Settlement Criteria for Steel

Oil Storage Tanks

Ali Akhavan-Zanjani
Research Student, Department of Civil engineering, University of Tehran, Iran
aakhavan@ut.ac.ir

ABSTRACT
This paper discusess the criteria of settlement in steel tanks which are used to storage oil or
gasoline. The steel tank is as a representative of many steel tanks constructed in south of
I.R.Iran, that has a ratio between the diameter and the height of order 4 with slenderness ratio
(radius to thickness) of the order of 1000 (first coarse) to 3750 (last coarse). weakness of the
site soil causes settlement to be more than usual so the most economical solution is to find how
much can the settlement be.
KEYWORDS: steel tank, settlement, tilt, shell

INTRODUCTION
The settlement of the foundation in large, thin walled shells has been of great concern in the
past and there is some codes and articles about it that are so useful. this paper wanted to show that
what is the Criteria of allowable settlement of a large and small steel tank. so the paper considered
some large and small steel tank that are the representative of many steel tanks constructed in south
of I.R.Iran.

According to D’Orazio and Duncan, examination of the settlement measured for the tanks
shows one fact clearly: Steel tank bottoms can undergo a wide variety of types of distortion as they
settle”. However, most analytical studies concentrate on just one type of distortion: a vertical
displacement pattern at the base of the shell that follows a harmonic shape. In another paper, the
same authors state: “Because their walls have significant stiffness and ability to span local soft
spots, the settlement profiles of tank walls tend to be smooth and free of sharp variations.

Here is one of the disasters that happen because of a tank failure that have been reported in the
literature notably is the report of the failure of a 26.15 m radius shell storing hot-oil in Japan in
1974. The consequences of this failure were manifold: “The contents flooded much of the refinery
property and flowed into the adjacent inland sea causing severe damage to the fishing industry. As
a result, the 270,000 bbl/day refinery was shut down for about nine months, largely because of
public reaction. By the time the refinery was permitted to resume operation. The accident had cost
 Vol. 13, Bund. B 2

the refinery more than $150,000,000. This shows how important and dangerous can the damage of
steel tank specially large steel tank be.

The paper is organized as follows: section 2 contains the most usual settlement that would
happen, section 3 is the case studies and review of literature, section 4 is about the comparison of
cited allowable settlement and section 5 wanted to refer recommended settlement.

DEFINITIONS
Various forms of settlements could take place so it is crucial to define all required variables at
the beginning of this chapter as follows:

• D = Diameter of the tank.

• R = Radius of the tank.
• H = Height of the tank.
• L = distance between two points with differential settlement.
• Δ max = Total maximum settlement: This type of settlement illustrates in Figure (2-1).

Figure 2-1. Total Maximum Settlement of Steel Tank Figure 2-2. Average Settlement of a Steel
Tank

• Δ ave = Average settlement: This type of settlement is an average of the settlement of all
points of a tank (Figure 2-2).

Figure 2-3. Tilt of a Steel Tank Figure 2-4. Bottom-Edge Differential Settlement of a ste
s steel tank
• w = Tilt: This component rotates the tank in a tilt plane (Figure 2-3).
• δ = Differential settlement between two points.
δ bottom = Edge settlement occurs when the tank shell settles sharply around the
•
periphery, resulting in deformation of the bottom plate near the shell-to-bottom corner
junction, (Figure 2-4), or the depth of the depressed area of the bottom plate, (Figure 2-
5).
 Vol. 13, Bund. B 3

• δ shell = This component of settlement at the bottom edge leads to the lack of
δ
circularity and creates stresses in the shell. shell is defined as differential outline
settlement between settlement of one measurement point with respect to the average of
settlements of its two adjacent points (Figure 2-2).
δ i = U i − (0.5 × U i +1 + 0.5 × U i −1 ) Eq. 2-1.
δ i = Differential settlement between one point and average settlement of its adjacent
points
U i = Settlement of each points in Figure (2-6).

