Drilling Operations Guide
Drilling Operations Guide
SECTION 5
ROUTINE DRILLING OPERATIONS
Wells may be broadly divided into vertical and deviated categories. This tends to suggest that
wells will drill vertically by default, but this is not strictly true. Directional drilling
originally had the task of maintaining vertical direction in locations where the formations'
orientation, density contrasts, etc cause the well to deviate naturally in response (for example)
to refraction created through major rock matrix contrasts.
The term straight hole can also be applied to sections of deviated wells where there is no plan
to alter the inclination of the well through that section. Section 5 deals with drilling
operations associated with straight-hole conditions, whether in vertical or deviated well
conditions.
When the driller places the bit onto the bottom of the hole, some of the string weight is
transferred from the hook onto the bit. This weight is computed by subtracting the measured
hook load from the simulated string weight. By lowering the travelling block further the
driller can compress the drill string even more, thus adding to the WOB.
String Weight
Adjuster
WOB
Hook Load
As the bit cuts the rock, the compression on the drill string will be reduced and weight will
transfer back onto the hook (drilling off). The driller reacts to this by lowering the travelling
block to recompress the drill string and maintain a constant WOB. The driller holds a long
handled brake lever. Releasing the brake causes the travelling block to move downwards,
compressing the string and re-transferring weight from the hook to the bit. When the desired
WOB is achieved the brake is applied again, stopping the block movement.
This data is displayed on the driller’s control panel (fig 5-1) and also within the mud logging
unit (fig 5-2).
The length of each pipe section is measured to the centimetre (pipe tally). When the bit is on
the bottom of the hole, the depth of the well can be measured by observing the length of pipe
currently above the drill floor (derived from the block height) and subtracting this from the
total length of pipe forming the drill string.
Well depth is referenced from the drill floor (measured depth) and expressed as depth below
rotary kelly bushings (BRKB). The block height is displayed on the driller’s control panel
and is also measured in the mud logging unit.
Travelling Block
Current Block Height
If the driller constantly adjusts the block position in the derrick in response to WOB change,
then the rate of change of block position exactly represents the rate of penetration of the bit
through the rock, as constant WOB means that the length of the drill string is constant.
Variations in this ROP can therefore be related to conditions at the bottom of the well, i.e.
formation changes and bit wear. These ROP changes (drill breaks) and their depth of
occurrence are logged in the mud logging unit, and when associated with drill cuttings and
other drilling parameters will allow a geological interpretation of the rocks drilled to be
made. Where there is no association between changes in ROP and lithology, then it is
possible to interpret the change as a potential drill string problem.
If the driller does not keep WOB constant, then the ROP will appear to be variable since the
block is either stationary, or it is being quickly lowered to reapply WOB. In these
circumstances the ROP changes are less significant as indicators of down-hole behaviour.
must therefore be interrupted to allow another length of drill pipe to be added to the drill
string. This is referred to as making a connection
Between connections, the driller does not require the other members of the drill crew. The
roughnecks will either carry out maintenance of drill floor equipment, or will look after the
solids control equipment used to filter rock cuttings out of the mud. The derrickman works
with the mud pumps and mud tanks (pits). When a connection occurs, they are called to the
drill floor so that the new pipe can be manhandled into position.
The connection between the top drive (or kelly) and the drill string must be broken to allow a
new pipe to be added. Thus the travelling block will no longer be able to support the drill
string. Before disconnection, the roughnecks insert a wedge apparatus (setting the slips)
between the drill pipe and the walls of the well below the drill floor such that the joint to be
opened is around one metre above the drill floor. Since the string is now completely
supported by the slips the joint can be broken. Two roughnecks are involved here. Each
clamps a large wrench (tongs) around the pipe, one above the joint and the other below. Both
tongs are attached to winches located at each end of the draw-works (catheads). The tongs
are also connected to fixed posts on the drill floor. The driller operates the draw-works such
that one tong is pulled to open the joint.
for the travelling block to raise the bottom of the stand off the drill floor. When this occurs
they guide the base of the pipe towards the open pipe held in the slips. The driller carefully
lowers the pipe in the travelling block until the threads meet. The pipes are then screwed
together, with the draw-works now pulling on the other tong to close the pipe joint to the
required torque. The top drive circulating system can now connect to the top of the new pipe.
The driller raises the string a fraction to allow the roughnecks to remove the slips,
transferring the weight of the string back to the travelling block.
Monkey Board
The driller raises the travelling block to pull the kelly out of the hole and expose the joint to
be opened. The roughnecks insert the slips as above to support the disconnected drill string.
The tongs are used as above to separate the kelly from the drill pipe.
The next length of pipe to be added (single) is located in a storage hole (mousehole) adjacent
to the well. The roughnecks pull the bottom of the kelly towards the top of the single, and
then the driller lowers the kelly until the threads meet. The kelly has an electric or hydraulic
swivel built in to allow the driller to quickly screw the kelly into this new pipe. The
roughnecks then apply the tongs to allow the full torquing of the connection.
The driller now raises the travelling block to withdraw the single from the mousehole. The
roughnecks then guide the base of the new pipe over the top of the existing drill string in the
slips, and the driller lowers the block until the threads meet. Finally this last joint is tightened
as before, and the slips are removed.
Drilling then recommences with the driller again applying the same WOB.
The rotation of the drill string causes a reactive force to be created that tends to oppose this
action. This is called torque. This defines the degree of resistance that the rotary system must
overcome in order to make the drill string rotate at the chosen speed.
Torque is largely a function of the bit/formation reaction, but is not confined to this. In
straight holes, there will also be contact between the formation and the more prominent parts
of the BHA, i.e. the stabilisers.
