SPE/IADC 67818
`
`Tortuosity versus Micro-Tortuosity - Why Little Things Mean a Lot
`Tom M. Gaynor, David C-K Chen, Darren Stuart, and Blaine Comeaux, Sperry-Sun Drilling Services, a Halliburton
`Company
`
`Copyright 2001, SPE/IADC Drilling Conference
`
`This paper was prepared for presentation at the SPE/IADC Drilling Conference held in
`Amsterdam, The Netherlands, 27 February–1 March 2001.
`
`This paper was selected for presentation by an SPE/IADC Program Committee following
`review of information contained in an abstract submitted by the author(s). Contents of the
`paper, as presented, have not been reviewed by the Society of Petroleum Engineers or the
`International Association of Drilling Contractors and are subject to correction by the author(s).
`The material, as presented, does not necessarily reflect any position of the SPE or IADC,
`their officers, or members. Papers presented at the SPE/IADC meetings are subject to
`publication review by Editorial Committees of the SPE and IADC. Electronic reproduction,
`distribution, or storage of any part of this paper for commercial purposes without the written
`consent of the Society of Petroleum Engineers is prohibited. Permission to reproduce in print
`is restricted to an abstract of not more than 300 words; illustrations may not be copied. The
`abstract must contain conspicuous acknowledgment of where and by whom the paper was
`presented. Write Librarian, SPE, P.O. Box 833836, Richardson, TX 75083-3836, U.S.A., fax
`01-972-952-9435.
`
`Abstract
`Tortuosity is commonly defined as the amount by which the
`actual well bore deviates from the planned trajectory.
`Elimination of excessive tortuosity has been regarded as a
`critical success factor in extended reach drilling operations.
`In this paper the authors will refer to “micro-tortuosity”, not
`measurable by survey data, in which the hole axis is a helix
`instead of a straight line. It is argued that this is Lubinski’s
`“crooked hole” described in the early 1950’s. The paper
`presents a study of micro-tortuosity using field data from
`hundreds of wells. The paper details how and why micro-
`tortuosity occurs and the negative impact micro-tortuosity
`can have on the entire drilling operation. The paper also
`presents a solution that eliminates or drastically reduces
`micro-tortuosity.
`Field results will be presented to demonstrate that micro-
`tortuosity is in fact the dominant component of the total
`tortuosity.
`
`Introduction
`Tortuosity has been recognized recently as one of the critical
`factors in extended-reach well operations1,2,3. The effects
`include high torque and drag, poor hole cleaning, drillstring
`buckling and
`loss of available drilled depth, etc.
`Conventional wisdom has always held that tortuosity is most
`often generated by steerable motors while attempting to
`correct the actual well trajectory back to the planned
`trajectory. However, in the early days of drilling in the mid-
`continent area of the United States, drillers observed a
`problem with running tubulars into wells. A vertical well
`drilled with a 12-1/4” bit would not drift 12-1/4”. This led
`
`Lubinski et al.4,5 to develop a formula for determining the
`minimum drift size for a hole drilled with a given collar and
`bit combination (or the reverse). This became known as the
`“crooked hole country” formula. Thus there was early
`recognition of the potential for problems due to the fact that
`the wellbore was not straight. This recognition predated the
`first use of steerable motors by some 30 years.
`Today, several types of drilling tools are targeted at
`achieving reduced hole tortuosity as measured by survey
`data, with a view to reducing torque and drag. Obvious
`examples are adjustable gauge stabilizers and adjustable
`gauge motors, and, more recently, rotary steerable systems.
`In parallel, it is commonly suggested that bent-housing
`steerable motors increase tortuosity as measured by survey
`data by mixing high dogleg sliding footage and low dogleg
`rotating footage. In brief, low dogleg equals low torque
`equals “good”, high dogleg equals high torque equals “bad”.
