Using a wattmeter
If youíre measuring input power using a wattmeter, first connect the wattmeter to the power-supply input and set the display to averaging mode for more stable readings. To properly snap on your power supply for this measurement, first configure your load and source settings. Turn on the AC input voltage and slowly increase it to the desired test voltage. Increase the load on the supply to full load. Then turn off the supply and snap it back on to complete the measurements.
In our example, the meters on the supplyís output showed 4.97 V and 4.005 A. The voltage reading on the electronic load showed an output voltage of 4.48 V. This means that 490 mV were dropped in the output cables and across the current sense element of the digital multimeter, further reinforcing why we measure output voltage at the supply terminals instead of at the load.
In this example, output power = 4.97 x 4.005 A, which equals 19.90 W. The input power read from the wattmeter display was 25.76 W, at an operating voltage of 115 V AC. Therefore, power supply efficiency = output power/input power = 19.90/25.76 = 77.3 percent.
If you donít have a wattmeter available, you can achieve approximately the same result using two multimeters to measure input power instead. Because we have no way to measure power factor, we take the input measurement after the diode rectifier stage has converted the AC input to DC. Weíll also take into account losses in components prior to the DC-bus stage to improve measurement accuracy. In this example, weíll add in the power losses of the diode rectifier bridge, which is typically the lossiest component in the input stage. Calculate these losses by multiplying two forward-conducting diode drops by the input current measured. In low-cost bridge rectifier diodes, this drop can be as high as 0.9 V. This approach can also be used to account for the power dissipated by other lossy elements whose resistance or voltage drops are significant and measurable.
Next, break the DC bus between the bridge rectifier and bulk capacitor using a razor blade or similar tool to cut the PCB trace. Itís important to make this break before the bulk capacitor, so the meter doesnít have to measure the power supplyís high-frequency switching currents. Thatís something a multimeter canít do accurately.
Once the bus is broken, solder leads to the PCB trace on either side of the break to connect the multimeters to the circuit. First, connect a high-resolution true RMS multimeter set to measure current across the break, in series with the circuit. Then, set a second multimeter to measure voltage and connect it across the bulk capacitorís DC positive and negative terminals.
To begin the test, turn on the AC source and slowly increase voltage to the desired test voltage. Increase the load on the power supply to full load. Set the input current meter to the highest current range. Then, turn off the AC input voltage and snap it back on.
In our example, the multimeters on the output indicated the power supply was delivering 4.97 V and 4.008 A, which gives us an output power of 19.92 W. On the input, the DC bus voltage was measured to be 151.6 V and input current was 0.166 A. We calculate input power as Vin times Iin = 151.6 x 0.166 = 25.1656 W.
Next, we must add the power lost in the rectifier bridge. The diode-bridge power-loss estimate = two times the worst-case diode drop times the input current. That comes to 2 x 0.9 x 0.166 = 0.299 W.
Therefore, total input power = 25.1656 W + 0.299 W = 25.46 W.
Using this technique, we calculate the efficiency of the power supply as 19.92 /25.46 = 78.2 percent.
When we compare this result to the 77.3 percent efficiency we calculated using a wattmeter to measure input power, we see that the multimeter method is 0.9 percent less accurate in this case.