This current flow is never steady, as the resistance builds up to peak value as the drill string
becomes more securely held by the formation (stick). This causes the drill string to become
twisted, and the rotation of the bit to reduce at the bottom of the hole. Current increases as the
system at surface reacts to this build up. Eventually sufficient power exists in the system such
that the formation's hold on the drill string is broken. The twist in the string is now able to
unwind, causing significant increase in rotation at the bit. Since this drastically reduces the
necessary power input requirement to maintain the chosen rotary speed, the current supply to
the system is reduced (slip).
Observation of torque at the surface then reveals a continuous stick-slip cycle. The size of the
peaks is determined by how sticky or abrasive the formations are.
Torque will also reveal failures in the drilling apparatus, such as bearings in tricone bits
seizing up or fragments of metal (junk) preventing proper rotation at the bottom of the hole.
Mud Density
Drilling fluid prevents influx through the pressure it exerts against the formations. However,
too much pressure can cause fluid to be lost to the formations through filtration or fracture.
Mud density is controlled by suspending fine-grained and dense material (usually barite)
within the drilling fluid. Significant chemical input is required to ensure that the weighting
agents remain in suspension even if the fluid is static for several hours.
Mud Viscosity
The drilling fluid also cleans the well of drilled material, allowing samples to be collected to
establish stratigraphy and also to prevent damage to the well caused by blockages in the
circulating system. To achieve this the drilling fluid must be capable of supporting rock
materials whose density is greater than that of the mud itself. This means that there will be a
tendency under the action of gravity for the cuttings to sink downwards through the mud
column. Viscosity is an expression of the thickness of a drilling fluid. This is controlled by
chemically adjusting the tendency of fine solid particles in the mud (barite and bentonite) to
attract each other electrostatically. The greater the attractive forces become, the more viscous
the mud becomes.
The viscosity can also create problems since a thicker mud requires more effort to circulate.
The frictional pressure losses are increased with viscosity, and hence the ECD will increase.
Mud Velocity
The rate at which fluid circulates is also vital to the hydraulic behaviour. Increases in velocity
cause the frictional pressures to increase. The nature of the fluid flow will also alter as the
velocity passes key threshold values.
The desired flow pattern is planar in nature, and is called laminar flow. Here the mud
behaves as parallel laminae of mud, sliding against each other. The velocity of each lamina
will depend on the friction it experiences with its neighbours. The greatest friction occurs
between the lamina adjacent to the borehole wall and the borehole wall itself, with the result
that this moves more slowly than those located towards the centre of the annulus.
The faster the fluid travels, the more likely it becomes that this laminar profile will begin to
decay into turbulent flow. This erratic fluid movement means that a highly erosive
environment is created beside the formations exposed in the open hole. Weaker formations
such as friable reservoir quality sandstones will wash away under turbulent flow conditions.
This results in poor quality wireline data since the diameter of the hole is critical to the
operation of many of the tools, particularly resistivity.
Mud velocity is controlled through the speed at which the mud pumps are operated, and by
the geometry of the different sections of the circulating system. For example pumping at
120spm will lead to vastly different velocities through the annulus, with the small drill
collar/open hole annulus travelling much faster than the mud flowing through the marine riser
annulus of a semi-submersible or drill ship.
The simplest and cheapest drilling fluid is water-based mud. Here the fluid required can be
pumped directly from the sea into the mud tanks before the chemicals are added to create
drilling mud. Water is relatively incompressible material, regardless of the pressure and
temperature to which it is subjected in the drilling environment. For water-based muds the
most effective model is the Power Law model, which is designed to assess the effects of
slow rates of circulation on the fluid behaviour.
An alternative is to use oil-based mud where the majority of the liquid component is highly
refined mineral oil. This is more costly to manufacture than water, but significant
improvements in drilling can be obtained through it's use, particularly where thick clay
sequences are being drilled. Many clay minerals will absorb water into their mineral lattice.
This causes the clay to break up and effectively liquify. The volumetric increase caused by
this process also pushes clay into the wellbore, reducing the hole diameter to a value less than
the bit or stabiliser diameters. While the drilling process may seem unaffected under these
circumstances, the swelling clay will prevent easy extraction of the drill string. This could
develop ultimately into stuck pipe, which may require sidetracks to be drilled.
Oil based muds are not absorbed by the clays, so the clays do not de-stabilise.
Use of oil-based mud means that the best model to select is that defined by API Bulletin 13D
(1985).
Environmental concerns regarding the effects of oil-based mud entering the sea as a coating
on drill cuttings led to the development of ester-based muds which gave similar protection
against swelling but were capable of bio-degrading once they were exposed to bacterial
action on the seabed. However, use of these muds has shown that some of the chemical
effects associated with this type of mud are just as problematic as the oil-based muds they
were intended to replace.
Some oil companies will transport oil-wet drill cuttings back to shore for treatment, rather
than use the bio-degradable mud systems.
Given the differences in the responses to these different fluids to pressure and temperature it
follows that their hydraulic behaviour will also be different. Given that esters are
manufactured from mineral and other oils (including fish oil) it follows that API Bulletin 13D
is a better selection than the water-based models.
Minimum mud velocity is determined by the rate of slip of cuttings through the mud in the
slowest moving annulus section, whereas maximum mud velocity is determined by the onset
of turbulence in the smallest (fastest flowing) annular section. Depending on the relative
values of pore fluid pressure and formation fracture pressure, the ECD effect (where the
annulus pressure loss is added to the annulus hydrostatic pressure) may cause the fracture
pressure to be exceeded at flow rates less than those at which turbulence may develop. In this
case the ECD will determine the maximum mud flow-rate.
Between these limiting factors will exist a flow rate at which the improved hole cleaning just
begins to interfere with the bit's contact with the hole bottom (optimum flow rate). This
interference is due to the enormous hydraulic impact force of the mud as it leaves the bit and
strikes the well bottom. At extreme levels this can oppose the WOB since it acts like a rocket
engine. Reducing the WOB artificially by this method takes the bit out of it's optimised WOB
and rotary speed combination.
