a graphical hole monitoring technique to improve drilling in high-angle

7/30/2019 A Graphical Hole Monitoring Technique to Improve Drilling in High-Angle

1/17


2/17

2 C. LENAMOND AADE 03-NTCE-24

22 of 15

understanding will improve the ability to makeinterpretations quickly.

Torque and Drag TheoryTorque and drag issues are particularly prominent indeviated and horizontal wells, due to the effects of theweight of the drillstring on the lower side of wellbore.

Excessive drag can prevent directional control withsteerable motor systems, create problems when tripping,cause stuck pipe, drillstring buckling and exceed rigcapabilities. Excessive torque can damage rigequipment, create shocks that damage drillstringcomponents and exceed the capacities of the drillstring.

Torque and drag modeling is performed using finiteelement models which divide the drillstring into individualelements. Sideforces are then calculated for eachelement. The sliding friction force F is calculated asfollows:

F = * N

where N is normal contact force between drillstring andwellbore, and is the coefficient of friction, or frictionfactor used to represent the average conditions in thewellbore. The rotating friction, or torque, is determinedfor each element as follows:

F = * N * r

where r is the radius of the drillstring element. Sum ofthe sliding forces for the entire drillstring is the hookloadweight and the sum of the rotating forces is the torquerequired to turn the drillstring. Since the modeling oftorque and drag requires a sideforce, these models arefor deviated wells only as no sideforce exists in vertical

wells.

Stated simply, drag is the resistance force that must beovercome to move the drillstring up or down in thewellbore. While torque is the moment force required torotate the drillstring. Both are related to the sideforcesand friction generated in the wellbore. Sideforces arethe normal, or perpendicular, forces exerted on eachelement of the drillstring due to:

Buoyant weight of each drillstringcomponent,

Tensile load across element (total is oftenreferred to as hookload),

Buckling and bending loads across eachdrillstring element, and

Drillstring component stiffness across holecurvature and micro-doglegs.

The two major contributors to torque and drag are the

weight of the drillstring and the tensile load across eachelement of the drillstring. From Figure 1, it can bclearly seen that as the inclination of the wellboreincreases, so does the normal force for each drillstringelement. Likewise, in figure 2, the normal force due ttensile load increases with both the weight of thdrillstring below each element and as the wellbor

curvature, or dogleg, increases. It is also obvious thathe wellbore sections in which the inclination is droppingresults in much higher tensile sideforces then in a buildor tangent section as the tensile and weight forcevectors are in the same direction.

Wellbore friction is generally represented as a coefficienof friction, friction factor, and is the force opposingmotion of the drillstring. Wellbore friction is a function othe materials interacting and the lubricity of the mud inthe wellbore. The materials interacting in the wellborare typically metal drillstring components and the metacasing and/or formations.

Drilling ProblemsDuring drilling, the surface measurements of torque andhookloads are invaluable indicators of hole conditionsBy plotting the pick-up, slack-off, rotating weights as welas drilling torque, trends can be identified. The key todetecting these trends quickly is to plot the actual dataversus theoretical values for each hole section. Thitechnique identifies abnormal trend changes fromnormal drilling loads as well depth and trajectorychanges.

Increases in torque and drag can be a warning sign oproblems such as:

the build-up of a cuttings beds on low-side ofthe wellbore

wellbore stability issues such as wellborebreak-outs and hole enlargements

tight hole conditions; ie, reactive shales, keyseats, differential sticking,

tortuosity in wellbore, especially micro-tortuosity and wellbore spiraling, and

rig equipment problems; ie, topdrive bearingfailures, high riser bending, torque gaugecalibrations, etc.

For wellbores in good condition, the primary source odrag is sliding friction due to contact between thedrillstring and the wellbore. Other surfacemeasurements such as flow rate and circulation timescan be added to the plot to aid interpretation.


