om

or

c
e.
t
di
!
W

chang
hange E

O
N
Y
U
B
to

d f- x
XC

k
lic
C
F-

.p
PD w
ww

APPENDIX B-EVALUATION OFTANK BOTTOM SETTLEMENT

B.1 Introduction Maximum spacing of 32 It around circumference.

8.1.1 In determining the effects of soil settlement on stor- There must be at least 4 equally spaced diametrical
measurement lines.
age tanks, it is common practice to monitor settlement of the
~ankshell o
tank bottom. In most cases, such a monitoring program is
initiated during the construction and continued during ~11
hydrostatic testing and operations. During operations, set-
tlement measurements should be taken at a planned fre- 10
quency, based on an assessment of soil settlement I
predictions. For existing tanks that do not have initial settle-

-+-
I
ment data, a program of settlement monitoring should be
9
based on prior service history.
8.1.2 If at any time settlement is deemed excessive, the I

tanks should be emptied and reI eve led. Releveling of a siz-

able tank is expensive and rather difficult to achieve. Thus, a 8
I 4~
decision to relevel a tank is a crucial one, and relies very
much on the proper interpretation and evaluation of the moni-
7 5
tored settlement data.
6
8.1.3 Approaches used to correct tank shell and bottom Notes:
settlement include techniques such as localized repairs of the
bottom plates, partial releveling of the tank periphery, and
I. There must he at least eight settlement points. The maximum spacing
of the settlement points is 32 ft around the circumference.
2. Points shall be equally spaced around the tank shell. See 12.5.1.2 for
Ir-
method of determining the number of measurement points.
major releveling of the entire tank bottom. Major releveling
of the tank, involving total lifting of the tank shell and bottom
at one time, can introduce highly localized stresses in the
structure and impair its integrity. Therefore, when choosing Figure B-1-Measurements of Shell Settlement
techniques for correcting settlement problems, an altemative (External)
to total lifting of the tank shell and bottom should be consid-
ered as a first choice. If it is decided to lift the entire tank shell
It may vary in magnitude, depending on the soil characteris-
and bottom at one time, it should be done by personnel with
tics. Uniform settlement of a tank does not induce stresses in
demonstrated experience in this technique.
the tank structure. However, piping, tank nozzles, and attach-
ments must be given adequate consideration to prevent prob-
B.2 Types of Settlement
lems caused by such settlement.
8.2.1 ELEVATION MEASUREMENTS
8.2.2.2 Rigid body tilting of a tank (planar tilt). This com-
The principal types of tank settlement consist of settle- ponent rotates the tank in a tilted plane. The tilt will cause an
ments that relate to the tank shell and bottom plate. These set- increase in the liquid level and, therefore, an increase in the
tlements can be recorded by taking elevation measurements hoop stress in the tank shell. Also, excessive tilting can cause
around the tank circumference and across the tank diameter. binding of peripheral seals in a floating roof and inhibit roof
Figures B-1 and B-2 show recommended locations on a tank travel. This type of settlement could affect tank nozzles that
shell and bottom plate for settlement measurements. Data have piping attached to them. Figure 8-3 shows that the set-
obtained from such measurements should be used to evaluate tled location of the tank shell, after rigid body tilt, can be rep-
the tank structure. Additional settlement readings may be resented by either a cosine or sine wave with respect to its
required to define local depressions. original position in a horizontal plane.

8.2.2 SHELL SETTLEMENT EVALUATION 8.2.2.3 Out-of-plane settlement (d(fferential settlement).

Due to the fact that a tank is a rather flexible structure,
Settlement of a tank is the result of either one, or a combi-
chances are great that the tank shell will settle in a nonplanar
nation of the following three settlement components.
configuration, inducing additional stresses in the tank shell.
8.2.2.1 Uniform settlement. This component often can be The out-of-plane settlements at the bottom edge lead to a lack
or
om predicted in advance, with sufficient accuracy from soil tests. of circularity at the top of the tank, and in the case of a float-
c
e.
t
di
!
W

