Failure Analysis Case History No.

001

Industry: Gas Treating

Specimen Description: Ball Valve Cover Capscrews
Material: ASTM A 574
Environment: Atmospheric
Background: A 1-inch ball valve completely separated without warning, releasing a
natural gas vapor cloud. Initial inspection of the failed valve showed that it
separated due to failure of the two body capscrews. Both capscrews failed
in a corroded thread root.
Time in Service: Unknown, but years.
Findings: Laboratory examination of the two failed capscrews showed that they had
corroded with brittle appearing fracture surfaces. No secondary cracks
below the fracture were observed. Hardness measurements showed the
capscrews averaged HRC 44, representative of that specified in A574.
SEM analysis showed that the fractures were intergranular and contained
secondary cracking, typical of that caused by hydrogen embrittlement.

Hydrogen Embrittlement of Valve Capscrew Fasteners

Body capscrews from a ball valve showing hydrogen embrittlement fracture. Exposed to atmospheric conditions in a gulf
coast environment, they corroded and fractured, separating the valve body. The left capscrew fractured due to hydrogen
embrittlement cracking through the complete cross section. The right capscrew initially fractured due to hydrogen
embrittlement then propagated due to tensile overload. Both fractures started at the bottom of the photograph.

SEM photograph of Hydrogen Embrittlement Fracture

Of A Fastener

Failure Analysis Page 1 of 45

Fracture surface of a capscrew that failed due to hydrogen embrittlement. The intergranular fracture and subsurface
intergranular cracks are characteristic of fractures due to hydrogen embrittlement. The pitted surface was caused during
cleaning to remove oxide from the fracture surface. 1000X

Failure Analysis Case History No. 002

Failure Analysis Page 2 of 45

Industry: Commercial
Specimen Description: Water heater inlet pipe
Material: Galvanized carbon steel
Environment: potable water
Background: Galvanized carbon steel nipples in residential water heaters were failing
after less than two years service. The nipples were downstream of, and
attached to, copper flexible hoses with electrically isolated, galvanized
couplings.
Time in Service: < 2 years.
Findings: Laboratory failure analysis of hot and cold water nipples from a water
heater showed that the nipples experienced corrosion under I.D.
tuburcles. Analytical analysis showed that the water chemistry was
classified as slightly corrosive, based on the Langlier Saturation Index, with
a calculated LSI = -0.32. Energy dispersive spectrometry (EDS) chemical
analysis of the tubercles showed that they contained copper, as well as
iron, zinc, chlorine and silicon. The hot water nipples experienced less
under-deposit corrosion than did the cold water nipples. This was due to
the increased scaling tendency of the higher temperature water.
The galvanized nipples experienced underdeposit corrosion at low velocity
areas due to a buildup of deposits. The oxygenated water and copper
deposition from corrosion of the upstream copper hose contributed to the
corrosion.

Corrosion Of Galvanized Potable Water Piping

Failure Analysis Page 3 of 45

I.D. of a galvanized carbon steel nipple from the cold water inlet connection of a residential water heater. Tubercles have
formed on the I.D. in low flow conditions. Underneath the tubercles are deep corrosion pits. The cold water chemistry
exhibited a Langelier Index of - 0.31 and contained 8 ppm of dissolved oxygen.

Failure Analysis Case History No. 003

Failure Analysis Page 4 of 45

Industry: Oil & Gas
Specimen Description: Downhole motor rotor
Material: AISI 1040
Environment: Corrosive drilling mud
Background: A hollow, fluted AISI 1040 carbon steel rotor in a downhole motor fractured
while drilling a directional well. The chromium plated rotor was exposed to
a corrosive, water-based drilling mud.
Time in Service: Unknown
Findings: Laboratory failure analysis findings showed that the rotor I.D. contained
many cracks at a 45o angle to the long axis of the rotor. Metallurgical
examination of a typical cracked area showed that they were torsional
fatigue cracks.Craze cracks in the chromium plating did not contribute to
the cracking.

