Summary This paper describes the use of two computer simulation utilities to design underbalanced or near-balanced coiled tubing drilling: one is a steady-state design module and the other is a transient wellbore simulator developed for coiled-tubing operations. The steady-state design module provides various design parameters for a drilling operation, and the transient wellbore simulator predicts the outcome from a particular design for the entire operation. Simulation results show the desired under or near balance can only be achieved by certain combinations of liquid and gas rates when using foam as the drilling fluid or using gas injection through gas-lift mandrel or parasite string. The effects of cuttings loading, depth of injection point and reservoir inflow on the downhole pressure or underbalance are presented and discussed. A procedure for designing under- or near-balanced drilling is described and demonstrated with an example using the computer utilities. A field example is used to show the accuracy of the simulator by comparing the measured and simulated downhole pressures. Introduction It has long been recognized that a key to improving the recovery of reserves is to minimize the reservoir damage created while drilling. Thus, the main objective of drilling close to or under balance is to increase the productivity of the well by reducing formation damage due to the invasion of the formation by the drilling fluid and/or fines. Additional advantages of underbalanced drilling are that higher penetration rates are obtained than with overbalanced drilling, the risk of differential sticking is reduced, and hole cleaning is improved. Coiled tubing is particularly suitable for underbalanced or near-balanced drilling because of the improved well control in a coiled-tubing system and because coiled tubing allows continuous drilling while maintaining the underbalance.1–4 When planning an underbalanced drilling operation, the magnitude of the desired underbalance is a primary consideration. If the downhole pressure is too low, there may be wellbore stability problems. The amount of underbalance also determines the rate of fluid production from the reservoir and the surface facility should be able to handle the production fluid in the return flow. After the desired amount of underbalance is determined, the next step is to determine how to achieve this underbalance. For normal or high pressure reservoirs, underbalance can be achieved by using water, brine or diesel as the drilling fluid. For low pressure reservoirs, nitrogen or produced gas is used to aerate the drilling fluid and reduce downhole pressure. Two injection methods are often used. One is pumping foam or nitrified fluid down the coiled tubing and into the annulus, and the other is injecting nitrogen or produced gas through a parasite string or through existing gas lift mandrels to aerate the liquid column in the annulus. The advantages and disadvantages of using foam and gas injection are discussed in Refs. 5 and 6. Aerated or foam drilling fluids are compressible. Downhole pressure or the amount of underbalance while circulating is more difficult to calculate for compressible fluids, and computer simulation tools therefore become necessary when designing under- or near-balanced drilling using compressible fluids.7–9 This paper is concerned with how to achieve the desired under or near balance when using foam or gas injection. Downhole pressure behaviors for various drilling conditions are studied using two computer simulation utilities. A design procedure is presented and demonstrated with a simulation example. Computer Simulation Overview When designing an under or near balanced drilling operation, questions like those listed below often need to be answered:What are the gas and liquid pump rates for achieving the desired under or near balance?What are the gas and liquid pump rates for achieving efficient hole cleaning?How much reservoir fluid will be produced in an underbalanced condition?How does the produced fluid and cuttings loading affect the downhole pressure?What is the maximum rate we can pump without exceeding the pressure limit of the coiled tubing or other well equipment? Two PC based computer utilities have been developed to assist the design of coiled tubing operations. One is a steady-state design aids module (DAM) and the other is a transient wellbore simulator.10,11 They are well suited for answering these questions when designing under and near-balanced coiled-tubing drilling operations. The DAM calculates the coiled-tubing circulation pressure, downhole pressure, foam quality, flow rate and solids concentration for given wellbore and coiled-tubing conditions. It provides liquid and gas rates for achieving the desired downhole pressure based on simplified steady-state conditions. The results are presented in the form of sensitivity plots and show the trend of change of the parameters, which is particularly useful when there is uncertainty in the design conditions. A pump schedule, as well as coiled-tubing size, fluid types, and injection method, can be determined based on the simulation results from the DAM. The operation design can then be tested using the wellbore simulator. The wellbore simulator predicts the transient results of pressure, velocity, cuttings removal, and foam quality in an operation when a designed pump schedule is executed. It includes the effect of the coiled-tubing movement, cuttings transport, inflow or fluid loss due to the reservoir, and change in pumping conditions during the operation. Based on the transient results, all requirements and limits can be verified for the entire operation.

