semiconductors - fuji electric

Whole Number 209

Semiconductors

The key technology to power electronics,Fuji Electric,s power devices.

The innovative technologies of Fuji Electric,s

power devices lead to market needs. Our power devices will contribute to the miniaturization and lower power consumption of such devices as inverters, industrial robots, air conditioners and elevators.

Cover photo:Accompanying the developments

of the times, the attributes of lowerpower consumption, compact sizeand higher functionality are in-creasingly being required of elec-tronic devices, electrical machinery,automobiles and the like. Semicon-ductor technology is essential to theevolution of these products and it isnot an overstatement to say thatsemiconductors are the “bread ofindustry.”

Fuji Electric is concentrating itsenergies on semiconductors for in-dustrial, automotive, IT and powersupply fields, and is involved at allstages, from research and develop-ment to production and marketing,in the power semiconductors andpower ICs that support the powerelectronics industry.

The cover photo shows images ofan IGBT module, power MOSFETand power IC, superimposed upona background depicting the indus-trial, automotive, IT, and consumerelectronic appliance fields in whichthese devices are applied.

Semiconductors

CONTENTS

Fuji Electric’s Semiconductors: Current Status and Future Outlook 38

U-series IGBT Modules 43

U-series of IGBT-IPMs (600 V) 48

Development of a Next-generation IGBT Module 52using a New Insulating Substrate

Low IR Schottky Barrier Diode Series 57

Medium-voltage MOSFETs for PDP-use 61

PDP Scan Driver IC 65

Head Office : No.11-2, Osaki 1-chome, Shinagawa-ku, Tokyo 141-0032, Japan

http://www.fujielectric.co.jp/eng/company/tech/index.html

Vol. 51 No. 2 FUJI ELECTRIC REVIEW38

Hirokazu KanedaAkinori Matsuda

Fuji Electric’s Semiconductors:Current Status and Future Outlook

1. Introduction

The development of broadband networks and ubiq-uitous computing are major trends of the 21st centuryand the technologies for supplying and controllingenergy are important technologies for supporting thiscause. Power electronics will certainly be an impor-tant technology for supporting the society of the future.

Fuji Electric’s semiconductor business has definedits business domain and market segment focus basedon this power electronics technology, and has beendeveloping proprietary semiconductor technology andsupplying commercial products to expand this busi-ness. In order to support various applications, FujiElectric is expanding its product line of power electron-ics semiconductor devices, which include insulatedgate bipolar transistors (IGBTs), metal oxide semicon-ductor field effect transistors (MOSFETs), diodes,control integrated circuits (control ICs) and the like,with a wide range of voltage and power capabilities,from low-voltage, low-power devices suitable for bat-tery power sources to high-voltage, high-power devicescapable of handling more than several thousand voltsand several hundred amps.

This paper describes Fuji Electric’s semiconductorproducts and summarizes the technical trends of eachmajor application.

2. Industrial-use Semiconductors

2.1 Fuji Electric’s IGBTsIGBT devices can be said to support the foundation

of power electronics in industrial applications, whichare typically inverters. Since the latter half of the1980s, IGBTs have increasingly been used as powerelectronics devices due to their high-speed switchingcapability and higher voltage and current capacitythan a conventional bipolar transistor.

Fuji Electric began commercial production ofIGBTs in 1988 and since then has continued to developproprietary leading-edge technology and to supplyproducts to the market that realize low loss and high-speed switching performance. Figure 1 shows theIGBT product history and main device technologies for

1st generation through 5th generation IGBTs. The 1stthrough 3rd generation IGBTs used epitaxial wafers,and their performance was enhanced through opti-mized injection efficiency and lifetime control and theadoption of submicron processing technology. Begin-ning with 4th generation IGBTs, a major advance inthe design concept was implemented to realize adramatic reduction in loss by employing a non-punchthrough (NPT) structure that achieves higher trans-port efficiency without lifetime control, instead of theconventional punch through (PT) structure that hadbeen utilized thus far, and by implementing innovativeprocessing technology that uses a thin floating zone(FZ) wafer having a lower thickness of nearly 100 µm.

Additionally, the recent 5th generation IGBT real-izes both lower loss and higher speed switchingcapability by employing a field stop (FS) structure anda trench-gate structure to achieve a large increase insurface cell density.(1) Figures 2 and 3 show thechanges in device structures for Fuji Electric’s 600 Vand 1,200 V IGBTs.

2.2 Increased performance from IGBTsIGBTs are expected to continue to evolve as the

main power devices used in industrial applications. In

Fig.1 Roadmap of Fuji Electric’s IGBTs

1985

1st gener-ation

Next gener-ation

2nd generation (L, F-series)

3rd generation (J, N-series)

4th genera-tion (S-series)

5th generation (T, U-series)

1990 1995 2000 2005

PT-type(epitaxial wafer, lifetime control)

Thin wafer technology

Submicron technology

Trench technology

NPT-type (FZ wafer)

FS-type (FZ wafer)


response to requests for increased performance fromIGBTs, Fuji Electric is vigorously pursuing technicaldevelopment from the following three approaches. Thefirst approach involves the application of leading-edgeprocess technology. Fine surface structures, such as atrench gate structure, enable a dramatic improvementto be realized in the tradeoff relation between on-voltage and turn-off loss. Device characteristics will beimproved by promoting the integration of technologyacquired from IC-related research and the adoption ofsubmicron processing technology. The second ap-proach is to find breakthrough solutions to the problemof decreased resistance to short circuit loads caused bythe application of submicron processing technology andto the problem of noise that becomes more prominentat higher switching speeds. To realize solutions,rather than focusing solely on the chip itself, compre-hensive technology that involves the materials, wiringstructure and cooling structure of the IGBT module isneeded.(2), (3) With the ultimate goal of providing noise-

free operation without any failures, Fuji Electric aimsto develop easy-to-use devices and supply them to itscustomers. The third approach is the creation of powerdevices based on a new concept that will help revolu-tionize the next generation of power electronics tech-nology. For example, Fuji Electric is actively promot-ing the research and development of a reverse-blockingIGBT which has the potential to revolutionize powerconversion methods.

In this manner, Fuji Electric intends to continue tosupply devices that realize advanced power electronicscapabilities.

3. Automotive Semiconductors

3.1 Fuji Electric’s automotive semiconductorsThe use of electronics in automobiles is rapidly

increasing in order to achieve such goals as higherenergy efficiency, lower emissions and improved safetyand comfort, and consequently, the number of semicon-ductors installed in automobiles is steadily increasing.Because automotive semiconductors are used in asevere operating environment, and for safety reasons,high reliability is required. Additionally, the at-tributes of small size, lightweight, a small footprintand low cost are also strongly requested.

Fuji Electric supplies a variety of application-specific automotive semiconductor products. For en-gine system applications, Fuji Electric is commercializ-ing pressure sensors for manifold air pressure mea-surement and atmospheric pressure compensation,smart IGBTs and hybrid ICs for use in igniter circuits,and high-voltage diodes to prevent premature ignition.For applications involving the chassis system, Fujiprovides MOSFETs, smart MOSFETs and diodes fortransmission control, traction control, brake control,suspension control, power steering, etc. In applica-tions involving the body system, Fuji Electric’s MOSFETs and diodes are used for power window, powerlock, automatic mirror and windshield wiper controlcircuits.

3.2 Intelligent functions of automotive semiconductorsThe performance of automotive MOSFETs has

been improved through innovations in the devicestructure and advances in submicron processing tech-nology. Figure 4 shows the roadmap of Fuji Electric’sautomotive MOSFETs. The application of a trenchgate structure and submicron processing technology tothe low-voltage class of products, which are used inmany applications such as power steering and airconditioning circuits, and for which use is likely toincrease in the future, enables the realization of a lowon-resistance per unit area of 0.8 mΩ, which is approx-imately 40 % of the on-resistance of conventionalproducts. Moreover, the development of a quasi-planarjunction structure enables the high-voltage class ofproducts, used in the DC-DC converters and electronic

Fig.3 Changes in the 1,200 V IGBT chip cross-sectionalstructure

Fig.2 Changes in the 600 V IGBT chip cross-sectionalstructure

G E

C

G E G E G E

n-n+

p+ substrate

n+

n+ buffer

p

N-seriesC

n-n- n-

p+ substrate

n+ buffer

S-seriesSubmicron

CT-series

NPT structure

CU-seriesTrench

structure

G E

C

G E EG

n-n+

p+ substrate

n+

n+ buffer

p

N-series

n-n-

CS-series

NPT structure

CU-series

FS trenchstructure


ballasts of hybrid automobiles, to realize an on-resistance per unit area and switching time that areboth less than one-half the corresponding values of aconventional device.

Meanwhile, semiconductor devices are increasinglybeing requested to provide higher reliability due to theincreasing scope and complexity of the electroniccontrol unit in response to requests for even higher-level electronic control, and also due to the year-after-year increase in temperature of the operating environ-ment resulting from installation space constraints. Inresponse to this request for higher reliability, Fuji

Electric has developed smart MOS technology thatintegrates a conventional MOSFET and a protectioncircuit, a status monitoring and output circuit, and adrive circuit all onto a single chip, and has providedthe device with intelligent functionality.

In an intelligent device that includes an integratedvertical power MOSFET and control circuit, the isola-tion structure is very important. As technologycapable of realizing stable isolation at minimal cost,Fuji Electric has developed and is applying its propri-etary self-isolation complementary MOS/double-dif-fused MOS (CDMOS) technology. Figures 5 and 6 showtypical examples of this self-isolation structure. Withthe self-isolation structure, the output-stage powerMOSFET is isolated from other devices integrated onthe same silicon substrate, such as a low/high voltageMOSFET or a protection zener diode, by the pnjunction of each device. Compared to the dielectricisolation or junction isolation structures, this self-isolation structure is advantageous because it enablesthe easy integration of peripheral circuitry with anexisting MOSFET, at a dramatically lower cost.

Automotive semiconductors are expected to evolveinto even smaller sizes in the future. Fuji Electricintends to respond such market requests by activelydeveloping unified IC technology capable of realizing

Fig.5 Cross-sectional structure of self-isolation power ICs (control circuit)

Fig.4 Roadmap of Fuji Electric’s automotive MOSFETs

Fig.6 Cross-sectional structure of self-isolation power ICs(power MOSFET)

1985 1990 1995 2000 2005

1st generation(FAP-3A series)(FAP-2 series)

2nd generation(FAP-3B series)(FAP-2A series)

3rd generation(FAP-T1 series)

(SuperFAP-G series)

Planar DMOS Trench gateQuasi-planar junction Super junction

6 µm 4 µm 3 µm 1.5 µm 0.8 µm 0.5 µm 0.35 µm

3.5 mΩ · cm2

800 mΩ · nC2.3 mΩ · cm2

540 mΩ · nC1.4 mΩ · cm2

260 mΩ · nC

130 mΩ · cm2

20 Ω · nC125 mΩ · cm2

15 Ω · nC76 mΩ · cm2

5.5 Ω · nC24 mΩ · cm2

3 Ω · nC

0.8 mΩ · cm2

175 mΩ · nC0.65 mΩ · cm2

125 mΩ · nC0.5 mΩ · cm2

90 mΩ · nC

Product changes

Device technology

Design rule

Figure-of-merit 60 VRon ARon Qgd

600 VRon ARon Qgd

Low-voltage NMOS Low-voltage PMOS Medium-voltage NMOS Medium-voltage PMOS Zener diode

n+ n+ p+ p+ n+ n+ p+ p+ n+ p+

p well p well p well p well

p zenern well

p epitaxial

p substrate

n zenerp zener

n offset

n+p+ n+ n+

n well

n zenern zener

p well

n well

p epitaxial

p substrate

Output stage MOSFET Vertical powerzener diode


multi-channel integration with lateral MOSFETs,chip-on-chip technology capable of reducing the re-quired installation area, and chip size package (CSP)technology. Environmental responsiveness is increas-ing in importance and Fuji Electric intends to givepriority to the elimination of lead in these semiconduc-tor products.

4. Semiconductors for Information Devices andPower Supply Systems

4.1 Fuji Electric’s semiconductors for information devicesand power supply systemsWith advances in information technology, the

popularity of PCs, digital household appliances andportable information devices is spreading rapidly. Inorder to conserve natural resources and energy, lowerpower consumption is becoming increasingly impor-tant, and the power supply systems in which thesedevices are installed are strongly requested to providehigh efficiency and low loss, as well as have a smallfootprint, low height and light weight. As semiconduc-tors for power supply system applications, Fuji Electrichas supplied a product series of AC-AC and DC-DCpower supply control ICs, MOSFETs, Schottky barrierdiodes (SBDs), and low-loss fast recovery diodes(LLDs).

