Double Pulse Testing Basics Series: Basic Principles and Applications

Double pulse is a basic experimental method for analyzing the dynamic characteristics of power switching devices, which runs through the development, application and design of drive protection circuits of devices. With the rational use of the double pulse test platform, you can easily debug the drive circuit, optimize the dynamic process and verify the short circuit protection in the system design.

The series of articles on the basics of double pulse testing includes basic principles and applications, requirements for voltage and current probes, and factors that affect test results.

Why perform a double pulse test?

In the past and even today, many engineers who use IGBT or MOSFET as inverters do not perform double pulse experiments, but directly run them under calibrated working conditions to see if the designed power can be achieved. Such tests are indeed necessary, but often the specific switching losses, voltage or current spikes, and parasitic conduction conditions cannot be seen. This will lead to blind spots in the understanding of some risks, which will in turn affect the long-term reliability of the final product in the future. Or too much design margin will increase costs and reduce the market competitiveness of the product. If the switching performance of the device can be accurately understood during the design and development stage, it will bring great benefits to the optimization of the entire product. For example, the switching loss can be obtained at different voltages, currents and temperatures to provide reliable data for system simulation; for example, the appropriate gate resistor can be selected by observing the waveform oscillation.

Double pulse test principle

As the name implies, the double pulse test is to give the device under test two pulses as the driving control signal, as shown in Figure 1. The first pulse is relatively wide to obtain a certain current. At the same time, the falling edge of the first pulse is used as the observation time of the shutdown process, and the rising edge of the second pulse is used as the observation time of the opening process.

figure 1

Considering the possible electric field interference, the best dual pulse platform is a full-bridge structure, as shown in Figure 2. The gate of tube 3 gives a 15V normally open signal, tube 4 is normally closed, and tube 2 is given a dual pulse signal as the device under test. Tube 1 is mainly used for freewheeling, so the gate can be a normally closed signal, or when using a MOS tube, the gate uses a synchronous rectification signal. When the first pulse comes, the current passes through tube 3 and the load inductor into tube 2. In order to obtain a desired current value, this pulse needs to last for a certain length of time, which can be obtained by T=I*L/V, where L is the load inductance value, I is the desired current, and V is the bus voltage. Of course, in practice, you can directly use an oscilloscope to observe the current value to adjust the pulse width.

At the end of the first pulse, tube 2 is turned off, showing the shutdown waveform of the device. After that, the current continues to flow in tube 3, the load, and tube 1. When the second pulse arrives, tube 2 is turned on, and the current on tube 1 flows back into tube 2. At this time, the reverse recovery characteristics of tube 1 and the turn-on characteristics of tube 2 can be measured. The current can be obtained by using a current detection resistor at low power, while a magnetic current detector is generally used for large currents, such as a Rogowski coil or a Pearson magnetic ring. The use of voltage and current probes has been introduced in detail in a special article before.

figure 2

During routine laboratory measurements, the structure can also be simplified to a half-bridge form, as shown in Figure 3. The working principle is similar to that of the full-bridge, so I will not go into details here.

image 3

Which parameters can be obtained through double pulse experiments?

Through double pulse testing, dynamic parameters including switching loss, voltage and current peak values, and slope change values ​​can be obtained. For example, IGBT parameters:

Each parameter can be directly measured on an oscilloscope, and is defined as shown in Figure 4:

Figure 4

And the anti-parallel diode parameters:

The measurement definitions of each parameter are shown in Figure 5:

Figure 5

The switching loss needs to be obtained with the help of the function operation function of the oscilloscope, or all waveforms can be saved as table points for later processing and operation. Figures 6 and 7 are the turn-on and turn-off waveforms, respectively. The black channel 1 is the VCE signal of the device under test, the red channel 2 is the bridge arm current, and the green channel 3 is the gate signal. Mathematical operation 1 is the product of the voltage and current signals, and mathematical operation 2 is the integral line of the product, that is, the loss value. According to the national standard definition, the integral time interval of the turn-on loss is the interval from the gate voltage rising by 10% to the VCE voltage falling by 2%; and the integral time interval of the turn-off loss is the interval from the gate voltage falling by 90% to the current falling by 2%. If there is some oscillation, the first boundary value touched shall prevail. However, if there is severe oscillation, it is recommended to adjust the drive parameters (such as increasing the gate resistance) and re-measure.

Figure 6

Figure 7

Summarize

Double pulse testing is very suitable for evaluating the loss of devices in the system during the R&D stage, and provides data support for system simulation. After all, there are too many factors that can cause a big difference between the actual value and the data specification. Using the values ​​in the specification as the basis for simulation is a bit rough, but of course it is still worth learning from when selecting devices. The more accurate the design, the more friendly it is to cost control.