NREL is a national laboratory of the U.S. Department of Energy, Office of Energy Efficiency and Renewable Energy, operated by the Alliance for Sustainable Energy, LLC.

Thermal Management and Reliability of Power Electronics and Electric Machines

Sreekant Narumanchi Organization: National Renewable Energy Laboratory Email: sreekant.narumanchi@nrel.gov Phone: 303-275-4062 NREL Team Members: Kevin Albrecht, Kevin Bennion, Emily Cousineau, Doug DeVoto, Xuhui Feng, Charlie King, Joshua Major, Gilbert Moreno, Paul Paret, Meghan Walters Advancements in Thermal Management Conference Denver, Colorado August 3, 2016 NREL/PR-5400-66754

This presentation does not contain any proprietary or confidential information.

2

Importance of Thermal Management and Reliability • Excessive temperature degrades the performance, life, and

reliability of power electronics and electric machines.

• Advanced thermal management technologies enable o Keeping temperature within limits o Higher power densities o Lower-cost materials, configurations, and system o Improve lifetime/reliability

• Predictive lifetime models help in time-and cost-effective design.

3

DOE Vehicle Technologies Office Electric Drive Technologies (EDT) Program Targets

(on-road status) • Discrete components • Silicon semiconductors • Rare-earth motor magnets

• Fully integrated components • Wide-bandgap (WBG) semiconductors • Non rare-earth motors

3

From DOE EV Everywhere Grand Challenge Blueprint, http://energy.gov/sites/prod/files/2016/05/f31/eveverywhere_blueprint.pdf

4

Advanced Packaging Reliability

Electric Machines Thermal

Management

Power Electronics Thermal

Management

NREL EDT Research Focus Areas

Research Focus Areas Will Reduce Cost and Improve Performance and Reliability

Enabling Materials

Photo Credit: Jana Jeffers, NREL

Photo Credits: Doug DeVoto, NREL Photo Credits: Doug DeVoto and Gilbert Moreno, NREL

5

Power Electronics Thermal Management

6

Power Electronics Thermal Management Strategy • Packages based on WBG devices

require advanced materials, interfaces, and interconnects o Higher temperature capability o Higher effective thermal conductivity

• Low-cost techniques to increase heat transfer rates o Coolants – water-ethylene glycol

(WEG), air, transmission coolant, refrigerants

o Enhanced surfaces o Flow configurations

• System-level thermal management

(capacitor and other passives)

WBG Module Packaging Design with Integrated Cooling

Device

Metalized Substrate Substrate Attach

Integrated Base/Cold Plate

Die Attach

Interconnect Encapsulant

Enclosure Terminal

Coolant Inlet

Coolant Outlet

7

The Challenge with Interfaces/Interface Materials • Interfaces can pose a major bottleneck to heat removal.

• Bond materials, such as solder, degrade at higher temperatures and are prone to

thermomechanical failure.

• Problem can become more challenging for configurations employing WBG devices.

2

2.5

3

3.5

4

4.5

100 110 120 130 140 150 160Dist

ance

acro

ss th

e Pa

ckag

e (m

m)

Temperature (°C)

New Degraded

Degraded Substrate Attach

132.5°C 156.7°C

Module Packaging Design with Integrated Cooling

Device

Metalized Substrate Substrate Attach

Integrated Base/Cold Plate

Die Attach

Interconnect Encapsulant

Enclosure Terminal

Coolant Inlet

Coolant Outlet

8

Thermal Resistance of Various Non-Bonded Thermal Interface Materials (TIMs)

• Red dashed line in the two figures above is the target thermal resistance (3 to 5 mm2K/W).

• Most non-bonded TIMs do not come close to meeting thermal specification of 3 to 5 mm2K/W at approximately 100-μm bond line thickness.

