PREDICATION OF RESIDUAL LIFE TIME OF STEAM REFORMER TUBES

Prof.BahaaZaghloul Prof Bahaa Zaghloul

ExPresidentofCentralMetallurgicalResearchand DevelopmentInstitute CAIRO,EGYPT

bahaazaghloul@cmrdi.sci.eg

DIFFERENT APPLICATIONS OF REFORMER TUBES

CMRDI

40

LimitoftubeMetal Applications pp

Pre essure(Kg g/cm2G)

30

H2
20

NH3 CH3OH
10

Town Gas D. R.
700 800 900 1000

GasTemperature(OC)

LIMITATION OF OPERATION TEMPERATURE

Inlet

CMRDI

Cata alyst Tube T

ControltoPrevent CarbonFormation

ControltheMax.Tube SkinTemperature

Outlet

CriticalPortion

DIFFERENT DESIGNS FOR REFORMER FURNACE 1 3

UP UP FiringType

CMRDI

Radiant Radiant WallType

Terrace Terrace WallType

Down FiringType

EFFECT OF FIRING SYSTEM ON TEMP. PROFILE

CMRDI

I . UP

II . Down D

III . DOWN & UP

Preferable Profile
IV . TERRACE WALL V . RADIANT WALL

IDEAL TEMPERATURE PROFILE

CMRDI

Inlet

Outlet

Gas Temperature

HeatFlux

TubeSkin Temperature

SolidLine:IdealProfile

Dottedline:PreferableProfile

REFORMER TUBES MATERIALS EVALUTION CHARTS

Si

CMRDI

HK40highSi 25Cr20Ni-highSi
Nb

HK40 25Cr20Ni
Ti

IN519 24Cr24NiNb
Nb

Ti

IN519mod. 24Cr25Ni-NbTi HIKa1b 25Cr20Ni-NbTiMS

MS: Mishmetal

HIKa1a 25Cr20Ni-TiMS

BSTAlloy 25Cr20NiTiNb

HT 15Cr35Ni HU 19Cr39Ni HTmod. 20Cr32NiC

Nb Nb

Alloy800 20Cr32Ni

Manaurite900 T52,PG2032 T52 PG2032 CR32W,KHR32C 20Cr32NiNbLowC Manaurite36XSu ParalloyH39W 25Cr35NiNb

Nb

Low C

Manaurite900B ParalloyCW39W 25Cr35NiNbLowC Manaurite36XM HPBST 25Cr35NiNbTi Manaurite36XS 25Cr35Ni-NbW SuParalloyH34CT 25Cr35Ni-5W2Co
Co

Ti

HP 26Cr35Ni

W W

MoRe1 25Cr35Ni-W
Mo

Nb

Co

Manaurite900 T52,PG2032 20Cr32NiNbLowC

Supertherm 28Cr35Ni-5W15Co

REFORMER TUBES MATERIALS EVALUTION TABLES

METALLURGY DEVELOPMENT
DATE 1960s 6 1970s 97 Early80s Mid80s Late80s COMMON NAME HK40 HK IN519 59 BST HP BST HPMicroalloy Cr % 25 25 5 23 25 25 Ni % 20 24 4 22 35 35 Nb % 1 1 1 1 OTHERS Ti Ti

CMRDI

RELATIVE STRENGTH 1.0 1.4 4 1.7 1.9

Ti,Zr,W,Cs 2.2

CREEP RAPTURE STRENGTH OF COMMERCIAL MATERIALS

20

CMRDI

Str ress (Kgf/mm2) g

10 8 6 4
BST - M 25 35Ni - Nb IN 519 HP HK 40

100

1000

10000

100000

Time to Rupture (h)

CREEP RAPTURE STRENGTH OF COMMERCIAL MATERIALS

CMRDI

CREEP RAPTURE STRENGTH OF COMMERCIAL MATERIALS

Rapture Stress for Generic Tube Materials
35 30 25 20 15 10 5 0

CMRDI

R RaptureSt tress(N/ /mm2)

