Test IV The “Townsend” Test The Townsend Test

Test IVTest IVTest IV Test IV The “Townsend” TestThe “Townsend” Test

Herb TownsendHerb Townsend

The Townsend TestThe Townsend TestTownsend Corrosion ConsultantsTownsend Corrosion Consultants

11 Townsend Corrosion Consultants

ScopeScopepp

• Fundamental definitions and concepts• Fundamental definitions and concepts

• Review of previous studiesReview of previous studies

• Test IV description and purposep p p

• Current status of testing


Specific RequirementsSpecific RequirementsFor Hydrogen EmbrittlementFor Hydrogen Embrittlementy gy g


Sustained, HighSustained, High--Tensile LoadTensile Load, g, g

• Hydrogen embrittlement requires a sustained, high tensile loadhigh tensile load.

• Cracking is a slow process• Cracking is a slow process

• Time needed for hydrogen to diffuseTime needed for hydrogen to diffuse to crack site

• Not detected at normal test speeds


Sources Of HydrogenSources Of Hydrogeny gy g

• Process HydrogenAcid cleaning– Acid cleaning

– Galvanizing– Phosphatingp g– Electroplating

• Corrosion HydrogenCorrosion Hydrogen – Cathodic protection– Galvanic corrosion of zinc


Fundamental ConceptsFundamental Conceptspp

• There is a critical threshold load for SCC

• In terms of fracture mechanics, the critical value is Kis KIscc

• KI decreases with increasing hardnessKIscc decreases with increasing hardness

• KIscc decreases with decreasing electrochemical Iscc gpotential


Key DefinitionsKey Definitionsyy

• Hydrogen embrittlement (HE) — loss of cohesion between atoms caused by H in steelbetween atoms caused by H in steel

• Internal hydrogen embrittlement (IHE) — HE caused by H entering the steel during processing also calledH entering the steel during processing, also called process hydrogen

• External hydrogen embrittlement (EHE) — HE caused• External hydrogen embrittlement (EHE) — HE caused by H entering from the environment

• In the case of exposure to water EHE is often called• In the case of exposure to water, EHE is often called corrosion hydrogen embrittlement, or stress corrosion cracking (SCC)


Electrochemical ReactionsElectrochemical ReactionsOn Galvanized SteelOn Galvanized Steel

• Galvanized coatingZn Zn+2 + 2eZn Zn+2 + 2e-

• Exposed steel2e- + 2H2O 2OH- + H2

• In solutionZn+2 + 2OH- Zn(OH)2 (white rust)

• Overall reactionOverall reactionZn + 2H2O Zn(OH)2 + H2


Results of Results of Previous StudiesPrevious StudiesPrevious StudiesPrevious Studies

1973 1973 —— Large research program conducted by Large research program conducted by BattelleBattelleg p g yg p g yfor the Research Council on Structural Connectionsfor the Research Council on Structural Connections


Battelle Bolt Study Battelle Bolt Study —— 19731973yy

• A325 (HRC 25-34) bolts

• A490 (HRC 33-39) bolts

• Black, ungalvanized

• Galvanized

• Laboratory and outdoor tests


Battelle Bolt Study Battelle Bolt Study 1973 Results1973 Results

• Bare and galvanized A325 bolts — no failures

• Bare A490 bolts — no failures

• Galvanized A490 — numerous failures

• Battelle suggested that failures were due to EHE caused by galvanic activity of galvanized y g y gcoatings


1975 1975 —— Fracture Mechanics TestsFracture Mechanics TestsPrecrackedPrecracked 11--in SQ Bars (ASTM E 1681)in SQ Bars (ASTM E 1681)( )( )


19751975Fracture Mechanics TestsFracture Mechanics Tests

• 4140 1-in. bars, HRC 32 to 48

• Notched cantilever-beam specimens

• Fatigue precracked• Fatigue precracked

• Bare and zinc-coated

• Rapid-load tests in air for baseline toughness

Sl l d t t i i f IHE• Slow-load tests in air for IHE

• Slow-load tests in 3.5% salt solution for EHE


FpcFpc Specimen and Specimen and the the KiesKies EquationEquationqq

The stress intensity factor, K1, is calculated from Equation 1, the Kies equation as given by Brown [8, 13] and modified to account for side grooves:

2/32/1

2/13

3

1

112.4

WBB

MK

(1)