Although cuttings samples allow the geologist to determine the rock types being drilled, it
should be remembered that the samples themselves are composites of (at best) a 10-foot
interval. Any stratigraphic variation within this interval cannot be detected by cuttings
analysis alone. This is illustrated in the example below.
A cuttings sample was collected which represented a-10 foot interval. Subsequent analysis
showed it to be comprised of 50% sandstone and 50% claystone. A number of interpretations
were possible, illustrated in figure 5-6.
However, the above do not allow for the presence of caved material, the effects of lost
circulation, the effects of drilling fluid hydraulics, etc., which would allow an even greater
range of (erroneous) interpretations to be made.
Clearly some way of interpreting the samples, which does not rely solely on the cuttings
themselves, must be found. A more frequent sample interval, e.g. 1 foot, would appear to
solve these problems but is impractical at even moderate drilling rates due to the time
required to correctly process each sample.
The cheapest and often most effective solution is to consider the response of drilling
parameters measurable at the surface. The key parameters are rate of penetration and
torque, but weight on bit, bit rotary speed and pump rate must also be considered.
If Logging While Drilling (LWD) tools are run, these will give more accurate interpretations
since there are fewer factors to affect the LWD output compared with ROP and torque.
However the LWD tool may be as much as 40 feet above the drill bit. Always check ROP
and torque data and compare with LWD data as drilling progresses.
ROP shows a general decrease with increasing depth in response to porosity reductions as a
result of compaction, but increases in ROP due to lithological changes or over-pressured
formations frequently occur.
Variations in ROP can therefore be used to detect transitions from one rock type to another,
or to detect change in the internal characteristics of a given formation.
These changes are referred to as drilling breaks, with transition to a faster rate being termed
a positive break, and to a slower rate being a negative break.
The depths at which drill breaks occur indicate the formation changes, and should be logged
onto worksheets and lithologs.
Positive
Negative
Progressive breaks will occur if the formation changes gradually, eg in a fining upward
sequence where coarse material is found at the base of the formation, but grades into finer
material at the top. In such a case, penetration would show a gradual increase in rate as the
formation is drilled from the top down.
All positive drill breaks must be investigated by taking a spot sample when the cuttings
associated with the break reach the surface. If the drill break occurs at a suspected reservoir-
quality zone, a flow check should be recommended before circulating bottoms up.
If the well program requires a core to be taken, it is imperative that the driller be informed
after no more than 5 feet are drilled at the faster rate. Since cores are taken in possible
reservoirs, it should be standard practice to perform a flow check prior to pulling out to alter
the BHA for coring. The wellsite geologist should be informed once the flow check is
underway. A decision to circulate the last drilled cuttings to the surface (CBU) before pulling
out or resuming drilling can then be taken.
Figure 5-7 above shows two drilling breaks as the well is drilled deeper. The first is a positive
drill break, since ROP increases as the bit passes through the boundary between clay and the
sandstone below. The drilling parameters show no change at this depth, indicating that the
effect is associated with change down-hole. As the bit enters the drill break, the geologist can
assume that a more porous rock has been entered. Note that this interpretation will only be
confirmed when cuttings have been collected at the surface. This lag time interval will
depend upon the geometry of the well, and the speed of the mud pumps.
A second drill-break, this time negative, occurs as the bit passes through the sandstone into
the underlying limestone. Again, the drilling parameters imply that the change is down-hole.
Only the examination of cuttings from the drill break depth will allow the geologist to
confirm that the ROP reduction is related to lithology rather than bit failure.
5.5.2 Torque
This is less clearly definable than ROP in terms of lithological indications. The importance in
the case of torque is not the numerical value, but more the change in character of the torque
trace on a chart recording (see Stick-Slip below).
Some rocks are sticky, others abrasive, still others may be hard and competent. All will
therefore cause different reactions as a drill bit is rotated within them.
In general, one can say that an increase in torque together with a positive drill break is
indicative of a sand or sandstone being penetrated, as illustrated in figure 5-7.
Torque increase coupled with ROP reduction is more ambiguous. This could indicate that
sticky shale is now being drilled, but equally could indicate problems developing with the bit
bearings, or swelling of the hole walls trapping the BHA.
Reduction in torque coupled with reduction in ROP may mean that a particularly hard,
competent rock such as chert is now being drilled. In figure 5-7, hard limestone is penetrated
below the sandstone. Alternatively the teeth may be extremely worn, although the bearings
still intact.
To summarise, all torque changes (as well as ROP change) should be investigated by
collecting a spot sample when the cuttings associated with the change reach the surface.
Torque variation should be used as a prompt for further investigation. All changes should be
noted on worksheets and lithologs, and significant changes reported to the driller and
company man immediately.
Stick-Slip
The torque trace recorded in mud-logging unit charts shows that torque is not a steady
parameter value. This variation can be very large in some situations. The average torque
value is therefore unhelpful, as a similar average value could be calculated from two
completely different torque patterns. It is the shape of the trace, rather than the values
themselves, which should be monitored.
This variation in value is due to a phenomenon called stick-slip. As the bit is rotated at the
surface, electrical power is supplied to the rotary motor to maintain revolutions at the set rate,
e.g. 100rpm. The formations at the bottom of the hole will interact with the drill bit according
to their character. Some will hold the bit, and prevent the bit rotating. This introduces a twist
in the drill string (stick phase). The surface system reacts to this by supplying increased
electrical current to the surface system. Electrical power (Watts) is the product of current and
voltage. Eventually, the power built up in the twisted pipe will become sufficient to overcome
the resistance of the formation, and the bit will be suddenly released. This allows the twist to
deplete itself, which adds energy to the bit beyond that supplied to the rotary system (slip
phase). As a result, it becomes easier to maintain the desired rotary speed so the electrical
current supply reduces.
