SPE/IADC 130324

Understanding Multiphase Flow Modeling for N2 Concentric Nitrogen

Injection Through Downhole Pressure Sensor Data Measurements While
Drilling MPD Wells
Juan C. Beltran, SPE, Corrado Lupo, SPE, Leiro Medina, SPE, Hermogenes Duno, SPE, Fernando Gallo, SPE,
Jose Tang, SPE, and Jimmy Rojas, SPE, Schlumberger-Optimal Pressure Drilling; Efraín Rodríguez,
Schlumberger-IPM; Hernán Melgares, Schlumberger-D&M; and Jorge Bedoya, SPE, and Gustavo Puerto, SPE,
Blade Energy Partners

Copyright 2010, SPE/IADC Managed Pressure Drilling and Underbalanced Operations Conference and Exhibition

This paper was prepared for presentation at the 2010 SPE/IADC Managed Pressure Drilling and Underbalanced Operations Conference and Exhibition held in Kuala Lumpur, Malaysia, 24–25 February 2010.

This paper was selected for presentation by an SPE/IADC program committee following review of information contained in an abstract submitted by the author(s). Contents of the paper have not been reviewed
by the Society of Petroleum Engineers or the International Association of Drilling Contractors and are subject to correction by the author(s). The material does not necessarily reflect any position of the Society of
Petroleum Engineers or the International Association of Drilling Contractors, its officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the written consent of the
Society of Petroleum Engineers or the International Association of Drilling Contractors is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may not
be copied. The abstract must contain conspicuous acknowledgment of SPE/IADC copyright.

Abstract
Hydraulic modeling is a fundamental piece of any Managed Pressure Drilling operation using multiphase gasified drilling
fluids. MPD Engineers rely on hydraulic flow modeling systems to predict equivalent circulating densities. Models also allow
designing and manipulating hydraulic parameters such as gas/liquid ratios, pressures and flow rates to achieve desired
conditions.

Therefore, the selection and calibration of the correct hydraulic model is critical for the success of any MPD Operation. The
ECD calculation in an MPD operation is not the solely objective of using a complex modeling system. Today’s downhole
drilling tools technology makes available a variety of sensors capable of measuring actual downhole pressure values.
Prediction of flow behavior is also an important step that will increase the ability of monitoring and keeping efficient hole
cleaning, cutting transport and heat transfer efficiency, which are critical for all multiphase drilling operations. Actual
measurement of downhole equivalent circulating density becomes now a critical new calibration value to compare hydraulic
models performance and approximation to reality.

This paper compares the two-phase hydraulic simulations results with the data gathered from a drillstring installed annular
pressure sensor used while drilling a highly deviated well in a low pressure reservoir using nitrogen injection through
concentric string technique in an MPD operation. This technique poses a series of new challenges for the MPD engineer that
needs to predict hydraulic behavior created by the typical transient “U” tube effect caused by connections, trips and surveys
in this kind of applications.

The paper details the model and calibration process, findings and best practices gathered from multiple runs, real time
transmission and high definition memory data. Actual results and conclusions and also discussed and analyzed in depth for
the benefit of any further concentric job applications associated with use of downhole pressure sensors.

Introduction
In the mature and severely depleted fields, the concentric nitrogen injection technique has been successfully applied in three
MPD wells in order to avoid the use of nitrogen injection thru the drill pipe to reach low equivalent circulating densities and
avoid massive loss circulation and related hole cleaning problems.

Some years ago, a large amount of nitrogen injection rates through the drill string was a frequent picture in these specific
fields many times reaching the technical limit of a multiphase Managed Pressure Drilling application and getting the friction
dominated side of pressure curves. The presence of high nitrogen injection rates thru the drill pipe represents a severe
difficulty for measuring and logging while drilling pulse type downhole tools due the increase in fluid compressibility and a
consequential reduction in the LWD/MWD/VPWD pressure signal amplitude for pressure pulse type conventional directional
downhole systems, frustrating the possibility of achieving a full directional control in the reservoirs section. Bottom hole
2 SPE/IADC 130324

temperatures are also higher than conventional drilling because of the presence of high gas vs. liquid ratios, reaching very fast
the technology limits.

