presentation - design, measurement and modeling of … measurement and modeling of ltcc embedded...

Design, Measurement and Modeling of LTCC Embedded Inductors and PCB

Balanced Devices

Prof. T.S. Horng (? ? ? )E.E. Dept., National Sun Yat-Sen Univ.Email: [email protected]

Outline

LTCC Embedded Inductors

PCB Balanced Devices

Conclusions

Design Trend

Planar Spiral Stacked Spiral Helical

≥ 222No. of Layers

HighestHigherLowQ(under the same Leff)

HighestHigherLowSRF(under the same Leff)

SmallestSmallerLargeArea(under the same Leff)

HelicalStacked SpiralPlanar SpiralLTCC Inductor

Measurement Techniques

Vector Network Analyzer

Test fixture

LTCCDevice

Reference plane after TRL calibration Electrode

Microstripline

LTCC Device(top view)

SG/GS-type probes

Test Fixture Microwave Probes

Test-Fixture Measurement vs. HFSS Simulation

Spiral inductor

The whole LTCC device

LTCC device on test fixture

Leff ≈ 6 nH

0 1 2 3 4 5 6 7 8 9 10

Frequency (GHz)

-50

-40

-30

-20

-10

0

dB

(S21

)

Mea#1Mea#2Mea#3Sim

0 1 2 3 4 5 6 7 8 9 10

Frequency (GHz)

-50

-40

-30

-20

-10

0

dB

(S11

)

Mea#1 Mea#2Mea#3Sim

Electrode

Microstrip line

Microwave-Probe Measurement vs. HFSS Simulation

Leff ≈ 6 nH

üSmaller areaüHigher SRFüHigher Q factorüBetter measured data repeatabilityüBetter agreement between simulation

and measurement

GS probe

0 1 2 3 4 5 6 7 8 9 10

Frequency (GHz)

-50

-40

-30

-20

-10

0

dB

(S21

)Mea#1Mea#2Sim

0 1 2 3 4 5 6 7 8 9 10

Frequency (GHz)

-50

-40

-30

-20

-10

0

dB

(S11

)

Mea#1 Mea#2Sim

A New Modified-T Model for Lossless Transmission Line

Lossless transmission lineZ0: Characteristic impedance

τ: Propagation delay

θ: Electrical length

pC

mL

sL sLsC

τ ,0Zθ

qωωπθ →=⇔→= 00

Equivalent π model

Equivalent Τ model

Equivalent modified-Τ model

≈?

ØIs it possible to create an equi-valent single-stage lumped model for a lossless transmission line having electrical length up to π ?

Derivation of Element Values of Equivalent Modified-T Model

+

+−+

+−

+−+

−

−

+−+

−+

+−+

+

−

−

p

smsms

ss

p

smsms

sm

p

smsms

sm

p

smsms

ss

Cj

CjLLjLLj

CjLj

Cj

CjLLjLLj

CjLj

Cj

CjLLjLLj

CjLj

Cj

CjLLjLLj

CjLj

Zj

Zj

Zj

Zj

2

)( )(

1

2

)( )(

1

2

)( )(

1

2

)( )(

1

cotcsc

csccot

00

00

ω

ωωω

ωω

ω

ωωω

ωω

ω

ωωω

ωω

ω

ωωω

ωω

θθ

θθ ≈

⇒ (F) (F),

(H) ) 1

41

( (H), ) 1

41

(

200

2020

π

ττπ

τπ

τ

ZC

ZC

ZLZL

ps

ms

≈≈

−≈+≈

Comparison of Bandwidth among Equivalent Models

π model

0 2 4 6 8 10 12 14 16

Freq(GHz)

-100

-90

-80

-70

-60

-50

-40

-30

-20

-10

0

S11

(dB

)

Lossless_TLModified-T_model

0 2 4 6 8 10 12 14 16

Freq(GHz)

-100

-90

-80

-70

-60

-50

-40

-30

-20

-10

0

S11

(dB

)

Lossless_TLPi_model

0 2 4 6 8 10 12 14 16

Freq(GHz)