chang
hange E

8-1
N
Y
U
B
to

d f- x
XC

k
lic
C
F-

.p
w

PD
ww
 om
or

c
e.
t
di
!
W

chang
hange E

O
N
Y
U
B
to

d f- x
XC

k
lic
C
F-

.p
PD w
ww

B-2 API STANDARD 653

it is called the uniform settlement component. A line through

Maximum spacing
/ of 32 It around
this point, parallel to the abscissa, provides a new base or
circumference. datum line for settlement measurements called adjusted
( There must be at settlements.
least 4 equally
spaced diametrical c. The plane of rigid tilt settlement is represented by the opti-
measurement lines. mum cosine curve between the maximum and minimum
values of the actual edge settlements. Several methods exist
Tank shell for determining the optimum cosine curve. The least accurate
o
method is by free hand drawing techniques, a kind of trial and
error procedure to fit the best cosine curve through the data. A
better method is to use the mathematical and graphical capa-
bilities of a computer.
d. The vertical distances between the irregular curve and the
cosine curve represent the magnitudes of the out-of-plane set-
tlements (Vi at data point i).
e. The most commonly used and accepted method is to use a
computer to solve for constants a, h, and c, to find the opti-
mum cosine curve of the form:

Maximum spacing Elevpred = a + h x cos (8 + c)

of 10 It across
diameter.
o Where Elevpred is the elevation predicted by the cosine
Note: See 12.5.1.2 for method of determining the number of 0
curve at angle theta. A typical starting point for a computer
measurement points. best-fit cosine curve is a least-squares fit where a, h, and c are
chosen to minimize the sum of the square of the differences
Figure B-2-Measurements of Bottom Settlement between measured and predicted elevations. The optimum
(Internal) Tank Out-of-Service cosine curve is only considered valid (i.e., accurately fits the
measured data) if the value R2 is greater than or equal to 0.9.
ing roof tank, the extent of the induced ovality may impede
the proper functioning of the floating roof in such a way that (S,.,,-SSE)
releveling is required. Also, such settlements may cause flat S'Y
spots to develop in the tank shell. This type of settlement
could affect tank nozzles that have piping attached to them. where
B.2.2.4 While uniform settlement and rigid body tilt of a Syy sum of the squares of the differences between
tank may cause problems as described above, the out-of- average measured elevation and the measured
plane settlement is the important component to determine and elevations,
evaluate in order to ensure the structural integrity of the shell
and bottom. Based on this principle, a common approach is to SSE sum of the square of the differences between
detennine the magnitudes of the uniform and rigid body tilt the measured and predicted elevations.
component (if any) for each data point on the tank periphery.
Once this is carried out, the plane of rigid tilt is then impor- Obtaining a statistically valid cosine curve may require
tant as a datum from which to measure the magnitudes of the taking more measurements than minimums shown in Figure
out-of-plane settlements. B-1. In many cases, the out-of-plane settlement may be con-
A graphical representation illustrating the effect of tank centrated in one or more areas and the least-squares fit under
shell settlement components is shown in Figure B-3. The con- predicts the local out-of-plane settlement and is not conserva-
struction of this settlement plot has been developed in accor- tive. In these cases, R2 will typically be less than 0.9. Appro-
dance with the following: priately selecting a, h, and c will result in very little difference
between predicted and measured elevation at all but one or
a. The actual edge settlement (in most cases an irregular two measurements points, and R2 will be greater than 0.9.
curve) is plotted using points around the tank circumference One method of selecting a, h, and c in these cases is to ignore
as the abscissa. one or two points that do not appear to fit the initially calcu-
b. The vertical distance between the abscissa and the lowest lated cosine curve, and recalculate the optimum cosine curve.
om
or point on this curve (point 22) is the minimum settlement, and The remaining points will show a good estimate of the true
c
e.
t
di
!
W

chang
hange E

O
N
Y
U
B
to

d f- x
XC

k
lic
C
F-

.p
w

PD
ww
 om
or

c
e.
t
di
!
W

chang
hange E

O
N
Y
U
B
to

d f- x
XC

k
lic
C
F-

.p
PD w
ww
TANK INSPECTION, REPAIR, ALTERATION, AND RECONSTRUCTION 8-3

\
\

Max. settlement at
(PI. 11)
Min.

~I
settlement
(PI. 22)

~I
--4"
il
represented by
Y
Plaoo of og'd ;,,;

optimum cosine curve

Act",1 ,dg'
settlement
~
E
::J
Uniform or minimum settlement:l Cii
o
3 5 7 9 11 13 15 17 19 21 23
2 4 6 8 10 12 14 16 18 20 22 1
N (23) data points, equally (L) spaced apart.

