Tektronix SiC,GaN Power Converter Analysis Kit User manual

SiC and GaN Power

Converter Analysis Kit

Instruction Guide

2

SIC AND GAN POWER

CONVERTER ANALYSIS KIT

1. 5 Series MSO: (1 GHz) Oscilloscope

2. 5-PWR Software: Power analysis software

3. TIVH08: (2.5 kV, 800 MHz) High voltage isolated

differential probes

4. TIVH05 (Optional): (2.5 kV, 500 MHz) High voltage
isolated differential probes

5. TIVM1 (Optional): (50 V, 1 GHz) High BW Isolated

differential probes

6. TPP1000 (Free): (1 GHz) High BW passive probe.
Comes standard with scope. One per channel.

7. MMCX Tip for TPP1000 (Optional): Passive probe

MMCX Tip for high BW performance.

8. GaN Half-bridge Demo Board Guide: Instruction
Guide for getting started.

* All products come standard with their own
accessories.

Index

3

SIC AND GAN POWER CONVERTER SWITCHING ANALYSIS

Topics

Page

Introduction

4

Evaluation Board Overview

5

Selecting the Right Sensor (Probe) Tips

6

Making Connections

8

Before Making Measurements

9

Powering Up the System

14

Tests:

Test 1: Analyzing dead

time and optimizing switching characteristics

16

Test 2: Making fast

-switching load/drain current measurements using a shunt

21

Test 3: Testing switching/conduction losses and magnetic properties

27

Specifications

34

Introduction

The SiC and GaN Switching Power Converter Analysis Kit
is the ONLY solution in the market that can accurately
characterize most of the critical parameters for optimizing
Power Electronics topologies that use ultra-fast, power
semiconductor switching technology such as SiC, GaN
and even some silicon-based MOSFET and IGBTs.

Characterize:

• VGS, VDSand IDmeasurements on high-side and low-
side switches.

• Dead time optimization, including accurate turn-on,
turn-off and gate-drive timing analysis.

• Switching and conduction loss measurements.

• Magnetic performance, losses and system efficiency.

4

GaN Half-bridge Evaluation Board Connections

USB Power

Main Power
Switch (S3)

MMCX Board
Power Switch (S1)

Square Pin Board
Power Switch (S2)

VGS_High Side (MMCX)

VDS_High Side (MMCX)

VGS_Low Side (MMCX)

VDS_Low Side (MMCX)

Shunt Current (MMCX)

VGS_High Side (Square Pin)

VDS_High Side (Square Pin)

VGS_Low Side (Square Pin)

VDS_Low Side (Square Pin)

Shunt Current (Square Pin)

• Two independent GaN-based half-bridge
circuits.

• Choose between MMCX connectors (top

half) or the Square pin connectors (bottom

half) for your evaluation.

• DC Bus Voltage 50 V; Gate Voltage 5 V

• USB Powered *

2X

*Use both ports of the provided dual-input USB cable. Using any other cable can cause serious damage to the oscilloscope’s USB port.

Selecting the Right Sensor Tip for TIVM

6

Note: Specifications are dependent on the probe tip cable.

ISOVU TIVH SERIES

Sensor Tip Cable

Connection to

DUT

Differential Voltage

Range

Offset Range Input Impedance

Most Sensitive V/div

Setting

Attenuation

SMA Input MMCX

IVTIP1X MMCX ± 1 V ± 2 V 50 Ω // <1 pF 1 mV/div 1X

IVTIP5X (Standard) MMCX ± 5 V ± 10 V 250 Ω // <1 pF 5 mV/div 5X

IVTIP10X MMCX ± 10 V ± 20 V 500 Ω // <1 pF 10 mV/div 10X

IVTIP25X (Standard) MMCX ± 25 V ± 50 V 1.25 kΩ // <1 pF 25 mV/div 25X

IVTIP50X (Standard) MMCX ± 50 V ± 100 V 2.5 kΩ // <1 pF 50 mV/div 50X

For this Demo, TIVM1 probe system is used to measure floating, high dv/dt Gate-Source (VGS) signal:

• Maximum dv/dt: 5 V/1 nS = 5 V/ns

• Maximum differential voltage at the test points: 7 V

pk-pk

• Minimum differential loading (input impedance) that the circuit can tolerate : 1 KΩ