3/17

Advanced Interpretation of Drilling ProblemsThe combination of advanced MWD/LWDmeasurements with hookload charts and a rigsite drillingmechanics expert will vastly reduce mechanical drillingrisks. MWD/LWD measurements such as annularpressure and gamma ray can be added to the plots to

improve interpretation. While adding such measuresimproves interpretation of drilling problems, thesemeasurements require broader skills by the user.Specially trained drilling mechanics engineers arebecoming increasingly utilized in high-cost and high-riskrigsites. PERFORM, Performance Thorough RiskManagement, engineers are especially trained in drillingmechanics and wellbore stability issues. Theseengineers concentrate solely on drilling mechanics,wellbore stability and pore pressure issues while drillingand risk assessment from offset well data.

Accurate Drilling Data

As with any analysis, junk data in equals junk data out.While the surface measurements of hookloads andtorque are simple measurements, these measurementsare not so easily obtained on the rig. The importance ofaccurate measurements is essential to monitoringtrends. Figure 3 is an example of poor drilling datacollection. The actual hookloads are the solid curvesand the stair-step and blocky patterns are indicatorsof poor data but also of the bigger issue of driller trainingand drilling practices. Even with the theoretical modelingproperly calibrated, there is little value in plotting thepoor quality data on the graph. One of the primaryobstacles for drilling mechanics experts, at the rig, arethe training of drill crews in proper drilling techniquesincluding:

proper speed control when raising andlowering the drillstring,

maintaining consistent drilling practicesbetween drill crews,

understanding swab/surge effects,

tripping practices,

proper hole cleaning practices, and

recognizing the symptoms of downholeshocks and vibrations.

Figure 4 is a plot of surface hookloads obtained properlyusing consistent drilling practices. Note how the actualhookloads overlap the theoretical curves and increasegradually with depth.

Hookload Charts and Friction Factor CalibrationThe rotating off-bottom, pick-up and slack-off hookloadare used to calibrate the torque and drag modelHookload points are taken after drilling out the casingshoe at every connection. These hookload and torquepoints are taken as follows:

accurate block weight must be known,

all measurements taken while circulating atdrilling flow rates,

rotating off-bottom hookload is taken atdrilling rotary speed and close to bottom withminimum rpm of 30,

pick-up and slack-off weights are measuredat constant block speed, which should beslow and continuous,

measurements read from drillers weightindicator to nearest 2 klbs,

torque measured off-bottom at drilling rotaryspeeds while circulating at the drillingflowrate with the traveling block stationary;this removes the variable bit torque frommeasurement.

If the rotating hookloads follow the theoretical rotatinghookloads, then the traveling block, BHA and drillstringweights are correct. The correlation should be exac

with same slope and accurate block weight must beincluded in the theoretical model. Rotating off-bottomnegates friction for the hookload measurement andrepresents the buoyant weight of the drillstringSecondly, in the first few connections out of the casingshoe, the hole is relatively clean of cuttings, in gaugeand is the best time to determine clean hole frictiofactors for the well. Once the rotating weight is correctthe friction factors used in the theoretical model aradjusted until the pick-up and slack-off hookloads matchthe theoretical curves. This friction factor is the cleanhole value with the current mud weight and muproperties. Once the theoretical and actual hookload

are in agreement at the beginning of the hole sectionthe modeled hookloads are only modified if there is amajor change in mud weight or lubricity of the mudsystem in that hole section. This simplifies the modelingprocedures and allows monitoring of trends easily.

The importance of the theoretical hookload values canbe seen in Figures 5 and 6. Figure 5 is a hookloadchart without the theoretical (modeled) curves. This is a62-degree well where the 12 hole section is the


4/17

tangent section. The rig capacity is the limiting factor forhole cleaning while drilling this section. The drillingtechniques are good for the first part of the hole section,but are not adapted as the increasing pump pressure,with the small drillpipe, reduces flow rates. It is difficultto determine if the hole conditions are deteriorating andthe interpretation is difficult. There are hole-cleaning

problems, but what depth did the problems begin?Adding the theoretical hookloads in figure 6, simplifiesthe interpretation and the hole cleaning problems areseen to begin at a depth of 14,700 feet and arebecoming worse. Corrective actions can be takenquickly before the situation results in additional non-productive time.