Torque is measured on the rig by monitoring the current supply in the cable linking the
generators to the rotary motor. Stick-slip therefore causes the current to continually move
between peak and trough values.
Most drilling rigs have sensors to monitor these parameters. Where a mud-logging unit is
available, charts of these parameter variations against time will be available. If changes in
ROP or torque are noticed, one’s first action should be to check these drilling parameters for
change. If the chart recordings show that the driller has not altered anything, then the change
in ROP or torque can be considered to be down-hole, and the necessary reports made.
Note that “down-hole” refers to any situation at the bottom of the hole. This includes
formation changes and problems like bit failure, sticking pipe, etc.
In figure 5-8 below, the variations in ROP and torque within the clay, sandstone, and
limestone could all be attributed to alterations in the formation. However, when the driller
enters the siltstone, he begins to adjust drilling parameters to obtain optimum ROP and torque
values. WOB is increased in two stages, with corresponding alteration to ROP and torque.
The eventual torque achieved is too great for safety, as increased torque can cause reduction
in the life of both the bit and the elements of the drill string. To reduce torque the driller has
increased RPM slightly, producing the desired torque reduction and (in this case) an
additional increase in ROP. The final operation is an increase in pump speed, which results in
improved bit cleaning and consequent increase in ROP. Had the pump increase been too
great, however, the increased impact force acting on the bottom of the hole would have
opposed the WOB and may have reduced ROP.
The values of RPM, WOB, and pump speed always represent a compromise to allow
maximum ROP without causing dangerously high drilling torque
Fig 5-8: Log Showing Effects on ROP and Torque of Driller Varying
Parameters
A vital component of the data gathering and monitoring systems at the wellsite is the safety
service. This is carried out in the first instance by the drilling contractor, and is generally
backed up by a mud-logging service.
Explosive Gases
The first and perhaps most obvious danger to the rig comes from the hydrocarbon itself. The
shorter the hydrocarbon molecule chain becomes, the more volatile it's behaviour. Methane
gas is the smallest of all hydrocarbons, and also the most commonly occurring.
It must be stressed that dangerous accumulations of gas may occur even when no obvious
reservoir unit is being drilled. Factors such as mud density and viscosity, and even ROP and
pump rate can increase the concentration of gas in the atmosphere.
Methane will burn, like any other fuel, when three conditions are met: there exists sufficient
quantity of fuel in the environment, the environment contains sufficient oxygen to support
combustion, and there is a heat source of sufficient energy to ignite the fuel in the
environment.
The first condition refers to the Lower Explosive Limit (LEL) of the gas. There is a
threshold concentration of any flammable gas that must be exceeded before it will burn in the
atmosphere. For methane this LEL value is 4.8% in air (48000ppm).
The supply of oxygen for combustion at a wellsite is simply that found naturally within the
atmosphere itself. There is no practical way to remove this factor.
The third criterion is the heat source that can cause ignition. This factor is controllable at the
wellsite by a system limiting activities that could cause spark production. Thus must be
rigidly followed. An efficient gas monitoring system must also be installed such that the
concentration of gas released by the drilling fluid at the surface can be quickly measured.
Such a system is called a permit to work system. Spark producing activity is classified as
hot work. Each job in this category must firstly be awarded a hot work permit before it can
commence. This allows the rig's control room personnel to track both the activities
currently underway together with their location. In the event of gas levels rising towards the
safety threshold, all hot work permits are immediately cancelled until the gas levels return to
safe levels.
Toxic Gases
The most significant toxic gas that may be encountered during drilling is hydrogen sulphide
(H2S). This can be lethal at even relatively small concentrations of a few hundred ppm in the
atmosphere. Concentrations as low as 10ppm can have lasting detrimental effects on
personnel exposed for even a few minutes.
H2S may be associated with any rock that has been deposited and maintained under anoxic
conditions. Organic matter decay in the presence of oxygen is harmless, but without oxygen
any sulphur in the environment will become important.
H2S release into the atmosphere can be controlled to some extent by the pH of the drilling
mud. At high pH levels the H2S exists within the drilling fluid in it's ionised form, as H+ and
HS- ions. In this state it is harmless. However, if pH were to fall below 3.5 all the ions would
recombine to form H2S molecules that may then be released from the drilling fluid into the
atmosphere.
Drilling fluid pH is therefore maintained as an alkaline solution with pH in the order of 10.
H2S is heavier than air, so will concentrate in low-lying areas around the rig-site. Efficient
monitoring systems should be located at all potential gathering zones whenever known H2S
areas are drilled. In areas with no expected or known H2S hazard, it is wise to at least employ
a sensor to measure H2S at the point where the mud returns to surface.
H2S is also particularly corrosive, forming acid when dissolved in water. This attacks both
steel and concrete structures.
Other gases which may be encountered, and which may pose a hazard to health, include
carbon dioxide (CO2) and nitrogen (N2). Although these are not poisonous, they act as
asphyxiants in sufficient concentration. Again efficient monitoring systems must be used
wherever the occurrence of these gases is considered to be possible.
Formation fluids (gas, oil, and water) can also pose significant threat to the rig if they are
allowed to enter the well as a flow. The primary control against this phenomenon is the
density of the drilling fluid in use. This creates a pressure (hydrostatic or ECD) which
should at all times exceed that of the fluids contained within the pore spaces of the formations
drilled. This situation is termed overbalance.
To be able to ensure that sufficient density is being used, it is essential to be able to evaluate
the formation fluid pressure in the well at all times. This can be done by a pressure engineer
using mathematical models of porosity behaviour in thick clay sequences, and by observation
of the well's behaviour in relation to the current mud weight (connection gas peaks, overpulls,
pressure cavings, etc).
