girth welding of api 5l x70 and x80 sour service pipes · pdf filethe api 5l x70 and x80 pipes...

8
Introduction Steel plates produced for pipelines are developed to ensure suitable proper- ties for each specific project. Particularly for the onshore part of the pre-salt gas pipelines in Brazil, the pipes are speci- fied with additional requirements for sour service application due to the pres- ence of H 2 S in the gas composition. This situation created a big challenge for gas pipeline transportation. The use of con- ventional API 5L (Ref. 1) grades up to X65 associated with the location class factors could result in heavy wall thick- ness pipes that are difficult to produce and are not usually used for onshore pipeline construction. The solution found to overcome this challenging sce- nario was the use of X70 and X80 sour service high-strength steels. High-strength steels such as X70 and X80 grades have been employed in high-pressure pipelines to avoid the use of thicker pipes, allowing the eco- nomic construction of long distance pipelines. However, the production of X70 and X80 for sour service applica- tion is a big challenge that requires a more complex production route to en- sure resistance against sulfide stress cracking (Refs. 2, 3). To apply these pipes in onshore pipeline construction, in addition to the challenge of producing a sour serv- ice resistance steel, it is necessary to carefully evaluate the girth welding ef- fects on heat-affected zone (HAZ) and weld metal properties. In high-strength steel welding, a hardness reduction has been observed at the HAZ, also known as HAZ soften- ing (Refs. 4, 5). It is related to the heat input effects on the strengthening mechanism of these steels. Since the sour service steels are produced by ac- celerated cooling and have lower levels of alloying elements, they may also be susceptible to HAZ softening. Duan, Lazor and Taylor (Ref. 6) commented that the softening effect in the HAZ will potentially cause highly localized deformation, which is undesirable in conditions where strain-based design is applicable. On the other hand, depend- ing on the heat input, the HAZ hard- WELDING RESEARCH OCTOBER 2016 / WELDING JOURNAL 363-s SUPPLEMENT TO THE WELDING JOURNAL, OCTOBER 2016 Sponsored by the American Welding Society and the Welding Research Council Girth Welding of API 5L X70 and X80 Sour Service Pipes The results of this investigation provide a better understanding of welding X70 and X80 pipe developed for sour service BY G. Z. BATISTA, L. DA P. CARVALHO, M. S. DA SILVA, AND M. P. SOUZA ABSTRACT The API 5L X70 and X80 pipes development for sour service application is very recent. Limited information can be found in literature about the girth weld- ing of these materials. This investigation presents the hardness, toughness, ten- sile, and sour resistance results of welded joints produced on API 5L X70 and X80 pipes developed for sour service. The joints were welded with the gas metal arc welding process for the root pass and flux cored arc welding process for the fill passes. The test methodology included tensile, bending, and nick-break tests as required by the API 1104 standard. The joints were also submitted to Charpy, crack opening displacement (CTOD), hardness, and stress corrosion tests. All joints met the requirements defined by API 1104 for tensile, bending, and nick- break tests. The Charpy tests presented high values of absorbed energy. The weld metal and HAZ of X70 and X80 presented good toughness according to the CTOD results, although one specimen for X80 exhibited low CTOD. All joints met the requirements of the stress corrosion cracking test. Considering all results, the weldability study demonstrated that both X70 and X80 sour service pipes were appropriate for sour service applications. For X80, it is recommended to perform an in-depth analysis of the heat-affected-zone (HAZ) toughness before the use of this material. The results presented here can be used for field girth welding of X70 and X80 sour service pipes and provides information leading to a better comprehension of the welding of these recently developed materials. KEYWORDS • Sour Service • Pipe • Girth Weld • API 1104 • Crack Tip Opening Displacement (CTOD) • High-Strength Steel G. Z. BATISTA is a metallurgical engineer and L. DA P. CARVALHO is a mechanical engineer with Petrobras, Rio de Janeiro, Brazil. M. S. DA SILVA is a metallurgical engineer with ATNAS, Rio de Janeiro, Brazil. M. P. SOUZA is a mechanical engineer with Tenaris, Rio de Janeiro, Brazil.