Tank circumference envelope (nO)

Out-of-plane deflection for point "1' is U j = out-of-plane settlement of point "i",

$j= Uj - (1/2 Uj _ 1+ 112 Uj + 1), for example (+) when above cos. curve;
$11 = U11 - (1/2U1O + 1/2Ud (-) when below cos. curve, for example:
U11 = (+)
U13 =(-)

Note: See 12.5.1.2 for detll1ition of "N."O

Figure 8-3-Graphical Representation of Tank Shell Settlement

om
or
c
e.
t
di
!
W

chang
hange E

O
N
Y
U
B
to

d f- x
XC

k
lic
C
F-

.p
w

PD
ww
 om
or

c
e.
t
di
!
W

chang
hange E

O
N
Y
U
B
to

d f- x
XC

k
lic
C
F-

.p
PD w
ww

B-4 API STANDARD 653

out-of-plane settlement at the worst point. Figure 8-4 shows 1.4.--------.--------,--------,--------,

an example where one point at 135 degrees is well off the ini-
tial curve, R2 is 0.87, and S is just less than the maximum 1.2+-~-----+--------+-------~_,~----~

allowed. By ignoring the point at 135 degrees, and recalculat-

ing the optimum cosine curve, R2 is increased to 0.98 (very
good), and the out-of-plane settlement at the point at 135 oS
C 0.8+-------~Hr-------+--~~--~------~
degrees is just above the maximum allowed. Q)
E
f. The vertical distances between the irregular curve and the Q)
:;:::: 0.6+-------~~~~~~------~--------~

optimum cosine curve represent the magnitudes of the out-of- OJ

(fJ

plane settlements (Vi at data point i). Si is the out-of-plane 0.4 + - - - - - + - - - V - - + - - - - - ] - - - - - j

deflection at point i. Refer to Figure B-3.
0.2 +-----+------+-----]---1::..-,;---1
g. Measurements for out-of-plane settlement must be taken
& x
carefully. In cases of distortion or corrosion of the tank bot- O+--------t-------~--~A~--~------~
tom extending beyond the shell, measurements taken near lap o 90 180 270 360
welds in the tank bottom can result in significant errors in Degrees
measured elevation. Repaired or replaced bottom plates, or
new slotted-in bottoms may not have been installed parallel to -+- Measured

the bottom shell course. In some cases, more consistent and Pred (all pts)

accurate results may be obtained by surveying the elevation 1::.. S/Smax (all pts)
Pred (ignore pt 135)
of the weld between the first and second courses.
x S/Smax (ignore pt 135)
Note: When using the optimum cosine curve described in B.2.2.4e,
taking measurements around the shell will result in a more accurate
cosine curve. However, using all of these measurement points in the Figure B-4-0ut-of-Plane Settlement
equation shown in B.3.2 will result in very small allowable settle- Measurement Points
ments, SII1{W since the arc length L between measurement points is
small. It is acceptable to use as many measurement points as needed
to develop the optimum cosine curve, but only use a subset of these
points spaced approximately 30 ft (8 minimum) when calculating S
and SIIJ(Lt. The points used must include the points furthest from the
R = Radius width of settled area
optimum cosine curve. For example, if 8 points are required, but 16 B = Settlement
measurements are taken, and the arc length between measurements
is only 15 ft, calculate the optimum cosine curve using all 16 points, I'l------ R -------1 Use straight edge
but use only 8 points to calculate S. The equations in Figure B-3 to determine the
Tank
would be revised to read: breakover point
shell
ori 9inal /
bottom
position

