N. Ames*, C. Frye**, K. Larsen***

* Edison Welding Institute, USA

** ExxonMobil, USA

*** Sandvik Materials Technology, Sweden

Keywords: Super Duplex, corrosion, GTAW, PE-GTAW, welding, orbital.

1. Introduction

The oil & gas industry uses a significant amount of small (<1.5-in) and intermediate (1.5-in to 6-in) diameter super duplex stainless steel (SDSS) tubing. One of key applications for these sizes of super duplex tubing is umbilical manufacturing. An umbilical can be defined as the lifeline for a subsea system. It supplies the necessary controls, through hydraulic fluid, electrical cables and fiber optics, for the subsea field as well as a vehicle to utilize gas lift-typically through the use of a larger diameter center tube.

Umbilcals are often exposed to the same harsh environment as the system it is being used to control and/or monitor. The SDSS tubing utilized in umbilical applications must be capable of enduring the high external pressures present on the ocean floor, high working pressures of the fluids inside them and withstanding the various types of corrosion that commonly occur in subsea systems. Industry specifications require that welded connections be utilized throughout the umbilical. These welds must perform equal to or better than the tubing they are joining.

The gas tungsten arc welding (GTAW) process is widely accepted and typically specified for welding of the tube to tube joints on umbilicals. GTA welding produces high-quality weld deposits, precise control of welding parameters, and has relatively low equipment costs. It is capable of producing full penetration single pass welds and high quality multipass welds on thick cross section weldments. Orbital GTAW, an “automatic” variant of manual GTAW, is a widely accepted method for creating repeatable, clean, high-quality, and documented welded connections on tubing. Many umbilical fabricators are now using orbital GTAW to join tubes during the fabrication of the umbilicals.

As the oil and gas industry transitions to deeper and deeper water, umbilical designs become increasingly difficult to fabricate. The operating pressures for deepwater umbilicals dictate that they are manufactured using increasingly heavy wall tubing. When welding heavier wall tubing, the ability to produce single pass full fusion welds while maintaining acceptable bead profiles becomes increasingly difficult. This often causes a shift from single to multi-pass welding. As multiple passes are introduced it becomes meeting the performance criteria (microstructure, mechanical properties, fatigue life and corrosion resistance) becomes more challenging.

Over the last five years, several advancements in the welding of SDSS have been made to help combat the increased welding difficulties. One of the most notable is the use of the SS-7 penetration enhancing (PE) compound to aid the GTAW process in controlling the weld metal microstructure, as well as mechanical and corrosion properties of SAF 2507.

The PE-GTAW process is still relatively new to the oil and gas industry and has only recently begun to see acceptance and implementation by major oil and gas companies. As a result of welding difficulties on recent gas lift umbilical projects, the PE-GTAW process has been investigated for use on SAF 2507 umbilical center tubes. Despite thousands of meters of butt-welded duplex and super duplex tubing, recent production trials have shown that weldments having

WTessWallThickn

== .004→ .0

ODOutsideDiameter

exhibit a reduced corrosion resistance in weld metal and heat affected zone (HAZ). This has been attributed to the ferrite content and the presence of intermetallics. High ferrite contents are typically caused by too fast of a cooling rate and reduced nitrogen content, effectively prohibiting the ferritic microstructure ample time and/or the driving force to transform to austenite. Intermetallics are undesirable brittle precipitates that form during slow cooling of high chromium containing stainless steels. The intermetallics

(σ and Χ) form when the cooling rate is sufficiently slow and are commonly seen in multipass welds.

2. Objective

The objective of this research initiative was to develop and deploy an alternative welding technique; capable of being implemented in a typical umbilical production environment, that yields improved corrosion properties and that meets all the mechanical and fatigue performance criteria for the umbilical. This was to be accomplished using off the shelf welding equipment that is readily available to the umbilical manufacturers. The goal was to complete all welding and testing in a laboratory environment then reproduce it using equipment and manpower within a leading umbilical manufacturing facility.