`Recent evidence suggests that any torque and drag benefits
`derived from reducing dogleg as measured by survey data
`(macro-tortuosity) are likely to be completely overwhelmed
`by the torque and drag generated by poor wellbore quality
`(micro-tortuosity).
`
`In the last two years, over 200 wellbore sections have been
`drilled using long gauge bits, primarily in pursuit of drilling
`improvements broadly encompassed by the term “hole
`quality”. Most of these bits have been run on steerable
`motors; some, on rotary steerable systems. Modeling,
`measuring, and comparing torque and drag values for
`sections drilled with long gauge bits and with short gauge
`bits immediately showed two surprising results. First, there is
`no dramatic difference between the resulting torque and drag
`values for steerable motors versus rotary steerables when
`both use similar bits. Secondly, there is a significant
`difference between torque and drag values for long gauge bit
`runs versus short gauge bit runs regardless of the method
`used to drive them. The use of long gauge bits also gives a
`clear improvement in activities that might be expected to
`benefit from improved hole quality or reduced micro-
`tortuosity. These include hole cleaning, logging operations,
`resultant log quality, casing runs, and cementing operations.
`
`Quantifying these differences by “back-calculating” the
`friction factors commonly used in the torque and drag model
`
`1 of 12
`
`Ex. 2067
`IPR2016-01517
`
`

`

`2
`
`T. M. GAYNOR, D. C-K CHEN, D. STUART, AND B. COMEAUX
`
`SPE/IADC 67818
`
`shows a general trend. The friction factors that give accurate
`results for long gauge bits are much lower than the values
`necessary for obtaining accurate results when using short
`gauge bits. Coupled with the observable field results, this
`suggests that attention to hole quality is likely to have a far
`greater effect on well design limits, particularly in extended
`reach drilling, than will minute attention to matching
`directional survey results to the ideal well proposal.
`
`Thus, we believe that micro-tortuosity is far more important
`than the commonly known “tortuosity” in determining the
`resulting torque and drag and overall wellbore quality. In
`addition, we will show that micro-tortuosity is highly
`dependent on the bit and we will discuss field results that
`support the contentions above.
`
`Tortuosity vs. Micro-Tortuosity
`To explain the difference between tortuosity and micro-
`tortuosity, we will first explain how tortuosities are defined
`and measured.
`
`Planned Tortuosity (Tp) is the summation of the total
`curvature (inclination and azimuth change) in the planned
`wellbore divided by the well depth. The result can be
`expressed by either the radius of curvature or, as its
`reciprocal, in degrees per 100 feet so as to be consistent with
`measurements of dogleg severity. For example, in a well that
`builds from vertical to 60 degrees with no change in azimuth,
`the total curvature is equal to 60 degrees. If the total depth of
`the well is 10,000 feet (3,048 meters) the Planned Tortuosity
`is 60/(10,000/100) or 0.6 degree/100 ft.
`
`Tortuosity (T) is computed from the final well survey by
`summing all the increments of curvature along the well and
`dividing by the well depth (total tortuosity), then subtracting
`the planned tortuosity. In conventional wisdom, tortuosity is
`approximately the same as macro-tortuosity created by the
`local dogleg severity associated with the use of steerable
`motors attempting to maintain or correct the actual well
`trajectory on course with
`the well plan. The recent
`development of rotary steerable drilling systems was to
`provide smooth wellbore curvature that potentially could
`minimize all
`the
`tortuosity. Thus conventionally,
`the
`tortuosity (T) of the wellbore is equal to the total tortuosity
`(TT) minus planned tortuosity (Tp) or
`T ≈ Macro-Tortuosity = TT - Tp
`
` (Conventional Wisdom)
`
`In their paper describing wellbore profile optimization,
`Banks, et al.1 stated that wells drilled without regard to
`“smoothness” could have tortuosity values as high as
`0.7°/100 ft while smoother wells could have values
`approaching 0.3°/100 ft. The “smoothness” to which Banks,
`et al. were referring had to do with the “kinks” imposed in
`the process of trying to steer the well back to the desired well
`plan with a steerable assembly.