Lr_ ---------:;;--r---+-'-----,
\
~~~~zzzi..

h. If out-of-measured out-of-plane settlement exceeds the

B

Breakover point
7 Tank bottom

limits described in B.3.2 using the optimum cosine curve

method, a more rigorous evaluation may be performed in lieu
of repairs. This evaluation must be done by an engineer expe-
rienced in tank settlement analysis. R
Tank- Original

,..j~-~~;~rn--t------.,
6.2.3 EDGE SETTLEMENT

6.2.3.1 Edge settlement occurs when the tank shell settles s...!h_e_II__
sharply around the periphery, resulting in deformation of the
bottom plate near the shell-to-bottom comer junction. Figure
-- ----- -~-t{m71Z'Z:".2'Z'11ZZZZZWzn;zza.ti
B
8-5 illustrates this settlement.
I
\
Breakover point
I
L Tank bottom
6.2.3.2 The formula given in 8.3.4 can be used to evaluate
om
or edge settlement. Alternatively, a rigorous stress analysis can Figure B-5-Edge Settlement
c
e.
t
di
!
W

chang
hange E

O
N
Y
U
B
to

d f- x
XC

k
lic
C
F-

.p
w

PD
ww
 om
or

c
e.
t
di
!
W

chang
hange E

O
N
Y
U
B
to

d f- x
XC

k
lic
C
F-

.p
PD w
ww
TANK INSPECTION, REPAIR, ALTERATION, AND RECONSTRUCTION 8-5

be carried out for the deformed profile. Measure the edge set- c. If the tank floor is cone up or cone down, the settlement B,
tlement carefully, taking into account the following: should be measured from a projection of the unsettled floor,
not from level. See Figure B-6.
a. Measurements taken when the bottom is not in contact
with the soil or foundation under the tank can overestimate or 8.2.3.3 The measured edge settlement B is defined as
underestimate edge settlement significantly. If the measured shown in Figure B-5. Bew is defined as the allowable edge set-
settlement is near the maximum allowable settlement, con- tlement in an area where there is a floor lap weld in the settled
sider repeating the measurement with the bottom forced area that is essentially parallel (± 20 degrees) to the shell. Be
down to the soil, e.g., standing on it, or take an additional set is defined as the allowable settlement in an area with no floor
of measurements in the same area, where the bottom is in welds, or only butt-welds in the floor, or lap welds in the floor
firm contact with the soil. that are essentially perpendicular (± 20 degrees) to the shell.
Refer to Figure B-4.
b. Locating the breakover point where the settled area begins
requires some judgment. Placing a straight edge on the unset- 8.2.3.4 Section B.3.4 provides methods for evaluation of
tled floor as shown in Figure B-5, and observing where the measured edge settlement B against allowable edge settle-
floor separates from the straight edge will help define the ment Bew and Be- Since Bew is more conservative than Be, the
breakover point. simplest approach is to initially evaluate measured settlement

Cone down Cone up

bottom bottom

Shell

Shell

Original bottom

Original
bottom

PO:U:"_\_--l~
Settled bottom

Settled bottom 8- Measured from original

position bottom position

8- Incorrect!
Measured from level line from
breakover point to shell

Figure B-6-Correction for Measured Edge Settlement

om
or
c
e.
t
di
!
W

chang
hange E

O
N
Y
U
B
to

d f- x
XC

k
lic
C
F-

.p
w

PD
ww
 om
or

c
e.
t
di
!
W

chang
hange E

O
N
Y
U
B
to

d f- x
XC

k
lic
C
F-

.p
PD w
ww

8-6 API STANDARD 653

B against Bew for all settled areas. If all areas meet this crite-
rion. the settlement is acceptable and no further evaluation is
necessary. If necessary, different settled areas can be evalu-
ated separately against Bew and Be. For areas containing lap
welds at an arbitrary angle to the shell, interpolation to find an
allowable settlement between Bew and Be based on the angle
of the weld to the shell is allowed.

B_2.4 BOTTOM SETTLEMENT NEAR THE TANK

SHELL

B.2.4.1 Figure B-7 illustrates bottom settlement near the

tank shell.

B.2.4.2 The formula given in B.3.3 can be used to evaluate

settlement near the tank shell. Alternatively, a rigorous stress
analysis can be carried out for the deformed profile.

B.2.5 LOCALIZED BOTTOM SETTLEMENT

REMOTE FROMTHETANK SHELL

B.2.5.1 Localized bottom settlement remote from tank

