Power Supply Report
Stephen Covrig
Nominal Current Test
In this test, the input voltage of the regulators on the JD power supply were varied using a
multimeter and the nominal current (a current level that can be sustained for 30 seconds) was
measured.
In the graph above, the nominal input and output current are shown with respect to the input
voltage of a 5V buck regulator (TPS54232). The data show a slight downward trend where
higher input voltages lead to lower nominal current levels. This trend, however, is not perfect.
Around 10V, the nominal output current jumps back up to a little over 2A. This could be due to
the behavior of the regulating chip, the behavior of the chip’s surrounding components, or the
way in which the current was measured.
Input Voltage (V) Output Current (A) Input Current (A)
7 2.17 1.92
8 2.02 1.54
9 1.84 1.24
10 2.2 1.25
11 2.17 1
12 2.11 0.94
13 2.09 0.82
14 2.01 0.81
In the graph above, the nominal input and output current are shown with respect to the input
voltage of a 3.3V linear regulator (LD1117S33TR). The data show a distinct downward trend
where higher input voltages lead to lower nominal current levels. Additionally, it is important to
note the close relationship between the input and output currents (the blue data points are almost
perfectly covered by the red data points). Nearly all of the input current is passed on to the output
of the regulator. This is because linear regulators act as variable resistors in series with the load
which vary their resistance to maintain a constant load voltage. And since the load is in series
with the variable resistance, the output current practically matches the input current. (The
missing current is used to control the variable resistance.)
Input Voltage (V) Output Current (A) Input Current (A)
5 0.9 0.901
6 0.64 0.643
7 0.44 0.445
8 0.33 0.335
9 0.26 0.265
10 0.23 0.236
11 0.2 0.206
12 0.17 0.176
Questions
1. The 5V buck converter is much more efficient than the 3.3V linear regulator. This is
because buck converters act as energy storage and redistribution mechanisms while linear
regulators dissipate all extra energy not needed by the load. By storing up and then redistributing
energy, a buck converter is able to match (or get close to matching) input power with output
power. In other words, if a buck converter is outputting 5V and 1A and its input voltage is 10V,
the input current will only draw around 0.5A. With a linear regulator, the input current must
match the output current regardless of the input voltage. In other other words, if a linear regulator
is outputting 3.3V and 1A and its input voltage is 5V, the input current will need be 1A (leading
to energy loss). For these reasons, buck/boost converters make more sense in applications where
energy efficiency is necessary.
2. Testing nominal current is important because many applications require that a power
supply provide a constant flow of current. These nominal current values let a designer know how
much he/she can continually draw. If peak current was the only current tested, designers would
not be able to know whether or not their device could safely draw from the power supply.
Peak Current Test
In this test, the input voltage of the regulators on the JD power supply were varied using a
multimeter and the peak current (a current level that can be sustained for 3 seconds) was
measured.
In the graph above, the peak input and output current are shown with respect to the input voltage
of a 5V buck regulator (TPS54232). These value are slightly higher than the nominal current test
in the previous section. The data show a very small downward trend where higher input voltages
lead to lower nominal current levels. Just like the nominal test, the peak output current jumps
back up a bit around the 10V mark.
Input Voltage (V) Output Current (A) Input Current (A)
7 2.35 2.2
8 2.24 1.7
9 2.13 1.4
10 2.51 1.4
11 2.5 1.4
12 2.48 1.3
13 2.46 1
14 2.46 1
In the graph above, the peak input and output current are shown with respect to the input voltage
of a 3.3V linear regulator (LD1117S33TR). The data show a downward trend where higher input
voltages lead to lower nominal current levels. Additionally, like the nominal test, there is a close
relationship between the input and output currents (the blue data points are covering the red data
points).
