Lab Experiment - Introduction To Switch Mode Power Supply

Download as pdf or txt
Download as pdf or txt
You are on page 1of 9

LABORATORY EXPERIMENT #14

AN INTRODUCTION TO SWITCHMODE POWER SUPPLIES

Discussion: AC voltage is very convenient to use for transmission and distribution


because of the ease of level-shifting using transformers. A transformer operates by
magnetically coupling two coils with typically different numbers of turns. If the number
of turns on the secondary winding is greater than the number on the primary winding,
then the transformer steps up the voltage. This is useful for transmission where, for a
given power being transmitted, we wish to have as large a voltage as possible so that the
corresponding current is small. A small current implies fewer I 2 R conduction losses and
less voltage drop between different points in the network. If the transformer has fewer
turns on the secondary, then the secondary voltage is less than the primary and we have a
step-down transformer. This is what we investigated in Experiment #13 as part of our
unregulated and regulated power supplies. Since in the 1890s DC had no comparable
mechanism for achieving voltage level shifting and therefore no comparable convenience
or efficiency, AC soon became dominate.
With the advent of switching technology, vacuum tubes and then semiconductors,
we now have the possibility of using controlled electronic switches, in concert with the
energy storage properties of inductors and capacitors, to realize DC voltage level shifting.
Now you might ask: cant we simply do this with voltage dividers to achieve lower DC
voltages? The answer is yes, but when significant amounts of power are required, this
approach quickly becomes very inefficient. A switchmode power converter that
efficiently steps down DC voltage is called a buck chopper. An alternative to the buck
chopper is the voltage regulator approach such as the LM317 that we investigated in Lab
#13. Now also interestingly, we have the capability of stepping up DC voltages
electronically. This device is called a boost converter and will be the focus of this lab
exercise.
iL

Vin

Rout Vout

Figure 1. Circuit Topology for Boost Converter

To appreciate the need for a boost converter, consider that you have a 9-volt
battery but that you require additional voltages to energize various electronic circuitry,
for instance 12V, 15V and maybe 24V. Now one solution could be to resize the battery
for the maximum voltage required (here 24V) and then use buck choppers to efficiently
generate the lower levels. If this approach is not an option, then we need to consider the
boost converter. A separate lecture on the boost converter documents its principle of
operation, shows representative current waveforms, and develops design relationships to

aid in the selection of circuit components. This will not be repeated here in detail, but
some key results will be highlighted.
iL
switch
open

switched closed

I max

I min

t
DTs

Ts

Figure 2. Boost Converter Steady-State Inductor Current

The circuit for the boost converter is illustrated in Figure 1. For a given switching
period denoted by Ts , the switch is closed for a portion of the time, ton , and is open for
the remainder. This ratio of on time to switching period is termed the duty cycle,
designated by D. We then found that for continuous inductor current that the output
voltage was related to the input voltage by the following relation:
Vout =

Vin
1 D

Since D is constrained to be between 0 and 1, this implies that Vout Vin , hence a boost
converter. A representative waveform for the inductor current is illustrated in Figure 2.
For this current to be in the continuous mode, it must not fall to zero (that is, I min > 0 A ).
As indicated in lecture, we prefer to operate the converter in this mode. As a
consequence, we derived a necessary and sufficient condition to ensure that the inductor
current would remain continuous for all anticipated load conditions. This relationship
helped us identify the minimum inductance required for the converter.
R
(1 D ) D
= out ,max
2 fs
2

Lcrit

1
is the switching frequency and Rout ,max is the largest value of output
Ts
resistance anticipated. In our exercise today, our choice of L will be constrained and we
will resolve the equation to identify the minimum permissible switching frequency.
In the above, f s =

R
(1 D ) D
= out ,max
2L
2

f s ,min

The corresponding equations for the minimum and maximum inductor currents then
become.
I min =

I max =

Vin

Rout (1 D )

Vin

Rout (1 D )

Vin D
2 Lf s

Vin D
2 Lf s

We also derived a design requirement on the output capacitance based on the permitted
peak-to-peak ripple in the output voltage. Here we found that the capacitor had to be
sized to satisfy the following minimum constraint.
Cmin =

Vout D
Rout ,min f sVout , pp ,max

We could rearrange this relationship to identify the output ripple given a value of
capacitor.
Vout , pp ,max =

