• • • TEST & MEASUREMENT • • • Common mode distortion of
wide-bandgap power electronics probing By Chun Soong Wong, Oscilloscope Product Manager, Rohde & Schwarz
n an ideal scenario, a probe accurately transmits a signal from the device under test (DUT) in its original form with no alterations. Differential probes are designed to eliminate common-mode (CM) distortion by cancelling out identical signal artifacts on each socket relative to earth. In practice, issues such as poor CM rejection, frequency response degradation and signal distortion can compromise signal integrity, particularly in high-speed power applications where signal fidelity is crucial.
I
When working on fast switching power semiconductors such as wide-bandgap (WBG) devices, fast and high voltage (HV) swings are common. The rise/fall time of devices such as those made from gallium nitride (GaN) have a range of < 2 ns, which require a higher bandwidth probing solution of up to 1 GHz. To optimise signal integrity in such power applications, probe setup performance and frequency response need to minimise distortion and signal degradation. HV differential probes are often used to measure signals without a ground reference in a high-side gate source voltage for a half-bridge converter. Note here the common mode rejection ratio (CMRR) of these probes, especially at high frequencies, help suppress fast environmental CM noise.
Limitation of conventional
HV differential probe A common misconception is that standard differential probes are fully earth isolated across their entire bandwidth. In reality, the differential amplifier in these probes is earth referenced, creating a return path for CM signals when used with a measurement instrument and inherently limiting their CMRR performance. While 4 mm
Fig 1: FFT setup
banana plug cables have HV ratings necessary for power measurements, they lack controlled impedance and proper shielding, especially at higher frequencies. Hence, conventional HV differential probes typically provide bandwidths of up to 400 MHz with limited CM noise suppression of about –20 dB. For every 10 V of CM signals, 1 V appears as noise in measurements.
Breaking the CM ground loop To effectively measure WBG power devices, which operate at higher voltages and faster switching speeds, full galvanic isolation between the DUT and
the measuring device is essential to limiting the CM return path and reducing noise. One popular approach for isolation is inductive or opto-coupling used in isolated-channel oscilloscopes. However, long and unshielded cables can still introduce noise along the signal path. The distance of the probe from the DUT’s measurement point must be minimised. Replacing traditional probes tips with coaxial tips can help maintain proper impedance matching and improve measurement accuracy.
Fig 2: Isolated power conversion design 44 ELECTRICAL ENGINEERING • MAY 2025
Isolated probing systems like the R&S RT-ZISO are designed with this measurement approach in mind. The probe head connects to the DUT over a short coax probe tip (MMCX) to shield the measured signals from CM noise and maintain stable and matching impedance, letting the probe operate even at high frequencies of up to 1 GHz. The probe head converts the measured signals into optical signals and transmits them to the receiver box, breaking the CM ground loop and reducing disturbance during the measurement. When characterising high-side WBG gate-source input, the switch node (C3) applies fast switching CM voltage to the measurement point. The high- side gate voltage (C4, screenshot below, left side) is highly affected by the CM noise when using conventional HV differential probes. The R&S RT-ZISO captures the same gate voltage (C2, screenshot below, right side) with slightly faster edges and less noise. The noise seen on HV differential probes (C4) has a huge perturbation and takes longer to settle down. Waveform details of gate voltage where the Miller plateau and charging time could be of interest can be highly disturbed and concealed by CM noise.
electricalengineeringmagazine.co.uk
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56