Fuji Electric’s ICs feature mixed analog-digitalpower technology that combines high-voltage powertechnology, high precision complementary metal oxidesemiconductor (CMOS) analog technology and CMOSdigital technology into a single chip. For relatively lowcapacity AC-DC conversion applications, Fuji Electricsupplies a power IC containing an integrated 700Vpower MOSFET, and this product can be used toconfigure a small-size AC adapter with a minimumnumber of parts. Moreover, for larger capacity AC-DCconversion applications, Fuji supplies all the semicon-ductor devices, including CMOS control ICs capable ofvoltage mode and current mode control, MOSFETs and

SBDs, required to configure a power supply system.For DC-DC conversion, there is a trend towards

using customized ICs for each portable device applica-tion. These devices often require different powersupply voltages for each system block, and multi-channel ICs that integrate multiple outputs on a singlechip are commonly used.

For information device applications, Fuji Electricsupplies LCD backlight control ICs and plasma displaypanel (PDP) driver ICs for flat panel display applica-tions. Future growth is anticipated for these flat paneldisplay applications.

4.2 Power IC technologyPortable information devices use batteries as their

power source. These devices are also equipped withimage displays and high-speed microprocessors, andtheir required supply voltage differs according to theirinternal circuitry. Accordingly, it is important for thepower supply of a portable information device to becapable of supplying different voltages at high efficien-cy, while realizing a small footprint and small volume.

Fuji Electric responded to this need early on bydeveloping a proprietary lateral CDMOS structurethat realizes mixed analog-digital power technologythat enables multiple power MOSFETs and CMOScircuits to be integrated in a single chip, and hassupplied various such products. Figure 7 shows FujiElectric’s roadmap for power IC technology. FujiElectric has supported the trends towards multi-channel, higher functionality devices by applying sub-micron technology and a proprietary lateral DMOSstructure to reduce the MOSFET on-resistance andenable large-scale control circuits to be integratedwithin a power IC chip.

As device technology that aims to realize evenlower on-resistance and higher switching performanceof power MOSFETs, Fuji Electric is researching anddeveloping technology for fabricating three-dimension-al devices on a silicon substrate and for achieving

Fig.7 Roadmap of Fuji Electric’s power IC technology

1985 1990 1995 2000 2005

Bipolar

Bi-CDMOS (epitaxial wafer)

CDMOS (non-epitaxial wafer)

Trench lateral MOS

4 µm 2 µm 1.5 µm 1 µm 0.6 µm 0.35 µm

300 mΩ · cm2 60 mΩ · cm2 20 mΩ · cm2 10 mΩ · cm2

Device technology

Design rule

Figure-of-merit 30 V Ron A


dramatically smaller DC-DC converters that integratepassive devices with a control IC (see Fig. 8). FujiElectric plans to develop commercial products in thenear future.(6), (7)

4.3 Reduction of the MOSFET lossIn a switching power supply for performing AC-DC

conversion, the turn-off loss and on-resistance loss ofthe power MOSFET account for the majority of thetotal loss. Therefore, in order for the switching powersupply to realize higher efficiency and lower loss, theturn-off loss and the on-resistance loss of the powerMOSFET must both be reduced simultaneously. More-over, in order to increase the switching frequency inresponse to requests for smaller size of the switchingpower supply, the reduction of switching loss, astypified by turn-off loss, is expected to become evenmore important in the future.

In response to these needs, Fuji Electric hasdeveloped and commercialized its propriety low-loss,high-speed power MOSFET technology as the Super-FAP-G series. The SuperFAP-G series uses quasi-planar junction technology to achieve a lower on-resistance that is within 10 % of the silicon limit.(5)

The SuperFAP-G series realizes high-voltage pow-er MOSFET performance that approaches the theoreti-cal performance limit, but Fuji Electric’s super junctionMOSFET (SJ-MOSFET) technology is attracting atten-tion as a breakthrough technology that, despite usingsilicon, achieves performance beyond the silicon limit.Fuji Electric is vigorously pursuing development ofthis SJ-MOSFET technology and plans to commercial-

ize this technology in the near future.

5. Conclusion

Fuji Electric is developing proprietary power de-vice and IC technology for power electronics applica-tions and is developing and supplying smart andintelligent products that provide solutions matched tothe customer. This paper has summarized FujiElectric’s efforts involving semiconductors and haspresented an overview of each application field. Foradditional details of the various relevant technologiesand products, please refer to the other papers in thisspecial issue.

In the society of the future, robotics and broadbandtechnology will be even more pervasive in our dailylives and accordingly, the importance of power elec-tronics will increase. Fuji Electric intends to continueto develop advanced, proprietary semiconductor tech-nology and to commercialize semiconductor devicesthat support power electronics. Additionally, basedupon our corporate motto of “ ,”Fuji Electric intends to continue to supply reliablequality products to its customers.

References(1) Otsuki, M. et al. Investigation on the Short-Circuit

Capability of 1200 V Trench Gate Field Stop IGBTs.Proceedings of ISPSD’02. 2002, p.281.

(2) Otsuki, M. 1200 V FS-IGBT Module with EnhancedDynamic Clamping Capability. Proceedings ofISPSD’04. 2004, p.339-342.

(3) Nishimura, T. et al. New Generation Metal Base FreeIGBT Module Structure with Low Thermal Resistance.Proceeding of ISPSD’04. 2004, p.374-350.

(4) Naito, T. et al. 1200 V Reverse Blocking IGBT withLow Loss for Matrix Converter. Proceedings ofISPSD’04. 2004, p.125-128.

(5) Kobayashi, T. et al. High-Voltage Power MOSFETsReached Almost to the Silicon Limit. Proceedings ofISPSD’01. 2001, p.435-438.

(6) Fujishima, N. A Low On-resistance Trench LateralPower MOSFET in 0.6 µm Smart Power Technology for20-30 V. International Electron Devices Meeting. 2002.

(7) Hayashi, Z. et al. High Efficiency DC-DC ConverterChip Size Module with Integrated Soft Ferrite. Pro-ceedings of 2003 IEEE International Conference Sym-posium. 2003.

Fig.8 Cross-sectional structure of trench lateral powerMOSFETs in power ICs

n+

n+

D S D

Device pitchGate polysilicon

Extended drain

p substrate

p base

n well

Oxide

Electrode


Shuji Miyashita

U-series IGBT Modules

1. Introduction

Power conversion equipment such as general-purpose inverters and uninterruptible power supplies(UPSs) is continuously challenged by demands forhigher efficiency, smaller size, lower cost and lowernoise. Accordingly, the power-converting elementsused in inverter circuits are also required to have highperformance, low cost and high reliability. At present,insulated gate bipolar transistors (IGBTs) are widelyused as the main type of power converting elementsbecause they exhibit low loss and enable the easyimplementation of drive circuitry. After commercializ-ing the IGBT in 1988, Fuji Electric has made efforts toimprove the IGBT further in pursuit of lower loss andhigher reliability.

This paper introduces the technology and productline-up of Fuji Electric’s fifth generation IGBT modules(U-series), which feature a large improvement inelectrical characteristics compared to the fourth gener-ation IGBTs (S-series).

2. Features of the New IGBTs

2.1 Trench gate IGBTFuji Electric is producing trench-gate type power

metal oxide semiconductor field effect transistors(MOSFETs), to which design and process technologieshave been applied in order to ensure sufficient reliabil-ity. The trench IGBT is the result of applying thesetechnologies to IGBTs.

Figure 1 shows a comparison of the planer andtrench IGBT cell structures. The trench IGBTachieves a drastic increase in cell density, enabling thevoltage drop at the channel part to be suppressed to aminimum. Since the distinctive JFET region, sand-wiched between channels of the planer type device,does not exist in the trench IGBT, the voltage dropacross this region can be completely eliminated. Onthe other hand, the high channel density of the trenchIGBT causes the problem of low capability to with-stand a short-circuit condition. However, the trenchgate structure developed by Fuji Electric optimizes thetotal channel length of the MOS device to realize high

short-circuit withstand capability without sacrificingthe saturation voltage.

2.2 NPT-IGBTThe unit cell structures of a non-punch through

(NPT) IGBT and a punch through (PT) IGBT are

Fig.2 Comparison of PT and NPT-IGBT cell structures

Fig.1 Comparison of planer and trench IGBT cell structures

V-pn V-pn

R-ch

R-acc

R-JFETR-drift

R-ch

R-acc

R-drift

n+

n+

pp

p+ p+

n-

n-

C C(b) Trench IGBT(a) Planer IGBT

EG EG

(b) NPT-IGBT(a) PT-IGBT

100 µm

350 µm

EG

n+

n+ Buffer

p

p+

p+

n-

C

n+

EG

n+p

n-

C

Epitaxial Si wafer was used.

FZ-Siwafer was used.

n+


shown in Fig. 2. Features of the NPT-IGBT are asfollows.(1) Since injection from the collector-side can be

suppressed, lifetime control is unnecessary andthe switching loss does not increase even at hightemperatures.

(2) Because the temperature dependence of outputcharacteristics is positive (the saturation voltageincreases at higher temperature), these devicesare well suited for parallel applications.

(3) Withstand capability, including the load short-circuit withstand capacity, is higher than that of aPT-IGBT.

(4) Use of a floating zone (FZ) wafer achieves bettercost performance and higher reliability owing toits low rate of crystal defects.

Also, it is important for NPT-IGBTs to suppresssaturation voltage while maintaining the collector-emitter (C-E) blocking voltage. The saturation voltagewill be lower when thinner wafers are used, however,the thickness of the depletion layer end must bemaintained sufficiently thick so there will be no punchthrough even when the maximum rated C-E blockingvoltage is applied, and this sufficient thickness is theminimum value. Therefore, the optimal thickness isthinner for devices having lower C-E blocking voltage,making their manufacture even more difficult. For600 V NPT-IGBTs, Fuji Electric has established massproduction technology to handle the optimal wafer

thickness by carefully choosing the optimal waferspecification and improving the precision of back-grinding process technology so that the reduction ofsaturation voltage could be achieved.

Thinner wafers also feature a reduction in turn-offswitching loss. Figure 3 shows a comparison of turn-offwaveforms. In the PT-IGBT, which has more carriersinjected from the collector side, lifetime control isimplemented to promote the recombination of carriersat the time of turn-off. However, this effect decreasesas the temperature increases, and therefore the turn-off switching loss tends to increase due to an increasein fall time. For the NPT-IGBT, on the other hand,lifetime control is not implemented and therefore thistemperature dependence does not exist and there is nochange in the turn-off waveform and no increase inturn-off switching loss. Accordingly, the trade-offrelationship between saturation voltage and turn-offswitching loss has been improved, and both can bereduced simultaneously with the use of an NPT-IGBT.

When the load is short-circuited, the NPT-IGBT,having a thick n- drift layer, can support the voltagewith its wide n- drift layer, thereby suppressing thetemperature rise which causes breakdown and achiev-ing high short-circuit withstand capability. Even athin-wafer 600 V NPT-IGBT can achieve a short-circuitwithstand capability of 22 µs.

2.3 Field stop (FS) structureFigure 4 compares the cross sections of NPT-IGBT

and FS-IGBT unit cells. The NPT-IGBT requires athick drift layer so that the depletion layer does notcontact the collector side during turn-off. The FS-IGBT does not, however, require as thick a drift layeras the NPT type because it is provided with a field stoplayer to block the depletion layer, and accordingly,saturation voltage can be lowered for the FS-IGBT.Furthermore, the FS-IGBT has fewer excess carriersbecause of its thinner drift layer. Moreover, the FS-IGBT can achieve reduced turn-off switching lossbecause the remaining width of its neutral region is

Fig.3 Comparison of turn-off waveforms

Fig.4 Comparison of NPT and FS-IGBT cell structures

600 V/50 Adevice

VDC = 300 V

VCE : 100 V/div

VCE : 100 V/div

Ic : 25 A/div

Ic : 25 A/div

VGE = ±15 V

RG = 51 Ω

Ic = 50 A

Tj = 125°C

VDC = 300 V

VGE = ±15 V

RG = 51 Ω

Ic = 50 A

Tj = 125°C

t : 200 ns/div

0

0

(a) PT-IGBT

600 V/50 Adevice

t : 200 ns/div

(b) NPT-IGBT

n+

p

p+

n-

C(a) NPT-IGBT

p+

n+

p

n-

C(b) FS-IGBT

n+ n+

FS layer

EG EG


small when its depletion layer is completely extended.Thus, by applying the FS structure to 1,200 V and1,700 V devices, their trade-off relationship betweensaturation voltage and turn-off switching loss has beenimproved, and both can be reduced simultaneously.

