Thermal and Mechanical Considerations for Design of Insulated Tubing

Abstract

Summary Of the several options for controlling temperature in a well, vacuum-insulated tubing (VIT) has proven useful in a number of applications. Use of VIT, however, necessitates unique design considerations, from both a thermal and mechanical perspective. This paper describes the application of VIT for a particular purpose - to minimize temperature change, and associated annular fluid-expansion pressures, in a deepwater well design. The specific application, however, lends background to more general design considerations associated with VIT. Thermally, the importance of heat loss at the couplings to the overall design is discussed, using both numerical analysis and experimental results previously reported in the redesign of wells for the Marlin tension-leg platform (TLP). Mechanically, a detailed examination of the loads to which VIT may be subjected also uncovers special considerations. Introduction Modern VIT typically consists of two tubes, welded together at the ends of the shorter tube (e.g., the tube that does not contain connection threads) so as to create an isolated annulus to which a vacuum is applied. The annulus may contain a variety of materials including insulation to affect a radiation shield and getter to absorb traces of gas, thus extending the life of the vacuum. A schematic of VIT having the outer tube threaded is presented in Fig. 1. VIT can be applied to control heat in either sense, that is, to confine heat within the tube (typically for flow assurance) or, alternately, to isolate outer regions of a wellbore from thermal effects (to minimize annular pressure buildup). The condition that inspired much of the work summarized in this paper is of the latter form. Having suffered a temperature-related failure of at least two tubular strings on the first Marlin TLP well, the fact that remaining wells on the platform were batch drilled and similar in design to the damaged wellbore necessitated that the outer annuli of these wellbores be protected from large excursions in temperature.1-3 The discussion to follow focuses on specific design issues that arose during the Marlin VIT design phase. Given the promising performance of VIT in large-scale experiments,2 it then becomes necessary to extend that performance to an entire wellbore to ensure that (a) the VIT will remedy the problem of annular pressure buildup, and (b) it will remedy that problem over the long term. The remaining issue, addressed elsewhere,3 is verifying that performance by field monitoring. VIT Configurations The insulated tubing discussed in this paper consists of two concentric tubes joined/sealed near either end with a weld. This construction raises the question as to which tube should host the threaded connection. Table 1 summarizes the primary advantages of each configuration option. Both configurations have appeared in well designs. The Marlin choice of a connection on the outer tube will influence some of the discussion to follow. Equations derived in the Appendix, however, can be applied to either configuration. Thermal Design An important step in designing a VIT completion is to verify, within the limits of one's modeling capabilities, that insulated tubing is actually a solution. For the Marlin wells, thermal design involved substitution of experimental results2 into a numerical wellbore simulator to ensure that thermally generated pressures in outer annuli would be sufficiently low to guarantee the survival of the tubular ensemble. Although the simulations executed for Marlin encompass the entire wellbore,3 here we shall concentrate on the all-important A annulus (e.g., tubing by production casing/tieback). Not only must the temperature in this annulus be constrained, it must be constrained at long time (steady state). This requirement for sustained temperature control places particular emphasis on natural convection. That is, given that the heat loss at the connection will be greater than opposite the vacuum chamber, confining the associated temperature increase to the vicinity of each connection is crucial. Connection Heat Loss. Fig. 2 displays the results of an axisymmetric thermal analysis of a VIT joint in the neighborhood of its threaded and coupled proprietary connection. The finite-element mesh actually serves a dual purpose. Stress analysis of the connection can be used in conjunction with tubing/riser dynamic analysis and fatigue testing to examine the long-term structural integrity of the connection and weld. Alternatively, the same mesh can be used to investigate heat flow at the connection. The results in Fig. 2 are for 5 1/2-in. (outer tube) VIT with an internal fluid at 212°F and a film coefficient of 12.23 Btu/hr-in. 2-°F and external fluid at 55 degree F and a film coefficient of 6.08 Btu/hr-in.2-°F. This simulates the experimental facility used to evaluate the Marlin VIT and coupling insulators, 2 with steam flowing internal to the joint and glycol flowing external to same. Once assembled, the finite-element model can be used to compare a variety of coupling insulation options. In the model, conduction is assumed to occur across the connection wherever solid components make contact. Heat flux through the VIT is essentially zero until the coupling is reached. For this example, the heat loss predicted by the finite-element model through a single coupling is 35,000 Btu/hr if the coupling is not insulated. The second curve in Fig. 2 shows the temperature at the outer surface of the VIT. In the vacuum chamber region, the outer surface remains at 55°F. The outer surface of the coupling, however, reaches temperatures as high as 115°F. A more difficult thermal analysis involves modeling the natural convection inherent in the VIT by casing annulus during service. Such work is beyond the scope of discussion of the current paper and involves use of either empirical correlations4 or computational fluid-mechanics software. Wellbore Simulation. Temperature modeling of the wellbore was conducted with a commercial thermal simulator capable of predicting both transient wellbore temperatures and the associated annular pressure buildup between outer casing strings. Initially, the VIT completion was modeled with a composite thermal conductivity intended to address the average behavior of the vacuum tube and connector. This presentation, however, employs a modified version of the software that permits detailing individual tubes and connectors.

A CFD study of steam injection tubing heat losses

“…The VIT technology consists in the use of two concentric tubes that are welded at their ends, creating an annular space containing a vacuum, which minimizes radial heat transfer [21,22,23,24,25,26,27,28]. However, since the junction (connector) between two tubings is not insulated, heat transfer may be significant at this location [29].…”

Mechanical performance experiments on rock and cement, casing residual stress evaluation in the thermal recovery well based on thermal-structure coupling

“…Insulated tubing and heat-resistant packing are installed in the wellbore before steam injection to reduce heat loss and protect the casing and cement sheath. However, when the packing fails, the insulated tubing is damaged or aged prematurely because the casing temperature can reach up to 250-300 C when the steam temperature is 250-350 C. Thus, thermal expansion is restricted, and the compressive stress of the casing exceeds the limit value (Pattillo et al, 2004). When steam injection is stopped, the wellbore temperature decreases, causing the casing to shrink.…”

Heat Transfer Modeling of Wellbore with Vacuum-Insulated-Tubing (VIT)

“…Over the past three decades, VIT has proven to be an effective thermal isolation management tool in increasingly challenging environments, including deepwater challenges, steam challenges, and onshore challenges (Purdy and Cheyne 1991;Azzola et al 2007;Bradford et al 2004;Ellis et al 2004;Ellis et al 2004;Ferreira et al 2012;Li et al 2010;Singh et al 2007;Pattillo et al 2004;Lively 2002;Feeney 1999;Karra et al 2014;Kutzak and Gunn 1989;Nowinka et al 2005;Gonzales and Bajwa 2010).…”