relative permittivity variation surrounding pcb via hole
TRANSCRIPT
1 SPI2008 Relative Permittivity Variation Surrounding PCB Via Hole Structures
Relative Permittivity Variation Surrounding PCB Via Hole StructuresSPI2008 Avignon France May 12-15, 2008Lambert [email protected]
2 SPI2008 Relative Permittivity Variation Surrounding PCB Via Hole Structures Lambert Simonovich
IntroductionIntroduction
Present IC technology advancements are allowing Present IC technology advancements are allowing data rates beyond 6 data rates beyond 6 Gb/sGb/s
PCB through hole via parasitics becoming more of a PCB through hole via parasitics becoming more of a factor affecting BER performancefactor affecting BER performance
Accurate via modeling and topology simulation are a Accurate via modeling and topology simulation are a mustmust
All models and methodologies require defining All models and methodologies require defining relative permittivity (relative permittivity (εεrr ) or dielectric constant () or dielectric constant (DkDk) of ) of material surrounding the via hole structurematerial surrounding the via hole structure
3 SPI2008 Relative Permittivity Variation Surrounding PCB Via Hole Structures Lambert Simonovich
PCB Fabrication OverviewPCB Fabrication Overview
Conventional PCB laminates are fabricated with Conventional PCB laminates are fabricated with a weave of Ea weave of E--glass fiber yarns (glass fiber yarns (Dk 6.6 @ 1MHzDk 6.6 @ 1MHz) ) and resin (and resin (Dk 3.2 @ 1MHzDk 3.2 @ 1MHz) [1]) [1]
Effective Effective DkDk is a function of glass to resin ratio is a function of glass to resin ratio of laminates used in the PCB stackupof laminates used in the PCB stackup
For FRFor FR--4 PCBs an effective 4 PCBs an effective DkDk of 4.3 is often of 4.3 is often used for trace etch impedance calculationsused for trace etch impedance calculations
4 SPI2008 Relative Permittivity Variation Surrounding PCB Via Hole Structures Lambert Simonovich
N4000-6 TM Dielectric Properties Table [1], [2]
Style Glass / Laminate ThicknessRatio(mils)
ResinContent(%)
YarnCount(threads/inch)
YarnPitch
(mils)
106 1.4/2.0 69.0 56x56 17.9x17.9
1080 2.3/3.0 62.0 60x40 16.7x21.3
2113 2.9/4.0 54.5 60x56 16.7x17.9
2116 3.8/4.0 43.0 60x58 16.7x17.2
7628 6.8/8.0 44.4 44x32 22.7x31.3
Resin Fiberglass Weave
Fiberglass weaves vs. resin content illustration. Smaller glass diameter and thread count results in higher resin content, while larger glass diameter and thread count results in lower resin content.
2113106 7628
Fiberglass Weave Fiberglass Weave vsvs Resin ContentResin Content
5 SPI2008 Relative Permittivity Variation Surrounding PCB Via Hole Structures Lambert Simonovich
FiberglassWarp/Fill
Resin
Via 1 Via 2
Dielectric Property Between ViasDielectric Property Between Vias
6 SPI2008 Relative Permittivity Variation Surrounding PCB Via Hole Structures Lambert Simonovich
++++
----
E-field
Higher glass to resin ratio between charges => Higher effective
dielectric constant
Higher glass to resin ratio between charges => Higher effective
dielectric constant
VSVS
ViasVias EtchEtch
E-field Resin
Fiberglass
Laminate Weave EffectLaminate Weave Effect
Lower glass to resin ratio between charges => Lower effective
dielectric constant
Lower glass to resin ratio between charges => Lower effective
dielectric constant
7 SPI2008 Relative Permittivity Variation Surrounding PCB Via Hole Structures Lambert Simonovich
Laminate Weave Effect of Via Hole StructureLaminate Weave Effect of Via Hole Structure
8 SPI2008 Relative Permittivity Variation Surrounding PCB Via Hole Structures Lambert Simonovich
Case Study ObjectiveCase Study Objective
To study the effect of relative permittivity To study the effect of relative permittivity surrounding a PCB via hole structuresurrounding a PCB via hole structure
Validate transmission line via model correlation Validate transmission line via model correlation to measured resultsto measured results
9 SPI2008 Relative Permittivity Variation Surrounding PCB Via Hole Structures Lambert Simonovich
PCB Coupon Under Test
Device Under Test
12
43
Agilent NS230A VNA
Via Pair1A, 1B
Via Pair2A, 2B
Test SetupTest Setup
10 SPI2008 Relative Permittivity Variation Surrounding PCB Via Hole Structures Lambert Simonovich
6”8-9-8 (mils) Diff Pair
Layer 2NELCO 4000-13TM
Stub270 mils
Via14 mils
Via1AVia1B
Via2AVia2B
Test Coupon TopologyTest Coupon Topology
11 SPI2008 Relative Permittivity Variation Surrounding PCB Via Hole Structures Lambert Simonovich
Data and ResultsData and Results
Resonant frequency nulls due to the via stub length occur at odd harmonics of the ¼ wave frequency null
Effective relative permittivity (a.k.a.
