Why are Optical-fiber Isolated Probe indispensable in dual-pulse testing

[Mcisig麦科信]

　　Wide Bandgap (WBG) semiconductor materials, such as Silicon Carbide (SiC) and Gallium Nitride (GaN), not only excel in high-temperature and high-voltage resistance but also possess characteristics of low loss and fast switching frequency. However, to fully leverage the potential of these advanced materials, precise testing and measurement techniques are crucial. Particularly in dual-pulse testing, Optical-fiber Isolated Probe not only ensure the safety of the test but also enhance the accuracy and reliability of the test measurements. This article will delve into the indispensable reasons for Optical-fiber Isolated Probe in dual-pulse testing.

　　The Function of Dual-Pulse Testing

　　Dual-Pulse Testing (DPT) is an experimental method used to evaluate the switching performance of power electronic devices such as IGBTs (Insulated Gate Bipolar Transistors) or MOSFETs (Metal-Oxide-Semiconductor Field-Effect Transistors). This test subjects the device to two short voltage pulses, simulating the switching process of the device in an actual circuit, thereby measuring and analyzing the switching characteristics of the device. It is used to optimize the device's drive and application design, and for fault diagnosis and verification of simulation models.

　　The Selection of Oscilloscopes and Probes in Dual-Pulse Testing

　　Taking the half-bridge gate drive circuit of a MOSFET as an example, we need to test the Vds, Id, and Vgs of the lower tube, while also observing the Vgs of the upper tube. By selecting the Micsig MHO high-resolution oscilloscope 3 series with a 500MHz bandwidth, 3GSa/s sampling rate, and an accuracy of ≤1%, the four channels can support the simultaneous observation of the switching of both the upper and lower tubes, perfectly meeting the testing requirements of DPT (Dual-Pulse Testing).

　　To accurately measure the Id waveform, it is essential to ensure that the current probe used has sufficient bandwidth. The Micsig high-frequency AC/DC current probe CP series can be considered, which has a bandwidth of up to 100MHz, an accuracy within 1%, a resolution that can reach 1mA, and provides a measurement range up to 30A. For testing requirements of larger currents, the Rogowski coil RCP series shown in the diagram can be used.

　　However, many users have raised doubts: "We have been using the Micsig MDP series high-voltage differential probes, which perform well when testing silicon devices, capable of measuring voltages up to 7000V, and the bandwidth is not low (500MHz). Now that we have switched to GaN and SiC devices, it should meet the bandwidth parameters of these devices. Testing the lower tube is also fine, but why is there always a problem when testing the voltage of the upper tube?"

　　After analyzing and comparing the data from the diagram, we found that the switching speeds of Silicon Carbide (SiC) and Gallium Nitride (GaN) have reached the nanosecond level. The significant advantage of this feature is the reduction of energy consumption in switching power supplies. However, this also poses a huge challenge for testing. In the half-bridge circuit, the upper tube Vgs voltage floats above the continuously conducting and switching off lower tube Vds. The Vds voltage can complete a transition from zero volts to thousands of volts within a few nanoseconds, and the superposition of high voltage with high frequency causes the high-order harmonic components to increase significantly. The differential mode voltage of our test object Vgs is often only tens of volts, and it is significantly affected by the common mode interference brought by the high-order harmonic components of Vds. When measuring, we need to suppress this common mode interference as much as possible, which requires the test equipment to still have a high common mode rejection ratio (CMRR) in the high-frequency range.

　　Taking the Micsig MDP series mentioned by the customer as an example, at 100KHz, the CMRR is greater than -70dB; at 20MHz, the CMRR is greater than -40dB; and at 120MHz, the CMRR is greater than -26dB. For a differential probe, this CMRR is already very excellent in the industry, but it is far from enough to meet our requirements for measuring the upper tube Vgs. We need a test device that still has a high CMRR in the high-frequency range.

　　Comparison of High-Voltage Differential Probes and Optical-fiber Isolated Probe in Practical Measurement

　　Regarding the impact of CMRR on testing, let's make a comparison to see the problems generated by high-voltage differential probes during testing, as well as the comparison of testing with probes that have a high CMRR:

　　Test Method: The SiC device under test has upper and lower tubes, with a Vce voltage of around 500V. Both high-voltage differential probes and Optical-fiber Isolated Probe (using the Micsig Optical-fiber Isolated Probe MOIP series) are connected to the upper tube Vge signal simultaneously, to perform dual-pulse testing.

　　The above figure is a test result chart. The white signal in the chart is the test result from the high-voltage differential probe. It can be seen that during the rise of Vge, there is a severe oscillation, making it almost impossible to discern the original waveform. We have previously used a high-voltage differential probe to test the upper tube Vge signal when the Vce voltage reached 800V, and the oscillation exceeded the turn-off voltage of the SiC, which would seriously affect the engineer's judgment.

　　The red waveform in the chart is the result of testing with an Optical-fiber Isolated Probe, and the signal interference is much less. If tested alone with an Optical-fiber Isolated Probe, there would be almost no interference. The interference seen here is the effect of the high-voltage differential probe on the Optical-fiber Isolated Probe. In fact, the baseline noise of the Optical-fiber Isolated Probe is lower compared to the high-voltage differential probe, with higher precision, and it can measure a larger common mode voltage. How is this achieved?

　　The Benefits of Optical-fiber Isolated Probe

　　Micsig utilizes its proprietary SigOFIT™ technology, allowing the selection of an appropriate attenuator before testing to ensure full-scale testing of differential signals ranging from ±0.01V to ±6250V. This not only accommodates a wide range of testing but also enhances testing accuracy (to 1%), reduces baseline noise, and improves the signal-to-noise ratio.

　　The Micsig MOIP series of Optical-fiber Isolated Probe offers a bandwidth that can reach up to 1GHz, with the minimum noise floor achievable at less than 0.45mVrms. At the 1GHz frequency band, the Common Mode Rejection Ratio (CMRR) remains high at over 100dB. Therefore, when using Optical-fiber Isolated Probe to measure the upper tube Vgs, there is no longer a need to consider the impact of common mode interference, perfectly resolving the issue of insufficient CMRR in high-voltage differential probes.

　　Additionally, due to the long leads of differential probes (usually around 20cm), these two input wires can be considered as an antenna, which can pick up interference from external magnetic fields. Given the extremely fast switching speed of Gallium Nitride (GaN), the magnetic field generated can cause oscillations when it passes through the input end of the high-voltage differential probes. Sometimes, these oscillations can exceed certain limits, leading to the instantaneous destruction or explosion of the GaN device. On the other hand, Optical-fiber Isolated Probe use MCX or MMCX connectors with extremely short leads, almost eliminating the antenna effect. The parasitic capacitance is within a few picofarads, effectively preventing safety hazards caused by parasitic effects during testing.

　　Summary

　　In summary, Optical-fiber Isolated Probe have actually surpassed differential probes in terms of overall performance, and for users who need to conduct dual-pulse experiments, Micsig's Optical-fiber Isolated Probe are the best choice.