)TC 6713 assessment of Conductor Setting Depth .R. Aldridge,
Fugro-McClelland
Ltd., and G. Haland, A/S Norske Shell
Copyright 1991, Offshore Technology Conference This paper waa presented at the 23rd Annual OTC in Houston, Texas, May O-9, 1991. This paper was selected for presentation by the OTC Program Committee followingreview of informationormtained in an abstract submitted by the author(s). Contents of the paper, as presented, have not been reviewed by the Offehore TechnologyConference and are subject to correctionby the author(a).The material, as presented, does not neceessrily reflect any positionof the Offshore Technology Conference or its officers. Permission to copy Is restricted to an abstract of not more than 300 words. Illustrationsmay not be copied. The abstract should contain mnapicuous acknowledgment of where and by whom the paper Is presented.
ABSTNACT This paper reviews the methode generally used by oil companies to determine the conductor eetting depth required to avoid hydrofracture of cohesive soils during drilling for the first casing string. Traditional approaches are compared with an approach developed by the authors, and the results of each method are compared with teat data obtained during geotechnical site investigations offshore. A case history is presented which showe the effect of the authors’ deeign method on the required conductor setting depth, and indicates the considerable possible cost savings and safety benefits available from effective conductor design.
INTRODUCTION The advancement of any kind of borehole ie dependent on the cuttinge being continually cleared from the bit face. This is usually achieved by direct circulation drilling, circulating fluid to the bit through the drillstring with the returning fluid and cuttinge passing up the annulus between the drillstring and the borehole or casing. If the marine casing is not set deep enough, the pressure of the drilling fluid may lead to formation breakdown and loss of circulation. Apart from difficulties in then advancing the borehole, this may also result in not being able to monitor and control shallow gas effectively. Formation breakdown can also lead to wash out and lees of support for the foundation of a structure. Correct assessment of the required conductor setting depth may therefore have not only economic but also safety implications for the well-drilling operation. One poesible method of determining required setting depth is to perform hydraulic fracture testing (HFT’s) in the field. This may however
References
and illustrations
prove to be both costly and in some cases difficult to perform. Where fracture test data are not available, analytical methods have traditionally been based on excess fluid pressures not exceeding the minor principal coil streae, These traditional methode are suspected to give unnecessarily deep setting depths, but no theoretically sound and practically proven method of calculating shallower setting depths is known to the authors. From an oil company’s point of view, the derivation of a reliable analytical tool that is less conservative than the traditional methods could lead to considerable savings. It ia, however, important that any such method should not underpredict the required setting depth, since the cost and safety implications of such underprediction may be considerable. A detailed review of field teste has therefore been performed to assesa the reliability of the new approach propoeed in this paper.
FIELD TESTS During the geotechnical site investigations performed for platforms in the North Sea hydraulic fracture tests (HFT’s) are often performed in order to determine ‘tin situ!! the fracture pressure which causes formation breakdown at various depths below mudline. The test is most frequently performed in hard clays using the type of apparatus preeented on Fig. 1. The typical procedure for performing euch tests in the North Sea is as follows: 1.
The borehole is advanced to the required depth using open hole drilling with returns to mudline, with the bottom assembly including the test apparatus as shown in Fig. 1.
2.
The string is pulled back to leave a length of predrilled hole below the bit as the test section,
at end of paper. 167
OTC 671:
ASSESSMENT OF CONDUCTOR SETTING DEPTN
2
3.
The packer is inflated to seal the test section, and a wireline dart is lowered to the bit to measure pressure during the test.
4.
The test section is pressurised by pumping fluid into the drillstring at a given flow rate and the test performed as outlined on Fig. 2.
Equation 1 represents the case of fracture occurring prior to blow off, and equation 2 represents fracture occurring following blow off of the soil from the piezometer. This assumption of a “perfect” installation and a Poisson’s ratio of 0.5 (undrained response) reduces equations 1 and 2, respectively, to:
The test is therefore generally performed in a pre-drilled section, and is flow controlled. The measurement made during the test include the fol– lowing: 1.
The initial fracture pressure.
2.
The steady rate.
3.
The close up pressure after initial fracture.
4.