Upload: ngoliem

Post on 21-Mar-2018

229 views

Category:

Documents


8 download

TRANSCRIPT

Page 1: Girth Welding of API 5L X70 and X80 Sour Service Pipes · PDF fileThe API 5L X70 and X80 pipes development for sour service application is very recent. Limited information can be found

Introduction Steel plates produced for pipelinesare developed to ensure suitable proper-ties for each specific project. Particularlyfor the onshore part of the pre-salt gaspipelines in Brazil, the pipes are speci-fied with additional requirements forsour service application due to the pres-

ence of H2S in the gas composition. Thissituation created a big challenge for gaspipeline transportation. The use of con-ventional API 5L (Ref. 1) grades up toX65 associated with the location classfactors could result in heavy wall thick-ness pipes that are difficult to produceand are not usually used for onshorepipeline construction. The solution

found to overcome this challenging sce-nario was the use of X70 and X80 sourservice high-strength steels. High-strength steels such as X70and X80 grades have been employed inhigh-pressure pipelines to avoid theuse of thicker pipes, allowing the eco-nomic construction of long distancepipelines. However, the production ofX70 and X80 for sour service applica-tion is a big challenge that requires amore complex production route to en-sure resistance against sulfide stresscracking (Refs. 2, 3). To apply these pipes in onshorepipeline construction, in addition tothe challenge of producing a sour serv-ice resistance steel, it is necessary tocarefully evaluate the girth welding ef-fects on heat-affected zone (HAZ) andweld metal properties. In high-strength steel welding, ahardness reduction has been observedat the HAZ, also known as HAZ soften-ing (Refs. 4, 5). It is related to the heatinput effects on the strengtheningmechanism of these steels. Since thesour service steels are produced by ac-celerated cooling and have lower levelsof alloying elements, they may also besusceptible to HAZ softening. Duan,Lazor and Taylor (Ref. 6) commentedthat the softening effect in the HAZwill potentially cause highly localizeddeformation, which is undesirable inconditions where strain-based design isapplicable. On the other hand, depend-ing on the heat input, the HAZ hard-

WELDING RESEARCH

OCTOBER 2016 / WELDING JOURNAL 363-s

SUPPLEMENT TO THE WELDING JOURNAL, OCTOBER 2016Sponsored by the American Welding Society and the Welding Research Council

Girth Welding of API 5L X70 and X80 Sour Service Pipes

The results of this investigation provide a better understanding of welding X70 and X80 pipe developed for sour service

BY G. Z. BATISTA, L. DA P. CARVALHO, M. S. DA SILVA, AND M. P. SOUZA

ABSTRACT The API 5L X70 and X80 pipes development for sour service application isvery recent. Limited information can be found in literature about the girth weld­ing of these materials. This investigation presents the hardness, toughness, ten­sile, and sour resistance results of welded joints produced on API 5L X70 andX80 pipes developed for sour service. The joints were welded with the gas metalarc welding process for the root pass and flux cored arc welding process for thefill passes. The test methodology included tensile, bending, and nick­break testsas required by the API 1104 standard. The joints were also submitted to Charpy,crack opening displacement (CTOD), hardness, and stress corrosion tests. Alljoints met the requirements defined by API 1104 for tensile, bending, and nick­break tests. The Charpy tests presented high values of absorbed energy. Theweld metal and HAZ of X70 and X80 presented good toughness according to theCTOD results, although one specimen for X80 exhibited low CTOD. All joints metthe requirements of the stress corrosion cracking test. Considering all results,the weldability study demonstrated that both X70 and X80 sour service pipeswere appropriate for sour service applications. For X80, it is recommended toperform an in­depth analysis of the heat­affected­zone (HAZ) toughness beforethe use of this material. The results presented here can be used for field girthwelding of X70 and X80 sour service pipes and provides information leading to abetter comprehension of the welding of these recently developed materials.

KEYWORDS • Sour Service • Pipe • Girth Weld • API 1104 • Crack Tip Opening Displacement(CTOD) • High­Strength Steel

G. Z. BATISTA is a metallurgical engineer and L. DA P. CARVALHO is a mechanical engineer with Petrobras, Rio de Janeiro, Brazil. M. S. DA SILVAis a metallurgical engineer with ATNAS, Rio de Janeiro, Brazil. M. P. SOUZA is a mechanical engineer with Tenaris, Rio de Janeiro, Brazil.

Batista -- 10-16_Layout 1 9/14/16 11:20 AM Page 363

Page 2: Girth Welding of API 5L X70 and X80 Sour Service Pipes · PDF fileThe API 5L X70 and X80 pipes development for sour service application is very recent. Limited information can be found

ness can increase to values above thelimits for sour service application. Another fact that needs to be stud-ied in the HAZ is the possibility of em-brittlement due to either grain coars-ening or brittle phase formation. Pep-pler et al. (Ref. 7) commented that alow toughness can be expected in thecoarse-grain HAZ due to grain growth.Furthermore, Terada et al. (Ref. 8)commented that the HAZ toughnesscan be reduced by the presence of brit-

