thin amorphous core material for power applications...•objective is to replace ferrite cores with...

19
Thin Amorphous Core Material for Power Applications Paul McCloskey, Santosh Kulkarni, Ansar Masood, Cian O’Mathuna Integrated Magnetics, Micro-Nano Systems Center, Tyndall National Institute, Cork, Ireland

Upload: others

Post on 26-Jun-2020

1 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Thin Amorphous Core Material for Power Applications...•Objective is to replace ferrite cores with high flux density thin film material with improved performance • Three different

Thin Amorphous Core Material for Power Applications

Paul McCloskey, Santosh Kulkarni, Ansar Masood, Cian O’Mathuna

Integrated Magnetics, Micro-Nano Systems Center, Tyndall National Institute, Cork, Ireland

Page 2: Thin Amorphous Core Material for Power Applications...•Objective is to replace ferrite cores with high flux density thin film material with improved performance • Three different

Outline

• Motivation

• Analysis of magnetic core losses

• Review of soft magnetic thin films

• Material performance comparison

• Post-processed nanocrystalline thin film

• New low loss alloy systems

• Conclusions

2

Page 3: Thin Amorphous Core Material for Power Applications...•Objective is to replace ferrite cores with high flux density thin film material with improved performance • Three different

• Desired performance: high efficiency and high power density

• Advances in power switches & controllers, GaN, SiC…

– frequency , inductance required

• Magnetic materials key challenge for advancing power conversion technology

Motivation

• Key Applications: o Power Factor Correctiono Flybacko Buck

• Complements advances in semiconductor technologies including wide-band gap devices.

3

Kolar et al*

Page 4: Thin Amorphous Core Material for Power Applications...•Objective is to replace ferrite cores with high flux density thin film material with improved performance • Three different

Ferrite Performance

Performance Bench Mark:

700 kW/m3 @500kHz & 100 mT Bpeak

4

Bsat = 420 mT at 100 OC

Page 5: Thin Amorphous Core Material for Power Applications...•Objective is to replace ferrite cores with high flux density thin film material with improved performance • Three different

Anomalous Loss–Inconsistencies in domain wall motion during magnetization reversal

–Variations in localized flux densities

– Model for estimating anomalous loss proposed by Bertotti (Book- Hysteresis in Magnetism)

Magnetic Core Losses

• Magnetic core losses can be broadly classified

Eddy Current Loss-Eddy currents resist change in applied magnetic field

-Skin depth, thickness at which the current density drops to (1/e) Js

- Skin depth, d = (2r /wmr)0.5

Where; r = resistivity, mr =

relative permeability, w =

angular frequency (=2pf )

-Eddy current loss (thickness less than one skin depth)

5

Page 6: Thin Amorphous Core Material for Power Applications...•Objective is to replace ferrite cores with high flux density thin film material with improved performance • Three different

• Commercial magnetic thin film alloys

o Very high permeability (typically > 20,000)

=> Lower skin depth & hence high eddy current losses at high frequencies

• Electrodeposited crystalline thin films

o Crystalline structure in plated NiFe alloys impedes domain wall motion

=> coercive fields (20-80 A/m)

• Electrodeposited amorphous thin films

o Elimination of magnetocrystalline anisotropy

=> low coercive fields (10-20 A/m)

• Nanocrystalline thin films

o Randomly oriented crystalline anisotropies cancelled by ferromagnetic exchange

o Opposite direction of magnetostriction of nanocrystalline and amorphous phases can => reduction or elimination of magnetostrictive anisotropy

=> Ultra-low coercivity (<2 A/m)

6

Review of Soft Magnetic thin films

Page 7: Thin Amorphous Core Material for Power Applications...•Objective is to replace ferrite cores with high flux density thin film material with improved performance • Three different

Review of Soft Magnetic thin films

Materials Research

polycrystalline

thin films (NiFe)

Research

Amorphous thin

films (CoP)

Nanocrystalline

thin films (Vitroperm,

MT, etc)

Thickness (um) 3~5 3~5 >16

Coercivity

(A/m)

20 ~ 80 10-20 <3

Resistivity

(W.m)

25 ~ 45 x10-8 >100 x10-8 ~110 x10-8

Saturation

Flux Density

(T)

0.8 ~ 1.5 0.8 ~ 1.2 1.2

Relative

Permeability

<1000 <1000 >15000

7

Page 8: Thin Amorphous Core Material for Power Applications...•Objective is to replace ferrite cores with high flux density thin film material with improved performance • Three different

• Objective is to replace ferrite cores with high flux density thin film material with improved performance

• Three different soft magnetic thin films evaluated:-o Electrodeposited research based thin films (NiFe, CoP etc)- thickness <5 µmo High permeability commercial thin film alloys (NiFe-Esong, Goodfellows)- thickness

<5 µmo Nanocrystalline thin films (Vacuumschmelze, Toshiba) - thickness >21 µm

• Wound core samples prepared with OD- 7.7 mm, ID- 7.5 mm along with 25 turn copper primary & secondary windings

• Performance of thin films compared to ferrite (3c90) at 100 kHz

Material performance comparison

Commercial magnetic thin films Amorphous and nanocrystaline tape-wound cores

8

Page 9: Thin Amorphous Core Material for Power Applications...•Objective is to replace ferrite cores with high flux density thin film material with improved performance • Three different