Input Voltage (V) Output Current (A) Input Current (A)
5 0.91 0.911
6 0.9 0.902
7 0.89 0.892
8 0.87 0.872
9 0.58 0.584
10 0.48 0.484
11 0.43 0.434
12 0.39 0.395
Questions
1. Below is a graph showing the efficiency of the two supplies when operating at peak
output current. This graph confirms what was discussed in the nominal current test section by
showing that buck/boost converters operate at higher efficiency levels than linear regulators. In
this graph, the linear regulator’s efficiency decreases as a larger input voltage is applied. Also,
this graph shows that the buck converter’s efficiency remains fairly constant throughout its entire
range of input voltages.
Nominal and Peak Current Comparison
In this comparison, the nominal and peak output current of the regulators on the JD power supply
are related to the input voltage.
Input Voltage (V) Nominal (A) Peak (A)
7 2.17 2.35
8 2.02 2.24
9 1.84 2.13
10 2.2 2.51
11 2.17 2.5
12 2.11 2.48
13 2.09 2.46
14 2.01 2.46
Input Voltage (V) Nominal (A) Peak (A)
6 0.64 0.9
7 0.44 0.89
8 0.33 0.87
9 0.26 0.58
10 0.23 0.48
11 0.2 0.43
12 0.17 0.39
In both of the above graphs, the peak output current exceeds the nominal output current. This
means that both buck and linear supplies are able to pulse more current than they are able to
continually supply.
Questions
1. Looking at the data above, each supply is able to pulse current at rates higher than they
can continually supply. Therefore, buck and linear regulators are similar in this regard. However,
when considering efficiency, it is better to use a buck/boost converter over a linear regulator.
Additionally, when needing precision, it is better to use a linear regulator than a buck/boost
converter (no voltage ripple).
Ripple Voltage Test
In this test, the output current of each regulator on the JD power supply was varied using a load
regulating machine and the output ripple voltage was measured using an oscilloscope.
In the graph above, the voltage ripple of a 5V buck regulator (TPS54232) is shown with respect
to the load current. The ripple remains fairly steady from no load to a load of 1A. However, as
the output current exceeds 1.5A, the ripple becomes very unstable. The reason this happens is
because the load is drawing charge from the supply at a quicker rate than the supply can handle.
Output Current (A) Output Ripple (mV)
7V Input 10.5 V Input 14 V Input
0 15 30.2 45.7
0.2 16.7 24.8 26.8
0.4 16.8 29.3 34.7
0.6 17.1 30 36.7
0.8 17.4 30.2 37
1 18.1 30.7 37.1
1.2 20.8 31.5 37.2
1.4 25.1 32.4 38.6
1.6 27.8 39.7 150
1.8 35.2 122 212
2 164 239
In the graph above, the voltage “ripple” of a 3.3V linear regulator (LD1117S33TR) is shown
with respect to the load current. Due to the extremely low magnitude of this “ripple” and the lack
of correlation between the “ripple” and the current, it is safe to say this “ripple” is just noise. The
reason linear regulators lack a ripple is because they do not use the charging/discharging
functionality seen in buck/boost converters. Instead they set a constant output voltage and vary a
resistance to dissipate unused energy.
Output Current (A) Output Ripple (mV)
5 V Input 7.5 V Input 10 V Input
0 3.8 3.3 3.4
0.1 3.6 3.5 3.3
0.2 3.9 3.4 3.2
0.3 3.6 3.2
0.4 3.7 3
0.5 3.4
0.6 3.6
0.7 3
0.8 3.3
Questions
1. As was mentioned in the descriptions under each graph, the 5V supply increases its ripple
voltage as its load increases while the 3.3V lacks any sign of a voltage ripple (the collected data
was just noise). The reason ripple increases as load increases with a buck converter is because a
higher loads make it harder for the converter to resupply the capacitor and inductor with energy
as the LC setup continually charges and discharges. The reason ripple is non-existent in a linear
power supply is because the supply acts like a simple resistor ladder that varies its resistance
with respect to the load resistance in order to maintain a constant output voltage (there is no
charging/discharging that would cause a ripple).
