Over-Temperature protection for IGBT modules

Similar documents
Uncertainty in measurements of power and energy on power networks

Physics 4B. A positive value is obtained, so the current is counterclockwise around the circuit.

Numerical modelization by finite differences of a thermoelectric refrigeration device of double jump". Experimental validation.

FE REVIEW OPERATIONAL AMPLIFIERS (OP-AMPS)( ) 8/25/2010

Advanced Circuits Topics - Part 1 by Dr. Colton (Fall 2017)

Sensors & Transducers Published by IFSA Publishing, S. L.,

Lab 2e Thermal System Response and Effective Heat Transfer Coefficient

Key component in Operational Amplifiers

R. W. Erickson. Department of Electrical, Computer, and Energy Engineering University of Colorado, Boulder

AGC Introduction

NUMERICAL DIFFERENTIATION

Dynamics of a Superconducting Qubit Coupled to an LC Resonator

Transfer Functions. Convenient representation of a linear, dynamic model. A transfer function (TF) relates one input and one output: ( ) system

Aging model for a 40 V Nch MOS, based on an innovative approach F. Alagi, R. Stella, E. Viganò

Singular Value Decomposition: Theory and Applications

Chapter 6. Operational Amplifier. inputs can be defined as the average of the sum of the two signals.

Thermal Analysis of Power Semiconductor Converters

( ) = ( ) + ( 0) ) ( )

THE CURRENT BALANCE Physics 258/259

Energy Storage Elements: Capacitors and Inductors

3.6 Limiting and Clamping Circuits

Physics 5153 Classical Mechanics. Principle of Virtual Work-1

Color Rendering Uncertainty

One-sided finite-difference approximations suitable for use with Richardson extrapolation

E40M Device Models, Resistors, Voltage and Current Sources, Diodes, Solar Cells. M. Horowitz, J. Plummer, R. Howe 1

Temperature. Chapter Heat Engine

Study on Non-Linear Dynamic Characteristic of Vehicle. Suspension Rubber Component

The Order Relation and Trace Inequalities for. Hermitian Operators

2010 Black Engineering Building, Department of Mechanical Engineering. Iowa State University, Ames, IA, 50011

Thermal Analysis of Power Semiconductor Converters

Pulse Coded Modulation

Numerical Transient Heat Conduction Experiment

ELECTRONIC DEVICES. Assist. prof. Laura-Nicoleta IVANCIU, Ph.D. C13 MOSFET operation

Lecture #06 Hotwire anemometry: Fundamentals and instrumentation

Scroll Generation with Inductorless Chua s Circuit and Wien Bridge Oscillator

Module 3 LOSSY IMAGE COMPRESSION SYSTEMS. Version 2 ECE IIT, Kharagpur

Pop-Click Noise Detection Using Inter-Frame Correlation for Improved Portable Auditory Sensing

APPLICATION OF EDDY CURRENT PRINCIPLES FOR MEASUREMENT OF TUBE CENTERLINE

Indeterminate pin-jointed frames (trusses)

N- and P-Channel 20-V (D-S) MOSFET

Lecture 13 APPROXIMATION OF SECOMD ORDER DERIVATIVES

Inductance Calculation for Conductors of Arbitrary Shape

2016 Wiley. Study Session 2: Ethical and Professional Standards Application

Clock-Gating and Its Application to Low Power Design of Sequential Circuits

Unit 1. Current and Voltage U 1 VOLTAGE AND CURRENT. Circuit Basics KVL, KCL, Ohm's Law LED Outputs Buttons/Switch Inputs. Current / Voltage Analogy

The optimal delay of the second test is therefore approximately 210 hours earlier than =2.

MAE140 - Linear Circuits - Winter 16 Midterm, February 5

Department of Electrical & Electronic Engineeing Imperial College London. E4.20 Digital IC Design. Median Filter Project Specification

CHAPTER 5 NUMERICAL EVALUATION OF DYNAMIC RESPONSE

DC Circuits. Crossing the emf in this direction +ΔV

Boise State University Department of Electrical and Computer Engineering ECE 212L Circuit Analysis and Design Lab

Linear Approximation with Regularization and Moving Least Squares

Erratum: A Generalized Path Integral Control Approach to Reinforcement Learning

Appendix B. The Finite Difference Scheme

ACTM State Calculus Competition Saturday April 30, 2011

B and H sensors for 3-D magnetic property testing

The internal structure of natural numbers and one method for the definition of large prime numbers

Odd/Even Scroll Generation with Inductorless Chua s and Wien Bridge Oscillator Circuits

1.4 Small-signal models of BJT

Copyright 2004 by Oxford University Press, Inc.