Corrosion Fatigue Cracks In AISI 1020 Steel In Oil And Gas Downhole
Service

Multiple corrosion fatigue cracks at the I.D. of a AISI 1020 carbon steel downhole tool. The tool was rotating with the I.D.
exposed to a water based drilling fluid. 2% Nital Etch, 50X

Failure Analysis Case History No. 004

Industry: Chemical Process

Specimen Description: Agitator Motor Bolts
Material: J429 Gr. 8
Environment: Atmospheric
Background: J429 hex-head bolts attaching an output shaft to the top motor thrust plate
of a chemical agitator failed, causing damage to the bottom steady
bearing.

Failure Analysis Page 5 of 45

Time in Service: Unknown
Findings: Laboratory failure analysis of the bolts showed that the bolt fracture
surfaces exhibited features attributed to rotating fatigue. No significant
corrosion was observed. It was suspected that the bolts became loose,
allowing the thrust plate to rock back and forth while turning. Gross non-
metallic inclusions were observed in one of the failed bolts. The
hardness of all failed bolts met J429 Gr. 8 specification requirements.

Fatigue Fracture Of A Grade 8 Bolt

Fracture surface of a Grade 8 bolt showing multiple rotating-bending fatigue origin features.

Non-Metallic Inclusions In The Thread Of A SAE J429 Gr. 8 Bolt

Failure Analysis Page 6 of 45

Longitudinal view of the microstructure of a Gr. 8 bolt showing entrapped foreign material in a thread crest. The material
is probably scale trapped during thread rolling. 2% Nital Etch, 100X

Failure Analysis Case History No. 005

Failure Analysis Page 7 of 45

Industry: Chemical Process
Specimen Description: U-bend exchanger tube
Material: 316 stainless steel
Environment: Shell-side - Hot Waste fuel
Tube side - Cooling water
Background: 316 stainless steel u-bend exchanger tubes in a waste fuel heeat
exchanger, resulting in a retube of the exchanger. The tube-side cooling
water had been blocked in for an unknown time with hot (420F), shellside
fuel oil still on the shell-side prior to the tube failures.
Time in Service: 5 years
Findings: Laboratory failure analysis findings showed that the 316 stainless steel
exchanger tubes suffered chloride stress corrosion cracking, in straight
sections as well as in the U-bend of exchanger tubes. Microhardness
measurements showed that the hardness of the U-bend was comparable
to that in the straight section (HV 158) and suggested that the U-bend had
been stress relieved after cold forming. It was believed that the root cause
for failure was blocking in the cooling water, allowing the hot oil to heat the
water to temperatures which initiated chloride stress corrosion cracking in
the approximately 800 ppm chloride containing cooling water.

Chloride Stress Corrosion Cracking Photograph

Chloride stress corrosion cracking (SCC) on the cooling water side of a 316L stainless steel
exchanger tube. The cooling water contained approximately 400 ppm chlorides had been blocked in
with the 350F shellside process still flowing. The black stringers are sulfide inclusions. 100X

Failure Analysis Case History No. 006

Industry: Chemical Process

Specimen Description: Pressure Relief Valve
Material: Carbon steel

Failure Analysis Page 8 of 45

Environment: High pressure ethane/propane with CO, CO2, O2 and H20
Background: A high-pressure relief valve (400 psig) failed at a weld while in service,
resulting in a flammable gas leak and autorefrigeration of the valve. The
leak caused a plant shutdown.
Time in Service: 3 years
Findings: Laboratory testing showed that the fracture surface of the failed weld was
typical of cleavage. Hydrogem embrittlement was considered; however,
the softness of the weld, appeared to preclude that failure mechanism. It
was believed that the most logical explanation for the PSV valve failure
was CO, CO2, H20 stress corrosion cracking. This little recognized cracking
mechanism occurs in a high pressure CO, CO2, H20 environments. PPM
levels of oxygen act as a cracking accelerator. Critical levels of CO are
neccesary for cracking to occur. Stress relieving will slow cracking but will
not eliminate it. Cracks can also occur in stress relieved base metal,
depending on pressure stresses.