Figure 2-5. Bottom-Center Differential Tank Figure 2-6. Shell Differential

S et setttlement of a Steel Settlement of Steel Tank

REVIEW OF LITERITURE AND CASE STUDIES FOR

THE SETTLEMENT OF STEEL TANKS
It is important to consider steel tanks suffered from excessive settlement in the past. Therefore
case studies in addition to the review of literature, design codes/standards and highly referred
papers, are presented to help a realistic judgment to be undertaken about allowable settlement.

Klepikov (1989) reviewed a large number of references related to allowable settlement.

According to Klepikov (1989), steel storage tank, with capacity less than 10000 m3 and between
20000~20000 m3, could ultimately tolerate 110 mm and 180 mm average settlement (Δavg)
respectively. For small tanks with dimensions of D=9 m and H=8 m in P.L.D area of this project,
the capacity is equal to 508.7 m3. For large tanks with dimensions of D=53.6 m and H=18.3 m,
the capacity is equal to 41271.5 m3. It should be noted that maximum total settlement (Δmax) is
often larger than average settlement (Δavg) which is referred by Klepikov (1989). This reference has
also recommended differential settlement ratio (δ/L) to be less than 0.004 for bottom of large tanks
and less than 0.008 for small ones (L=D), and allowable outline shell settlement is equal to 0.01
and 0.008 for large and small tanks respectively (L=6m). And finally tilt (w/H) of all tanks should
be less than 0.007. The limit for visible tilt is equal to 0.004.

(B) USACE (1990) published an engineering manual, EM 1110-1-1904, for geotechnical

procedures. It suggests that allowable differential settlement, (δ/L) for circular steel tanks on
flexible base, either with fixed or floating roof, could be consider as equal to 0.008. For large tanks
in Mahshahr oil product terminal revamp project L=D/2. The mentioned ratio suggests 213 mm
and 36 mm as differential settlement between center and edge for large and small tanks
respectively. As mentioned by a number of authors (Bowles, 1996), differential settlement is
conservatively equal to 75% of maximum total settlement, (Δmax). Therefore the allowable total
settlement could be estimated.
 Vol. 13, Bund. B 4

(C) API 653 is the most related document to this report because it only concentrates on oil
tanks. Klepikov (1989) and USACE (1990) are general documents for various structures. API 653
(1995) includes an appendix for the evolution of tank bottom settlement, (appendix B). The
following points should be noted from API 653 (1995):

• Uniform settlement: This component often can be predicted in advance, with sufficient
accuracy from soil tests, and does not induce stresses in the tank structure. However,
piping, tank nozzles, and attachments must be designed with adequate consideration to
prevent problems caused by such settlement.

• Planer tilt: This type of settlement could affect tank nozzles which have piping
attached to them. The tilt will cause an increase in liquid level.

• Outline settlement of the shell: Use the following formula to calculate the maximum
δ = ( L2 × ε × 5.5) /( H ). δ =
allowable outline settlement: shell y
Where: shell deflection
in meter, L= arc length between measurement points in meter. It is equal to 6 m for
ε =
large tanks (Figure 2-6) and equal to 3.53 m for small tanks. y yield strain,
(conservatively equal to 0.001), and H=tank height in meters. According to this
formula the allowable outline settlement between adjacent measurement points is equal
to 10 mm for large tanks with length of 6 m is equal to 0.8 mm and with length of 3.53
m on small tanks.
• Bottom-Edge differential settlement: The maximum allowable bottom-edge settlement
δ =B
is shown in Figure (2-7). Conservatively bottom − edge ew
could be assumed and it
means areas with bottom lap welds approximately parallel to the shell.
• Bottom-Center differential settlement: Use the following formula to calculate the
maximum allowable bottom plate settlement:
δ bottem−center = 0.031× R .

Figure 2-7. Allowable Bottom-edge differential settlement, (From API 653 (1995), Appendix-B).

