Data Engineer Manual

SG

DATA ENGINEERING COURSE NOTES

NAME:

DATE:

AN INTRODUCTORY GUIDE TO DATA ENGINEERING

1 of 53 LOGGING SYSTEMS

Data Engineering

Data Engineering Course

CONTENTS I ntr oduction _______________ ______________________ ______________ ______________ _______________ _______________ _____________4 ______4 F orm ation Pressures Pressures and H ydrostatics___________ ydrostatics__________________ ______________ ______________ ___________5 ____5 Equivalent mud weight___________________ weight_____________________________ _____________________ _____________________ ____________ __ 5 Problems caused by abnormal pressures _____________________ ________________________________ ________________6 _____6 Problems associated with excess overbalance _____________________ _______________________________ ____________ __ 6 Problems associated with deficient overbalance _____________________ _______________________________6 __________6 Problems associated w ith underbalance _____________________ _______________________________ _________________7 _______7

Overbu Overbu rden Gradi ents ______________ _____________________ ______________ _______________ _______________ _____________8 ______8 Calculation of overburden _____________________ ________________________________ _____________________ _________________8 _______8

F orm ation Pressures Pressures ______________ _____________________ ______________ ______________ _______________ ______________11 ______11 Pressure mechanisms _____________________ ________________________________ ______________________ ____________________11 _________11 Mechanisms _____________________ _______________________________ _____________________ _____________________ __________________11 ________11 Tectonic movement ____________________ _______________________________ _____________________ _____________________ _____________ __ 12 Underpressure ____________________ _______________________________ ______________________ _____________________ ________________13 ______13

Dr il li ng Expon ent _______________ ______________________ ______________ ______________ ______________ _______________14 ________14 Calculation of overpressure values from dc _____________________ ________________________________ _____________ __ 16

Pore press pressur e evalu evalu ation wh il e dri ll in g ______________ ______________________ _______________ ____________18 _____18 Depth of Seal _____________________ ________________________________ ______________________ _____________________ ________________18 ______18 Gas levels _____________________ _______________________________ _____________________ ______________________ ____________________21 _________21

Post Post dri ll in g an alysis ____________________ ___________________________ ______________ ______________ _______________25 ________25 Pore pressure from sonic logs ______________________ ________________________________ _____________________ _____________ __ 25 Pore pressure from Density logs _____________________ ________________________________ _____________________ ___________ _ 28

F ractu re Pressure Pressure _______________ ______________________ ______________ ______________ ______________ _______________29 ________29 Mechanisms _____________________ _______________________________ _____________________ _____________________ __________________29 ________29 Leak off tests _____________________ ________________________________ ______________________ _____________________ ________________29 ______29 Poissons Ratio ( µ) _____________________ ________________________________ _____________________ _____________________ _____________ __ 33

Basic Dr il li ng F lu id _______________ ______________________ ______________ ______________ _______________ ______________35 ______35 Functions of drilling fluid________________________ fluid__________________________________ _____________________ _______________35 ____35 Definitions of some drilling fluid terms _____________________ _______________________________ ________________35 ______35 Clay chemistry_______________ chemistry__________________________ _____________________ _____________________ _____________________ ___________ _ 36 Basic types of drilling fluids ____________________ _______________________________ _____________________ ________________37 ______37 2 of 53 LOGGING SYSTEMS

Data Engineering

Drilling f luid additives ____________________ _______________________________ ______________________ ____________________37 _________37

Dr il li ng Hydrau li cs __________ _________________ ______________ _______________ _______________ ______________ ___________39 ____39 Flow regimes _____________________ ________________________________ ______________________ _____________________ ________________40 ______40 Pressure losses and hydraulic horsepower ____________________ _______________________________ _______________41 ____41 Annular pressure losses and ECD ____________________ _______________________________ _____________________ ___________ _ 41 Hole cleaning _____________________ ________________________________ ______________________ _____________________ ________________41 ______41 Swab / Surge _____________________ ________________________________ ______________________ _____________________ ________________41 ______41 Problems associated with swab / surge ______________________ ________________________________ ________________42 ______42

Well Contr ol _______________ ______________________ ______________ ______________ _______________ _______________ ____________45 _____45 Procedures for killing a well _____________________ _______________________________ _____________________ _______________47 ____47

PWD and E CD - a qui ck gui de ___________ __________________ ______________ ______________ ______________ _________49 __49 Swab and Surge _____________________ ________________________________ _____________________ _____________________ _______________51 ____51 Hole cleaning _____________________ ________________________________ ______________________ _____________________ ________________52 ______52


Data Engineering

Data Engineering Course Introduction Data Engineers are responsible for the safe and efficient operation of the surface acquisition system. This includes monitoring rig activity, pore pressure and fracture gradient estimation, hydraulics optimisation, systems maintenance, reporting and supervising logging geologists. Crossing the floor to the “dry” end requires a change of focus. Data engineers must know what is happening on the rig at any time and what will be happening. They have much more contact with rig and operator personnel and should be able to give advice as required. Morning meetings came into vogue in the early 1990s, but they really only served to formalise an existing forum. Many rig meetings are now video confrenced by satellite to the beach. Data engineers should present a professional image and be prepared to be pragmatic. The learning curve for a newly promoted data engineer is very steep. By this stage a senior logger should be able to run the logging unit on their own and have some idea of ADT work. Hopefully this course will at best clarify this work or at worse introduce a different aspect of Logging Systems’ activities.


Data Engineering

Formation Pressures and Hydrostatics In ADT work, engineers use a variety of conversion factors and constants to calculate the various pressures and gradients (and to convert between them) involved in drilling wells. The most common is 0.052. A column of fresh water water 1” square and and 12” tall will will exert a pressure pressure of 0.433 psi/ft: psi/ft: 3

P = 62.4 lb/ft 62.4 lb/ft 2 2 144 (in 144 (in per ft )

= 0.433 psi/ft 0.433 psi/ft

So, a fluid column of 8.33ppg (fresh water) exerts a pressure of 0.433psi/ft ∴a column of any fluid of any density will exert a pressure of

0.433 = 0.0519 0.0519 psi/ft per ppg 8.33

The metric equivalent (for use with specific gravity and metres) is 1.421

E q u i v al al e n t m u d w e i g h t Very few operators use psi/ft to express pressure gradients. It is easier to deal with gradients expressed as equivalent mud weights. Pounds per gallon (ppg), specific gravity (sg), psi per thousand feet (pptf) are the most common. Shell are particularly keen on pptf, which is basically psi/ft multiplied by 1000. Calculating pressures and gradients is a frequent task in ADT work and the ability to work in the various units involved is rapidly acquired. The basic equations are as follows: To convert a mud weight to a pressure in psi : ppg x 0.0519 x TVDft sg x 1.421 x TVDm To convert a pressure to a mud weight (EMW) psi / TVDft / 0.052 psi / TVDm / 1.421 To convert a EMW to a gradient : ppg x .052 sg x 1.421 NOTE : NOTE : when calculating pressures using these equations always always use True Vertical Depth. Using the above, the ADT engineer can calculate the pressure exerted at a specific point in the well. The use of these constants and formula soon become second nature.


Data Engineering

P r o b l em em s c a u s e d b y a b n o r m a l p r e s s u r es es An optimum mud weight is required to drill a hole in the most safe and efficient manner. High mudweights lead to lost circulation, differential sticking, poor quality wireline / MWD logs and formation damage which has an adverse effect on any future production. Low mud weights lead to borehole instability kicks and problems with casing. Knowledge of the pore pressure regime in the well allows this optimum mud weight to be run. Ideally the mud weight should be close to BUT above the formation pressure. The point where pore pressure equals the mud weight is called the balance point. point. When the mud weight is greater than the pore pressure this is an overbalanced overbalanced situation, while when mud weight is below the pore pressure this is underbalanced. underbalanced.

P r o b l e m s a s s o c i a t e d w i t h e x c e s s o v e r b a l an an c e Lost circulation circulation - probably the commonest hole problem encountered. Lost circulation can range from the slow seepage losses which can be treated with a little addition of LCM (Lost Circulation Material) into the mud, to catastrophic losses which can be incurable and may ultimately lead to a kick due to loss of hydrostatic head. Mud is lost into natural fractures and fissures in the formation, but these may be opened or even generated by the mud weight exceeding the fracture pressure of the formation. Once these fractures have been opened it is not always possible to close them by reducing the mud weight. Formation damage damage - reservoir flushing and loss of porosity and permeability are also associated with high overbalance. Borehole erosion and washouts may also occur leading to poor quality logs. Differential sticking sticking - thick filter cake may build up on the borehole increasing the area in contact with the drill string. Low ROP ROP - rock bits work by initiating and propagating fractures in the rock. If the differential between the rock and the mud in the borehole is low, the rock chips fly off into the borehole. If the differential is too high the chips will not fly off and the ROP will be reduced. This is called the chip hold down effect.