MEASUREMENT OF MOMENT OF INERTIA

Week3, Chapter 4. Position and Displacement. Motion in Two Dimensions. Instantaneous Velocity. Average Velocity

Output Stages and Power Amplifiers

Flyback Converter in DCM

Linearity. If kx is applied to the element, the output must be ky. kx ky. 2. additivity property. x 1 y 1, x 2 y 2

Electrical Circuits 2.1 INTRODUCTION CHAPTER

Lecture Notes on Linear Regression

Pressure Measurements Laboratory

Physics 114 Exam 2 Fall 2014 Solutions. Name:

CHAPTER 9 CONCLUSIONS

Massachusetts Institute of Technology Department of Electrical Engineering and Computer Science Circuits and Electronics Spring 2001

Gravitational Acceleration: A case of constant acceleration (approx. 2 hr.) (6/7/11)

CHAPTER 14 GENERAL PERTURBATION THEORY

CLARKSON UNIVERSITY. Block-Based Compact Thermal Modeling of Semiconductor Integrated Circuits

Week 11: Differential Amplifiers

Tensor Smooth Length for SPH Modelling of High Speed Impact

Psychology 282 Lecture #24 Outline Regression Diagnostics: Outliers

Supplementary Notes for Chapter 9 Mixture Thermodynamics

A Hybrid Variational Iteration Method for Blasius Equation

CONTRAST ENHANCEMENT FOR MIMIMUM MEAN BRIGHTNESS ERROR FROM HISTOGRAM PARTITIONING INTRODUCTION

Department of Electrical and Computer Engineering FEEDBACK AMPLIFIERS

DUE: WEDS FEB 21ST 2018

3.2 Terminal Characteristics of Junction Diodes (pp )

Frequency dependence of the permittivity

Design and Optimization of Fuzzy Controller for Inverse Pendulum System Using Genetic Algorithm

ASSOCIATION POUR LA CERTIFICATION DES MATERIAUX ISOLANTS

Study on Active Micro-vibration Isolation System with Linear Motor Actuator. Gong-yu PAN, Wen-yan GU and Dong LI

Thermal State Investigations of Two-stage Thruster with Anode Layer

Composite Hypotheses testing

DEGREE REDUCTION OF BÉZIER CURVES USING CONSTRAINED CHEBYSHEV POLYNOMIALS OF THE SECOND KIND

Kernel Methods and SVMs Extension

Physics 114 Exam 2 Spring Name:

Power law and dimension of the maximum value for belief distribution with the max Deng entropy

STUDY OF A THREE-AXIS PIEZORESISTIVE ACCELEROMETER WITH UNIFORM AXIAL SENSITIVITIES

MAE140 - Linear Circuits - Fall 13 Midterm, October 31

x i1 =1 for all i (the constant ).

System in Weibull Distribution

6.01: Introduction to EECS I Lecture 7 March 15, 2011

Convexity preserving interpolation by splines of arbitrary degree

Thermal-Fluids I. Chapter 18 Transient heat conduction. Dr. Primal Fernando Ph: (850)

Transcription:

Over-Temperature protecton for IGBT modules Ke Wang 1, Yongjun Lao 2, Gaosheng Song 1, Xanku Ma 1 1 Mtsubsh Electrc & Electroncs (Shangha) Co., Ltd., Chna Room2202, Tower 3, Kerry Plaza, No.1-1 Zhongxns Road, Futan Dstrct, Shenzhen 2 Shenzhen Inovance Technology Co., Ltd., Chna E-mal:WangKe@mesh.chna.meap.com Abstract Ths paper wll summarze the approaches and applcatons of temperature sensor for IGBT modules, ncludng drect measurement of mountng a sensor on the chp and the NTC nsde IGBT modules. Addtonally, the calculaton methods of the maxmum juncton temperature T jmax for low output frequences are ntroduced. Ths paper wll also show how to desgn temperature sensng crcut and set over-temperature protecton for Mtsubsh CM450DX-24S1. 1. Introducton to the approaches of temperature sensor for IGBT modules 1.1. Dode as a temperature sensor on IGBT chp Forward voltage of dode lnearly vares wth temperature. By utlzng ths characterstc, we can use dode as a temperature sensor. For Mtsubsh J-Seres T-PM IGBT, dode ntegrated on IGBT chp s used to detect chp temperature drectly and enable to turn off IGBT safely and fast for over-temperature protecton. Fg.1 shows the IGBT chp ntegrated temperature sensng dode used for T-PM IGBT. Fg. 1. Temperature sensng dode nsde T-PM IGBT 1.2. NTC ntegrated as a temperature sensor nsde IGBT Fg.2 shows the NTC nsde IGBT modules, whch s used n Mtsubsh NX6 and NX6.1 IGBT modules. The NTC mounted on the ceramc substrate makes the thermal measurements smple n nverter.