On this way, a directional drilling operation using multiphase MPD but with only liquid phase inside the drill pipe to preserve
directional and logging tools mud pulse and nitrogen injection into the concentric casing string was a mandatory solution not
only from the directional drilling point of view but also to maintain bottom holes circulating pressures within the operational
window required to avoid circulation losses, assure good hole cleaning and control risk of hole instability.

In July, 2008 after a long planning, technical and economical feasibility period, the first application of MPD using nitrogen
injection through concentric string was completed successfully. At that time, that well was constituted not only in first
drilled in the country with that concentric injection technique, but in the world-wide record with regard to the nitrogen
injection ports depth used to lighten the hydrostatic column.

In February, 2009, a second well was drilled in the same field with a great performance related to drilling times, well
construction and directional positioning. In both mentioned wells, the productivity obtained was higher than expected by the
client and for that reason and based on the operational success, the technique was selected also to be applied in other field,
located in different area.

In June, 2009 a high angle well was initiated, this well nowadays is related as the world-wide record as far as due deepest
well and deepest injection point for concentric string application. This well had the nitrogen injection ports located at 5,398
m MD, being reached a total drilled depth at 6,174 m MD with 76° of inclination (5,666 TVD). This well is the analyzed
case used for this paper.

One of the first steps in the MPD planning previous to the multiphase MPD simulation is to perform a complete analysis for
critical gas injection volume for concentric annular application. The critical gas injection volume is influenced by inclination
and depth of injection ports; concentric casing annulus volume and mud rate and as well as the density of liquid phase in the
annulus.

Once the critical gas injection rate analysis is done to assure that nitrogen rate required reaching the desired equivalent
circulating density (ECD) accomplishing the minimum nitrogen rate for pressure stability; a multiphase steady state software
is used to find out the operational window. It means the different mud and nitrogen rates that satisfied the desired bottom hole
pressure conditions, adequate hole cleaning and equipment constraints.

With the well trajectory, the wellbore geometry, the BHA and drillstring components and the liquid and gas fluid defined,
multiphase modeling can be done to determine the optimum fluid rates that would meet all of the design parameters. These
include:

• Desired degree of near-balanced across the reservoir section,

• The minimum hole cleaning requirements,
• The minimum and maximum motor limits are met,
• Minimize well slugging tendency and bottom hole pressure instability

However, to have confidence in a flow model during the planning of a MPD multiphase operation, it is important to have
examined previous cases and compared actual drilling data with the design stage calculations; however the best option to
validate any modeling is use the VPWD© -Vision Pressure While Drilling (Down Hole Annular and internal Pressure Sensor)
to adjust in real time.

Several commercial wellbore multiphase hydraulics simulators are available in the industry. The flow modeling software
used to conduct the two-phase simulations in the three concentric injection cases was Wellflo 7.300.2. This software has the
steady state module of OLGA (mechanistic two-phase flow model) and it is accepted as industry standard for multiphase
flow modeling in MPD/UBD applications. The predictions from simulator were compared during the execution of the jobs
with the recorded downhole pressure while drilling. In this paper, the Equivalent Circulating Density (ECD) is used for the
comparisons.

The data from VPWD© additionally, allow to a better confirmation and understanding related to pressure regimes that the
“U” tube follows during the drill pipe connections, survey and trip. The results confirmed some theories that were used in the
first well drilled but they also showed many lessons learned to be applied in future wells where concentric casing designs will
be used.
SPE/IADC 130324 3

Concentric Injection Technique, The solution for High Angle, Deeper and Depleted Reservoirs

The reservoir here in mention is conformed by fractured carbonates and dolomites from Lower Cretaceous and the common
productive zones are at 5,550-5,600 meters TVD range. Formation pressure has been decreasing over the last 25 years with
current values down to 0.4 g/cm3 SG.

The use of Managed Pressure Drilling technique along with multiphase fluids had been widely used in this region to reduce
loss circulation and differential sticking problems. N2 is reportedly injected directly into the standpipe, thus transforming the
usual light oil based mud in a multiphase system inside the drill pipe and over the annulus back to surface separation systems.
In the last 4-5 years the amount of N2 injected increased up to levels were the usual pulse MWD tools lost their signals
seriously affecting directional drilling capabilities. A secondary damaging effect was also created by the diminished heat
transfer capabilities of the drilling fluids unable to cool down downhole tools reaching their maximum temperatures ranges
causing catastrophic failures. Several attempts to solve the signal issue have been pursued by different service companies
with reasonable success, but leaving the temperature issue unsolved.