(D) Case histories data presented by D'Orazio and Doncan (1987) are very valuable because
they are real data of settlement for steel tanks which are used to store oil materials. D'Orazio and
Duncan (1987) wrote a paper on differential settlement of steel tanks which has been extensively
cited by design codes and guidelines. Based upon 31 case histories and finite element simulation,
D'Orazio and Duncan (1987) concluded that allowable bottom settlement of steel tanks depends on
the shape of the deformation. Steel tanks could deform into 3 profiles, as shown in Figure (2-8).
Tanks with settlement shapes of a settle most at the center, and their settlement decrease smoothly
 Vol. 13, Bund. B 5

toward the edge. Tanks with settlement profile of Shapes B have relatively flat interiors, with
settlements decreasingly rapidly toward the tank wall. Tanks with settlement profile shapes of C
settled most at locations about 2/3 of the distance from the center to the edge of the tank. Shape A
is the least severe with respect to distortion, and Shape C is the most severe. Because of the
importance of settlement profile shape and resulting bottom distortion, it is of interest to examine
what factors control the shape. From this report, a number of useful facts may be noted:

• Tanks may be stable even though the minimum factor of safety against undrained
failure is less than unity. This has occurred in cases where loading was slow, or
drainage was rapid, or both.

• Tanks with Fmin (based on undrained strength) greater than 1.1 and De / T < 4 , had
settlement profile A, (De is the effective diameter of the tank, De=D+Tp, D=Actual
diameter of tank, T=Thickness of clay layer and Tp= Thickness of any granular layers
or compacted clay pad between the base of the tank and the top of the clay layer
beneath the tank).

• Most tanks with Fmin (based on undrained strength), greater than 1.1 and De / T > 4 ,
had settlement profiles shape B.

Figure 2-8. Differential Settlement Profiles of Bottom Plate of Steel Tank

• Most tanks with Fmin (based on undrained strength), less than had settlement profiles
of shape C, whenever the tank was stable or unstable when filled.

Using information in this paper, criteria have been selected tolerable amount of differential
(δ bottom −center / D) = 0.025 ;
settlement, as follows: profile shape A profile shape B
(δ bottom −center / D ) = 0.015 ; and profile shape C (δ bottom −center / D) = 0.005 .

It should be noted that large tanks of tank farm (D=53.6m, H=18.3m) and small tanks of the
T = 2 .5 m
P.L.D area (D=9m, H=8m), confirm the profile A, shown in Figure (2-8), because p and
depth of clay layer is more than 50m. Therefore tolerable settlement could be considered.

Iran Khak Company (IKCE, 2006) performed a geotechnical investigation in the P.L.D area of
this project. They recommended total measured settlement should not exceed 1.4(mm week) in
 Vol. 13, Bund. B 6

the last week of hydrostatic test of the tanks. By this criteria, it is ensured that much settlement will
not take place after the end of hydrostatic test of tanks. However this rate might be used to estimate
total allowable settlement. It is not necessary to add that settlement rate much faster at the
beginning of hydrostatic test. Conservatively, it could be assumed that, settlement rate of each
week is equal to 60% of settlement rate of its past 7 days. By this conservative assumption, total
settlement in 120 days of hydrostatic test is about 150(mm). It should be noted that, total settlement
of tanks happens in years. So the measured settlement in 120 days of hydrostatic test is less than
total settlement of the tanks in their service period. Therefore the recommended value of IKCE
(2006) could be used only for hydrostatic test.

COMPARISON OF CITED ALLOWABLE SETTLEMENTS

FOR STEEL TANKS
According to references cited in Section (2-2), a comparison of cited allowable settlements for
the steel tanks of Mahshahr oil product terminal revamp project is presented in Tables (2-1) to (2-
3).

