Crespo (2013) PDF
Crespo (2013) PDF
Crespo (2013) PDF
A simplified surge and swab pressure model for yield power law fluids
Freddy Crespo 1, Ramadan Ahmed n
Mewbourne School of Petroleum and Geological Engineering, University of Oklahoma, OK, USA
a r t i c l e i n f o a b s t r a c t
Article history: Surge and swab pressures can be generated in different stages of well construction including tripping
Received 15 January 2012 operations and reciprocation of drillstring in the wellbore. Significant surge and swab pressures can
Accepted 28 October 2012 lead to a number of costly drilling problems such as lost circulation, formation fracture, fluid influx,
Available online 10 November 2012
kicks, and blowouts. This phenomenon is of economic importance for the oil industry.
Keywords: Theoretical and field studies indicate that pressure surges strongly depend on drillpipe tripping
Surge and swab pressures speeds, wellbore geometry, flow regime, fluid rheology, and whether the pipe is open or closed.
Drilling Although a large number of studies were conducted in the past to investigate surge and swab pressures,
Tripping experiments under controlled laboratory conditions have never been reported. This paper presents
Fluid displacement
results of an experimental study aimed at investigating the effects of pipe speed (i.e., tripping speed),
Well-bore hydraulics
fluid properties and borehole geometry on surge and swab pressures under laboratory conditions.
Other phenomena such as fluid gelling and pipe eccentricity effects were also examined. Experiments
were performed in a test setup that has the capability of varying the tripping speed and accurately
measure the surge or swab pressures. The setup consists of fully transparent polycarbonate tubing and
inner steel pipe, which moves axially using a speed-controlled hoisting system. Experiments were
conducted using mineral oil and polymeric fluids.
A new regression model has been developed to calculate surge and swab pressures under steady-
state flow conditions. The model is based on the results of approximate numerical solutions obtained
by considering the annulus as a narrow slot. Model predictions were compared with experimental
measurements and predictions of existing models. A satisfactory agreement has been obtained.
Experimental results and model predictions confirm that the trip speed, fluid rheology, annular
clearance and pipe eccentricity significantly affect the surge and swab pressures.
& 2012 Elsevier B.V. All rights reserved.
1. Introduction A number of field studies (Clark, 1956; Moore, 1965; Clark and
Fontenot, 1974; Lal, 1983; Wagner et al., 1993) were undertaken to
Recently, surge and swab pressure modeling has received investigate the effects of fluid properties and wellbore geometry on
increased attention as deepwater drilling and new technologies surge and swab pressures. These studies demonstrated that surge
such as slim-hole and casing drilling techniques have emerged in and swab pressures increase with tripping speed and wellbore depth.
the industry. As thousands of wells are drilled every year applying In addition, it was reported that surge and swab pressures with
these new techniques, challenges associated with downhole pres- closed-ended pipe are higher than those with open-ended pipe. More
sure management have become more important. Surge and swab recent studies (Bing et al., 1995; White et al., 1997; Ward and
pressures are usually generated during many phases of well Andreassen, 1998; Rudolf and Suryanarayana, 1998; Isambourg et al.,
construction such as tripping, running casing and cementing opera- 1999; Samuel et al., 2003; Rommetveit et al., 2005) using highly
tions. Surge pressure becomes more problematic when excessive accurate downhole tools re-affirmed the strong impact of tripping
tripping speeds and narrow and constricted annular clearance speeds, wellbore geometry, flow regime, fluid rheology on the surge
situations are involved in wells. This condition is more exacerbated and swab pressures.
by the fact that most of deepwater wells have very narrow margin Downhole pressure-recording tools are limited by temperature
between pore and fracture pressure gradients. and capacity to transmit data while making connections. The drilling
industry greatly relies on surge/swab models and simulators. Com-
monly used steady-state models have been developed by Burkhardt
(1961), Schuh (1964), and Fontenot and Clark (1974).
n
Corresponding author. Tel.: þ1 405 325 0745; fax: þ1 405 325 7477.