The re-fracture pressure.
state pressure under a given flow
Whilst all stages of the HFT test may be used to infer geotechnical soil parameters, this paper concerns the pressure required to cause initial fracture, which is generally adopted as the limit of allowable excess fluid pressure during drilling operations. Initial fracture pressures from such tests have been used to check the theoretical method presented in the following.
THEORETICAL BACKGROUND
Minor Principal Stress Approach fluid and in situ The hydrostatic, drilling soil pressures during drilling for the first casing string are shown schematically on Fig. 3. The drilling fluid pressure which may be expected to fracture the soil formation has been the subject of analysis by Bjerrum et al. (1972). This approach was developed following obaervationa of fracture occurring whilst installing push-in piezometers. The excess pressure, Au, required to cause a vertical crack in the soil as derived by.- Bierrum et al. is given by: Au or Au where
v CY=
= Po’(l/v-l) [(l-~)ko + Pt’/po’] ... (1) = po’ (l-v) [(2+&c@ko
+ pt’/pol] ... (2)
= Poisson’s Ratio of the soil effect of installation on circumferential stress effect of installation on radial stress
or
coefficient at est
of
lateral earth pressure
tensi e stress sustainable by the soil
= ko.p ‘ + pt’ o
... (3)
Au
= ko.p ‘ + pt’/2 o
... (4)
Bjerrum also notes that there is a possibility of a horizontal crack forming if the excess head exceeds p ‘. In his recommendations on allowable pressures,” derived from theory and field and laboratory observations, the tensile stress ptt was conservatively ignored. These results then reduce to the assumption adopted by many oil companiea in estimating required conductor setting depth, that an excess pressure equal to the lower of the principal stresses in the ground should be assumed to cause hydrofracture, i.e.: Au=p’ Au
o
= ko.p ‘ o
(ko>l)
... (5)
(ko
... (6)
The analyses presented by Bjerrum et ai, are based on observations of push-in piezometer installation and subsequent applied excess water pressure. However, in offshore operations the borehole is advanced by drilling rather than pushing the string into the soil. The method generally followed in HFT’s performed offshore also forms a pre-drilled section of borehole prior to testing. The approach adopted by Bjerrum et al. to account for the disturbance caused by a cavity expansion into the soil may therefore not be applicable for this case.
Shear Failure Approach The in situ total stress condition in the soil and the changes in total stress caused during an idealised drilling and subsequent pressurisation operation for the HFT test are presented on Fig. 4. The stress changea in the radial and circumferential directions have been calculated from elasto-plastic theory (Den Hartog (1972)) on the assumption that: 1.
the SOil response ia linear-elastic perfectly plastic.
2.
the permeability of the soils is low enough that pressure changea due to flow are minimal.
3.
a Poisson’s Ratio of 0.5 is sssumed for, cohesive soil under undrained loading.
Po’ = vertical effective stress in-situ k= o
Au
It may be seen from Fig. 4 that for undrained loading the application of any increment of radial stress results in a reduction of the same magnitude in circumferential stress. Baaed on these atresa changes it is possible to datermine the excess fluid pressure in the borehole at which the circumferen-
168
WC
6713
ALDRIDGE AND HAIMD
tial stress falls to zero or to any given value of tensile stress, pt’, as follows: Au where
‘h
= 2,ko.p ‘ + uh + ptt o
RESULTS OF FIELD TESTS A review has been made of the reeults of 34 hydraulic fracture tests ‘(HFT’s) performed in predrilled sections in geotechnical boreholes performed during platform site-investigations in the North Sea, The teste were performed at aix sites, at depths of between 40 and 140 metres below mudline, in hard clay strata. Of the 34 tests, three resulted in very high fracture pressures close to those expected from cavity expaneion theory, as previously reported by Overy and Dean (1986). Three resulted in anomalously low pressures, believed to have been due to leakage around the packer. Data from the remaining 29 tests are reviewed below.
... (7)
= hydrostatic pressure at the given depth
Equation 7 is consistent with the results given by Jaeger (1969) for a porous elastic medium, if the permeability of the medium is set to zero. It may therefore be expected that hydrofracture will occur at the preseure given by equation 7; unless a general shear failure of the clay at the wall of the borehole occurs at a lower pressure. Examination of the three principal stresses given on Fig. 4 may be used to calculate the maximum deviator streea in the clay material. If the maximum deviator stress exceeds twice the undrained shear strength of the clay, then a plastic failure of the borehole wall may be expected to occur. The deviator stresses derived from the vertical (v), radial (r) and circumferential (c) principal etresses are as followe: = Au-
or-u
p , o
.,.