Stress Corrosion Cracking Fracture Of A Carbon Steel Weld

Cleavage fracture surface of a carbon steel weld from an emergency relief valve pipe. Fracture was
believed to be due to C0-CO2-h30 stress corrosion cracking. (~250X)

Failure Analysis Case History No. 007

Industry: Chemical Process

Specimen Description: 310 stainless steel thermal oxidizer rain hat
Material: 310 stainless steel
Environment: Atmospheric with hot oxidizer flue gases
Background: A 310 stainless steel "rain hat" shielding the opening of a thermal oxidizer
flue gas stack failed from cracking at attachement welds while in service.
Time in Service: Unknown
Findings: Laboratory failure analysis results showed that failure of the 310 stainless
steel rain hat was due to hot cracking and lack of fusion of the attachment
welds. Many polythionic acid stress corrosion cracks were also

Failure Analysis Page 9 of 45

 observed in the microstructure. It was believed the polythionic acid stress
corrosion cracking occurred when sulfur containing flue gases condensed
on the rain hat.

Polythionic Acid Stress Corrosion Cracking Of 310 Stainless Steel

Polythionic acid stress corrosion cracking of type 310 stainless steel. The item was exposed to sulfur containing natural
gas in a continuous flare. (100X)

Failure Analysis Case History No. 008

Industry: Oil & Gas

Specimen Description: Downhole Motor Transmission Coupling
Material: AISI 9310
Environment: Drilling Mud
Background: A downhole motor failed while drilling at 3300 meters. Inspection revealed
that two female AISI 9310 case hardened couplings in the ball drive shaft
had fractured. Corrosion was evident on the coupling and the fracture.
Time in Service: Hours
Findings: Laboratory analysis showed that the couplings had experienced several
types of damage, including pitting corrosion, fatigue and hydrogen
embrittlement cracking. The maximum case hardness measured 675
HK, while the core hardness was 29 HRC.

Fatigue And Hydrogen Embrittlement Fracture Of AISI 9310 Low Alloy

Steel

Failure Analysis Page 10 of 45

Fatigue and hydrogen embrittlement fracture of a female coupling from the ball transmission of a downhole motor.

Hydrogen Embrittlement Crack In AISI 9623 Low Alloy Steel

Hydrogen embrittlement crack initiated by corrosion of a AISI 9623 alloy in downhole service. The microstructure is
unetched, case hardened, quenched and tempered martensite at HRC 29. 2% Nital Etch, 50X

Failure Analysis Page 11 of 45

 Failure Analysis Case History No. 009

Industry: Petrochemical
Specimen Description: Hex Head Capscrews
Material: Unknown low alloy steel
Environment: Atmospheric
Background: Hex head capscrews that attached valves to vertical piping unexpectedly
failed. No inspection or maintenance history was known.
Time in Service: 7 years
Findings: Laboratory analysis showed that the capscrews most probably failed due
to pre existing quench cracks. Non-metallic inclusions, believed to be
aluminum oxide dross particles from the steelmaking killing operation, are
thought to have contributed to quench cracking. Static load and vibrational
stresses are believed to have caused total separation of the capscrews
over time. The hardness of the capscrews measured 44 HRC.