When the pore fluid pressure exceeds the drilling fluid pressure the situation is called
underbalance. This differential pressure tends to force fluids through the available
permeability of the formations until they reach the wellbore where they mix with drilling
fluid and are circulated towards the surface.
If the rock permeability is very small (e.g. in clays) the underbalance will cause the formation
itself to swell into the wellbore, thus reducing the effective well diameter. This can lead to
stuck drill pipe if the diameter of the well becomes less than stabiliser or bit diameter. The
consequences of stuck pipe can be extremely costly, especially if the down-hole equipment
cannot be retrieved and the well is sidetracked.
Influxes of drilling fluid into the well are termed kicks. Most kicks are quickly recognised by
the crew on the rig, and are consequently dealt with very quickly and relatively easily.
Kicks that are allowed to grow out of control, through failure to recognise them or by poor
kick management, may eventually lead to unrestrained flow of formation fluid to the surface.
This is termed a blow-out. If the blow-out fluid is hydrocarbon, there is the additional risk of
explosion and fire (fig 5-9).
Kick Detection
Detection of the entry of fluid into the well-bore depends on the interpretation of surface
measured parameters such as mud flow into and out of the well, volumetric behaviour in the
mud tanks, and pressure variations in the circulating system.
Since the volume of the well is finite, any fluid entry will displace an equivalent volume of
drilling fluid at the surface.
The rate of fluid flow from the formation will be added to that supplied by the mud pumps.
Gas expansion will occur as the influx travels to surface since the pressure of mud above the
influx is reduced as the surface is reached. This will further accelerate the mud flow-rate out
of the hole.
The first surface sensor to react to a kick will be the mud flow out (MFO) system. This is
usually a paddle device, located in the flow-line (or ditch), which creates a variable electrical
current in response to the degree of paddle deflection by the mud. This is compared with the
flow-rate going into the hole by studying the pump speed. If the driller increased pump rate,
a near immediate corresponding increase in flow out should occur. A kick would be seen as
an increase in flow rate out of the hole with no corresponding alteration to pump rate.
This sensor reacts first because it is located as close as possible to the top of the well annulus.
All other sensors are further downstream so the mud surge will reach them at some later time,
depending on the design of the rig circulating system.
For example, on a simple land rig the distance that a surge must travel between mud flow out
sensor and active pit is likely to be as little as 20 metres. On a tender-assisted platform, where
tanks and pumps are located on a barge alongside the main drilling platform, this separation
may be as much as 100 metres. All personnel responsible for monitoring well behaviour must
be fully aware of the effects on sensor response related to the rig design.
During drilling, the well becomes deeper so that the volume of drilling fluid required to keep
it full will grow at a rate controlled by well geometry and ROP. This behaviour can therefore
be easily calculated since both the geometry of the well and drill-string are well known.
Comparison of actual volumetric change with calculated allows the driller or mud-logger to
determine whether the well is under control or not. If the reduction in surface tank volume is
less than that expected, or worse the tank volume is increasing as the well gets deeper, then
there is likely to be an influx of material into the well which is displacing an equivalent
volume of mud into the tanks.
However, since the active pit volume is expected to decrease over time, eventually it will
have to be re-filled. This process is called a transfer, and involves fluid with the same
properties as the active pit mud being pumped from another tank (reserve tank) over a
period of time. The transfer is not rushed so as to ensure that the new mud mixes as
efficiently as possible with the existing mud to smooth out any discrepancy in density or
viscosity that may exist.
Such an activity will cause active pit volume to increase regardless of the behaviour of the
well. By monitoring the sum of all surface tank volumes (total pit) this can be used to factor
out the effects of the transfer since this simply involves fluid within the measured system
changing location within that system. The change in total pit volume over this period should
therefore be related entirely to the behaviour of the well.
Tank volumes can also be affected indirectly, particularly on floating rigs. Semi-
submersibles have a large deck storage capacity for containers bringing supplies to the rig via
supply boats. The crane operation will alter the draft (trim) of the rig as containers are added
to or subtracted from the deck load. Periodically the ballast control operators will adjust the
rig to compensate for this and so keep the deck level. This is referred to as trimming the rig.
The resulting effect in the tanks can mimic gains and losses in volume since the mud will
flow according to the orientation of the rig. Depending on the pit sensor location, the mud can
either gather below the sensor or move away from the sensor. Good communication between
the rig control staff and the monitoring departments is essential here.
Transfers of fluids between storage tanks (diesel, water, etc) can also affect rig trim.
Standpipe Pressure
This term is used to describe the total circulating system friction pressure loss as fluid is
circulated around the well. It is measured at the base of the standpipe, on the standpipe
manifold.
The friction pressure in the well is affected by changes in fluid properties (density and
viscosity), fluid velocity, and conduit geometry. When an influx occurs, the intruding fluid
will generally have density and viscosity less than that of the drilling fluid. There will be a
reduction in friction pressure generated in the annulus segment of the circulating system as a
result of this.
Since the annulus pressure is only around 5% of the total system pressure loss, the alteration
caused by the intrusion of fluid will be difficult to detect initially, but will become more
pronounced the longer the well is permitted to flow without being brought under control. It is
unlikely that a kick will be first recognised by changes to standpipe pressure, but the pressure
behaviour may help to confirm that a kick has taken place.
Drilling fluid will be lost from the well for a variety of reasons, including filtration and
fracturing.
Filtration Losses
An overbalanced mud creates a differential pressure that tends to force the drilling fluid into
the formation. If the formation is impermeable, then fluid loss should be around zero, but
permeable rocks like sandstone can accept a large volume of fluid into the formation in
response to this overbalance effect.
To combat this the drilling fluid is designed such that the mud solids (bentonite and barite)
suspended in the mud have a particle size larger than the anticipated pore throat diameters.