`
`Micro-Tortuosity (Tm) is defined as the tortuosity that occurs
`on a much smaller scale as compared to macro-tortuosity. We
`will demonstrate that the primary source of micro-tortuosity
`is borehole spiraling, where the hole axis is helix instead of a
`straight line. (Despite this the authors have stuck with the
`commonly used term spiralling.) Micro-tortuosity differs
`from macro-tortuosity in that (i) it occurs on conventional
`assemblies as well as motor assemblies (and rotary
`steerables, for that matter), and (ii) it creates a uniform
`spiraled wellbore that can only be detected by advanced
`wireline survey techniques or MWD caliper tools. Unlike
`more randomly occurring (and easily measured) localized
`washout, a spiraled borehole can last several thousand feet
`and can occur across a range of different formations. The
`effect of washout
`is
`therefore considered minor
`in
`comparison to the impact of thousands of feet of spiraled
`borehole. The authors also suggest that what has historically
`been classified as “rugose”, “corrugated”, or “ledged” hole, is
`more likely spiraled hole.
`
`Spiral hole was first mentioned by MacDonald and Lubinski
`in a paper in 19514. They reported that a spiral hole, though it
`has no objective rate of change in angle, could develop
`serious key seating difficulties, drill pipe wear on
`intermediate casing, etc. Lubinski used the term “tight
`spiral” to emphasize the high torque and drag associated with
`the spiraled wellbore.
`
`We believe that the tortuosity (T) of the wellbore should
`consist of the macro-tortuosity and the micro-tortuosity as
`
`T = Macro-Tortuosity + Micro-Tortuosity
`
`In the past, the micro-tortuosity associated with a spiraled
`hole has been lumped into the crude “friction factor” value in
`torque and drag models. As a result, even with the
`introduction of new rotary steerable drilling systems which
`should have minimized all the local dogleg severity (macro-
`tortuosity), the observed field friction factors are still much
`higher than the coefficient of friction between steel and rock
`measured in a laboratory. This suggests that a significant
`portion of the torque and drag created by micro-tortuosity
`still exists downhole. We believe that micro-tortuosity occurs
`in most of the wellbore in the form of hole spiraling. Only
`by recognizing and removing micro-tortuosity can one drill a
`truly smooth wellbore. Based on the above hypothesis, the
`torque and drag (and the associated friction factor) in a
`wellbore with little to no micro-tortuosity should approach a
`level that is lower than has ever been seen before. We will
`demonstrate that in the following sections.
`
`Mathematical Model of a Spiral Hole
`The geometry of a spiral wellbore as defined in a Cartesian
`coordinate system is:
`
`X= r *cos θ
`
`------- (1)
`
`2 of 12
`
`Ex. 2067
`IPR2016-01517
`
`

`

`SPE/IADC 67818
`
`TORTUOSITY VERSUS MICRO-TORTUOSITY - WHY LITTLE THINGS MEAN A LOT
`
`3
`
`Y = r *sin θ
`and
`Z = P*θ /(2π)
`
`------- (2)
`
`------- (3)
`
`In which r and P are the radius and pitch of the spiral,
`respectively. The wellbore depth S can be calculated as
`
`S = [P2 +4 π2 r2 ]½ * z/P
`
`------- (4)
`
`and the curvature of the spiral hole can be expressed as
`
`K = 4 π2 r / (P2 +4 π2 r2)
`
`------- (5)
`
`Thus, for a typical 5’ pitch and 0.5” radius helix, using Eq.
`(5) the wellbore curvature K is calculated to be 0.0656 1/ft or
`a dogleg severity of 376 deg/100’.