Figure 6a is a torque chart that is similar to the hookloadcharts. The theoretical torque curve is on the chart andactual off-bottom torque values are plotted forcomparison. Similar to the modeled hookloadcalculations, the drillstring sideforces are used for the

theoretical torque calculations as well. Unlike thehookload calculations, the torque modeling iscomplicated due to the effects of drillstring rotationspeed. As the drillstring rotation speed changes, thesideforces change along the drillstring due to centrifugalforces that are not possible to model accurately.Theoretical torque values are best used to detectdownhole vibrations that waste drilling energy and showup on the rig floor as increased torque and/or erratictorque values. Off-bottom torque values are thepreferred values to use as the bit torque is removed fromthe simulations. The bit torque is difficult to model as thebit torque varies with formation strength, weight on bit,bit speed and bit design.

Poor Hole Cleaning ExampleAn extended reach well, with hole cleaning problems, isshown in Figure 7 using oil-based mud. LWD gammaray curves were added to the hookload chart to assistthe interpretation of hole cleaning and wellbore stabilityproblems. The increasing pickup hookloads areindicating problems in the well below 14,500 ft. Thegamma ray curve is used by the PERForm engineers atthe rigsite to monitor sand formations drilled and theiraffect on hole cleaning. The MWD annular pressurewhile drilling (APWD) sensor also indicated holecleaning problems but not until below 15,000 ft.

Because the equivalent circulating density (ECD)calculation is a function of true vertical depth (TVD), inhigh angle wells, the ECD measurement does notrespond as quickly to cuttings buildup on the low-side ofthe wellbore as does the pick-up hookloadmeasurements. This phenomenon has been seen inmany high-angle wells and is one of the primary reasonsfor the increasing use of hookload charts to improve thedetection of hole cleaning problems in high angle wells.In this well, hole cleaning could not be improved while

continuing to drill. Controlling the penetration rates aftethe problems occurred did not improve pick-uphookloads or ECD issues. Corrective action requiredstopping drilling and circulating while rotating thedrillstring until the cuttings beds were removed. Oncethe cuttings beds were removed, drilling continued withcontrolled penetration rates and optimizing circulating

times between connections to prevent further problems.

Another hole cleaning problem is illustrated in Figure 8Hole cleaning problems begin below 11,000 feet in a 67degree well. Problems increase rapidly to 12,500 feewhere a trip is made to clean the hole mechanicallyCirculation times between connections are added to theplot to optimize circulation times between connectionsMultiple friction factor hookload curves were added tothe plot as relative measure of deteriorating holconditions. Hole cleaning was related to penetratiorates in this well as flow rate was limited due to surfacpump pressure limitations. Penetration rates exceeding

100 fph resulted in cuttings beds on the low-side of thehole. Back-reaming at connections did not improve holecleaning or removal of cuttings beds. The friction factorfor the hole had increased from the 0.17 value for aclean hole to more then 0.22 or a 30% deterioration ohole quality. Recommendations to discontinue drillinand circulate the hole clean with high drillstring rotationwere not accepted by the client. As a result, when totadepth was reached, the rig spent the next 2 daystrying to pull out of the hole and became stuck andpacked-off in the hole several times.

The effects of drilling different formations on hole

cleaning is illustrated in Figure 9. This is a deep well to22,000 feet using water based mud and a rotarsteerable tool to improve hole cleaning by allowingcontinuous drillstring rotation while drilling. Circulationtimes between connections are plotted on the right sideof the hookload chart and multiple theoretical hookloadsare plotted for two different friction factors. In thprevious hole section, the clean hole friction factor was0.20. After drilling out the cement and shoe track of the9 5/8 casing, the pickup hookloads are indicating poohole cleaning as the cement cuttings were nosufficiently circulated out of the 73-degree well due topoor mud quality. At 17,800 feet, a trip was made due ta downhole tool failure, the mud was conditioned andcuttings bed was removed.