• Required measurement sensitivity (V/div) <50 mV/div

• Sensor tip selected: IVTIP25X: ±25 V; 1.25 kΩ // <1 pF; 25 mV/div

Selecting the Right Sensor Tip for TIVH

7

Note: Specifications are dependent on the probe tip cable

ISOVU TIVH SERIES

Sensor Tip Cable

Connection to

DUT

Differential Voltage

Range

Offset Range Input Impedance

Most Sensitive V/div

Setting

Attenuation

Direct Connection into SMA Input of the Sensor Head

SMA Input MMCX ± 1 V ± 25 V 1 MΩ // 20 pF 1X

Requires MMCX Style female connectors on the DUT

MMCX10X MMCX ± 10 V ± 250 V 10 MΩ // 6 pF 10mV 10X

MMCX50X (Standard) MMCX ± 50 V ± 250 V 10 MΩ // 3 pF 50mV 50X

MMCX250X MMCX ± 250 V ± 250 V 10 MΩ // 2 pF 250mV 250X

Requires 0.100” Pitch (2.54 mm) Square Pins on the DUT

SQPIN100X Square pin ± 100 V ± 600 V 10 MΩ // 3.5 pF 100mV 100X

SQPIN500X (Standard) Square pin ± 500 V ± 600 V 10 MΩ // 3.5 pF 500mV 500X

Requires 0.200” Pitch (5.08 mm) Square Pins on the DUT

WSQPIN1000X Square pin ± 1000 V ± 1000 V 40 MΩ // 3 pF 1.0V 1000X
WSQPIN2500X Square pin ± 2500 V ± 1000 V 40 MΩ // 3 pF 2.5V 2500X

For this Demo, TIVH probe system is used to measure floating, high dv/dt Drain-Source (VDS) signal:

• Maximum dv/dt: 50 V/5ns = 10 V/ns

• Maximum differential voltage at the test points: 50 V

pk-pk

• Minimum differential loading (input impedance) that the circuit can tolerate: 10 kΩ

• Required measurement sensitivity (V/div): 10 V/div

• Sensor tip selected: MMCX50X: ±50 V; 10 MΩ // <3 pF; 50 mV/div

Make Connections – V

DS

and V

GS

Measurements

8

Channel 1: Low Side VGS- TPP1000 and MMCX adapter tip.

Channel 2: High Side VGS- TIVM1 and MMCX25X sensor tip.

Channel 3: High Side VDS- TIVH08 and MMCX50X sensor tip.

Channel 4: Low Side VGS- TPP1000 and MMCX adapter tip.

*Do not use TPP1000 passive probe for any high-side, floating measurements. This can cause serious damage to the oscilloscope input.

Vgs_hi

Vgs_lo

Vds_hi

Vds_lo

Before Making Measurements

1. Once all the connections are made, turn on the
oscilloscope by pushing the power button in the lower
left corner of the front panel.

2. Press ‘Default Setup’ button on the oscilloscope.

3. Turn on Ch1, Ch2, Ch3, Ch4 on the oscilloscope by
double-clicking on the respective channels’ badges on
the bottom ribbon.

4. Before turning the GaN evaluation board ON or making
any measurements, run following settings on all active
channels. (Details on the following pages):

• Self Calibration (IsoVu only)

• Auto Zero (IsoVu only)

• Deskew

• Probe Compensation (passive probes only)

9

Self Calibration for IsoVu Probes

The SELF CAL sequence should always be run on an IsoVu probe after it is
first powered on and has warmed up for 20 minutes.

1. Make sure there is no differential signal present at the sensor tip cable.

2. Press the SELF CAL button on the controller to adjust the operating

point of the measurement system.

3. The indicator blinks orange during the self calibration process; it turns
solid green when the operation completes or solid red when the
operation fails.

• Always run Self Calibration in following situations:

• The measurement system is first attached to the oscilloscope.

• Changes are made to the range (1X|2X), internal compensation or

clamp (ON|OFF) setting.

• The temperature in the sensor head changes more than 10°C.

• The sensor tip cable is changed.

10

Auto Zero for IsoVu Probes

Run Autozero on start or when the displayed waveform is
not centered correctly (for example, due to a small DC
offset error).

