Society of Petroleum Engineers

SPE 26558

A New Look at Sucker Rod Fatigue Life

N.W. Hein Jr., Conoco Inc., and D.E. Hermanson, LTV Energy Products (Retired)
SPE Members

II
Copyright 1993, Society of Petroleum Engineers, Inc.

This paper was prepared for presentation at the 68th Annual Technica! Conference and Exhibition of the Society of Petroleum Engineers held in Houston, Texas, 3-6 October 1993.

This paper was selected for presentation by an SPE Program Committee folloWing review of information contained in an abstract submitted by the author(s). Contents of the paper,
as presented, have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material, as presented, does not necessarily reflect
any position of the Society of Petroleum Engineers, its officers, or members. Papers presented at SPE meetings are subject to publication review by Editorial Committees of the Society
of Petroleum Engineers. Permission to copy is restricted to an abstract of not more than 300 words. Illustrations may not be copied. The abstract should contain conspicuous acknowledgment
of where and by whom the paper is presented. Write Librarian, SPE, P.O. Box 833836, Richardson, TX 75083-3836, U.S.A. Telex, 163245 SPEUT.

ABSTRACT the development of new allow,ible

stress curves for used rods.

The ability of a sucker rod to last

in a well lifted with beam pumped INTRODUCTION
equipment depends on many factors.
One of the principle factors is the
fatigue life of the rod. Previous A sucker rod string used in beam
rod life expectations are based on pumped wells is considered the
the modified Goodman Diagram that connecting link between the surface
resulted in the expected life of pumping unit and the subsurface,
approximately 10 million cycles. downhole rod pump. The sucker rod
These expectations are based on string transmits the reciprocating
fatigue work that is approximately motion generated by the pumping unit
fifty years old. This paper will to activate the pump and lift the
discuss the major factors that fluid to the surface. The rods in
affect fatigue life, the the rod string have to be selected
improvements in the quality and and the string designed to withstand
manufacturing processing of rods, the loads generated from the
and the use of other stress range upstroke and downstroke motions.
diagrams over the last number of
years. These combined factors have The upstroke loads normally consist
now resulted in the expected fatigue of the weight of the rod strin'J in
life of steel sucker rods of the fluid in the tubing, the fluid
approximately 50 million cycles. load over the gross plunger area and
Additionally, discussed in this the dynamic effects on the upstroke.
paper are the application of fatigue The loads on the downstroke include
life for used steel sucker rods and the weight of the rods in fluid
References at end of paper minus the dynamic effects on the
*Retired downstroke. Compounding the loading

439
2 A NEW LOOK AT SUCKER ROD FATIGUE LIFE
on the rods further are the time
changing of the loads due to net
lift effects as the well fluids are
pumped down and the dampening
effects of the system, along with
unaccounted friction effects due to
the produced fluids, any well bore
deviation, buckling of the rods
and/or tubing, etc. The net effect
is that the loads imposed on the
rods fluctuate and are repeated
many, many times until a failure
occurs. Loading of this nature is
called "fatigue loading".
Failures of fatigue loading are
characterized by two features; the
first is the development of a crack
normal to the plane of the load, and
the second by a relatively sudden
fracture. These failures occur even
though the stresses are below the
tensile strength and, normally, even
below the yield strength of the
material. Often the fatigue crack
initiates at a point of high stress
concentration. 1 For steel sucker
rods, this point may be a small,
preexistent crack or flaw, an
inclusion, or a corrosion pit.
There are many factors that effect
the fatigue life of a material.
This paper will specifically address
the fatigue of steel sucker rods and
the major factors that will
influence the life and acceptable
stress that a sucker rod can
withstand in a beam pumped well.
Additionally, new performance
estimates and allowable stress
ranges for used steel sucker rods
will be developed.
DISCUSSION
Modified Goodman Diagram
440
SPE 26558
Original work on fatigue strength of
materials was addressed over 60
years ago. 2,3 One of the key
developments of these works was the
Goodman Diagram. This diagram
compares any condition of
alternating load or stress between
the extremes of complete reversal to
a static stress equal to the tensile
strength of the material. A curve
(or curves) generated usually shows
the expected stress range for a
given life or cycle time of the
material. An example of a generic
Goodman diagram is shown in Figure
1.
The potential use of the Goodman
diagram for selection of maximum
allowable steel sucker rod stress
and stress range were recognized in
the 1950 IS. 4 This diagram showed
that for every minimum load, there
was a maximum load to provide the
required cycle life. Additionally
shown was the smaller the load
range, or difference between the
maximum and minimum load, the higher
the maximum allowable load can be.
The basic diagram was developed
using short, polished metal
specimens to reduce the variables
and achieve relatively consistent
results. This was done in a short
period of time by running tests at
very fast cycle rates (approximately
1750 rpm) to develop an endurance
limit/fatigue life expectation of 10
million cycles.
Fatigue diagrams normally assumed an
infinite life if the threshold of
10,000,000 cycles is achieved. This
contributed to the practice of
stopping fatigue tests since the
assumption was that the specimen
will not fail even upon continued
loading. In an oil well, the
SPE 26558 N. W. HEIN, JR. AND D. E. HERMANSON 3