Table 2-1. Comparison of Allowable Settlements from Different References

Total Settlement (mm) Differential Settlement Tilt
(Figure2-
δbottom δshell w (Figure2-3)
(Figure2-2)
Reference

Tank type

Δ max
1)
Δ ave

Visib
Cent

Cent

Outli
ure2

ure2

ure2
Edg

Edg
(Fig

(Fig

(Fig

mat
Ulti
-5)

-4)

-6)
ne
er

er

le
e

e
API 653 Large Figure2-
- - - 0.031(R) 0.0055(L2)/H - -
(1995) Small 7
Klepikov Large - - 180 0.004(D) - 0.01(L)
0.004(H) 0.007(H)
(1989) Small - - 110 0.008(D) - 0.008(L)
USACE Large
- - - 0.008(R) - - - -
(1990) Small
D'Orazio
and Large
- - - 0.025(D) - - - -
Duncan Small
(1987)

As it can be seen in Table (6-1), the allowable settlement depends on the dimensions of steel
tanks. Therefore different values could be calculated for large tanks in the tank farm area
(D=53.6m, H=18.3m) and small tanks in P.L.D area (D=9m, H=8m). Tables (2-2) and (2-3)
compare various values for large and small tanks respectively.

Table 2-2. Allowable Settlements from Different References for large tanks of the tank farm.

Differential Settlement
Total Settlement (mm) Tilt (mm)
(mm)
Δ max δbottom δshell w
Reference: Δ ave
Center* Edge Center Edge Outline Ultimate Visible
API 653
1106 - - 830 170 10 - -
(1995)
Klepikov
285 - 180 214 - 60 128 73
(1989)
USACE
285 - - 214 - - - -
(1990)
 Vol. 13, Bund. B 7

D'Orazio
and
1786 - - 1340 - - - -
Duncan
(1987)

Table 2-3. Allowable settlements from different references for small tanks of the P.L.D area.
Differential Settlement
Total Settlement (mm) Tilt (mm)
(mm)
Δ max δbottom δshell w
Reference: Δ ave
Center* Edge Center Edge Outline Ultimate Visible
API 653
186 - - 139.5 75 2.4 - -
(1995)
Klepikov
- - 110 36 - 60 56 32
(1989)
USACE
48 - - 36 - - - -
(1990)
D'Orazio and
Duncan 300 - - 225 - - - -
(1987)
*Conservatively assumed as: δ bottom −center = 0.75(Δ max −center )

PROPOSED ALLOWABLE SETTLENEMTS

API 653 (1995) and D'Orazio and Duncan (1987) recommended values which are more related
to large tanks used for oil material so they could be mainly used to select allowable in this project.
Based upon Tables (2-1) to (2-3) the allowable settlements are conservatively proposed in Table 2-
4 for the steel tanks of Mahshahr oil product terminal revamp project. Although API 653, (1995) is
the most used references for the design of steel tanks, it may gives fairly un-conservative values for
this project. The writer recommends API values to be used.Therefore about half of the API values
for allowable total and differential settlements, as shown in Table 2-4, have been conservatively
considered for presented design. Allowable tilt is proposed based upon visible limit of tilt.
Table 2-4. Proposed Allowable Settlements for Steel Tanks in Mahshahr Oil Export Port.

Tape of Total Differential Tilt

D(m) H(m)
Tank Settlement (mm) settlement (mm) (mm)

Large 53.6 18.3 500 375 73

Small 9 8 100 75 32

Allowable settlement for ordinary buildings, such as residential and office buildings, water
towers and shelters, is more restricted but the proposed values are commonly agreed in the
literature and text books and Table (2-5) is used in different projects.
Table 2-5. Proposed Allowable Settlements by Skempton and MacDonald for Conventional
Buildings.