Problems associated with deficient overbalance Borehole instability - Swelling clays can be caused by low overbalance, as can caving of formation into the borehole. These will manifest themselves as drag or fill. Stuck pipe - borehole instability can cause the formations to swell or cavings to enter the borehole which can then stick the drillstring mechanically. High gas levels levels - usually associated with lack of overbalance, especially if background and trip gasses are high. Connection gas is also a good indicator of low overbalance. Low overbalance is not necessarily a bad thing, but requires vigilance to prevent the situation moving to the underbalanced underbalanced case.


Data Engineering

Problems associated with underbalance Kicks Kicks - if porosity and permeability are encountered when in an underbalance condition, or underbalance is created by swabbing, a kick may occur. ADT work can allow this overbalance and therefore optimum mud weight to be achieved. Accurate and timely analysis analysis of information can allow allow engineers to have a good handle handle on the formation pressure regime.


Data Engineering

Overburden Gradients Much of the ADT and data engineers work in establishing the pressure regime of a well involves the use of overburden gradients. Overburden plays a pivotal role in the analysis of the pressure regime of a well. Overburden (for our purposes) is the cumulative pressure exerted pressure exerted by the, air, water and rock formations above the point of interest. Without an accurate value for the overburden, calculations calculations relying on overburden (such as pore and fracture pressure) will be correspondingly inaccurate.

AIR GAP WATER

LITHOLOGY

DEPTH

OVERBURDEN GRADIENT

As can be seen in the diagram (where the curve represents a gradient) air has a gradient of 0.028ppg, sea water a gradient of 8.6ppg / 1.04 SG. These gradients are fairly constant for an offshore well. Water depth can have a great effect on the overburden, especially in the more recent deepwater projects west of Shetland. On the other hand, the diagram shows that the gradient of the lithology increases with depth following the Normal Compaction Trend. As compaction of the sediment occurs with depth, dewatering of the formation causes the density of that formation to increase. In deep or mature basins the density of the formations may reach a maximum as the diagenetic process runs it course.

C a l c u l a t io io n o f o v e r b u r d e n As noted above, overburden is a function of formation density. density. To calculate overburden requires an estimation of formation density. There is a variety of methods to acquire density data: Bulk density density - obtained by using drilled cuttings and a mud balance. Shale density can also be used, but involves potentially hazardous chemicals.


Data Engineering

Sonic - using wireline data to calculate formationdensity from the wireline sonic log. The transit time in the rock is related to the density.

Density = 2.75-(2.15*(T-47)/(T+200)) This converts the transit time to a density by using known velocities for the lithology. Density log log - data from the wireline density log. These values, usually in sg, can be input directly without any conversion Of these the use of the density log is easiest, especially if you can get the data in ASCII format. Bulk and shale density can be suspect if hydration of the cuttings is occurring. Bulk and shale density do have the advantage of being used while drilling. This data is acquired as an interval average, for example over 10 metres or 50 feet. If ascii data is available it can be imported to XL and an averaging macro run. If ASCII data is unavailable and all you have is a log, eyeball an average for the interval. Take a piece of paper and copy the track scale. Use this as a scale for your estimation of the interval average. Having acquired your interval averages, input the values into a spread sheet. If using sonic you will also need a column for the relevant lithology transit times. Refer to the mudlog for the lithology over the interval and input the relevant transit time. Next, use the XL spreadsheet to produce an overburden data sheet similar to that shown below

OVERBURDEN DATA TABLE W Well ell 399/1 399/12b-2 2b-2 (Format (Format ion density from petrophysic petrophysical al log) Depth (mMD) 0 26 861 1450 1460

Depth (mTVD) 0 26 861 1450 1460

Interval (mTVD)

Density (SG)

26 835 589 10

0.23 1.04 1.90 1.95

Interval Cum ulative Density verbur verburden den (psiOv (psiOverb erburd urden en (psiGradie (psiGradient nt (psi/m) (psi/m) 15 1234 1590 28

15 1249 2839 2867

0.58 1.45 1.96 1.96

EMW (SG) (SG) 0 0.41 1.02 1.38 1.38

In the above example the units are metric (metres, Specific Gravity or Grammes per cubic centimetre), the density data is taken from the wireline log and is averaged over 10 metres. The first interval (from RKB, 0m) to sea level (26m) is the airgap. This has a density of 0.23 sg, multiply the density by 1.421 and multiply that by the interval producing an interval overburden of 15 psi. The next interval is from sea level (26m, to seabed (861m). This interval comprises 835m of seawater with a density of 1.04 sg with the interval overburden being 1234 psi. These interval overburden values are added together to produce an cumulative overburden with a value of 1249 psi. Since logs were not run until 1450m and there were no surface returns, there is an information gap from 861m to 1450m. An estimation of the formation density is used to calculate the missing interval. In this case an estimated density of 1.9 sg over an interval of 589m gives an interval overburden overburden of 1590 psi. This is added to the cumulative to produce a new cumulative overburden of 2839 psi.


Data Engineering

From 1450 m the wireline density data is available and this is input into column 4 of the table. The interval overburden and cumulative overburden are calculated for each 10m interval for the entire section or well. Up to this point all the overburden data is expressed as a pressure in psi. For our purposes the overburden should be expressed as a Gradient. Column 7 of the table converts cumulative overburden (in psi) to a density gradient in psi per metre (psi/m) by dividing the cumulative overburden by the vertical depth. The last stage, in column 8 is to convert the density gradient to an equivalent mud weight in sg by dividing the gradient in psi/m by 1.421. This is the data used in pressure evaluation work and imported into the INSITE database for use in logs.


Data Engineering

Formation Pressures Work on formation pressures by Terzaghi and Peck established the following relationship:

S = σ + P Where :

S is the overburden pressure P is the pore pressure

σ is the formation matrix stress This relationship is the foundation on which all pressure engineering is built. During normal compaction, water in the pore space will be squeezed out and the rock will be supported by the matrix. If the fluid is unable to escape, the rock will become supported by the fluid in the pore space. The overlying overlying rock will cause this pore fluid to become overpressured. Matrix stress is stress is the difference between the pore pressure and the overburden pressure. Under normal compaction the porosity of the f ormations should decrease and the bulk density should increase with depth. Plotting these on semi-logarithmic graph paper produces a straight line. This line is called the Normal Compaction Trend. Trend. If shale density is plotted any points on the Normal Compaction Trend will be normally pressured.

Pressure mechanis ms Normal formation pressure is expressed as :

P = ρ x g x D Where

ρ = average density of the fluid g = gravitational acceleration D = height of the column

e.g. fresh water would give a pressure gradient of 0.433psi/ft. In most drilling environments the salinity of the pore fluid ensures that normal pressure gradients will rarely be equal to that of fresh water. Normal pressure gradients (for the sake of overpressure detection) vary from area to area, e.g. the North Sea has a normal gradient of 0.452psi/ft whereas the Gulf Coast has a typical normal normal gradient of 0.465psi/ft. Underpressure is any formation pressure below the recognised normal gradient of the area. Overpressure development relies on the inhibition of fluid flow, both laterally and vertically, within the rock column. This seal prevents the dewatering sediments undergo as they are compacted and buried. Generation of hydrocarbons hydrocarbons may also produce overpressure.

Mechanisms Hydrocarbon reservoirs reservoirs - Oil and gas bearing formations may be overpressured through being in communication with the deepest formations of the sequence. Undercompaction Undercompaction - sometimes called sedimentary loading. Low permeability and rapid deposition restricts the escape of pore fluid from argillaceous sediments thus preventing the 11 of 53 LOGGING SYSTEMS

Data Engineering

establishment of hydrostatic equilibrium. These mechanisms usually result in overpressured shales such as those found in the tertiary of the Central North Sea. The rate of overpressure increase within these shale sequences depends on the integrity of the seal. Some sequences contain thin limestones that act as a perfect seal, producing very rapid increases in overpressure. Sequences with imperfect seals are characterised by a gradual increase in overpressure with depth, possibly over hundreds of metres. Another form of undercompaction is tectonic loading. If a thrust fault moves rock over an uncompacted sequence, an effect similar to sedimentary loading my generate overpressure. Aquifers Aquifers - While not strictly geopressure, the overpressure in this case is the result of the hydrostatic hydrostatic effect of the water column.

Aquifer

T ec ec t o n i c m o v e m e n t Faulting Faulting - allows communication between deeper, pressured formations and shallow formations. If these fluids cannot escape to the surface, overpressure will result. Overpressure can also be produced if a permeable zone is faulted against an impermeable zone, thus sealing the permeable zone and preventing de-watering. Associated uplift may also cause pressure

Faulting and or uplift Uplift Uplift - if a formation, which is normally compacted and pressured at depth, is uplifted, the original formation pressure may be maintained. Any uplift of a sequence usually results in some erosion of the overlying formations (overburden). For the same degree of movement,


Data Engineering

uplift from shallower depths produces a greater increase in overpressure than from much deeper.

Uplift Aquathermal pressuring pressuring - a formation that is totally isolated will generate overpressure as it is buried as a result of the geothermal gradient. Charged Sands Sands - a shallow sand sequence can be charged up by gas migrating from a deeper formation. This will normally be encountered in areas where wells drilled previously have suffered subsurface blow out or have been poorly plugged and abandoned.