calculated by [5];
= 6.46
Where: c = Speed of light (1.18E10 inches/sec); Stub_Length (0.270) inches; f = ¼ wave frequency (13.00GHz/3)
dielectric constant Dk) can be
2
*_*4 ⎥⎦
⎤⎢⎣
⎡=
flengthStubcDk
12 SPI2008 Relative Permittivity Variation Surrounding PCB Via Hole Structures Lambert Simonovich
StriplineStriplineLayerLayer
¼¼ WaveWaveFrequencyFrequency
(GHz)(GHz)
33rdrd
HarmonicHarmonicFrequencyFrequency
(GHz)(GHz)
Via/StubVia/StubLengthsLengths(Mils)(Mils)
Calculated Calculated DkDk
0202 4.34.3 13.013.0 14/269.914/269.9 6.466.46
1010 6.36.3 18.918.9 105.7/178.1105.7/178.1 6.906.90
2020 18.418.4 N/AN/A 219.6/64.3219.6/64.3 6.206.20
L2
L10
Stub = 269.9 milsStub = 178.1 mils
L20
Stub = 64.3 mils
Via = 14 milsVia = 105.7 mils
Via = 219.6 mils
Data and ResultsData and Results
13 SPI2008 Relative Permittivity Variation Surrounding PCB Via Hole Structures Lambert Simonovich
Differential Pair Via ModelDifferential Pair Via Model
Modeled as differential pair transmission lines
Analogous to a twin-round-wire wire structure
Zvstub
Zv
Dk
⎟⎟
⎠
⎞
⎜⎜
⎝
⎛−⎟
⎠⎞
⎜⎝⎛+
Ω= 1
22ln120 2
rs
rs
DkZdiff
Where: Zdiff = Differential impedance; Dk = Dielectric constant; s = center-to-center space of vias; r = radius of drill [6]
s
r
2ZdiffZv =
Odd mode via impedance (Zv)
14 SPI2008 Relative Permittivity Variation Surrounding PCB Via Hole Structures Lambert Simonovich
Via Impedance CalculationVia Impedance Calculation
⎟⎟
⎠
⎞
⎜⎜
⎝
⎛−⎟
⎠⎞
⎜⎝⎛+
Ω= 1
22ln120 2
rs
rs
DkZdiff
⎟⎟
⎠
⎞
⎜⎜
⎝
⎛−⎟
⎠⎞
⎜⎝⎛+
Ω= 1
2859
2859ln
4.6120 2
Ω= 65
Ω==== 5.32265
2ZdiffZoddZv
15 SPI2008 Relative Permittivity Variation Surrounding PCB Via Hole Structures Lambert Simonovich
ADS Simulation Correlation TopologyADS Simulation Correlation Topology
Measured
Model
Probe
Via 1A,B
Via 2A,B
Probe
Diff Pair Etch
16 SPI2008 Relative Permittivity Variation Surrounding PCB Via Hole Structures Lambert Simonovich
ADS Simulation Correlation ResultsADS Simulation Correlation ResultsDk = 6.4 Dk = 3.8
-- Measured -- Simulated-- Measured -- Simulated
17 SPI2008 Relative Permittivity Variation Surrounding PCB Via Hole Structures Lambert Simonovich
Conclusions Conclusions
Due to the fiberglass weave effect surrounding Due to the fiberglass weave effect surrounding differential pair vias, the relative permittivity is differential pair vias, the relative permittivity is higher than the dielectric surrounding wiring etch higher than the dielectric surrounding wiring etch within the same PCB stackupwithin the same PCB stackup
Using the correct value of dielectric constant Using the correct value of dielectric constant has shown better accuracy for via modeling has shown better accuracy for via modeling during topology simulationsduring topology simulations
18 SPI2008 Relative Permittivity Variation Surrounding PCB Via Hole Structures Lambert Simonovich
ReferencesReferences
1. Scott Mc Morrow, Chris Heard, “The Impact of PCB Laminate Weave on the Electrical Performance of Differential Signaling at Multi-Gigabit Data Rates”, Design Con 2005.
2. Park Electrochemical Corp. , http://www.parkelectro.com3. Gustavo Blando, Jason R. Miller, Istvan Novak, “Losses Induced by Asymmetry in Differential
Transmission Lines”, DesignCon 2007.4. Howard Johnson, Martin Graham, “High Speed Digital Design, a Handbook of Black Magic”.5. Howard Johnson, Martin Graham, “High Speed Signal Propagation, Advanced Black Magic”.6. Eric Bogatin, “ Signal Integrity Simplified”7. Stephen C. Thierauf , “High-Speed Circuit Board Signal Integrity”8. Lee Ritchey, “"Right the First Time, A Practical Handbook on High Speed PCB and System
Design, Volume 1 & 2"9. Taras Kushta, Kaoru Narita, Tomoyuki Kaneko, Takanori Saeki, and Hirokazu Tohya,
“Resonance Stub Effect in a Transition From a Through Via Hole to a Stripline in Multilayer PCBs”, IEEE MICROWAVE AND WIRELESS COMPONENTS LETTERS, VOL. 13, NO. 5, MAY 2003.
10. Miroslav Pajovic, Jinghan Yu, and Dragan Milojkovic, “Analysis of Via Capacitance in Arbitrary Multilayer PCBs”, IEEE TRANSACTIONS ON ELECTROMAGNETIC COMPATIBILITY, VOL. 49, NO. 3, AUGUST 2007.