As drilling resumes following the trip, hole cleaning icontinuously improved while drilling shale formationsand the circulation times are reduced to find theoptimized hole cleaning time. This trend is reversed afine sand formations are drilled below 19,000 feet asseen on the gamma ray curves and pick-up hookloadsTwo stands were racked back at 19,200 feet and thehole circulated to assist removal of the cuttings beds on


5/17

the low-side of the hole. As more sand is drilled, holecleaning is not improved and a short trip is made at20,500 feet to mechanically clean the low side of thehole and check hole conditions. The pink curve is theTOOH hookloads during the short trip overlaid on thehookload chart. Increasing tripping hookloads indicatedwellbore stability problems causing additional hole

cleaning problems. Hole cavings, as shown in Figure10, were seen on the shale shakers by the PERFormengineers that confirmed wellbore breakouts and thatadditional mud weight was needed to control wellborestability.

Tripping Hookload ExampleFigure 11 is series of two BHA trips in a 17 section ofa directional well. Theoretical tripping hookloads areplotted as dashed lines for friction factors of 0.20 and0.30. For water-based muds, tripping friction factorsbelow 0.30 are considered a relatively clean hole in goodcondition. The first TOOH occurred at 4600 ft and

experienced high drag between 3900-4200 feet. TheTIH experienced similar drag going back in the hole. At4050 ft, the hole was conditioned and large amounts ofcuttings were seen at surface. The drag was reduced onthe next stand in the hole, but returned on the followingstands until below 4200 ft. These high drag sectionscorrelated with a sand section with laminated shalestringers which washed-out leaving ledges in the hole.The last trip out of hole, at casing point deoth, had muchimproved drag over previous trips as seen on the purplehookloads curve.

Casing Running Hookload ExamplesHookload modeling charts can be produced to monitorcasing running operations as well as drilling operations.Figure 12 is a hookload chart for the running of a 13 3/8casing string in a 38-degree well. The actual hookloadstrack a constant friction factor profile indicating goodhole conditions through the buildup curve to total depth.Figure 13 is the hookload chart for running the 9 5/8casing string. Multiple theoretical hookloadscorresponding to various friction factors are shown bythe colored bands. Increasing drag, and increasingfriction factors, are to the left-side of the chart. As the 95/8 casing enters open hole at 7600 feet, the hookloadsare indicating negative friction factor values, which isdue to casing beginning to float until fill-up is achieved.

Below 9000 feet, the casing begins hanging on ledgesas seen on the gamma ray curve. Measured frictionfactors increase three hundred percent to 0.65. Oncecirculation is established below 11,000 feet to clean upthe low side of the hole, the drag is reduced and thehookloads are reduced less then a 0.20 friction factor.Washing of the casing continued, but below 13,000 feetthe hookloads remained constant. As the theoreticalcurves indicate, the casing weight should be increasingas depth increases. In effect, drag is increasing even

though the casing hookloads are relatively constanvalues. The casing had already exceeded the point ono return. This is the depth in which, once exceededthe casing could not be pulled out of the hole due toeither rig or casing strength limitations and the rig wacommitted to continuing to wash the casing in the holeThe casing string became stuck off-bottom at 15,100

feet and was cemented in place. The use of hookloacharts for running casing strings can be used todetermine the depth of no return and to plan contingencyoperations before reaching that depth.