1. Make sure there is no differential signal present at

the sensor tip cable.

2. Press the MENU button on the IsoVu controller to
view the Probe Setup menu* on the oscilloscope.

3. Press the Autozero button in the Probe Setup menu
of the oscilloscope.

4. Repeat this for all the connected IsoVu probes.

11

Deskew All Probe Channels

Running Deskew is very important to get accurate delays

between the signals. Double-tap the Ch1 badge and go to

‘Other’ Menu.

1. Set the Ch1 Deskew to 0 by pressing ‘Set to 0’ soft
button.

2. To set other channels’ Deskew, press the ‘Multi Channel’
soft button. This will open an Deskew setting popup.

3. Select ‘From Source’ as Ch1, ‘To Source’ as Ch2, and

press the ‘OK, Deskew’ button.

4. This will deskew the second channel using the default
propagation delay for each probe.

5. Repeat this step for all channels and deskew them with
respect to Ch1.

12

Probe Compensation for Passive Probes

Probe compensation optimizes probes to a
specific channel and stores the results to the
oscilloscope.

IsoVu probes are compensated at the factory
and do not need any additional compensation.

For all other passive probes, run the probe
compensation under the Probe Setup menu by
double-tapping the respective channel badges.

Follow the instructions under the ‘Probe

Compensation’ popup.

13

Powering on the GaN Half-bridge Evaluation Board

1. Once all the pre-requisite settings are applied,
connect the supplied dual-input USB cable to two of
the oscilloscope USB ports before connecting the
cable to the USB Power input on the demo board.

Using any other cable can cause serious damage

to the oscilloscope’s USB port due to high inrush

current on the board.

2. Now, connect the GaN Half-bridge board power
(USB power) to the USB cable.

3. The green LED labeled ‘USB PWR’ on the board

will turn on and remain steady when the USB cable
is connected.

4. Turn Switch S3, next to the USB input, to the ON
position to turn on the main board power.

5. Next, turn ON Switch S1 to power the upper half of
the board with MMCX test points.

6. Now, you are ready to make the measurements.

14

GaN Half-bridge Evaluation Board

USB Power

Main Power
Switch (S3)

MMCX Board
Power Switch (S1)

Square Pin Board
Power Switch (S2)

*Use both ports of the provided dual-input USB cable. Using any other cable can cause serious damage to the oscilloscope’s USB port.

Test 1:

Gate-charge Characterization and Effective Dead Time Analysis

• Measuring high-side V

GS

and low-side V

GS

simultaneously

• Measuring high-side V

DS

and low-side V

DS

simultaneously

Vgs_hi

Vgs_lo

Vds_hi

Vds_lo

Gate-charge Characterization and Effective Dead Time Analysis

Getting the best performance out of a SiC- or GaN­based power converter topology requires optimizing
dead time and gate charge characteristics. This
requires accurate measurement of all high-side and
low-side gate-source (VGS) and drain-source (VDS)
voltages.

Measuring the signals on the low-side, ground
referenced switch is possible with a passive probe
setup, but ground return currents and the ground lead
inductance of a traditional passive probe can degrade
the measurement. The high-side, high frequency,
floating measurements are almost impossible to
measure using traditional differential measurement
systems due to the presence of high common-mode
voltage.

The following section will demonstrate how to make all
gate-source (VGS) and (VDS) measurements accurately.

Ground referenced – Easy

Floating Measurements – Very Difficult

On-Screen Setup – All Measurements

17

1. Start by pressing the ‘Autoset’ button on
the front panel. This will adjust the
screen to show all the waveforms
simultaneously.

2. Press trigger badge and set the trigger
on Ch2 (Vgs_High Side) rising edge and
set trigger to 50%.

3. For the most accurate measurements,

the probe’s offset function should be

used to center the signals on the display;
for instance, the 0-50 V VDSsignals
should be offset 25 V).

4. Enable cursors to analyze dead time and

switch characteristics.

High-Side and Low-Side V

GS

Measurements

18

1. Turn off Ch3 and Ch4 by double-tapping the respective
channel badges and tapping the display button to OFF.
This will give a clear view of gate-source (VGS) signals on
the low side and high side.