` Because of this high equivalent dogleg value, the drill collars
`and drill pipe cannot possibly conform to the spiral. If the
`BHA is slick, the collars will lie on the crests but tool joints
`will tend to hang up. If the collars are stabilised, either the
`BHA distorts
`to accommodate micro-tortuosity or
`the
`stabilisers attempt to ream the hole straight. Either way
`increased torque and drag is probable and this is not
`accounted for in T & D models
`
`The collars will act to limit the amount of lateral movement
`of the bit off the centerline of the hole. Thus the spiral
`amplitude will be determined by the relative size of the bit
`and collars. This is exactly the function described by Woods
`and Lubinski5 in determining the maximum wellbore “drift”
`of a “crooked hole”. Lubinski calculated the maximum drift
`created in a crooked hole as
`
`Drift = (Bit Diameter + Collar Diameter) /2 ------- (6)
`
`Figure 1 shows the two-dimensional schematic of Eq. (6) and
`the drill collars in a spiral hole.
`
`Figures 2 and 3 show two spiral borehole images taken from
`the wireline CAST (Circumferential Acoustic Scanning Tool)
`tool in a well in South America. The evidence of hole
`spiraling is presented in the strong diagonal response of the
`CAST images running across the compressed and expanded
`2-D images presented in tracks 1 and 2. The reverse 3-D
`image presented in track 3 clearly indicates the wellbore
`spiraling while it was being drilled. Note that the spiral
`seemed to change its direction from time to time and had a
`pitch length was about 2 feet.
`
`Figure 4 shows a spiral hole detected by a differential caliper
`tool on a wireline density measurement at a well in Gulf of
`Mexico. The log indicates that the hole is under gauge
`approximately by 1.5” every 4 feet and rarely over gauge.
`This phenomenon is repeated over thousands of feet on this
`log. This section was drilled by a 9-7/8” bit and 6-3/4”
`
`collars. Using Eq. (6) the drift (new wellbore) is calculated to
`be 8.31”, a 1.56”(16%) reduction in wellbore OD which is
`exactly the same magnitude measured by the wireline tool.
`The reduction in the cross section area (drift vs hole size) is
`calculated to be 22.32 in2 (29%). As a comparison, Figure 5
`shows a perfectly gauge hole drilled with a new steerable
`system (a matched long gauge bit and positive displacement
`mud motor). The entire 12,000 ft interval was drilled in only
`2.7 days with no short trips
`
`Although a spiral hole creates higher torque and drag, the
`extra wellbore length due to spiraling is usually negligible.
`For example, using the same parameters above in Eq. (4), S =
`1.014 z, representing a 1.4% increase of wellbore length
`drilled by the bit. Only for cases of very large radial
`clearances (17-1/2” bits and 9-1/2“ collars) can the additional
`length increase to perhaps 3%, or an extra 30 feet drilled per
`1000 feet of hole.
`
`Solution for Micro-Tortuosity: Long Gauge Bits
`The ability of any bit to move off the centerline of the
`wellbore is determined by the gauge length on the bit, the
`amount of side cutting structure on the bit, and the
`stabilization of the bit and BHA. Other factors may play a
`role in reducing the tendency to move off center, such as
`anti-whirl feature, but these factors are addressing symptoms
`rather than causes.
`
`The concept of preventing side-cutting to improve hole
`quality is not new as machinists have taken advantage of it
`for years. A conventional twist drill for drilling through metal
`is furnished with a cutting structure that cuts only in the
`direction of the tool’s long axis, and the spiral flutes serve
`only to stabilize the cutting structure, and “burnish” the sides
`of the hole. Until the flutes begin to stabilize the cutting
`structure, the drill will tend to precess in the same direction
`as drill rotation. This can readily be observed using a
`domestic electric drill, and explains why the hole being
`drilled is often triangular until the stabilizing flutes begin to
`control this movement. If they did not exist, the drill would
`continue to precess, and the resulting hole would be
`triangular in section, following a helical path.
