SiC Power Device｜Application

Method for Connecting a Probe

2024.08.07

Points of this article

・A measured waveform is greatly affected by how the probe is connected.

・When an extension cable is long, a loop formed between the gate and source terminals and the measurement instrument may cause a waveform that is completely different from the actual waveform to be observed. Hence a connecting method should be used that makes this loop as small as possible.

Points to Note When Measuring SiC MOSFET Gate-Source Voltages: Method for Connecting a Probe

As indicated in “General Measurement Methods“, a measured waveform is greatly affected by the method used to connect the probe. In order to confirm this fact, we compare the differences in measured waveforms resulting when employing each of the probe connecting methods shown below, which are widely used.

(a) Directly connect the probe head to the terminals of the DUT
(b) Use twisted wires to connect the probe head
(c) Place a 100 Ω damping resistor in the middle of each long twisted wire which is used to connect the probe head
(d) Place a 100 Ω damping resistor in the middle of each short twisted wire which is used to connect the probe head

In method (a), the head of the voltage probe is directly connected to the DUT. In (b), the end of the twisted extension cable approx. 12 cm long, formed from twisted wires, is soldered to DUT terminals, and the other end is connected to the head of a voltage probe. In (c), 100 Ω resistors are inserted midway in the extension cable of (b). And in (d), the extension cable of (b) is shortened to about 4 cm, and 100 Ω resistors are inserted midway. Fig. 5 shows the extension cables actually used, and Fig. 6 shows how the probe head is connected.

Actual extension cables. From left to right: (b) 12cm long extension cable, (c) 12cm long extension cable + 100Ω, (d) 4cm long extension cable + 100Ω
Fig. 5. Actual extension cables

Fig. 6. Photos of actual voltage probe connection

Fig. 7 compares the gate-source voltage waveforms in double pulse tests using the connecting methods of (a) to (d) in Fig. 6. Focusing on gate-source voltages on the LS (low side), which is the commutation side, we see that the waveforms are strikingly different when using different connecting methods.

In the case of (a), when at turn-on the HS MOSFET switching operation begins and a change in current occurs, a change occurs in the magnetic flux that links the loop formed by the voltage probe head shown in Fig. 8. This change in magnetic flux induces an emf in the clockwise direction in the loop of the head, so that it appears as if a negative surge occurs in the observed waveform. In the actual waveform, a surge on the positive side appears (*3), as in (c) and (d).

In the case of (b), the impedance due to the extension cable induces ringing caused by high-speed switching, and a large surge appears to be occurring.

Fig. 7. Comparison of gate-source voltage waveforms in double pulse tests using the connecting methods (a) to (d) in Fig. 6

Emf due to the probe on the commutation side

Fig. 8. Emf due to the probe on the commutation side

In this way, the loop formed between the gate and source terminals and the measurement instrument reacts to changes in the magnetic flux caused by current changes in the main circuit, so that the waveform observed is completely different from the original waveform, and so in order to observe the actual waveform, it is necessary to minimize the closed loop formed between the measurement instrument and the gate and source terminals. It should be noted that in (b), (c), and (d) of Fig. 6, the extension cables are soldered to terminal parts very close to the SiC MOSFET packages, so that the loop formed is minimized.

*3. Reference material: “Improvement of Switch Losses by a Driver Source Pin“, Application Note (No. 62AN039J Rev. 002）, ROHM Co., Ltd., April 2020

SiC Power Device

Basic

Application

Product Information

FAQ

SiC Power Devices FAQ