The use of electromagnetic directional tools is also limited by high bottom hole temperature of these reservoirs and formation
resistivity issues. Wired drill-pipe systems emerged as another good option, but also the tools would still be affected by
higher temperatures as a consequence of using multiphase systems.

This situation has restricted the implementation of high angle and horizontal drilling due to reduced possibility to control the
trajectory to reach the target in this kind of wells.

The nitrogen concentric injection technique was analyzed as a solution to make feasible the use of conventional mud pulse
MWD/LWD tools and keeping at the same time bottom holes circulating pressures within the operational window required to
avoid circulation losses, assure good hole cleaning and control risk of hole instability.

Well Configuration
In the concentric nitrogen injection technique, a temporary liner tieback is set on top of the intermediate liner and gas is
pumped through the injection ports from the liner tieback micro annulus to the tieback-drill pipe annulus as the well section is
drilled through the reservoir. Figure 1 shows the final well configuration for concentric casing nitrogen injection. When the
concentric annulus is used to inject the nitrogen, the well is virtually drilled as a conventional well with only liquid going
inside the drill pipe, hence eliminating the limitations of the conventional pulse telemetry to certain fraction of nitrogen
inside drill pipe.
4 SPE/IADC 130324

Figure 1. Well Configuration for Concentric String Nitrogen Injection

Fluid Selection
Once the well profile was defined, the second step was to define the fluids to be used. Taking into consideration that the pore
pressure is 0.4 g/cm3, the liquid selected as the drilling fluid fulfilled the following criteria:

• A compatible fluid with the formation that generates the minimum possible damage to the formation.
• The fluid had to avoid extreme adverse conditions in down hole as well as in surface like generation of stable
emulsions or rigid foam difficult to break.
• The fluid weight must facilitate to obtain the Equivalent Circulating Density required and it must keep the
designed weight during all the phase, to optimize the nitrogen volume needed.
• The fluid rheology must guarantee the hole cleaning without creating excessive frictional losses and facilitates
the separation process of the phases at surface.

Considering these aspects, the following drilling fluid system was chosen to drill the 6 ½” hole section of the well:

• Inverse emulsion (80/20)

• Density: 0.92 g/cm3
• Plastic viscosity: 8 to 9 cp
• Yield Point: 5 to 8 lbs/100 in2
• Percentage of solids: 2-4 %
• API Filtrate (ml) 30 min: <6
• Salinity > 180000

The gas phase selected was on-site generated nitrogen by membrane units which normally has a purity varying between 95%
to 99% and 0.97 SG (air = 1 SG). This is the gas source more commonly used in Underbalanced Drilling and as well as
Managed Pressure Drilling Operations due to the high required volumes, by logistic facility and because it has a lower cost
by unit in comparison with cryogenic nitrogen. Additionally, the use of nitrogen as lifting gas, minimizes the associated risks
to down hole fire and explosions and as well as the corrosion process by its inert character. Also, membrane nitrogen is
environmentally friendly does not produce dangerous waste.
SPE/IADC 130324 5

The two-phase simulation indicated that to reach the values of ECD proposed with the base fluid selected, around 140-150
m3/min of nitrogen were required. In order to obtain and to handle these volumes of nitrogen at the estimated injection
pressure, several packages of membrane nitrogen generation need to be installed.

Drill String and BHA

The drill pipe used was 3 ½” OD, S degree of 15.5# and combined with S, G and X of 13.3#.
The BHA used consisted basically of the following elements:

• Downhole Motor (the first BHA use a Power Drive)

• Signal Booster MWD
• Monel
• VPWD© (ran with the BHA #1, 3 and 5)
• HWDP
• Drill Collars
• Drilling Jar
• Drill Collars

Only two PDC bits were used to drill the 5 7/8” hole section.

Different combinations of nozzles and signal to booster were used in order to have the configuration that generated the best
signal of MWD at surface.