0

50

100

150

200

250

300

350

400

0 0.05 0.1 0.15 0.2Thickness (mm)

Ther

mal

resi

stan

ce (m

m2 K

/W)

Arctic Silver 5

Thermalcote-251G

Wacker Silicone P12

Keratherm KP77

Dow Corning TC5022

Shinetsu-X23-7783D-S

Chomerics XTS8010

Chomerics XTS8030

3M AHS1055M

Honeywell PCM45

172 kPa, ~ 75 C sample temperature TIM thickness in all cases is 100 μm

0

25

50

75

100

125

150

175

200

225

250

275

Ther

mal

cote

25

1G

Wac

ker

Sili

cone

P12

Arc

tic S

ilver

5

Ker

athe

rm

KP

77

Cho

mer

ics

XTS

8010

Cho

mer

ics

XTS

8030

Hon

eyw

ell

PC

M45

Dow

Cor

ning

TC

-502

2

Shi

nets

u X-

23-

7783

D-S

3M A

HS

105

5M

Res

ista

nce

(mm

2 K/W

)

Bulk resistance Contact resistance

9

Pump Laser

Function Generator Lock-in Amplifier

Multi-mode optical fiber

Collimation lenses

Lens tube

Modulated pump laser

Optical chopper

Probe laser

Reflecting mirrors

Photodiode

Modulated pump laser

Heat transfer

Probe laser

Probed temperature

variation

Sample

Thermal Resistance of Thermoplastics with Embedded Carbon Fibers

• Thermoplastics with embedded carbon fibers show very good thermal performance

Frequency-domain transient thermoreflectance experiment configuration Thermoplastic

film HM-2

Bondline thickness (µm)

60

Bulk thermal conductivity (W/m·K)

37.5 ± 6.8

Contact resistance (mm2·K/W)

3.1 ± 1.1

Total thermal resistance (mm2·K/W)

7.5 ± 1.9

Photo: Courtesy of BtechCorp

10

• Bonded interface resistance in the range of 0.4 to 2 mm2K/W is possible. Materials developed in the DARPA programs are in this range

o Copper nanowires o Boron-nitride nanosheets (0.4 mm2K/W for 30- to 50-µm bondline thickness) o Copper nanosprings (1 mm2K/W for 50-µm bondline thickness with very good reliability) o Graphite solder o Nanotube-based

Other Bonded Interface Materials

11

Integrated Module Heat Exchanger

• Up to 100% increase in power per die area

• Up to 34% increase in coefficient of performance (efficiency)

NREL integrated module heat exchanger Patent No.: US 8,541,875 B2

Photo Credit: Kevin Bennion, NREL

12

Liquid Jet-Based Plastic Heat Exchanger

Metalized substrate

Base plate

Plastic manifold

Device

Bonded interface material (BIM)

Wire/ribbon bonds

WEG jets

Enhanced surface

• Up to 12% increase in power density

• Up to 36% increase in specific power

Photo Credit: Doug DeVoto, NREL

Photo Credit: Gilbert Moreno, NREL

13

Two-Phase Cooling for Power Electronics

Fundamental Research

Module-Level Research

Inverter-Scale Demonstration

Vapor Liquid

Evaporator

Characterized performance of HFO-1234yf and HFC-245fa

Achieved heat transfer rates of up to ~200,000 W/m2-K

Reduced thermal resistance by over 60% using immersion two-

phase cooling of a power module

Quantified refrigerant volume requirements

Dissipated 3.5 kW of heat with only 250 mL of refrigerant

Predicted 58%–65% reduction in thermal resistance via indirect and

passive two-phase cooling

Power electronics modules

Evaporator

Air-cooled condenserPhoto Credit: Bobby To, NREL

Photo Credit: Gilbert Moreno, NREL Photo Credit: Gilbert Moreno, NREL

14

WBG Power Electronics Thermal Management

Create thermal models of an automotive inverter

Simulate WBG operation using the inverter model

Explore advanced cooling strategies

Quantify the inverter component temperatures

under elevated device temperatures

Validate the thermal models

Evaluate different module topologies

Experimentally validate some key thermal management

concepts

cold plate

copper copper-molybdenum direct-bond copper

cold plate

Develop thermal management concepts to

enable WBG power electronics

Identify the primary thermal paths through which heat is

conducted from the devices to the other components

0

10

20

30

40

50

60

70

80

90

100

0 2 4 6 8 10 12 14 16

Spec

ific

ther

mal

resi

stan

ce: R

" th

,j-l (m

m2 -

K/W

)