Thus, the suspension of drilling fluid is separated into clean fluid (filtrate), which enters the
formation, and deposited solid (filter cake). The filter cake should build up to a thickness of
around 2 mm, which should then prevent further filtration losses from occurring.
Should the mud solid particles prove insufficient to close the rock's pore throats through
normal filtration, lost circulation material (LCM) may be added to the drilling fluid to more
efficiently block the available permeability of the formation.
If overbalance is excessive (up to 200psi is accepted as ideal) then filtration may continue,
causing greater fluid loss and consequently thicker cake development. This can pose a stuck
pipe hazard (differentially stuck pipe). The filter cake material is very sticky, and the
differential pressure will hold the pipe against the cake at any point where the drill string
comes into contact with the cake. This is particularly difficult to rectify since in many cases
the stuck point is located above the jars in the BHA. Jars require large tensions (overpull) to
operate. The stuck point represents the point below which there will be no tension as the
driller pulls on the drill string. Differentially stuck pipe often results in the stuck portion of
the drill string being blown off, and the well side-tracked around the obstruction.
Where movement has occurred along the fracture plane this is termed a fault. The action of
the fault can often cause the rocks along the plane to be reduced to powder and so reduce
permeability to virtually zero. Fault planes of this type pose no fluid loss hazard.
However, many planes will remain as open fractures through which drilling fluid can be lost
in extreme volumes. The only control that may be applied to control these problems is to
reduce drilling fluid density to balance point, and then pump cement into the fracture to
create a seal. Should the density of the drilling fluid be reduced to the point of underbalance,
the formation fluids initially occupying the fault plane will instead flow at great rate into the
wellbore. Hazards of this type are encountered in gas fields in Morecambe Bay, England.
Where no movement has occurred in the plane of fracture, this is termed a joint. The absence
of a grinding action means that joints tend to remain open. They may also have been enlarged
through water action at some point in the rock's history. Jointed rocks may act as excellent
hydrocarbon reservoirs if the joints are connected in two planes (for example clints and
grykes found in granite formations). Commercial hydrocarbon reservoirs in granite are
known in areas such as offshore Vietnam. As with faults, this high fracture permeability can
cause significant fluid loss if drilled with overbalanced drilling fluids, while posing a severe
kick threat if drilled with underbalanced fluids. Rocks of this type include Horizon B
limestones found in Algeria.
The surface pressure supplied by the pumps is added to the actual drilling fluid hydrostatic
pressure and converted to equivalent mud density by rearranging the hydrostatic pressure
equation.
However, the casing shoe formation is always selected as a formation capable of bearing the
stresses associated with circulation of heavy cement, and so is less likely to remain as the
weakest open hole formation as various sandstones and fractured limestones are penetrated as
the well progresses.
SICP
Rupture
Leak
Off
Propagation
SICP
Linear
Build Up
Volume Pumped
Fractures of this type may be small enough to be controllable using lost circulation material
(LCM).
Pressure exerted by drilling fluid depends on both the density of the fluid and the vertical
height of the fluid column. When lost circulation occurs, there is a danger that the height of
mud in the hole will reduce. This will only occur if the losses are 100%, ie the flow rate into
the formation exceeds the flow rate of the mud pumps. Should this occur, it is possible that
one or more formations exposed in the open hole will become underbalanced, even though
they were originally drilled without difficulty. Underbalanced permeable formations will
begin to flow into the well-bore.
Normal kick detection relies on interpretation of flow rates and volumes in the mud tanks.
With 100% losses, this is no longer available, so fluid entry into the well during losses will be
undetectable until the fluid reaches the surface. The potential for blow-out under these
circumstances becomes enormous.
Partial losses will allow the well to remain full, so there is no loss of hydrostatic pressure.
The consequences of partial losses are related to cost (many drilling fluids are formed of
expensive chemicals) and geological data interpretation (the lag time of the cuttings
becomes impossible to assess accurately).
The difficulty with LCM lies mostly with the efficiency of down-hole tools located in the
BHA, such as directional survey tools or formation evaluation tools. The LCM can plug up
the working parts of the tools, preventing them from communicating with the surface. This
usually necessitates retrieval of the tools and so causes lost drilling time.
LCM is also sometimes confusing to the wellsite geologist as it becomes mixed in with true
drill cuttings samples. Most LCM will be obvious in identity, bit some materials are easily
confused with cuttings, e.g. nut plug (walnut shell material) mica flakes, etc.
The nature of the ROP and torque variations will depend upon the bit type in use (tricone,
PDC, diamond, etc). The critical analysis is whether or not any surface induced effect or
lithological effect can be recognised at the same point where ROP and torque alter.
In general terms, ROP will reduce with any failure in the bit, whether teeth or bearings have
failed. Obviously, alterations in formation porosity, WOB, rotary speed, or pump rate may
have similar effects, so no decision regarding pulling the bit out of the hole should be made
until the cuttings have been studied.
With tricone bits, failures in the cone bearings inevitably result in increases in torque since
the bearing failure causes a cone to stop rotating (or at least resist rotation).
In PDC bits a cutter may fall off the body of the bit. This causes a ridge of undrilled
formation to project upwards. This will soon come into contact with the bit body causing
greater frictional drag to be developed on the bit.
Worn teeth on a tricone bit will cause torque to reduce, since the cutting action of the tricone
bit is intended to develop a large amount of energy at the cutting edges of the teeth. If the
formation is relatively hard (e.g. crystalline) then the worn cones will rotate like wheels
against the bottom of the well.
With all bit failures, there will be no alteration in lithology over the interval where the failure
becomes apparent on the drilling responses.
In all cases, the opening up of a new aperture through which drilling fluid may pass will
cause a reduction in the amount of energy that must be supplied to overcome the frictional
resistance within the circulating system.