In equation 1, M is the applied moment, W is the overall specimen height, B is the overall

WBB n

the overall specimen height, B is the overall specimen width, Bn is the net specimen width

at the side grooves, and α is defined:

a1

Where a is the overall crack length, notch depth included.

w1


g , p

Dry, SlowDry, Slow--Load Tests Galvanized Load Tests Galvanized Bars In Air For IHEBars In Air For IHE


Greater Susceptibility to EHEGreater Susceptibility to EHEp yp y


Effect of HardnessEffect of Hardnesson EHE Thresholdon EHE Threshold

70Kscc vs Hardness

dKscc/dHRC = -3.8 50

60

KSI

in^0

.5

30

40

y Fa

ctor

, K

20

30

ss In

tens

ity

0

10

30 35 40

Stre

s

H d HRC


Hardness, HRC

Effect of Electrochemical Effect of Electrochemical Potential on EHE ThresholdPotential on EHE Threshold

100

Kscc vs E (HRC=35)

80

90

100

SI in

^0.5

Steel

dK/dE = - 0.134 Ksi-in0.5/mv

60

70

y Fa

ctor

, KS

40

50

ss In

tens

ity

Zi

20

30

0.6 0.65 0.7 0.75 0.8 0.85 0.9 0.95 1 1.05 1.1

Stre

s Zinc


Potential, -Vsce

Townsend 1975Townsend 1975Summary of ResultsSummary of Resultsyy

• Dry slow-step load Loss of strength due to hydrogen from galvanizing processhydrogen from galvanizing process

• Wet slow-step load Larger loss of strength• Wet slow-step load Larger loss of strength due to hydrogen from galvanic corrosion of zinc coatingg

• Led to prohibition of galvanizing for A490 bolts and HRC 32 maximum for galvanized A325


Model for Hydrogen Embrittlement Model for Hydrogen Embrittlement of Highof High--Strength Galvanized BoltsStrength Galvanized Boltsgg gg

1. Galvanized coatings crack at the thread roots of A490 bolts with high pretensionof A490 bolts with high pretension (using turn-of-the-nut or equivalent)

2. Hydrogen is galvanically deposited on bare steel exposed at these cracks

3. If the load is greater than the threshold for hydrogen embrittlement, slow crack growth begins

4. The crack grows until a critical length is


4. The crack grows until a critical length is reached, and the bolt breaks

Test IV ObjectivesTest IV Objectivesjj

• Determine the threshold load levels for• Determine the threshold load levels for hydrogen embrittlement of full-size galvanized A354 BD rods exposed to salt waterA354 BD rods exposed to salt water

• Use the results as a guide to identifying safe g y gload levels and if necessary, suggest remedial action for galvanized fasteners on the SAS


Application of Application of PrecrackedPrecracked Results Results to Rods is a Significant Challengeto Rods is a Significant Challengeg gg g

• Fracture mechanics equations for threaded sections are approximationssections are approximations

• Uniform hardness for bars vs. hardness profiles for rods

• The 1975 analysis for bolts assumed turn-of-The 1975 analysis for bolts assumed turn ofthe-nut tightening, which generally leads to stresses greater than yield strength

• Rods on the SAS are tensioned by jacking to lower-than-yield stress levels


lower than yield stress levels

BuecknerBueckner Equation Equation for Stress Intensity Factor, K, of a for Stress Intensity Factor, K, of a Circumferentially Notched Round BarCircumferentially Notched Round Baryy

K= σ (πD1/2) f(d/D)Where:

d = diameter of notchσ = tensile stressD = diameter of barD = diameter of barf(d/D) is a constant

K ( t t) 0 80(%F ) fK = σ x (constant) ~ 0.80(%Fu) for 3-inch rods


Differences Between Differences Between 1975 Lab Tests and Test IV1975 Lab Tests and Test IV

• Cantilever bending vs. tensioning for rods

• No fatigue precrack for Test IV

• K calculations for rods are approximations

• Uniform HRC vs. HRC gradients

• 1 in bars vs 2 in to 4 in rods• 1-in bars vs. 2-in to 4-in rods

• Low-iron vs. high-iron coatings for rods


Low iron vs. high iron coatings for rods

Test IV Common Design FeaturesTest IV Common Design Featuresgg

• Slow-step, load increases to determine thresholdthreshold

• Exposure to 3 5% NaCl solution• Exposure to 3.5% NaCl solution

• Measurement of electrochemical potentialMeasurement of electrochemical potential and pH