3. Features of Fuji’s New FWD

As IGBTs are improved, free wheeling diodes(FWDs), which are packaged together with IGBTs intoIGBT modules are then installed in inverter circuits,are also subject to improvement. The FWDs arerequired to have lower conduction loss, which is causedby forward voltage, and lower reverse recovery loss.Also, soft reverse recovery of FWDs, which correlatesto the faster turn-on switching of IGBTs, is anespecially important characteristic in order to suppressa rise in surge voltage, protect the IGBT from damage,and to suppress malfunction of peripheral circuitry.By optimizing the wafer specifications, applying injec-tion control from the anode at the chip’s front structureand implementing optimal lifetime control, Fuji Elec-tric has developed a new FWD having superior soft

reverse recovery characteristics.Figure 5 compares the cross sectional structures of

a conventional and a new FWD. The new FWD has astructure that is able to suppress carrier injection.Peak current during reverse recovery is reduced, andthe new FWD achieves not only a softer reverserecovery characteristic but also less reverse recoveryloss. Figure 6 shows an example of the trade-offrelationship between forward voltage and reverserecovery loss. The new FWD shows the better result ofimproved reverse recovery loss compared to that of theconventional FWD. On the other hand, the forwardvoltage was designed to be approximately 1.6 V at ahigher temperature (Tj = 125°C). Therefore, the con-duction loss in an inverter circuit application would belower due to the lower forward voltage of the newFWD. Furthermore, since the output characteristic ofthe new FWD has a positive temperature coefficient,similar to the U-series IGBT, the current imbalanceamong parallel-connected devices will be smaller.Therefore, in a parallel connection application, for

Fig.5 Comparison of conventional and new FWD cell struc-tures

Fig.7 Trade-off relationship for 1,200 V devices

Fig.8 Examples of packages for U-series IGBT modules

Fig.6 Trade-off between forward voltage and reverse recoveryloss

p

n-

Cathode

(a) Conventional FWD

n+

p pp

n-

Cathode

(b) New FWD

n+

Anode Anode

1.4 1.6 1.8 2.0 2.2 2.4

15

10

5

0

Rev

erse

rec

over

y lo

ss (

mJ/

puls

e) 1,200 V/150 A FWDs

Conventional

Tj = 125°C

Forward voltage (V)

New

12

10

8

6

4

2

01.4 1.6 1.8 2.0 2.2 2.4

Saturation voltage [Tj =125°C] (V)2.6 2.8 3.0 3.2 3.4

1,200 V/50 A device

Tu

rn-o

ff s

wit

chin

g lo

ss :

Eof

f (m

J/pu

lse) Tj = 125°C

VDC= 600 VIc = 50 ARG = recommendedVGE = ±15 V

P-series

N-series

4th generationS-series

5th generationU-series/FS trench


Table 1 U-series IGBT modules line-up

Fig.9 Module package (6 in 1) with shunt-resistersexample, high power inverter circuits will be easier touse.

4. Introduction of Fuji Electric’s Product Line-up

Fuji Electric has combined the above IGBT andFWD technology while continuously employing thesame packaging technology as used in fourth genera-tion IGBT modules which have higher power cyclingcapability, and has finally completed the developmentand product line-up of U-series IGBT modules thatexhibit much improved characteristics in comparisonto fourth generation S-series IGBT modules. Figure 7shows an example of the trade-off relationship for a1,200 V device. Both the saturation voltage and theturn-off switching loss are simultaneously reduced. Itcan be seen that the trade-off relationship is dramati-cally improved in comparison to that of fourth genera-tion IGBT modules, and that almost 20 % less powerdissipation loss can be expected in the case of aninverter circuit application. Figure 8 shows examplesof packages for these U-series IGBT modules, Table 1lists Fuji Electric’s line-up of U-series IGBT modules.Three blocking voltage ratings of 600 V, 1,200 V and1,700 V, a wide current range from 10 A up to 3,600 A,and many package variations have been prepared toenable application to various types of power conversionequipment. Also, new module packages with shunt-resisters have been developed as an addition to the U-

series product line-up. A schematic drawing of thepackage and its equivalent circuit are shown in Fig. 9.The modules are designed for vector control inverters,

General purpose Inverter

600 V

Small

1,200 V

1,700 V

Small

Package3,600 A2,400 A1,600 A1,200 A10 A 15 A 20 A 30 A 50 A

(5.5 kW)75 A 100 A

(11 kW)150 A 200 A

(22 kW)300 A 400 A

(40 kW)600 A 800 A

3,600 A2,400 A1,600 A1,200 A10 A 15 A 25 A(5.5 kW)

35 A 50 A(11 kW)

75 A 100 A(22 kW)

150 A 200 A(40 kW)

300 A400 A450 A

(75 kW)600 A 800 A

VCES

rating

PIM

PIM/6 in 1

6 in 1

2 in 1

2 in 1/1in1

IC rating(Inverter rating)

PackageVCES

ratingIC rating(Inverter rating)

PIM 6 in 1

PIM 6 in 1

EP2 EP3

PIM EP2EP3

6 in 1

6 in 12 in 11 in 1

NewPC2NewPC3

M232

NewPC3(with Shunt R)

For vector control

EconoPack-Plus* (6 in 1)

EconoPack-Plus (6 in 1)M248

M233 M247M234 M142 M143

M142 M143

M249M235 M127

M138

NewPC2

NewPC3

M232 M233 M247

High performance Inverter *EconoPack-Plus : A registered trademark of Eupec GmbH, Warstein

14 1

5

33 3

2

25 2427 2630 21 20

215

6U

29,30

22

78

R2

271

2

28

34

R1

179

10

18

W

1112

R3

15,16

13,14

31,32

33,34

23,24 19,20V

3.81

20.5

17

62 50

122

110

1 2 3 4 5 6 7 8 9 10 11


and the shunt resisters and their voltage drop detec-tion terminals are installed at the AC output terminalsof a three-phase inverter bridge. In contrast to theconventional current detection method that uses anexternal current detector, a voltage detection methodthat uses these modules will be able to control themotor output current, thereby enabling inverter equip-ment to be made simpler and smaller.

5. Conclusion

IGBT and FWD technology of Fuji Electric’s U-series IGBT modules, its features and product line-uphave been presented. Through using the latestsemiconductor technology and packaging technology,

these products achieve superior low loss characteris-tics, and we believe they will make important contribu-tions to the realization of smaller size and lower lossinverter circuit equipment.

Fuji Electric intends to continue working to im-prove this technology further, with the goals of realiz-ing higher performance and higher reliability devices,and to contribute to the development of power electron-ics.

References(1) Laska, T. et al. The Field Stop IGBT (FS IGBT) A New

Power Device Concept with a Great ImprovementPotential. Proc. 12th ISPSD. 2000, p.355-358.


Kiyoshi SekigawaHiroshi EndoHiroki Wakimoto

U-series of IGBT-IPMs (600 V)

1. Introduction

Intelligent power modules (IPMs) are intelligentpower devices that incorporate drive circuits, protec-tion circuits or other functionality into a modularconfiguration. IPMs are widely used in motor driving(general purpose inverter, servo, air conditioning,elevator, etc.) and power supply (UPS, PV, etc.)applications.

The equipment that uses these IPMs are requiredto have small size, high efficiency, low noise, longservice life and high reliability.

In response to these requirements, in 1997, FujiElectric developed the industry’s first internal over-heat protection function for insulated gate bipolartransistors (IGBTs) and developed an R-IPM seriesthat achieved high reliability by employing an all-silicon construction that enabled a reduction in thenumber of components used.

Then in 2002, Fuji Electric changed the structure ofits IGBT chips from the punch through (PT) structure,which had been in use previously, to a non-punchthrough (NPT) structure, for which lifetime control isunnecessary, in order to realize lower turn-off loss athigh temperature, and also established finer planar gate

and thin wafer processing technology to develop an R-IPM3 series that realizes low conduction loss.

With the goal of reducing loss even further, FujiElectric has developed an IGBT device that employs atrench NPT structure to realize lower conduction lossand has developed a new free wheeling diode (FWD)structure to improve the tradeoff between switchingnoise and loss. Both of these technologies are incorpo-rated into Fuji Electric’s newly developed U-seriesIGBT-IPM (U-IPM) which is introduced below.

2. U-IPM Development Concepts and ProductLine-up

The concepts behind the development of the U-IPMare listed below.(1) Realization of lower loss

Lower loss can be realized by developing newpower elements and optimizing the drive performance.Increasing the carrier frequency of the equipmentcontributes to improved control performance. Also,larger output can be obtained from the equipmentduring the operation at the same carrier frequency.(2) Continued use of the same package as prior

products

Table 1 Product line-up, characteristics and internal functions of the U-IPM series

No. ofelements

Inverter part Brake part Internal function

Packagetype

Both upper and lower arms Upper arm Lower armVDC

(V)VCES(V)

4506 in 1

7 in 1

600

450 600

IC(A)

PC(W)

IC(A)

PC(W)

Dr: IGBT driving circuit, UV : Under voltage lockout for control circuit, TjOH: Device overheat protection, OC: Over-current protection, ALM: Alarm output, TcOH: Case temperature over-heat protection*6MBP20RUA060 uses a shunt resistance-based over-current detection method at the N line.

Model

Dr UV OC ALM OC ALMTjOH TcOH

6MBP 20RUA060 20 84 – – Yes Yes Yes None None None P619Yes Yes

6MBP 50RUA060 50 176 – – Yes Yes Yes Yes None Yes P610Yes Yes

6MBP 80RUA060 80 283 – – Yes Yes Yes Yes None Yes P610Yes Yes

6MBP100RUA060 100 360 – – Yes Yes Yes Yes None Yes P611Yes Yes

6MBP160RUA060 160 431 – – Yes Yes Yes Yes None Yes P611Yes Yes

7MBP 50RUA060 50 176 30 120 Yes Yes Yes Yes None Yes P610Yes Yes

7MBP 80RUA060 80 283 50 176 Yes Yes Yes Yes None Yes P610Yes Yes

7MBP100RUA060 100 360 50 176 Yes Yes Yes Yes None Yes P611Yes Yes

7MBP160RUA060 160 431 50 176 Yes Yes Yes Yes None Yes P611Yes Yes


The continued use of the same package as withprior products makes it possible to improve equipmentperformance by replacing the IPM without having tomodify the design of the equipment.

Table 1 lists the product line-up, characteristicsand internal functions of Fuji Electric’s 600 V U-IPMseries. The U-IPM series maintains internal functionsand a package size that are interchangeable with theR-IPM series; its rated current is 20 to 160 A for the “6in 1” pack and 50 to 160 A for the “7 in 1” pack(containing an internal IGBT for braking use).Figure 1 shows an external view of the packages.

3. Characteristics of the Power Devices

A fifth-generation U-series IGBT (U-IGBT) is usedas the power device. This U-IGBT combines trenchgate technology with a basic design comprising FujiElectric’s floating zone (FZ) wafer technology, thinwafer processing technology, carrier injection controltechnology, and transportation factor improving tech-nology.

Figure 2 compares the structures of the convention-al planar IGBT and the trench IGBT. The adoption of

Fig.1 External view of U-IPM packages

a trench gate structure results in a smaller voltagedrop at the channel (R-ch) due to increased surface celldensity and results in a lower saturation voltage due tothe smaller voltage drop resulting from the eliminationof the planar device’s characteristic JFET region (R-JFET). Moreover, short circuit immunity capability isrealized through optimization of the design of thesurface structure. Figure 3 illustrates the changes thathave occurred in the cross-sectional IGBT structure inthe transition from the conventional IGBT to the U-IGBT, and Table 2 compares their applied technologies.

The FWD, in accordance with the U-IGBT, incorpo-rates a new design featuring optimized wafer specifica-tion, control of anode-side injection and optimal life-time control technology to realize the characteristics oflow peak current during reverse recovery operation,low generated loss, and soft recovery.

4. U-IPM Loss

4.1 Comparison of total lossThe marketplace requires that new IPM products

achieve lower levels of loss. (1) Increased carrierfrequency to enhance controllability and (2) largeroutput current at the same carrier frequency arenecessary for the achievement of the goal. The loss

Table 2 Changes in IGBT technology

Fig.2 Comparison of planar IGBT and trench IGBT chip crosssections

Fig.3 Change in cross-sectional structure of 600 V IGBT chip

P619

P610

P611

n+source

p-channel

Emitterelectrode

(a) Planar IGBT (b) Trench IGBT

Insulationlayer

Gate electrode

Gate oxidelayer

n- siliconsubstrate

n+source

p-channel

R-ch

R-accR- J

FE

TR

- dri

ft

R- d

rift

R- a

cc R- c

h

V- p

n

V- p

n

p+ layer

Collectorelectrode

IGBT technologyR-IPM R-IPM3 U-IPM

N-IGBT T-IGBT U-IGBT

EpitaxialWafer FZ

350 µmWafer thickness 100 µm

PTStructure NPT

PlanarGate structure Trench

YesLifetime control None

HighCarrier injection Low

LowTransportation factor High

n+ buffer

n-pn+ n+

p+ substrate

n- n-

p p

U-IPMTrench NPT

structure

R-IPM3Planar NPT

structurethinner surface

R-IPMPlanar PT structure

C

G EG EG E

C

C


Fig.4 Comparison of total loss (at same current) for the U-IPM, R-IPM3 and R-IPM series

Fig.5 Current vs. total loss (at same frequency) for U-IPM, R-IPM3 and R-IPM

generated by existing models and by the U-IPM isdescribed below.

