor for
bipolar), which is controlled by a low
power, logic level switch, Q2 (NPN
transistor for bipolar). With a logic ‘low’
at the input to Q2, Q1 is held off via R1.
When logic ‘high’ is at the input to Q2,
Q1’s gate is pulled to zero volts and is
switched on, allowing power to the
subsystem. Note that it is important,
when using MOSFETs as load switches,
to ensure that the input voltage is higher
than the output voltage. The intrinsic
body diode will conduct if forward
biased which results in significant
current flow. This could be a problem in
multiple power source systems such as
battery chargers where excess battery
drain could be introduced.
Thermal Considerations.
Load switches are selected by their
power handling capability and low
on-state losses. Optimisation for the
load switch is critical. If the junction
temperature is increased the device
could be damaged. Therefore it is
important to ensure that the load switch
is designed for its operating
Design Example
This is a design example using the Zetex
High Density MOSFET, ZXM64P02X.
(Note 1)
Ambient temperature, Ta = 70°C.
Isteady-state(max) = 1A.
VGS = 5V.
Rth(j-a)max = 69.4°C/W (supplied by
Zetex) (Note 2).
DN 59 - 2
Design Note 59
Issue 1 December 2000
Now some general calculations can be
Tj = Ta + (Rth(j-a) x Pd(practical)) = °C
Tj = 70°C + (69.4°C/W x 135mΩ) = 79.4°C.
Power Dissipation,
Pd(maximum allowable) =
(Tj(max) – Ta(max)) / Rth(j-a) = Watts.
Pd(maximum allowable) =
(150°C – 70°C) / 69.4°C/W = 1.152W.
Note 1. For bipolar transistors the same
procedure can be used except that
Pd(practical) = Isteady-state x VCE(sat) @
F o r a M O S F E T t h e o n - r e s i s t a nc e
increases by 50% when the junction
temperature reaches 150°C. Therefore
assume worse case on-resistance.
RDS(on)worse-case =
1.5 xRDS(on)max at 25°C = mΩ
Tj = 79.4°C giving a 47% design margin.
Note 2. This figure is measured by
mounting the device on a 25 x 25mm
area of FR4 PCB of full copper. If you do
not use the same area, derate using
thermal derating graphs supplied by
Inrush Current.
RDS(on)worse-case =
1.5 x 90mΩ= 135mΩ.
Where Pd(practical) =
Isteady-state2 x RDS(on)worse-case = W.
Pd(practical) = 1A2 x 135mΩ = 135mW.
By comparing the absolute allowable
power dissipation and the practical
power dissipation you can see that the
MOSFET would be suitable for this
Load switches which turn on into low
ESR capacitors can have high inrush
current, this can be detrimental to
overall system performance and
reliability. The inrush current transients
are limited by the ESR of the output
capacitor and the on-resistance of the
pass element and can be in the order of
several amperes. There are a number of
ways in which to slow down the turn on
of the device and resolve this problem.
The simplest is to add a capacitor, C1,
between the gate and drain (base and
collect or f or bipolar) of the pass
element, Q1, as shown in fig 2.
It is important to know the design
margin in the event of future
environment changes.
DN 59-3
Design Note 59
Issue 1 December 2000
In Figure 3. a comparison is performed
between MOSFET and Bipolar
technologies with equal die area.
The results clearly indicate that for a
given area of silicon there is a point
where the two technologies crossover.
At the lower levels the base drive
required by PNP Bipolar transistors
dominates the los ses in the pass
element. As the load current increases
and the ratio of base current to load
current reduces the efficiency is seen to
improve and hold up.
In a MOSFET the drive losses are
negligible and therefore the device
performs best at lower currents. As the
load current increases the on-resistance
has a larger effect on the losses and the
efficiency falls off.
Efficiency (%)
Iout (A)
Figure 3.
P-Channel and PNP. Optimised for 1A
operation with equal die area.
DN 59 -4