Although nitrogen injection through the drill pipe was not considered, always a wireline retrievable NRV was installed when
the bit reached the 9⅝” casing shoe. The real purpose of this valve was helps to reduce the times related to pressure drainage
in the drill string during connections. Additionally, it serves as barrier in case of drillstring washout below the depth of its
installation.

Once again it was possible to confirm that IMPulse is the best MWD tool available until the moment to work to the extreme
suction conditions of flow inside the drill string by action of nitrogen injection in the annulus. The answer in signal
transmission was reached inclusive with values from 30 to 50 psi inside drill pipe. The Kit initially designed for the Power
Drive showed not to be adapted for the conditions of low liquid flow necessary to reach lower ECD in this type of well. With
the obtained data from the MWD tool, an analysis was made in order to redesign the kit for and extend rank of work with low
liquid flow.

Methodology
Multiphase Flow Modeling
Since most operations in low pressure reservoir involve the use of a gas-liquid mixture in place of conventional drilling mud,
the results from multiphase flow model is a key in the planning stage, where is important indentify a target bottom hole
pressure and assess how to reach this target. The application of multiphase flow modeling to UBD/MPD has been detailed by
previous authors (e.g. C. Mykytiw. P.V. Suryanarayana1) and includes the following items. However this paper only will
focus in monitoring and determine changes in the ECD planned vs. real.

• Determining the feasibility of reach target pressure

• Optimizing injection rate and pressure
• Establishing operation parameters
• Ensuring adequate hole cleaning and motor performance
• Identifying fluids required to achieve the target pressure
• Identifying the technical specifications of equipment
• Defining the operating envelope for the Operation
• Monitoring drilling to determine changes to operation parameters

Data and Assumption

Looking at the proposed versus actual data, the items that most affect the accuracy of the model include:

• Wellhead Pressure
• True Vertical Depth
• Drill pipe, casing, and open hole size
• Fluid properties (liquid density, viscosity, etc)
• Pump rates
• Reservoir inflow rates
6 SPE/IADC 130324

• Measured depth
• N2 injection Depth.
• Bottom hole circulating temperature

The data and assumptions for the planning phase were obtained from the original drilling program and source material
provided for the well and offset wells. Where enough data are not available, assumptions were made based on experience and
accepted practices.

Table 1 shows the comparison between proposed and as-drilled TVD, liquid density, well head pressure, Injection point
depth. The operation was designed to be drilled with the minimum surface backpressure possible. The best approximation
for a steady-state well head pressure was chosen from the record data, but fluctuations in the wellhead pressure are always
expected while drilling with multiphase flow.

Planned Real
5400 m MD 5386m MD
7 5/8" Tie-Back
5227 m TVD 5222m TVD
Injection Point
42 ° 47 °
5755 m MD 5830m MD
7" Liner Setting
5468 m TVD 5513m TVD
Point
49 ° 47 °
6331 m MD 6174m MD
Total Depth 5683 m TVD 5666m TVD
77 ° 76 °

Mud Weight 0.92 SG 0.92 SG

Well Head
60 psi > 120 psi
Pressure

Table 1. Planned Data vs. Real Data

The true vertical depth for the case was extremely close to the actual as-drilled case. The largest variation from the proposed
to the actual data was about 17 m (56 ft). Also, the liquid phase density used was stable in 0,92 SG (7,66 ppg) in almost all
operation.
For roughness, all of the design stage and onsite modeling was performed using the Neotec defaults for pipe roughness
(whether in new, or used condition), and the default value for rough concrete was applied in the open hole section.

Other differences in the proposed, to the actual on site modeling, including things such as the length of BHA, wellhead
temperature, build rate, depth of injection point etc, appear to have a very small effect on the prediction of BHCP, for the
case analyzed.

Operational window Proposed

For the determination of the operational window was necessary to make a multiphase simulation that satisfied the limits
imposed by the downhole motor, the limit required for a suitable hole cleaning, to maintain the flow below the limits of the
MPD surface equipment and the bottomhole pressures over the limit of well stability, but within a degree of overbalanced
minimum in order to diminishing the losses of circulation.