Figure 4 compares the loss of the U-IPM and theexisting R-IPM and R-IPM3 devices in the case ofoperation at carrier frequencies of 4, 8 and 16 kHz, anda current of 50 Arms (1/3 of the rated current). As canbe seen in the figure, the newly developed U-IPMrealizes a total loss that is approximately 22 to 28 %lower than that of the R-IPM and approximately 11 to12 % lower than that of the R-IPM3. In particular, itcan be seen that the loss generated when using the U-IPM at a carrier frequency of 8 kHz is less than the

loss generated by a R-IPM operating at a carrierfrequency of 4 kHz, and therefore, the carrier frequen-cy can be increased from 4 kHz to 8 kHz by replacing aR-IPM with a U-IPM of the same size package.Moreover, according to Fig. 5 which shows the relation-ship between current and total loss at fc = 4 kHz, togenerate the same amount of loss (50 W) as the R-IPM,the output current of the U-IPM can be increased by24.5 % compared to that of the R-IPM, or increased by13.7 % compared to that of the R-IPM3.

These techniques for reducing loss were focused onreducing the conduction loss, which accounts for morethan 50 % of the total loss, and on reducing the turn-on

Fig.6 IC-VCE characteristics for U-IPM, R-IPM3 and R-IPM

Tot

al lo

ss (

W)

20

0

40

60

80

100

R-IPM : 6MBP150RA060R-IPM3 : 6MBP150RTB060U-IPM : 6MBP160RUA060

Tj = 125°C, Ed = 300 VVCC = 15 V, Io = 50 ArmsPower factor = 0.85, =1λ

Prr

Pf

Poff

Pon

Psat

R-IPM

45.50

26.9

6.13

7.01

4.121.34

R-IPM3fc = 4 kHz

40.55

26.3

5.76

3.033.901.59

U-IPM

22.2

4.213.853.891.22

R-IPM

59.80

26.7

12.3

14.0

4.122.68

R-IPM3fc = 8 kHz

50.90

26.2

11.6

6.04

3.903.18

U-IPM

44.57

22.1

8.46

7.68

3.892.44

R-IPM

88.79

26.7

24.7

27.9

4.12

5.37

R-IPM3fc = 16 kHz

71.67

26.1

23.2

12.1

3.89

6.38

U-IPM

63.08

22.0

17.0

15.3

3.89

4.89

35.37

100

150

50

0

Tot

al lo

ss (

W)

120100806040200

R-IPMR-IPM3

U-IPM

53 A58 A66 A

Io (Arms)

R-IPM : 6MBP150RA060R-IPM3 : 6MBP150RTB060U-IPM : 6MBP160RUA060

Tj = 125°C, Ed = 300 Vfc = 4 kHz, Vcc = 15 VPower factor = 0.85, =1λ

VCE (sat) (V)

I c (

A)

1.5 2 2.5 3 3.510.50

50

0

100

150

R-IPM3

R-IPM

U-IPM

Tj = 125°C, VCC = 15 VVCE (sat) at IPM pin


the emission noise.(1) Application of the new soft recovery FWD sup-

presses dv/dt.(2) The capacitance between the gate and emitter is

optimized in order to reduce di /dt, which in-creased as a result of the lower gate resistance,without reducing dv/dt

Through application of the above techniques, evenif currents of all sizes are controlled with the samegate resistance, emission noise will be maintained atthe same level as that of the R-IPM3 as shown inFig. 8, and lower loss can be realized. Accordingly, thetotal loss generated in all these products is linearlyproportional to the current, and the total loss andtemperature rise that occur during actual use caneasily be estimated.

5. Conclusion

Fuji Electric’s 600 V U-IPM that uses a U-seriesIGBT chip having a trench NPT structure has beendescribed above. This U-IPM provides suitable perfor-mance to satisfy the marketplace in which lower loss isrequired. In the future, Fuji Electric intends tocontinue to develop new IPMs that will satisfy marketrequirements.

Fig.7 Characteristics of turn-on waveform and emission noise

Table 3 Characteristics of gate resistance and turn-onwaveform

Fig.8 Emission noise

loss, which accounts for a large percentage of theswitching loss of the R-IPM3. Each type of lossreduction is described below.

4.2 Reduction of conduction lossFigure 6 shows IC-VCE(sat) characteristics for U-

IPM, R-IPM3 and R-IPM devices. It can be seen thatwhen IC = 150 A, the VCE(sat) of the U-IPM is 0.45 Vless than that of the R-IPM and 0.55 V less than thatof the I-RPM3. This is the VCE(sat) reduction effect dueto the trench IGBT described in chapter 3.

4.3 Turn-on loss and emission noiseFigure 7 shows a schematic drawing of the current

(IC) and voltage (VCE) at the time when the device isturned on. As can be seen in the figure, typically, losscan be reduced by making dv/dt larger and emissionnoise can be reduced by making di /dt smaller. Howev-er, in the case where turn-on operation is controlled bythe typical method of gate resistance only, there is atradeoff relation as shown in Table 3, and it is difficultto establish both high dv/dt and low di/dt simulta-neously.

In the newly developed U-IPM, the following twotechniques suppress the emission noise that usuallyincreases when gate resistance is decreased and di/dtis increased, thereby enabling di /dt to be increased andturn-on loss to be decreased without any increase in

Low noise

Low loss

VGE

VCE

IC

t1 t2

t1 : Time from IC = 0 until IC = ICP

t2 : Time from IC = ICP until VCE = 0

di /dt is small and t1 is long low emission noisedv/dt is large and t2 is short low loss

Gate resistance RG

High Low Low Increases Decreases

Low High High Decreases Increases

Turn-ondi /dt

Turn-ondv/dt Loss Emission

noise

Measurement conditions: Distance between servo amplifier and antenna is 2 m, vertical direction, standby state

Em

issi

on n

oise

leve

l (dB

µV/m

)

Frequency (MHz)(a) R-IPM3 (150RTB)

(b) U-IPM (160RUA)Frequency (MHz)

8030

50

100

Em

issi

on n

oise

leve

l (dB

µV/m

)

80

130

13030

50

100


Yoshitaka NishimuraEiji MochizukiYoshikazu Takahashi

Development of a Next-generationIGBT Module using a New Insulating Substrate

1. Introduction

In response to the recent demand for energy-efficient electronic appliances, insulated gate bipolartransistor (IGBT) modules are being utilized in a widerscope of applications, ranging from conventional indus-trial applications to home-use electronic applianceapplications and the like. There is strong demand forlow-capacity IGBT modules, which are used mainly forthese home-use electric appliances, to be low cost,lightweight and have a compact size.

In response to this demand, Fuji Electric hasdeveloped and has begun producing its “Small Pack”series of IGBT modules. This series utilizes a heat-dissipating base-free structure (which does not containa heat-dissipating metal base) in order to realize low-cost and lightweight IGBT modules.

This paper introduces a heat-dissipating base-freestructure that utilizes a new insulating substrate toachieve an additional 30 % decrease in thermal resis-tance.

2. Design Concept (1), (2)

Heat-dissipating base-free structures, which typi-cally have larger thermal resistance than heat-dissi-pating metal-base-equipped structures, have been diffi-cult to use in industrial applications and other suchapplications where the usage conditions are severe.Heat-dissipating base-free structures are presentlyused only in low power applications and, unless a newtechnological approach is employed, their applicationto medium and large capacity applications is seen as

unlikely.Figure 1 shows a cross-sectional view of typical

IGBTs. In the heat-dissipating metal-base-equippedstructure of Fig. 1(a), the DCB substrate (substratewith ceramic insulation) is soldered to a heat-dissipat-ing metal base, and a cooling fin is attached to thebase. In the heat-dissipating base-free structure ofFig. 1(b), a cooling fin is attached directly to the DCBsubstrate.

In order to transfer the heat efficiently to a coolingfin, a thin thermal compound must be used to fill anygaps between the fin and module. In actuality, it isimportant that a screw be tightened in order tomaintain the pressure between the fin and modulesurfaces, and in order to prevent damage to the modulewhile the screw is tightened, the module must main-tain its mechanical strength. In the conventional heat-dissipating base-free structure, the alumina ceramicportion of the DCB substrate is made thicker in orderto maintain the mechanical strength. As a result,however, the thermal resistance becomes larger thanin the case of a heat-dissipating metal-base-equippedstructure, and this is a factor which limits applicationsfor heat-dissipating base-free structures.

Figure 2 shows Fuji Electric’s “6 in 1” IGBT mod-ule. In terms of the power per unit area of an IGBTmodule, the heat-dissipating metal-base-equippedstructure of Fig. 2(a) has a capacity of 4.7 W (max)/

Fig.2 Fuji’s 6-pack module

Fig.1 Cross section of an IGBT module

IGBT chip

Cooling fin Cooling fin

DCB substrate

DCB substrate

Heat-dissipating metal base

(a) Structure with heat-dissipating metal base

(b) Heat-dissipating base-free structure

IGBT chip

PC2

(a) Heat-dissipating metal-base-equipped structure

22 kW (75 A 1,200 V) Area : 107 × 44 (mm)

22 kW/4,708 mm2=4.7 W (max)/mm2

Small Pack

(b) Heat-dissipating base-free structure

(Low capacity, low cost)7.5 kW (35 A 1,200 V) Area : 63×34 (mm)7.5 kW/2,142 mm2=3.5 W (max)/mm2

Development of a Next-generation IGBT Module using a New Insulating Substrate 53

mm2, while the heat-dissipating base-free structure ofFig. 2(b) has a capacity of 3.5 W (max)/mm2.

Given these circumstances, we investigated andthen developed a DCB substrate having low thermalresistance and high strength in order to increase theusable power per unit area of an IGBT module whilerealizing lighter weight, smaller size and lower cost.

3. Design of a New Alumina Insulating Substrate

3.1 A comparison of characteristics according to IGBTmodule structureTable 1 lists the characteristics of the conventional

heat-dissipating base-free structure and of the heat-dissipating metal-base-equipped structure. The heat-dissipating base-free structure uses a 0.635 mm-thickalumina ceramic layer in order to maintain mechanicalstrength of the IGBT module. On the other hand, theheat-dissipating metal-base-equipped structure hassufficient mechanical strength and the thickness of itsalumina ceramic layer can be reduced to 0.32 mm.

Due to this difference, the thermal resistance ofthe heat-dissipating base-free structure is 1.6 timesthat of the heat-dissipating metal-base-equipped struc-ture.

3.2 Factors that inhibit thermal conduction and measuresfor improvementFigure 3 shows cross sections of IGBT modules

having a heat-dissipating base-free structure. Heatgenerated at the junction in an IGBT chip passesthrough the DCB substrate and is transferred to aheat-dissipating fin. Because the thermal conductivityof the alumina ceramic of the insulated portion of theDCB substrate is 20 W/(m·K) and the thermal conduc-tivity the copper used for the electronic circuitry is 390W/(m·K), the alumina ceramic layer of the DCBsubstrate acts as a thermal resistance layer throughwhich heat generated from the IGBT chip has difficul-ty in passing through.

In order to make it easier for heat to pass throughthis layer, it is efficient to make the thermal resistancelayer smaller and to decrease the heat flow (heatdensity) per unit area. Specifically, the following twocountermeasures are proposed.(1) Decrease the thickness of the alumina ceramic

layer(2) Increase the thickness of copper foil in order to

disperse the heat and decrease the heat flow perunit area of the alumina ceramic layer

Accordingly, reducing the thermal resistance of theDCB substrate is an efficient way to lower thetemperature of the IGBT chip. Moreover, by increas-ing the thickness of the copper foil, an improvement inthe mechanical strength of the DCB substrate itselfcan be expected.