Quench Cracking in AISI 4140 Low Alloy Steel Fastener

Failure Analysis Page 12 of 45

 Quench cracking along a line of aluminum oxide dross particles in a hex head capscrew. The aluminum oxide particles
were probably added as a "killing" agent in the steeling practice. The gray material in the crack interior is high-
temperature oxide formed during tempering. The microstructure is 44 HRC quench and tempered martensite. 2% Nital
Etch, 200X

Failure Analysis Case History No. 010

Industry: Oil & Gas

Specimen Description: Elastomer Ram Body Seal in a Blow Out Preventer
Material: Viton B (FKM)
Environment: Oil Well Workover Fluids
Background: A portable blow out preventer used when servicing oil and gas wells failed
due to loss of a ram body seal.
Time in Service: Months
Findings: The ram body rear seal failed due to blistering from explosive
decompression, thought to have occurred when pressure in the blow out
preventer (BOP) was released rapidly. Examination of the interior of a
blister showed that rupture of the elastomer occurred at two initiation sites
due to pressure buildup at voids in the material.

Explosive Decompression Blistering of Viton B

Failure Analysis Page 13 of 45

Viton B elastomer showing blisters resulting from explosive decompression failure.

Explosive Decompression Blistering of Viton B

Viton B elastomer showing the interior of a blister occurring due to explosive decompression failure.

Failure Analysis Page 14 of 45

 Failure Analysis Case History No. 011

Industry: Gas Treating

Specimen Description: 6" Furnace Tube
Material: Carbon Steel
Environment: Rich Oil containing C1-3 paraffin hydrocarbons
Background: Horizontal, carbon steel furnace tubes in a Rich Oil Fractionator furnace
failed causing a fire.
Time in Service: 30 years
Findings: Laboratory analysis showed that the carbon steel tube sample had
experienced long term overheating under a heavy coke layer. I.D. pitting
was observed under the coke. The pits were believed to be due to metal
dusting, based on a dark etching, carburized layer below the pits.
Observation of the microstructure showed that the dark etching layer
was100% pearlite. The undegraded pearlite proved that metal
temperatures had exceeded the AC3 transformation temperature at some
time. Long term metal temperatures were estimated at approximately
1000F, based on the presence of graphite nodules in the completely
decarburized ferrite matrix.

Graphitization Of A Carbon Steel Furnace Tube

Failure Analysis Page 15 of 45

I.D. pit in a carbon steel furnace tube due to metal dusting after long term overheating and coking in rich oil service.

Metal Dusting Photograph -

Graphitization Of A Carbon Steel Furnace Tube

Pitting due to metal dusting in a carbon steel furnace tube in gas treating rich oil service.

Graphitization Of A Carbon Steel Furnace Tube

Failure Analysis Page 16 of 45

Carburization of a carbon steel furnace tube associated with metal dusting due to long term overheating and coking rich
oil service.

Graphitization Of A Carbon Steel Furnace Tube

Graphitization of a carbon steel furnace tube due to long term overheating and coking in rich oil service.

Failure Analysis Page 17 of 45

 Failure Analysis Case History No. 012

Industry: Electric power generation

Specimen Description: Turbine gearbox pinion gear
Material: 17CrNiMo6 low alloy steel, case hardened by gas carburizing
Environment: Lubricating oil
Background: A double-helical pinion gear in a gas turbine gearbox developed pits on the
loaded teeth. A laboratory failure analysis was conducted to determine the
cause for the gear teeth pitting.
Time in Service: several years
Findings: The pinion gear suffered random, open progressive pitting on the loaded
flanks of the teeth. Other areas of the loaded gear teeth exhibited visual
images of subsurface fatigue cracks that had not developed into open pits
yet. The companion bull gear had not pitted. Microstructural investigation
of a representative subsurface pit showed that it propagated parallel to
the case hardened surface and contained white etching areas (WEA's).
Other subsurface, intergranular cracks were also present that had
initiated at inclusions.

Results of the failure analysis showed that the gear tooth pitting was due to
surface-contact fatigue cracking. The WEA's have been described as
"butterfly wings" and white bands of altered martensite. They reportedly
occur in gear teeth that have experienced heavy shear or impact loads.

Gear Tooth Fatigue Pit

Failure Analysis Page 18 of 45

Progressive pit on the loaded surface of a case hardened pinion gear tooth due to surface contact fatigue.