-100

-90

-80

-70

-60

-50

-40

-30

-20

-10

0S11

(dB)

Lossless_TLT_model

T model

Modified-T model

θ = π θ = 2π θ = 3π

Transmission-line circuit

S11

(dB

)

S11

(dB

)

S11

(dB

)

ps 100 , 1000 =Ω= τZΩ 50 Ω 50


0 2 4 6 8 10 12 14 16

Freq(GHz)

-200

-150

-100

-50

0

50

100

150

200

S11

(phas

e)

Lossless_TLPi_model

0 2 4 6 8 10 12 14 16

Freq(GHz)

-200

-150

-100

-50

0

50

100

150

200

S11

(phas

e)Lossless_TLT_model

0 2 4 6 8 10 12 14 16

Freq(GHz)

-200

-150

-100

-50

0

50

100

150

200

S11

(ph

ase)


π modelModified-T model

T modelθ = π θ = 2π θ = 3π

Transmission-line circuitS

11 (p

has

e)

S11

(ph

ase)

S11

(ph

ase)

ps 100 , 1000 =Ω= τZΩ 50 Ω 50


0 2 4 6 8 10 12 14 16

Freq(GHz)

-20

-18

-16

-14

-12

-10

-8

-6

-4

-2

0

S21

(dB

)


0 2 4 6 8 10 12 14 16

Freq(GHz)

-20

-18

-16

-14

-12

-10

-8

-6

-4

-2

0

S21

(dB

)

Lossless_TLPi_model

0 2 4 6 8 10 12 14 16

Freq(GHz)

-20

-18

-16

-14

-12

-10

-8

-6

-4

-2

0

S21

(dB

)

Lossless_TLT_model


T model

S21

(dB

)S

21 (d

B)

S21

(dB

)

θ = π θ = 2π θ = 3π

Transmission-line circuit

ps 100 , 1000 =Ω= τZΩ 50 Ω 50


0 2 4 6 8 10 12 14 16

Freq(GHz)

-200

-150

-100

-50

0

50

100

150

200

S21

(ph

ase)


0 2 4 6 8 10 12 14 16

Freq(GHz)

-200

-150

-100

-50

0

50

100

150

200

S21

(phas

e)

Lossless_TLPi_model

0 2 4 6 8 10 12 14 16

Freq(GHz)

-200

-150

-100

-50

0

50

100

150

200

S21

(phas

e)

Lossless_TLT_model


T modelθ = π θ = 2π θ = 3π

Transmission-line circuitS

21 (p

has

e)S

21 (p

has

e)

S21

(ph

ase)

ps 100 , 1000 =Ω= τZΩ 50 Ω 50

Distributed Modified-T Model

τ ,0Z

πθ 3=

3/pC

3/mL

3/sL 3/sL3/sC

3/pC

3/mL

3/sL 3/sL3/sC

3/pC

3/mL

3/sL 3/sL3/sC

≈

3-stage modified-T model3π-long transmission line

0 2 4 6 8 10 12 14Frequency (GHz)

-180

-120

-60

0

60

120

Ph

ase(

S11

)

Lossless_TL 3-stage modified-T_model

0 2 4 6 8 10 12 14Frequency (GHz)

-3

-2.5

-2

-1.5

-1

-0.5

0

dB

(S21

)


0 2 4 6 8 10 12 14Frequency (GHz)

-300

-180

-60

60

180P

has

e(S

21)


0 2 4 6 8 10 12 14Frequency (GHz)

-60

-50

-40

-30

-20

-10

0

dB

(S11

)


θ = π θ = 2π θ = 3π

Modified-T Models for Spiral Inductors

Leff ≈ 6 nH

pC

mL

sL sLsC

pL

1pC 1pL

pR

gCgL0 2 4 6 8 10 12 14

Frequency (GHz)

-35

-30

-25

-20

-15

-10

-5

0

dB

(S21

)

zero

1st by-pass 2nd by-pass

Ground Resonance

Shunt Cap

loss

Modified-T model sCsC effLsR

pCConventional π model

Comparison of Bandwidth between Two Inductor Models

0 2 4 6 8 10 12 14 16

Freq(GHz)