shell are depressions (or bulges) that occur in a random man-
ner, remote from the shell (see Figure B-8).

B.2.5.2 Acceptability of these localized settlements is

dependent on localized stresses in the bottom plate, design
and quality of the lap welds (single-pass or multi-pass), and SECTION A-A
voids below the bottom plate. The formula given in B.3.3 can
R ~ Radius of inscribed circle in bulged or depressed area
be used to evaluate localized settlement remote from the tank B ~ Settlement or depression depth or height of bulge
shell. These limits are applicable to tank bottoms that have
single-pass lap-welded joints. Figure B-7-Bottom Settlement Near Shell

B.3 Determination of Acceptable

to calculate the maximum permissible out-of-plane deflection
Settlement
(see note below.)

B.3.1 GENERAL
2
ISI:s;(L xYxll)
For existing tanks with history of successful service, it may
2[(£ x H)]
be possible to accept greater settlement and distortion of the
foundation from a true plane than new tank construction stan-
where
dards allow. Each tank must be evaluated based on service
conditions, materials of construction, soil characteristics, tank S deflection, in ft (out of plane distortion),
foundation design, and tank service history. The methods dis- L arc length between measurement points, in ft,
cussed in following sections are not mandatory and approxi-
mate the maximum permissible settlement. However, Y yield strength, in Ibf/in. 2,
experience has shown that if settlements exceed the following
£ Young's modulus, in Ibflin. 2 ,
requirements, further assessment or repair is required.
H tank height, in ft.
B.3.2 SHELL SETTLEMENT Note: This fonnula is based on "Criteria for Settlement of Tanks,"
W. Allen Marr, M. ASCE, Jose A. Ramos, and T. William Lambe, F.
From the measurements described in B.2, determine the ASCE, Journal of Geotechnical Engineering Division, Proceedings
maximum out-of-plane deflection. Use the following formula of the American Society of Civil Engineers, Vol. 108, August 1982.
om
or
c
e.
t
di
!
W

chang
hange E

O
N
Y
U
B
to

d f- x
XC

k
lic
C
F-

.p
w

PD
ww
 om
or

c
e.
t
di
!
W

chang
hange E

O
N
Y
U
B
to

d f- x
XC

k
lic
C
F-

.p
PD w
ww
TANK INSPECTION, REPAIR, ALTERATION, AND RECONSTRUCTION 8-7

Figure B-9 is a graphical representation of this formula.

Note: This formula is based on "Criteria for Settlement of Tanks,"
W. Allen Marr, M. ASCE, Jose A. Ramos, and T. William Lambe, F.
ASCE, Journal (~l Geotechnical Engineering Division, Proceedings
of the American Society of Civil Engineers, Vol. 108,August 1982.

B.3.4 EDGE SETILEMENT

B.3.4.1 Maximum allowable settlement Bew is shown in
Figure B-10 for settled areas that include floor lap welds
essentially parallel to the shell (± 20 degrees). In settled
areas where the measured settlement B exceeds 75 percent
of allowed settlement Bew, all shell-to-bottom welds and
floor welds should be inspected visually and with magnetic
particle examination or liquid penetrant examination. All
indications should be repaired, or evaluated for risk of brit-
tle fracture, and/or fatigue failure prior to returning the tank
to service.

B.3.4.2 For settled areas where measured settlement B

exceeds 75 percent of Be,,' any welds within 12 in. of either
side of the breakover area (see Figure B-5) should be exam-
Localized depressions ined visually. Any suspect areas should be examined with
or bulges in tank either magnetic particle examination or liquid penetrant
bottom plate
examination. All indications should be repaired or evaluated

~'
for risk of fatigue prior to returning the tank to service.

~
c:::==::::::;:1 I
B.3.4.3 Maximum allowable settlement Be is shown in
Figure B-Il for areas of edge settlement with no welds, butt
welds, or lap welds in the floor that are essentially radial to
the shell (± 20 degrees). In settled areas where the measured
settlement exceeds 75 percent of the allowed settlement, all
R = Radius of inscribed circle in B = Settlement depth or
bulged or depressed area height of bulge shell-to-bottom welds and floor welds should be inspected