A BHCP between 0.4 and 0.5 g/cc was settled down. The simulations made during the technical Feasibility phase indicated
that minimum ECD attainable would be of 0.43 g/cc, using a mud weight of 0.88 g/cc. However, during the front end
engineering phase was determined that the weight fluid to use during the implementation would be an inverse emulsion of
0.92 g/cc, which limited the minimum ECD attainable of 0,45 g/cc. This represented a degree of overbalanced of 0.1 to 0.15
g/cc with respect to the equivalent pore pressure reported by the client of 0.35 g/cc (185 kg/cm2).

Different combinations from liquid fluids volume by drillstring with gas injection flows by concentric annulus and as well as
by drillstring were analyzed with the intention to find the ideal relation to fulfill the limitations above mentioned but that at
the same time generated stable conditions of circulation. For the analysis it was considered that the total capability available
for gas generation onsite was 150 m3/min of membrane nitrogen.

From the multiphase simulations considering injection via drillstring, it was encountered that the system was near to the zone
where the bottomhole pressure is dominated by the friction and additional increases of nitrogen injection will not have a
SPE/IADC 130324 7

positive effect in the reduction of the ECD. Therefore it was decided that in case of requiring nitrogen injection by the
drillstring, it will be not to reduce the ECD but to diminish the effect of “U” tube between the fluid inside the drillstring and
the main annular.

The maximum nitrogen volume to pump by drillstring depends on the mud volume used, but for all the situations it never
could exceed 15 m3/min. The Figure 2 shows the operational window proposed for the 5 ⅞” hole section and all the Hole
cleaning criteria used.
Planned Multiphase MPD Operating Window – Bit @ 6325 m (TD)
Annular Injection: 7 5/8” x 9 5/8” Csg & 7” Liner – 5 7/8” Hole Section (3 ½” DP)
150 GPM + 0 m3/min 160 GPM + 0 m3/min 190 GPM + 0 m3/min
210 GPM + 0 m3/min Formation Pressure (0.33 gr/cc) = 2660 psia Min BHCP Proposed (0.42 gr/cc) = 3390 psia
Max BHCP Proposed (0.5 gr/cc) = 4040 psia Min Liquid Vel. 55 m.min Min Liquid vel. 70 m/min
Citric Volume of N 2 Gas 130% Citric Volume of N 2 Gas 146% Citric Volume of N 2 Gas

N2 Critic Vol. 30% of Additional N2 Critic Vol. 46% of Additional N2 Critic Vol. Min Vel. = 70 m/min

FRICTION DOMINATED
Min Vel. = 55 m/min

Injection Ports @ 5440 m

Mud Weigh = 0.92 gr/cc
WHP = 60 psia

N2 Gas Injection Rate (m3/min)

Figure 2. Planned Operational Window
For the accomplishment of the operative window, hydraulic simulations were performed using the following premises:
• Fluid System Type: Simplified
• Case Type: Drilling
• Gas Viscosity: Lee, Gonzales and Eakin
• Dead Oil Viscosity Equation: General Equation
• Dead Oil Viscosity Correlation: Beal
• Drilling Profile Type: 3 Dimensional (Min Curvature)

The parameters that determined the operative window (shaded area of graph in Figure 2) were established by the liquid
velocity, pressure gradients and technical considerations of downhole tools.

Simulations of Transient State

To generate a more robust and reliable design, a simulator of transient state was used to predict the behavior of the flow and
the pressures and consequently to help in the selection of the surface equipment adequate for the MPD operation. These
simulations were used to confirm the operational window and the calculation of the critical gas injection volume.

Basically, between the things determined from these transient simulations were to know which would be the maximum
nitrogen injection pressure at surface through concentric micro annular when the initial displacement of mud will be made.
This would allow in the determination of the minimum capability for the surface compression equipment. Also they allowed
to know the minimum flow volume of returns generated by the displacement and to determine as well if the specifications of
the separation equipment were adapted to flow expected. Finally, the results served to confirm the calculation of the critical
volume of gas necessary to reach a concentric annular injection pressure stable and to have an idea for the stabilization time
of the circulation if the displacement became just by nitrogen.