(8)
2.Au - ko.po’
...
(9)
v
Ur - IJC = Uv-u
‘Au+p
c
o
’-2,ko.p
0
’
The predicted and meaaured test results are compared for the six test sites on Fig. 5. The dashed line representa the calculated minor principal stress and the solid line is the lowest of the preseures derived from equations 11 to 13. The results chow graphically that the “shear failure!! approach gives a closer fit to the HFT test data than the traditional “minor principal stress!!method at these sites. The ratio of measured to calculated fracture pressure has been plotted for the traditional approach, i.e. equations 5 and 6, on Fig, 6, and for the pressures given by the lowest of equations 11 to 13 on Fig. 7, for all 29 sites. The valuee given by equation 7 are not plotted, since they are alwaYs higher than the values given by equations 11 to 13, and the “shear failure” mechanism may therefore be assumed to control. It may again be seen from Figs. 6 and 7 that the “shear failure!’ approach represented by equations 11 to 13 gives a significantly better overall fit to the data than the “minor principal stress” method.
... (lo)
Shear failure will occur when any of these deviator stresses exceeds twice the undrained compreaaive ahear strength of the soil. Using equations 81 to 10 it is therefore possible to derive defining the eqUatiOnS 11 to 13, respectively, eXCeSS fluid pressure which would cause a shear failure in the borehole wall: Au = 2.su + Po’ Au=
S
u
+
kO.PO’
Au = 2.au + po’(2ko-1) where
... (11) . . .
(12)
... (13)
s = undrained ehear strength in compression u
An alternative possible mode of failure to that considered above is a uniform cavity expansion, for which the excees pressure to cause failure is of the order of 5.7 to 6.3 times the undrained shear strength, depending on the overconsolidation ratio of the clay (Randolph et al. (1979)). It ia poaaible that a ahear failure, as given by the lowest of equations 11 to 13, will occur prior to the tensile failure given by equation 7. There are assumptions inherent in both theoretical approaches, however. Observations made during drilling for the first casing string are therefore reviewed below to assess the validity of each approach. The results of these approaches are also compared with the method traditionally used, as given by equations 5 and 6.
?
Observations of drilling mud pressures and returns or lack of returns during offshore drilling operations are not generally available to the geotechnical consultant. Records obtained during drilling from a semi-submersible drilling rig at the Draugen site in the Norwegian Sea are presented on Fig. 8. This figure shows the mud preaeure actually applied during drilling and the estimated fracture pressure based on the “minor principal stresstt and “shear failure” methods. Again this data confirms that the “ahear failure” approach gives results which are more coneiatent with observations, since although excess drilling fluid pressures exceeded those given of by the “minor principal stress” approach, no IOS5 returns waa encountered. The results of a statistical analysis of the data are also presented on Figa. 6 and 7, and show that the measured test pressures are on average almost exactly double those given by the “minor principal strees” method but only 34 per cent higher than given by the “shear failure” method. The statistical correlation, as measured by the standard deviation, is also better for the ‘rehear failure” method.
169
ASSESSNE??YOF CONDUCTOR SETTING DEPTN
4
Whilst the above data is limited, the statistical correlations indicate that the “shear failure’ approach presented here is the more appropriate method for calculating setting depth, From the data analysed here, it is suggested that the excess pressure calculated using the “shear failure” appreach should be divided by a factor of safety of 1.3 to give an allowable drilling fluid pressure for assessing the required setting depth. This approach should result in a greater than 95 per cent statistical confidence of avoiding hydrofracture. Records from actual well-drilling operations and further in situ HFT tests may allow this factor of safety to be reduced with time.