visually and with magnetic particle examination or liquid
SECTION A-A penetrant examination. All indications should be repaired or
evaluated for risk of brittle fracture and/or fatigue prior to
returning the tank to service.
Figure B-8-Localized Bottom Depressions
or Bulges Remote from Shell B.3.4.4 Maximum allowable settlement for areas of edge
settlement with a lap weld at an arbitrary angle to the shell
may be interpolated from Be and Bew from Figures B-8 and
8-3.3 INTERNAL BOTTOM SETTLEMENTS OR
B-9, and the following formula:
BULGES

Measure the bulge or depression. The permissible bulge or

depression is given by the following formula (see note
below). Where a is the angle of the weld to a tank centerline and
Ba is the allowable settlement for an area with a weld at that
BB=0.37R angle (see Figure B-12).
B.3.4.5 In general, settlement occurs slowly, and for most
where existing tanks, the majority of settlement is presumed to have
maximum height of bulge or depth of local occurred in the first few years of service. Significant addi-
BB
tional settlement will not be expected after the initial inspec-
depression, in.,
tions. Therefore, typical practice is to compare the measured
R radius of inscribed circle in bulged area or local edge settlement with the maximum allowable edge settlement
or
om depression, in ft. Bew and Be' and not include allowance for additional settle-
c
e.
t
di
!
W

chang
hange E

O
N
Y
U
B
to

d f- x
XC

k
lic
C
F-

.p
w

PD
ww
 om
or

c
e.
t
di
!
W

chang
hange E

O
N
Y
U
B
to

d f- x
XC

k
lic
C
F-

.p
PD w
ww

8-8 API STANDARD 653

15

14
Repairs or a rigorous assessment
13 should be conducted if the depth
of the observed bulge or
entll 12
depression is above this line
.r:
u
.'=
Q; 11
C
tll
U 10
a;
tll
Q> 9
:::J
..0
0 8
c
.Q
C/l
C/l
tll 7
0..
tll
"0 6 Partial ring-
a type depression,-- --...f
.r: or bulge
15. 5
tll (at edge only)
0 where R< 2R
11
In 4

3
Depression
2
or bulge
(anywhere)

o
o 5 10 15 20 25 30 35 40 45
R = Radius of inscribed circle in depression or bulge (feet)

Figure B-9-Localized Tank Bottom Settlement Limits for Single Pass Welds

ment during subsequent operation. Note that erosion of the than 3/16 in. are smaller than 12 in. in diameter and the thin
pad adjacent to the tank may cause local settlement. In this area does not include a weld.
case the settlement will continue unless the pad is repaired d. Settlement is presumed to be slow, and a small amount of
and future erosion prevented. For cases where significant additional settlement is expected to occur prior to the next
additional settlement is expected, an engineer experienced in inspection.
tank settlement evaluation should evaluate the settlement
8_3.4.7 Edge settlement increases secondary stress at the
expected at the next inspection with the limits in B.3.4. This
bottom-to-shell weld. If weld repairs are made to the bottom-
is analogous to a corrosion allowance for components
to-shell weld in an area where settlement exceeds 1/2 of Be'
expected to corrode.
these additional stresses should be evaluated by an engineer
8.3.4.6 The edge settlement limits described in 8.2.3.1 experienced in tank settlement evaluation before waiving a
through B.2.3.5 were developed for typical 1/4 -in. thick tank hydrostatic test per 12.3.
bottoms with minimal corrosion.
a. Edge settlement limits can be applied with reasonable 8.4 Repairs
accuracy to 5/ 16- and 3/8-in. thick tank bottoms. 8.4.1 If it is determined that settlements have occurred
b. Edge settlement limits can be applied with reasonable which are beyond the permissible limits established in the
accuracy to bottoms with general corrosion, as long as the previous sections, then consideration should be given to mak-
areas near all welds are thicker than 3/ 16 in. ing repairs or a rigorous stress analysis should be performed
c. Edge settlement limits can be applied with reasonable to evaluate the deformed profile. Various repair techniques
accuracy to bottoms with local corrosion, if all locally thin have been discussed above. The judgment on repairs should
om