environmental conditions usually discussion, a consensus was reached

resulted with corrosion pits, that the Y-intercept should have a
handling problems and stress factor of safety of two. This
concentration areas that do not reduced the intercept to the tensile
allow infinite life. It generally strength divided by four.
was accepted from field observation
that the original life expectations The remaining issue that had to be
will apply to sucker rods. settled was the factor of safety on
The sucker rod is a hot-rolled bar the tensile strength apex. Everyone
and does not have a polished seemed to have a different opinion.
surface. The bar is not machined Values ranged from 1-1/2 to in
perfectly straight, so bending excess of 2, with positions in
stresses, in addition to tensile, between. Finally, after much
can occur. Furthermore, the rod can deliberation, an agreement was
be subj ected to dents and surface reached. An arithmetical average of
discontinuities caused by handling, all the proposed safety factors was
and to corrosion pits. Both of calculated. Since the resulting
these can cause added stress average was reasonably close to 1-
concentrations in the rod. 3/4, it was recommended that this
value be used. The results from
Application of the diagram from the this agreement was the development
short, polished specimens to sucker of the "modified Goodman Diagram"
rods caused factors of safety to be that has been/still is published in
applied. Sucker rod manufacturers the API recommended practice (RP) on
handled the factor of safety problem care and handling of sucker rods. 5
in a variety of ways. Some The expected fatigue life was still
manufacturers used the yield assumed to be 10 million cycles.
strength of the material instead of
the tensile strength for the apex.
Various factors of safety were used Fatigue Factors and Effect
for both the tensile strength and
left hand portion of the diagram It has been over 30 years since the
instead of using the Y-intercept of API committee reached their
one-half of the tensile strength. consensus agreement, and the diagram
Each manufacturer vigorously was published in API Recommended
defended his company's approach. Practice 11BR. During this time
period, many changes have been made
In an effort to bring some in sucker rod steel chemistry and
consistency in the application of manufacturing. In addition, the
the Goodman diagram to sucker rods, knowledge bank in field performance
an American Petroleum Institute for steel rods has dramatically
(API) task group was formed to improved, as well as a better
investigate the situation and understanding of the parameters that
develop a consensus on its use, if influence fatigue performance. A
possible. The committee met in the review of these conditions and their
early 1960' s at the Mayo Hotel in impact suggest a new application
Tulsa, Oklahoma. It did not take diagram to take advantage of
long for the members to reach increased performance and an
agreement that the apex of the increased fatigue life.
diagram should be the tensile
strength. After considerable
441
4 A NEW LOOK AT SUCKER ROD FATIGUE LIFE SPE 26558