Vout D
Rout ,min Cf s

The final element that we need to discuss before pushing on to the exercise is the
implementation of the switch in Figure 1. This can be achieved by suitable selection of
Bipolar Junction Transistor (BJT), but today we will introduce a new friend called a
Power FET (Field-Effect Transistor). This is a voltage-controlled device (vice the
current-controlled BJT switch) that has very fast switching characteristics. Like the BJT,
the Power FET has three terminals: the gate, the drain, and the source. To make it look
like a switch, we need to apply an 8V signal to the gate to turn it on and a zero or
negative voltage (not less than 8V) to turn it off. The Tektronix CFG function generator
will be used to realize the pulse waveform and will enable us to adjust the duty cycle so
we can adjust the output voltage.
Section A. Theoretical Design
Given: The boost converter that you will build and evaluate must satisfy the following:
the input voltage is +5V, the output voltage must be adjustable for values of +10V, +15V,
and +20V, you must use a 1mH inductor, the output must stay continuous for load

resistances ranging from 500 Rout 2k , and the output capacitor has been selected
to be a 47 F aluminum electrolytic.
1. For each output voltage, identify the required duty cycle, calculate the minimum
permissible switching frequency, and then evaluate the theoretical output peak-topeak voltage ripple.

Vout
10V
15V
20V

fsmin

Voutpp

2. From the above, identify the switching frequency that will lead to continuous
current mode for each output voltage condition (round the value up to the nearest
integer multiple of 5kHz; for example, 145kHz or 150kHz). Calculate the
theoretical switch on-time based on the required duty cycle and the identified
switching frequency. Recalculate the anticipated maximum peak-to-peak output
ripple for the three output voltages. Finally, calculate the maximum and minimum
output power.
Switching Frequency: _______________

Vout
10V
15V
20V

Ton

Voutpp

Pmin

Pmax

3. Have the instructor verify the above calculations.


INSTRUCTOR VERIFICATION: _____________________

1mH

UP1004
IRF510 FET
TOP VIEW

5V
Protoboard
Supply

D
G

47uF

Rout

S
Function
Generator

500
or
2k

Electrolytic
G D S

Figure 3. Boost Converter Evaluation Circuit

Section B. Building the Boost Converter


1. Build the circuit illustrated in Figure 3. DO NOT hook up the function generator
signal or apply proto-board power until explicitly instructed.
a. Use the 1mH, 1A toroidal core inductor for L.
b. Use the proto-board +5V supply for the input.
c. The pin-out for the IRF510 FET is indicated in Figure 3. Note, to
understand what is the top, consider (but dont do) bending the device
over and mounting it to a heatsink.
d. A fast-recovery diode is provided to you for D (a fast-recovery diode is
one specifically made for fast-switching applications). Make the
connection from the FET drain to the diode as short as possible to
minimize the wiring inductance between the two components.
e. Remember that the aluminum electrolytic capacitor is polarized and must
have its positive terminal connected to where the output voltage will be
positive.
f. We will implement the load resistance by either paralleling two 1k
resistors (to give 500 ) or by series connecting two 1k resistors (to
give 2k ). Determine the required wattage of the resistors.
Required Resistor Wattage: _____________
g. Set-up the circuit initially for a 500 load.
h. Use the Tektronix function generator and your scope to establish to a +8V
to 8V pulse waveform at the desired frequency with the duty cycle
necessary to produce +10V out.
2. Have the instructor verify your function generator waveform and your circuit
layout.
INSTRUCTOR VERIFICATION: _________________
3. Set up the oscilloscope to measure the output AC ripple and the Digital MultiMeter (DMM) to measure the DC output voltage.

Section C. Evaluating the Boost Converter Output


1. Energize the proto-board power supply. The output DC voltage should be one
diode drop below the input voltage (we have not yet applied the switching
command signal).
2. Apply the function generator output to the FET gate. Record the output DC
voltage and peak-to-peak ripple (note, do not include the switching transient
voltage in your measurement of peak-to-peak ripple).
Output DC Voltage with Theoretical Duty Cycle: _____________
Measured Peak-to-Peak Output Ripple: _____________
3. Adjust the symmetry knob on the function generator to vary the duty cycle and
achieve an output voltage of +10V. Put the function generator output on CH2 of
the scope and measure the actual duty cycle required to achieve +10V. Measure
the peak-to-peak output ripple. Repeat the measurements for output voltages of
+15V and +20V. The discrepancy between theory and practice comes about
because of our assumption of ideal components in the theoretical derivation. In
practice, a real inductor has some series resistance, a real switch and diode
have some voltage drop across them, and the dynamics of switching will also
create additional heat loss in both the transistor and the diode.