Hydraulic analysis of concentric annular flow with axial motion
E-mail address: [email protected] (R. Ahmed). of the inner pipe is used to model surge and swab pressures (Lin and
1
Now with Halliburton Energy Services Hsu, 1980; Chukwu and Blick, 1989; Malik and Shenoy, 1991;
0920-4105/$ - see front matter & 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.petrol.2012.10.001
F. Crespo, R. Ahmed / Journal of Petroleum Science and Engineering 101 (2013) 12–20 13
Nomenclature e
x Dimensionless x-coordinate
y y-coordinate
A Geometric parameter e
y Dimensionless y-coordinate
b Constant y1 Lower limit of Region II
B Geometric parameter y2 Upper limit of Region II
dh Hole/casing diameter e1
y Dimensionless lower boundary limit of Region II
dpo Pipe outer diameter e2
y Dimensionless upper boundary limit of Region II
dR Diameter of the guiding rod
f Friction factor Greek Letters
H Slot thickness
k Consistency index p1 Dimensionless pressure
K Diameter ratio p2 Dimensionless plug thickness
L Pipe length in the wellbore r Fluid density
n Fluid behavior index t1 Shear stress profile in Region I
qa Flow rate in the annulus t3 Shear stress profile in Region III
ea
q Dimensionless annular flow rate tw Shear stress acting on the borehole wall
Re Generalized Reynolds number t0 Yield stress
Ren Modified Reynolds number DPS Surge or swab pressure
v Local fluid velocity dv/dy Shear rate
v1 Local fluid velocity in Region I
v3 Local fluid velocity in Region III Acronyms
e1
V Dimensionless velocity profile in Region I
e2
V Dimensionless velocity profile in Region II
e3 PAC Polyanionic cellulose
V Dimensionless velocity profile in Region III
XG Xanthan gum
Vp Pipe velocity (tripping speed)
YPL Yield power law
W Slot width
YS Yield stress
x x-coordinate
Haige and Xisheng, 1996; Filip and David, 2003). In order to obtain a applying the narrow-slot approximation technique. As reported by a
simple analytical solution, a number of previous studies (Schlichting, number of studies (Guillot and Dennis, 1988; Chukwu and Blick,
1955; Wadhwa, 1966; Flumerfelt et al., 1969; Chukwu and Blick, 1989; Guillot, 1990; Kelessidis et al., 2007; Bourgoyne et al., 1986),
1989) applied the narrow slot method. This method approximates this approach is valid for diameter ratios greater than 0.3.
the annular flow as a slot flow between two parallel plates, one Based on the narrow-slot approach, a theoretical surge pressure
stationary and the other moving at a constant speed. model presented in Appendix A has been formulated for yield power
Although a large number of field and modeling studies were law fluids. The model is developed assuming: (i) incompressible
conducted in the past to investigate surge and swab pressures, fluid; (ii) laminar, steady state and isothermal flow conditions; (iii)
experiments under controlled laboratory conditions have never been constant pipe velocity; (iv) close-ended drill pipe; (v) no-slip
reported. In addition, surge and swab pressure models have been conditions at the walls; and (vi) concentric annulus. The theoretical
limited to Newtonian, power law and Bingham plastic fluids. Often model requires cumbersome numerical procedure to obtain the
the yield-power-law (Herschel–Bulkley) model best fits the rheolo- solution. To develop a simple correlation, extensive parametric
gical properties of most of drilling fluids and aqueous clay slurries study was carried out using the model. Numerical solutions were
(Fordham et al., 1991; Hemphil et al., 1993; Merlo et al., 1995; obtained varying different fluid and geometric parameters. Subse-