2. Set trigger on Ch2 and adjust the horizontal scale to
10 nS/div to fill the screen with the turn-on (rise)
sequence on Ch2 (Vgs_hi).

3. Adjust the vertical scale on Ch2 and Ch1 to 1 V/div.
Alternatively, use pinch and zoom to view the gate-charge
response on the gate signals.

4. Miller plateau and dead time should be clearly visible for
analysis. Use cursors for further evaluation.

5. Use the Measure button on the right soft panel to enable

Rise Time, Fall Time, Slew rates and other critical

measurements for each channel. Note that the rise time
measurement might include the time to charge the Miller
capacitance, depending on where the reference level is
set.

High-Side and Low-Side V

Ds

Measurements

19

1. Turn on Ch3 (Vds_hi) and Ch4 (Vds_lo) and turn off
Ch1(Vgs_lo) and Ch2(Vgs_hi) to get a clear view of
drain-source (VDS) signals on the high side and low side.

2. Set the trigger on Ch3 (Vds_hi) falling edge.

3. Adjust the horizontal scale to 200 ns/div to view the
dead time between the (Ch3) Vds_hi falling edge and
Ch4 (Vds_lo) rising edge.

4. Adjust the vertical scale on Ch3 and Ch4 to 10 V/div.

Alternatively, use pinch and zoom to view the drain­source response on the full screen.

5. Dead time should be clearly visible. Use cursors to

evaluate dead time and rise-fall time characteristics.

6. Use the Measure button on the right soft panel to
enable Rise Time, Fall Time, Slew rates and other
critical measurements for each channel.

Note: Turn on Ch2 (Vgs_hi) and Ch3 (Vds_hi) to view VGS, Miller

charge time and VDSturn-on correlation (not demonstrated).

Test 2:

Fast-Switching Load Current or Drain Current Analysis

• Measuring high di/dt switching, drain/load current using a shunt

Fast-Switching Load Current or Drain Current Analysis

Testing high di/dt load and drain current is critical for optimizing power
converter performance. The most common methods of measuring these

currents involves inserting either a wire loop for a current probe or a

resistive current shunt at the test point.

The added inductance of a wire loop for a current probe can have a

negative impact on the performance of the circuit. Using a low-inductance

resistive current shunt can be a better choice in many applications.

The following section will demonstrate how to make fast-switching load

current measurements on a floating shunt using the IsoVu probe.

NOTE: The GaN Evaluation board is not designed to carry any significant
current, so use the following steps as representative test only.

I_Load
(shunt)

I_Load
(loop)

Make Connections – Current Measurements Shunt vs. Loop

22

Channel 3: Load current (IL) using a shunt – TIVM1 and MMCX25X sensor tip.

Channel 4: Load current (IL) using a loop – TCP0300A current probe.

* The GaN Evaluation board is not designed to carry any significant current, so use the following steps as representative tests only.

I

Load

(shunt)

I

Load

(loop)

Fast-Switching Load Current or Drain Current Analysis

• Make sure the Ch3 and Ch4 are deskewed using the steps noted in the Deskew

section.

• Because Ch3 -TIVM1 (IsoVu) probe is testing current through a floating shunt,

change the measurement unit to current: ‘A’.

• Double-tap the Ch3 badge and under ‘OTHER,’ toggle the ‘Alternate Units’
switch On and set the Ratio to 1 A/V. (Current shunt value on the board is 1 Ω).

• To compare the traditional loop-based measurement with the shunt-based

measurement, add current loop using a standard square pin lead on the

‘Current probe’ test-point of the lower GaN Half-bridge circuit.

• Use a high BW current probe (TCP0030A) to measure load current through the

loop. Degauss the probe before making the measurement.

23

Measuring Fast-Switching Load Current

• Average current through the shunt in
on-state is quite low (~50 mA). The
parasitic elements on output circuit

causes ringing as high as 4 A

pk-pk

,

which makes it difficult to measure the
on-state current.

• To see the on-state current, switch to

averaging mode by tapping the
Acquisition badge and selecting
Average from Acquisition mode.

• The vertical scale might have to be
extended beyond the clipping range to
view the average current.

24

Current Measurement – Shunt vs. Loop

• It can be clearly seen from the
waveform that adding a loop adds
quite a lot of ringing and settling time

(~275 ns) vs. (~50 ns) on the shunt-

based measurement.

