Professional Documents
Culture Documents
Scholars' Mine
International Conference on Case Histories in (1998) - Fourth International Conference on Case
Geotechnical Engineering Histories in Geotechnical Engineering
A. Desai
Bechtel Corporation, Gaithersburg, Maryland
H. Senapathy
Bechtel Corporation, Gaithersburg, Maryland
L. W. Young Jr.
Bechtel Corporation, Gaithersburg, Maryland
Recommended Citation
Clemente, J. L. M.; Desai, A.; Senapathy, H.; and Young, L. W. Jr., "Foundations Performance of Large Diameter Tanks" (1998).
International Conference on Case Histories in Geotechnical Engineering. 53.
http://scholarsmine.mst.edu/icchge/4icchge/4icchge-session01/53
This Article - Conference proceedings is brought to you for free and open access by Scholars' Mine. It has been accepted for inclusion in International
Conference on Case Histories in Geotechnical Engineering by an authorized administrator of Scholars' Mine. This work is protected by U. S. Copyright
Law. Unauthorized use including reproduction for redistribution requires the permission of the copyright holder. For more information, please contact
scholarsmine@mst.edu.
265
Proceedings: Fourth International Conference on Case Histories in Geotechnical Engineering, St. Louis, Missouri,
March 9--12, 1998.
ABSTRACT
The paper presents a detailed case history of foundation performance of six 60-m diameter, 15-m high, floating roof fuel oil tanks and
six 96.8-m diameter, 20-m high, fixed roof process water tanks built for a large power plant. Tank walls were supported by concrete
ringwall footings. General subsurface conditions at the site are discussed, along with proposed site grading and the rationale for tank
foundation selection. Because vibro-replacement improvement of site soils had been used beneath settlement-sensitive structures.
there was skepticism regarding the decision to support the tanks on unimproved soils. To allay doubts about the adequacy of tank
foundation performance, a staged hydrotesting procedure and an extensive settlement monitoring program were developed and imple-
mented. The excellent tank hydrotesting results demonstrated that ground improvement was not needed due to the more settlement-
tolerant nature of the tanks.
KEYWORDS
"'"'
::r
Wf-
_j<( -15 ..
58
Jl
~
B
30
21
CORAL ,, ..........-15
" wo .. ···2J ~· :1)' " ' " " • •U21
~ f-
37
J8
•D !)
52
5l
77
"" z -20.... ~_7"'. ~1.,_. 46-·· ... ........-2C
"'...
-§, ~
"
" s -25...
-'
. ............-25
(]] ~
L
QJ 5 5
~
PWT 9 PWT 7 PVfT 5
QJ
Finish Grade [1. 2.4r1
~ E Q 0
~ BH-12 ·siR.UCtURi\L FILL. . BH:_·13 BH'-H
.,
'-'
~ -5. ~ .... ~ Y SAND
G
- l~
9 ..
..... -5
~
::?:
::0
8 SILTS AND CLAYS '15 "
13
~ -- -- . -- -- f---. . --
... --
... --
... --
... --. -10
"'~ f-
~
0
-1Q 2
,,"
27
-- . . -- .. 38.
27
" .
23 .•
"
""'"'"
26 -15
f- -15. .. 22 . . . ..... . . 20 . . .
z J5
CORAL
2B
""' "44
57 33
~
~
_J
(l_
-2G .. 46
7f "46· - 39· -20
""'
"(l
~
z
0
-25 . . . ..""
J7
.0
=-
... 20. . . . . !10 .
"
50
27
32. -25
" f-
" HO.,IZONTAL SCALE- METERS " -30
~
>
-30 7Q
~
..
' . . . ..
"
46. N
Fourth International Conference on_Jw
Case Histories in Geotechnical Engineering
Missouri University of Science and Technology
w -35
""--
-35
"'_,
http://ICCHGE1984-2013.mst.edu
268
erecting and hydrotcsting tanks, and the development of a where: ui, ui·l> ui+l ::::; out-of-plane differential movement, in
comprehensive staged hydrotesting program with extensive mm, for three neighboring, equally
settlement monitoring. spaced tank shell points
L = distance, in mm, between equally
spaced points i, i-1, i+ 1
Tank Settlement Criteria
The maximum tolerable out-of-plane distortion is 1/450, or
The following tolerable settlement criteria were adopted as a 0~22 percent. For a 96.8-m (96,800-mm) diameter tank and
basis for evaluating tank performance during hydrotcsting. eight equally-spaced settlement markers, such as the process
These criteria are based on published literature (Rosenberg water tanks, the maximum tolerable value of ~Sop is 84 nun,
and Journeaux 1982) and have been used extensively for tank and for a 60-m (60,000-mm) diameter tank and eight equally-
settlement performance evaluation. spaced settlement markers, such as the fuel oil tanks, the
maximum tolerable value of ~Sop is 52 mm.