The first thing to consider is that improvements should resul t in the

current sucker rod bar stock is a minimum cycles to failure increase
hot rolled product. Thirty years from 10 million cycles (over 30 to
ago these bars were produced by 60 years ago) to over 50 million
major steel mills from ingots. cycles today. When these cycle
Today, domestic manufacturers use times are applied to a typical beam
mini-mills with continuous casting pump operation producing
and up-graded rolling processes. continuously at 10 strokes per
This produces a smoother, more minute, for 24 hours a day, and for
consistent rod product with less 12 months in the year, the expected,
inclusions or discontinuities. theoretical fatigue life will have
Additionally, improved workmanship risen from about 23 months for 10
requirements were developed by an million cycles to 115 months for 50
API task group approximately 10 million cycles.
years ago. This provided tighter
requirements for straightness, A verification of these improvements
smoother surface, and limited can be obtained from other
surface discontinuities. Current applicable literature .11,12 These
day rod manufacturers also have papers discuss the development of a
better control on the sucker rod system and tracking of production
chemistry, increased cleanliness of operations equipment life. The
the steel, finer grain size, and results from these papers show that
better heat treatment requirements current day steel sucker rods can
to produce the final strength obtain a mean time to failure in
requirement for a grade of rod. operation before the first failure
is experienced improvement from 20
The comparative effect for some of months to 75 months. Although the
these improved requirements on steel anticipated minimum for 50 million
sucker rods can be seen in fatigue cycles has not been reached, the
curves or descriptions in a number improved performance shows that the
of text books. 6,7,8 Figures 3 thru 6 old reference of 10 million cycles
show the effect of some of these is no longer applicable.
factors on the actual fatigue Additionally, with improve steel rod
performance of steels similar to or manufacturing, improved/tighter
actually from sucker rod bars. specification requirements, and
Figure 3 shows the effect of surface better tracking/operation of
roughness. Figure 4 shows the production failures, the theoretical
effect of inclusion size. Figure 5 minimum fatigue life may be able to
shows the effect of cycle rate and be obtained. It also needs to be
environmental condition. Figure 6 mentioned that this theoretical
shows the effect of environmental minimum may not be obtained in the
condition on steel sucker rods with field since our operating practices
and without proper corrosion will prevent this from occurring.
inhibition. As an example, this value assumes
perfect corrosion inhibition, with
All of these factors, when properly no excessive-unaccounted for loads
applied, will result in improved such as from pounding fluid or
fatigue life. Specifically, for tapping bottom.
current day steel sucker rods, these

442
SPE 26558 N. W. HEIN, JR.
Alternate Design Consideration
Although the Modified Goodman
Diagrams is considered a useful tool
to determine the maximum allowable
stress for a given minimum stress,
it is also recognized that it is a
very conservative theory of the
effect of mean stress on fatigue
life. More recent fatigue work and
experiments have shown that the non-
linear theories do a better job of
predicting failure since they better
account for the data. 6,7,8,13
The one theory that has been most
widely used is the Gerber Parabolic
Relation. The generalized curve for
this relation and a comparison to
other linear, theories is shown in
Figure 7. This curve should allow
increased maximum allowable stresses
for the same given minimum stress
when compared to the modified
Goodman diagram. However, it is
generally not used unless all the
applicable strengths are accurately
known. The following are the usual
forms for comparing these
relationships.
Modified Goodman Theory:
Sa= S[l- (Sm/Su)]
Gerber Theory:
Sa= S [1- (Sm/Su) 2]
where Sa is the alternating, maximum
allowable stress associated with the
mean stress Sm, S is the alternating
fatigue strength, and Su is the
tensile strength. More fatigue data
needs to be developed for steel
sucker rods in the current industry
manufacturing and quality
requirements to develop a better S.
However, the currently published
data, such as in Figure 6, may be
443
AND D. E. HERMANSON 5
used to provide the required fatigue
strength. Coupling this factor with
the other known parameters should
allow higher loads and associated
stresses to be put on the steel rods
while still maintaining the expected
50 million CYCle, minimum fatigue
life.
Used Sucker Rods Consideration
The use of used sucker rods when
fatigue life is considered becomes a
complex situation. The problem with
using used sucker rods is that we do
not know "how used are they?"
Normally records are not kept
showing or tracking the number of
cycles each rod in a rod string has
to know when it may be approaching
the fatigue limit. Additionally,
operating abuses such as over-
stretching the rod string when
trying to unseat the pump, buckling
rods due to over-pumping and/or not
properly setting tubing anchors or
not using rod centralizers, allowing
corrosion pits to occur, etc., will
all reduce the allowable life for a
rod.
In order to obtain the most return
on the investment for buying sucker
rods, some operating companies will
have their used rods inspected. The
acceptable used rods are usually
placed back in service while the
rejected rods are usually down-
graded or scrapped. Other operating
companies just place limits on the
number of failures in a rod string
section before replacing the
section. The typical number of
failures for a section is three in a
year's period of time. However,
what is usually found after
replacing the section that has been
failing is that another section
(usually the next size larger rods)
starts to fail. When these rods
6 A NEW LOOK AT SOCKER ROD FATIGUE LIFE SPE 26558