Gear Tooth Rolling Contact Fatigue Crack

Subsurface rolling contact fatigue crack eminating from inclusions in a case hardened pinion gear tooth.

Gear Tooth White Etching Area

Failure Analysis Page 19 of 45

White etching area (WEA) at the surface of a case hardened pinion gear tooth.

Gear Tooth Rolling Contact Fatigue Crack

Failure Analysis Page 20 of 45

Subsurface rolling contact fatigue crack eminating from inclusions in a case hardened pinion gear tooth.

Failure Analysis Case History No. 013

Failure Analysis Page 21 of 45

Industry: Shipping
Specimen Description: ship anchor
Material: KSC 40 high carbon cast steel
Environment: seawater, river water
Background: A ship's anchor failed due to complete separation just above the flukes
while hosting the anchor in a US river. The river temperature at the time of
the anchor failure was approximately 25C.
Time in Service: Unknown
Findings: The anchor fracture was flat, brittle and contained chevron markings. After
polishing a section at the fracture initiation point, it became evident that the
anchor casting had been repair welded at some time. Cracks were
following the heat affected zone (HAZ) of the weld at the surface.
Observation of the microstructure showed that the cracks were
intergranular, following prior austenite grain boundaries in the martensitic
matrix. The maximum hardness in the cracked area was measured at
Vickers 653.

Hydrogen Embrittlement Failure Of A Ship Anchor

Fracture surface of a KSC 42 ship's anchor that failed due to hydrogen embrittlement at a repair weld.

Hydrogen Embrittlement Failure Of A Ship Anchor

Failure Analysis Page 22 of 45

Macro etched specimen of a KSC 42 ship's anchor that failed due to hydrogen embrittlement (HE) at a repair weld.
Hydrogen embrittlement cracks are on the right heat affected zone (HAZ) of the weld at the surface.

Hydrogen Embrittlement Failure Of A Ship Anchor

Micro showing hydrogen embrittlement cracks in a ship anchor repair weld, see macro photograph. The maximum
Vickers hardness in the cracked heat affected zone was HV 653.

Hydrogen Embrittlement Failure Of A Ship Anchor

Failure Analysis Page 23 of 45

KSC 42 high carbon steel microstructure showing hydrogen embrittlement (HE) cracks in a ship anchor repair weld, see
mounted micro. The maximum Vickers hardness in the cracked heat affected zone (HAZ) was HV 653.

Failure Analysis Case History No. 014

Failure Analysis Page 24 of 45

Industry: Wastewater Processing Plant
Specimen Description: Vertical Pump Impeller
Material: C95300 Nickel Aluminum Bronze
Environment: Filtered water
Background: The blades from aluminum bronze pump impellers were failing after a short
time in service due to complete separaton fracture. The blade failures
resulted in disruption to the plant operation and expensive repairs.
Time in Service: Unknown
Findings: Dye-penetrant inspection of the impeller revealed many areas of random,
PT indications over much of the impeller blade and hub. Visual
examination of blade fractures revealed thumbnail areas where the
fracture was a different color from that of the overload fracture. Fractures
initiated from the "loaded" side of the impeller blade. SEM examination
of a fracture revealed fatigue striations at the initiation edge of the blades.
Microstructural examination of a blade at a fracture revealed that
dealuminification had occurred in service,. reducing the strength of the
blade material in the dealuminified areas. The impeller blades had suffered
fatigue at weak, dealuminified areas near the blade hub. The
dealuminification was due to exposure to the filtered water environment.

Waste Water Pump Impeller

Nickel aluminum bronze pump impeller after dye-penetrant inspection. The red dye represents areas of surface voids.

Fractured Pump Impeller Hub

Failure Analysis Page 25 of 45

Nickel aluminum bronze impeller showing fracture at the hub radius. The reddish, thumbnail areas represent areas of
dealuminification.

