Module L: More Rock Mechanics Issues in Drilling Argentina SPE 2005 Course on Earth Stresses and Drilling Rock Mechanics Maurice B. Dusseault University of Waterloo and Geomec a.s.

“Predicting” Onset of Instability Now, we have methods of estimating in situ stress conditions Also, we have methods of measuring or estimating strength Furthermore, we have methods of calculating stresses around a circular opening, subject to several assumptions… Putting this together allows prediction of shearing initiation on the borehole wall …An estimate of “breakouts initiation”

Linear Poroelastic Borehole Model… Eqn: Where: pw]cr critical wellbore pressure, shear initiation pi pressure just inside the borehole wall σ1, σ3 largest, smallest ppl σ in borehole plane A = α(1-2)/(1-) ( = Poisson’s ratio) α Biot’s coefficient (1.0 for soft rocks) N friction coefficient = (1 + sin’)/(1 - sin’) UCS,  Unconfined Compressive Strength, friction angle (MC yield criterion) Δp “drawdown” = pi - po

Discussion of Parameters pw pi pw – pi is support pressure Usually, we ignore effects of “α”, except in low porosity, stiff shales (E > 30-40 GPa) UCS and N are equivalent to the c’, ’ of the linear MC yield criterion for shear Poisson’s ratio for shales, 0.25 to 0.35 σ1, σ3 are computed using equations converting 3-D stress to stresses in the plane of the borehole (90° to hole axis) po radius - r

Control Parameters in Drilling Mud weight, mud rheological properties, the geochemistry of the filtrate, cake quality, mud type (WBM, OBM, foam, etc.) LCM content, type and gradation Tripping and connection practices: Surging (run-in), swabbing (pull-out) pressures Drilling parameters: ROP, bit type… Hydraulics and hole cleaning ECD (BHA characteristics, mud properties) Well trajectory, and maybe a few others

Defining Limits in Our Well Plan Pressure or stress Gradient Predicted MW for onset of unmanageable sloughing shmin, danger of LC sv Onset of ballooning in shale zones sv po, onset of blowout if in a sand zone Depth Depth

How are the Limits Defined? Lower MW limit Pressure control Rock Mechanics stability, experience, use of correlations to predict stability line, etc. How much sloughing can we live with? Underbalanced Drilling is a good example of RM Upper MW limit Avoiding massive lost circulation Fracture gradient, earth stresses analysis Effects on ROP The new concept of overbalanced drilling is an example of RM extending this envelope

Are All Limits Absolute? No, and here are examples: Drilling underbalanced? OK as long as it is shales or lower permeability sands, and if the shales are strong (little sloughing) Drilling overbalanced? OK for up to ~1000 psi with properly designed LCM in mud! Drilling below sloughing line? OK if good hole cleaning, use increased MW for trips… Pushing the envelope is typical in offshore drilling, HPHT wells… (e.g. mud cooling…) Vigilance and RM understanding needed…

Example: Drilling Underbalanced It is a Rock Mechanics issue, a pore pressure issue, and a fluids type issue If the shale is strong enough to be self supporting in a bore hole with a negative r If the pore pressure is not so high that it “blows” sand and shale into the borehole If the fluids that enter the hole are “safe”, i.e., not oil and gas in large quantities Excellent for drilling through depleted zones, fast drilling through good shale, entering water sensitive gas-bearing strata, reservoirs that are easy to damage

Underbalanced Stress Conditions s – stress sq High shear stress at the borehole wall shmin = sHMAX sr po pw < po pw radius Some tensile stress exists near the hole wall in underbalanced drilling because po > pw

Mud Rheology High gel strength can cause mud losses on connections, trips Increases surge and swab effects when BHA is in a small dia. Hole Also affects ECD Mud rheology & density can be changed for trips to sustain hole integrity Hydraulics is a vital part of borehole stability! Mud Rheology Diagram Static condition Shearing resistance m – mud viscosity Yield point YP Dynamic conditions Shearing rate

Effect of Mud Weight Increase , shear stress MC failure line yield  Mohr’s circle of stresses no yield c r a n, normal stress Increasing MW (with good cake) reduces the stresses on the wall

Effect of Loss of Good Filter Cake , shear stress MC failure line  failure Mohr’s circle of stresses c r a n, normal stress With loss of mudcake effect, radial support disappears, shear stress increases