fail, they start damaging the new related failures in the rod string
section of ~ods that was just in a 12 month period of time.
replaced. These failures can
usually be associated with the
allowable stress range and fatigue CONCLUSIONS
life for each section of rods. The
next size rods start to fail since
they have been in the well longer 1. The modified Goodman diagram
and are exposed to produced fluids, and associated 10 million cycle
operating practices, etc. that will fatigue limit does not
prevent them from lasting. represent current sucker rod
capabilities.
Since used sucker rods do not have
odometers like automobiles to keep 2. Current day sucker rods have
track of the number of cycles a rod improved fatigue strength due
has received nor can rod inspection to improved manUfacturing,
tell the fatigue experience or processing, and specification
remaining life of a sucker rod, new requirements and could resul t
allowable stress curves have been in approximately 50 million
developed. These are based cycles before failure.
primarily on decreasing the
allowable stress associated with the 3. The Gerber parabola theory for
acceptable rod diameter loss found estimating allowable stresses
in the API RP IlBR for the three should be considered for
classes of rods. current day sucker rods if the
other required stresses are
Figures 8 and 9 present the accurately known.
recommended curves for API grade C
and D rods, respectively. API K 4. Used sucker rods should not be
rods can use the C rod curves, if K expected to last as long as new
rods are used. These curves will sucker rods for the same
allow used, inspected rods to be application and require
used in current string designs derating depending on
provided that the allowable stresses inspection class for the
are still below the derated values. appropriate grade of rod.
If the derated stress is below this
value, used rods should not be used
for this application. It should be ACKNOWLEDGEMENTS
noted that derating the used rods
will not show when the rod string
should be replaced. It is The authors thank their respective
recommended that in order to management for permission to publish
determine when the entire string this paper.
should be replaced, a full-cycle,
economic analysis needs to be done. 14
In lieu of doing these full cycle REFERENCES
analyses, it has generally been
found that the entire rod string
should be replaced after three 1. Polakowski, N. H. and Ripling,
E. J.; Strength and Structure
444
SPE 26558 N. W. HEIN, JR. AND D. E. HERMANSON 7