CASE STUDY The “shear failure” method described above has been used in the establishment of setting depth for the Draugen Field offshore Norway. A/S Norske Shell is the operator for the field on behalf of their to partners Statoil and BP Norway, and proposed install a concrete gravity base structure supporting 10 well slots with six producing platform wells and some subsea wells. The conductor arrangement at the gravity base structure is shown schematically on Fig. 9. The water depth at the Draugen platform site is 252 metres, and the drilldeck is approximately 313 metres above mudline. It is planned to install the platform in the field during the summer of 1993 with conductor setting starting a few days after platform installation. The soil investigation covering the upper 130 metres of the soil revealed clay layers with varying shear strength. The strength in the most critical layers, i.e. between 50 and 150 metres below mudline, varies between 200kPa and 1200kPa as shown on Fig. 10. NO hydraulic fracture tests were performed, partly due to cobbles within the clay layers, which could have made the use of testing equipment very time consuming, For the gravity base structure, analyses were performed relating to the condition following installation of the structure. Following placement of the GBS structure, the increase in the total horizontal and vertical stresses beneath the structure were calculated using elastic theory (Poulos and Davis (1974)). Equations 5 and 6 were then modified to incorporate the increased stresses directly, and equations 11 to 13 required modification ae follows: AU=2.Su+p
’+Ap o
where
. . . v
Installation of the 26 inch conductor into the clay layer between 65 metres top of the very hard and 95 metres below mudline was considered feasible using a drill/drive sequence with an IHC S90 pile hammer, or equivalent. Installing the conductor below this layer would have called for a larger hammer and thicker wall conductor to give the same confidence in reaching the required setting depth. The presence of a aand layer at 130 metrea below mudline would also have led to an increased risk of encountering shallow gas during conductor installation.
CONCLUSIONS The data presented in this paper indicate that the traditional “minor principal stress” method of estimating conductor setting depth is generally conservative, and may result in much deeper setting depths for the conductor than are actually required to avoid hydrofracture during drilling for the first casing string. The “shear failure” approach as presented in this paper ia considered to give a more realistic assessment of the actual required setting depth, and ita use, in conjunction with an appropriate factor of safety, will often result in significant savings in the casing programme. The use of the “shear failure” approach has led to a saving of more than 50 metres on the Draugen conductor design, and has avoided the need for a more expensive installation method using thicker wall conductors and heavier plant. The shallower setting depth has also eliminated the requirement for special procedures to install the conductors through a sand layer which would have presented a significant increased risk of encountering shallow gas.
APh
... (15)
ACKNOWLEDGMENT
po’(2ko-1) + 2.Aph
.... (16)
The Authors wish to thank A/S Norske Shell, Statoil and BP Norway for permission to publish this paper.
ko.p ‘ + 0
=
S
Au
=
2.su+
Apv
=
increase in vertical total stress
Aph
=
increase in horizontal stress
+
shown on Fig. 11, whereas the “shear failure” appreach indicated that a setting depth of only 70 metres below mudline would provide an acceptable safety level of over 300kPa difference between the calculated fracture pressure and the estimated mud pressure. This difference was equivalent to a factor of safety of 1.39 on excess pressure, resulting in a greater than 97 per cent statistical confidence of avoiding hydrofracture. This probability, in conjunction with the confidence given by the apparent lack of hydrofracture during drilling operations at the site, resulted in a 70 metre penetration being considered an acceptable conductor setting depth for this project.
(14)
Au
u
OTC 6713
Using these modified stresses, the traditional “minor principal stress” method led to a setting depth in excess of 130 metres below mudline, aa
----I [u
)TC 6713
10
A.LDRII!GE ANI
.
—SIGNAL
Bjerrum, L., Nash, J.K.T.L. , Kennard, R.M. and Gibson, R.E. (1972), “HydraulicFracturing in Field Permeability Testing!!, Geotechnique Vol. 22, No. 2, pp. 319-322.
—~
CABLE
F T DART
2.
Den Hartog, J.P. (1952), “AdvancedStrength of Materialsit,McGraw-Hill.
3.
Jaeger, J.C. (1969), “Elasticity, Fracture and Flow”, Halsted Press.
_SLllJING VALVE FOR PACKER
4.
Randolph, M.F., Carter, J.P. and Wroth, C.P. (1979), llDri~en pilee in Clay - ‘he Effects of Installation and Subsequent Consolidation”, Geotechnique Vol. 29, No. 4, pp. 361-393.
—PRESSURE DROP VALVE
5.
6.
—PRESSURE SENSOR
—ROUGH HOLE PACKER
Overy,
R.F. and Dean, A.R. (1986), “Hydraulic Fracture Testing of Cohesive soil”. Proc. Offshore Technolozv Conference. -. Paper No. OTC 5226.