Charged formation

Clay diagenisis diagenisis - Montmorillonite alters to Illite during diagenisis. The interlayer bound water in Montmorillonite becomes free and is released into the pore space, causing over pressure. Salt domes domes - the plastic movement of salt formations can cause pressure anomalies due to the faulting and folding accompanying diapirism. In addition, formations within the salt such as dolomites may become overpressured due to uplift within the salt, e.g. the Plattendolomit. Salt may also seal clay formations, preventing dewatering.

Underpressure Also referred to as subnormal pressures due to the pressure gradients being less that 0.433 psi/ft. The main form of underpressure encountered in the North Sea area is caused by depletion of reservoirs due to production. The gas reservoirs of the southern North Sea can be depleted to 6ppg EMW. 13 of 53 LOGGING SYSTEMS

Data Engineering

Drilling Exponent While indicators like gas, drag or fill allow the ADT engineer to suggest the presence of overpressure, an operator needs to know how to restore an effective overbalance. This requires quantitative evaluation to produce an estimate of the overpressure as an equivalent mud weight. In short the operator wants to know how much to increase the mud weight. This would normally happen while drilling, so drilling parameters supply the data required. The data is used in a drilling model, which attempts normalise all the parameters to produce a figure independent of parameter variations, but generating a value that represents the formation characteristics. There are a few drilling models, Modified Log Normalised Drill rate being Sperry-Sun developed, Instantaneous Drilling Evaluation Log was developed by Anadril and Sigmalog, developed by AGIP abd Geoservice. MLNDR and Sigmalog suffer from being area specific (MLNDR to the Gulf of Mexico, Sigmalog to the Po valley). The commonest and most understood drilling model is d exponent. Various workers had looked at the problem of using drilling data to generate pore pressure values, but until 1966 the results were unsatisfactory. Jordan and Shirley derived an earlier drill rate equation to produce an equation for drillability, d, or the d exponent. exponent. The d exponent varies inversely with ROP, but will increase with depth in a normally compacted argilaceous formation. The next step was to modify the equation to incorporate mud weight changes, using the normal pore pressure divided by ECD (Effective Circulating Density). This is called the Corrected (or Modified) d exponent, dc. The days of hand calculating d exponents are long gone, so suffice to say that d c is plotted on semilog paper against TVD and allows the ADT engineer to calculate pore pressures based on deviation from the normal compaction trend.

Plot of dC 14 of 53 LOGGING SYSTEMS

Data Engineering

As can be seen in the plot above, plots of d c require careful interpretation. Having to rely on picking up deviations from the trend line is fraught with problems. The initial problem is where to place the trend line. D c is also sensitive to changes in formation, hole size and bit type. Shales and claystones give the most reliable d C, so the trendline is placed so that it intercepts as many “shale points” as possible. Normal compaction should increase to the right, while the deviations that indicate indicate overpressure should be seen as a “cut back” to the left. D c exponent is often criticised for inaccuracy and its reputation is not enhanced by poor interpretation. D c should only be used in conjunction with other indicators.

Dc plot showing the information required for accurate interpretation There are many debates about the movement of trend lines. In the case above had the trend line not been moved to the left for the 12 1/4” hole the calculated pore pressures would have been wildly inaccurate. Also visible in the plot above is the effect of using PDC bits, which also move the dc to the left. As d C is in effect an indicator of drillability, sandy formations may drill faster making d c shift to the left while hard limestones shift it to the right. A fining up sequence will generate a d c trace that appears to cut back, but is in fact due to the formation becoming more arenaceous. The dc exponent is unreliable when coring. Most important feature of the above plot is that the clay section where the PDC bit was run is overpressured. The d c can be seen cutting back to the left before entering the limestone sequence. Bit dulling causes the d c to trend to the right, due to decrease in ROP. This may mask a cut back due to overpressure, but is only a pronblem towards the end of the run. Thin carbonates will effect the trend, but can also seal an overpressured zone. Marls cause the trend to shift to the right, which can give the impression that any overpressure is lower. In summary, the d c exponent plot shows where overpressure probably is, but must be interpreted with care (especially when shifting trend lines) and with reference to other parameters.


Data Engineering

Calculation of overpressure values from d c c The commonest method of calculating pore pressure while drilling is the Eaton Equation, using the corrected d exponent (d c).

p =

S -

(S - pn) x (dco / dcn) [ (S -

Where

1.2

]

p =

Pore Pressure

S =

Overburden

pn =

Normal Pore Pressure

dco =

Observed dc exponent

dcn = Normal dc exponent

dc exponent v depth plot

0 Normal Compaction Depth

Trend

Ft

TVD 5000 Abnormally Pressured Shale Dco = 1.0 Dcn = 1.25

10000 0.5

1.0

1.2

Dc Exponent

Using the data from the d c plot: p =

Calculated Pore Pressure

S =

Overburden

=

19 ppg 16 of 53

LOGGING SYSTEMS

Data Engineering

pn =

Normal Pore Pressure =

8.7 ppg

dcn =

Norm Normal al Dc Dc expo expone nent nt =

1.25

dco = Observ Observed ed dc expone exponent nt

[

=

1.0 1.2

p=

19 -

(19 - 8.7) x (1.0 / 1.25) 1.25)

p=

19 -

p=

19 - 7.879

p =

11.12 ppg 11.12 ppg

(10.3 x 0.765)


]

Data Engineering

Pore pressure evaluation while drilling

There are other m ethods of estimating formation pressure while drilling in addition to d c. They do, however, depend on samples of formation being available. Bulk density density - Cuttings are washed and placed in a Baroid balance until the balance reads 8.33ppg / 1.0sg. Top up with fresh water (stirring to expel air) and replace cap. After wiping off any excess water, reweigh and take the reading from the scale. Use the reading in the relevant formula. 1 sg =

------ ------------------

8.33 ppg =

2 - Final W eight

------ ------------ ---- ---- ----16.66 - Final W eight

Values tend to be lower than shale density due to absorption of drilling fluid and the washing water by hydroturgid shales, but trends can be seen. Useful in argillaceous rocks. Watch for decrease in trend, this is indicative of overpressure. Shale density density - a graduated column filled with variable density fluid and 4 - 5 glass beads of different (known) density is used to test the density of shale fragments. A calibration chart is drawn with density against against the graduations on the column. Shale fragments are dropped in the column and the point where they stop is read off and checked against the calibration chart. Again, the density is plotted against depth and any deviation from the trend may indicate overpressure. Shale factor - also known as the Methylene Blue TesT (MBT) test. Measures the cuttings’ Cation Exchange Capacity. As diagenesis proceeds, the Montmorillonite in the sediments converts to Illite, so the amount of Montmorillonite should decrease with depth. Overpressured zones are more likely to have increased Montmorillonite due to normal dewatering not having taken place and the pore fluid supports more of the overburden. Diagenesis, being pressure related, does not occur to the same extent due to this f luid supported condition, condition, so the ratio of Montmorillonite to Illite in a given formation will increase. Montmorillonite has a higher Cation Exchange Capacity than Illite, so the higher the CEC, the more Montmorillonite. Shale factor should decrease with depth, following following a normal compaction trend. Deviations Deviations from the normal compaction trend should indicate overpressured zones. Calculating pore pressure from the above involves plotting the data against depth, identifying the normal compaction trend and any deviations from that trend. The method used to calculate a formation pressure is called the Depth of Seal method.

D e p t h o f S e al al May be used with formation density, seismic velocity, formation resistivity, interval transit time, Dc Exponent or shale factor. Based on the density of an overpressured shale at the depth of interest being the same as the density at the "depth of seal" on the normal compaction trend. As a result of this this relationship, relationship, the Matrix Stress at these depths depths will will be the same.


Data Engineering

Formula

S = σ + P

where S = Overburden σ = Matrix Stress

P = Pore Pressure

EXAMPLE

This example uses a plot of shale density.

To calculate the Pore Pressure at 5400ft

Depth of seal (ds)

1400ft

Overburden at 1400ft S (ds)

0.68 psi/ft


Data Engineering

Overburden information from the well overburden table or offset data Pore pressure pressure at 1400ft P(ds) = normal 0.45 psi/ft Depth of interest (di) =

5400ft

Overburden at 5400ft S(di)=

0.83 psi/ft

Using the relationship S = σ + P σ = S(ds) - P(ds) σ=

0.68 - 0.45 psi/ft

σ=

0.23 psi/ft

Since all pressure variables should be in units of absolute pressure, they must all be converted to psi. Calculate the matrix stress s at depth of seal (ds) σ psi = σ x (ds)ft σ psi = 0.23psi/ft 0.23psi/ft x 1400ft 1400ft σ psi = 322psi 322psi

Calculate S(di)the overburden at depth of interest, S(di) = S(di) psi/ft x (di ( di)ft )ft = S(di) psi S(di) = 0.83 psi/ft 0.83 psi/ft x 5400ft 5400ft = 4482 psi 4482 psi To calculate the pore pressure at 5400ft, P(di), use: P(di) = S(di) - σ P(di) = 4482 psi 4482 psi - 322 psi 322 psi P(di) = 4160 psi 4160 psi Convert to ppg = 4160 psi 0.052 x 400ft = 14.8ppg

Calculated Pore Pressure at 5400ft = 14.8ppg 20 of 53 LOGGING SYSTEMS

Data Engineering

Other Pore Pressure Indicators

Gas levels, drag, fill, temperature, cavings and torque can indicate the presence of overpressured formations. They can been used qualitatively to lend support to information obtained from quantitative methods such as d c.