Evaluating Mud Additives Example

In figure 14, the torque chart illustrates the effects omud lubricants on surface torque and downhole drillingmechanics. High stick-slip can be seen in the build-uand beginning of the tangent sections to 8000 ft. Mulubricants were added to the water-based mud at 8,000feet to reduce erratic torque values. Immediate resultare seen on both torque and stick-slip values. While thstick-slip vibrations remain greatly reduced over the 12 hole section, drilling torque values increase within thenext 1000 ft. Evaluation of the mud lubricants basedsolely on surface torque readings would suggest verylimited success, but based on the reduction in severestick-slip conditions, the mud lubricants preventedrillstring failures. To improve the effects on surfacetorque, mud lubrication concentrations should bincreased and further improvements in hole cleaningwere recommended. Erratic surface torque values aractually indicating that more then lubricity improvementsare required to reduce the surface torque. Hole cleanin

issues in the 56-degree tangent section are contributingto the torque problems. Typically, hole inclinations from40-60 degrees are the most difficult to clean due tocuttings beds sliding downhole when flow rates arreduced. Annular pressure sensors can be used toimprove interpretation and evaluation of hole cleanintechniques.

Conclusions1. A graphical presentation of simple drilling hook loads

and torques versus theoretical calculations vastlyimproves real-time adjustments to drilling practicesas hole conditions change and provides earlywarning of poor hole cleaning and wellbore stressfailures in inclined wellbores.

2. Integration of annular pressure-while-drilling (APWDand logging-while-drilling measurements improvethe interpretation of root causes of hole problemhigh-angle wells.


6/17

3. The graphical presentation of hookload and torquedata improves communication of hole issues in aneffective manner to both rig and office personnel inan easily understood format.

4. The graphical technique, used while drilling, allowsthe accurate determination of friction factors that can

be used to ascertain casing running loads while alsoproviding a quick method to monitor these actualloads and points of no return.

5. The graphical output provides a simple visualmethod of checking the accuracy of the torque anddrag model and a qualitative check of the drillersability to monitor hole conditions using rigfloorgauges.

6. The graphical torque charts can be used to evaluatethe effects of lubricants to the mud system.

NomenclatureAPWD = annular pressure while drilling measurementBHA = bottom-hole assemblyECD = equivalent circulation densityEMW = equivalent mud weightFPH = feet per hourLWD = logging while drilling toolsMWD = measurements while drilling toolsN = normal force (sideforce)ROP = drilling rate of penetrationrpm = revolutions per minuteTOOH = trip out of holeWOB = weight on bit

AcknowledgmentsThe author would like to thank the many field engineersand Drilling Engineers that provided data to make thispaper possible.

AuthorChris Lenamond is currently the Drilling EngineeringManager for Latin America South for SchlumbergerDrilling and Measurements (formally Anadrill). He hasbeen in the oil field for 18 years and holds a BS degreein Petroleum Engineering from Texas A&M University.

Chris began his career with a drilling contractor beforemoving to Schlumberger to learn MWD/LWD operationsand a directional drilling career before entering thedrilling engineering centers.


7/17

Figure 1- Sideforce for the weight component of a drillstring element

incl

weight

Figure 2- Sideforce for the tensile component of a drillstring element

tensile

load

weight

tensile

tensileresultant

Dropping Section

weight

resultant

Building Section

tensile

load


8/17

Me

asure

dDep

th(ft)

Measure

dDep

th(ft)

135/8"

Cas

ing

String

12

.25

133/8"

Cas

ing

String

121

/4OH

Figure 3- Hookloads Chart - poor surface data measurements example

Drilling Loads FF Calibration

4,000

5,000

6,000

7,000

8,000

9,000

10,000

11,000

12,000

13,000Slack-Off Wt. Rotating Wt. Pick/Up Wt.

100 125 150 175 200 225 250 275 300 325Hookloads (klbs)

Figure 4- Hookloads Chart - good surface data measurements example

Drilling Loads FF Calibration

8,000

8,500

9,000

9,500

10,000

10,500

11,000

11,500

12,000

12,500

Slack-Off Wt. Rotating Wt. Pick/Up Wt.

175 200 225 250 275 300 325Hookloads (klbs)


9/17

Measure

d

Dep

th

(ft)

Measure

d

Dep

th

(ft)

133/8"

Casing

String

133

/8"

Casing

String

121

/4OH

121/4OH

Figure 5- Hookloads Chart - hole cleaning example without theoretical hookloads

D r il li ng Lo ad s F F C a l i b r atio n

8,000

9,000

10,000

11,000

12,000

13,000

14,000

15,000

16,000

17,000

18,000

Rotating Wt.