Flow rate (liters per minute)

Experiment

Model: maximum

Model: average

Photo Credit: Scot Waye, NREL

15

Advanced Packaging Reliability

15

16

Bonded Interface Material Reliability

• Thermoplastics yield very good reliability

• Reliability of

sintered silver is better than solder

Photo Credit: Doug DeVoto, NREL

17

Bonded Interface Material Reliability

• Thermoplastics yield very good reliability

• Reliability of

sintered silver is better than solder

18

Bonded Interface Material Finite Element Modeling

-80

-40

0

40

80

120

160

0 2000 4000 6000 8000 10000 12000 14000

Tem

pera

ture

(°C)

Time (s)

Temperature Cycling Profile • Temperature cycling parameters: – -40°C to 150°C – 5°C/minute ramp rate – 10 minute dwell/soak time

• Anand viscoplastic material model applied to solder layer

• Temperature-dependent elastic material properties incorporated for base plate and substrate

Quarter Symmetry Model

Substrate

Solder

Base Plate

19

Strain Energy Density

Top view of strain energy density contour plot in the solder layer

Corner fillet region

• Volume-averaged strain energy density/cycle (ΔW) in the corner fillet region is the final output

20

Impact of Geometric Variations on Reliability

Profile Joint Thickness (µm) ΔW (MPa) Predicted experimental

cycles to failure – Nf

Thermal cycle

50 2.65 465

100 1.36 1,400

150 0.93 2,600

• Joint Thickness

• Substrate Variation

Profile Substrate CTE (x 10-6/°C) ΔW (MPa)

Predicted experimental

cycles to failure – Nf

Thermal cycle

Si3N4 2.8 1.36 1,400

AlN 4.5 1.08 2,000

Al2O3 8.1 0.44 9,000

Predictive Lifetime Model, Nf = 2312.5 (ΔW)-1.645

CTE: coefficient of thermal expansion Si3N4 : silicon nitride AlN: aluminum nitride Al2O3: aluminum oxide

21

Electric Machines Thermal Management

22 DOE APEEM FY14 Kickoff Meeting

Stator Laminations

Slot Winding

Rotor Laminations Rotor

Hub

Case

Air Gap

Stator-Case Contact

Nozzle/Orifice

ATF Impingement

Shaft ATF Flow

ATF Flow

Stator Cooling Jacket

End Winding

Motor Axis

Electric Machines Thermal Management Strategy

Problem • Multiple factors impacting heat

transfer are not well quantified or understood.

Contributing Factors 1. Direction-dependent thermal

conductivity of lamination stacks 2. Direction-dependent thermal

conductivity of slot windings and end windings

3. Thermal contact resistances (stator-case contact, slot-winding interfaces)

4. Convective heat transfer coefficients for ATF cooling

5. Cooling jacket performance

4

1

1

2

3

5

Motor Cooling Section View

ATF: Automatic Transmission Fluid

23

Transmission Oil Jet Heat Transfer Characterization

Heat transfer coefficients of all target surfaces at 50°C inlet temperature

• Surface features increase heat transfer

18 AWG surface target

Note: Heat transfer coefficient calculated from the base projected area (not wetted area)

Top View

Side View 50°C Inlet Temperature

0

2000

4000

6000

8000

10000

12000

0 2 4 6 8 10 12

Heat

Tra

nsfe

r Coe

ffici

ent [

W/m

2 K]

Velocity [m/s]

18 AWG22 AWG26 AWGBaseline

Photo Credits: Gilbert Moreno, NREL

24

• Enclosure allows direct impingement on motor for heat transfer measurements and flow visualization