tle phases, including the formation ofmartensite-austenite (MA), which ac-cording to Babu et al. (Ref. 9) may de-teriorate or improve the properties de-pending on its morphology or concen-tration of carbon. For weld metal, it isimportant to evaluate if the highstrength levels required do not impairthe toughness or increase the hard-ness above the limits for sour serviceapplication. Although studies on high-strength

steel welding are easily found in theliterature, the same cannot be saidabout high-strength steel for sourservice, especially for heavy wall thick-ness pipes. This investigation showshow the girth welding of API 5L X70and X80 pipes developed for use insour environments affects weld metaland HAZ properties. The test method-ology included tensile, nick-break, sidebend, Charpy, crack tip opening dis-placement (CTOD), hardness, andstress corrosion tests. The results pro-vide technical information to supportthe application of these materials foronshore gas pipelines from presaltfields.

Material and ExperimentalProcedure

Pipe Material

The plates for pipe manufacturewere produced by a thermomechani-

WELDING RESEARCH

WELDING JOURNAL / OCTOBER 2016, VOL. 95364-s

Fig. 1— Bevel dimensions (ASME B31.8, Ref. 12).

Fig. 3 — Hardness indentation locations (ISO 15156­2, Ref. 15).

Fig. 2 — X70 pipe welding.

Fig. 4 — Specimen orientation and notches position(based on DNV­OS­F101, Ref. 17).

Table 1 — Chemical Composition

C Mn Si Nb V Ti Cr Cu Mo X70 0.04 1.23 0.237 0.048 0.002 0.010 0.32 0.007 0.09 X80 0.05 1.62 0.270 0.033 0.005 0.013 0.17 0.011 0.11

Ca Al N P S Ni B CE IIW Pcm X70 0.0016 0.03 0.009 0.007 0.002 0.366 0.0002 0.35 0.14 X80 0.0018 0.03 0.007 0.009 0.003 0.032 0.0003 0.38 0.16

CEIIW = C + (Mn/6) + [(Cr + Mo + V)/5] + [(Ni + Cu)/15] (Ref. 1).CEPcm = C + (Si/30) + (Mn/20) + (Cu/20) + (Ni/60) + (Cr/20) + (Mo/15) + (V/10) + 5B (Ref. 1).

Batista -- 10-16_Layout 1 9/14/16 11:20 AM Page 364

Page 3: Girth Welding of API 5L X70 and X80 Sour Service Pipes · PDF fileThe API 5L X70 and X80 pipes development for sour service application is very recent. Limited information can be found

cal-controlled process followed by ac-celerated cooling to produce a fine-grain microstructure. The chemicalcomposition (Table 1) was specially de-signed to provide a low amount of in-clusions and segregation, and thus en-sure resistance to stress-corrosioncracking and hydrogen-induced crack-ing. According to Kalwa (Ref. 2), thesteel works have to provide a leanchemistry avoiding alloying elementsthat tend to segregate, which wouldgive absorbed hydrogen the chance torecombine to molecular H2. It is important to use low-C andlow-Mn steel. A low-Mn content pre-

vents the formationof MnS inclusionsand a low-C contentof below 0.05% con-tributes to a more ho-mogeneous mi-crostructure by low-ering the volume frac-tion of pearlite andreducing the severityof centerline segrega-tion (Ref. 10). The Scontent has to be re-duced to values below0.001%, and the steel

must be Ca treated to control the re-maining inclusions. In addition, central segregation shallbe avoided during the casting process,which is also minimized by keeping thelowest possible content of the elementsC, Mn, S, and P (Refs. 2, 11). To com-pensate for the low levels of C and Mn,alloying elements must be added, andan accelerated cooling technique needsto be used. The target is to achieve a re-fined ferritic or ferritic/bainitic mi-crostructure with high strength and re-sistance to sulfide stress cracking andhydrogen-induced cracking. The API 5L X70 and X80 pipes were

formed by the UOE process, where theplate is conformed to an U shape fol-lowed by an O shape, joined by sub-merged arc welding, and then cold ex-panded (E), according to API 5L stan-dard (Ref. 1). The diameter and thewall thickness were 20 ¥ 1 in. (508 ¥25.4 mm) for X70 and 24 ¥ 1.25 in.(610 ¥ 31.8 mm) for X80.