CT drilling has proven to be a commercially viable technique for drilling horizontal drainholes with over 100 directional wells drilled in 1996. The horizontal reach of these wells is much less than with rotary drilling. The various weight transfer devices and methods currently used (or conceived) to enhance the reach with coiled tubing drilling are compared in detail. For example, larger OD coiled tubing (CT) is the most effective way to extend the drilling reach if transport of the reel is not a problem. Tractors would be very effective to extend reach, but reliability and vulnerability to hole conditions may be problematic. A rotator (second downhole motor) to rotate the CT has the potential to drill a 20,000 ft horizontal drainhole. Composite CT is shown to have much less extended reach drilling capability than steel. The results from field testing a solid bottomhole assembly (BHA), bumper sub (thruster), and weight on bit (WOB) equalizer are presented. The field test conditions were specifically selected for the difficulty of weight transfer. The WOB equalizer provided the highest rate of penetration (ROP) by a factor of 2. The equalizer also indicated (via a change in pressure drop) to the CT operator when the downhole WOB exceeded a preset value. Stalling of the downhole motor was experienced only with the solid BHA (without WOB equalizer or bumper sub). The first known analysis of stick-slip motion with CT drilling is presented. A method is illustrated to deduce both static and dynamic coefficients of friction from pickup and slackoff data. These are used in the analytical model to calculate the motion of the bit as it drills off and illustrate both the problems of stick-slip drilling and the theory behind the effectiveness of the WOB equalizer. Introduction Since the introduction of CT drilling in 1991, the business has grown dramatically as shown in Fig. 1. As the drilling of 2,000 ft lateral re-entries becomes routine and 3,300 ft is shown to be achievable, the question often arises "what is the maximum drainhole length achievable." Leising and Newman went into great detail to show what is possible and how this is determined for straightforward re-entries. The purpose of this study is to indicate what can be done to maximize the drainhole length possible today and what may be done to extend the reach of CT drilling in the future. The potential is obvious; the record CT drilled well has a horizontal section of 3,256 ft. The record rotary drilled well (as of March '96) has a horizontal displacement of 26,361 ft with a hole twice the diameter of the record CT drilled well. Many techniques have been studied with rotary drilling to extend the reach; some of these apply to CT drilling and some do not (e.g., eccentric drillpipe). Only those which apply to CT drilling are considered here. The various means of extending reach are discussed with respect to potential problems, cost, risk, and potential benefit. The primary methods as graded by the authors' opinions are summarized in Table 1. These methods are discussed below. Current Technology to Extend the Reach of CT Drilling Larger CT/Smaller Liner. The options of increasing the CT diameter or installing a liner to reduce buckling were previously analyzed. The results are shown in Fig. 2 for a geometry similar to that of Fig. 3 (without tubing in well). Fig. 2 covers the conventional re-entry application where the tubing is pulled from the well. With no tubing in the well, the CT will buckle inside the casing. Note the increase in drainhole length possible by increasing the CT diameter or running inside a smaller casing/liner. In many cases, this is the easiest way to extend the reach. P. 677^

It has long been known that CT rotates during its life because the orientation of the seam weld at the end of the CT varies. This rotation changes the bending plane and thus the fatigue life of the CT. None of the current CT fatigue and deformation (elongation and diametrical growth)prediction models considers this rotation. An experimental rotational-orientation measuring device (ROMD) was developed to measure the rotational orientation of the seam weld as the CT runs in and out of the hole. This device was used to record the rotation of CT strings in various applications. Results from these runs are presented. The affects of rotation on fatigue life are quantified using an existing fatigue model. Some fatigue test results are presented in which the CT was rotated between bends. Both the modeling results and the experimental results show that the fatigue life increases due to rotation. Introduction Most fatigue and deformation modeling and testing that has been done on CT has assumed that the CT remains in one rotational orientation throughout its life. This means that the bending on the reel and around the guide archal ways occurs about the same axis orthogonal to the CT axis. Thus the point of maximum tensile strain is always the same point, typically designated as the "top" of the CT cross-section. Likewise the point of maximum compressive strain is always the same point typically designated as the"bottom" of the CT cross-section. Observation of the ends of used reels of CT has shown that the location of the longitudinal seam weld tends to be random. In one case a 20 ft section of CT was cut from the end of a reel, and it was noted that the longitudinal seam weld was rotated about 30 degrees in this short section. Other non-documented stories from the CT field operations indicated that significant rotation does occur in the CT. As part of a project for GRI, a CT test machine (CTTM), discussed in Ref. 1, was developed that allowed the CT sample to be rotated between bending cycles. Some tests were performed using the CTTM, which showed that rotation significantly impacted the fatigue life.