Gas levels After dc exponent, gas is probably the best indication that an overpressured formation is being penetrated. Although affected by lithology and ROP, gas remains a good qualitative indicator. indicator.

The diagram above shows the effect of increasing pore pressure on gas levels

B a c k g r o u n d g a s - The amount of gas in the mudstream is dependent on the ROP, amount of gas in the drilled formation and the pressure differential in the borehole. As the differential between pore pressure and mud hydrostatic is reduced by increasing pore pressure, background gas gradually increases. - as the differential approaches balance, additional gas can be Connection gas introduced to the well bore by the action of pipe movement at connections causing swab pressures. Connection gas appears as sharp peaks above the drilling background gas level. The peaks will appear at bottoms up from the connection. Connection gas can be used to quantify the pore pressure by using the swab/surge application to calculate swab pressures. NB connection gas is reported as a peak a b o v e background. background.


Data Engineering

Snapshot showing connection gas - the size and arrival time of trip gas can give an idea of the downhole environment. environment. Trip gas Although also caused by the the swab action of the pipe, trip gas is thought to be caused by poor filter cake formation, due to low overbalance. overbalance.

N o n - d r i l l i n g b a c k g r o u n d g a s - a very useful tool as it indicates the state of the hole when the bit is off bottom and circulating. On high pressure / High temperature wells it is good practice to circulate off bottom to establish a non-drilling background. As this is unaffected by drilling effects, variations in NDB can indicate changes down hole. - as balance is approached the confining pressure on the drilled formation may be Cavings reduced enough to allow fragments of rock to slough off and cave into the borehole. The cavings will be carried to the surface by the mud, but if too large to be lifted, will remain in the hole to cause further problems. When examined, pressure cavings are normally larger than drilled cuttings, splintery and curved. Note some formations may react with mud to produce similar cavings e.g. the Kimmeridge Clay.

D r ag ag o n c o n n e c t i o n s - indicates that the borehole is unstable and the shale behaving plastically. As overpressured (undercompacted) shales can behave in a plastic manner the borehole wall encroaches on the bit and stablisers, showing as drag when the string is picked up. Pressure cavings can also enter the hole and accumulate around the bit. Note - in high angle directional holes drag is normal, so it is unreliable as a pore pressure indicator.

T o r q u e - As with drag, torque is caused by the borehole encroaching on the drilling tools. Again torque is is unreliable unreliable in high angle angle holes. - on going back on bottom after a connection cavings that have accumulated at the Fill bottom of the hole have to be "drilled out". T he caving process may be aggravated by the lack of additional overbalance when the pumps are shut down.


Data Engineering

The snapshot above shows the effect of hole fill after a short trip.

T e m p e r a t u r e - overpressured zones are characterised by having a increased thermal gradient. Flow line temperature and ∆T (temp out - temp in) are normally plotted and should give an indication that an overpressured zone is being approached. Entering an overpressured zone will normally be associated with an increase in flow line temperature.

Plot showing temperature against depth when drilling into an overpressured zone. Note how the temperature gradient decreases in the cap rock. 23 of 53 LOGGING SYSTEMS

Data Engineering

Drilling the cap rock above the zone may show a decrease in flowline temperature due to its higher thermal conductivity, but as the undercompacted zone is entered, flow line temp should increase. Temperature readings can be affected by many surface factors such as mud additions or riser length, so much so that in some cases it is useless.


Data Engineering

Post drilling analysis

Wireline logs allow the ADT to check the pore pressure values obtained while drilling. The main parameters used are Sonic, Density and Resistivity. Direct formation pressure readings can be obtained from RFT (Repeat Formation Tester) tools, which measure the pressure in the formation by clamping a pressure tool to the borehole wall. RFTs require some porosity and permeability to obtain good readings, so are not suitable for measuring presures from claystones.

P o r e p r es es s u r e f r o m s o n i c l o g s Post drilling pore pressure analysis can be done using sonic logs, again using the Eaton Equation. Sonic transit time, ∆T, will decrease decrease with depth as the density density of the formation increases due to reduction in porosity. Sonic is plotted against depth on semilog paper and a normal compaction trend identified. Deviations Deviations from the trend m ay be due to overpressured formation. p = S - (S - pn) pn) x ( ∆Tn / Tn / ∆To) To)3.0 Where p

= Pore Pressure

S

= Overburden

pn

= Normal Pore Pressure

∆Tn

= Normal Sonic ∆T

∆To

= Observed Sonic ∆T

Pressure variables used in these calculations calculations should be expressed as gradients (psi/ft)

0

Normal Compaction

Depth

Trend

Ft ∆ Tn

∆ To Abnormally

5000

Pressured Shale

∆ Tn = 138 ∆ To = 191

10000

80

100

200

Sonic Velocity ( sec/ft) µ

Using the data from the sonic plot: p

= Calculated Pore Pressure 25 of 53

LOGGING SYSTEMS

Data Engineering

S

= Overb erburd urden

= 0.68 psi/ft 0.68 psi/ft

pn


= 0.452 psi/ft 0.452 psi/ft

∆Tn

=Normal Sonic ∆T

= 138 usec/ft 138 usec/ft

∆To

= Observed Sonic ∆T

= 191 usec/ft 191 usec/ft

p = 0.68 - (0.68 (0.68 - 0.452 0.452)) x ( 138 / 191 ) 191 )3.0 p = 0.68 -

(0.228 x 0.377) 0.377)

p = 0.59 psi/ft 0.59 psi/ft p = 0.59 / 0.59 / 0.052 ppg 0.052 ppg p = 11.42 ppg


Data Engineering

Pore pressure from resistivity logs Post drilling pore pressure analysis can be done using resistivity logs, again using the Eaton Equation. If MWD resistiivity tools are being run it is also possible to calculate the pore pressure while drilling, in a similar manner to that employed for d c. Deep resistivity is the most relevant for pore pressure detection as it will ideally be reading uninvaded formation. Resistivity should increase with depth through a normally compacted shale section, due to the decrease in the volume of pore fluid. Deviations from the normal trend may be due to undercompacted sediments, but be aware that resistivity changes with the salinity of the pore fluid. 1.2

p = S - (S - pn) x ( Rn / Ro)

Where p

= Pore Pressure

S

= Overburden

pn


Rn

= Normal Resistivity

Ro

= Observed Resistivity

Pressure variables used in these calculations calculations should be expressed as gradients (psi/ft)


Data Engineering

p

= Calculated Pore Pressure

S

= Overb erburd urden

pn

= Normal Pore Pressure = 0.452 psi/ft

Rn

=Normal Resistivity

= 1.1 Ohm/m 1.1 Ohm/m

Ro

= Observed Resistivity

= 0.6 Ohm/m 0.6 Ohm/m

= 0.68 psi/ft 0.68 psi/ft

1.2

p = 0.68 - (0.68 (0.68 - 0.452) 0.452) x ( 0.6 / 0.6 / 1.1 ) 1.1 )

p = 0.68 -

(0.228 x 0.438) 0.438)

p = 0.59 psi/ft 0.59 psi/ft

p = 0.59 / 0.59 / 0.052 ppg 0.052 ppg

p = 11.15 ppg

Pore pressure from Density logs

Uses the same depth of seal method as described for “while drilling”, but with wireline density data.


Data Engineering

Fracture Pressure

Fracture pressure determines the maximum mud weight which can be run in a particular hole section. While high mud weights are used to control overpressured zones, severe problems can be encountered if the m ud weight, or rather its hydrostatic pressure, exceeds the fracture pressure of the form ation. Accurate knowledge of the formation fracture pressure is necessary to prevent problems such as lost circulation or subsurface blow outs.

Mechanisms Applying Applying pressure to a formation will initiate fractures along the line of least resistance within the rock. To propagate these fractures the pressure must be greater than that of the least principal stress. Fractures will propagate normal to the direction of least principal stress. This direction can be ascertained by looking at the region’s faults. Normal faults indicate the least principal stress is horizontal. Reverse faults indicate least principal stress is vertical. Transcurrent faults, although indicating horizontal, means that the stress is higher than for normal faults but insufficient to cause reverse faulting.

Leak off tests LOTs are performed to establish the formation breakdown gradient after a casing string has been set. After drilling out the casing tools and a few metres of new formation the hole will be circulated clean and the annular preventer closed. Using the cement pump, the pressure is increased by pumping in increments until leak off is attained.