Welding

The pipes were cut in rings around300 mm in length and beveled accord-ing to Fig. 1, taken from ASME B31.8(Ref. 12). The girth welding was donewith the pipe fixed in the horizontalposition. The gas metal arc welding (GMAW)process using short circuit transferwas used for the root pass due to itshigher productivity and the thickerweld metal that can be deposited withone pass (usually around 4 mm), whencompared with processes such asshielded metal arc welding (SMAW) orgas tungsten arc welding (GTAW). Forfill and cap passes, the gas-shieldedflux cored arc welding (FCAW) processwas applied since it also has good productivity. Consumables with differentstrength levels were selected accordingto manufacturer’s recommendation tomeet the strength requirements ofeach pipe grade and also to complywith sour service requirements (Table2). Procario and Melfi (Ref. 13) com-mented that for sour service applica-tions, weld metal alloy systems shouldbe carefully selected to provide a bal-ance between toughness and hardness.The chemical composition of the con-sumables can be seen in Table 3. Thevalues were obtained from the certifi-

WELDING RESEARCH

OCTOBER 2016 / WELDING JOURNAL 365-s

Fig. 5 — Four­point loading method (ISO 7539­2, Ref. 18).

Fig. 7 — X70 weld metal microstructure.

Fig. 6 — Fatigue precrack position (BS 7448, Ref. 20).

Table 2 — Processes and Consumables

Pipe Pass Process Progression Shielding Gas Consumable Diameter (mm)

Root GMAW Downhill 80% Ar/20%CO2 AWS A5.18 ER70S­6 1.2 X70 Fill/ FCAW Uphill 80% Ar/20%CO2 AWS A5.29 1.2 Cap E81T1­Ni1C

Root GMAW Downhill 80% Ar/20%CO2 AWS A5.18 ER70S­6 1.2 X80 Fill/ FCAW Uphill 80% Ar/20%CO2 AWS A5.29 Cap E101T1­G 1.2

Batista -- 10-16_Layout 1 9/14/16 11:20 AM Page 365

Page 4: Girth Welding of API 5L X70 and X80 Sour Service Pipes · PDF fileThe API 5L X70 and X80 pipes development for sour service application is very recent. Limited information can be found

cates issued by the manufacturers andcorrespond to the chemical analysis ofthe same heat of the consumable usedin the tests. For each consumable, thechemical analysis was performed ac-cording to the requirements of the ap-plicable AWS standard. Figure 2 shows the API 5L X70 pipeduring welding. The pipe girth weldingwas done with two welders represent-

ing the welding conditions found inonshore pipeline construction. Weld-ing parameters are seen in Table 4. Apreheating of 100°C and interpasstemperature of 175°C were also applied. After welding, the coupons were in-spected by radiographic examinationand ultrasonic tested, and they met therequirements of API 1104 (Ref. 14).

Mechanical Tests and Metallographic Analysis

Specimens were taken from theweld coupons to perform metallo-graphic analysis and mechanical testsof tensile, nick break, and bending, re-quired by API 1104 (Ref. 14), and alsoCharpy, hardness, CTOD, and stresscorrosion testing.

WELDING RESEARCH

WELDING JOURNAL / OCTOBER 2016, VOL. 95366-s

Fig. 8 — X80 weld metal microstructure. Fig. 9 — Tensile results.

Fig. 11 — Weld metal CTOD.Fig. 10 — HAZ CTOD.

Table 3 — Consumables Chemical Composition

Process Grade Consumable C Mn Si P S Ni Cr Mo Cu V

X70 0.140 1.52 0.89 0.016 0.010 0.010 0.016 0.028 0.033 0.001 GMAW ER70S­6 X80 0.089 1.44 0.84 0.016 0.014 — — — — — X70 E81T1­Ni1C 0.046 1.47 0.41 0.010 0.013 0.820 0.030 0.020 — 0.02 FCAW X80 E101T1­G 0.050 1.57 0.38 0.010 0.020 0.900 0.040 0.170 — 0.02

Batista -- 10-16_Layout 1 9/14/16 11:21 AM Page 366

Page 5: Girth Welding of API 5L X70 and X80 Sour Service Pipes · PDF fileThe API 5L X70 and X80 pipes development for sour service application is very recent. Limited information can be found

Samples for metallographic analysiswere etched with nital 2%, and the im-ages were taken with magnification of500¥ by optical microscopy. The test methods and samples fortensile, nick break, and bending testsof the girth welds were done accordingto API 1104 (Ref. 14) requirements. A Vickers hardness test was per-formed with a load of 10 kgf. The spec-imen, removed from the 6 o’clock posi-tion, was prepared according to Fig. 3extracted from ISO 15156-2 (Ref. 15)that defines the hardness require-ments for sour service. Charpy impact test specimens weresampled at 2 mm from the inner surface(close to the root pass) and at 2 mmfrom the outside surface (close to thecap pass) of the pipe for both weld met-al and HAZ. The specimens were re-moved from both one upper and one