HK40

IN519

BST

HP BST

HPMicro alloy

FAILURE MODE OF REFORMER TUBE

CMRDI

CROSS SECTION OF FAILED REFORMER TUBE

CMRDI

PROGRESS OF CREEP FAILURE

CMRDI

START Cracks30%from innerwall

GROWTH Cracksgrowto breakinnerwall

FAILURE Cracksprogressto outerwall

EFFECT OF MACROSTRUCTURE ON CRACK PROPAGATION

CMRDI

TYPICAL CREEP DAMAGE MECHANISM IN REFORMER TUBES

CMRDI

FRACTURE

CreepStr rain

C B A
I, II I II, III : CreepRanges

ExposureTime

Microstructure Evaluation of Cast Steel Tubes During Operation Time in Reformer Furnace
ChangeInTubeMicrostructure DuringOperationinReformer Furnace

CMRDI

MATRIX
Chemicalcomposition Chemical composition modifications&accompanied heterogentyinsolidsolution Changesindislocationstructure anddislocationdensitygrowth Changesinaustenitegrain Ch i t it i boundarymicrostructuredueto thecarbidescoalescenceand voidsformation

PRIMARYAND PRIMARY AND SECONDARYPHASES

Changeinprimaryinnerdendritic Change in primary inner dendritic austenitecarbideseutectic Secondarycarbidesprecipitationand Secondary carbides precipitation and interactionswithprimarycarbides Continuouscarbidesnetwork formationaroundaustenitegrains formation around austenite grains Coalescenceofsigmaphaseinside andaroundaustenitegrains g

CMRDI

EffectsOfMicrostructural Eff t Of Mi t t l ChangesInReformerTube DecreesInCreep Resistance

MicrovoidsInitiationOn InterphaseBoundaries

DeteriorationOn MechanicalProperties

VoidsPropagation id i VoidsCoalescenceInto Fissures

COMMON DAMAGE MECHANISM IN REFORMER TUBES

CMRDI

1. Overheating (Cumulative) 2. Thermal Cycling ( cyclic operation ) 3. T b B di Tube Bending ( i improper counter weight balance i h b l for tube linear will increase stress on tube) 4. Thermal shock 5. Stress Corrosion Cracking

OVERHEATING 1 2 3 4 5

CMRDI

TemperatureIncreaseToMaintainYieldWhenCatalystBecomes LessActive. FeedstockOrSteamSupplyFailure

TheAbsenceOfCoolingEffectOfTheEndothermicReaction

BurnerMisalignmentCanResultInOverTemperature Burner Misalignment Can Result In Over Temperature

RestrictedFlowOfProcessDueCatalystChoking

EFFECT OF OPERATION TEMPERATURE AND PRESSURE ON TUBE LIFE

CMRDI

COMMON DAMAGE MECHANISM IN REFORMER TUBES

CMRDI

1. Overheating (Cumulative) 2. Thermal Cycling ( cyclic operation ) 3. T b B di Tube Bending ( i improper counter weight balance i h b l for tube linear will increase stress on tube) 4. Thermal shock 5. Stress Corrosion Cracking

THERMAL CYCLES 1 2 3 4
ThroughWallStressesareTemporarilyIncreased

CMRDI

TubeWallTemperatureGradientsCanBeSignificant Tube Wall Temperature Gradients Can Be Significant

DuringOperation,TheStressProducedByDifferentExpansion RelaxesThroughCreep TemperatureChangesUpOrDownWillReintroduceSome Temperature Changes Up Or Down Will Reintroduce Some Stress.