`
`Since the mechanics that governs machining metal is
`identical to that in rock, there is no reason to expect that
`preventing fixed cutter bits contacting the side of a wellbore
`will have any different result. Any tour of a machine shop
`will immediately reveal that a drill press, designed only for
`cutting tools that do not side cut (twist drills or reamers), is
`relatively slender. Milling machines, designed to cope with
`side-cutting tools, are massive, with stiff, well-supported
`spindles to give them the stiffness to resist side-cutting
`forces. This offers a possible explanation for the observation
`that PDC bit tests carried out in laboratories (on rigs that are
`much more like milling machines than like drilling rigs),
`produce results that have never been seen in the field. A drill
`string can never approach the lateral and torsional stiffness of
`
`3 of 12
`
`Ex. 2067
`IPR2016-01517
`
`

`

`4
`
`T. M. GAYNOR, D. C-K CHEN, D. STUART, AND B. COMEAUX
`
`SPE/IADC 67818
`
`a milling machine. However, a drill press, designed only to
`drill holes – which is what we want to do – has no great
`torsional, and often very little lateral stiffness. Instead, the
`cutting tool provides the solution. The drilling equivalent of a
`twist drill is a long gauge bit.
`
`There is now abundant evidence that hole spiraling exists,
`whether this evidence is anecdotal, visual (from imaging
`tools) or by inductive reasoning from logs with an otherwise
`inexplicable periodic variation or tools with an otherwise
`inexplicable wear pattern. There is also abundant evidence
`that long gauge bits minimize or more often eliminate hole
`spiraling. Since the drilling industry depends on steering, if
`“long gauge bits do not steer” then this piece of information
`is interesting, but has no practical application. Once it is
`discovered that long gauge bits can be made to steer, initially
`on specially designed motors, and subsequently on point-the-
`bit rotary steerable tools, then the information is worth re-
`examining.
`
`This solution can be demonstrated by recourse to Lubinski’s
`crooked hole equation in Eq. (6). It demonstrates that the
`drift of a hole is controlled by the diameter of the drill collar
`directly above the bit. A non-spiraled, high quality hole will
`have a drift diameter equal to its gauge, presumed to be the
`nominal bit diameter. It is easy to demonstrate that if the
`collar directly above the bit is in fact the same diameter as
`the bit, then drift equals hole gauge, and the hole must
`possess no spiraling. Running 12 ¼” collars in 12 ¼” hole
`would pose problems. Running a bit with a 12 ¼” sleeve
`directly above the cutting structure (a long gauge bit) does
`not.
`
`When this thinking has been applied to oil field bit design,
`the results have been surprising. Straighter holes have
`resulted
`in
`friction
`factors
`that defy conventional
`expectations. Bit life has been extended greatly. Circulation
`time as a percent of below-rotary hours has been reduced to
`10-12% on average, demonstrating the efficiency with which
`the cuttings are being circulated out of the well. Short trips
`have been reduced or eliminated. Log quality and ease of
`running logging tools has been improved. Cement job
`success rate has been nearly 100%, with cement bond logs to
`demonstrate the high quality of the job. MWD and LWD
`failures have been reduced due to the drastic reduction in
`downhole vibration. Lost-in-Hole risk has been reduced.
`Entire hole intervals have been drilled in record times
`repeatedly.
`
`It is important to note that the drilling system employed
`required changes to the bit design as well as changes to the
`positive displacement mud motor design.
`
`While these benefits apply to the vast majority of wells being
`drilled today, the reduction in friction delivered by this new
`system is of particular value for pushing the extended reach
`envelope even further than previously thought possible.
`
`Quantifying Tortuosity and Micro-Tortuosity by the
`Friction Factor
`There are several ways to quantify tortuosity, such as using
`the surface torque1 or using the friction factor in the torque
`and drag modeling as proposed in this paper. More than one
`hundred wells have been analyzed where the friction factors
`were back-calculated, that is, the value of friction factor
`necessary to generate model results that matched observed
`field data was calculated. All of the wells were drilled with
`conventional BHA’s, including motor and rotary assemblies.