• The effects of lower bandwidth on Hall

effect-based current probe are also

visible on the low frequency ringing on
the loop-based (Ch4) waveform.

• Adding a loop can also affect the
performance of gate drive, along with
other timing characteristics and can be
seen in VGSand VDSmeasurements.

25

Test 3:

Testing Switching/Conduction Losses and Magnetic Properties

(This test does not use the GaN Evaluation Board)

Switching/Conduction Loss and Magnetic Measurements

Getting the best efficiency and power density out of any power
converter requires accurate measurements of switching, conduction
and magnetic losses.

To measure switching and conduction losses, measure the drain­source voltage (VDS) and drain current (ID) across the power switch
using appropriate probes, as shown in the connection diagram on
the next page. For magnetic properties, measure voltage (VL)
across the coil and current (IL) through the coil.

The following section uses the built-in demo to demonstrate the
capabilities of the 5 PWR software for switching and conduction
losses.

Use your own DUT to test any of the demonstrated parameters.

ID(I_drain)

Vds_hi

IL(I_Coil)

VL(V_Coil)

Make Connections – Switching/Conduction/Magnetic

28

Switching/Conduction Losses & R

DSON

:

Channel 1: Voltage across the switch (VDS) – TIVH08 and MMCX50x sensor tip.

Channel 2: Drain current (ID) (Through shunt) – TIVM1 and MMCX25X sensor tip.

Magnetic Measurements:

Channel 1: Coil Voltage (VL) – TIVH08 and MMCX50x sensor tip.

Channel 2: Coil current (IL) (Through shunt) – TIVM1 and MMCX25X sensor tip.

Representative circuit for demonstration purposes only. GaN Evaluation board does not have test points for drain current measurements.

ID(I_drain)

Vds_hi

IL(I_Coil)

VL(V_Coil)

Power Measurements Demo

• No connections are required to enable
the 5-PWR scope demo.

• Tap ‘Utility→Demo’ on the top ribbon.

• Tap ‘Power Measurements’ under the

Demo popup.

• Select between the available choices.

• The following pages show the

screenshots and capabilities of these

functions:

◦ Switching/Conduction Loss
◦ Magnetic Measurements

29

Switching/Conduction Losses Demo

30

Magnetic Measurements Demo

31

Power Measurements Software Help

32

For detailed set­up steps and a full

description of

5-PWR software
features, please
refer to the on­screen help by
tapping the Help
button.

Specifications

IsoVu Technology Specifications

34

ISOVU TIVM SERIES ISOVU TIVH SERIES

Bandwidth

Up to 1 GHz Up to 800 MHz

Rise Time

Down to 350 ps Down to 450 ps

Differential Voltage Range

±5 mV up to ±50 V ±5 mV up to ±2.5 kV

Common Mode Voltage Range

60 kV 60 kV

Common Mode Rejection Ratio

DC – 1 MHz: 160 dB (100 Million to 1)

1 MHz – 100 MHz: 120 dB (1 Million to 1)

1 GHz: 80 dB (10,000 to 1)

DC – 1 MHz: 160 dB (100 Million to 1)

1 MHz – 100 MHz: 120 dB (1 Million to 1)

800 MHz: 80 dB (10,000 to 1)

Input Impedance

Up to 2.5 kΩ

< 1 pF

Up to 40 MΩ

As low as 2 pF

Fiber Cable Length

3 meters or 10 meters 3 meters or 10 meters

Power Over Fiber

Powered over the fiber connection –

no batteries required

Powered over the fiber connection –

no batteries required

Input Offset

Up to ± 100 V Up to ± 1000 V

Input Coupling

DC DC or AC

Note: Specifications are dependent on the probe tip cable.

Find more valuable resources at TEK.COM

Copyright © Tektronix. All rights reserved. Tektronix products are covered by U.S. and foreign patents, issued and pending. Information in this publication supersedes that in all previously published material.
Specification and price change privileges reserved. TEKTRONIX and TEK are registered trademarks of Tektronix, Inc. All other trade names referenced are the service marks, trademarks or registered trademarks of
their respective companies.
020719 LAB 48W-61538-0