Maglione and Ferrario, 1996; Kelessidis et al., 2005, 2007). This paper quently, a simple regression model (correlation) has been developed
presents a new model, which is applicable for yield-power-law fluid. by analysis the numerical results. The correlation predicts surge or
The performance of model has been rigorously tested by direct swab pressure conveniently for yield-power-law fluids without
comparison with experimental data and other existing models. requiring an iterative procedure. Predictions can be made for New-
tonian (t0 ¼0; n¼1), Power-law (t0 ¼ 0) and Bingham plastic (n¼1)
fluids as well. Accordingly, the surge or swab pressure (DPS) is
2. Modeling determined using friction factor as:
The annular fluid flow induced by axial motion of the inner pipe DPS 2rV 2p f
¼ ð1Þ
can be mathematically modeled to predict surge and swab pressures L H
where r and L are fluid density and pipe length in the wellbore. As
Motor
shown in Fig. 1, Vp and H are trip speed and annular clearance, Polymer Water
respectively. The relationship between the friction factor and
generalized Reynolds number is established methodically to resem-
ble the pipe flow equation. Hence, the friction factor is expressed Cable Guide
as: f¼16/Re. The expression for generalized Reynolds number is Motor
Controller
given as:
Mixing Tank
16Ren
Re ¼ ð2Þ
2:83 þ 3:88 1e2:62F
where Ren is the modified Reynolds number. The yield stress factor,
F is a dimensionless parameter, which is greater than one for any
fluid that exhibits yield stress. This parameter is defined as:
n Pressure
H H t0 Transducer
F¼ þ1 ð3Þ Transparent
dh Vp k
Polycarbonate ΔP
Tube
The modified Reynolds is expressed as: 30 cm
Drillpipe
n
rV 2p
Re ¼ n n ð4Þ
t0 þk A þB=n Vp Guiding Rod
100
PAC 1.00% PAC 0.75% PAC 0.56% Mix PAC 0.28% + Xantan Gum 0.44% b Xantan Gum 1.00%
Xantan Gum 0.67%
100
80
10
60
40
20
0 1
0 200 400 600 800 1000 1 10 100 1000
Shear Rate (1/s) Shear Rate (1/s)
3.3. Test materials Heavy mineral oil Newtonian 19 0.000 0.204 1.00
Light mineral oil Newtonian 20 0.000 0.024 1.00
Experiments were performed using Newtonian fluids (high and 1.00% PAC Power law 20 0.000 2.260 0.57
low viscosity mineral oils) and non-Newtonian polymeric fluids, 0.75% PAC Power-law 20 0.000 0.670 0.67
0.56% PAC Power-law 20 0.000 0.170 0.80
which are aqueous suspensions of polyanionic cellulose (PAC) and 0.28% PACþ 0.22% Power-law 20 0.000 0.774 0.50
xanthan gum (XG). Fig. 3 presents flow behavior of the polymeric xanthan gum
fluids. Rheological parameters of all the test fluids are presented in 1.0% xanthan gum YPL 20 18.6 0.767 0.50
Table 1. Polymer based fluids show considerable shear thinning. 0.67% xanthan gum YPL 20 7.80 0.553 0.50
0.44% xanthan gum YPL 20 3.44 0.359 0.52
Three different concentrations of PAC (1.00%, 0.75% and 0.56% by
weight) and xanthan gum fluids (1.00%, 0.67% and 0.44% by weight)
were tested. In addition, a polymer mix of 0.28% PAC and 0.22% XG
by weight was considered in the study. The solid lines in Fig. 3a which is represented by the straight line (f¼16/Re). The strong
represent power-law model rheology curves obtained using the agreement between the regression model line and experimental
fluid rheological parameters ‘‘n’’ and ‘‘K’’ calculated from viscometric data points verify the accuracy of the model. In addition, the
measurements. agreement indicates that the prevalence of laminar flow condi-
tions during the experiments. Because turbulent flow data points
3.4. Data processing are expected to deviate from the regression model (i.e., laminar
flow) line as regularly observed during the onset of turbulent in
Fig. 4 shows measured surge pressure as a function of time when pipe flow or annular flows.