3.3 FEM analysis resultsNext, as a first step to verify the effectiveness of

the above, we performed a steady-state thermal analy-sis using the finite element method (FEM). Theanalysis was performed under the conditions of a DC80 A current applied to a 3-dimensional half-scalemodel of a 9.25 mm-square IGBT chip, a 33 × 30 (mm)DCB ceramic substrate, and a 25 × 17 (mm) copper foilon top of the DCB surface. In the steady-stateanalysis, we varied the thickness of the aluminaceramic layer and the thickness of the copper foil toanalyze their effect on IGBT chip temperature. The

Fig.3 Cross-section of DCB structures and thermal character-istics

Table 1 Comparison between the metal-base free structureand the structure having a metal base

Fig.4 Thermal simulation results for each structure (DC 80 Asteady state condition)

DCB substrate characteristics

Conventional heat-dissipating base-free structure

Alumina thickness 0.635 mm 0.32 mm

Bending strength 108 N 53 N

Thickness of metal base – 3.0 mm

Thermal resistance Rth(j-c) 100 62

Heat-dissipating metal-base-equipped structure

IGBT chip

IGBT chip

Thick

Thick

Thin

Cooling finThermal grease

Copper 390 W/(m·K)

Alumina 20 W/(m·K)

(a) Conventional heat-dissipating base-free structure

(b) New structure

t : 0.25 mm

t : 0.635 mm

t : 0.25 mm

Heatconduction

IGBT chip

q Make ceramic thinner

Conventional heat-dissipating base-free

structure

Alumina thickness:0.32 mm

Copper foil thickness:0.6 mm

Alumina thickness:Copper foil thickness:

0.635 mm0.25 mm

0.32 mm0.25 mm

0.32 mm0.6 mm

Tj = 181°C Tj = 158°C Tj = 126°C

∆ 23 deg ∆ 32 deg

w Make copper foil thicker

Top copper foil


results are shown in Fig. 4.From the results of this analysis, it can be seen

that by reducing the thickness of the aluminumceramic layer from 0.635 mm to 0.32 mm, the IGBTchip temperature decreases by 23°C. Moreover, whilemaintaining the thickness of the alumina ceramiclayer at 0.32 mm and increasing the thickness of thecopper foil from 0.25 mm to 0.6 mm, it can be seen thatthe chip temperature decreases by an additional 32°C.

Next, we performed a steady-state heat analysis inwhich the IGBT chip temperature was fixed at 126°C,and we analyzed the relationship between copper foilthickness and heat conduction. Those results areshown in Fig. 5.

Compared to the DCB substrate copper foil thick-ness of 0.25 mm, a copper foil thickness of 0.6 mmexhibited greater conduction of heat. From this result,it can be understood that increasing the thickness ofthe copper foil decreases the density of heat flowthrough the alumina ceramic layer.

Additionally, we performed a steady-state heatanalysis to investigate the correlation between copperfoil thickness and chip temperature for the twoalumina ceramic layer thicknesses of 0.32 mm and0.635 mm. Those results are shown in Fig. 6.

It can be seen that decreasing the alumina ceramic

layer thickness and increasing the copper foil thick-ness are effective measures for lowering the IGBT chiptemperature. Moreover, in a steady-state heat analy-sis under the same conditions, the IGBT chip tempera-ture was 125°C in the case of a heat-dissipating metal-base-equipped structure. To realize an IGBT chiptemperature of 125°C with a heat-dissipating base-freestructure, the same thermal resistance was obtainedby selecting an alumina ceramic layer thickness of0.32 mm and a copper foil thickness of 0.6 mm.

4. Results of Testing and Verification with ActualMachines

In order to verify the above analysis results, wemeasured the transient thermal resistance and steady-state thermal resistance of a DCB test piece andmeasured the mechanical characteristics of a DCBsubstrate.

4.1 Transient thermal resistanceWe input a DC 80 A current to the IGBT chip and

investigated the relationship between current conduc-tion time and IGBT chip temperature for three differ-ent types of DCB substrates (using the same conditionsas the FEM simulation of Fig. 4).

Figure 7 shows the relationship between the cur-rent conduction time and chip temperature.

First of all, one second after the start of currentconduction, it can be seen that the conventional heat-dissipating base-free structure (alumina ceramic thick-ness of 0.635 mm, copper foil thickness of 0.2 mm) hadan IGBT chip temperature which was 85°C higherthan that of the heat-dissipating metal-base-equippedstructure (DCB: alumina ceramic layer thickness of0.32 mm, copper foil thickness of 0.2 mm, and basethickness of 3 mm).

Secondly, while maintaining the heat-dissipatingbase-free structure but reducing the thickness of thealumina ceramic layer in the DCB substrate from0.635 mm to 0.32 mm, it can be seen that the IGBTchip temperature decreases by approximately 20°C.(See q of Fig. 7.)

Fig.6 Relationship between copper foil thickness and chipjunction temperature Fig.7 Time dependent temperature rise of three different types

of substrates

IGB

T c

hip

tem

pera

ture

(°C

)

Copper foil thickness (mm)0.50.40.30.20.1 0.6 0.7 0.8

120

140

160

180

200

100

0.635 mm

0.32 mm

Alumina ceramic thickness:q

w

0.1 1 100.010.001

80

100

120

140

160

180

200

220

60

4020

IGB

T c

hip

tem

pera

ture

(°C

)

Current conducting time (s)

Alumina thickness: 0.32 mm

Copper foil thickness: 0.6 mm



Fig.5 Relationship between copper foil thickness and heatconduction area

IGBT chip

Chip temperature: 126°C steady-state condition

Copper foil thickness: 0.25 mm

Copper foil thickness:0.6 mm

Surface copper foil

Applied power

Alumina thickness

Copper foil thickness

109 W

0.32 mm

0.25 mm

150 W

0.32 mm

0.6 mm

Tj = 126°C Tj = 126°C

Development of a Next-generation IGBT Module using a New Insulating Substrate 55

Thirdly, by reducing the thickness of the aluminaceramic layer in the DCB substrate to 0.32 mm andincreasing the copper foil thickness to 0.6 mm, it canbe seen that the IGBT chip temperature decreases by66°C compared to the case where the copper foilthickness is 0.25 mm (see w of Fig. 7).

Fourthly, by reducing the thickness of the aluminaceramic layer to 0.32 mm and increasing the copper foilthickness to 0.6 mm, it can be seen that the IGBT chiptemperature decreases by 86°C compared to the con-ventional heat-dissipating base-free structure.

This value is nearly the same as the IGBT chiptemperature of the heat-dissipating metal-base-equipped structure. Accordingly, the above resultsverify that decreasing the thickness of the aluminaceramic layer in the DCB substrate and increasing thecopper foil thickness are extremely effective measuresfor improving the transient thermal resistance charac-teristic.

4.2 Steady-state thermal resistanceWe measured the steady-state thermal resistance

under the same conditions as used in the measurementof transient thermal resistance. Figure 8 showsthermo-photographs of the IGBT chip in a steady-statecondition. With a new heat-dissipating base-freestructure using an alumina ceramic layer thickness of0.32 mm and a copper foil thickness of 0.6 mm, theIGBT chip temperature decreased by 62°C compared tothe conventional heat-dissipating base-free structure,and this IGBT chip temperature is approximately thesame as that of the heat-dissipating metal-base-equipped structure. From these experimental results,it was verified that the use of a thinner alumina layerin the DCB substrate and a thicker copper foil areeffective measures for improving the steady-state ther-mal resistance.

4.3 Mechanical characteristics of the DCB substrateFigure 9 shows cross-sectional photographs of DCB

substrates. In the figure, (a) shows a DCB substratethat uses the conventional heat-dissipating base-freestructure, and (b) shows the newly-developed thickcopper foil DCB substrate.

When a copper foil of thickness 0.4 mm or greateris directly bonded to a alumina ceramic layer having a

typical purity of 96 %, differences in the coefficients ofthermal expansion between the alumina ceramic mate-rial and the copper cause cracking to occur in thealumina ceramic layer near the bonded junction. Thebending strength of 96 % pure alumina ceramic mate-rial is approximately 400 MPa, and as a DCB sub-strate, this strength is insufficient for bonding to athick copper foil.

Therefore, by adding zirconia to the aluminaceramic material to increase its mechanical strengthand increase its bending strength to 700 MPa, wesucceeded in bonding a 0.6 mm-thick copper foil to thisalumina ceramic (Table 2). Compared to the conven-tional heat-dissipating base-free structure, a structurethat uses this new DCB substrate achieves an approxi-mate 30 % total improvement in mechanical strengthdue to the increased thickness of the copper foil,despite the thin alumina ceramic layer.

By making the copper foil thicker and by using athin zirconia-doped alumina ceramic layer, a new heat-dissipating base-free structure is possible, realizinggreater mechanical strength and lower thermal resis-tance of the module compared to the conventional heat-dissipating base-free structure.

4.4 Evaluation and results of reliability testingThere were concerns that the increase in thickness

of the copper foil would lead to an increase in thecoefficient of thermal expansion of the copper circuitry,deterioration of the solder at the bottom of the silicon

Table 2 Fabrication limits for DCB

Fig.9 Cross-sectional photographs of DCB substrates

Fig.8 Thermo-photographs (at steady-state condition)

Conventional heat-dissipating base-free

structure

186°C

Newly developedproduct

124°C

Heat-dissipatingmetal-base-equipped

structure

120°C

(a) DCB substrate of conventional heat-dissipating base-free structure

Alumina = 0.635 mm

Copper foil = 0.25 mm


(b) Newly developed DCB substrate

Alumina = 0.32 mm



Copper foil thickness (mm)

Alumina

Ceramic thickness: 0.32 mm : Possible × : Not possible

0.3 0.4 0.5 0.6 0.7

× – – –

Zirconia-doped alumina ceramic

×


Fig.11 Fuji Electric’s 1,200 V IGBT module line-up

chip, cracking of the alumina ceramic layer, and so on.Therefore, we performed heat cycle tests and powercycle tests to investigate the reliability of a heat-dissipating base-free structure that uses this new DCBsubstrate.

A heat cycle test was performed for 500 cyclesunder test conditions of -40 to +125°C, and it wasverified that there was no degradation of the solder atthe bottom of the IGBT chip, no change in thermalresistance due to solder deterioration, and no crackingof the alumina ceramic layer.

A power cycle (intermittent operation) test wasalso performed under the test condition of ∆Tj-c = 75

deg. No changes in thermal resistance or electricalcharacteristics have been observed through 350 k cy-cles of the test. The above results demonstrate that anIGBT module using the new DCB substrate achievesthe same reliability characteristics as a conventionalIGBT module.

5. Product Evaluation Results

Table 3 shows the results of an evaluation of FujiElectric’s Small Pack which uses the new heat-dissipating base-free structure.

The use of the new heat-dissipating base-freestructure achieves 30 % lower thermal resistance thanthe conventional Small Pack. The new heat-dissipat-ing base-free structure also achieves sufficient modulemechanical strength and reliability.

Figure 10 shows the characteristics (∆Tj-c andoutput) of Fuji Electric’s latest IGBT chip used in theSmall Pack. A Small Pack that employs the conven-tional heat-dissipating base-free structure can be usedin inverter systems of up to 7.5 kW maximum. Howev-er, a Small Pack that employs this new DCB substratecan be used in inverter systems of up to 11 kWmaximum.

Figure 11 shows Fuji Electric’s 1,200 V IGBTmodule line-up. The application of a new heat-dissipating base-free structure enables the power ap-plied per unit area to be increased from 3.5 W/mm2 to5.1 W/mm2.

6. Conclusion

By adding zirconia to a DCB substrate of aluminaceramic material, decreasing the thickness of thealumina ceramic layer and increasing the thickness ofthe copper foil, a lightweight, compact and low costIGBT module that uses a low thermal resistance heat-dissipating base-free structure has been developed.

Fuji Electric efforts in developing this new IGBTmodule structure have been described above. FujiElectric intends to continue to expand the range ofapplications for this technology and to develop IGBTmodules that satisfy increasingly severe customerneeds and new demand.

References(1) Nishimura, Y. et al. Improvement of Thermal Resist-

ance in Metal Base Free Structure IGBT Modules byThicker Cooper Foil Insulation Substrate. Publicationin Industry Applications Society — Annual MeetingIAS’03. Salt Lake City, USA, October 2003.

(2) Nishimura, Y. et al. New generation metal base freeIGBT module structure with low thermal resistance.The 16th International Symposium on Power Semicon-ductor Devices & ICs (ISPSD’04). Kitakyushu, Japan,May 2004.

(3) US PAT. 5675181, 5869890.

Fig.10 Characteristics of Fuji Electric’s IGBT chips

Table 3 Evaluation results of Fuji Electric’s new product thatuses the new DCR structure

Thermal resistance Rth(j-f) 100


71

New structure

∆Tj-

c (de

g)

Io (Arms)20100 30 40 50

5

0

10

15

20

25

30

35

40

45

50


Newstructure

7.5 kWinv.150 %

11 kWinv.150 %

Cap

acit

y

Module size

5.3 W/mm2PC2

PC3100 A(22 kW)

50 A(11 kW)

35 A(7.5 kW)

75 A

Heat-dissipating base-free structure

25 A(5.5 kW)

10 A

4.7 W/mm2

3.5 W/mm2

21.5 48.4 75.6 (cm2)



Mitsuhiro KakefuMasaki Ichinose

Low I R Schottky Barrier Diode Series

1. Introduction

Representative of the recent trends towards small-er size and higher functionality of portable devices andtowards higher speed CPUs for computers, electronicdevices are rapidly becoming smaller in size, lighter inweight and are achieving higher performance, and it isessential that their circuit boards and switching powersupplies be made to consume less power, are moreefficient, generate less noise and support higher densi-ty packaging. Moreover, in order to suppress the surgevoltage that is applied across a diode during switchingand the noise generated by a steep dv/dt characteristic,snubber circuits, beads and the like are used, but as aresult the number of components increases, leading togreater cost. In order to achieve better portability, ACadapters for notebook computers are being miniatur-ized; however, the trend toward higher power con-sumption results in higher internal temperatures,increasing the severity of the environment in whichthese semiconductor devices are used. Consequently,semiconductor devices are strongly required to providethe characteristics of lower loss, improved suppressionof thermal runaway, higher maximum operating tem-perature and lower noise. In particular, an improve-ment in the characteristics of the secondary sourceoutput rectifying diode, which accounts for nearly 50 %of the loss in a switching power supply, is stronglydesired.