-OPEN
Poulos, H.G. and Davis, E.H. (1974), “Elastic Solutions for Soil and Rock Mechanicst$. Series in Soil Engineering, John Wiley and Sons.
,,.iil
Fig.1
I
I
1
,,
1000
I
I
I
TJUF
Hydraulic FractureTestEquipment
RF
I
I
II
800 k
I
I
I
I
CLOSE UP PRESSURE
I ‘i
I
STEAOY STATE PREsSUBE
1
I
BIT
1’
I 4
/ “
I
600
\
?w
uP
\
<
‘
T
400
i
200
PUMP OFF - +
PUMP O Jb
+
PUMP OFF
o 0
4
a
12 TIME
16 AFTER
20 START
24 OF TEST
28 (rein)
Fig. 2 Hydraulic Fracture Test Procedure
171
32
36
40
I
1’
‘1 ‘1 ‘1 ‘1 ‘1 ‘1 ‘1 ‘1 ‘1 ‘1 ‘1 ‘1 ‘1 ‘1 ‘1 ‘1 ‘1 ‘1
I I I
,,l!!
,1,1,
I
I
I
I
1!:’
I
,,!!!
I I I
,,
II
I
I
I
I
1!:
I
1,1,
I “’”
,,11, II II II II
,
II
I
I
““
‘,
‘,
1,
II
J
I
I
1,
1,
I
I
I
I
I
1, i II II II II II II II II II II II
I I I I I,;!!:”I I II II II 1,1, II II
I
I “’’”
I
l,l,
II
1,i, 1, 1; 1, 1, 1, 1,
1,
I,,!! I I I
I
I
1!!:
1,1, 1,1, i,lll, II
1,1, l,
,,
~
~
~
1,
1,
1,
i,
II
~
~
I
I,,!! I
I
1!!’
I
I
I
I “’’”
I
I
1,1,
I
I
i,lll,
i,i,
l,l,
II
II II
I
I
I
I “’’”
I
I
I
~
~
II
;
I
I “’’”
I
1,
1,
1,
1, i, 1, 1, 1 1, 1, 1,
1,
1,
1,
1,i, 1, 1; 1, 1, 1, 1,
‘1
II
II II II II
koPo’+ Uh 1,
;“’’’’l lllllllllll!l ,,
l,l,
II
l,l,l,i,i,l~l~,~,~,
I
TOTAL STRESSESIN-SITU
i,l, ill,
l,i,
)
I
I
1!:
1,1,
I
~
II II II II II II II II
I I I
I “’”
I
I
I
~
1’
I
II II II II II II II II II II II II II
I I I!!1!:I
~
,,
II II II II II II II II II II II I I I
~
1 ! 1, 1, 1, I
i,
““
,,
II II II II II
,,l!!
~~
,’,,,,
1,
‘1
,1
I
I
II!!
,,,
‘1
i,l~ll,
l,i,
11,,,,1,,1’~ I
‘1
II II II II
I,,!!!!:I I I
II
‘,
““““““““““““““ 1,1, l,i, !!!!!:
II
II II II II II II II II II II II II
200
1,1, l,
i,lll,
“’’”
~
0 ,,!!
I I I
I
1!!:
I
I
I
I
I
I “’’”
1,
I
1,
1,
i,
1 ~ 1, 1, 1, I
i,
II
II II II II II II II II II II II
I
‘1
. -1 IO
TOTAL
STRESSES
DURING
ORILLING
-200
I,,!! I
I I I
1!!!
I
I
I
I
I
I “’’”
1,1,
I
1,111,
i,i,
l,l,
II
II II II II II II II II II II II
I,,!!1!1’ I I I
I
I
-a
I
I
I
II II II II II II II II II II II II II I ,,, I I I!!1!:
II II
,,!!!
i,l, ill,
l,i,
l,l,
II
II
~
-ma I
I “’’”
1,1,
I
i,lll,
l,i,
1,1, l,
II
‘1
I
I
1!:
-ma I
I
I
I “’’”
I
1,1, 1,1, i,llll,
I
II
I
I
II
-1000
I
I
!,
I
I
I “’’”
I
1,1, 1,1, i,lll,
I
1,1, l,
II
I
1
1
!
!/l
,1
,,,,,,,
I I I I I,;!!:”I I II II II 1,1, l’, II
I
I
I
I
1,
1,
1,
i,
i,
““
II!!