the pipe moved downward. As the pipe begins to move downward, For horizontal and inclined wells, eccentricity of the drillpipe
first the fluid particles accelerate and the pressure loss abruptly and thickness of the cuttings bed need to be considered in the
increases with time for a short period. Then, the pressure stabilizes analysis to optimize the trip speed. Eccentricity has been recog-
becoming approximately constant. Average pressure reading under nized to have significant effect on annular pressure losses (Singh
steady state condition was determined for each tripping speed. High and Samuel, 2009; Sorgun and Ozbayoglu, 2010). During the
speed experiments require a longer stroke length to reach steady experimental investigation, it was observed that when the inner
state flow conditions. As a result, the maximum trip speed during pipe was eccentric, surge pressures measurements were reduced
the investigation was limited to 0.21 m/s. For tripping speed ranging as much as 42% percent compared with a fully concentric test
from 0.03 m/s to 0.21 m/s, stabilized steady-state pressure reading (Fig. 6). Adequate modeling of eccentricity effects on the surge
were obtained (Fig. 4a). Stead state condition needs enough time to and swab pressures enable higher trip speed, which can reduce
establish. When the trip speed increases a constant stroke length, non-productive time and drilling cost.
the trip time reduces. As a result, at high tripping speeds (i.e., greater The effect of gelling (static time) on surge pressure were also
than 0.21 m/s), it was not possible to reach the state flow conditions examined. The 1.0% XG based fluid exhibited very high yield
(Fig. 4b). stress. This fluid was chosen for the gelling investigation. Two
identical tests were conducted at 0.015 m/s trip speed after
3.5. Experimental results allowing the fluid to stay under quiescent condition for different
time durations (15 s and 6 min). Approximately same surge
Experimentally measurements obtained from all the tests are pressure values were measured in both cases (Fig. 7) indicating
presented in Fig. 5 as the friction versus generalized Reynolds minimal gelling effect on surge pressure with XG based fluids. In
number. Results are highly correlated with the regression model, general, a highly gelled drilling fluid can create significant swab
16 F. Crespo, R. Ahmed / Journal of Petroleum Science and Engineering 101 (2013) 12–20
2.0 2.5
Readings Readings
Average Average
2.0
1.5 Unsteady-State
Surge Pressure (kPa)
1.0
1.0
0.5
0.5
0.0 0.0
0 20 40 60 80 100 10 20 30 40 50 60
Time (s) Time (s)
Fig. 4. Measured surge pressure vs. time: (a) 1% PAC (k¼2.26 Pa sn and n¼ 0.57) and VP ¼ 0.061 m/s and (b) Mineral Oil (m ¼204 cp) and VP ¼ 0.21 m/s.
100.00 7.0
Test After 15 s
6.0
4.0
1.00 Light Mineral Oil
Mineral Oil 3.0
1.0% PAC
0.75% PAC
0.56% PAC
2.0
0.10 Mix 0.56% PAC + 0.44% XG
1.0% Xantan Gum 1.0
0.64% Xantan Gum Yield Stress Effect
0.22 Xantan Gum
f = 16 / Re 0.0
0.01 0 2 4 6 8 10 12
0 1 10 100 1000
Time (min)
Generalized Reynolds Number (Re)
Fig. 7. Effect of static time on surge pressure measurements (1.0% Xanthan Gum,
Fig. 5. Friction factor vs. generalized Reynolds number. t0 ¼ 18.6 Pa, k¼0.767 Pa sn, n¼ 0.5, VP ¼ 0.015 m/s).
2.0
the testing of fluids without yield stress. Due to the yield stress,
Concentric Annulus after the pipe movement, the XG based fluids showed pressure
1.5 difference even under static condition.
Surge Pressure (kPa)
8 1.00
Measurements
Measurements
4 0.50
2 0.25
0 0.00
0.00 0.04 0.08 0.12 0.16 0.00 0.05 0.10 0.15 0.20
Pipe Velocity (m/s) Pipe Velocity (m/s)
Fig. 8. Surge pressure gradient vs. trip speed: (a) high-viscosity mineral oil (m ¼204 cp); and (b) low-viscosity mineral oil (m ¼24 cp).