Stresses and Drilling sv sHMAX shmin sv sv sHMAX sHMAX shmin shmin sHMAX ~ sv >> shmin To increase hole stability, the best orientation is that which minimizes the principal stress difference normal to the axis 60-80° cone sHMAX shmin Favored hole orientation sv sv Drill within a 60°cone (±30°) from the most favored direction sHMAX sHMAX shmin shmin sHMAX >> sv > shmin sv >> sHMAX > shmin

Uncontrollable Parameters Constrained trajectory (no choice as to the wellbore path) Sequence of rock types (stratigraphy) Rock strength and other natural properties Fractured shales Clay type in shales (swelling, coaly, fissile) Salt, etc. Formation temperatures and pressures, plus other properties such as geochemistry Natural earth stresses and orientations

Can You Live with Breakouts? Yes, in most cases the breakouts are a natural consequence of high stress differences, and can be controlled In exceptional cases, the breakouts are so bad that massive enlargement takes place If hole advance is necessary, there are special things that can be done: Some new products, silicates, polymers that set in the hole and can even be set and then drilled Increase MW, even to the point of overbalance Gilsonite and graded LCM can help somewhat In desperation, set casing!

Some Diagnostic Hole Geometries General sloughing and washout Swelling, squeeze b. sHMAX drill pipe Keyseating e. shmin Breakouts Fissility sloughing Induced by high stress differences c. c. f. Only breakouts are symmetric in one direction with an enlarged major axis

Equivalent Circulating Density Viscous resistance increases the apparent mud weight at the bottom of the hole This is a kinematic (viscosity) effect, and takes place only as the mud is circulating ECD can lead to fracture at the bit though static pressure of mud column is below PF As high as 2.0#/gal recorded in 4¾” hole! Real-time BHP pressure data allow it to be measured and managed (offshore drilling) This leads to early warnings of high ECD This leads to better control and mitigation

Pressure gradient plot ECD Pressure gradient plot 15 16 17 18 19 ppg PF (shmin) MW = 16.7 ppg (static value) reamers and stabilizers mud rings also increase ECD Dynamic pressure (ECD) because of friction, hole restrictions, high mud m A hydraulic fracture is induced at the base of the hole where the ECD exceeds PF (shmin). When the pumps stop, much of the mud comes back into the hole! BHA and collars High ECD! Depth

ECD pBH = mud weight plus friction Dp loss High ECD values (>0.5 ppg) are related to: High mud viscosities and gel strengths (evident on connections and trips as “breathing” of hole) Rapid slim hole drilling leading to large cuttings loads in the drilling fluids near the bit Limited clearance with BHA (MWD system), reamer system, extra large collars… Sloughing of shales leading to partial mud rings or high cavings loads in the mud Reducing ECD is the same as expanding your safe MW window for drilling!

High ECD Effects Gradient plot po 15 16 17 18 19 ppg Gradient plot po PF = shmin Top of restrictive BHA MW = 16.7 ppg (static value) reamers mud rings Dynamic pressure (ECD) because of friction, hole restrictions, high mud m Cannot reduce the MW much because of borehole instability uphole or blowout danger on trips, connections, gas cutting… BHA and collars stabilizers Large mud losses at hole bottom because of fracturing High ECD! Depth

Reducing High ECD Values High ECD: excessive ballooning, high losses, increased risk, reducing the drilling window The high ECD values can be reduced in several ways, here are a few examples: Reduce the mud weight (careful about gas cuts!) Reduce the viscosity and gel strength Avoid sloughing above bit (increases ECD) Circulate out cavings and cuttings as needed Use less restrictive BHA, reduce ROP Use an off-center bit (lower friction losses) Redesign well plan (one less casing, larger hole) OBM probably somewhat better than WBM

North Sea ECD Example Serious ECD problems, but extra depth needed Very long & restrictive BHA was being used Drill (mud motor) to Z with 8.5” hole size Trip out, replace bit with eccentric 9¾” bit Ream to bottom & trip Drill to TD with the 8.5” drill bit size Set 7” casing to TD 10¼” casing High ECD Underream Drill to TD

Some Other Comments on ECD If high drill chip loads from rapid ROP are contributing to ECD, reduce ROP Lower viscosity and gel strength during drilling, but increase it a bit for trips Break the gel strength of the mud during trips by pumping, rotating pipe as you are breaking circulation Be careful in inclined and horizontal holes where pipe is not being rotated much, better to rotate more aggressively Use LCM in mud to plug fractures