Standpipe pressure is the primary sensor that will reveal the creation of a washout in the drill
string. The standpipe pressure is the sum of all the circulating system pressure losses (surface
pipes, drill string, bit nozzles, and annulus) Any change in any or all of these sections will
cause the standpipe pressure to react.
Since it becomes easier to pump fluid through the combination of bit jets and washout
aperture, the standpipe pressure is expected to reduce should a washout occur.
However, standpipe pressure also reduces whenever losses to the formation occur, or when
influx of lower density or viscosity formation fluids occurs. To distinguish between these
events, the flow rates in and out of the well must be compared, as must the volumes in the
surface tanks. A pipe washout introduces no change in flow rates at the surface, and no fluid
is gained or lost within the circulating system. Any pressure reduction associated with no
change in pump speed, no change in fluid return rate, and no change in active tank volume
should be interpreted as a pipe washout.
Since all the mud is no longer passing around the bit, there is a loss in lubrication at the
cutting faces that may result in increased torque. Flow rate through the bit jets affects
hydraulic impact, and hence hole cleaning, so penetration may also be reduced.
If standpipe pressure loss is continuously increasing in rate, the hole or crack in the pipe must
be increasing in size. If this is allowed to continue, the pipe will eventually separate into two
pieces. This phenomenon is referred to as a twist-off. The consequence of twist-off is a
period of lost drilling time since the broken pipe must first be retrieved before drilling may
resume.
Pipe washouts should result in drilling being suspended, and the drill string being slowly
pulled from the hole and inspected to located all the hole which may be present. Gas analysis
can allow the number of holes to be estimated by performing a check where a pulse of
acetylene gas is circulated through the system. The gas will leak a proportion of its total
through all the apertures in the pipe, resulting in gas peaks being detected for each hole. The
last peak should be interpreted as that exiting the drill string through the normal bit jets.
In poorly cemented rocks such as reservoir quality sandstones, any turbulence within the
fluid circulated through the annulus will cause the rock to break down as the cement bonds
between the sand grains are broken. The washing away of rock material widens the hole, with
the eventual result that wireline data quality will be compromised. Washouts are revealed by
the presence of otherwise unexplained formations within the cuttings samples. For example,
if all data implies that the bit is cutting through shale yet the associated cuttings have a
significant percentage of sand within them, then the conclusion must be that the sand is being
derived from a formation already exposed higher up in the well.
To reduce the washout effect, the pump rate must be reduced to bring the drilling fluid back
into the laminar flow regime.
Other formations are soluble in water, and so will be washed away in the presence of water-
based drilling fluids. Rocks in this category include rock salt (halite) and chalk (relatively
soft limestone). The only cure for this condition is to use oil-based drilling fluids in
subsequent wells.
Apart from wireline quality, the quality of the cementing operation that binds casing in place
may be seriously affected when the formations are badly washed out. Cement must continue
inside the previous casing shoe if all the exposed formations in the open hole are to be
isolated from the surface.
The wireline caliper log should be checked to determine the locations where formation
washout has occurred. During drilling, a gas check using acetylene pulses (carbide check) can
be made to compare calculated circulation time with that of the acetylene pulse. The
difference in the two values may be related entirely to the diameter of the open-hole section
of the well.
The first option to release stuck pipe is to pull the drill string with a force greater than that
simply required to raise the drill string through drilling fluids (overpull).
At critical overpull values, a special section of the BHA known as the jars will operate. This
is keyed by the tension in the drill string. When operated, it expands suddenly causing a
shock wave to develop, which is intended to break the hold that the formation has on the drill
string. Since the widest sections of drill string are located in the BHA, the jars are located at
the top of the drill collar section. Jarring may take place for several days before the string is
released.
Failure of the jars to release the drill string results in the string being blown apart by a small
explosive charge just above the stuck section. A wireline tool capable of sensing tension in
the steel drill string is lowered down the inside of the string to locate the top of the stuck zone
(free point) prior to the explosive operation.
After withdrawing the free drill string, the trapped section must be sealed in place using
cement. A deviated drilling system must them be employed to change the well direction to
allow the new path to move around the abandoned section. This process is called
sidetracking.
Underbalance
In impermeable formations, underbalance causes the formation itself to swell into the
wellbore in response to the differential pressure. Associated with this one can expect to
observe overpulls at drilling connections, increased drilling torques between connections, and
the presence of pressure cavings at the shale shakers.
Stuck pipe in response to underbalance can be avoided simply by reducing the degree of
underbalance.
Overbalance
When highly permeable formations are drilled with significant overbalance, filtration of the
drilling fluid will occur resulting in the deposition of excessively thick filter cake. At any
point where the well is not vertical, the drill string will not be central in the hole. If the pipe
comes into contact with the filter cake, it may become stuck. The differential pressure will
tend also to push the pipe harder against the filter cake, resulting in the phenomenon referred
to as differentially stuck pipe.
Overpull and increased torque will be observed, but there will be an absence of pressure
cavings. The cuttings should reveal the presence of permeable rather than impermeable
formations.
Chemicals may be added to reduce filtration, or the degree of overbalance may be reduced.
Casing can be used to isolate the potential sticking zone at the first opportunity. Careful well-
planning to avoid making direction turns in permeable zones is also important.
Water-Swelling Clays
Smectite clays (montmorillonite, etc) display a tendency to absorb water into their mineral
lattice. This causes the volume of the clay mineral to increase. Since vertical expansion is
impossible due to the weight of overburden, this expansion will occur laterally into the
wellbore. The resultant reduction in well diameter allows the pipe to become trapped.
As with all stuck pipe situations, overpull will be observed at connections, while torques will
increase during drilling. Clays in the cuttings will have the appearance of soft, plastic mush
(gumbo clays).