Fg. 2. NTC nsde IGBT module 2. Usng the NTC as a temperature sensor and OT protecton 2.1. Defnton of the NTC The NTC s negatve temperature coeffcent thermstor, whch s located on the same ceramc substrate as the IGBT and dode chps for Mtsubsh NX6 and NX6.1 IGBT modules. The IGBT chp temperature can be calculated by usng a thermal model and measurng the temperature of NTC n steady state. 2.2. The approaches of calculatng T jmax for low output frequences The NTC s desgned for detecton of a long term overload condtons, whle t s not sutable for OT protecton durng short crcut condton or very short term overload. The prncple of calculatng the maxmum juncton temperature T jmax under short pulse s shown as the follow. 2.2.1. How to calculate T jave and T jmax Refer to Fg.3, wth the ad of the thermal resstances defned by reference ponts (h - heat snk, c - case, j juncton), the average juncton temperature T jave and the maxmum juncton temperature T jmax are calculated usng the followng two equatons Eq.1 and Eq.2. The average losses can be calculated usng Mtsubsh s Melcosm Ver.5.0.1 T = T + P * R + P * R Eq.1 jave h ave th( c- h) ave th( j - c ) T T + P * R +ΔT jmax h ave th( c- h) j- c_ max = T +ΔT C j- c _max Eq.2 wth R th(j-c) as thermal resstance juncton to case for the IGBT, R th(c-h) as thermal resstance case to heat snk. 2.2.2. The methods of measurng T h and T C The heat snk temperature T h s measured underneath the module n a borehole of up to 2mm nto the heat snk surface. The case temperature T C s measured drectly beneath the chp va a drll hole n the heat snk as descrbed n Fg.3. Besdes, T C can also be measured by dggng a hole of Ø0.8mm at just under the chp on the base plate of the modules as descrbed n Fg.4, and ths specal sample can be customzed by Mtsubsh.

Fg. 3. Measurng ponts of T h and T C Fg. 4. Sample wth thermo couple suppled by Mtsubsh 2.2.3. The prncple of calculatng ΔT j-c_max Under the actual operaton for IGBT modules, the chps nsde the modules are heated and cooled by the dfferent ampltude and pulse duraton n a PWM chopped current. The juncton temperature T jmax oscllates wth the frequency of the output current. The prncple of calculatng the maxmum temperature dfference between juncton and case ΔT j-c_max s shown n Fg.5. Fg. 5. (a) PWM chopped current (b) Input power pules

(c) Power pulses separated nto postve and negatve pulses (d) T j-c change caused by postve and and negatve pulses (e) ΔT j-c composte Based on the above superposton prncple, the followng equaton Eq.3 s used to calculate the temperature at the end and start of every pulse. Δ T ( jc - ) [@ t1] = P1 * R [@ t1] Δ T ( jc - ) [@ t2] = P1 * R [@ t2] - P1 * R [@ t2 - t1] Δ T ( jc - ) [@ t3] = P1* R [@ t3] - P1* R [@ t3 - t1] + P2 * R [@ t3 - t2] n t τ R [@ t] = r (1 - e ), t = t1, t2 - t1, t3 - t2 th( j - c) = 1 = r * c τ The r and c can be easly extracted from a measured coolng curve of the IGBT modules, and the coeffcents of τ and r are provded n the datasheet. At output frequences hgher than 5Hz, an approxmaton of usng rectangular shaped block to calculate the average losses. A smple method of calculatng ΔT j-c_max, as a functon of the output current frequency f O, s as the followng formula Eq.4. 1-2* f o * τ n 1- e Δ T jc - _max = 2* P av * r = 1 1 - Eq.4 f o * τ 1- e 2.2.4. Temperature dfference between juncton and case for low output frequences For example, usng CM450DX-24S1 to develop 110kW two-level G/P nverter, the rated applcaton condtons are that: V DC =540V, I O =210Arms, overload=315arms, f c =3 khz, R g(on) = R g(off) =5Ω.The average loss for upper or bottom IGBT s 306.53W. Based on ths loss, Eq.3