Magnetic Core loss measurement set-up

• Test samples characterized as toroidal transformers• Current sensor measures primary current (I)• Oscilloscope measures secondary voltage (U)• Power loss, P= U.I• Air-core contribution compensated for accurate core loss measurement

sample

9

Page 10: Thin Amorphous Core Material for Power Applications...•Objective is to replace ferrite cores with high flux density thin film material with improved performance • Three different

Material performance comparison

10

Page 11: Thin Amorphous Core Material for Power Applications...•Objective is to replace ferrite cores with high flux density thin film material with improved performance • Three different

Measurement & Discussion

• Thinned nanocrystalline films produced by post-processing of commercial ribbons• Assembled into wound core transformers • Similar air-core transformer for air core compensation

Magnetic Core

Air-Core3F35(datasheet)

Test Frequency 500 kHz

11

Page 12: Thin Amorphous Core Material for Power Applications...•Objective is to replace ferrite cores with high flux density thin film material with improved performance • Three different

Measurement Results

Eddy current

Anomalous

Hysteresis

Losses 21 µm 5 µm

Eddy Current 1650 kW/m3 93 kW/m3

Hysteresis 31 kW/m3 31 kW/m3

Anomalous 419 kW/m3 76 kW/m3

Total Loss-2100 kW/m3

Total Loss-200 kW/m3

80 %

46 %

12

Page 13: Thin Amorphous Core Material for Power Applications...•Objective is to replace ferrite cores with high flux density thin film material with improved performance • Three different

Potential Impact on magnetic components

• Lower loss density Higher efficiencySmaller core sizeShorter conductor lengthHigher power density

• Higher Bsat Greater design flexibility

40% reduction in device volume

25% reduction in magnetic loss

13

Page 14: Thin Amorphous Core Material for Power Applications...•Objective is to replace ferrite cores with high flux density thin film material with improved performance • Three different

14

LM22679 Evaluation Board

Wurth inductor:

o 4.8 μH

o DCR 12mΩ

Tyndall gapped inductor:

o 4.7 μH

o DCR 15 mΩ

Efficiency the same at currents > 2 A

Tyndall Inductor

Performance evaluation- Tyndall vs Wurth

• Input Voltage : 5 V

• Output Voltage: 3.3V

• Output Current Range: 1 A to 3.5 A

• Frequency of Operation: 500 kHz

82

84

86

88

90

92

94

96

0 1 2 3 4

Eff

icie

ncy (

%)

IOUT (A)

Wurth

Tyndall litz+solid wire

Page 15: Thin Amorphous Core Material for Power Applications...•Objective is to replace ferrite cores with high flux density thin film material with improved performance • Three different

Key Summary points

• Very low loss density achieved using post-processing of nanocrystalline thin film material:

o 200 kW/m3 @500kHz & 0.1T Bpeak vs. 700 kW/m3 for 3F35 o 30 kW/m3 @ 100kHz & 0.1T Bpeak vs. 70 kW/m3 for 3C90

• Demonstrated performance of post-processed nanocrystalline thin film material

o 40% reduction in device volumeo 25% reduction in magnetic loss

• Challenge still exist in relation to costs:-o Elemental costso Processing costs

15

Page 16: Thin Amorphous Core Material for Power Applications...•Objective is to replace ferrite cores with high flux density thin film material with improved performance • Three different

New Alloy Systems

Tyndall’s low loss material systems

VitroVAC alloy: Co67Fe4B11Si16Mo2

Alloy # 1: CoFe- alloy- New elemental mixture to minimize

cost & also achieve in-situ film thinning

Alloy # 2: CoFe- alloy- New CoFe alloy system to minimize

elemental costs

17

Page 17: Thin Amorphous Core Material for Power Applications...•Objective is to replace ferrite cores with high flux density thin film material with improved performance • Three different

New alloy systems

100 kHz

18

Page 18: Thin Amorphous Core Material for Power Applications...•Objective is to replace ferrite cores with high flux density thin film material with improved performance • Three different

Key Conclusions

• Ultra-low loss performance demonstrated with as-deposited magnetic thin film

• Further, loss performance improvements under investigation

o In-situ magnetic field induction

o Creep & Magnetic Anneal

• Alloy 1 shows a power loss density of 200 kW/m3 at 100 KHz (same as Vitrovac and ferrite); significant volume and cost reduction

• Tyndall looking for partners to develop this technology further

19

Page 19: Thin Amorphous Core Material for Power Applications...•Objective is to replace ferrite cores with high flux density thin film material with improved performance • Three different

Acknowledgements

- Dr. George Young, Dr. Trong Tue Vu - Icergi Ltd.

- Mr. Diarmuid Hogan, Mr. J J Wilkinson - Excelsys Ltd

- Mr. Hugh McCafferty - Nuvotem Talema Ltd

- Dr. Plamen Stamenov, Trinity College, Dublin

19

The authors would like to acknowledge European Union, Enterprise Ireland,

Icergi Ltd & Competence Center for Applied Nanotechnology (CCAN) in

funding this work.

The authors also acknowledge the support from Vacuumschmelze, Toshiba

and E-song in providing thin film material for this work.