2. Overview

Schottky barrier diodes (SBDs) exhibit the proper-ties of low forward voltage (VF), soft recovery and lownoise, and are widely used in the secondary sourcerectifying circuits of switching power supplies. FujiElectric has previously developed a product line ofconventional 20 to 100 V SBDs (low VF type) and 120to 250 V SBDs [low reverse current (IR) type] as adiode series available in a variety of packages andsupporting various output voltages and current capaci-ties in order to be applicable to a wide range of powersupply applications. However, when the conventionallow VF type SBD operates at high temperatures, its IR

becomes large, and as a result reverse loss increases,efficiency decreases and thermal runaway may occur,making it difficult to use this low VF SBD in a smallpower supply packages such as an AC adapter.

The newly developed low IR-SBD is considered tobe the ideal diode for secondary source rectification ina switching power supply, and is especially well suitedfor rectification in a high temperature environment.Figure 1 shows the cross-sectional structure of the SBDchip. The chip design incorporates a guard ring toprevent premature breakdown, and the doping density,specific resistance and thickness of the epitaxial layer(n- layer), diffusion depth, and barrier metal that havebeen optimized to develop a low IR-SBD series thatprovides not only low IR, but also breakdown voltagesof 40, 60 and 100 V, comparable to the conventional VF.Compared to a conventional SBD having the samebreakdown voltage, this product achieves an approxi-mate single-digit decrease in IR, a large decrease inreverse loss, a higher temperature at which thermalrunaway occurs, and a higher maximum operatingtemperature. Moreover, this new series has a highavalanche breakdown voltage and is expected to becapable of withstanding the large surge voltage thatoccurs when a power supply is turned on. The newseries is also expected to enable the design of switchingpower supply circuits that realize increased efficiency,smaller size and greater flexibility. Table 1 lists theabsolute maximum ratings and electrical characteris-tics of this low IR-SBD series and Fig. 2 shows external

Fig.1 Cross-sectional structure of SBD chip

Guard ring Schottky electrode (barrier metal) SiO2

Epitaxial layer

Si substrate


Table 1 Absolute maximum ratings and electrical characteristics of low IR SBD

Fig.2 External view of the packages

Model number Package

YG862C04R TO-220F 45 45 10 125 330 0.61 150 3.50

YA862C04R TO-220 45 45 10 125 330 0.61 150 2.00

TS862C04R T-Pack 45 45 10 125 330 0.61 150 2.00

YG862C06R TO-220F 60 60 10 125 330 0.68 150 3.50

YA862C06R TO-220 60 60 10 125 330 0.68 150 2.00

TS862C06R T-Pack 60 60 10 125 330 0.68 150 2.00

YG862C10R TO-220F 100 100 10 125 330 0.86 150 3.50

YA862C10R

TS862C10R

TO-220 100 100 10 125 330 0.86 150 2.00

T-Pack 100 100 10 125 330 0.86 150 2.00

YG865C04R TO-220F 45 45 20 145 330 0.63 175 2.50

YA865C04R TO-220 45 45 20 145 330 0.63 175 1.75

TS865C04R T-Pack 45 45 20 145 330 0.63 175 1.75

YG865C06R TO-220F 60 60 20 145 660 0.74 175 2.50

YA865C06R TO-220 60 60 20 145 660 0.74 175 1.75

TS865C06R T-Pack 60 60 20 145 660 0.74 175 1.75

YG865C10R TO-220F 100 100 20 145 660 0.86 175 2.50

YA865C10R TO-220 100 100 20 145 660 0.86 175 1.75

TS865C10R T-Pack 100 100 20 145 660 0.86 175 1.75

YG868C04R TO-220F 45 45 30 160 1,000 0.63 200 2.00

YA868C04R TO-220 45 45 30 160 1,000 0.63 200 1.25

TS868C04R T-Pack 45 45 30 160 1,000 0.63 200 1.25

YG868C06R TO-220F 60 60 30 160 750 0.74 200 2.00

YA868C06R TO-220 60 60 30 160 750 0.74 200 1.25

TS868C06R T-Pack 60 60 30 160 750 0.74 200 1.25

YG868C10R TO-220F 100 100 30 160 750 0.86 200 2.00

YA868C10R TO-220 100 100 30 160 750 0.86 200 1.25

TS868C10R T-Pack 100 100 30 160 750 0.86 200 1.25

Absolute maximum ratings Electrical characteristics

VRRM(V)

VRSM(V)

IO(A)

IFSM(A)

PRM(W)

Rth(j-c)(°C/W)

VFM (V)IF = 0.5 × IO(Tj = 25°C)

IRRM (µA)

VR = VRRM

views of the packages. The current ratings are 10 A,20 A and 30 A and the product packages are availableas the TO-220, the TO-220F full-mold type, and the T-

Pack (S) surface mount type.The newly developed low IR-SBD is described

below.

YG868C

YA868C

TS868C

Model number : YG868CR Model number : YA868CR Model number : TS868CR

10 10 10

10

9.5

15

4.5

2.7

4.5

1.3

4.5

1.32

13

15 9.5

1.5

3

13.5

See view from arrow direction P

Arrow direction P


3. Device Characteristics

Figure 3 compares the forward characteristics ofthe low IR-SBD with those of conventional products,and Fig. 4 compares their reverse characteristics. TheSBD loss is the sum of the forward and reverse losses,and it is desirable that this loss be reduced within theactual operating temperature range. In particular, thereverse loss caused by increased IR at higher tempera-tures must be considered. A tradeoff relation existsbetween VF and IR, however, and VF typically increases

when IR is reduced. The newly developed 40 to 100 VSBD achieves a dramatic decrease in loss at hightemperatures through the use of a new barrier metalas described in chapter 2 and optimized crystal specifi-cations in order to achieve an approximate 10 %increase in VF at rated current compared to a conven-tional product, and an IR that is reduced to approxi-mately 1/10th that of the conventional product.

4. Consideration of the Generated Loss

A simulation was performed to calculate the loss

Fig.3 Comparison of forward characteristics

Fig.4 Comparison of reverse characteristics

1

For

war

d cu

rren

t I F

(A

)

Forward voltage VF (V)0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.00

0.01

0.1

1

10

YG805C04R 100°CYG805C04R 25°CYG865C04R 100°CYG865C04R 25°C

For

war

d cu

rren

t I F

(A

)

Forward voltage VF (V)0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.00

0.01

0.1

1

10


For

war

d cu

rren

t I F

(A

)

Forward voltage VF (V)0.2 0.4 0.6 0.8 1.0 1.20

0.01

0.1

10


Reverse voltage VR (V)10 20 30 40

100

101

102

103

Rev

erse

cu

rren

t I R

(µA

)

Tj = 100°C

Tj = 25°C

YG805C04R

YG865C04R

YG805C04R

YG865C04R

Reverse voltage VR (V)10 20 30 40 50 60

101

100

102

103

Rev

erse

cu

rren

t I R

(µA

)

Tj = 100°C

Tj = 25°C

YG805C06R

YG865C06R

YG805C06R

YG865C06R

Reverse voltage VR (V)10 20 30 40 50 60 70 80 90 100 1100

101

100

10–1

102

103

Rev

erse

cu

rren

t I R

(µA

)

Tj = 100°C

Tj = 25°C

YG805C10R

YG865C10R

YG805C10R

YG865C10R


increases, and IR becomes more noticeable as itincreases at higher temperatures. As a result, avicious cycle ensues in which the increase in IR leads toan increase in loss, which generates heat in theelement, leading to an increase in IR, etc. In somecases, this phenomenon ultimately leads to thermaldamage (thermal runaway) of the element. Figure 6shows thermal runaway data of the ambient tempera-ture vs. reverse voltage for a 40 V/30 A product. Forthe sake of comparison, a conventional SBD is alsoshown. Compared to the conventional product, it canbe seen that the allowable operating temperaturerange has been expanded due to the lower IR. Table 2shows the estimate thermal runaway temperatures fora 60 V/20 A product in 24 V output power supplies(Vdc = 380 V, I = 3.5 A or 5 A) for 23-inch and 30-inchLCD TVs, which approximate actually installationconditions. Compared to the conventional product, thethermal runaway ambient temperature is estimated tobe 32 % higher (98°C) at the forward side and 28 %higher at the flyback side (108°C) in the case of the 23-inch LCD, and 34 % higher (97°C) at the forward sideand 29 % higher at the flyback side (100°C) in the caseof the 30-inch LCD. With a high maximum allowableoperating temperature, these new devices are wellsuited for high temperature applications.

6. Conclusion

An overview of the low IR-SBD and its applicationto secondary source rectification applications in switch-ing power supplies have been presented.

In response to the anticipated future requests forpower supplies that are smaller in size, generate lessloss and have higher efficiency, Fuji Electric intends tofurther improve SBD characteristics and to develop aproduct line of small package products. Fuji willcontinue to make additional improvements in order todevelop high quality products and enrich this productseries.

generated in the case of a 24 V power supply(Vdc = 380 V, I = 5 A) for a liquid crystal display (LCD)TV. Figure 5 shows the relationship between junctiontemperature (Tj) and estimated loss (Wo) for a 60 V/20 A product. For the sake of comparison, a conven-tional SBD is also shown. In the region of low Tj, theconventional product has less loss, but because IR hasa large effect on loss at high temperatures, the low IRdevice achieves less loss than the conventional deviceat high temperatures, and at Tj = 150°C, the low IRproduct achieves approximately 76 % less loss than theconventional product, and its application to higherefficiency power supplies is anticipated.

5. Consideration of the Thermal Runaway Tem-perature

The temperature of an element rises as its loss

Fig.5 Junction temperature vs. estimated loss (60 V/20 A)

Fig.6 Thermal runaway data (TS868C04R, TS808C04R)

Table 2 Ambient temperature when beginning thermalrunaway at LCD-TV 24 V output power supply

Est

imat

ed lo

ss W

O (

W)

Junction temperature Tj (°C)10080604020 120 140 160 180

1

2

3

4

5

6

7

8

0

YG805C06R Forward sideYG805C06R Flyback sideYG865C06R Forward sideYG865C06R Flyback side

New device: YG865C06R

Conventional device: YG805C06R

Rev

erse

vol

tage

VR (

V)

Ambient temperature Ta (°C)500 100 150 200

5

0

10

15

20

25

30

35

40

45

DC

DC21

New device: TS868C04R

Conventional device: TS808C04R

Model number

Conventional device : YG805C06R

Approx.74°C

Approx.84°C

Approx.72°C

Approx.77°C

New device : YG865C06R

Approx.98°C

Approx.108°C

Approx.97°C

Approx.100°C

23-inch LCD-TVpower supply

(+24 Vout/3.5 A)

30-inch LCD-TVpower supply

(+24 Vout/5.0 A)

Condition : installation cooling fin (30°C/W)

Forward Flyback Forward Flyback


Fig.1 PDP circuit configuration

Yukihito HaraMasanori Inoue

Medium-voltage MOSFETs for PDP-use

1. Introduction

Flat-screen televisions are a digital consumerappliance of modern convenience, and plasma displaypanel (PDP) flat-screen televisions are capable ofdisplaying high definition images on a large screen.With the evolution in PDP technology towards higherquality images, larger screen size and lower cost, andwith the partial start-up of digital terrestrial televisionbroadcasting in Japan, the popularity of PDPs isincreasing at a rapid rate.

PDP technology is trending toward lower powerconsumption, higher brightness, quieter (fan-less) op-eration, and thinner panel sizes. Similarly, theattributes of higher efficiency and smaller and thinnersize are required of the sustain circuit that controls theplasma light emission. Figure 1 shows the basic circuitconfiguration of a PDP. The sustain circuit consists ofan on/off circuit, a main circuit (X/Y sustain circuits)and a power recovery circuit, and also utilizes severaldozens of power MOSFETs (metal oxide semiconductorfield effect transistors). Because a large current flowoccurs instantaneously, it is important for this sustaincircuit to have low on-resistance. In addition, theattributes of small size (reduced number of parallelelements) and thinner profile (surface mounting) arealso required.