Although it is not presented in this paper, also were performed calculations to know the influence of nitrogen injection
volume by the drillpipe. The result showed that the value of pressure in the drillstring was increased, that is to say that there
is a smaller effect of suction due to the minor hydrostatic not balanced between the fluid column inside the drillstring and the
mixture of fluids in main annulus.
8 SPE/IADC 130324

Results
The VPWD© tool allow analyzed several phenomena present when is drilled with nitrogen injection directly in the annulus.
Due VPWD© constantly is recording the BCHP (ECD) and the pressure inside the drill pipe, many transients situations were
recorded in memory and in real time for analysis.

Due drilling process there is always transient situations; only drilling situations were compared in pseudo steady stable
according to the steady stable planning, and describing the “U” tube effect present in connections and surveys and their
impact in the Bottom Hole Pressure.

Comparative Analysis of ECD Simulated Versus ECD Measured by the Bottomhole Sensor VPWD©
Figure 3 shows the real behavior of the ECD measured by the bottomhole sensor VPWD©, installed in the BHA #5 at 26 m
of the bit. This graph shows the measurement from 5916m until the total depth drilled at 6174m.

An ECD around 0.6 SG can be seen in the left part of the graph, which is the value that was had when the BHCP was still
controlled due to hole instability uncertainty. Later it is possible to appreciate how the ECD is reduced, remaining during all
the drilling phase within the planned operational window from 0.4 SG to 0.5 SG, but with sporadic values slightly above of
the superior limit.

In the graph, two zones with strong tendency can be considered towards the inferior limit of the operational window. In both
zones the bit was pulled from the bottom and positioned into the cased hole by safety issues, while other operations were
made. In the first case it had to be performed due to the mandatory necessity of change out the washpipe and in the second
case to change out the Rotating Head stripper rubbers, which reached the limit of their life utility.

ECD (VPWD) & Bit Depth

ECD Measured with VPWD (g/cc)

Final Zone of
TD @
Controlled ECD
6174 m

Bit Depth (m)

Stand
Reamed

5916 m
@ 6013 m Trip to @ 6128 m Trip to
Casing Shoe Due Casing Shoe Due to
to TD Wash pipe RFH Stripper Rubber
Change out Change out

ECD Measured VPWD (g/cc) Bit Depth (m)

Figure 3. ECD Measured by Bottomhole Sensor VPWD©

The graph was made with the stored information recovered from the memory of the VPWD©, for that reason, the ECD data of
the two zones already mentioned need a small adjustment by the depth of the bit.

The picks that leave from the normal tendency of the bit depth curve correspond to the reaming performed each stand drilled.

Perhaps the variable that more uncertainty generates in the multiphase simulations is the equivalent circulation density (ECD)
and it is understandable by the endless number of factors that affect it, by the same complexity of the mathematical
calculations and by the selection of the used correlations for the calculations.
SPE/IADC 130324 9

The equivalent circulation density by definition is consequence of the hydrostatic pressure, of the fictional pressure losses in
the annulus generated by the circulation of the fluids and by the exerted back pressure at surface. In very simple terms, the
hydrostatic component is the adding up of the same weight of the fluids in the annulus, in this case mud and nitrogen and the
weight of drilling cuttings, whereas the frictional losses are consequence of the annular geometry or flow area, flow volume
of fluids and their rheology properties. On the other hand, the back pressure is consequence of the closing given to choke
manifold at surface.

Finally, the effect of temperature again used inverse emulsion or nitrogen cannot be forgotten, therefore is comprehensible
that exists a natural uncertainty in the mathematical results delivered by the software. In order to fit the simulations, a
bottomhole sensor VPWD© was used and the obtained data allowed to adjust the configuration of the calculation, obtaining
with this adjustment results with a deviation smaller than 2% in most of the cases. For the comparative analysis, always
conditions of stabilized flow were expected for being the software used in field a simulator of stable state. Figure 4 shows
ECD VPWD© vs. Wellflo simulated ECD.