bottom pipe quadrant. Three full-sizespecimens of 10 ¥ 10 ¥ 55 mm weretested according to ASTM A370 (Ref.16) for each position. The test was car-ried out at 0ºC, which is below the mini-mum design temperature for onshorepipelines. Figure 4, based on DNV-OS-F101 (Ref. 17), shows the notch posi-tions for the girth weld Charpy test. The stress corrosion test of threespecimens was performed according toISO 7539-2 (Ref. 18) standard usingthe four-point loading method — Fig.5. During the test, a stress level corre-sponding to 80% of the actual yieldstrength of the pipe was applied. Thestressed samples were exposed to thetest solution B of NACE TM 0284standard (Ref. 19) for 30 days withcontinuous bubbling of H2S. This solu-tion was chosen because its severity iscloser to that recommended by ISO

15156-2 (Ref. 15) to test an opera-tional condition similar to the presaltgas characteristics. The girth weld rootwas placed in the tensioned area of thespecimen and was not machined totest the real weld profile found in cir-cumferential weld roots. For CTOD tests, single-edge,notched bend specimens were used ac-cording to BS 7448 (Ref. 20). The weldmetal test was conducted on a throughthickness specimen of rectangular sec-tions (Bx2B) with crack plane orienta-tion correspondent to NP, where Nmeans normal to weld direction and Pmeans crack propagation parallel toweld direction, in accordance with BS7448 (Ref. 20). The HAZ test was con-ducted on surface-notched specimensof square sections (BxB) with crackplane orientation corresponding toNQ, where N means normal to welddirection and Q means crack propaga-tion in the weld thickness direction ac-cording to BS 7448 (Ref. 20), to placethe fatigue precrack at the graingrowth region of the HAZ — Fig. 6. Nine specimens were tested for

both the weld metal and HAZ, threefrom the 12 o’clock position, threefrom the 6 o’clock position, and threefrom the 3 o’clock position. The testtemperature was –10°C for X70 and0°C for X80. In fact, the test tempera-ture depends on the minimum tem-perature of the environment wherethe pipe will be applied. During theX80 CTOD tests, the minimum tem-perature had been set to 0°C, later on,it was decided to adopt a more conser-vative condition for the X70 pipe, andthe test temperature was resetto –10°C.

WELDING RESEARCH

OCTOBER 2016 / WELDING JOURNAL 367-s

Fig. 11 — Weld metal CTOD.Fig. 10 — HAZ CTOD.

Table 4 — Process Parameters

Pipe Pass Number Polarity Voltage Current Speed Heat Input of Passes (V) (A) (mm/s) (kJ/mm)

Root 1 CC + 15–16 140–190 1.6–2.1 0.9–1.8 X70 Fill 22 CC + 20–23 180–220 1.8–4.6 0.7–2.3 Cap 5 CC + 21–23 180–200 3.0–5.5 0.8–1.0

Root 1 CC + 15–17 120–140 1.6–2.1 0.8–1.5 X80 Fill 25 CC + 21–23 190–220 2.1–5.5 0.9–2.1 Cap 5 CC + 21–23 190–210 3.3–4.7 0.9–1.2

Table 5 — Stress Corrosion Cracking Testing Results

Pipe Grade pH before H2S Saturation / Result pH at End of Test

X70 8.14/5.25 No cracks X80 8.17/5.23 No cracks

Batista -- 10-16_Layout 1 9/14/16 11:21 AM Page 367

Page 6: Girth Welding of API 5L X70 and X80 Sour Service Pipes · PDF fileThe API 5L X70 and X80 pipes development for sour service application is very recent. Limited information can be found

Results and Discussion The weld metal microstructure ofX70 and X80 can be seen in Figs. 7 and8, respectively. The X70 microstruc-ture presented acicular ferrite, grainboundary ferrite, and polygonal fer-rite, while the X80 presented a higheramount of acicular ferrite and alsosome polygonal ferrite. The results of the weld tensile testfor each quadrant are shown in Fig. 9.The horizontal dashed line expressesthe minimum acceptable values of 570MPa for X70 and 625 MPa for X80. Eight side bending specimens foreach joint were tested. No one speci-men presented a crack or other discon-tinuity greater than 3 mm, so that allspecimens of both materials met therequirements for the bending test. Forthe nick-break test, four specimensfrom each joint were tested, one foreach quadrant, and all of them met theAPI 1104 (Ref. 14) acceptance criteria.