Pump Impeller Blade Fracture

Close-up of a pump impeller blade fracture at the hub radius showing the initiation sites and fast fracture areas.

SEM photograph of a Pump Impeller Blade Fracture

Failure Analysis Page 26 of 45
SEM photograph of a nickel aluminum bronze pump impeller showing fatigue striations at the fracture initiation site.

Microstructure of a Nickel Aluminum Bronze

Microstructure of a nickel aluminum bronze impeller that has experienced dealuminification in filtered water. ~170X,
ammonium hydroxide etch.

Failure Analysis Case History No. 015

Failure Analysis Page 27 of 45

Industry: Cement Manufacturing Plant
Specimen Description: 310SS Refractory Anchors
Material: 310 Stainless Steel
Environment: High-Temperature Flue Gases
Background: Type 310SS refractory anchors were failing due to complete fracture,
resulting in loss of refractory insulation in a cement kiln.
Time in Service: Unknown
Findings: As-received 310SS refractory anchors were grayish and scaley but
otherwise showed no significant surface scale resulting from the high
temperature service. The anchor fractures were brittle appearing with no
macroscopic ductility. Metallographic examination of the anchor fractures
revealed intergranular fracture, intergranular cracking and grain dropping,
suggestive of polythionic acid stress corrosion cracking (PASCC) from
exposure to a condensed aqueous phase containing oxygenated sulfur
compounds. The PASCC suggested that the anchors were cracking during
downtimes. No significant oxidation or subsurface sulfidation was
observed.

Type 310SS Refractory Anchors

As-received type 310SS refractory anchors for analysis, one used and one new. Note the fracture at the corner of the top
anchor.

Microstructure of a Type 310SS Anchor

Failure Analysis Page 28 of 45

Microstructure of a 310SS anchor at a brittle fracture. The grain dropping and intergranular cracking and corrosion result
from polythionic acid stress corrosion cracking. 57X, glyceregic atch.

Failure Analysis Case History No. 016

Failure Analysis Page 29 of 45

Industry: Oil & Gas
Specimen Description: Offshore Platform Drilling Riser Bolt Insert
Material: AISI 4340 Quenched and Tempered Low Alloy Steel
Environment: Seawater
Background: AISI 4340 drilling riser bolt inserts were cracking after exposure to
seawater and cathodic protection currents.
Time in Service: Unknown
Findings: Visual examination of the as-received insert showed circumferantial
cracking at the body to flange radius. The fracture face was brittle
appearing with a single cracking initiation site. SEM imaging of the
fracture revealed a deposit high in calcium obscuring the fracture features.
After removing the deposit with inhibited hydrochloric acid, the true,
intergranular fracture could be observed. The integranular fracture and
knowledge of the service environment suggested failure from hydrogen
embrittlemnt.

AISI 4340 Riser Insert

Failure Analysis Page 30 of 45

As-received insert from an offshore drilling riser.

Fracture of a Riser Insert

Failure Analysis Page 31 of 45
Brittle fracture of an AISI 4340 riser insert. The fracture initiation site is at the 6 o:clock position.

Fracture initiation site of a riser insert

Failure Analysis Page 32 of 45

O.D. fracture initiation site of a offshore drilling riser.

Calcarous Deposits on a Riser Insert Fracture

Calcium deposits on the fracture surface of a AISI 4340 riser insert.

Intergranular Fracture of AISI 4340

Failure Analysis Page 33 of 45

Intergranular fracture of a AISI 4340 drilling riser insert resulting from hydrogen embrittlement.