ECD Services Example of output from BHI service MWD gauges used Gives ECD, MW, annular pressure, connection effects… This data can be used in a diagnostic manner during drilling to manage ECD and aid well performance This website gives many useful formulae http://www.tsapts.com.au/formulae_sheets.htm

Drilling and Shale Fissility If a hole is within 20° of strong fissility… Sloughing is more likely Shale breaks like small brittle beams Breakouts can develop deep into strata In this GoM case, in the “tangent” section, the hole angle was 61° Vertical offset hole, no problems bedding direction Courtesy Stephen Willson, BP

Coping with Fissile Shale Sloughing If possible, stay at least 30° away from the fissility dip direction (see sketch) Otherwise, keep your mud properties excellent, keep circulation rate & ECD low, gilsonite and fn-gr LCM in mud may help… Keep the drillhole within this cone to avoid severe fissility sloughing problems 100-120° cone Normal to bedding planes

DENSITY NEUTRON IMAGE OF 12500’ MD SHALE BREAK OUT From: Bruce Matsutsuyu SECTION OF SHALE BREAKOUT Note that the majority of the shale sloughing appears to be from the top of the borehole. PHOTOELECTRIC FACTOR CURVES DENSITY CURVES BOTTOM OF BOREHOLE GR Density Neutron Image

Drilling Through Faults The fault plane region is often: Broken, sheared, weak shales and rocks It may have a high permeability It can be charged with somewhat higher po First, expect the faults from your data: Seismic data analysis Near salt diapirs, especially shoulders Accurate mud DV(t) measurements can be of great value to good drilling Cavings monitoring MWD (ECD, resistivity, bit torque…)

Borehole Shear Displacement High angle faults, fractures can slip and cause pipe pinching Near-slip earth stresses condition High MW causes pw charging Reduction in sn leads to slip BHA gets stuck on trip out Can be identified from borehole wall sonic scanner logs (profile logs) Raising MW makes it worse! Lowering MW is better… Also, LCM materials to plug the fault or joint plane are effective sn pw

Slip of a High-Angle Fault Plane borehole sv = s1 casing bending and pinching in completed holes sh = s3 high pressure transmission pipe stuck on trips slip of joint surface slip of joint (after Maury, 1994)

Slip Affected by Hole Orientation! OFFSET ALONG PRE-EXISTING DISCONTINUITIES FILTRATE TYPICAL MUD OVER-PRESSURE Courtesy Geomec a.s.

Diagnostics for Fault Slip Problems In tectonic areas, near salt diapirs… On trips, BHA gets stuck at one point Easy to drop pipe, hard to raise it Borehole scanner shows strange shapes: not the same as keyseating or breakouts drill pipe Start of keyseat Serious keyseat Evidence of fault plane slip

Curing Fault Plane Slip Problems Usually occurs up-hole in normal faulting regimes that are highly faulted, jointed, as MW is increased to control po downhole May occur suddenly near the bit when a fault is encountered Back-ream through the tight zone High pw contributes to the slip of the plane, thus reduce your MW if possible Condition the mud to block or retard the flow of mud pressure into the slip plane: Gilsonite, designed LCM in the mud Use an avoidance trajectory for the well

Mud Volume Measurements Extremely useful, but, accurate DV/Dt needed Case A: fracture/fault encountered, quickly blocked, now analyze data for k and aperture! Case B: fractured rock not healed by LCM Other cases have their own typical response curves (ballooning, slow kick…) Diagnostics! Losses - gpm 20 A 15 10 5 Filtration fluid loss Hole deepening rate Time - min 5 6 7 8 9 B Losses - gpm 20 15 10 5 Filtration fluid loss Hole deepening rate Time - min 5 6 7 8 9

A Precise Mud Volume Installation Outlet mud line Precision flow meter (taken from SPE 38177 - Agip well )

Actual Field Example of Analysis 2890 2910 2930 2950 2970 2990 3010 3030 3050 0.5 1 Hydraulic Aperture (mm) Depth (m) 20 40 60 Average permeability (D/m) Courtesy Geomec a.s This information proved extremely valuable for reservoir engineers in this case, as a gas reservoir was found