Considering the results of tensile,bending, and nick-break tests, thewelding procedure was consideredqualified according to API 1104 (Ref.14). However, to meet the specific re-quirements for presalt gas pipelinesand perform a deep evaluation, othertests such as Charpy, CTOD, hardness,and stress corrosion were necessary. Charpy and CTOD tests were car-ried out to verify the weld metal andHAZ toughness. Figure 10 presentsthe CTOD results of the HAZ and Fig.11 the CTOD of the weld metal. Onlythe CTOD valid specimens were plot-ted so that less than three results mayappear for some positions. The hori-zontal dashed line represents the min-imum acceptable value of 0.15 mmrecommended by Hopkins and Denys(Ref. 21) to allow the use of a fitness-for-purpose criteria. This minimumvalue is required to ensure that plasticcollapse equations can predict defectfailure stresses. This provides more

possibilities from the engineeringpoint of view to deal with situationswhere specific analysis is necessary,such as defects found during in-service inspection and to avoid unnec-essary repairs (Ref. 21). Figure 10 shows that one X80 spec-imen presented a CTOD value below0.15 mm for the HAZ. Barnes (Ref. 22)commented that local brittle zones ex-ist within the HAZ, especially withinthose regions of the grain-coarsenedHAZ, reheated into the intercriticalregime by subsequent weld passes.These brittle zones are formed due tothe formation of MA constituent. According to Barnes (Ref. 23), thescatter observed in the test data can beexplained by the inhomogeneity of themicrostructure and by the fact that thelocal brittle zones are small so that thecrack tip can be, or not, at these regions. Barnes (Ref. 23) suggested that thethermal simulation techniques can re-duce the inherent variability and allow

WELDING RESEARCH

WELDING JOURNAL / OCTOBER 2016, VOL. 95368-s

Fig. 12 — X70 weld metal CTOD. Fig. 13 — Weld metal microstructure of X70 welded with 100%CO2.

Fig. 14 — Charpy absorbed energy of weld metal. Fig. 15 — Charpy absorbed energy for HAZ.

Batista -- 10-16_Layout 1 9/14/16 11:21 AM Page 368

Page 7: Girth Welding of API 5L X70 and X80 Sour Service Pipes · PDF fileThe API 5L X70 and X80 pipes development for sour service application is very recent. Limited information can be found

close matching of the thermal condi-tions and microstructure at various lo-cations in a multipass HAZ. This, inturn, allows easier and more accuratetesting and examination. To perform a more accurate evalua-tion, a study based on the API RP 2Z(Ref. 24) standard could be used to eval-uate the HAZ toughness with differentheat inputs. Moreover, a microstructur-al evaluation would help to check if thepresence of MA in the HAZ is contribut-ing to a toughness reduction. The weld metal CTOD values pre-sented at Fig. 11 are very close to theminimum recommended value of 0.15mm. Generally, the alloying elementsadded to meet high strength levels in-creases the challenge to meet hightoughness requirements for weld metal.Felber (Ref. 25) observed that to meetthe high strength requirements it isnecessary to supply a weld metal withadditions of nickel and molybdenum.These elements are included in thechemical composition of E81T1-Ni1Cand E101T1-G used for X70 and X80welding, respectively. According to Pro-cario and Melfi (Ref. 13), althoughmolybdenum is used to increase tough-ness in the as-welded condition by pro-moting the formation of acicular ferrite,it can reduce the toughness when re-heated because these reheated regionsdo not retain their fine-grained acicularferrite microstructure. For X70 grade, some results werebelow the minimum. One alternativeto improve weld metal toughness for

X70 is changing the shielding gas com-position from the mixture of 80% ar-gon and 20% CO2 to 100% CO2. Alloy-ing elements used for deoxidizationsuch as Mn and Si can remain in theweld metal when the gas mixture of Arand CO2 is used and therefore reducetoughness (Ref. 26). The use of ashielding gas of 100% CO2 reduces theamount of Mn and Si and also increas-es the oxygen level in the weld metal,promoting inclusion formation thatcan increase the volume fraction ofacicular ferrite and improve tough-ness. In order to test the effect ofshielding gas, another X70 weldedjoint was produced with a shieldinggas of 100% CO2. Figure 12 shows thatthe CTOD results were improved, andFig. 13 shows that a large amount ofacicular ferrite was obtained, whencompared with Fig. 7. Charpy results are shown in Figs.14 and 15 for the weld metal and theHAZ, respectively. As the onshorepipeline welding standard API 1104(Ref. 14) does not specify Charpy re-quirements for a girth weld, the valueof 45 J specified by DNV-OS-F101(Ref. 17) for X80 will be used as a ref-erence. This value is represented by ahorizontal dashed line. For both X70 and X80, the ab-sorbed energy values were above theminimum required. The HAZ present-ed higher absorbed energy valuesthan the weld metal. The hardness results for X70 andX80 are shown in Fig. 16. The hori-