Type of Soil Type of Foundation Differential Settlement Total Settlement

Sand Isolated 25 mm 40 mm
Sand Raft 25 mm 40~65 mm
Clay Isolated 40 mm 65 mm
 Vol. 13, Bund. B 8

Clay Raft 40 mm 65~100 mm

Although clayey soil exists in site but considering that it is likely that a layer of compacted
gravel and sand to be constructed beneath the foundations so the allowable total settlement
presented in Table (6-6) have been considered for buildings. Where a large loaded area is founded
on a relatively incompressible stratum (e.g. dense gravel) overlying compressible soil, settlement of
the structure will occur due to the consolidation of the latter layer, but it will not take the form of
the bowl-shaped depression. The effect of the dense layer, if thick enough, is to form a rigid raft
which will largely eliminate differential settlement. Therefore, the allowable settlement of ordinary
buildings is suggested in this project as if they are founded on sandy layers.

Table 2-6. Proposed Allowable Settlements for Buildings in P.L.D Area.

Type of Foundation Total Settlement

Isolated 40 mm
Raft 65 mm

For pump stations and heater, no allowable settlement is recommended by design codes like
API 610, but it is noted that the differential settlement values should be very small. According to
definitions of NIOEC-SP-00-01 (2006) pumps in this project are heavy machinery because total
weight of as pumps is greater than 23kN. It suggested that the weight of the heavy rotary
machinery foundation like the pumps in Mahshahr project shall be at least 3 times the weight of
machinery. Therefore the pump itself should be placed on a fairy rigid raft foundation to avoid
relative movement between pump supports. However tilt and total settlement could be a problem
for the connections of pump and pipes so the use of expansion loop or expansion joint or both of
them is highly recommended. Design of such expansion loop needs flexible analysis. Therefore the
allowable settlement of pumps is a function of pipe-pump connections design, but it assumed as 2
cm for foundation selection and land reclamation design.

REFERENCES
1-API 653, Appendix-B. TENTH EDITION, NOVEMBER 1998, ADDENDUM 1, JANUARY
2000 ADDENDUM 2, NOVEMBER 2001

2-Timothy B.D’orazio, A.M. ASCE and James M. Duncan, F. ASCE (1987) "Differential
settlements in steel tanks"journal of Geotechnical Engineering,vol. 113,NO 9,pp 967-83

3- D’Orazio T, Ducan JM, Bell RA. (1989) Distortion of steel tank due to settlement of their walls.
Journal of the Geotechnical Engineering Division ASCE 115(6):871–90.

4-T.Y. Wu, G.R. Liu (2000)" Comparison of design methods for a tank-bottom annular plate and
concrete ringwall", International Journal of Pressure Vessels and Piping,NO 77,pp. 511-517

5-L.A. Godoy, E.M. Sosa (2003) "Localized support settlements of thin-walled storage tanks"Thin-
Walled Structures,NO 41,pp 941–955

6- Brown GD, Peterson WG. (1964) Failure of an oil storage tank founded on sensitive marine
clay. Canadian Geotechnical Journal;1:205–14.
 Vol. 13, Bund. B 9

7- Green PA, Hight DW (1964) The failure of two oil storage tanks caused by differential
settlement, Proc. of the Conf. Settlement of Structures, British Geotechnical Society, Cambridge,
UK, 353-60.

8- Clark JS (1969) Survey of oil tank failure. Annales de l’Institute Belge du Petrol; 6:15–24.

9- Myers P. (1997) Aboveground storage tanks. New York: McGraw-Hill

10-D'Orazio, T. B and Duncan J. M. (1982) “CONSAXA: Computer Program for Axisymmetric

Finite Element Analysis of Consolidation." Research report No UCB/GT/82-01, Dept. of Civil
Engineering. Univ. of California, Berkeley, Calif

11-Kamyab H, Palmer S C (1991) Displacements in oil storage tanks caused by localized

differential settlement. J Pressure Vessel Technol, Trans ASME; 113:71–80.

12-Kamyab H, Palmer S C (1989) Analysis of displacements and stresses in oil storage tanks
caused by differential settlement. J Mech Eng Sci, Proc IMechE Part C 203:61–70.

13-Teng J G (1996) Buckling of thin shells: recent advances and trends. Appl Mech
Rev;49(4):263–74.

© 2009 ejge