Leak off test graph

To calculate the leak off as an EMW, convert the mud weight to a hydrostatic pressure, add the surface pressure, then convert back to a mud weight. 29 of 53 Data Engineering LOGGING SYSTEMS

Casing is normally set in a competent formation like a claystone and the LOT is usually performed in a similar formation. Unfortunately if the formation changes to a less competent lithology such such as a sandstone, the maximum allowable mudweight mudweight (obtained from the LOT) is no longer valid. In order to maintain borehole integrity the ADT engineer calculates fracture gradient while the hole is being drilled. This requires a model which can be updated with depth. Daines Fracture Gradient The Daines method is the most widely used used technique for the estimation of fracture gradients. Its use is recommended by most operators operators and is specified by BP on all of their their wells. Our systems offshore are configured to produce fracture gradients using this method. Fracture gradients can be estimated by employing the Daines technique which incorporates the use of a tectonic stress which is either known for a region or is calculated from Leak Off Test values. Tectonic stresses are calculated at each Leak Off Test using the Daines relationship: Tectonic Stress ( T ) = F - [ (S-P) x (

µ )+ P ]

-µ Where,

S (sg emw) =

Overburden pressure

P (sg emw) =

Pore pressure

F (sg emw) =

Fracture pressure

µ

=

Poissons Ratio.

Poissons ratio (µ) is a value between 0 and 1 that attempts to reflect the relationship between elasticity and rigidity of different lithologies. There is a predefined list of figures available that have been collated from the work of various authors, notably Wuerker (1961) and Daines (1982). Fracture gradients gradients can then be calculated at regular regular depth intervals, intervals, inputting appropriate values for Poisson's Ratio at each lithology change. As each leak off test /formation integrity test is performed the equation above above can be rearranged to give :Fracture Gradient Gradient ( F ) = T +

[ (S-P) x (

µ )+P]

-µ Daines’ method uses the actual actual formation fracture pressure as a basis for further calculations as the well deepens. This fracture pressure is the Leak Off Test Pressure gradient. By knowing the formation at the test depth (from the log) it is possible to look up the Poisson’s ratio of that formation. Since the overburden gradient and the formation pressure is known, the tectonic stress ratio for the well can be calculated. The tectonic stress ratio is then used to calculate fracture pressures as drilling proceeds, only requiring the Poisson’s ratio to be changed. Good geological control is required for Daines to work as advertised. As it is the rock matrix which is involved in the fracturing, attention should 30 of 53 LOGGING SYSTEMS

Data Engineering

be paid to any matrix or cement in the formation. The Poisssons ratio should be selected on this basis. One feature of the Daines method is that it does require a Leak Off Test. Many operators only conduct formation integrity tests. An FIT is conducted to ensure that the formation will be able to hold a specific pressure without breaking down. As a result this FIT does not represent the true fracture properties of the formation at that depth. This can produce spurious results when the FIT values are applied to fracture gradient prediction. prediction. The diagram below shows the effect of using FIT data compared with LOT data. Engineers should be aware of this.

AIR GAP WATER

30” CSG 20” CSG D E

FIT =

*

LOT = 1.74 SG

1.56

P

L

T

I

H

T H O

FRACTURE GRADIENT


Data Engineering

Calculation of fracture gradients in practice As in the case of overburden, the fracture gradient calculations calculations are best done on a spreadsheet. The first step is to calculate the tectonic stress ratio from the leak off data. This uses the overburden, pore pressure gradient gradient and the Poisson’s Ratio Ratio at the leak off depth. Having calculated the tectonic stress ratio, it is used as the reference cell for the rest of the drilled interval. Set up the next section of the spreadsheet using the same depth interval as the overburden. Copy the overburden gradient data into the fracture sheet as column 2 , and input the pore pressure in the next column. The fourth column calculates the matrix stress which is overburden minus the pore pressure. Column five contains the Poisson’s ratio, and is input manually depending on the lithology. Column 6 contains the µ ratio that is the Poissons ratio divided by 1 minus Poisson’s ratio. The horizontal stress is calculated in column 7 and is the matrix stress multiplied by the µ ratio. Column 8 calculates the tectonic stress, which is the tectonic stress ratio multiplied by the matrix stress. the final column calculates the fracture gradient gradient using the Daines equation. The fracture gradient data can then be used for reports and plots. Daines fracture gradient spreadsheet

FIRST LEAK OFF TEST DATA

LOT DEPTH:

1 6 9 8 mMD

LOT

Overburden

Pore Press.

Matrix Stress

Daines

Ratio

Horiz.

Tectonic

Tectonic/Mat rix

sg

S

P

S-P

µ

µ/ ( 1−µ)

Stress

Stress

Stress Ratio

1.31

1.48

1 .0 4

0 .4 4

0.17

0.20

0 .0 9

0.18

0.41 (Const. for the w ell)

GIV EN:

where

F = P + ( S - P ) ( µ / 1−µ ) + σ

σ = Tectonic Stress

Well 399/1 2b-2 Depth

Ov ve er burde n

Pore Press.

Matrix

m

S (sg)

P (sg)

S-P

1 7 10 1 7 20 1 7 30 1 7 40 1 7 50

1. 3 9 1. 3 9 1. 4 0 1. 4 1 1. 4 1

1.04 1.04 1.04 1.04 1.04

0. 35 0. 35 0. 36 0. 37 0. 37

µ

0 . 17 0 . 17 0 . 17 0 . 17 0 . 17


µ

Horiz.

Tectonic

Fracture

Ratio

Stress

Stress

Pressure

0. 20 0. 20 0. 20 0. 20 0. 20

0. 07 0. 07 0. 07 0. 07 0. 08

0.14 0.14 0.15 0.15 0.15

1.25 1.26 1.26 1.26 1.27

Data Engineering

P o i s s o n s R a t io io (µ )

Clay (v. wet)

0.5

C lay

0.17

Coal

0.19

Conglom erate

0.2

Dolom ite

0.21

Greywacke

Limestone

Sandstones

Shale

course

0.07

fine

0.23

medium

0.24

fine micritic

0.28

medium calcarenitic

0.31

porous

0.2

stylolitic

0.27

fossiliferous

0.09

bedded fossils

0.17

shaley

0.17

coarse

0.05

coarse, w cemented

0.01

fine

0.03

very fine

0.04

medium

0.06

poorly sorted, clayey

0.024

fossiliferous

0.01

calcareous (<50% CaCO3

0.14

dolomitic

0.28

siliceous

0.12


Data Engineering

silty (<70% silt)

0.17

sandy (<70% sand)

0.12

kerogenous

0.25

Siltstone

Slate

0.08

non calc mudstone

T uff Glass

0.13

0.34


Data Engineering

Basic Drilling Fluid What follows is a basic introduction to drilling fluids, taken from the ADT manual. It covers the fundamental uses, properties and types of mud (water based mainly) and is by no means meant to be comprehensive or definitve. For further, and more current information, ask the mud engineer.

Functions o f drilling fluid •

Remove cuttings from the the hole.

•

Suspend cuttings during trips.

•

Allow cuttings to settle settle in the surface system

•

Form wall wall cake on the the formation formation

•

Prevent caving of the the formation

•

Control formation pressures

•

Control corrosion of drilling drilling tools

•

Lubricate and Cool Bit

Definitions of some drilling fluid terms plastic viscosity (PV) - plastic viscosity is a measure of the flow resistance of fluids. PV is the difference between the θ600 and θ300 readings. PV is related to the number and size of the solid particles in the drilling fluid as well as the viscosity of the fluid phase. Yield Point (YP) - a measure of the attractive forces existing between particles and between particles and fluid. YP is the θ300 reading minus the PV. Gel Strength (GELS) - measures the attractive forces in a mud related to time. Normally measured at 10 seconds and 10 minutes, gel strength represents an expression of thixotropy. PV, YP and gel strength are generated by using a rheometer, and the values obtained are used in hydraulics calculations. Apparent Viscosity (VIS) - the time in in seconds for a quart of mud to drain drain from a Marsh Marsh funnel. Vis is a rough guide to mud properties much loved by American toolpushers, but not used in hydraulics calculations. calculations. Its main function is to exend the time the shaker hand spends on the PA. Filtration (fluid Filtration (fluid loss) - is a measure of the amount and rate at which mud loses its liquid to the formation. This is measured using a standard filter press at 100 psi, and is reported in ml/30mins. The higher the fluid loss, the thicker the filter cake and the greater the dehydration of the mud. Filter Cake Cake - this is the thickness of the residue expressed in 32nds or mm. It is invariably equal to 2. 35 of 53 LOGGING SYSTEMS

Data Engineering

Clay chemistry Most water based muds are generally composed of clays. The principal clay is Bentonite, which is Sodium or Calcium Montmorillonite. Montmorillonite. When added to water, or affected by the ions in water, the following reactions reactions will occur: Dispersion Dispersion - the clay platelets separate, the Na+ goes into solution, leaving a net negative charge on the faces of the clay platelets. The resultant repelling repelling forces disperse the platelets. Aggregation - the opposite of dispersion, where the platelets remain or become clumped together. Flocculation Flocculation - edge-to-edge or edge-to-face clumping of platelets when attractive forces predominate. Deflocculation Deflocculation - results when the attractive forces between edge-to-edge or edge-to-face are neutralised. Causes, description and remedies for clay behaviour Dispersion Dispersion - results when dry clay aggregates are added to fresh water. Caustic Soda is usually added to assist in disaggregation. Na+

OHWATER CAUSTIC

Na+ OH-

Na+ OH-

Properties Of Dispersion - attractive forces minimal, PV is high to very high, YP is relatively low and gels are low. Filtration rate is low. Flocculation - in the presence of salt or divalent ions, eg when drilling halite or anhydrite with fresh water mud, flocculation and aggregation of clay platelets results.