zontal dashed line represents the lim-it of 250 HV defined by ISO 15156-2(Ref. 15). From Fig. 16, it can be noted thatall hardness impressions do not ex-ceed the limit of 250 HV10 definedby ISO 15156-2 (Ref. 15). Figure 16 also shows an HAZ soft-ening, which is more pronounced forX80. According to Denys and Lefevre(Ref. 4), the susceptibility to HAZsoftening increases with the reduc-tion of alloying elements and with in-creasing heat input. Due to the lowamount of alloying elements in sourservice steels, hardness reduction was expected in this region. Accord-ing to Duan, Lazor, and Taylor (Ref.6), it can be critical for submerged arcwelding where the softening effect in the HAZ will cause highly localizeddeformation, which is undesirablewhere strain-based design is applica-ble. Duan, Lazor, and Taylor (Ref. 6)also commented that a narrow HAZ isnot expected to have a high level oflocal deformation due to the high lat-eral constraint from the adjacentparts of the weld metal and the unaf-fected pipe body. As the materialsevaluated in this study were designedfor use in a stress-based design, andthe heat input is lower than that fromthe submerged arc welding process, itcan be concluded that the HAZ soft-ening found is not critical for X70 andX80 sour service pipe welding. The stress corrosion cracking testfor both X70 and X80 girth weldsconcluded that the weld joints aresuitable for sour service. After 30days immersed in a test solution, nocracks were found. Table 5 summa-rizes the test results. As mentioned by Procario andMelfi (Ref. 13) for sour service appli-cation, the weld metal chemical com-position must be carefully selected toprovide a good balance betweentoughness and hardness. Further-more, for X70 and X80 grades, thewelded joint must also meet the highstrength levels required. The tests re-sults obtained in this study show thatthe girth welding of high-strengthsteel sour service pipes is feasible.

Conclusions The weldability study using the

WELDING RESEARCH

OCTOBER 2016 / WELDING JOURNAL 369-s

Fig. 16 — X70 and X80 hardness.

Batista -- 10-16_Layout 1 9/14/16 11:21 AM Page 369

Page 8: Girth Welding of API 5L X70 and X80 Sour Service Pipes · PDF fileThe API 5L X70 and X80 pipes development for sour service application is very recent. Limited information can be found

GMAW process for the root pass andthe FCAW process for fill passes wereevaluated for both X70 and X80 sourservice pipes. According to API 1104requirements, the welding procedurescan be considered approved. The X80 HAZ presented lowtoughness according to one specimenevaluated by CTOD tests. In fact, theCTOD test is not required to meet theworkmanship criteria of API 1104.However, a low toughness can restricta fitness- for-purpose analysis toavoid unnecessary repairs of defectsfound during in-service inspection oreven the pipeline construction. It isrecommended to perform an in-depthevaluation to identify the reasons be-hind this low CTOD value presented. The X70 weld metal presented lowtoughness when welded with the mix-ture of 80% Ar and 20% CO2. Howev-er, the shielding gas of 100% CO2 in-creased the CTOD results to valuesabove 0.15 mm. According to the liter-ature, this improvement can be attrib-uted to two main factors: the reduc-tion of deoxidizing elements, such asMn and Si in the weld metal, and theincreased volume fraction of acicularferrite due to inclusions formationcaused by the increased oxygen level inthe weld metal. The Charpy impact test presentedhigh values of absorbed energy and allspecimens met the requirements. The stress corrosion cracking testdemonstrated that the welded jointsare appropriate for sour service appli-cation. Both the welding parametersand consumables applied showed to besuitable for X70 and X80 girth weldingfor sour service. Considering all results presented,the weldability study demonstratedthat both X70 and X80 sour servicepipes are appropriate for sour serviceapplications.

This paper presented the results ofa weldability study applied in X70 andX80 pipes developed for sour service.The results obtained can be used forfield girth welding of these materialsand provides information that leads toa better comprehension of the welding

of these recently developed materials. The authors wish to acknowledgePetrobras and Tenaris for the financialand technical support, and Senai-CTSfor the welding of test joints. The au-thors also would like to thank RichardBravo and Harold Leon, Tenaris, Mar-cy Saturno de Menezes, Petrobras, andSuelen Navarro, Consulpri for theircontributions, technical guidance, andsupport provided during this research.

