Power supply compliance testing according to energy efficiency standards

Over the past few years, several major international standards organizations, including the U.S. Environmental Protection Agency (EPA) Energy Star program, the European Commission's Code of Conduct (CoC), and the California Energy Commission (CEC), have established new external power supply efficiency requirements. These standards require designers to test their products' active mode efficiency and no-load power consumption with high accuracy. This article will introduce a relatively simple measurement method to determine the compliance of external power supplies with these changing energy efficiency standards, and will also provide some useful testing tips.

Basic Requirements

Let’s start by looking at the equipment you’ll need to perform these tests. To accurately measure the efficiency of your product, you’ll need four tools:

1. A wattmeter;

2. A programmable AC power supply;

3. An electronic load;

4. Two digital multimeters (one of which must be a high-precision ammeter);

Next, there are some general guidelines you need to follow to get accurate measurements. First, because measuring the energy efficiency compliance of a power supply is a system-level test that also measures the power dissipation in the input and output cables, you must ensure that the cables used for the test are the same as those used in the final product. Second, it is important to note that these tests require long periods of temperature stabilization during changes in the output line voltage and input line voltage. Therefore, expect these tests to take several hours to complete. Finally, these tests must be repeated every time you modify your product design to ensure accurate results.

Achieve the highest precision

This test requires measuring both the no-load input power and the loaded mode efficiency of the external power supply. To calculate the efficiency, both the input and output power must be measured. When measuring the input power, it is critical to select the proper power supply. As shown in the graph in Figure 1, the raw AC power from the wall outlet and the autotransformer results in inaccurate measurements. To ensure that the test is performed with an accurate input voltage, a programmable AC power source is required.

The graph in Figure 2 shows the output from a programmable AC power source. Note that the input waveform is a true sine wave.

We need to use a wattmeter when making this measurement because it can measure the power factor, which is the cosine of the phase angle (cos) between the voltage waveform and the current waveform, and take this factor into account. It is important to note that energy efficiency standards require that the uncertainty of input power measurement should be less than 2% when measuring power of 0.5W or more, and the uncertainty should be 10mW when measuring power below 0.5W.

Wattmeters from several major meter manufacturers can be configured to meet these requirements. Since wattmeters contain both current sensing and voltage sensing elements, the voltage sensing element can be configured to be located before or after the input current sensing element. Information on how to configure the meter can usually be found in its user manual. When making low load or no-load measurements, more accurate measurements can be obtained by configuring the voltage sensing element to be located before the current sensing element. This prevents the current from the voltage sensing element from being measured by the current sensing element. Considering that the current consumed by the voltage sensing element is usually greater than the 10mW allowed at 230VAC, this configuration is critical in meeting the low measurement uncertainty requirements of energy efficiency standards.

High power applications

When designing for higher power, different issues need to be addressed. In these applications, the power loss of the voltage sense element is very small, and the voltage sense element can be connected after the current sense element, close to the voltage input. This approach prevents the voltage drop across the current sense element and the internal wiring of the wattmeter from being erroneously included in the power measurement, resulting in a lower calculated efficiency value. Setting the wattmeter to average 32 (or 64) samples can obtain more stable results (Figure 3).

To measure output power, you need two multimeters: one to measure the output voltage and the other to measure the output current. Use a multimeter with higher accuracy to measure the output current. Since the output power is pure DC power, it can be calculated by multiplying the output voltage by the output current.

To simplify the task of determining if a power supply specification complies with global regulations, PowerIntegrations has developed a useful Energy Efficiency Compliance Calculator. Target values ​​for no-load and active-mode efficiency can be calculated from the nominal rating of each power supply. This rating is simply the rated output value marked on the power supply housing and represents the minimum rated output power of the power supply at room temperature and rated input voltage. For example, a constant voltage constant current charger with a nominal rating of 5V, 350mA can provide a minimum of 5V, 350mA.

Simply enter the nominal output power rating of your power supply and the calculator will automatically tell you the compliance target value for energy efficiency standards relevant to your power supply design. Designs that accept universal input voltages require measurements at both 115VAC, 60Hz and 230VAC, 50Hz. For single input designs, measurements should be made at the nominal input voltage of either 115VAC or 230VAC.

Now, let's take a 5 V, 350 mA mobile phone charger rated for a universal input range as an example to demonstrate the test procedure. First, we will perform a series of tests to measure the load mode efficiency of this power supply at 115VAC, 60 Hz. The load mode efficiency is the average of the efficiency measured at 25%, 50%, 75%, and 100% load levels of the nominal rating at the rated input line voltage and rated line voltage frequency. Since the nominal load rating of this charger is 350 mA, the efficiency must be measured at the following currents: full load 350mA, 75% load 262 mA, 50% load 175 mA, and 25% load 88mA.