3. Experimental Procedure

The first step was to assess the current welding practices and equipment utilized by industry to weld SAF 2507. Based on this assessment, use commensurate welding practices and equipment to develop and optimize weld procedures utilizing the PE-GTAW process. When possible the procedures should be designed around a square butt joint thus reducing the time and cost to machine complicated end preparations on the tubing for welding. The final tube size to be evaluated in this program was selected based on availability of comparable test data (2.78-in outside diameter by 0.140-in wall thickness). This corresponded with a size utilized in a recent umbilical, which had sufficient data available and was problematic from a welding standpoint.

Upon development of an acceptable weld procedure, the PE-GTA welding technique would be transferred to an umbilical fabrication facility. Following training of the production welding personnel, the welding procedures would be dialed in to work on the production equipment in their facility. Following welding in the laboratory and in the production environment, all weld coupons would be evaluated in accordance with ASME Section IX, to ensure they met the minimum acceptance criteria for SAF 2507. All testing was duplicated by two independent laboratories to ensure that the results were unbiased. In addition to the required weld qualification testing, additional work was done to validate acceptable weld microstructure was attained. Upon completion of qualification testing and microstructural evaluation, all welds were subjected to corrosion testing. The number of weld from each facility was selected such that statistically sound data would be expected from the corrosion testing.

Two corrosion tests were employed, ASTM G 48 Method A and ASTM A 293 (often refered to as modified ASTM G 48). Both tests were to be completed using standard industry practices for determining corrosion resistance. ASTM G 48 Method A employs a 24 hr immersion into 6% ferric chloride at 40°C. For SAF 2507, acceptable performance is determined as a minimum weight loss of 1.5 g/mm² and no visible signs of pitting when examined using a 10x magnification. The ASTM A 293 employs the same 6% ferric chloride solution with 2.5°C (or 5°C) temperature steps between weight loss and pitting measurements, each incremental temperature is maintained for 24 hours.

4. Discussion and Results

4.2 Corrosion Testing

Corrosion resistance is always a concern for umbilicals since the average design life is a minimum of 25 years. The fact that the weight of the umbilical is typically supported by a vessel or oil platform means the wall thickness of the umbilical tubing must be minimized in order to reduce the weight (and cost) exerted on the floating vessel. By reducing the wall thickness, the corrosion allowance must be reduced if not eliminated completely. Umbilicals are thus designed to operate under zero corrosion conditions, which demand the use of super alloys like SAF 2507. To assess the corrosion resistance of SAF 2507 and other SDSS, which are nearly impervious to seawater corrosion, an accelerated, more aggressive test must be used. Most standards and specifications require either ASTM A 293. In this test the weld metal and surrounding HAZ are submerged in ferric chloride at incrementally increasing temperatures. The temperature at which corrosion initiates is termed the critical pitting temperature (CPT). Most global subsea specifications require a CPT of at least 40�C. Many specifications allow ASTM G 48 at 40°C to be used in place of A 293. If the weld performs acceptable at the specified CPT, then no additional test temperatures are required.

Corrosion testing provides significant insight into the ability of the weld metal, heat-affected zone (HAZ), and base metal to perform acceptably in-service. While the test solutions may not be the same as the subsea conditions, their correlation to real-world has historically been applied as a means of ranking materials. The 6% ferric chloride solution provides an electrolyte similar to that of seawater, but with accelerated corrosion, allowing for the corrosion testing to be completed in a relatively short period of time.

The acceptance criteria (1.5-g per cm² of weight loss at 40�C) was easily met by the PE-GTAW weldments. The specific CPT for the PE-GTAW welds was determined to be 56°C (see Table 2 and Figure 1). Similar to CPT, the critical crevice temperature (CCT) is the temperature at which corrosion begins in a tight crevice in a given solution. The CCT, which is determined in the same solution and manner as CPT under an artificially induced crevice, was determined to be 37.5°C.

ExxonMobil provided data from the traditional GTAW qualification testing for the selected tube size. The qualification data met all the requirements specified by ASME and industry standards. Figure 2 graphically shows the CPT comparisons for the two welding techniques. The average CPT of the PE-GTAW process is 4.5°C higher than that of the traditional GTAW process. The standard deviation of the PE-GTA welds is 1.3 compared to that of the conventional GTA welds, which was calculated to be

3.5. Notably, more welds were analyzed for the PE-GTAW process. However, if the two most dissimilar data points from the PE-GTA welds are averaged (55°C) and used to calculate the standard deviation (3.5) they still out perform the traditional GTA welds. It can thus be assumed that had a small sample size been used for the PE-GTAW process, the process would have still outperformed the conventional GTA welds.