Na+

OHNa+ Na+

OH-

Ca++ Na+

NaCl Na+

CaSO4 Ca++

Ca++

OH-

Properties Of Flocculation - attractive forces at a maximum, PV is high,YP - is very high, gels are high, filtration rate very high.


Data Engineering

Basic types of drilling fluids Muds range from spud muds to the latest high technology oil free muds. They differ in composition depending on the conditions they will be used in. The wrong mud in the wrong formation can lead to serious problems. Spud Mud / Native Mud - uses bentonite or native clays from drilled formations, has little or no chemical treatment and has no resistance to contamination. contamination. Organic Thinned Freshwater Mud - contains bentonite for viscosity lignin, tannin or lignosulphanate for gel control, Carboxymethylcellulose (CMC) for filtrate control and NaOH for pH control. These muds have fair resistnce to contamination. Lime Muds - inhibited freshwater mud using Ca(OH) 2 to inhibit shale hydration, organic thinners and NaOH to control pH. These muds have good resistance to contamination. Gyp Muds - inhibited freshwater muds containing Gypsum to inhibit shale hydration. Lignosulphan ate thinners and filtratio n agents are used, with NaOH used to control pH. These muds have good resistance to Anhydrite and salt contamination. KCl Polymer Muds - Use KCl salts to inhibit shale hydration with cellulosic polymers for viscosity. Bentonite and CMC can be used to control fluid loss, while NaOH is used for pH control. These muds have fair resistance to salt contamination. Low Solids Muds - used extended or beneficiated bentonite to reduce the concentration of bentonite. CMC is used for filtration control, with NaOH for pH. These muds have fair resistance to contamination due to the lower reactive solids content. Salt Water Muds - built using sea or other salty water. bentonite for viscosity, viscosity, CMC to control fluid loss, lignosulphanate or lignins for thinning and NaOH for pH control. Saturated Salt Muds - uses NaCl to saturate the mud for drilling Halite sections. ZEOGEL, a betonite for use in salt muds is used as a viscosifier, starches for fluid loss, lignosulphanate for thinning and NaOH for pH control. A common type you may encounter is Thixal. Oil Based Muds - use refined oils for the fluid phase with emulsifiers added to form an emulsion with the water phase. CaCl 2 is added to the water phase,

D r i l l in in g f l u i d a d d i t i v e s Additives are used used to alter the chemical chemical and physical physical characteristics characteristics of the mud producing the desired effect on the drilled formation Weighting Agents - to control formation pressures a weighting agent like barite is used. barite has a specific gravity of 4.3. Other weighting agents include: Calcium Carbonate, Calcium Chloride, Galena and Haematite. Thinning Agents - thinners reduce the attractive forces between clay particles in the mud, resulting in reductions in YP, gel strength and vis. Organic thinners include lignins, tannins and lignosulphanates, lignosulphanates, although the toxicity of some thinners has led to the development of others. Filtration Control Agents - these absorb and and hold water water in the mud and form a filter filter cake, which reduces the rate of water loss. These include CMC, starches and polysacharides. 37 of 53 LOGGING SYSTEMS

Data Engineering

pH control - Caustic Soda (NaOH) is added to mud, keeping the pH above 8 to aid the solubility of thinning agents. Caustic Soda also gives better yield to bentonite. LCM LCM - (Lost Circulation Material). This may consist of varying types and sizes of particulate matter. Often you may see Baracarb, (Calcium Carbonate of known ‘mesh’ size) and “Nut Plug” (literally ground-up walnut shells). There are however, many differing types of LCM, each useful in its own particular situation. Other Additives - Soda Ash to treat Ca++ contamination, Bicarbonate of Soda to treat cement contamination. Defoamers and corrosion inhibitors may also be added


Data Engineering

Drilling Hydraulics Moving mud around the circulating system requires a certain amount of pressure input at the pumps. As the mud moves around the system, friction from the pipe work, drill string, jets and annulus dissipates this pressure. Hydraulics analysis allows the ADT engineer to check where these pressure losses occur and optimise them. By optimising hydraulics, the drilling action can be made more efficient. System pressure loss loss - The input pressure (which is visible as the standpipe pressure) is the sum of the losses in all the sections of the system. The main areas of loss are: surface lines, drill string, bit and the annulus. The pressure loss across the bit is the most important, it being desirable to expend as much energy as possible at the bit, thus increasing the efficiency of the drilling action. Factors affecting hydraulics hydraulics - the size of the system system components and the the properties of the muds have the most effect on hydraulics. Fluid behavior models Hydraulics analysis requires some knowledge of fluid behavior and some form of model is necessary. The simplest is that of a Newtonian Newtonian fluid : W ater

SHEAR STRESS

τ µ

SHEAR RATE

γ

Newtonian fluid behavior Mud, being a suspension of material in water or oil, does not behave as a Newtonian fluid, so another model was developed, the Bingham Plastic. Bingham Plastic assumes a straight line relationship with the slope of the line names Plastic Viscosity, PV, and the intercept with the Y axis, YP, the Yield Point. SHEAR

PV

STRESS

τ

YP

300

600

SHEAR RATE γ Bingham Plastic model


Data Engineering

Fortunately, Fortunately, improvements in rheological test equipment revealed that the relationship relationship was not a straight line, but a curve. By plotting the data on log-log paper the linear relationship is restored and the PV and YP values replaced by n and k. This is the basis of the Power Law model

LOG

n

SHEAR STRESS

τ

k

600

300 LOG SHEAR RATE

γ

n and k are directly related to the mud viscosity characteristics by the following following formulae n = 3.32 log θ600 / θ300

k = θ300 / 511

n

Having established the rheological basics, they can be used to model the flow and pressure environment in the circulating system.

Flow regimes In the circulating system three distinct flow regimes can be identified: laminar, transitional transitional and turbulent. Laminar - as the name suggests involves concentric layers (shear planes) of mud flowing through the pipe work, with little intermixing of the layers. Velocity increases toward the centre of the flow, with the boundary layer being stationary. Transitional Transitional - as its name suggests, this is the transition zone between between laminar and turbulent. NB - Planit and the LS2000 hydraulics applications do not identify transitional flow. Turbulent Turbulent - at high flow rates the shear planes have broken down and the mud tumbles through the pipe work, with velocity constant across the flow area, apart from the thin boundary layer. Laminar flow is desirable in the annulus as turbulent flow causes borehole erosion. Turbulent flow may be desirable in high angle holes, to break up cuttings beds. The regimes are defined by Reynolds Reynolds numbers. W hen calculated for a particular hole section, if the Reynolds number is greater than 2000, then the flow regime is probably turbulent.


Data Engineering

P r e s s u r e lo lo s s e s a n d h y d r a u l i c h o r s e p o w e r Power law enables the pressure losses in the different sections of the system including the all important bit pressure loss, to be calculated. As mentioned above it is desirable for the maximum amount of energy in the system to be expended at the bit. This energy is used to clean the bit and provide jetting of the formation. This is usually expressed as a percentage of the total system energy, or total system pressure loss. Optimising for HHP HHP - When calculating pressure losses to optimise drilling performance it is usual to talk about Hydraulic Horsepower HHP. Hydraulics are said to be optimised when 65% of the available HHP is used up at the bit. In practice this percentage cannot always be maintained due to the presence in the string of tool such as MWD and motors. Optimising for HHP, by using smaller jets and lower flow rates, r educes annular velocities, annular pressure drops and ECD. These help to prevent possible hole damage. Optimising for Impact - when 49% of the available energy is expended at the bit, by way of increasing the jet size and flow rate. This is desirable in situations where hole cleaning is most important, such as top hole.

A n n u l a r p r e s s u r e l o s s e s an an d E C D Effective Circulating Density Density - dependent on the mud properties, flow rate and annular configuration. ECD is the pressure generated by the friction losses in the annulus acting on the mud column, thereby increasing the effective density of the mud.

Hole cleaning Drilling generates cuttings which must be removed from the hole. This is dependent on the lifting capacity of the mud and the annular velocities of the various sections. The lifting capacity of the mud varies depending on the flow rate as most m uds are thixotropic, the faster they are sheared the thinner they become, thus reducing their lifting capacity. Cuttings slip velocity velocity - a particle in a fluid will fall under the force of gravity, but its slip velocity is governed by its density and the rheology of the fluid. As mentioned above high flow rates thin mud, reducing the lifting capacity, so a particle in a high flow rate annulus will have a higher slip velocity than a particle in a lower flow rate annulus. The actual cuttings velocity in the annulus is the fluid velocity minus the slip velocity. Attention should be paid to the largest annulus section, particularly marine risers, as the slip velocity may exceed the annular velocity. If this is the case it may be necessary to use a riser boost pump.