ECD Simulated CARDENAS

(Wellflo) Vs. ECD 839Measured (VPWD)
ECD Simulado (Wellflo) vs ECD Medido (VPWD)

1000
0.60
1500

0.55
2000

0.50 2500
ECD (g/cc)

BHP (psi)
ECD (g/cm^3)

BHP (psi)
3000
0.45

3500
0.40
4000

0.35
4500

0.30 5000
5906 5906 5907 5908 5910 5911 5914 5916 5923 5954 5981 6014 6072

Depth (m)
Profundidad (m)

ECD Wellflo ECD VPWD BHP Wellflo BHP VPWD

Figure 4. ECD VPWD© vs. Wellflo Simulated ECD

The simulator runs were configured during the planned phase having the following premises:
• Input of the detailed well geometry, where all casings, liners and tubular section were included.
• The fluid system was characterized as inverse emulsion, specifying the density of the oil to standard conditions
and modified density.
• The model used was OLGAS 1992.
• The temperature profile was calculated with the initial circulation parameters.

Initiated the MPD operation the following adjustments at the field were made in the simulation:
• The modified density of mud was varied and adjusted manually in the simulator in order to meet the real density
of the mud at entrance of the system and calibrating the same value of mud density at the returns.
• The real temperature profile provided by VPWD© tool was used to re-input this data to the simulator.
• In the BHA section, the real pressure losses for the downhole tools at different flow volumes were specified.
• The outflow points were specific as circulation losses points once they appeared.

These adjustments allowed to obtain the ECD values show in the Figure 4, where the veracity of such is determined by the
VPWD© tool readings.
10 SPE/IADC 130324

Connection- Survey Strategy

Because the well was designed to be drilled in near balance condition, the bottom hole pressure should be stable in the
connections to avoid any influx. Additionally due the “U” effect caused by the difference between annular pressure and inner
pressure in the connection, the time to the connection had to be optimized to reduce its impact in the bottom hole pressure.

During the connection, simultaneously while pumping off and stop nitrogen injection, the well is closed until reach a
maximum well head pressure value, similar to the loss friction pressure predicted by the model to this depth. Once the mud
and nitrogen pumping are reestablished the WHP is reduce until the previous value in order to keep a BHP constant
according to the simulation. However, as per is represented in the Figure 5, some increases in the ECD are obtained due
mainly to the “U” tube effect present in this kind of concentric injection operation, where extra fluid coming from drill string
is moved to the annulus when the mud and nitrogen pumping are stopped.

The connections, unlike an operation of two-phase fluids where the nitrogen is injected by the drillstring and the standpipe
pressure must bleed-off, for this concentric application were made in conventional way. Once reached the connection depth,
the complete stand is reamed in order to eliminate any torque or drags abnormal and the hole circulated to confirm the clean
returns.

0.7
Connection 4532m
Connection @ 5954m
500
WHP - 550 psi 0.6
WHP (psi) - Mud Rate (gpm) - N2 Rate (m3/min)

400 0.5

ECD = 0,5 SG
0.4
300

ECD (sg)
0.3
200

0.2

100
0.1

0 0
19:12:00 19:26:24 19:40:48 19:55:12 20:09:36 20:24:00 20:38:24 20:52:48 21:07:12 21:21:36 21:36:00
N2 Rate (m3/min) Mud Rate (gpm) WHP (psi) ECD (sg)

Figure 5. ECD – WHP – Mud – Nitrogen Rate - Connection at 5954 m

While the mud pumps are stopped, the nitrogen injection is kept in the micro annulus and at the same time to avoid charge
the annular space, the MPD choke is closed slowly, monitoring all the time the pressure of concentric annulus. If the annular
pressure increases much, the MPD choke operator proceeds to open the choke in order to reduce the returns pressure and the
concentric one. Done this, the string is set in the rotary slips and the connection is performed in conventional way, trying to
do it as rapidly as possible. Once connected the new stand and raised the rotary slips, the MPD supervisor will give
instructions to the driller so the liquid injection being resumed at high pump strokes around 360 gpm, in order to accelerate
the filling of the empty space in the drill pipe created by the time spent during the connection (“U” tube effect). Once
confirmed by the drillpipe pressure sensors that the string is full, the pump strokes are adjusted immediately to around 170
gpm. The MPD supervisor orders the opening gradual of the MPD choke preventing an abrupt fall in the annular and
concentric space. Subsequently the directional engineer has to confirm good communication with the tool, in order to proceed
with the survey taking. Confirmed the survey and having the established parameters of mud pumping, continuous nitrogen
injection, and stabilized variables of pressure and returns at surface the drilling operation is resumed.