12 8.0
6.0
8
4.0
Fig. 9. Surge pressure gradient vs. trip speed: (a) 1.0% PAC (k¼ 2.26 Pa s , n¼ 0.57) and (b) 0.75% PAC (k¼0.670 Pa sn, n ¼0.67). n
Regression Model
Theoretical Model observed (Fig. 11a). However, for fluids with higher yield stresses
Schuh, 1964
(1.00% and 0.67% xanthan gum suspensions) predictions are slightly
2 higher (10% to 13%) than measurements (Fig. 11b). One possible
explanation for the discrepancies could be overestimation of the
yield stress resulting from the regression technique that uses very
limited data points at very low shear rates. More accurate visco-
1 metric data is necessary to reduce these discrepancies.
5. Parametric study
0
0.00 0.04 0.08 0.12 0.16
The relationship between surge pressure and pipe velocities
Pipe Velocity (m/s)
depends on a number of drilling parameters including fluid rheology
Fig. 10. Surge pressure gradient vs. trip speed for polymer mix (0.28% PAC and and borehole geometry. Using the new correlation, the sensitivity of
0.22% XG, k¼0.774 Pa sn, n¼ 0.5). the surge pressure to the variations of fluid yield stress and diameter
ratio has been investigated. Base case input parameters for this
with the theoretical model and existing surge calculation techni- analysis were: n¼0.5, k¼0.6 Pa sn, t0 ¼9 Pa, r ¼1000 kg/m3, dP ¼
que developed for power law fluids (Schuh, 1964). Discrepancies 0.127 m, and dh ¼0.272 m. Fig. 12 presents predictions of the new
between model predictions are minimal. model showing the effect of yield stress on the surge pressure at
To evaluate the performance the new model with YPL fluids, different pipe speeds. As anticipated, at high yield stress values, the
experimental measurements obtained using XG based fluids are surge pressure becomes less sensitive to the increase in pipe velocity
compared with model predictions (Fig. 11). For test fluid with the as the fluid becomes more shear thinning.
18 F. Crespo, R. Ahmed / Journal of Petroleum Science and Engineering 101 (2013) 12–20
3 5
2
3
2
1
Measurements Measurements
1
Regression Model Regression Model
Theoretical Model Theoretical Model
0 0
0.00 0.04 0.08 0.12 0.16 0.00 0.04 0.08 0.12 0.16
Pipe Velocity (m/s) Pipe Velocity (m/s)
Fig. 11. Surge pressure gradient vs. trip speed: (a) 0.44% XG (t0 ¼3.44 Pa, k¼ 0.359 Pa sn, n¼ 0.52) and (b) 0.67% XG (t0 ¼ 7.8 Pa, k¼ 0.553 Pa sn, n ¼0.5).
Guillot, D., Dennis, J. 1988. Prediction of Laminar and Turbulent friction pressures control tests with extensive instrumentation: field tests and data analysis. SPE
of cement slurries in pipes and centered annuli, SPE Paper 18377 presented at Drill. Completion, 251–257, December.
the European Petroleum Conference, 18–19 October, London. Rudolf, R., and Suryanarayana, P. 1998. Field validation of swab effects while
Haige, W., Xisheng, L., 1996. Study on surge pressure for yield-pseudoplastic fluid tripping-in the hole on deep, high temperature wells, paper SPE-39395
in a concentric annulus. Appl. Math. Mech. 17 (1), 15–23. (January). presented at the IADC/SPE Drilling Conference, Dallas, Texas (3–6 March).
Hemphil, T., Campos, W., Tehrani, M.A., 1993. Yield-power-law model accurately Samuel, R., Sunthankar, A., McColping, G., Bern, P., Flynn, T., 2003. Field validation
predicts mud rheology. Oil Gas J. 91, 45–50. of transient swab-surge response with real-time downhole pressure data. SPE
Isambourg, P., Bertin, D., Brangetto, M., 1999. Hydraulic tests improve HPHT Drill. Completion, 280–283, December.
drilling safety and performance. SPE Drill. Eng. J. 14 (4), 219–227. Schlichting, H., 1955. Boundary Layer Theory. McGraw-Hill, New York, pp. 60-62.