Slack Off

Weight Pick/Up

w eight


175 200 225 250 275 300 325 350 375 400 425 450Hook loads (klbs

)

Figure 6- Hookloads Chart - hole cleaning example with theoretical hookloads

D rilling Lo a ds F F C a libra t io

n

8,000

9,000

10,000

11,000

12,000

13,000

14,000

15,000

16,000

17,000

18,000

Rotating Off-Btm Theoretical

Hkld FF=0.0

Rotating Wt.

Trip-In Theoretical Slack-off

FF=0.15

Slack OffWeight

Trip-Out Theoretical Pick-up

FF=0.15

Pick/Up weight

Hole Cleaning

Problems


175 200 225 250 275 300 325 350 375 400 425 450Hook loads (klbs


10/17

)


11/17

Figure 6a- Torque Chart off-bottom and theoretical torque curves

Actual off-bottomtorque

Theoretical off-

bottom torque

12 Tangent Section@ 72 degrees


12/17

//

Measure

dDep

th(ft)

121/4OH

133/

Figure 7- ERD poor hole cleaning example in 12 hole size

12 Tangent Section

6,000

7,000

8,000

9,000

10,000

11,000

12,000

13,000

LWD Gamma Ray Curve

14,000

15,000

16,000

17,000

18,000

19,000

20,000

21,000

Pick/UpWt.

Slack-Off Wt.RotatingWt.

Gamma

Ray

Pick-up hookloads indicatingpoor hole cleaning in tangent

section

175 200 225 250 275 300 325 350 375 400 425 450 475 500 525

Hookloads(klbs)

Figure 8- Poor hole cleaning example in 12 hole size


13/17


14/17

Figure 9- ERD poor hole cleaning example in 8 hole size

Circulation time

during connections

Gamma ray curvesindicating increasingamount of sand formations

being drilled Increasing pick-uphookloads as more

sand formations drille

Figure 10- Hole cavings from high angle well


15/17

Measured

Depth

(ft)

20"

Casing

String

17.5

Figure 11- Tripping Hookloads Chart

Tripping Hookloads

3,000

3,200

3,400

3,600

3,800

4,000

4,200

4,400

4,600

4,800

5,000

5,200

Excess drag 80 Klbsat 3986 ft MD

Slack-off

Tripping

Circulate BU

Wash

down

Pick up

Tripping Out

Max. overpull 65 Klbs

at 4030 ft MD

5,400

120 140 160 180 200 220 240 260 280Hookload (klb s)

Figure 12- Casing Running Chart good hole conditions

Actual casing

running hookloads


16/17

Figure 13- Casing Running Chart poor hole conditions

Gamma ray

Increasing drag running 9 5/8

casing due to hanging in ledges

in wellbore

Drag improves once circulationis established to clean hole

Progressively increasing drag

Hookload remaining constant while

running in hole, indicating increase

drag. Casing becomes stuck off-

bottom at 15,100 feet.


17/17

16"Section

12

1/4"Section

8

1/2"

Section

M

easuredDepth(ft)

Figure 14- Torque Chart mud lubricants evaluation

Torqu eAna ly s i s

0

100 0

200 0

300 0

400 0

500 0

Torque( 1 0 00 l b s )

0 5 10 1 5 20 2 5 3 0 35

High Stick-slip due to

wellbore friction

700 0

800 0

900 0

1000 0

Addition of mud lubricants

1100 0

1200 0

1300 0

1400 0

1500 0

1600 0

1700 0

1800 0

1900 0

2000 0

MWD

DTOR

MWD Stick-slip/10 MWD RPM/10 Surface

Torque

Inclination/2

Theoretical

Torque

a graphical hole monitoring technique to improve drilling in high-angle

Documents