Swab / Surge Moving pipe in the hole causes the drilling fluid to move, generating pressure. Hydraulics analysis allows allows these pressures to be calculated and their effects controlled. These pressures are referred to a swab (a reduction in pressure) and surge (an increase in pressure) Mechanism Mechanism - Whenever pipe is moved in the hole both swab and surge effects occur, due to the flow of mud around the pipe. Pressure is generated in the opposite direction to pipe movement, mainly due to mud being displaced by the pipe or moving into the space formerly occupied by the pipe. Speed of pipe movement has a great effect on swab/surge pressures, with the maximum pressure generated at maximum pipe speed. If the pipe is closed ended 41 of 53 LOGGING SYSTEMS

Data Engineering

(e.g. blocked jets or a float valve) the pressures are greater still due to lack of equalisation equalisation between fluid around the string and fluid in the string. A - pipe lifted out of slips producing swab pressure

D B

+ve

B - surge pressure to break mud gel C - minimum run speed

Pressure

C

D - Maximum run speed

- ve

E

E - swab as pipe stops abruptly

A Time Pressure effects of lowering a joint of pipe into the hole Breaking gels can also add add to the surge effect of moving pipe. It is good practice to break circulation when running in the hole to break the gels to prevent problems when tripping in the hole.

Problems associated with sw ab / surge As described above, when the pipe is moved when tripping or running casing, pressures are generated in the mud column. These act to increase or decrease the static mud weight, even if momentarily. These can affect the bore hole in a variety of ways: Swab Swab - if the negative pressure is great enough, the mud hydrostatic can drop below that of the pore pressure. If there is porosity and permeability in the formation, pore fluid can be swabbed into the annulus.


Data Engineering

In the snapshot above, the string has become stuck and in the course of freeing the pipe fluid has been swabbed into the annulus. This shows as a gain in the trip tank. Surge Surge - if positive pressure is great enough to exceed the fracture pressure of the formation, fractures may be initiated and propagated, causing losses or formation damage. Before every trip out or in the ADT engineer should produce a report of the expected swab / surge pressures with maximum running speeds. Especially prior to casing runs, where good swab / surge information can prevent potentially serious hole problems. Liase with Operator personnel to define the pressure limitations and constraints.


Data Engineering

In the case above, It was thought that the formation was broken down by surging (mud weight 820 pptf, minimum open hole fracture gradient calculated to be 843pptf, Leak Off at 13 3/8” shoe was 863 pptf. Operator insisted that the LOT was the lowest fracture gradient. The casing surged, fractured the formation and 110% losses occurred. Very expensive. Conclusion Hydraulics optimisation is one of the main tasks of the ADT engineer. Modern applications make the calculations easy, but some interpretation and experience is required. Running bits with poor bit hydraulics is a costly business. The ADT engineer must be aware of the limiting factors when optimising hydraulics. Factors such as pore pressure and fracture gradient are the obvious ones, but the pump pop-off settings must always be considered, as should the pressure drops due to mud motors and MWD tools. Hole cleaning is important, but care must be taken to prevent borehole erosion. Occasionally the guidelines are broken, e.g. in high angle holes where cuttings beds must be broken up. Both swab and surge effects are present whenever pipe is moved in the hole and again various constraints may limit the scope for run speeds. Recent developments with PWD (pressure while drilling) allow downhole pressures to be measured, leading to a new field of hydraulics analysis.


Data Engineering

Well Control

This is a very brief introduction to well control. If a well control situation is encountered, ensure that the senior engineer is aware of it and takes charge in the unit. The work load on the unit in this situation can be high and as much information as possible should be recorded. A kick is a situation where the formation pressure has exceeded the hydrostatic hydrostatic pressure exerted by the mud i.e. there is a hydrostatic hydrostatic underbalance. underbalance. In a formation where there is little or no porosity or permeability, this manifests itself by swelling shale, caving and high gas levels. In a porous and permeable formation, the pore fluid can enter the annulus. This fluid is called an influx. Influxes can be gas, oil, f resh water, salt water or a mixture. - Kicks are usually the result of a series of events which lead to negative differential differential C a u s e s (underbalance) between the hydrostatic head of the mud and the formation pressure. Most kicks are caused by the f ollowing: ollowing: 1) Swabbing during a trip 2) Failing to keep the hole full while tripping 3) Lost circulation 4) Drilling into a high pressure zone 5) Shallow gas

- On detecting a possible kick, e.g. if a flow check showed that the well was Shutting in flowing, the driller would space out the pipe to ensure that no tool joints will foul the BOP rams, then shut the well in. Depending on the situation the well may be shut in hard or soft. On a hard shut in, the annular preventer is closed with the choke shut. Hard shut in can cause formation breakdown due to “water hammer”, but is quicker than the soft shut in. On a soft shut in, the choke valve is opened prior to the annular preventer being closed. The disadvantage of the soft shut in is that it allows additional influx to enter the annulus. BOPs can have 4 types of closure devices - Annular preventer (Hydril or “the bag”) which is a rubber doughnut shape which which is compressed to fit around any object object in the BOP. Pipe rams which are shaped to fit around drillpipe or collars. Shear rams cut the pipe while blind rams close against each other to seal the hole. Most BOPs are operated pneumatically by the Koomey unit. - if a kick is taken while off bottom, it is advisable to attempt to run back in to Stripping in bottom. Obviously running in conventionally will allow more influx into the annulus, so the pipe is “stripped” through the annular preventer, with the displaced mud bled off as the pipe is run in. Taking the pressures - having shut the well in, the next step is to take the various pressures. These pressures will govern how the well is killed. Having allowed the pressures to stabilise, the Shut In Casing Pressure and Shut In Drill Pipe Pressure are required. Normally the influx will enter the annulus, so reducing the hydrostatic hydrostatic head of the mud. Therefore the SICP will be higher than SIDPP. 45 of 53 LOGGING SYSTEMS

Data Engineering

Calculating the kill mud weight - by using the formula

Kill mud weight weight = (SIDPP/(0.O52 x TVDft))+ TVDft))+ MW The density of the mud to balance the formation pressure can be established. Usually Usually a safety factor is added to this mud weight to produce an overbalanced situation. Overkilling the well with excessive mud weight can damage the well. - the manner in which the well will be killed will be governed by the nature of Influx density the influx. Gas kicks can produce some dramatic effects during a kill operation, so prior knowledge of how the well should react does help. The height of influx is required to calculate density, as is the SICP and SIDPP. Height of influx = PIT GAIN(bbls) / ANNULAR VOLUME(bbls/ft) VOLUME(bbls/ft)

Next the influx density can be calculated. Kicks are always assumed to be gas until proven otherwise. SICP - SIDPP Gradient of influx (psi/ft) = MUD GRADIENT -

---------------------INFLUX HEIGHT

INFLUX GRADIENTS

GAS

1 - 3 ppg

GAS/OIL/WATER

3 - 5 ppg

OIL/W ATER

5 - 7 ppg

SALT W ATER

8 - 12 ppg

The aims of well control control : maintain constant bottom bottom hole pressure, thus preventing preventing further influx, remove the influx from the well bore and return the well to equilibrium. By introducing a new mud to the well, the hydrostatic overbalance can be regained. To kill the well by whatever method, the initial and final circulation pressures must be calculated. It is normal drilling practice to record Slow Circulation Rates (SCRs) once a tour, or when mud properties change. The pumps are run in turn at the rate to be used to kill the well and the pressures recorded. These are the circulating system system pressure losses. Before killing the kick, the initial circulating pressure is calculated:


Data Engineering

Initial Circulating Pressure = SIDPP + KRP* *From SCRs The final circulating pressure is then calculated: Final Circulating Circulating Pressure = KRP x (KILL MW / MW) These pressures allow a Drill Pipe Pressure schedule to be drawn up. This has pump pressure on the Y axis and Stroke pumped on the X axis. As the drillpipe pressure will fall at a constant rate during the kill, the graph can be drawn using the ICP and FCP alone.

While the kill operation is underway, the choke valve should be opened or closed to ensure that the drill pipe pressure follows the line on the graph.

Procedures for killing a w ell There are 3 methods of killing a well, differences between methods are generally determined by the number of circulations required and pressures generated. a) Wait And Weight Method : Method : the well is shut in, kill mud is weighted up, then pumped. Kill is achieved in one circulation. When the kill mud has reached the bit, the well is dead. Also known as the Engineer's method. Circulation continues until the influx is out of the annulus. W+W Produces the lowest casing shoe and annular pressures, but requires longer noncirculating time. b) Driller's Method : the well is shut shut in, the influx is circulated circulated out. Kill mud is then pumped pumped with the well being dead when the kill mud reaches the bit. The well is killed in two circulations with higher casing shoe and annular pressures than the Engineer's method c) Concurrent Method : the well is shut in, circulation begins as the mud is weighted up in increments. May require a number of circulations to kill the well. Requires more on-choke circulating time and produces higher casing shoe and annular pressures than engineer's method. May be used in the event of rapid gas migration.


Data Engineering

MAASP MAASP - the pressure effects on the well bore are due to gas expansion as the influx is circulated out. If the pressure exerted by the gas and mud column exceeds the fracture pressure of the formation, a sub surface blow out may occur. When the formation fractures, the influx may escape into the formation, leading to a loss of hydrostatic head. This in turn leads to more influx entering the annulus. To ensure that this does not occur the Maximum Allowable Allowable Annular Surface Pressure (MAASP) is is calculated. calculated. MAASP = 0.052 x TVDft CASING x (LOT - MW ) Gas migration migration - If a gas kick has occurred and the pumps are off, a gas influx will start to migrate upwards. As it migrates, the mud hydrostatic will be reduces. As the hydrostatic falls, the gas will expand. This can be dramatic, especially near the surface, and may displace mud from the annulus, (annulus unload) allowing more influx into the annulus. If the well has been shut in, migration still takes place, but expansion will not occur as the system is closed.