-45

-40

-35

-30

-25

-20

-15

-10

-5

0

S22

(dB

)

measurementmodeling

0 2 4 6 8 10 12 14 16

Freq(GHz)

-200

-150

-100

-50

0

50

100

150

200

S22

(ph

ase)

measurementmodeling

0 2 4 6 8 10 12 14 16

Freq(GHz)

-35

-30

-25

-20

-15

-10

-5

0

S11

(dB

)

measurementpi_model

0 2 4 6 8 10 12 14 16

Freq(GHz)

-200

-150

-100

-50

0

50

100

150

200

S11

(ph

ase)

measurementpi_modelmeasurementPi_model

measurementPi_model

measurementModified-T_model


S11

(dB

)

S11

(dB

)

S11

(ph

ase)

S11

(ph

ase)

Conventional π modelModified-T model

Comparison of Bandwidth between Two Inductor Models

0 2 4 6 8 10 12 14 16

Freq(GHz)

-30

-25

-20

-15

-10

-5

0

S21

(dB

)

measurementmodeling

0 2 4 6 8 10 12 14 16

Freq(GHz)

-150

-100

-50

0

50

100

150

200

S21

(ph

ase)

measurementmodeling

0 2 4 6 8 10 12 14 16

Freq(GHz)

-30

-25

-20

-15

-10

-5

0

S21

(dB

)

measurementpi_model

0 2 4 6 8 10 12 14 16

Freq(GHz)

-200

-150

-100

-50

0

50

100

150

200

S21

(ph

ase)

measurementpi_model

measurementPi_model

measurementPi_model



S21

(dB

)

S21

(dB

)

S21

(ph

ase)

S21

(ph

ase)

Conventional π modelModified-T model

Outline

LTCC Embedded Inductors

PCB Balanced Devices

Conclusions

Measurement Systems for Multiportand Mixed-Mode S-Parameters

ØPure Mode Network AnalyzerØMultiport Network Analyzer Using

Full-N Port CalibrationØTwo-Port Network Analyzer Using

Renormalization Techniques

Port Termination Problem

DUT

Port 3 terminated

1a

1b

2a

2b

3a 3b

Port 1 Port 2

Port 3

3332321313

3232221212

3132121111

aSaSaSb

aSaSaSb

aSaSaSb

++=++=

++=

Three-port Network Three-port S parameters

Reflection due to port termination

3

33 b

a=Γ

Measured two-port S parameters

Γ−Γ

+Γ−Γ

+

Γ−Γ

+Γ−Γ

+=

=

333

3322322

333

3312321

333

3321312

333

3311311

p3t22

p3t21

p3t12

p3t11p3t

11

11

][

SSS

SSSS

S

SSS

SSSS

S

SS

SSS

Partial Renormalization

[ ] [ ] [ ]( ) [ ] [ ]( ) [ ] [ ][ ]( ) [ ] [ ]( )

2,1for , where

00

00

11

=+−

=

−Γ−Γ−−=′ −−

kZZ

G

SUSUSSUS

kk

kkk ζ

ζ

DUT

Port 3 terminated

1a

1b

2a

2b

3a 3b

Port 1 Port 2

Port 3

1oZ 2oZ

3oZ

DUT

Port 3 terminated

1a

1b

2a

2b

3a 3b

Port 1 Port 2

Port 3

1oζ 2oζ

3oζ

⇒][S ][ 'S

Renormalization Transforms

=

=

= p1t

33p1t32

p1t23

p1t22p1t

p2t33

p2t31

p2t13

p2t11p2t

p3t22

p3t21

p3t12

p3t11p3t ][ , ][ , ][

SS

SSS

SS

SSS

SS

SSS

ØThree partial 2-port S-parameter measurements

ØAfter partial renormalizations

=

=

= m3

33m332

m323

m322m3

m233

m231

m213

m211m2

m122

m121

m112

m111m1 ][ , ][ , ][

SS

SSS

SSSSS

SSSSS

[ ] [ ] m233

m332

m231

m323

m122

m121

m213

m112

m111

S

SSS

SSSSSS

S ⇒

=′

ØConstruct the S matrix of the three-port network normalized to [ζ0] and then transform it back to the S matrix normalized to [Z0] :

renormalized to

Determination of [ζ0]

