low thermal resistance igbt structure · low thermal resistance igbt structure daniel benhammou...

Low Thermal ResistanceIGBT Structure

Daniel Benhammou Email: [email protected] Phone:303-275-4458 National Renewable Energy Laboratory

Vahab Hassani (NREL)

FY05 Budget: $ 400K FY06 Budget: $ 500K Project Duration: FY05 to FY06

FreedomCAR APEEM FY05 Wrap-up/FY06 Kick-off Meeting Oak Ridge National Laboratory National Transportation Research Center

November 2, 2005

November 2, 2005 FreedomCAR FY05 Wrap-up/FY06 Kick-off 2

Where Does This Project Fit?

Power

Electronics

Electrical

Machines

Thermal

Systems

Wide Band-Gap Mat’ls

High-Temp. Capacitors

Wide Band-Gap Mat’ls

High-Temp.Capacitors

Magnets

Reduce Thermal Resistance

Air Cooling105C and 85C WEG

Increase Heat-Transfer Coefficient

Low Thermal Resistance IGBT Structure

/ i 3

What is This Project? • Maximize heat transfer from IGBT structures by

removing layers of greatest thermal resistance. By doing this, we can cool inverter power electronics by means of jet impingement cooling directly on DBC.

November 2, 2005 FreedomCAR FY05 Wrap-up FY06 K ck-off

/ i 4

Technical Approach

123

Design Process

Layout

geomoetry

Modify

Geomoetry

Hand

calculations

Fabricate

Prototype

Design

Simulation

Experimentation

Legend

Computer Model

Physical Model

YES

YES

NO

NO

Run

performance

experiments

Analyze

failure criteria

Needs

Revising?

YES

NO

Import

GeometryGenerate Mesh

Q uality

Mesh?Refine Mesh

Apply Boundary

ConditionsSolve Model

Converged

Results?

Meets

Design

Criteria?

NO

NO

YES

YES

Validate

Simulation

Data

Design

Complete

New Design?

Meets

Design

Criteria?

NO

YES


/ i 5

Technical Approach

• li ili ll ltipl i isiwi i i i i l

li i

• l i ii l i l

Positives • i

• i l

ltipl i i i i l l lt in l l i i

Negatives • Ti i

How do we address positives and negatives?


Mode ng capab ty a ows mu e des gn rev ons thout the h gh cost of fabr cat on and exper menta

va dat on.

Low therma res stance IGBT structures prov de a cost effect ve so ut on to meet DOE program goa s.

Cost Effect ve

Fewer phys caprototypes

Mu e des gn terat ons and numer ca mode s resuonger ead t me to f rst run prototypes. me Consum ng

Design

Simulation

Experimentation

/ i 6

Goal

• l isili i l

• ll iiali i

–

–

– / 2


Reduce therma res stance between con d e and coo ant

Produce a fu y operat ng prototype ready for commerc zat on

105°C inlet temperature

125°C maximum die temperature

Heat flux: 200 W cm

/ i 7

Approach for FY05

• li i

• l i– l

– l

–

• i

• i


Estab sh ndustry partner

Mode phys cs Structura analysis

Therma analysis

CFD simulations

6 res stor proof of concept

Iterate des gn

/ i 8

Major Accomplishments for FY05

• i l i i

• i l si l iibili

• i l Hi ll il i l li i

• ly i i i


CRADA n p ace w th Sem kron

Numer ca mu at ons corroborate feas ty of cut-through IGBT structure

Exper menta gh Heat F ux (HHF) test oop bu t for exper menta va dat on

Current n 8th generat on of des gn

2

/ i 9


l

l


Cross Section of IGBT Considered

Case A : Solid Spreader Plate

Case B : Spreader Plate with Square Holes

Therma grease

So der

/ i

Conventional IGBT Structure

November 2, 2005 FreedomCAR FY05 Wrap-up FY06 K ck-off 10

/ i90.00

100.00

110.00

120.00

130.00

140.00

150.00

160.00

170.00

0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00

Distance from Top of IGBT, mm

Te

mp

., C

Conventional IGBT Structure


/ i



/ i


104.00

105.00

106.00

107.00

108.00

109.00

110.00

111.00

112.00

113.00

0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80 Distance from top of the IGBT

f


Temp. for cut thru

Temp. or cut thru

/ i

Exploded view of design


/ i

• l

– Refinement needed in areas of steep gradients

•


Finite element mesh of Semikron inverter phase

Mesh must represent physicabody

Optimize mesh to reduce computation time

/ i

Structural Analysis

• – Applied

– Thermally induced

• – T = 125

– T if = -40

– P = 15 MPa


Analyze stress

Boundary conditions

ref °C

un °C

/ i

Thermal Analysis

• – Inlet Temp: 105

– hj =65,000 W/m2

– h ll 2

•

• – O-ring overlaps IGBT

– Diminishes cooling capability


Boundary Conditions °C

et °C

wa =15,000 W/m °C

Max Temp: 142.2°C

Why?

/ i

Proposed modification

•

– O-ring no longer blocks flow

•

– Fl /

– l

–


DBC extended 5mm

Increased cooling

ux: 100 W cm^2

In et: 105°C

Max Temp: 125.8°C

/ i

CFD modeling

• /

• •


Single multiple jets

Visualize flow paths Identify problematic geometry

/ i

Approach for FY06

• l i i

• ipl i l li i

• i imi i i–

–

–


Comp ete 6-res stor exper ments

Fabr cate 1st prototype heat exchanger ate for exper menta va dat on

Cont nue to opt ze conf gurat on Jet dimensions

Nozzle configurations

Impingement geometry

/ i

Semikron heat exchanger plate


/ i

Integrated heat exchanger plate


Timeline for FY06

FY06

Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep

Build / test small scale IGBT Structure

Build / test 1st generation prototype Complete

Laboratory Testing

Analyze experimental data

Modify design as requ

uild / tes

ired Functional Prototype Inverter

B t generation 2

Deliver final design

/ iNovember 2, 2005 FreedomCAR FY05 Wrap-up FY06 K ck-off 23

Key task Milestone/ Deliverable

/ i

Barriers/Challenges

• i i i i–

• Cast parts

• Minimize machining time

• Minimize components

• ial Limi

• i


Fabr cat on cons derat ons Cost effective manufacturing solutions

Mater ts

Eros on

/ i

Interactions and Collaborations

• NREL • Design, model, test

components

• Semikron • Assemble Prototype


National Labs Universities Industry Partners

Questions

low thermal resistance igbt structure · low thermal resistance igbt structure daniel benhammou...

Documents