Allow the power supply to warm up for 30 minutes before starting the test. Before taking the first measurement at full load, connect the power supply to an AC mains and apply a 60Hz, 115VAC input. Increase the load on the power supply gradually to full load, allowing at least 30 minutes for the circuit to reach thermal equilibrium and for the input power reading to stabilize. Make sure no oscilloscope probes or other instruments are connected to the circuit before recording the measurement. In some cases, it may be necessary to manually set the voltage and/or current range of the wattmeter to prevent it from automatically changing ranges and causing unstable results. The same measuring instrument will often produce different accuracy when used on different ranges. Usually, the operating manual will indicate the range that produces the highest accuracy.

When the circuit reaches thermal equilibrium, record the initial power reading from the wattmeter. Wait five minutes and take a second reading. If you find that the two readings differ by less than 5%, record the second reading. If they differ by more than 5%, wait another five minutes and continue this process until the difference between two consecutive readings is within 5%. Alternatively, you can use the "integration mode" available on most wattmeters, which is described below.

Calculate the integrated input power

To measure the input power of a design that varies with time, the following steps need to be taken:

1. Set the wattmeter to integrating mode;

2. Set the integration interval for the wattmeter to capture approximately one full cycle of the variable input power; (the longer the duration, the more accurate the measurement. We recommend 1 minute of integration for most applications)

3. Read the wattmeter input power reading in Whr;

4. Divide this number by the integration interval. Make sure to adjust the time units so that they cancel each other out. For example: Input Energy (W hr) / Measurement Time Interval (min) × 60mins/1hr = Input Power (W); In this example, based on the multimeter readings, we recorded the output current as 0.35A and the output voltage as 6.124V, which gives an output power of 2.14W.

The efficiency is calculated as follows:

Full load efficiency = Pout/Pin = 2.14W/3.14W = 68.2%

Before taking the next measurement, adjust the load level to 75%, or 262mA. When using a wattmeter to calculate an average, remember to allow at least one minute for the reading to stabilize. Then record the input power from the wattmeter. Wait five minutes, record again, and use the <5% difference rule to determine whether to use this value or calculate the integrated input power.

In this example, we record the output current as 0.262A and the output voltage as 6.502V, which gives an output power of 1.704W. The efficiency is then calculated as:

Efficiency at 75% load = 1.704W/2.42W = 70.4%

The above measurement procedure should then be repeated for 50% and 25% load levels.

Measuring no-load input power

Before measuring the no-load input power, disconnect the output load and all output multimeters from the power supply. Next, turn off the AC input and configure the wattmeter so that the voltage sense element precedes the current sense element. Record the initial power reading of the wattmeter, then wait five minutes and record the second reading. As we described in the previous test procedure, if the difference between the two readings is less than 5%, record the second reading. If the difference is greater than 5%, the integral of the input power must be calculated and then divided by the integration interval according to the previous procedure.

You are now done with the 115 VAC test. Next, you must reconnect the load and output multimeter and raise the input voltage to 230VAC. Repeat all of the tests you performed previously at the new input voltage. Remember to set the wattmeter to the appropriate power range first. When the test is complete, you should have a complete test data that summarizes the measurements for all four load levels at both input voltages.

The Power Integration calculator will then calculate the load mode efficiency, which is an equally weighted average of the efficiencies at all load levels. In the field on the right side of the screen, you can see many rows of different energy efficiency standards. The calculator will automatically calculate the compliance requirements for the load mode efficiency and no-load input power, and compare the test results with the standard requirements. To simplify the analysis of the results, it will display the test results that meet the requirements in green and the test results that do not meet the requirements in red.

Supplementary content 1:

Testing Tips: Manual Calculation

If you decide to calculate these values ​​manually, it is important to note that the no-load requirements and active-mode efficiencies of these standards are rounded to double digits. This may seem insignificant, but it has a significant impact on standard compliance. Here is an example of calculating the minimum efficiency of a 12V, 1.1A power supply using the Energy Star formula:

Supplementary content 2:

Testing Tips: Changing Input Voltage

Are you measuring no-load power consumption and planning to change the input voltage during the test? Always apply a full load to the output during this transition. Why? In no-load mode, the input bulk capacitors take a long time to discharge, and if there is no load connected to the output after the input voltage drops from high to low, the capacitors will support the DC bus voltage for a long time before drawing power from the AC input. This in turn causes the no-load input power of the power supply to be 0W for a long period of time. To avoid this problem, bring the power supply to full load before starting the measurement, and then disconnect all output loads.