Conclusions

• MWD/LWD/ VPWD© information was received all the time at surface allowing keeping directional control and adjusted
the multiphase simulator runs and calibrations.
SPE/IADC 130324 11

• Detailed design, modeling and simulations are keys for success in the application of concentric casing injection
techniques.

• The use of VPWD© tool in this well allows real-time downhole reading like internal and annular pressures, on this way it
was possible to perform a comparison with the results delivered by the simulator and adjust and calibrate the model to
the real well conditions.

• Circulation and bottom hole pressures stability are very affected by connection time that influence gas and liquid phases
segregation.

• It is recommended to continue investigating the best option to obtain a greater integrity of the MPD/LWD/ VPWD©
signal for low liquid injection volumes in order to be able to reduce more the equivalent circulation density (ECD). It is
possible to be thinking about the use of a smaller internal diameter drillpipe, a different combination of drillstring signal
boosters or inclusive the use of small nitrogen injection volume through the drillstring, keeping the percentage of gas
into the drillstring as low as possible and without exceed the limits which can debilitated or attenuated the MWD/LWD/
VPWD© signal.

Acknowledgements
The authors thank Schlumberger IPM Villahermosa, Optimal Pressure Drilling and Blade Energy Partners for allowing the
publication of this work.

References
Mykytiw, C.G., Davidson, I.A., Frink, P.J.: “Design and Operational Considerations to Maintain Underbalanced Conditions
with Concentric Casing Injection”, SPE 81631, Presented at the IADC/SPE Underbalanced Technology Conference and
Exhibition held in Houston, Texas, U.S.A. March 2003.

C. Perez-Tellez, A. Urbieta, C. Lupo, J.M. Castellanos, O. Ramirez, G. Puerto, J. Bedoya, O. Gabaldon, J. Beltran, G.
Castiblanco. “MPD Concentric Nitrogen Injection Used to Drill a Successful Horizontal Well in Fractured and Depleted
Mature Reservoir in Mexico South Region”. SPE 122982, Presented at the 2009 SPE Latin American and Caribbean
Petroleum Engineering Conference held in Cartagena, Colombia, 31 May–3 June 2009.

J. Alfredo Rios J., Luis Angel Ayala G., R. Rayon Casiano. Perforación y Mantenimiento de Pozos, División Sur, PEMEX.
Successful Drilling in Highly Depleted Formations, Villahermosa, Mexico, 2002.

IADC well classification system for underbalanced operations and Managed pressure drilling, March 2005.

Urbieta, A., Perez Tellez, C., Lupo, C., Castellanos, J.M., Ramirez, O., Puerto, G., Bedoya, J. “First Application of
concentric nitrogen injection technique for a managed pressure drilling depleted well in southern Mexico”, SPE 122198,
Presented at the IADC/SPE MPD & UBD Conference and Exhibition held in San Antonio, Texas, U.S.A. Feb. 2009.

Nomenclature / Acronyms

MPD = Managed Pressure Drilling

ECD = Equivalent Circulating Density
LWD = Logging While Drilling
MWD = Meassuring While Drilling
VPWD = Vision Pressure While Drilling
TVD = True Vertical Depth
MD = Messured Depth
UBD = Underbalanced Drilling
BHA = Bottom Hole Assembly
HWDP = Heavy Weight Drillpipe
BHCP = Bottom Hole Circulating Pressure
BHP = Bottom Hole Pressure
SG = Specific Gravity
ppg = Pound Per Galon
WHP = Wellhead Pressure
PDC = Polycrystalline Diamond Compact
12 SPE/IADC 130324

Appendix A

Pressure in Concentric Annulus

Nitrogen Hydrostatic Pressure
Drilling Fluid Hydrostatic Pressure
In the concentric Annulus
Drilling Fluid Hydrostatic Pressure
In the Main Annulus

Injection Ports Pressure = P4 = P1 + P2+ P3

Figure 6. Hydrostatic Pressure in the Main Annulus During a Concentric Casing Injection Operation