COMPARISON OF STRESS DISTRIBUTION IN CYCLIC OPERATION

CMRDI

HK40

IN519

COMMON DAMAGE MECHANISM IN REFORMER TUBES

CMRDI

1. Overheating (Cumulative) 2. Thermal Cycling ( cyclic operation ) 3. T b B di Tube Bending ( i improper counter weight balance i h b l for tube linear will increase stress on tube) 4. Thermal shock 5. Stress Corrosion Cracking

THERMAL SHOCK

CMRDI

CreatesExtremelyHighStressesAsTheTubeAttemptsTo ContractUnderRestraint. C t tU d R t i t

2 3

ItResultsInCircumferentialTearingOrShatteringOfTheTube.

CanOccurThroughBoilerWaterCarryOverOnInsideTheTubes

EFFECT OF OPERATION TEMPERATURE ON MICROSTRUCTURE OF IN519 MATERIAL IN519

CMRDI

MICROSTRUCTURE OF IN519 CAST STEEL TUBE

Nondegradedsampleafter24000h N d d d l ft 24000 h

after 30000h

OperationinReformerfurnace 86000h

FAILURE ANALYSIS OF PRIMARY REFORMER TUBE

CMRDI

OPERATION PERIOD = 6 Years

TopFlange

RaptureZone RuptureZone p

TopFlange

GENERAL AND ENLARGED VIEWS OF OUTER AND INNER SURFACES OF RUPTURE ZONE OF REFORMER TUBE.

CMRDI

NOTE: NONUNIFORM WIDTHOF BRANCHEDCRACK.

SUMMARY OF MICROSTRUCTURE CHANGE THROUGH THICKNESS, ALONG REFORMER TUBE LENGTH G

CMRDI

S1: 30cmfrom topflange

1000X 1000X

S2: 114cmfrom topflange

S4: Rupturezone 320cmfrom topflange

200X 1000x

S8: 1070cmfrom topflange

S4: 1255cmfrom topflange

500X 1000x

S12: 1330cmfrom topflange

FAILURE OF PRIMARY REFORMER TUBE

OPERATION PERIOD 18 Years
S4 S3 S2 S1 RUPTURE S5 RUPTURE

CMRDI

S1
B C

1000 X

1000 X

1000 X

1000 X

1000X

1000X

Header

200 x Header

1000X

50x

PRIMARY REFORMER TUBE

CMRDI

Tube: 135.6mm,12.8mm Material: HPNb 0.4C,35Ni,25Cr,1.24Nb

Pressure : 35.7 bar

OPERATION PERIOD = 9 Years

Skin Temp.: 917oC Max

860oC Min

Test Coupon 5

Test Coupon 3
Inner Surface S e Outer Surface S e

100X

TEST COUPON 5

100X

TUBE WALL CENTER

400X

TEST COUPON 3

RESULTS OF CREEP RUPTURE TEST OF USED PRIMARY REFORMER TUBE

CMRDI

Creep Rupture Strength Master Curve for Used y o & Unused Primary Reformer Tube Materials

CMRDI

LMP=T(C+tr )103
T:TestT(OK), tr :Rup.time(hr) C:Constant

This form of d t plot i used Thi f f data l t is d to extrapolate shortterm test data to rupture stresses up to 100,000 100 000 hr

Working S is a function of mainly circum., long., and thermal S. The most critical one is the circum. S & it was assumed to be 1.8kg/mm2 based on the following equation. (ASME Section VIII Division 1)

S=(PR+0.6Pt)/Et ( )/
S:Maximumallowablestressvalue. P:InternaldesignPormax.allowableworkingP P I t ld i P ll bl ki P R:Insideradius E:Jointefficiency t:Min.thick

SkinTwasassumedtobe917 OC, ki d b RemaininglifetimeiscalculatedfromdataofLarsonMillar parameter ofusedreformertube. ofusedreformertube TheremaininglifewasestimatedbasedonS &skinT.tobe~20,000hr at1.8kg/mm2workingS. at1 8kg/mm2workingS Thelifetimewillbeincreasedatlowerskin T