Based on these positive results a literature review was completed to determine some typical CPT values for comparison purposes. Two papers published by Sandvik ^[5,6] (the manufacturer of SAF 2507) detailed weld procedures and their corresponding CPT values, see Table 3 and Figure 3. The PEGTAW technique produced welds with nominal CPT values between 2% and 29% better than those produced in laboratory conditions The critical pitting temperature of super duplex welds produced using the PE-GTAW technique showed between 2% and 11% improvement in CPT over filler based welds, and a 29% improvement over autogenous welding procedures.

Interestingly, the welds produced in a laboratory setting and those produced in the production facility resulted in test results with no visible variance. This was noted as key indicated to the robustness of the PE-GTAW technique.

Additional work was done in an attempt to quantify why this improved corrosion resistance occurred. The early hypothesis was that the PE-GTAW process altered the composition sufficiently to influence austenite and ferrite equilibrium. Upon evaluation, see Table 4, it was observed that the weld metal was nearly identical to the base material and was well within the specified limits for the SAF 2507. In actuality the PE-GTAW welds had a composition closer aligned with the base metal than the anticipated composition of the traditional welds (produced with 25 Cr, 10 Ni, 4 Mo—25.10.4L filler). Previous research ^{[1, 2, 3, 4]} has shown that most duplex and super duplex welds lose a notable portion of the base metal nitrogen content during welding. As shown in Figure 4, this generalization was not true for the PE-GTA welds. It is believed that the retained nitrogen content, coupled with the small grain size (shown in Figure 4) and the reduced heat input were the main driving forces behind the improved corrosion properties.

5. Conclusions

EWI and Sandvik have been working for a number of years to assess the capabilities of the PE-GTAW process as it pertains to duplex and super duplex stainless steel welding. A majority of this research has been focused on optimization of the technique and quantification of the resultant properties. This program, however, was driven by a specific industry need to obtain higher quality welds. The current state of the art welding technology was either cost prohibitive or did not provide adequate properties. ExxonMobil teamed with EWI and Sandvik to take the PE-GTAW process to the real world. With the intent being to let the results fall out where they may and thus create a starting point for additional research and optimization.

The PE-GTAW process outperformed all expectations. The project team had hoped to produce either a higher quality weld or one with a shorter cycle time. The final results of the laboratory and field welding were such that the process yielded both higher quality, more corrosion resistant welds on SAF 2507 and did so with a shorter production cycle time than typically attainable. While ExxonMobil, Sandvik, and EWI were very happy with the results, each has committed to continue research on this topic. The ongoing work suggests that the PEGTAW process may be applicable to other avenues such as filler metal manufacture, submerged arc welding fluxes, and steel production. Work continues in these areas to push the limits of applicability, as well the continued development of new PE-GTAW applications.

References:

[1] P. Woolin: ‘Pitting Corrosion of Weldments in High-Nitrogen Austenitic Stainless Steel’, Stainless Steel World 1999 Conference.

[2] N. Ames, M. Holmquist, M.Q. Johnson: ‘Orbital Welding of Small-Bore Super Duplex Tube Using GTAW Flux’, Duplex Stainless Steel 2000 Conference, November 2000.

[3] N. Ames, M. Ramberg, M. Johnson, T. Johns: ‘Comparison of austenitic, super austenitic and super duplex weld properties produced using GTAW flux’, Stainless Steel World America Conference, 2002.

[4] S.R. Collins and P.C. Williams: ‘Weldability and Corrosion Studies of AISI 316L Electropolished Tubing’, www.swagelok.com, 1999.

[5] S-A. Fager, L. Odegard, U. Ekstrom: ‘Welding of SAF 2507’, Welding Reporter-S-WR291.

[6] C-O Pettersson, S-A. Fager: ‘Weldign Practice for the Sandvik Duplex Stainless Steels SAF 2304, SAF 2205 and SAF 2507’, R&D Centre, August 1995.