Data Engineering

PWD and ECD - a quick guide

This guide assumes absolutely no knowledge of drilling hydraulics, so be prepared to be patronised. Apologies in advance. PWD allows the Equivalent Circulating Density of the drilling fluid to be observed in real time and recorded for end of run analysis. What is ECD? ECD? - The Equivalent (or effective) Circulating Density is the actual pressure exerted on the borehole by the drilling fluid while the fluid is being circulated . When a drilling fluid is static in the borehole it exerts a pressure which is dependent on: a) density of the fluid b) depth (always True Vertical Depth) To calculate this pressure, a constant is used. This constant varies depending on the units of measurement in use: Pounds per Gallon (ppg) and feet (ft) the constant is 0.0519 Specific Gravity (sg) and Metres (m) the constant is 1.421 To calculate a pressure in psi, ppg x 0.0519 x ft or if metric units are in use, sg x 1.421 x m This gives a static pressure. pressure. Pressure losses losses - Friction in a circulating system is seen as a pressure on a gauge. The magnitude of this pressure depends on the point at which the measurement is taken.

B

A

PUMP

C

FLOW LINE

In the case above, the pressure gauge at A will show the frictional pressure loss for the entire circulating system from A to C. The pressure gauge at C will show zero psi as there is no frictional loss upstream from C, as this is the outfall to the flow line. The gauge at B will only show the frictional loss from B to C. Consider this diagram as being representative of the well with A to B representing the drill string and bit, B to C representing the annulus. The gauge at B represents the PWD sensor and therefore can only measure frictional loss in the annulus plus the static pressure of the drilling fluid. This pressure is the annular circulating pressure. Equivalent Mud Weight Weight - Having measured this effective circulating pressure, it can be converted to an Equivalent Mud Weight (EMW) by using the same constants as above:


Data Engineering

ppg = psi / TVD ft / 0.0519 or if metric units are in use, sg = psi / TVD m / 1.421 The system uses a pair of high accuracy quartz gauges, one open to the annulus, the other open to the bore of the collar. The annular gauge measures the frictional pressure loss in the annulus. Being a sum of the static mud pressure and the annular frictional pressure loss, this represents the actual pressure being exerted on the borehole. The PWD software converts that pressure to an Equivalent Mud Weight, but only if the relevant survey information is available to calculate True Vertical Depth, which is then used in the calculation of EMW. When converted to an EMW, this is the Equivalent Circulating Density (ECD). Why use Equivalent Circulating Density? ECD allows comparison of the actual pressure exerted on the formation while circulating and the static pressure exerted by the m ud. ECD will be greater than the static mud weight by a factor which depends on the drill string and hole geometry, mud properties, amount of cuttings suspended in the mud and flow rate. The PWD tool allows direct measurement of ECD and actual static mud weight, which will vary from that reported by the mud engineer as soon as any cuttings enter the mud column. Before PWD tools became available the ECD had to be modeled from hydraulics hydraulics models. Now that a real time measurement of ECD is available, it is possible to assess its impact on the down hole environment. The mud weight should be sufficient to control formation pressures without exceeding the fracture pressure of the f ormation. Fracture Pressure Pressure - Is the pressure required to initiate fractures within the formation matrix. These fractures may propagate if the applied pressure exceeds the initiation pressure. fractures lead to lost circulation, subsurface blow outs and ultimately to severe drilling problems. The fracture pressure is calculated by ADT engineers as the well is drilled using the Daines Fracture pressure model. This gives a calculated value for fracture pressure, but an actual value of fracture pressure can be obtained from Leak Off Tests (LOT). These are usually done after drilling out from a new casing string. Casing is normally set in a high strength formation such as a claystone, claystone, providing a maximum fracture pressure at that depth, but if a low strength formation such as a sand is encountered the fracture pressure may decrease. Some operators consider the leak off value to be the lowest fracture gradient in the open hole section, but this may not always be the case. More wells are being drilled under High Pressure / High temperature (HP/HT), these being characterised by having high formation pressures close to the formation fracture pressure. This means that to balance the formation pressure without fracturing the formation, the mud weight must not fall below the formation pressure or exceed the fracture pressure. The mud weight and the ECD must be in a “window” defined defined by pore pressure and fracture pressure. PWD allows the ECD to remain within this window by providing the information to allow modification of drilling pactices. If the EMW falls below the formation pressure, the hole may become unstable and collapse or if circumstances permit, the well may kick. If the EMW exceeds the fracture pressure, the formation may break down, causing losses, which may ultimately lead to a kick.


Data Engineering

Mud weight

A

ECD Fracture

DEPTH

B C

In the above diagram, the relationship between Pore pressure, fracture pressure, static mud pressure and ECD can be seen. At point A the “window” between the ECD and fracture pressure is large enough and poses no problem. At B the window has reduced to the extent that if the ECD rises, the fracture pressure may be exceeded and problems occur. At C the ECD is the same as the fracture pressure and may have already fractured the formation. Note how fracture pressure changes with lithology type.

Swab and Surge The equivalent mud weight is also affected by movement of the drill string whether the pumps are on or off. The simplest analogy is to consider the bit as a piston and the borehole as the cylinder. As the string is moved up and down, pressures are generated at the bit. Moving the bit up will reduce the pressure under the bit, reducing the EMW of the mud. This is called swabbing. Moving the bit down creates positive pressures, called surging. Swab / Surge mechanism and effects Moving pipe in the hole causes the drilling fluid to move, generating pressure. Hydraulics analysis allows allows these pressures to be calculated and their effects controlled. These pressures are referred to as swab (a reduction in pressure) and surge (an increase in pressure). These pressure changes can be seen on the PWD tool (recorded only at present) and allow direct comparisons to be made with calculated values.


Data Engineering

Mechanism Mechanism - Whenever pipe is moved in the hole both swab and surge effects occur, due to the flow of mud around the pipe. Pressure is generated in the opposite direction to pipe movement, mainly due to mud being displaced by the pipe or moving into the space formerly occupied by the pipe. Speed of pipe movement has a great effect on swab/surge pressures, with the maximum pressure generated at maximum pipe speed. If the pipe is closed ended (e.g. blocked jets or a float valve) the pressures are greater still due to lack of equalisation between fluid around the string and fluid in the string.

A - pipe lifted out of slips producing swab pressure

D +ve

B

B - surge pressure to break mud gel

Pressure

C - ve

E

C - minimum run speed

Time Pressure effects of lowering a joint of pipe into the hole Breaking gels can also add to the surge effect of moving pipe. It is good practice to break circulation when running in the hole to break the gels to prevent problems when tripping in the hole. Problems associated with swab / surge As described above, when the pipe is moved when tripping or running casing, pressures are generated in the mud column. These act to increase or decrease the static mud weight, even if momentarily. These can affect the borehole in a variety of ways: Swab Swab - if the negative pressure is great enough, the mud hydrostatic hydrostatic can drop below that of the pore pressure. If there is porosity and permeability in the formation, pore fluid can be swabbed into the annulus. A kick. Surge Surge - a positive pressure can induce and propogate fractures with the same consequences consequences as increased ECD. Formation fracture, lost circulation.

Hole cleaning As mentioned above the amount of cuttings in the annulus can effect the static EMW and the ECD. From this the annular pressure can be used to determine the amount of cuttings in the annulus and therefore how well these cuttings are being removed from the well bore. Hole cleaning is a function of cuttings size and density, mud rheology and flow rate. Flow rate is probably the most important factor, but in high angle holes string rotation becomes important. In a vertical hole an increase in the PWD EMW would indicate that the cuttings cuttings load is building up (usually in the largest annular section eg the riser) and this can be cleared by increasing the flow rate. If cuttings do build up in an annular section, there is a possibility that a pack-off will occur, exerting excess pressure on the wellbore and causing formation fracture.


Data Engineering

In a high inclination hole (50 ° +) cuttings beds (or dunes) can form on the low side of the hole, especially when the drillstring is not being rotated during long periods of sliding. This shows on the PWD EMW as a gradual decrease. When rotation resumes a sharp increase in PWD EMW will be observed as the cuttings beds are stirred up into the mud column.

As with most drilling parameters trends and changes rather than absolute values are more relevant to PWD monitoring.

The above log shows how annular pressure increases due to variations in the cuttings load in the annulus. In this case ROP had increased (due to increasing formation pressure) over the previous 30 minutes. The hole was being under-reamed and as the cuttings entered the smaller ID of the casing the ECD increased, exceeding the formation fracture gradient. Lost circulation followed followed and remedial action had to be taken. Conclusion PWD allows realtime measurement and monitoring of the ECD. This allows the ECD to be managed with the aim of minimising hole problems.


Data Engineering

Data Engineer Manual

Recommend Documents