ESTIMATED RUPTURE TIME OF USED PRIMARY REFORMER TUBE UNDER DIFFERENT TEMPERATURES & STRESSES

CMRDI

1.8kg/mm2

2.5kg/mm2

917oC

NON DESTRUCTIVE TESTING METHODS FOR DAMAGE EVALUATION EVALUATION

CMRDI

Dimensionalchanges; diameterandwallthicknessduetocreep Replica R li metalography; spot check f micro structural changes t l h t h k for i t t l h duetooverheatingandsometimescreepdamage Radiography; largevoidsandmicrocracks Eddy Eddy current; chromium migration due overheating and chromium migration due overheating and conductivitychanges Ultrasonic; attenuationandscattering Newtechniques suchasLOTIS andIESCO H SCAN New techniques such as LOTIS and IESCO HSCAN

CREEP DAMAGE EVALUATION METHOD

CMRDI

NewProjectfor

PREDICTION OF REMAINING LIFE OF REFORMER TUBES

& CMRDI Egypt PARTICIPATING COMPANIES FraunhoferIWM Germany

LIFE TIME PREDICATION FOR STEAM REFORMER TUEBS

Possible Industrial Partners: Egyptian Fertilizers Co., Ain El-Sukhna, EFC

CMRDI

ALEXFERTAlexandriaFertilizersCo.,Alexandria DeltaFertilizerCo.,Talkha D lt F tili C T lkh ApuQuirFertilizerCo.;Alex Schmidt+Clemens,CentrifugalCastingDiv.,Kaiserau UHDEEngineeringEgypt,Cairo

LIFE ASSESSMENT REQUIREMENTS

CMRDI

Requirements For Analysis

Toundertakeafurnacelifeassessmentthefollowing inputinformationisrequired
Furnace design drawings / material records / Operation history (turn-around and process conditions) Process temperature and pressure records as available Tube samples for material specific assessment Appropriate NDT t h i A i t techniques data d t

Collection/SelectionOfTubeMaterials,Geometries, Plant,ConnectionsAndProcessConditions

DevelopmentOfTestSetUpAndMeasurement TechniquesForSpecimensAndModelTubes.

ContinuedTestingOf SpecimensFromUsed TubesAndFailureCases

LIFE TIME PREDICTION OF REFORMER TUBES

TestingOfVirginMaterialSpecimensInTension, CreepAndThermalFatigueAcc.ToService

NDEOfUsedTubes Profilometry Visual Vi l UT,EC, Xray

RepeatedExaminationOfDamage NDEE.G.EddyCurrent,UT Hardness Metallography,Microstructure. Sem&Edx EvaluationOfMechanisms AndDegreeOfDamageIn UsedTubes MaterialDescriptionfor Material Description for deformation damage failureincl.overlappingeffects ModelComponentTestAnd Simulation(ShortSectionOfTube ExaggeratedLoading,ShorterTimes)

ImplementationInFECode

Verification ComparisonOfCalculation(Material DegradationComponentBehavior) WithMeasurementResults

Material/microstructurebasedlifetime predictionappliedtoaspecificplant

LIFE TIME PREDICATION FOR STEAM REFORMER TUEBS

STEP 1 : COLLECT & SELECT INPUT DATA

CMRDI

- Tube material (properties, behaviour), -dimensions - Plant design: inlet/outlet pigtails,firing system, welds, - Operational conditions: feedstock rate pressure fluid temperature p skin temperature through wall temperature start up/shut down cycles process upsets - service experience

LIFE TIME PREDICATION FOR STEAM REFORMER TUEBS

STEP 2 : DEVELOP TEST SET UP AND MEASUREMENT TECHNIQUES FOR ROUND BAR SPECIMEN TESTS ON REFORMER TUBE MATERIALS Cyclic strain- and temperatures tests