Losses Identify Fractured Zones Mud Loss Rate – litres/min Likely, each event involved filling a single fracture Depth - m

Problems in Coal Drilling OBM are worse than WBM in Coal Filtrate penetrates easily (oil wettability) Coal fractures open easily if pw > po Coal is extremely compressible Difficult to build a filter cake on the wall Fissure apertures open with surges Sloughing on trips, connections, large washouts, … Packing off of cuttings and sloughed Coal around the pipe, even during trips

Deep pore pressure penetration because of coal fractures Drilling in Coal stresses around wellbore sq sr Mud rings and pack-off caused by slugs of cavings and cuttings Deep pore pressure penetration because of coal fractures Massive sloughing fracture-dominated coal

Drilling Fractured Coal Safely Keep jetting velocities low while drilling through the coal (avoid washouts) Keep MW modest to avoid fractures opening and coal pressuring, low ECDs while the BHA is opposite the coal seams Drill with graded LCM in the mud to plug the fractures and build a cake zone Avoid swabbing and surging on trips See Appendix to Module H for some results on drilling overbalanced with LCM

A Case History of Salt Diapir Drilling in the North Sea

North Sea Case, Shallow Depth Well A 1a Shallow Gas 2000 m Gas Pull Down Courtesy Geomec a.s.

Above a Deep Diapir, North Sea Normal faulting observed well above the top of the diapir, these will likely be zones of substantial mud losses (low shmin) Beds are distorted, likely shearing has occurred along the bedding planes (weaker) Seismic data show strong “gas pull-down effect”, lower seismic velocities because of free gas in the overlying shales and high po Free gas zones are noted in the strata, and these will increase gas cuts (Gas “pull-down” refers to the effect of free gas on seismic stratigraphy)

Deeper, Around the Diapir This region avoided Well A 1b Gas Pull Down 2000 m Mid-Miocene regional pressure boundary Top Balder Top Chalk Intra Hod/Salt 3000 m Courtesy Geomec a.s.

MWD RESISTIVITY LOG SIGNATURE (OBM) Invaded Zone Time-lapse and different spacing resistivity logging data identified fractured zone clearly Courtesy Geomec a.s.

INVADED ZONE Symmetry(O-B) Courtesy Geomec a.s.

What Was Done to Improve Drlg? A trajectory was chosen to avoid the worst of the crestal faulting and gas pressures Shales also intersected at ~ 90 to fissility Mud losses were carefully monitored with depth in the critical zones, then analyzed Designed LCM in the mud allowed a bit of overbalance in a critical region Of course, gas cuts, shale chip geometry, total cutting volumes, etc., and many other things were monitored in “real-time”

Statfjord Case: North Sea 1 2 3 4 5 6 Mud Pressure minus stress in MegaPascals B-06B B-23AT2 B-39A B-39BT2 Well STATFJORD OVERBALANCED! -800 psi These wells were drilled with overbalance: a MW above the lowest estimated shmin in the zone Courtesy Geomec a.s.

Conclusions Fracturing pressure can be increased by several 100 psi by graded LCM, analysis Young’s modulus (E) is the control parameter Induced fractures or even natural fractures encountered open up almost immediately to their final width: This aperture controls LCM design The plugging happens rapidly with right LCM The effect is enhanced with high viscosity mud and slower ROP Design tools are available for this

A Well Plan, North Sea classical mud weight window is too narrow; cannot avoid instability low mud weight  breakouts high mud weight  destabilized fractured zones & losses breakout problems are controllable by good hole cleaning; fracture zones are uncontrollable Strategy: keep mud weight low manage breakouts with good hole cleaning before increasing mud weight during trips monitor cavings and mud losses for warning of fractured zones Courtesy Stephen Willson, BP

Executing this Difficult Well Background gas controlled by ROP, not MW Monitoring greatly reduced “wiper trips” Continuous ECD and mud volume monitoring to avoid destabilization (+”charged” faults) Chip analysis to identify fractured shales Strength profile modified “on-the-fly” using ISONIC MWD + behavior + prognosis Ballooning analysis refined shmin data Hole condition from CRD scan on trips Weighted pills placed for trips Mud properties well maintained (ECD…)