Ø Use TRL Calibration

Øi

ii Z

Γ−Γ+

=11

00ζ

DC-Block Branch-Line Coupler

000 2ZZZ ao

ae =−

000 2ZZZ ao

ae =−

Port 1 Port 2

Port 4 Port 3

Design guide Photo of component

Comparison between Ensemble Simulation and Measurement

1E+009 2E+009 3E+009 4E+009 5E+009 6E+009 7E+009

Frequency

-30

-25

-20

-15

-10

-5

0

S11

(d

B)

EnsembleMeasured

1E+009 2E+009 3E+009 4E+009 5E+009 6E+009 7E+009

Frequency

-30

-25

-20

-15

-10

-5

0

S21

(dB

)

EnsembleMeasured

1E+009 2E+009 3E+009 4E+009 5E+009 6E+009 7E+009

Frequency

-45

-40

-35

-30

-25

-20

-15

-10

-5

0

S31

(dB

)

EnsembleMeasured

1E+009 2E+009 3E+009 4E+009 5E+009 6E+009 7E+009

Frequency

-35

-30

-25

-20

-15

-10

-5

0

S41

(d

B)

EnsembleMeasured

Impedance Transform Branch-Line Coupler

Design guide Photo of component

][) ][ ][ ( ) ][][ (][] [ :][][

][) ][][ () ][][ ( ][][ :][][

01

001-

0''

10

10

ξξξξ −

−−

+−=⇒

+−=⇒

ZZSSZ

ZSUSUZZZS

Generalized S-parameter Transform

4λ

4λ

Port 1Port 2

Port 4 Port 3

1Z

1Z

2Z

2Z

1Z 2Z

221ZZ

221ZZ

50O 25O

Comparison between Ensemble Simulation and Measurement

1E+009 2E+009 3E+009 4E+009 5E+009 6E+009 7E+009

Frequency

-40

-35

-30

-25

-20

-15

-10

-5

0

S11

(dB

)

EnsembleMeasured Before transformMeasured After transform

1E+009 2E+009 3E+009 4E+009 5E+009 6E+009 7E+009

Frequency

-40

-35

-30

-25

-20

-15

-10

-5

0

S22

(dB

)


1E+009 2E+009 3E+009 4E+009 5E+009 6E+009 7E+009

Frequency

-30

-25

-20

-15

-10

-5

0

S21

(dB

)


1E+009 2E+009 3E+009 4E+009 5E+009 6E+009 7E+009

Frequency

-50

-45

-40

-35

-30

-25

-20

-15

-10

-5

0

S41

(d

B)


Mixed-Mode S parameters

Common-Mode Rejection Ratio

21

21CMRRCC

DDSS

=

Mixed-Mode S Matrix

11DDS 12DDS

21DDS 22DDS

11CDS 12CDS

21CDS 22CDS

11DCS 12DCS

21DCS 22DCS

11CCS 12CCS

21CCS 22CCS

Port 1 Port 2 Port 1 Port 2Differential-Mode Common-Mode

Stimulus

Port

1P o

rt2

Por t

3Po

rt4

Diff

eren

t ial- M

ode

Com

mo n

-Mod

e

Resp

onse

[ ][ ]

[ ] [ ][ ] [ ]

[ ][ ] [ ] [ ]

[ ]

=

=

C

Dmm

C

D

CCCD

DCDD

C

D

aa

Saa

SSSS

bb

Mixed-Mode Transform

−

−

=

4

3

2

1

2

1

2

1

110000111100

0011

21

aaaa

aaaa

C

C

D

D

−

−

=

4

3

2

1

2

1

2

1

110000111100

0011

21

bbbb

bbbb

C

C

D

D

]][[][ semm aMa =

]][[][ semm bMb =

1]][][[][ −= MSMS semm

Lange-Type Marchand Balun

Input2λ

Output

Lange Coupler

Lange Coupler

InputOutput

11S

DDS

1CS CDS

CS 1

DCS

CCS

Port 1Balanced Port 2

Stimulus

Res

pons

e

Differential Common

Port

1B

alan

ced

Port

2D

iffer

entia

lC

omm

on

DS1

1DS

[ ]