Failure Analysis Case History No. 017

Failure Analysis Page 34 of 45

Industry: Chemical
Specimen Description: 309 stainless steel wedge divider
Material: AISI 309 stainless steel
Environment: hot flue gases
Background: A 309 stainless steel wedge divider in a thermal oxidizer failed due to
catastrophic metal loss after a short time in service. Failure caused
shutdown of the oxidizer to replace the dividers.
Time in Service: ~ Two years
Findings: A 309SS wedge divider was submitted for laboratory analysis due to
significant metal loss in a high temperature exhaust gas environment. The
wedge divider acted to contain catalyst and was exposed to hot exhaust
gases from a rotary dryer that dried wood shavings and sawdust. The
offgas from the wood contained combustion products of methanol,
formaldehyde, acrolyn, acetaldehydes, and wood dust. The wood and
resin reportedly contained sodium and potassium chloride salts and
calcium sulfates. It was reported that the greatest metal loss occurred
where the exhaust temperatures were the lowest. Metallographic
examination of the wedge divider in the area of greatest metal loss (lowest
temperature) revealed that the material was almost completely
consumed and contained massive, multi-layered corrosion products.
Metallographic analysis of the wedge divider at the top (highest
temperature) end showed that the microstructure contained massive
nitride needles, suggesting of exposure to an disassociated ammonia or
high-temperature nitrogen gas environment. Observation of sigma phase
in the top portion microstructure suggested exposure to temperatures
between 1000F and 1600F. All evidence suggested that the wedge divider
experienced metal loss due to exposure to molten fuel ash containing low-
melting point vanadium, potassion and sodium salts.

309SS Thermal Oxidizer Wedge Divider

Failure Analysis Page 35 of 45

As-received 309SS wedge divider illustrating the variable metal loss from top to bottom.

Hot End MIcrostructure of a 309SS Wedge Divider

Failure Analysis Page 36 of 45
Unetched microstructure of the 309SS wedge divider at the hot end showing masive, layered surface scales, 57X.

309SS Microstructure

309SS Microstructure with Surface and Subsurface Corrosion Products, 760X.

Failure Analysis Page 37 of 45

309SS microstructure with Nitride Needles

Massive nitridie needles in the 309SS matrix, 380X

Sigma Phase in 309SS

Sigma phase in the 309SS thermal oxidizer microstructure, indication exposure to temperatrues from 1000-1600F,
~450X.

Failure Analysis Case History No. 018

Failure Analysis Page 38 of 45

Industry: Refining
Specimen Description: LPG Product Hose
Material: Rubber Composite
Environment: LPG
Background: Several LPG product loading hoses failed due to bulging and tearing of the
outer rubber layer. The hoses contained LPG (propane, butane, isobutane)
at approximately 150-200 psig.
Time in Service: Less than 3 months
Findings: The as-received hoses all contained deflated bulges, splits and tears to
the outer rubber cover. Researching the manufacture and use of rubber
hoses in LPG service indicted that LPG hoses were supposed to contain
pin pricks, holes, or other means of venting gases that permeated through
the hose to the outer cover. As the aneurisms and tears suggested build
up of pressure under the outer cover, a test procedure was devised to test
good sections of the hoses for permeability. A gouge was made on the I.D.
of the hoses to a depth of the outer cover. End caps were placed on the
cut ends of the hoses and they were pressured up using shop air with the
hose under water to facilitate observation of air bubbles. All submitted
hoses bulged indicating that the hose outer covers did not contain
adequate gas venting means.

LPG Hoses

As-received photograph of submitted failed, LPG product hoses.

LPG Hose Tear

Failure Analysis Page 39 of 45

Tear in the outer cover of a LPG product hose showing the twine braided reinforcement layer.

Bulged LPG Product Hose

Bulge in the outer cover of a LPG hose from trapped air that built up under the cover while the hose was pneumatically
tested.