In response to these requirements concerning PDPsustain circuits, Fuji Electric has used its existinghigh-voltage SuperFAP-G series technology to developa new SuperFAP-G series, ranging from 150 to 300 V,for use in PDP sustain circuits. Additionally, inresponse to requests for even smaller size and higherefficiency, Fuji Electric is developing trench MOSFETscapable of realizing even lower resistance. It isanticipated that the application of these products willenable small mounting area and more efficient mount-ing due to the reduced number of MOSFET elementsin the sustain circuit and also small cooling elements(heat sinks) and higher operating efficiency due to thelower loss.

The product line and characteristics of the SuperFAP-G series and the trench MOSFETs are describedbelow.

2. Characteristics of PDP Power MOSFETs

It is important that the power MOSFETs used inPDP sustain circuits have low on-resistance. Charac-teristics of the SuperFAP-G series and the trenchMOSFETs are described below.

2.1 Characteristics of the SuperFAP-G seriesCompared to the conventional MOSFET, the Su-

perFAP-G series features an improved gate-draincharge (Qgd) tradeoff relation, the charging time con-stant determined by the on-resistance (Ron) and turn-off loss, also a dramatic improvement in the Ron · QgdMOSFET figure-of-merit. The following technologywas adopted to realize these characteristics.(1) QPJ technology

Resistivity of the epitaxial layer is the predomi-nant component of on-resistance in a MOSFET andsimply lowering this resistivity causes the drain-sourcebreakdown voltage to decrease. A conventional MOSFET has a 3-dimensional cellular structure, and highelectric fields exist locally in portions of the structure.As an improvement to the conventional structure, thequasi-plane junction (QPJ) shown in Fig. 2 was devel-oped. The QPJ features a bonded planar cellularstructure realized by fabricating a dense arrangement

PDPpanel

Addressdriver

PFC DC-DC

Signal processing™ Interface circuit™ Logic circuit

Su

stai

n o

n/o

ffci

rcu

it

Sca

n d

rive

r

Sca

n d

rive

r

Y-s

ust

ain

circ

uit

X-s

ust

ain

circ

uit

Pow

erre

cove

ryci

rcu

it

Pow

erre

cove

ryci

rcu

it


of low concentration shallow p- wells instead of thedeep p+ wells used conventionally. Accordingly, theelectrical fields in the cellular structure are reduced,enabling the use of an epitaxial resistive layer havinglower resistivity than that of the epitaxial resistivelayer in a conventional MOSFET.

By employing the QPJ structure, the width of then- silicon region (current path) becomes narrower andshorter, and Qgd can be decreased. On the other hand,

the narrowing of the n- silicon region is correlated toan increase in on-resistance, and a tradeoff relationexists between the width of this region and Qgd. Theincrease in on-resistance is caused by narrowing of thecurrent path in the n- silicon region due to anexpanding depletion layer. In order to limit thisexpansion of the depletion layer, the concentration ofimpurities in the n- silicon region was optimized andthe tradeoff relation improved. By applying theseenhancements, Ron · Qgd was decreased by approxi-mately 60 % compared to Fuji Electric’s conventionalMOSFET. Table 1 lists characteristics of a SuperFAP-G and a conventional product.(2) Guard ring

By employing a design that uses the QPJ struc-ture, the device will achieve a reduced cellular electricfield and it will be possible to use a low-resistanceepitaxial layer. However, by continuing to use theconventionally designed breakdown structure, a prob-lem occurs in that the electrical field of the breakdownstructure becomes greater than that of the cellularstructure, and the breakdown voltage cannot be en-sured. It is essential for the electric field of the

Fig.3 Planar chip structure and trench chip structure

Fig.2 Comparison of SuperFAP-G chip structure (QPJstructure) and conventional MOSFET

Table 1 Comparison of characteristics of SuperFAP-G andconventional product

n+ n+

p+

n+

n-

n+

n-

n+

n+ n+

Gate

Gate

Drain

Drain

Source

Source

p+ p- p- p+p- p-p-

(a) Conventional power MOSFET

(b) SuperFAP-G

p-p+n+ n+

p-p+n+ n+

p-p+n+ n+

p-p+n+ n+

p-p+n+ n+

p-p+n+ n+

n+

n+

n-

p+

p p p p

Gate oxide layer

Gate oxidelayer

Trench

Gate Source Gate

Drain

Source

(Rch) (Rch)(Rch)

(Repi)

(RjFET)

(Rsub) (Rsub)

(Repi)

n+ n+ n+ n+ n+p+ p+

p

(a) Planar chip structure

n+

n-

Drain

(b) Trench chip structure

Series SuperFAP-G Conventional product

2SK3535Parameter 2SK2254

250 V 250 VVDS

±25 A ±18 AID

270 W 80 WPD

3 to 5 V 2.5 to 3.5 VVGS (th)

75 mΩ 130 mΩRDS (on)(typ)

50 nC 52 nCQg

16 nC 16 nCQgd

1.20 Ω · nC 2.08 Ω · nCRon · Qgdfigure-of-merit


Table 4 Fuji Electric’s product line of power MOSFETs for PDP-use

breakdown structure to be lower than that of thecellular structure, and the development of a break-down structure capable of reducing the electric fieldwas necessary. While using an epitaxial layer of lowresistance, an irregularly-pitched optimized guard ring(OGR) was applied in order to reduce the electric fieldof the breakdown structure. The electric field wassimulated in order to determine the optimal pitch andnumber of guard rings for a design which realizes a

lower electric field in the breakdown structure than inthe cellular structure. Moreover, in order to insurereliability, the design was made resistant to anyinfluence from charge accumulation.

2.2 Characteristics of trench MOSFETsFuji Electric has previously promoted its Super-

FAP-G series that realizes low on-resistance. Howev-er, in response to requests from the PDP field for evenlower on-resistance, Fuji Electric is concentrating on

Table 2 Comparison of ON-resistance of conventionalMOSFET and trench MOSFET Table 3 Sheet resistance (calculated value: TO-220 series)

Fig.4 Comparison of the ON-resistance components of a200 V conventional planar chip and a trench chip

Fig.5 Internal structure of MOSFET (surface mount type)

On

-res

isti

vity

(%

)

Planar MOSFET

Repi

Rsub

RjFET

20

0

40

60

80

100

120

Trench MOSFET

Rch + Racc

Drain-source breakdown voltage

200 V 66 50.6 23.3 %

250 V 100 84 16 %

On-resistance Ron (mΩ)

ConventionalMOSFET

TrenchMOSFET

Ron reduction rate

* on development

TO-220 TO-220F T-Pack(D2-Pack) TFP TO-247 TO-3PF

–57 A 41 mΩ 2SK3590 2SK3591 2SK3592 2SK3593 –

–65 A 28.5 mΩ *FMP65N15T2 *FMA65N15T2 *FMB65N15T2 – –

2SK378892 A 26 mΩ – – – – 2SK3789

2SK3882100 A 16 mΩ – – – – –

BreakdownBVDSS

150 V

–45 A 66 mΩ 2SK3594 2SK3595 2SK3596 2SK3597 –

–37 A 100 mΩ 2SK3554 2SK3555 2SK3556 2SK3535 –

–32 A 130 mΩ 2SK3772 2SK3773 2SK3774 2SK3775 –

–49 A 50.6 mΩ *FMP49N20T2 *FMA49N20T2 *FMB49N20T2 – –

–38 A 84 mΩ *FMP38N25T2 *FMA38N25T2 *FMB38N25T2 – –

2SK377653 A 72 mΩ – – – – 2SK3777

2SK388586 A 40 mΩ – – – – –

2SK378073 A 36 mΩ – – – – 2SK3781

2SK377859 A 53 mΩ – – – – 2SK3779

2SK3883100 A 20 mΩ – – – – –

2SK3884100 A 30 mΩ – – – – –

200V

250V

300V

Rated current

ID

ON-resistance

RDS (on)

Package

= 400 µm×2 wires

0.4 mΩ

= 400 µm×2 wiresStitch bondingφ φ

0.2 mΩSheet resistancecalculated value

Epoxy resin

Semiconductorchip

Lead wire

Base frame


Fig.6 External view of packagesdeveloping next generation products and, based onFuji’s track record of success with the trench gatestructure technology (60 V and 75 V devices for auto-motive use), is optimizing the trench depth and waferspecifications in order to achieve higher performance.

Figure 3 shows a cross-sectional comparison of theplanar chip and trench chip structures. In the trenchchip structure, a gate is formed on a groove (trench)that passes through the channel, and this enables thecell to be made smaller and the channel resistance(Rch) component and JFET resistance (RjFET) compo-nent to be reduced, which had been difficult toimplement with the planar chip structure shown inFig. 4. Table 2 compares the on-resistance of theconventional MOSFET with that of the trench MOSFET. As can be seen, adoption of the trench gatestructure achieves a large decrease in on-resistance.

2.3 Reduction of the internal wiring resistanceBecause the MOSFETs used in PDPs must have

low on-resistance, it is important to reduce the on-resistance of the MOSFET chips and the resistance ofwiring inside the package. For low on-resistance chipsof the150 V drain-source voltage class, the sheet resis-tance of the source aluminum electrode on the chip’ssurface increases as a percentage of the total on-resistance of the product. Figure 5 shows the internalstructure of a T-Pack package. A reduction in sheetresistance is achieved by bonding the source aluminumwire to the chip’s source electrode in several locations.Table 3 shows the effectiveness of reducing the sheetresistance.

3. Product Line and External Appearance

Table 4 lists a summary of Fuji Electric’s powerMOSFET series for PDP-use. Figure 6 shows theexternal appearance of these packages. The productline contains drain-source voltages ranging from 100 to

300 V and a series of TO-220, TO-3PF and TO-247packages. A series of T-Pack (D2-Pack) and TFPsurface-mount products are also available and areanticipated to contribute to the production of thinnersustain circuits.

4. Conclusion

This paper has described features of Fuji Electric’sSuperFAP-G series for PDP-use and Fuji’s medium-voltage trench MOSFETs, which are still under devel-opment. In the future, Fuji Electric intends tocontinue to develop products optimized for PDPs andto contribute to the development of the PDP industry.

References(1) Kobayashi, T. et al. High-voltage Power MOSFETs

Reached Almost to the Silicon Limit. Proceedings ofISPSD’01. 2001, p.435-483.


Fig.1 PDP module drive circuit

Hideto KobayashiGen TadaHitoshi Sumida

PDP Scan Driver IC

1. Introduction

A shift in consumer electronics from analog todigital technology is underway, and the televisionindustry is transitioning from CRT to flat paneldisplay (FPD) technology. With the increasing popu-larity of FPDs, the market for plasma display panels(PDPs) has also grown rapidly. The FPD market isdivided according to screen size: sizes of 30 inches andsmaller use liquid crystal display (LCD) technology,sizes from 40 to 50 inches use PDP technology, andlarger sizes use projector technology. However, compe-tition among the different FPD technologies is intensi-fying and the market division according to screen sizeis becoming less prevalent. Within this context, PDPtechnology is being required to provide such perfor-mance improvements as lower power consumption andhigher luminous efficient, as well as lower cost.

There are two types of PDP driver ICs, a scandriver IC that selects scan lines and an address driver(or data driver IC) that selects data lines. Thecharacteristics of the driver ICs affect the quality ofthe image display, and because many IC devices areused in a single panel, these driver ICs are required toprovide high performance at a low cost.

Fuji Electric develops both scan driver ICs andaddress driver ICs and this paper describes thetechnology used in Fuji Electric’s FD3284F PDP scandriver IC which features high current and low on-stateresistance.

2. Features of the PDP Scan Driver IC

Figure 1 shows the drive circuit of a PDP module.The scan driver IC has 64-bit output lines, and 12 ofthese scan driver ICs are used in an extended graphicsarray (XGA) panel. The basic operation of the scandriver IC is shown in Fig. 2.(1) Scan period

During the scan period, the scan driver IC selects ascan line, and according to signal from an addressdriver IC, outputs a 100 V address discharge to the cellto be displayed.(2) Sustain period

During the scan period, the scan driver IC alter-nates operation with the sustain driver IC, andoutputs a 160 V sustain discharge to the cell thatreceived an address discharge during the scan period.

This sustain discharge operation is repeated toimplement a grayscale display.

The scan driver IC must be able to supply a largecurrent at a high voltage during address and sustaindischarges.

Fig.2 Scan driver IC operation

Su

stai

n d

rive

r IC

Scan driver IC

PDP panel

Scan driver IC

Scan driver IC

Scan driver ICA

ddre

ss d

rive

r IC

Add

ress

dri

ver

IC

Add

ress

dri

ver

IC

Add

ress

dri

ver

IC

Add

ress

dri

ver

IC

Add

ress

dri

ver

ICScan driver IC

Scan driver IC

Scan driver IC

Sustain driver IC

Scan period Sustain period


3. PDP Scan Driver IC Technology

3.1 Process and device technologyFuji Electric has been using a silicon-on-insulator

(SOI) method of dielectric isolation technology. Al-though the SOI process has the disadvantage of anexpensive wafer cost, it features small device isolatedareas and latch-up free operation, and therefore SOIprocess technology is well suited for scan driver ICsthat require high voltage and high current.