Stator Thermal Management with Transmission Oil

Enclosure with stator and ATF cooling Jet impingement on target surface

Photo Credits: Kevin Bennion, NREL

25

• Performing thermal analysis on passive thermal materials

Motor Passive Thermal Materials Characterization

Slot windings

Slot liner or ground insulation

Stator laminations

Measure cross-slot winding thermal conductivity

Measure winding-to-liner thermal contact resistance

Measure liner-to-stator thermal contact resistance

26

Lamination Stack Effective Thermal Conductivity

Measured Stack Thermal Resistance

Lamination-to-Lamination Thermal Contact Resistance

Effective Through-Stack Thermal Conductivity

0.6

0.8

1.0

1.2

1.4

1.6

1.8

2.0

2.2

0 200 400 600

Asym

ptot

ic S

tack

The

rmal

Con

duct

ivity

[W

/m-K

]

Pressure [kPa]

M19 26 Gauge

M19 29 Gauge

HF10

Arnon7

Error bars represent 95% confidence level

27

Transverse Winding Thermal Conductivity

0.00.20.40.60.81.01.21.41.61.8

3 4 5 6 7

Ther

mal

Con

duct

ivity

[W/m

-K]

Sample Thickness [mm]

22 AWG Transverse

data

FEA

0.00.20.40.60.81.01.21.41.61.8

3 4 5 6 7

Ther

mal

Con

duct

ivity

[W/m

-K]

Sample Thickness [mm]

19 AWG Transverse

data

FEA

• Error bars represent 95% measurement uncertainty (U95)

• Finite element analysis (FEA) model results based on measured sample copper fill factor

• FEA assumes hexagonal or closed-pack wire pattern

Photo Credit: Emily Cousineau, NREL

28

Slot Liner Paper and Thermal Interface Resistances

• Limited sample size for measurements

• Slot liner paper thermal conductivity of 0.18 W/m-K (thickness 0.29 mm)

• Winding thermal conductivity measured to be 0.88 ± 0.11 W/m-K (U95)

• FEA estimate of thermal conductivity is 0.99 W/m-K for measured copper fill factor

0

1000

2000

3000

Slot Liner Paper Paper to Windingbond

Paper to Flat Copper

Laye

r The

rmal

Res

ista

nce

[mm

²K/W

]

29

• Industry-led inverter development with VTO and AMO funding – Delphi inverter (VTO) – GM inverter (VTO) – Wolfspeed WBG inverter (VTO) – John Deere WBG inverter (AMO)

• UQM Technologies motor development (VTO funding)

Participation in Industry-Led Projects

AMO: Advanced Manufacturing Office VTO: Vehicle Technologies Office

30

Summary

• Low-cost, high-performance thermal management technologies are helping meet aggressive power density, specific power, cost, and reliability targets for power electronics and electric machines.

• NREL is working closely with numerous industry and research partners to help influence development of components that meet aggressive performance and cost targets through: o Development and characterization of cooling technologies o Thermal characterization and improvements of passive stack materials and interfaces.

• Thermomechanical reliability and lifetime estimation models are important

enablers for industry in cost- and time-effective design.

NREL is a national laboratory of the U.S. Department of Energy, Office of Energy Efficiency and Renewable Energy, operated by the Alliance for Sustainable Energy, LLC.

For more information, contact:

NREL APEEM Section Supervisor Sreekant Narumanchi sreekant.narumanchi@nrel.gov Phone: (303) 275-4062

Acknowledgments:

Susan Rogers and Steven Boyd EDT Program Vehicle Technologies Office Advanced Manufacturing Office U.S. Department of Energy

Industry and Research Partners Industry OEMs Ford, GM, FCA, John Deere, Tesla, Toyota

Suppliers/Others 3M, NBETech, Curamik, DuPont, GE Global Research, Semikron, Kyocera, Sapa, Delphi, Btechcorp, ADA Technologies, Remy/BorgWarner, Heraeus, Henkel, Wolverine Tube Inc., Wolfspeed, Kulicke & Soffa, UQM Technologies, nGimat LLC

Agencies DARPA

National Laboratories Oak Ridge National Laboratory, Ames Laboratory

Universities Virginia Tech, University of Colorado Boulder, University of Wisconsin, Carnegie Mellon University, Texas A&M University, North Carolina State University, Ohio State University, U.S. Naval Academy