Vout
+10V
+15V
+20V

Dtheo

Dmeas

Voutpp

4. De-energize the function generator and then de-energize the proto-board supply.
Reconfigure the output load resistance so that the 1k resistors are in series. Reenergize the proto-board supply. Re-apply the function generator pulse waveform.
Repeat the measurements of step 3. Note, with different output power and current
flowing, we will have different losses and require a new duty cycle.

Vout
+10V
+15V
+20V

Dtheo

Dmeas

Voutpp

5. Demonstrate your last set of measurements to the instructor.


INSTRUCTOR VERIFICATION: _______________
6. De-energize the function generator and then the proto-board power supply.

Section D. Measuring the Inductor Current


1. In order to measure the inductor current, we will place a known resistor in series
with the inductor and then measure the voltage across that resistor using a
difference amplifier. Note, we cannot use a conventional inverting or noninverting amplifier since we are looking at the voltage across the resistor and not
at a voltage point referenced to ground.
2. Modify your circuit to include the elements shown in Figure 4.
a. Note that the LM741 op-amp will be powered from the +15V and 15V
supplies of the proto-board.
b. Precision 1% resistors are used in the differential amplifier to improve the
accuracy of the measurement.
c. Return the boost converter output resistance to 500 (two 1k in
parallel).
1 , 5%

1mH
Rest of circuit, same as Figure 3.
10k, 1%

5V
Protoboard
Supply

10k, 1%

+15V

7
6
10k, 1%

Vout

-15V
10k, 1%

Figure 4. Circuit to Measure Inductor Current

3. Have the instructor verify your circuit.


INSTRUCTOR VERIFICATION: _______________
4. Energize the proto-board power supply. Re-apply the function generator input.
Re-adjust the duty cycle to obtain +20V at the output. Calculate the DC output
current.
DC Output Current: ______________

For a 100% efficient converter, the following must hold.


Vin, DC I in, DC = Vout , DC I out , DC
Calculate the theoretical input DC current.
Theoretical DC Input Current: ______________
With the DMM, measure the DC value of the output of the differential amplifier.
Measured DC Input Current: _______________
Any differences are attributed to losses in the actual circuit.
5. Place the output of the differential amplifier on the scope (DC coupled). Note the
large spikes in the measured voltage waveform (also the current waveform since
this is a 1 resistor) at the switching instances. Attempt to ignore these spikes
and try to measure the values of I max and I min for this full-load condition.
Measured I max : ____________
Measured I min : ____________
6. Have the instructor verify your measurements.
INSTRUCTOR VERIFICATION: _______________
7. De-energize the function generator and then the proto-board power supply.
Reconfigure the output resistance for 2k .
8. Energize the proto-board power supply. Re-apply the function generator input. If
necessary, re-adjust the duty cycle to obtain +20V at the output. Calculate the DC
output current.
DC Output Current: ______________
For a 100% efficient converter, the following must hold.
Vin, DC I in, DC = Vout , DC I out , DC
Calculate the theoretical input DC current.
Theoretical DC Input Current: ______________

With the DMM, measure the DC value of the output of the differential amplifier.
Measured DC Input Current: _______________
9. Place the output of the differential amplifier on the scope (DC coupled). Note the
large spikes in the measured voltage waveform (also the current waveform since
this is a 1 resistor) at the switching instances. Attempt to ignore these spikes
and try to measure the values of I max and I min for this full-load condition.
Measured I max : ____________
Measured I min : ____________
10. Have the instructor verify your measurements.
INSTRUCTOR VERIFICATION: _______________
11. De-energize your function generator and then your proto-board supply. Remove
all piece components and return them to the appropriate bin drawer. Return all
proto and measurement equipment to the appropriate cabinet.

You might also like