or areas in the settled area (closer than "R" to the shell) thinner be tempered with knowledge of tank service history, previous
c
e.
t
di
!
W

chang
hange E

O
N
Y
U
B
to

d f- x
XC

k
lic
C
F-

.p
w

PD
ww
 om
or

c
e.
t
di
!
W

chang
hange E

O
N
Y
U
B
to

d f- x
XC

k
lic
C
F-

.p
PD w
ww TANK INSPECTION, REPAIR, ALTERATION, AND RECONSTRUCTION 8-9

10.00

Tanks with larger edge settlements

9.00 are to be repaired, or have detailed -l------+-----+-- Diameter, It
analysis of floor, and floor-to-shell
junction
8.00
Welds in tanks with settlement
'"
Q)
..c
greater than or equal 75% of Bew'
() 7.00 and larger than 2 in., are to be
.'=' inspected with Magnetic Particle or 160 & Larger
iQ)
Liquid Penetrant Examination
E
Q) 6.00 120

~
100
'"
Q)
:0 5.00
l1l
~
.2
'iii
E 4.00
::J
E
'xl1l
::2:
I;, 3.00
Q)
co 20
2.00

Tanks with settlement less than 75% of

1.00 -l-------+-----+-----+-- Bew may be retumed to service ----I

0.00 +---------r--------+--------+-------~r_------_r--------+_------~
o 2 3 4 5 6
R-Radius of settled area, feet

Figure B-1 O-Maximum Allowable Edge Settlement for Areas with Bottom Lap Welds
Approximately Parallel to the Shell

repairs, previous inspections, tank foundation conditions, soil unacceptable settlement should be corrected. Depending
characteristics, the material of construction, and estimates of on the severity and location of the settlement, required
future settlement. See 9.10 for suggested repair details. repairs may include regrading the soil under the tank bot-
tom, and/or repairing the foundation. Jacking and relevel-
B.4.2 For tanks with edge settlement exceeding the lim-
ing the shell may be required to prevent additional
its and assumptions given in B.2.3.1 through B.2.3.7, the
settlement damage. Jacking and leveling are usually done
tank should be repaired. Any plate exceeding acceptable
in conjunction with, not instead of, replacing damaged
strains (typically 2 percent - 3 percent) should be
plate and welds. In lieu of repairs, a detailed analysis of
replaced. Releveling the tank will not remove the plastic
the settled area may be performed by an engineer experi-
strain, so leveling the tank without replacing the strain
enced in tank design and settlement evaluation. The analy-
may not be a sufficient repair. Welds in the area of the high
sis should consider primary and secondary stress and the
strains should be removed and replaced, or be subjected to
risk of brittle fracture.
a fitness-for-service evaluation by an engineer experienced
in tank settlement evaluation. The condition leading to the

om
or
c
e.
t
di
!
W

chang
hange E

O
N
Y
U
B
to

d f- x
XC

k
lic
C
F-

.p
w

PD
ww
 om
or

c
e.
t
di
!
W

chang
hange E

O
N
Y
U
B
to

d f- x
XC

k
lic
C
F-

.p
PD w
ww
B-10 API STANDARD 653

10.00

Tanks with larger edge settlements

9.00 are to be repaired, or have detailed Diameter, ft
analysis of floor, and floor-to-shell
junction
8.00
160 & Larger
Welds in tanks with settlement
(J)
Q) greater than or equal 75% of Bew'
.r:: 7.00 and larger than 2", are to be
u
.~ inspected with Magnetic Particle or 100
EQ)
Liquid Penetrant Examination
E 6.00
Q)

~(J)
Q)
:0 5.00
C\l
:;:
.2
""iii
E 4.00
::::J
E
·x
C\l
::2: 20
I 3.00
Q)
CO

2.00

Tanks with settlement less than 75% of

1.00 +------+-----+-----f-- Bew may be returned to service ----1

0.00
0 2 3 4 5 6
R- Radius of settled area, feet

Figure B-11-Maximum Allowable Edge Settlement for Areas with Bottom Lap Welds
Approximately Perpendicular to the Shell

om
or
c
e.
t
di
!
W

chang
hange E

O
N
Y
U
B
to

d f- x
XC

k
lic
C
F-

.p
w

PD
ww
 om
or

c
e.
t
di
!
W

chang
hange E

O
N
Y
U
B
to

d f- x
XC

k
lic
C
F-

.p
PD w
ww
TANK INSPECTION, REPAIR, ALTERATION, AND RECONSTRUCTION 8-11

Settled

~
~I;hl: area

"- , /----'--.",...'--'--+----'"----'1.

r-~~~--~~---_r~

Tank
centerline

Floor plate
layout

Figure B-12-Edge Settlement with a Lap Weld at an Arbitrary Angle to the Shell

om
or
c
e.
t
di
!
W

chang
hange E

O
N
Y
U
B
to

d f- x
XC

k
lic
C
F-

.p
w

PD
ww
 hange E hange E
XC di XC di
F- t F- t

PD
PD

or
or
! !
W W
O O
N N
Y Y
U U
B B
to to
k k
lic lic
C C

om
om

ww
ww
w c w c
.p e. .p e.
d f- x d f- x
chang chang