Trajectory Variations Example Erskine HPHT field Deviated holes need MWD, better control, the dashed line path was abandoned Instead, reach was established above HTHP zone, then the well turned vertical No MWD used, hole cleaning was better, lower ECD, etc… Also, low flow rates, low surge-swab, etc… S-profile trajectory 5000 m Reach section Top of HTHP zone A vertical trajectory in the HTHP zone proved to be cheaper and faster, rather than steering an inclined well trajectory

Real-Time Wellbore Stability For deep, difficult, costly holes only Quality prognosis is needed – po(z), shmin(z) Diagnostic tools used: Real-time pressures (ECD management) Caliper and resistivity data, D-exponent data Borehole imagery (on trips) Accurate mud loss gauges & ballooning analysis Cuttings volumes and visual classification Prevention and and remediation options: Mud properties and special chemicals Hydraulics, drilling parameters, reamers… Special cures… (pills, LCM,,,)

Tests on the Rig Floor on Chips Performed on “intact” cuttings Brinnell hardness is related to strength The dielectric properties can be related to the shale geochemical sensitivity Sonic travel time can be related to strength and stiffness empirically You can use dispersion tests in water of different salinities to assess swelling Even some others can be used These can be taken regularly and plotted as a log versus depth (very useful)

Mud Cooling to Increase Borehole Stability in Shales

Heating and Cooling in the Hole in tanks Heating occurs uphole, cooling downhole. The heating effect can be large, exceptionally 30-35°C in long open-hole sections in areas with high T gradients. Heating is most serious at the last shoe. The shale expands, and this increases sq, often promoting failure and sloughing. At the bit, cooling, shrinkage, both of which enhance stability. Commercial software exists to draw these curves mud up annulus casing heating geothermal temperature shoe open hole +T mud down pipe drill pipe mud temperature BHA -T cooling depth bit

DT Effects in the Borehole Mud goes down the drillpipe fast: ~5 to 10  faster than it returns up the annulus It picks up heat from rising mud in annulus At the bit, still 10°-40°C cooler than rock in HT wells with long open-hole sections Rising uphole, the mud picks up heat from formation, and heats rapidly till the cross-over point (T diff. Is as large as 30°-40°C) Then, it cools all the way to the surface It gets to the tanks hot, and loses some heat, but usually goes back in quite warm

A Simple Quantitative Example… Change in sq at the wall is given by: Dsq]ri ~ (DT·b·E)/(1-n) E = Young’s modulus = 1 to 5106 psi b = Thermal expan. coef. = 10-1510-6/°C n = Poisson’s ratio = 0.30 – 0.35 DT = Temperature change Reasonable values are: E = 3106 psi, b = 12 10-6/°C, n = 0.35, DT = +25°C This increases sq at the wall by ~1400 psi! Not good for shale stability!

Heat Also Reduces Strength a Bit 20 40 60 80 0.5 1 1.5 2 2.5 3 3.5 Strain (%) Deviatoric stress (MPa) Temperature = 20°C Temperature = 60°C 3 = 2.5 MPa Mancos shale About 10% strength loss for this DT, so this is a secondary effect

More Temperature Effects +T reduces strength, increases stress +T also makes adsorbed water more mobile Absorbed water layer thickness is reduced Either water is expelled, or stresses must change because the pore pressure changes In either case, additional V takes place, in addition to thermoelastic effects Furthermore, reaction rates change w. DT Boy! Does this make modeling difficult!

Cooling the Mud Reduces +DT Cooling mud T The mud is cooled at surface through heat exchangers and sea water. As much as -30°C to -40°C is feasible in some cases. Now, the amount of heating at the shoe is very small, only a few degrees. Also, the shale remains stronger by virtue of the cooling. There are other benefits as well… BHA mud up annulus +T -T cooling depth

Benefits of Mud Cooling Increases shale stability throughout hole! Low temperature reduces the rate of negative geochemical reactions between the mud filtrate and the shale Generally, mud properties are far easier to maintain with cooler mud, lower cost Tanks are less hot (in some areas, mud can exit the hole almost boiling!) BHA is “protected” from high T Use it when appropriate!

Lessons Learned Stability in drilling involves many factors Rock mechanics information, cavings and cuttings information, rig site tests… Hydraulics management Lithostratigraphic knowledge MWD in difficult offshore cases (ECD) Temperature management MW and rheology management The key is rock mechanics behavior, as stability is mainly a stress issue But… All factors must be considered together in difficult wells