Test IV Design FeaturesTest IV Design Featuresgg

• Step load increases applied by hydraulic tensioningtensioning

• Displacement measurementDisplacement measurement


Test IV Additional Design Test IV Additional Design FeaturesFeatures

• Loads measured by strain gages

• Full-size rods

• Intentional coating holidays

• Continuous monitoring of temperatures

• Acoustic emission monitoring


Test IV Load ScheduleTest IV Load ScheduleLoad, % Fu K, KSI-in1/2 Hold time , days

30 24 230 24 240 32 250 40 255 44 260 48 265 52 270 56 275 60 280 64 2


85 68 6Fu = minimum specified ultimate strength

Test IV Average Load Rate Compared Test IV Average Load Rate Compared to 1975 Testing ksito 1975 Testing ksi--inin1/21/2 per Hourper Hourgg pp

1975 Test IV

0.2 ksi-in1/2/hr 0.1 ksi-in1/2/hr


Test IV Rig for FullTest IV Rig for Full--length length Bridge Rods Bridge Rods gg


Test IV Phase 1 (2010 E2 Rods)Test IV Phase 1 (2010 E2 Rods)( )( )

• 2, Full-length 3-in dia shear key bottom anchor rodsanchor rods

• 2 Full-length 3-in dia bearing bolts bottom• 2, Full-length 3-in dia bearing bolts bottom housing rods

• Tests completed — analysis in progress


Setting Up Test IV RigSetting Up Test IV Rigg p gg p g


Tent Enclosure During TestingTent Enclosure During Testingg gg g


After FractureAfter Fracture


Fracture SurfaceFracture Surface


Test IV Preliminary ResultTest IV Preliminary ResultHighlightsHighlightsg gg g

4, 3-inch 2010 rods fractured at loads of,0.80 Fu, 0.85 Fu, 0.85 Fu, and >0.85 Fu


Test IV SlowTest IV Slow--Step Step Load ApplicationLoad Applicationpppp

FRACTURE AT 0 80 FAT 0.80 Fu


Test IV Preliminary Results Test IV Preliminary Results yy

• Scanning electron microscopy shows evidence of intergranular cracking on the fracture facesof intergranular cracking on the fracture faces for 3 out of 4 rods

• This is consistent with failure by hydrogen embrittlement

• The fracture morphology was similar to that of the failed 2008 rods


Test IV Results to DateTest IV Results to DateIntergranularIntergranular Cracking X500Cracking X500gg gg


Test IV Test IV Results to DateResults to Date

• 2 of 4 rods broke at end without coating holiday

• This suggests that coating defects are not critical in the mechanism of fracturecritical in the mechanism of fracture


Galvanized Coating CrackGalvanized Coating CrackRod 4 X500Rod 4 X500


Test IV Results to DateTest IV Results to Date

• Electrochemical potentials were 0.15 V to 0.20 V less negative than expected for typicalless negative than expected for typical galvanized

• Galvanized rod coatings were generally higher in iron content

• The high iron content accounts for the reduced galvanic activity of the Test IV rods


Test IV Results to DateTest IV Results to Date


Test IV Phase 2Test IV Phase 2

• 2-in dia. Upper Bearing Rod

• 3-in dia. Tower Base Anchor Rod

• 4 in dia Tower Saddle Tie Rod• 4-in dia. Tower Saddle Tie Rod

• 3.5-in dia. Main Cable Anchor Rods – with rolled threads

• 3.5-in dia. Main Cable Anchor Rods – with cut threads

• Tests completed — analysis in progress


Tests completed analysis in progress

Test IV Phase 3Test IV Phase 3

• 3-in 2008 shear key bottom anchor rods – (2 Samples)( p )

• Test set-up in progress


Test IV Phase 4 Test IV Phase 4 ——Planning PhasePlanning Phasegg

Plan to Test:

– 3-in dia. shear key anchor replacement rods (2013 upgrade)

– Black and galvanized

– Effect of Second Heat treatmentEffect of Second Heat treatment


Thank youThank youThank youThank you


Test IV The “Townsend” Test The Townsend Test

Documents

Test IV The “Townsend” Test The Townsend Test