Shell Settlement. Uniform settlement of the concrete ringwall
footing is generally not used as a tolerable settlement criterion, Bottom Plate Settlement. Edge-to-center distortion is defined
because uniform settlements do not cause detrimental effects as the maximum difference in measured settlement between
to either the tank shell or bottom. Uniform settlement of the the center of the tank bottom over the radius of the tank, i.e.,
concrete ringwall footing can generally be accommodated by
providing flexible tanklpipe connections. Edge-to-center distortion = (Sc - SE)/R = ~SEcfR (3)
Planar tilt is defined as the difference in measured settlement where: SE = settlement under the edge of the tank, in
between two diametrically opposed points on the tank shell mm
divided by the diameter of the tank, i.e., Sc = settlement under the center of the tank, in
mm
Planar tilt= (S 1 - S2)/D = <l.S.,ID (1) R tank radius, in mrn
ll.SEC difference between Sc and SE = differential
where: St. S2 = settlement of two diametrically opposed settlement between the edge and center of
points of the tank shell, in mm the tank, in mm
D tank diameter, in mm
~So = difference between S1 and S2 = differential The maximum tolerable edge-to-center distortion is 1/50, or 2
settlement between two diametrically percent. For a 48.4-m (48,400-mrn) radius tank, such as the
opposed points of the tank shell, in mm process water tanks, the maximum tolerable value of
differential settlement (ll.SEcl is 968 nun, and for a 30-m
The maximum tolerable planar tilt is 1/200, or 0.5 percent. (30,000-mm) radius tank, such as the fuel oil tanks, the
For a 96.8-m (96,800-mm) diameter tank, such as the process maximum tolerable value of differential settlement (llSEc) is
water tanks, the maximum tolerable value of differential 600 mm.
settlement (ll.S 0 ) is 484 mm, and for a 60-m (60,000-mm)
diameter tank, such as the fuel oil tanks, the maximum
tolerable value of differential settlement (l.\S 0 ) is 300 mm. Settlement Analyses
When a perfect planar tilt occurs, all points along the concrete
ringwall footing remain on a plane with a slight tilt from the Settlement analyses were performed using the typical consoli~
horizontal. dation parameters previously described for the silt/clay layer,
and elastic parameters for the granular soils (including struc-
Out-of-plane distortion is illustrated in Fig. 4 (Rosenberg and tural fill to be placed in the process water tank area). Based on
Joumeaux 1982). If a perfect planar tilt of the tank occurs, a the SPT N-values, an elastic modulus of 17,500 kPa was
plot of the settlements along the perimeter of the tank would selected for the natural granular soils in the process water tank
result in the cosine-shaped curve shown in Fig. 4. When the area, and a value of 13,500 kPa was selected in the fuel oil
tilt is not perfect, points of the concrete ringwall footing move storage area. The elastic modulus of granular structural fill
away from the slightly tilted plane described in the previous was selected to be 22,500 kPa, based on previous experience.
paragraph. The result is that when settlements along the
perimeter of the tank are plotted, they do not fall on the Consolidation settlement analysis of the silt/clay layer was
cosine-shaped curve shown in Fig. 4. An out-of-plane performed using the TCON Version 4.99 software package
differential movement (u) described in Fig. 4 develops, and the (TAGA 1993) that allows the simulation of load application
out-of-plane distortion is defined as: with time.
Settlement Measurement
Point (Typical) ---..........
-------.. 8 2
3
1----+--- Reference Radius
Out-of-Plane Distortion
At Measurement Point 7
Observed Settlement
Fig. 4 Planar & out-of-plant tilt evaluation (Rosenberg & Journeaux 1982)
Analysis results indicated settlements of 55 rnrn at the edge Staged Hydrotesting Program
and 100 mm at the center of the fuel oil storage tanks at the
end of hydrotesting. Calculated settlements were 140 mm at The following staged hydrotesting procedure was developed
the edge and 260 mm at the center of the process water tanks and implemented:
at the end of hydrotcsting. The TCON analyses also indicated
that the settlements in the silt/clay layer would stabilize within Step 1 - Install settlement monitoring markers at eight equally
a short period of time (weeks rather than months). spaced locations along the perimeter of the tanks.
The calculated settlements would result in edge-to-center dis- Step 2 - Obtain the "zero-loading" reading of each of the
tortions much smaller than the maximum tolerable values pre- settlement monitoring markers.
viously described. The calculated settlement values were also
within the range of tolerable limits included in the authors' Step 3 - Fill tank to 50 percent capacity. Obtain one set of
database of tank settlement measurements during hydrotesting readings immediately before filling the tank, one set of
(Senapathy et a/. 1994 ).
Detailed settlement monitoring result.;; arc presented for FOT The data in Fig. 6 indicate that the maximum out-of-plane
#4 and POT #4, i.e., the first fuel oil tank and the first process differential settlement for FOT #4 was about 23 mm. The
water tank to be hydrotested. Changes to the hydrotesting pro- maximum out-of-plane distortion was about 1/2,100, or 0.047
cedure based on the settlement behavior of FOT #4 and PWT percent. This value is about five times smaller than the 0.22
#4 are discussed. Remarks are offered regarding the settle- percent allowable.
ment behavior of the remaining tanks.