Kelessidis, V.C., Christidis, G., Makri, P., Hadjistamou, V., Tsamantaki, C., Mihalakis, Schuh, F.J., 1964. Computer makes surge-pressure calculations useful. Oil Gas J. 62
A., Papanikolaou, C., Foscolos, A., 2007. Gelation of water-bentonite suspen- (31), 96–104.
sions at high temperatures and rheological control with lignite addition. Appl. Singh, A., and Samuel, R. 2009. Effect of eccentricity and rotation on annular
Clay Sci. 36, 221–231. frictional pressure losses with standoff devices, Paper SPE-124190, presented
Kelessidis, V.C., Mihalakis, A., and Tsamantaki, C. 2005. Rheology and rheological at the Annual Technical Conference and Exhibition, 4–7 October, New Orleans,
parameter determination of bentonite-water and bentonite-lignite-water LA.
mixtures at low and high temperatures, In: Proceedings of the 7th World Sorgun, M., Ozbayoglu, M., 2010. Predicting frictional pressure loss during
Congress of Chem. Engg., Glasgow. horizontal drilling for non-Newtonian fluids,. Energy Sources Part A 33 (7),
Lal, M. 1983. Surge and swab modeling for dynamic pressures and safe trip 631–640. (December).
velocities, paper SPE-11412, presented at the IADC/SPE Drilling Conference, Wadhwa, Y.D., 1966. Generalized couette flow of an ellis fluid. AlChE J. 12 (5),
20–23 February, New Orleans, LA. 890–893.
Lin, S.H., Hsu, C.C., 1980. Generalized couette flow of a non-Newtonian fluid in Wagner, R.R., Halal, A.S., and Goodman, M.A. 1993. Surge field tests highlight
annuli. Ind. Chem. Eng. Fundam. 19 (4), 421–424. dynamic fluid response, Paper SPE 25771 presented at the IADC/SPE Drilling
Maglione, R., Ferrario, G., 1996. Equations determine flow states for yield- Conference, 23–25 February, Amsterdam.
pseudoplastic drilling dluids. Oil Gas J. 94, 63–66. Ward, C., Andreassen, E., 1998. Pressure-while-drilling data improve reservoir
Malik, R., Shenoy, U.V., 1991. Generalized annular couette flow of a power-law drilling performance. SPE Drill. Completion 13 (1), 19–24. (March).
fluid. Ind. Eng. Chem. Res. 30, 1950–1954. Ward, C., Beique, M., 2000. Pore and fracture pressure information from PWD data.
Merlo, A., Maglione, R., and Piatti, C. 1995. An innovative model for drilling fluid AADE Drill. Technol. Forum. (February 9–12).
hydraulics, Paper SPE-29259 presented at the Asia Pacific Oil & Gas Conf., White, Z., Zamora, M., Svodoba, C., 1997. Downhole measurements of synthetic-
Kuala-Lumpur, Malaysia. based drilling fluid in an offshore well quantify dynamic pressure and
Moore, P.L., 1965. Pressure surges and their effect on hole conditions. Oil Gas J., 90. temperature distributions. SPE Drill. Completion, 149–157, September.
(13 December). Zoellner, P., Thonhauser, G., Lueftenegger, M., and Spoerker, H. 2011. Automated
Rommetveit, R., Bjorkevoll, K., Gravdal, J., Goncalves, C., Lage, A., Campos, J., real-time hydraulics monitoring. Paper SPE 140298 presented at the SPE/IADC
Aragao, A., Arcelloni, A., Ohara, S., 2005. Ultradeepwater hydraulics and well- Drilling Conference, 1–3 March, Amsterdam.