`
`Table 1 shows the results from the study. As can be seen
`from the table, the friction factors can vary considerably,
`depending on mud type and whether the hole is open or
`cased. The friction factors are normally less in casing than in
`open hole. It has always been assumed that this was due
`primarily to the lower relative coefficient of friction between
`steel on steel (drill pipe on casing) compared to steel on rock
`(open hole). We propose that a larger effect is the
`elimination of micro-tortuosity (spiraling) once the casing
`has been run. Our reasons for believing this will become
`clear shortly.
`
`These friction factors have been used for some time now for
`the purpose of predicting torque and drag on planned wells.
`However, with
`the
`introduction of
`this new positive
`displacement mud motor and long gauge PDC bit drilling
`system6, it became apparent that the generic friction factors
`used for everyday wells were no longer applicable to wells
`drilled using this new system. The pick-up, slack-off and
`torque values predicted using the conventional friction
`factors were considerably higher than those observed in the
`field, indicating that the friction factors were set too high for
`accurate torque and drag prediction for the new drilling
`system.
`
`In order to improve predictions when designing future wells
`that would utilize this new drilling system, we analyzed
`several North Sea runs that had been drilled with the system
`to determine an accurate friction factor. We found that on
`wells drilled with the new system and using pseudo oil based
`mud, the average friction factor value was 0.12. This
`compares to 0.17 for conventional assemblies with the same
`mud type, a dramatic 30% reduction. As can be seen from
`Tables 1 and 2, the open hole friction factor value using this
`new system is actually lower than the calculated casing
`friction factor in conventional assemblies. This was a
`surprising revelation. The authors suggest that the drag
`measurements on which the casing friction factor is based are
`normally recorded soon after the casing is run, perhaps on the
`shoe drill-out run. This initially high drag value can be
`expected to drop with every rotating hour as the inside of the
`casing becomes polished. Hence the friction will reduce with
`time giving a lower friction factor value. We have found
`from further study that the open hole friction factor using this
`new system is almost identical to cased hole friction factor
`after polishing, further proof that the reduction must be down
`
`4 of 12
`
`Ex. 2067
`IPR2016-01517
`
`

`

`SPE/IADC 67818
`
`TORTUOSITY VERSUS MICRO-TORTUOSITY - WHY LITTLE THINGS MEAN A LOT
`
`5
`
`to micro-tortuosity. Figures 6 and 7 show “typical” calculated
`friction factor analyses using this new drilling system at two
`North Sea wells.
`
`In order to simulate tortuosity on well plans, a tortuosity
`scale factor is applied to the back-calculated actual friction
`factors (See Table 2). Normally for conventional assemblies,
`Halliburton has used a tortuosity scale factor value of 1.34.
`Using this new drilling system, it was noted that the
`tortuosity scale factor was reduced to 1.14. This would mean
`that in our example of Pseudo Oil Based Mud the planning
`friction factor when using this new system will only be 0.14
`(0.12 x 1.14) as opposed to the “normal” value of 0.23 (0.17
`x 1.34) for conventional assemblies.
`
`Field Examples of Micro-Tortuosity
`Micro-tortuosity affects almost every aspect of drilling and
`completing a well. Due to space limitations we have focused
`on the most important areas, based on the significance of the
`impact on drilling time and cost.
`
`•••• Bit Life and MWD/LWD Tool Reliability
`Bits that designed to cut away the side of the hole as well as
`the hole in front of it tends to drill a spiral hole. They include
`short gauge length bits or bits with side-cutting structure7.
`These types of bit are also prone to vibrations and “whirl”.
`As most drillers are aware that impact damage is a primary
`cause of PDC bit damage. Thus, spiral hole is often
`associated with bit vibrations resulting in shorter bit life.
`
`The same vibration that destroys the bit also travels up the
`drill string and can lead to a premature MWD/LWD failure.