Time (days)
0
5 30 35 40 45
-20
-40
e
.s -60
E
~ --¢--#113
E --£:r#114
~
E -80
-6-#115
~
Fig. 5 Time vs. settlement curve for fuel oil tank No. 4
Fourth International Conference on Case Histories in Geotechnical Engineering
Missouri University of Science and Technology
http://ICCHGE1984-2013.mst.edu
271
.,
c F 0 T t 4
'0 -' - co Bin e c u ·~ .. 6 P W Tf 4
'w ' eosin• c u r • •
_.,
"
•" ·"
ffi
"~
~
·"
-
1---
~-------
-~
~
i
'
.
----
;!;/
-v v
---------
"~
w
>
-1 0 Q
-------------------
. _, 2 0 ~--~-.;__
... -
' '
-
-i .-----;;-
w - I I _____ ...,
-1.0
-~----~-4-
-I - - - A
- - - - I'
-1 eo
'" 13S
ANGLAil DISTANCE
tao 225
degrooo '" ,"
The maximum planar tilt for FOT #4 was 69 mm, which at the center of the tank was damaged while filling the tank to
corresponds to about 1/870 or 0.12 percent. This value is 50 percent capacity. Also, the holding period at 50 percent
more than 4 times smaller than the allowable. loading did not fully stabilize before the tank was filled to 75
percent capacity.
Figure 7 shows a plot of settlements of two points on the
perimeter where maximum and minimum edge settlements The data in Fig. 8 indicate that the maximum settlement at the
were measured, as well as the center of tank settlement for edge of PWT #4 was 154 mm and the minimum settlement
FOT #4 and PWT #4. The data in Fig. 7 indicate that the was 119 mm. The average settlement along the edge of the
maximum edge-to-center differential settlement for FOT #4 tank was about 136 mm, which is almost identical to the
was 87 mm. The maximum edge-to-center distortion was predicted 140 mm and well within tolerable limits. The data in
about 11870, or 0.29 percent. This value is more than 6 times Fig. 8 also show how quickly the settlements stabilized after
smaller than the allowable. the 75 percent and 100 percent loading stages were reached.
Based on these results, the hydrotesting procedure was
The settlement behavior of the remaining fuel oil tanks was changed to allow holding the 100 percent load for a period of
similar to that of FOT #4. no more than 2 weeks for the remaining process water tanks.
F0 TI 4
T
_, 1"----
~
~
w
);;
~ ·1 0 0
5
0
z 1---
,..
0
·l 50
-- ---
w
>
w
"
~ --- -- --
w ·2 0 0 1------------~-+------"-',__,_____
- - -------~~--=-'==----+------~---1
·2 50
0.2 5 0 5 0 '7 5
RATIO OF iDIAMETAIC DISTANCE ALONG TANK·BASE)/OIAMETER
-20
-40
-120
--·5
--e--.4
---··
--.--.7
-140
--•a
------ C en Ia r
-1 6 0
Fig. 8 Time vs. settlement curve for process water tank No. 4
The maximum planar tilt for PWT #4 was 23 mm, which served as the basis for the foundation selection strategy for 12
corresponds to about 1/4,400 or 0.023 percent. This value is large diameter tanks. The available data and rationale indi-
more than 22 times smaller than the allowable. cated that the tanks could be built without the then-perceived
notion that ground improvement would be required. The tanks
The center-of-bottom-plate settlement shown in Fig. 7 was were erected and hydrotcsted without using ground improve-
calculated based on the settlement analysis results and edge ment, and excellent settlement performance was observed. It
settlements shown in Fig. 8 (the settlement marker at the center was confirmed that vibro-replacement ground improvement
of the bottom plate was damaged, as shown by readings on was not needed, due to the more settlement-tolerant nature of
Fig. 8). The data in Fig. 7 indicate that the maximum edge-to- these tanks.
center differential settlement for PWT #4 was 94 mm. The
maximum edge-to-center distortion was about 1/515, or 0.19 REFERENCES
percent. This value is more than 10 times smaller than the
allowable_ Rosenberg, P_, and N. L. Journcaux [1982]. "Settlement
limitations for cylindrical steel storage tanks." Canadian
The settlement behavior of the remaining process water tanks Geotechnical Journal, Ottawa, Canada, 19:232-238.
was similar to that of PWT #4, except that edge settlements of
slightly more than 200 mm were observed for PWTs #7, #8 Senapathy, H., J_ Davie, and L. W. Young, Jr. [1994]_
and #9. However, maximum planar tilts, out-of-plane "Performance of a steel tank founded on a marine clay." Proc.
differential settlements, and distortions remained well below of Settlement '94, ASCE Geotechnical Special Publication No.
tolerable limits. 40, New York, Vol. I, pp. 830-84 L