CMRDI

LIFE TIME PREDICATION FOR STEAM REFORMER TUEBS

STEP 3 : DEVELOP TEST SET UP AND MEASUREMENT TECHNIQUES FOR MODEL REFORMER TUBE TESTS

CMRDI

Optical Strain Measurement

HAZT91 5 510mm

HAZT91 5 510mm

NiCr20Nb

Austenite

10mm 5mm

LIFE TIME PREDICATION FOR STEAM REFORMER TUEBS

STEP 4 : TESTING OF VIRGIN AND USED REFORMER TUBE MATERIAL SPECIMENS
Grobkorn-WEZ 550C Grobkorn-WEZ 625C

CMRDI

300

Spannung in MPa

Grobkorn-WEZ 650C

LCFtestsatRTuptoservicetemp.(950C) p p ( ) cyclichardening/softening hystereses lifetime

F

100 interkritische
WEZ 625 C 625C interkritische WEZ 650C

TMFwithrealisticTecycles in/outofphase holdtimes heatingandcoolingtimes

1000 Zeit in h 10000

50

WEZ3 GK WEZ2 FK WEZ1 IK GW FZJ GW MFI

T in C 550 625 650

10

100

Creeptests

LIFE TIME PREDICATION FOR STEAM REFORMER TUEBS

CMRDI

STEP 5 : MEASUREMENT OF DEFORMATION, DAMAGE & FAILURE; EVALUATION OF MECHANISMS IN THE TUBE MATERIALS
NDE:dimensionalmeasurements,UT,EC, Hardness Metallography,microstructure,replicatechniques SEM&EDX

CyclicPlasticDeformation

Micro/MacroCracking

CreepDamage

Corrosion

LIFE TIME PREDICATION FOR STEAM REFORMER TUEBS

CMRDI

STEP 6 : MATERIAL DESCRIPTION FOR DEFORMATION, DAMAGE AND FAILURE INCL OVERLAPPING EFFECTS INCL. Materialmodelsforcyclicplasticdeformation (Chaboche,Jiang) - IWM Software FITIT

950C

Damageparameter

DTMF

I2,eff 2,4 I2 pl 1,45 = + e F (T , t ) cy E 1 + 3 N cy e

LIFE TIME PREDICATION FOR STEAM REFORMER TUEBS

STEP 7 : IMPLEMENTATION OF NEW MATERIAL STRUCTURAL (FE) MODELS (EVALUATIONOFSPECIMENTESTRESULTS)

CMRDI

ApplicationOfDamage ParameterDTMF ToLCF andTmftests

LIFE TIME PREDICATION FOR STEAM REFORMER TUEBS

STEP 8 : VALIDATION BY

CMRDI

Postevaluationofwelldocumentedfailurecasesfromfertilizerplants (parameterstudies!) ModelcomponenttestsandFEsimulation(shorttubesection, exaggeratedloading,shortertimes,)

Tube Tests
600 500 400 300 200 100 0 1 Cycles 2 3 Axial load 41.8 kN Internal pressure: 480 bar 600

Temperatur [C] T re

Intern. Pre Internalessure,in bar, p pressure bar Axial lo load in kN axial oad, kN

500 400 300 200 100 0

FE - Simulation Of Stresses & Strains

Cycles

Temp peratur [C]

LIFE TIME PREDICATION FOR STEAM REFORMER TUEBS

STEP 9 : APPLICATIONS Materialtests/microstructuralanalysisandFEM simulation based ValidatedProcedure t V lid t d P d toassesscomponentbehaviorundertransient tb h i d t i t severeserviceconditions

CMRDI

ReadyToBeUsedFor LifeTimeManagement ofcatalyststeamreformertubes ProcessOptimisation InspectionStrategy development ReconsideringplantDesign andOperatingConditions

YOURKINDPARTICIPATIONINTHEPROJECT WILLBEHIGHLYAPPRECIATED
ManyThankForEFCWh I i dM F Thi P M Th kF WhoInvitedMeForThisPresentation i

&
THANKS FOR YOUR ATTENTION