`By stopping
`the vibration before
`it can
`initiate,
`the
`MWD/LWD system reliability should improve. Data from
`multiple incidents where vibration-related failures have
`occurred, utilizing the new drilling system has had a dramatic
`impact on eliminating or reducing the frequency of tool
`failure.
`
`•••• Hole Cleaning
`Due to the rugosity of the spiral wellbore, cuttings will travel
`a tortuous path and will encounter a trough every 2 to 10 feet
`(dependent on the actual pitch of the spiral). This will lead to
`additional circulating time as well as extra
`time for
`backreaming and short trips in an attempt to dislodge the
`trapped cuttings. When using the new drilling system
`(utilizing the long gauge bit), entire intervals have been
`drilled without short
`trips and with greatly reduced
`circulating hours. In one instance, a 12,000-foot open hole
`interval in the Gulf of Mexico was drilled with no short trips.
`The entire interval was drilled in only 2.7 days.
`
`•••• ROP
`Stabilizers will tend to hang up in a spiral hole, especially in
`a non-rotating (“sliding”) mode. This mechanism explains
`the reduction in sliding rates of penetration (ROP) relative to
`
`rotating ROP that is generally recognized as a universal
`phenomenon. If spiraling could be eliminated, one would
`expect to see a resulting increase in sliding ROP relative to
`rotating ROP. In fact, this is exactly what has been seen
`when using the new drilling system that utilizing the long
`gauge bit. In some areas sliding ROP has been increase to
`within 80% or more of the rotating ROP. Thus, the penalty
`for sliding is reduced. This opens the door for the directional
`driller to spend more time keeping the well closer to the well
`plan while achieving a respectable ROP, thus reducing the
`macro-tortuosity in the well also.
`
`•••• Stabilizer Wear
`Stabilizer hang up in spiral holes would also result in the
`excessive wear on the leading and trailing edge of stabilizers
`that has been observed on numerous wells. This is the area
`that would contact the spiral every pitch, and also would tear
`out the new formation when backreaming is done. The short
`gauge bit that originally allowed the spiraling to occur would
`not perform this function, because at any point in the hole,
`the bit will prefer to follow the relatively gauge hole it
`originally cut, and so will follow the spiral in and out of the
`hole. The job of backreaming is left to the stabilizers. A key
`identifier for spiralled hole is that stabiliser wear advances
`along the hole axis, not perpendicular to it.
`
`•••• Torque and Drag
`The torque and drag in the wellbore often determine the
`success of drilling extended reach or horizontal wells. Torque
`and drag data gathered from the new drilling system show a
`40% reduction in the friction factor value required for the
`modeling. This is a result of eliminating micro-tortuosity.
`
`•••• Logging Tool Response
`Spiraled boreholes have long plagued wireline and LWD
`service companies and have led to totally ambiguous
`responses from resistivity, density, neutron, and other
`logging sensors. This is due to the fact that the logging tool
`will be supported on the low side of the hole by the peaks in
`the spiral. If the hole is in fact spiraled instead of corrugated,
`then the opposing side of the hole will be “in phase” rather
`than “out of phase” as would be expected for a corrugated
`hole. In simple language this means that opposite every peak
`on the low side will be a peak on the high side, not a valley.
`See Figures 1 to 3.
`
`Figure 4 illustrates a spiral hole detected by a differential
`wireline caliper tool. The borehole fluctuates between almost
`perfectly gauge and 1.5” under-gauge. This is due to the fact
`that the caliper arm is regularly moving from a peak to a
`valley on the high side. When it measures the peak, the tool
`has its “back against the wall” at that point, so the distance is
`exactly the bit diameter. At all other times its “back” is
`spanning the valley between two peaks, and it is therefore
`unable to conform to the borehole center, and thus measures
`
`5 of 12
`
`Ex. 2067
`IPR2016-01517
`
`

`

`6
`
`T. M. GAYNOR, D. C-K CHEN, D. STUART, AND B. COMEAUX
`
`SPE/IADC 67818
`
`an undergauge hole. As a comparison, Figure 5 shows a
`perfectly gauge hole drilled with the new steerable system.