The output device uses insulated gate bipolartransistors (IGBTs) connected in a totem pole configu-ration. The output circuit of a scan driver IC occupies60 % of the IC area and therefore the output devicesize has a large impact on cost. The IGBT, which isable to output a large current from a small area, isoptimally suited as an output device for a scan driverIC. As shown in Fig. 3 and Table 1, the newlydeveloped third-generation SOI-IGBT device is smallerand achieves greater drive capability than a conven-tional IGBT.

3.2 Circuit technologyFigure 4 shows the output stage circuit of a scan

driver IC. N-channel IGBTs are connected in a totempole output configuration. The high-side IGBT (N2 inthe figure) is controlled by a level shifter, and becausethe IGBT gate is driven at 5.5 V, a 5.5 V zener diode isinserted between the gate and source for protection.

The operation of the output stage circuit is summa-rized below.(1) Scan period

IGBTs N1 and N2 operate to output selectedwaveforms and unselected waveforms. During anaddress discharge, the N1 IGBT supplies a largecurrent.(2) Sustain period

The N1 IGBT and the D1 diode operate to providea sustain discharge current supplied from both the N1and D1 devices.

As the size of the display panel increases, a largerdischarge current is required for address discharge andsustain discharge, and the on-state resistance of thedevice becomes an issue. If the device has a large on-state resistance, it will emit a large quantity of heatand the resulting temperature rise will lead to degra-dation of the device characteristics and a correspond-ing degradation of display quality. Comparing the on-state resistance of the N1 IGBT and the D1 diodereveals that when the on-state resistance of the D1diode is 1.4 V/400 mA, the N1 IGBT will have a largeon-state resistance of 6.0 V/400 mA. Because the N1IGBT operates during both the scan period and thesustain period, its on-state resistance has a dominanteffect on the amount of heat generated. Accordingly,the key to the development of a successful scan driverIC is to provide an N1 IGBT device with high drivecapability.

Table 1 FD3284F characteristics

Fig.3 IGBT device comparison

Fig.4 Output circuit

Conventional2nd gen. IGBT

Newly developed3rd gen. IGBT

Area ratio: 90 %

Parameter

Absolute maximum voltage (logic circuit) 7.0 V 7.0 V

Absolute maximum voltage (output circuit) 165 V 165 V

Operating voltage (logic) 5.0 V 5.0 V

Operating voltage (output) 30 to 130 V 30 to 130 V

High output source or sink current

- 200 mA/ +1,000 mA

- 200 mA/ +1,500 mA

High output diode current - 1,000 mA/ +1,000 mA

- 1,200 mA/ +1,500 mA

FD3284FConventional IGBT

D1 (diode)

N2 (IGBT)

N1 (IGBT)

Level shifter

Gatecontroller

(Zener diode)

Current flow during address period

Current flow duringsustain period


Fig.7 FD3284F block diagram

Fig.6 FD3284F’s low-level RON (N1)

Fig.5 Address discharge operation

3.3 IGBT gate control technologyIf the current density of an IGBT device is

increased so that large current can be obtained in asmall area, the safe operating area (SOA) will becomesmaller and if an unusual discharge occurs duringaddress or sustain discharge operations and causes anoverload or short-circuit state, the device will exceedits SOA and be damaged.

Similarly, if waste metal adheres between theoutput terminals and the device unexpectedly entersan overload state, the device will exceed its SOA andbecome damaged.

On the other hand, in an attempt to expand theSOA, if the current density is decreased, the devicearea will increase due to the tradeoff relation thatexists between current density and cost will alsoincrease. In order to resolve this tradeoff betweencurrent density and SOA, Fuji Electric has developedtechnology for controlling the gate voltage of the N1IGBT in accordance with the output timing. Thisoperation is shown in Fig. 5.

During the scan period, in the case of a 5 V supplyvoltage VDL, so as to supply sufficient current for theoutput to drop, a voltage of approximately 4.5 V isapplied to the gate of the N1 IGBT, causing the outputvoltage to drop from 100 V to 0 V and a scan line to beselected.

The address discharge begins, and when a largercurrent is required, the gate voltage of the N1 IGBT isincreased to 7 V to boost the current drive capability ofthe N1 IGBT.

After the address discharge, the gate voltage isagain lowered to 4.5 V to reduce the drive capability.Then, 1.5 µs after the scan period, the voltage of theN1 IGBT gate is gradually decreased until the IGBTturns off. This control prevents the N1 IGBT fromoperating outside its normal operating period, therebyprevent possible damage due to an unexpected over-load condition caused by an unusual discharge, shortcircuit or the like.

Figure 6 shows the characteristics of an N1 IGBTthat incorporates this gate control technology. Even ifthe device area is 10 % smaller than a conventionalIGBT, implementation of this gate control enablestwice the output current capacity compared to the casewithout gate control.

4. Application to a PDP Scan Driver IC

Fuji Electric’s FD3284F PDP scan driver IC, whichuses this newly developed third-generation SOI-IGBTdevice and gate control circuit technology, is describedbelow.

4.1 Features(1) 64-bit bidirectional shift register (15 MHz, with

CLR function)(2) Absolute maximum voltage: 165 V (output circuit),

7 V (logic circuit)(3) Output operating voltage: 32 to 130 V(4) Logic operating voltage: 5 V(5) High output current: -0.2 A/+1.5 A (source/sink)(6) High output diode current: -1.2 A/+1.5 A (source/

N1: Gate voltage

N1: IGBT output voltage

N1: IGBT output current

Scan period

100 V

0 V

1 A500 mA

4.4 V7.0 V

1.6

1.4

1.2

1.0

0.8

0.6

0.4

0.2

00 5.0 10.0 15.0

Ou

tpu

t cu

rren

t (A

)

Output voltage (V)

With gate control

Without gate control

OC1

DO1to

DO32

OC2

LE

CLK

DA

CLR

A/B

DB

VDL VDH1 VDH2

DO33to

DO64

VDH1

GND

VDH2

GND

GND

LEQ1 L1

Q2

Q3

Q64 L64

CLKDATA Q1

Q2

Q3

CLR

A/B

DATA Q64

64-b

it s

hif

t re

gist

er

64-b

it la

tch

Sel

ecto

rS

elec

tor


4.3 Characteristics compared to those of a conventionalICTable 1 lists the main differences in characteristics

between the conventional scan driver IC and theFD3284F. The FD3284F, which has been designed tobe capable of driving large PDP panels, featuresdramatically improved driver output current and diodeoutput current capabilities. Figure 8 shows externalviews of the FD3284F which uses a heat-radiatingexposed pad TQFP 100-pin package.

5. Conclusion

This paper has described characteristics of PDPscan driver IC technology and of Fuji Electric’sFD3284F PDP scan driver IC. Competition amongFPD technologies will intensify in the future and FujiElectric plans to continue to advance device, circuitand process technologies in order to satisfy marketrequirements for higher performance and lower costPDPs.

References(1) Kobayashi, H. et al. IDW’04. PDP3-3.

sink)(7) Package: TQFP 100-pin (exposed pad)

4.2 Circuit configurationFigure 7 shows a block diagram of the circuit

configuration. The circuit is configured from a 64-bitbidirectional shift register, a 64-bit latch, data selectorcircuits, and a totem pole output drive circuit.

Fig.8 FD3284F package

Top sideTop side

Back side (exposed pad)

: Representative Office : Sales Bases : Manufacturing Bases

Global Network

AMERICA

EU

FUJI ELECTRIC CORP. OF AMERICAUSATel : +1-201-712-0555 Fax : +1-201-368-8258U.S. FUJI ELECTRIC INC.USATel : +1-732-560-9410 Fax : +1-732-457-0042FUJI HI-TECH, INC.USATel : +1-510-651-0811 Fax : +1-510-651-9070FUJI SEMICONDUCTOR, INC.USATel : +1-972-733-1700 Fax : +1-972-381-9991GE FUJI DRIVES, USA INC.USATel : +1-540-387-5925 Fax : +1-540-387-8580GE FUJI DRIVES AMERICA S.A. de C.V.MEXICOTel : +52-8-154-7000 Fax : +52-8-154-7007

FUJI ELECTRIC HOLDINGS CO., LTD.Erlangen Representative OfficeGERMANYTel : +49-9131-729613 Fax : +49-9131-28831FUJI ELECTRIC FA EUROPE GmbHGERMANYTel : +49-69-6690290 Fax : +49-69-6661020FUJI ELECTRIC (SCOTLAND) LTD.U.K.Tel : +44-1355-234111 Fax : +44-1355-238810FUJI ELECTRIC FRANCE S.A.FRANCETel : +33-4-73-98-26-98 Fax : +33-4-73-98-26-99

ASIA

FUJI ELECTRIC HOLDINGS CO., LTD.China Representative Office (Shanghai)CHINATel : +86-21-6471-0897 Fax : +86-21-6471-4903FUJI ELECTRIC HOLDINGS CO., LTD.China Representative Office (Beijing)CHINATel : +86-10-6505-1263 Fax : +86-10-6505-1851FUJI ELECTRIC FA (SHANGHAI) CO., LTD.CHINATel : +86-21-6466-2810 Fax : +86-21-6473-3292FUJI ELECTRIC (CHANGSHU) CO., LTD.CHINATel : +86-512-5284-5629 Fax : +86-512-5284-5640FUJI GE DRIVES (WUXI) CO., LTD.CHINATel : +86-510-528-1932 Fax : +86-510-528-2052FUJI ELECTRIC DALIAN CO., LTD.CHINATel : +86-411-8762-2000 Fax : +86-411-8762-2030SHANGHAI FUJI ELECTRIC SWITCHGEAR CO., LTD.CHINATel : +86-21-5718-5740 Fax : +86-21-5718-1448

SHANGHAI FUJI ELECTRIC TRANSFORMER CO., LTD.CHINATel : +86-21-5718-5740 Fax : +86-21-5718-5745DALIAN FUJI BINGSHAN VENDING MACHINE CO., LTD.CHINATel : +86-411-8730-5908 Fax : +86-411-8730-5911DALIAN JIALE VENDING MACHINE OPERATION CO., LTD.CHINATel : +86-411-8596-2721 Fax : +86-411-8596-2732SHANGHAI GENERAL FUJI REFRIGERATION EQUIPMENT CO., LTD.CHINATel : +86-21-6921-1088 Fax : +86-21-6921-1066HANGZHOU FUJI REFRIGERATING MACHINE CO., LTD.CHINATel : +86-571-8821-1661 Fax : +86-571-8821-0550FUJI ELECTRIC (SHENZHEN) CO., LTD.CHINATel : +86-755-2734-2910 Fax : +86-755-2734-2912HONG KONG FUJIDENKI CO., LTD.HONG KONGTel : +852-2664-8699 Fax : +852-2664-8040FUJI ELECTRIC FA (ASIA) CO., LTD.HONG KONGTel : +852-2311-8282 Fax : +852-2312-0566FUJI ELECTRIC SYSTEMS CO., LTD.Taipei Representative OfficeTAIWANTel : +886-2-2561-1255 Fax : +886-2-2561-0528FUJI ELECTRIC TAIWAN CO., LTD.TAIWANTel : +886-2-2515-1850 Fax : +886-2-2515-1860FUJI/GE TAIWAN CO., LTD.TAIWANTel : +886-2-2556-0716 Fax : +886-2-2556-0717ATAI FUJI ELECTRIC CO., LTD.TAIWANTel : +886-3-321-3030 Fax : +886-3-321-7890FUJI ELECTRIC KOREA CO., LTD.KOREATel : +82-2-780-5011 Fax : +82-2-783-1707

East Asia

Southeast Asia

FUJI ELECTRIC SYSTEMS CO., LTD.Bangkok Representative OfficeTHAILANDTel : +66-2-308-2240, 2241 Fax : +66-2-308-2242FUJI ELECTRIC SYSTEMS CO., LTD.Jakarta Representative OfficeINDONESIATel : +62-21-572-4281 Fax : +62-21-572-4283FUJI ELECTRIC (MALAYSIA) SDN. BHD.MALAYSIATel : +60-4-403-1111 Fax : +60-4-403-1496FUJI ELECTRIC PHILIPPINES, INC.PHILIPPINESTel : +63-2-844-6183 Fax : +63-2-844-6196FUJI ELECTRIC SINGAPORE PRIVATE LTD.SINGAPORETel : +65-6535-8998 Fax : +65-6532-6866FUJI/GE PRIVATE LTD.SINGAPORETel : +65-6533-0010 Fax : +65-6533-0021

Printed on recycled paper

semiconductors - fuji electric

Documents