1. API 5L. 2012. Specification for LinePipe, API Specification 5L. American Petro-leum Institute. 45th Edition. 2. Kalwa, C., and Hillenbrand, H. G.2012. Europipe’s experience and develop-ments on pipe material for sour service ap-plications. International Seminar MicroalloyedSteels for Sour Service. São Paulo, Brazil. 3. Ishikawa, N. et al. 2012. Material de-sign of high strength/heavy gaugelinepipes for sour service. InternationalSeminar Microalloyed Steels for Sour Service.São Paulo, Brazil. 4. Denys, R. M., and Lefevre, T. 2000.Effects of Welding on HAZ Softening ofX70/X80 TMCP Linepipe Steels. Pipeline Re-search Council International. 5. Chen, Y., and Wang, Y. Y. 2007. Mi-crostructure modeling of HAZ softening inmicroalloyed high strength linepipe steels.Proceedings of the Seventeenth InternationalOffshore and Polar Engineering Conference.Lisbon, Portugal. 6. Duan, D., Lazor, R., and Taylor, D.2010. Evaluation of double jointing girthwelds of high grade line pipes. 8th Interna-tional Pipeline Conference. Calgary, Alberta,Canada. 7. Peppler, E. et al. 2009. Suitable HAZtesting to predict linepipe safety. PipelineTechnology Conference. Ostend, Belgium. 8. Terada, Y. et al. 2004. High strengthlinepipes with excellent HAZ toughness.Nippon Steel Technical Report, No. 90. 9. Babu, S. S. et al. 2009. A generalizedmicrostructure model for microalloyedhigh strength pipeline weldments. PipelineTechnology Conference, Ostend, Belgium. 10. Stallybrass, C. et al. 2012. Alloy de-sign for UOE linepipe material for stan-dard and non-standard HIC conditions. In-ternational Seminar Microalloyed Steels forSour Service. São Paulo, Brazil. 11. Nieto, J. et al. 2012. Process andquality controls for production of line pipeslabs for sour service applications atArcelorMittal Lazaro Cardenas, Mexico. In-ternational Seminar Microalloyed Steels for

Sour Service. São Paulo, Brazil. 12. ASME B31.8. 2010. Gas Transmis-sion and Distribution Piping Systems. TheAmerican Society of Mechanical Engineers. 13. Procario, J. R., and Melfi, T. 2011.Weld metal alloy systems for seam weldingof niobium micro-alloyed pipe steels. Weld-ing of High Strength Pipeline Steels Interna-tional Seminar. 14. API 1104. 2010. Welding of Pipelinesand Related Facilities. American PetroleumInstitute. 20th Edition. 15. ISO 15156-2. 2009. Petroleum andNatural Gas Industries — Materials for Usein H2S-Containing Environments in Oil andGas Production. Part 2: Cracking ResistantCarbon and Low-Alloy Steels, and the Use ofCast Irons. 2nd Edition. 16. ASTM A370. 2010. Standard TestMethods and Definitions for Mechanical Test-ing of Steel Products. American Society forTesting and Materials. 17. DNV-OS-F101. 2012. SubmarinePipeline Systems. Det Norske Veritas. 18. ISO 7539-2. 1989. Corrosion of Met-als and Alloys — Stress Corrosion Testing.Part 2: Preparation and Use of Bent-BeamSpecimens. 1st Edition. 19. NACE TM 0284. 2011. Evaluation ofPipeline and Pressure Vessel Steels for Resist-ance to Hydrogen-Induced Cracking. NationalAssociation of Corrosion Engineers. 20. BS 7448. 1991. Fracture MechanicsToughness Tests. British Standard. 21. Hopkins, P., and Denys, R. 1993.The European pipeline research group’sguidelines on acceptable girth weld defectsin transmission pipelines. Eighth Sympo-sium on Line Pipe Research. 22. Barnes, A. M. 1995. The Effect ofThermomechanically Controlled Processing ofSteels on Heat Affected Zone Properties. TheWelding Institute. 23. Barnes, A. M. 1990. Local brittlezones in C-Mn steel multipass welds. TWIBulletin. 24. API RP 2Z. 2005. RecommendedPractice for Preproduction Qualification forSteel Plates for Offshore Structures. API Rec-ommended Practice 2Z. American PetroleumInstitute. Fourth Edition. 25. Felber, S. 2009. Pipeline Engineering.Welding Research Council. 26. The James F. Lincoln Arc WeldingFoundation. 2000. The Procedure Handbookof Arc Welding. 14th Edition.

WELDING RESEARCH

WELDING JOURNAL / OCTOBER 2016, VOL. 95370-s

References

Acknowledgments

Batista -- 10-16_Layout 1 9/14/16 11:21 AM Page 370