ΔT j-c_max s calculated at low output frequences (Table.1). ΔT j-c_max can be calculated usng the followng formula Eq.5. T T T T f P j max C +Δ j c_ max = C + corr * ave Eq.5 The factor f corr can be obtaned usng the followng formula Eq.6. f corr ( f ) o ΔT j c_ max = Eq.6 Pave f o (Hz) ΔT j-c_max f corr f o (Hz) ΔT j-c_max f corr 0.1 60.37 5.08 2 53.30 5.75 0.2 60.21 5.09 3 50.06 6.12 0.3 59.97 5.11 4 47.44 6.46 0.4 59.66 5.14 5 45.29 6.77 0.5 59.31 5.17 6 43.51 7.05 0.6 58.92 5.20 7 42.00 7.30 0.7 58.52 5.24 8 40.70 7.53 0.8 58.11 5.27 9 39.57 7.75 0.9 57.69 5.31 10 38.57 7.95 1 57.27 5.35 Table.1. ΔT j-c_max and f corr for low output frequences For output frequences above 10Hz, ΔT j-c_max can also be calculated usng Melcosm Ver.5.0.1. 2.2.5. The dfferences between T NTC and T C For IGBT modules and heat snk, the below pcture Fg.6 shows the thermal dstrbuton and postons of dfferent reference ponts for T NTC and T C. Fg. 6. Thermal dstrbuton and postons of T NTC and T C Under constant average losses, T NTC and T C are slghtly nfluenced by output frequences, so the dfferences between T NTC and T C can be calculated wth equaton Eq.6.

T T = k * P NTC C ave T = T k * P C NTC ave Eq.7 The factor k can be obtaned by testng T NTC and T C under dfferent P ave. 2.2.6. Usng the NTC as over-temperature protecton for IGBT modules Based on Eq.5 and Eq.6, Eq.7 s derved as the followng. T max T + ( f k)* P Eq.8 j NTC corr ave Takng CM450DX-24S1 as an example, T jmax has to be less than 150 to ensure SCSOA and RBSOA. T 150 C ( f k) * P Eq.9 NTC corr ave Usng the NTC temperature T NTC as OT protecton, ths OT protecton value can be fnally acheved by checkng the relaton of f corr, k and P ave n software. 3. Desgn detals of temperature sensng crcut usng dode and the NTC 3.1. Usng dode as a temperature sensor to desgn temperature sensng crcut To mnmze the nfluence of the flowng current nsde dode, a small current s employed.fg.7 shows the applcaton crcuts for T-PM IGBT. Fg. 7. (a) Usng pull-up resstor (b) Usng constant current 3.2. Usng the NTC as a temperature sensor to desgn temperature sensng crcut 3.2.1. The solaton consderatons between the NTC and chps Snce the NTC nsde IGBT modules could be exposed to a hgh voltage level durng the falure, a functonal solaton for the NTC couldn t be suffcent that renforced solaton s often requred n nverter. In accordance wth EN 50178, addtonal solaton has to be done externally. 3.2.2. Desgn detals of temperature sensng crcut Based on EN 50178, proper solaton levels have to be guaranteed for all parts of a pece of

equpment that can be touched by a person. The solaton aganst hgh voltage can be acheved by an opto-coupler to make the NTC solate from the control logc. To lmt the self heatng of the NTC by the flowng current, 3 to 4mA of flowng current are recommended. Fg.8 s a common sensng crcut. Fg. 8. Applcaton crcut for temperature sensng 4. Concluson The NTC s ntegrated nsde IGBT modules as a temperature sensor to make the desgn of an accurate temperature measurement easy. Based on the relaton of T jmax and T NTC, t s easy to set the OT protecton pont to protect IGBT modules. 5. References [1] Tetsuya Ueda, Smple, Compact, Robust and Hgh-performance Power module T-PM (Transfer-molded Power Module), n Proc. of the 22nd Internatonal Symposum on Power Semconductor Devces & IC's, Hroshma, Japan, June 2010, pp.47-50. [2] ON Semconductor, AN569/D:Transent Thermal Resstance-General Data and Its Use