II
#
I
1!!:
I
I
I,,!! I I I
I
1!!’
I
I
I
~
‘1 I
I
I “’’”
II II II II II 11111111111111
1,1,
I
I
I
i,lll,
i,i,
l,l,
II
)
‘1
‘1
I
I
“’”
1,1,
l,i,
1, I,
I,
i,l, ill, II
l,l,
II II II II II II II II II II II I
I
I
1!!’
I
I
I
I
I “’”
I
I
i!
I
I ~ 1, 1, 1, I
II II II II II
I,,!! I I I
I
,’,,,
!1111!!
11111,!
I,,!! I I
UhdJ
l,l,
IIII II II
I,,, I I!!!1!:I
II
1,1,
~
I
I
I
II II II II
I I I II II
I
I “’”
i,
i,
1, ~ 1 I 11 ,,!11
I
Fig.3
Hydrostatic,Mud-umand InsituSoilStresses
Fig.4
Changes in TotalSoilStresses DuetoDriUingOperations
EXCESS FRACTURE PRESSURE
0
1
(MPal
EXCESS FRACTURE PRESSURE
z
o
40
40
(MPa)
1
2
1-
● 1
I 60
60
●
I I
‘Y\\ 1
I
\
80
)
1
\
\
80
100
/
I I 1 I
100
120
●
I I I I
120
140
●
●
I I
m
140
o
i
0
2
50
2
4
60 I I I
70
I
\ \ t
80
I 1
●
✼ ✼
‘*
I
90
●
1
I I , I
,. 100
1
1 x
I
●
110
120
0
i
o
2
00
100
●
2
A
80 ●
1 \
●
\ \
\
100 \ \
\ ●
\ \ 120
\
120
w
\ \
o I \ \ .
\
o
D
140
140
‘-’-—
HFT RESULTS MINOR PRINCIPAL STRESS PREDICTION SHEAR FAILURE PREDICTION
Fig.5
Comparison ofHFI’Reaults with Predictions
173
(B
RATIO o
1
RATIO
MEASURED/CALCULATED
2
3
4
5
I
I
o
6
i
MEASURED/CALCULATED 3
2
4
5
I
I
6
40 -1
SD [
60
MEAN
+1 I
●
I
I
I
4
I
SD
●
●1
I
●
b
1:
I
MEAN
=
1.99
S.D.
=
0.52
I
I
SD
MEAN
I
I
I
I
100
I I 1° ● P I I
120
‘1
I I
100
●
18
●
I
-0 ● 1-
120
q I I I
140
f Ill 140
Fig. 6 Ratio of Measured to Calculated Fracture PresaureMinor Principal Stress Method
1.34
S.D.
=
0.32
1
-1
J
r
1
●
10! I
I
I
I
I a
I
I
●
‘1 ●
I ●{ I
●
●1
I
1010’ dw I ● D
I*
=
1°
● I
m
●
I
MEAN
I
,
10
I
.1 ●
ao
SD
I
L
60
~
●
-1
+1
●
I
I
I
I
I
I
I
I
I
Fig. 7 Ratio of Measuxwl to Calculated Fracture Pres.sureShear-Failure Method
1
I
(-ah
—._ — & — r
—.
Y
.“
n
‘H
-: !1/
1! H--l Hl--Ill
.—
—.
—.
g
:::--H-I-W+H47H:H :4 :-.-—. ,._.LLmtmTl i tilt< . ._. . “-’”:F “l’’t-+’=t’;--’l:+:F+l<-l<4 2? – c!”
. .: :.— ;. .. .- —= .-.
~>
.
08
.
.-.
8
_. ._.,, .
.
e
.
___
.1 .
__.
8
.
.
.,.
.,.
-
.
_ ..
_
8 [W
z-. 3N11OW
M013B
175
__
.
.
----
.._
__ .—
s H1d30
g
~
.——.
~
_.
~
m . t?
. . . —.—.. ..— —....—. =n
‘“~.-—-– —— –— !#
g“ ~
z 8 :.
..-_._ . . —.
—..——
“
.
.._ .
8
—.
$
s s. ;
) “
,/’
-=
-..
“8
.>=
—
.
.
-—
.-
-
s W
%11MV4 lk0130 H1d30
gl e r%
0
1+ .
2
.-
176