`
`•••• Cementing
`The spiral borehole described above is equivalent to a
`continuous thread in the well. As previously described, a
`spiraled wellbore will have a drift diameter substantially less
`than the bit gauge diameter. Running casing into a wellbore
`will create an annular space with a varying annular clearance
`at any given cross section. This might not create too many
`problems as long as it is not continuous. Unfortunately, by
`definition
`the spiraled borehole
`is continuous.
` The
`consequences of this are that there could potentially be an
`area of minimum cement thickness that wraps around the
`casing in a continuous path from shoe to shoe. There are no
`cased hole logging tools in existence today that can measure
`this feature in a cement job, so it has escaped detection so far.
`
`•••• Gravel Packing
`In gravel packing wells, and especially inclined wells, a good
`distribution of gravel over the entire gravel pack interval is
`the design goal. A gravel pack screen that has been run in a
`spiral wellbore faces the same annular clearance issues
`described above. The tortuous path the gravel is required to
`travel may well be the root cause of many of the early sand-
`out problems that have been experienced.
`
`Conclusions
`1. Tortuosity has been
`two
`redefined as having
`components, macro- and micro-tortuosity.
` Macro-
`tortuosity can be detected by examination of survey
`results. Micro-tortuosity is a smaller scale of dogleg that
`will not show up in MWD survey data. It can only be
`definitively detected by advanced wireline survey
`techniques or MWD caliper tools.
`
`2. Micro-tortuosity commonly exists in the form of hole
`spiraling. The pitch of the spiral appears to range
`between 2 and 10 feet. Analysis of hundreds of wells
`indicates that micro-tortuosity exists in many of the
`wells being drilled today
`
`3. Friction factors back-calculated in the torque and drag
`model are used to quantify the micro-tortuosity. The data
`indicate
`that Micro-tortuosity
`is a very
`important
`component in total tortuosity, perhaps even the dominant
`component.
`
`4. Many factors contribute to hole spiraling or micro-
`tortuosity but the most significant issue is the bit design.
`There
`is abundant evidence
`that
`long gauge bits
`eliminate hole spiraling. To exploit the benefits of long
`gauge bits, a new motor system has been designed to be
`able to steer the long gauge bit. The same benefits are
`available from “point-the-bit” rotary steerable tools.
`
`5. Field data using the new drilling system have shown a
`much lower friction factor compared to that from any
`existing drilling system. This suggests that only by
`removing micro-tortuosity, can one drill a truly smooth
`wellbore, regardless of the technology employed to steer
`the well.
`
`6. By eliminating or reducing spiraling, nearly every facet
`of the drilling operation is quantifiably improved. None
`of these improvements can be realized by a likewise
`reduction in steering-related macro-tortuosity
`
`7. Micro-tortuosity should be routinely considered
`in
`torque and drag modeling exercises. Until then the
`industry will make decisions on field development that
`are based on ERD limits susceptible to dramatic
`improvement at little cost.
`
`Acknowledgements
`The authors would like to thank members of senior
`management from Sperry-Sun and Halliburton Energy
`Services for supporting the team during the development of
`the new drilling system and for permission to prepare and
`present this paper.
`
`References
`1. Banks, S. M., Hogg, T. W., and Thorogood, J. L., “Increasing
`Extended-Reach Capabilities Through Wellbore Profile
`Optimization”, IADC/SPE #23850. 1992 IADC/SPE Drilling
`Conference in New Orleans, Louisiana.
`
`2. Payne, M. L., and Abbassian, F. ”Advanced Torque and Drag
`Considerations in Extended-Reach Wells” IADC/SPE #35102,
`1996 IADC/SPE Drilling Conference
`in New Orleans,
`Louisiana.
`
`3. Guild, G. J., Hill, T. H., and Summers, M. A. ”Designing and
`Drilling Extended Reach Wells”, Petroleum Engineer
`Int