These clays do not share their affinity for water with oils, so drilling subsequent wells with
oil-based drilling fluids will control this phenomenon.
Drill string is always removed from the hole as stands. All the drill crew will be involved in
the tripping operation throughout, which may take several hours depending on the well depth.
The kelly must be completely removed from the travelling block hook, and stored in it’s own
location called the rathole. The weight of the kelly makes this operation awkward.
Once disconnected, the tripping process becomes repetitive as each stand is withdrawn and
stacked neatly into the derrick. As each stand is put into place, one of the roughnecks will
write the stand number directly onto the pipe in wax crayon. The stands will be re-inserted
into the hole in the exact opposite order in which they were removed, so that the position of
each pipe joint is precisely known in relation to the Blow Out Preventer (BOP) systems.
To remove a stand, the driller lowers the travelling block to a point just above the drill pipe in
the slips. The roughnecks clamp the elevators suspended from the travelling block around the
pipe just below the joint. Drill pipe joints (tool joints) are intentionally wider than the main
body of the pipe. When the elevators are engaged, the diameter of the elevator aperture is
equal to the main pipe diameter, but much less than the joint diameter. Thus when the slips
are removed, the driller can pull the pipe out of the hole by raising the travelling block. The
elevators support the pipe. The driller will pull out one stand, leaving the next joint to be
opened at around one meter above the drill floor.
The roughnecks apply the tongs so that the draw-works can open the joint. Once free, they
can then push the bottom of the stand into it’s storage location. The driller lowers the block
until the base of the stand rests on the drill floor.
On the monkey board, the derrickman will throw a rope or chain around the top of the stand
as it appears before him. He can then pull the stand close enough so that he can disengage the
elevators and pull the top of the stand into it’s location. The monkey board is slotted for this
purpose (fingers) so that stands can be stored in neat rows (figure 5-12).
Elevators
Driller's
"Dog House"
Draw Works
Pipe Stacked in
Slips
Racks
Between the drill pipe and the drill collars, there is usually a section of stronger drill pipe
with the same pipe body diameter as normal drill pipe but much smaller internal diameter.
Since this also has wider joints, the pulling process for heavy weight drill pipe (HWDP) is
identical to that described above.
Derrickman
Hook load can also be observed when looking for tight spots in the hole, i.e. locations where
the diameter of the hole has been reduced making it difficult to move the pipe through. As
pipe is withdrawn, the hook load is expected to fall by the weight of one stand as each stand
is removed. Increases in hook load above this trend suggest that extra force is required to
raise the string above that needed to overcome the weight of the string only. This effect is
called overpull.
If a tight spot is encountered as pipe goes into the hole, the obstruction will cause weight to
be transferred to the bit, reducing the hook load measured at surface to a value less than
expected.
Note that the buoyant action of the mud causes the hook load when pulling out to be greater
than when running the same drill string in, as the buoyancy supports the drill string as it is
run into the hole, but has little effect as the string is withdrawn.
The direction of travel is also essential, as the hydraulic effect reduces effective mud density
when pulling out (swab), but increases the effective density due to compression of the mud
(surge) when running in. This also affects the volume calculation as mud is displaced from
the hole if pipe goes in. Direction can be gauged by looking at hook load as the pipe moves.
If the pipe moves upwards with a high hook load, then pipe is being pulled. If hook load is
low when the block moves up then the driller is raising the blocks up with the string
stationary in the slips. Similar reasoning can be applied when the blocks are lowered in the
derrick.
Swabbing
Due to the hydraulic problems caused in the well during tripping by the mud’s response to the
pipe movement (swabbing) it is possible that gas could be drawn into the well below the bit.
It is essential to monitor the pipe speed as the pipe is moved as this is the only means of
controlling the degree of swab pressure reduction.
d × vel 2 × l × F
Pann =
( Dh − D p ) × units const
where
d = mud density
l = length of drill string
F = Fanning Friction Factor (viscosity dependent)
vel = mud velocity (calculated from pipe speed)
Dh = hole or casing ID
Dp = outer diameter of drill string
Provided the overbalance in the drilling fluid is greater than the swab pressure effect, there
will be no fluid entry into the well.
Surging
Surge pressure is the equivalent effect to swabbing encountered when the drill string is run
into the hole. In this case the surge pressure is added to the drilling fluid hydrostatic pressure,
with the corresponding problem that there may be sufficient increase in pressure to cause
fracturing in the open hole below the drill bit.
The shape of drill pipe and heavy weight drill pipe introduces some complexity into the
calculation of pipe displacement since the joint sections of the pipes have different diameter
to the main body of the pipe. Although apparently insignificant when considering a single
pipe, in a drill string with over 400 such pipes the cumulative effect can be significant in
terms of the number of pump strokes.
To combat this, the size of the joints at each end of a piece of pipe should be measured for
length and outer diameter. These measurements are consistent between pipes, unlike
Fig 5-14: Pipe Joints Showing Greater Diameter at Joint Compared with
Pipe Body
the total pipe lengths. The average diameter of a length of drill pipe or HWDP can then be
computed.
For example, conventional drill pipe has a nominal 5” OD. Allowing for the joints, this
effective diameter becomes 5.14”.
Detection of Influxes
As each stand of pipe is pulled from the hole, a volume of mud will be transferred from a
specialised low volume tank (called the trip tank) into the well to replace the pipe. The
purpose of this is to keep the hole full at all times, and hence keep the hydrostatic pressure at
maximum.
The volume of pipe removed should be compared with the volume of fluid transferred to the
well. If the hole remains full, but the volume transferred to replace the pipe is less than the
pipe volume, then the only conclusion is that fluid has entered the well from another location,
ie from the formations below the bit. The total of the transferred mud volume plus the influx
volume should be equal to the pipe volume. Thus the influx is detected and it's volume
measured.