−=110110

002

21

M

Planar type Lange type

Design guide

,)2

1( 1in0

out0 −+=

Z

ZC

Photo of component

Mixed-mode [S]

⇒

Comparison between HFSS Simulation and Measurement

1E+009 2E+009 3E+009 4E+009 5E+009 6E+009 7E+009 8E+009

Frequency

-30

-25

-20

-15

-10

-5

0

SD

1 &

S1D

(dB

) SimulatedMeasure

1E+009 2E+009 3E+009 4E+009 5E+009 6E+009 7E+009 8E+009

Frequency

-2.5

-2

-1.5

-1

-0.5

0

SC

C (

dB

)

SimulatedMeasured

1E+009 2E+009 3E+009 4E+009 5E+009 6E+009 7E+009 8E+009

Frequency

-35

-30

-25

-20

-15

-10

-5

0

SC

1 &

S1C

(d

B)

SimulatedMeasured

1E+009 2E+009 3E+009 4E+009 5E+009 6E+009 7E+009 8E+009

Frequency

-25

-20

-15

-10

-5

0

S11 (dB

) SimulatedMeasured

S11

(d

B)

Comparison between HFSS Simulation and Measurement

1E+009 2E+009 3E+009 4E+009 5E+009 6E+009 7E+009 8E+009

Frequency

-30

-20

-10

0

10

20

30

CM

RR

SimulatedMeasuredSD1

1E+009 2E+009 3E+009 4E+009 5E+009 6E+009 7E+009 8E+009

Frequency

-25

-20

-15

-10

-5

0

5

SD

D (d

B)

SimulatedMeasured

1E+009 2E+009 3E+009 4E+009 5E+009 6E+009 7E+009 8E+009

Frequency

-2.5

-2

-1.5

-1

-0.5

0

SC

C (d

B)

SimulatedMeasured

1E+009 2E+009 3E+009 4E+009 5E+009 6E+009 7E+009 8E+009

Frequency

-45

-40

-35

-30

-25

-20

-15

-10

SC

D &

SD

C (

dB

) SimulatedSimulated

Semi-Automatic Measurement System

VNA setup via Vee

Measure, renormalize and Transform

Single-ended S parameters

Mixed-Mode S Parameters

Application to On-Wafer Measurement

$JLOHQW

$JLOHQW

91$

$JLOHQW

Our proposed System System made by NIST

Conclusions

Ø Various types of LTCC Embedded Inductors have been designed and measured. Simulated results agree with measurements quite well.

Ø A new modified-T equivalent circuit has been proposed to model LTCC inductors over an extremely large bandwidth successfully.

Ø Very cost-effective multiport network analyzer system based on renormalization techniques has been developed to measure balanced devices.

Ø Several examples of multiport and balanced devices on PCB have been designed and measured. Comparison between simulation and measurement shows excellent agreement.

References

1. K. Lim, et. al., “RF-system-on-package (SOP) for wireless communications,” IEEE Microwave Magazine, pp. 88-99, March 2002.

2. J.C. Tippet and R.A. Speciale, “A rigorous technique for measuring the scattering matrix of a multiport device with a 2-port network analyzer,” IEEE Transactions on Microwave Theory and Techniques, pp. 661-666, May 1982.

3. L.-Q. Yang, Design and modeling of embedded inductors and capacitors in low-temperature cofired ceramic technology, Master’s Thesis, National Sun Yat-Sen University at KaohsiungTaiwan, 2002.

4. D.-C. Tsai, Measurement of balanced devices using vector network analyzers, Master’s Thesis, National Sun Yat-Sen University at Kaohsiung, Taiwan, 2002.

presentation - design, measurement and modeling of … measurement and modeling of ltcc embedded...

Documents