Failure Analysis Case History No. 019

Failure Analysis Page 40 of 45

Industry: Chemical
Specimen Description: Failed Scaffold Clip
Material: Carbon Steel
Environment: Atmosphere
Background: Welds attaching a scaffold clip to a scaffold failed causing the scaffold
boards to fall to the bottom of a API 650 tank.
Time in Service: ~ Less than a week.
Findings: The as-received scaffold clips were made by fillet welding one side of a
bent bar onto the I.D. of the tank wall with approximately one-inch of weld.
On one sample, the clip still remained. On the other sample both welds
attaching the bar to the tank wall had failed. Visual observation of the fillet
welds showed that they had been made with the SMAW process and
apparently without removing rust, etc. from the tank wall or the bar. The
weld fracture appeared dendritic and contained significant voids. SEM
examination of a fractured weld revealed significant unfused areas
amongst areas of ductile overload. A polished and etched profile of the
unfailed clip showed that the weld contained large interdendritic cracks and
a large untempered weld heat-affected zone. The weld microstructure
contained many interdentric cracks suggestive of hot tear cracks.

Tank Scaffold Clips

As-received API 650 tank scaffold clips. The clips were made by fillet welding a bent bar onto the I.D. of the tank wall.

Failed Tank Scaffold Clip Weld

Failure Analysis Page 41 of 45

Failed tank scaffold clip weld showing the SMAW weld on a rusted tank wall plate.

Fractured SMAW fillet weld from a tank scaffold clip

Close-up of a fractured SMAW fillet weld from a tank scaffold clip.The weld is dendritic appearing and contains several
large voids.

SEM photograph of a Fractured Tank Scaffold Clip Weld

Failure Analysis Page 42 of 45

SEM photograph of a Fractured Tank Scaffold Clip Weld showing unfused metal and areas of ductile overload fracture.

Polished Profile of a Tank Scaffold Clip

Polished and etched macro of an unfailed tank scaffold clip showing large interdendritic cracks and a large, untempered,
weld heat-affected zone.

Failure Analysis Case History No. 020

Failure Analysis Page 43 of 45

Industry: Pipeline
Specimen Description: Rich/Lean Amine exchanger tube
Material: Carbon Steel
Environment: Rich monoethanolamine (MEA) amine
Background: Carbon steel exchanger tubes in a rich/lean exchanger failed due to
circumferential cracking near the tubesheet.
Time in Service: ~ Unknown
Findings: Carbon steel exchanger tubes experienced circumferential cracking and
leaking near the tubesheet of a Rich/Lean Exchanger in a natural gas
sweetening unit. Carbon steel tubes in the top half of the exchanger had
been replaced with stainless steel tubes due to failures of an unknown
type. The fracture face of the failed tube samples appeared blocky and
brittle appearing. It was heavily scaled and the tube I.D. was corroded.
After chemically removing the scale/deposits from the fracture, SEM
examination showed that the fracture was intergranular. The polished and
etched microstructure revealed surface cracking, an intergranular crack
path in a matrix of annealed ferrite grains. Based on the pearlite content,
the carbon content of the tube was low. the intergranular fracture, plus
knowledge of the MEA service suggested failure of the tubes due to amine
stress corrosion cracking, or alkaline stress corrosion cracking. Amine
SCC in a exchanger tube was considered unusual in that amine cracking
of carbon steel components usually occurs in non stress-relieved welded
components. It was hypothesized that differential thermal stresses created
the high tube stresses responsible for cracking.

Circumferential Fracture of a Carbon Steel Exchanger Tube

Circumferential Fracture of a Carbon Steel Exchanger Tube from a Rich/Lean amine exchanger. The fracture is blocky
and the I.D. corroded.

SEM Photograph of a Carbon Steel Exchanger Tube Fracture

Failure Analysis Page 44 of 45

SEM Photograph of a Carbon Steel Exchanger Tube Fracture showing intergranular fracture features.

Microstructure of a Carbon Steel Exchanger Tube

Microstructure of a carbon steel exchanger tube at the intergranular fracture. The photograph shows a blocky, brittle
fracture profile in a matrix of annealed ferrite grains with secondary, intergranular cracking off the main fracture face.

Failure Analysis Page 45 of 45