of Engineering Materials, 10. Chittum, J. F. , "Corrosion

Prentice-Hall, Inc., New Fatigue Cracking of Oil Well
Jersey, 1966. Sucker Rods, " Materials
Protection, National
2. Goodman, J.; Mechanics Applied Association of Corrosion
to Engineering, 9th Ed., Engineers, Houston, (April,
Longmans, Green, and Co. , 1968), 30-34.
London, p. 634.
11. Bucaram, S. M. and Sullivan, J.
3. Kommers, J. B.; "Effect of H.; "A Data Gathering and
Range of Stress and Kind of Processing System to Optimize
Stress on Fatigue Life," Producing Operations," Journal
Proceedings ASTM, Vol. 30, of Petroleum Technology,
1930, Philadelphia, p. 272. (February, 1972), 185-192.
4. Sucker Rod Handbook, Handbook 12. Bucaram, S. M. and Yearly, B.
489, Bethlehem Steel Company, J.; "A Data Gathering System to
New York, 1958. Optimize Producing Operations:
A 14 Year Overview," Journal of
5. API Recommended Practice 11BR; Petroleum Technology, (April,
"Recommended Practice for Care 1987), 457-462.
and Handling of Sucker Rods,"
American Petroleum Institute, 13. Shigley, J. E. and Mitchell, L.
Washington, D. C. , Eighth D.; Mechanical Engineering
Edition, October 1, 1989. Design, McGraw-Hill Book
Company, New York, Fourth
6. Rice, R. C. , ed. ; Fatigue Edition, 320-325.
Design Handbook AE 10,
Society of Automotive 14. Powers, M. L. i "Optimization of
Engineers, Inc. , Warrendale, Sucker Rod Replacement," SPE
Pa. , Second Edition, 1988. 3470, presented at the 46th
annual meeting, New Orleans,
7. Boyer, H. E., ed.; Atlas of 1971.
Fatigue Curves, American
Society for Metals, Metals
Park, Ohio, 1986.
8. Boyer, H. E., ed.; Metals
Handbook - Failure Analysis and
Prevention, American Society
for Metals, Metals Park, Ohio,
Eighth Edition, Vol. 10, 1975.
9. Bardes, B. P., ed.; Metals
Handbook Properties and
Selection: Irons and Steels,
American Society for Metals,
Metals Park, Ohio, Ninth
Edition, Vol. 1, 1978.

445
 ~--------1.0 ;:.E ..!.T-'=E~N:.::S~IL:lE=_S~T.:.,;R""E=_N=GT~H____"!S~u'__ _::aB

O.ll

Z ('.
0
~
~'
iii
z ~~
w
..... ~~
C ~"
,e,~
~~
.$
0.2

LINE OF" ZERO STRESS

0
z
0
iii Above diagram is meant to show a trend
UJ
w only. As the maximum stress increases
a: the permissible range of stress must
ll.
::l: decrease.
0
U
0 ----- 0.33

Fig. 1-General Goodman diagram showing linear relationship

between allowable load/stress, mean stress, and stress
range (Ref. 4).

...--~so
o
S, =({- + M S"" ) SF •• __ •••••••••••••••••••••••• I
S, = (0.Z5T
6$" = SA -
+Sv.IN
0.5615 S"" ) SF.••••••••••••••••••.••• 2
__
~ • • __ • 3

Where:
S, = MAXIMUM AVAILABLE STRESS, PSI (N/mm')
6S, = MAXIMUM ALLOWABLE RANGE OF STRESS, PSI (N/mm'l
M = SLOPE Of S, CURVE = 0.5625
S",,= MINIMUM STRESS, PSI IN/mm'l ICALCULATEO OR MEASURED)
Sf =~ SERVICE fACTOR
T = MINIMUM TENSILE STRENGTH, PSIIN/mm')

Fig. 2-Modified Goodman diagram for sucker

roads in noncorrosive service (Ref. 5).

446
 1000
900 .... ---

~ 600
800
700
~
~
-'-=-----~ t"---- (
I"--...
-

round
100 rn
90
x
vi
vi
rn
~
U;
"0
~ 400
500 ~~
~ ~otrolled ----- r---r--::-r--
;:--.... r----
Machinecy
80 rn
70
60
~
U;
"0
Q)

~
Q)

~
~~
Q)
> F::::: 50
>
~
~ >-
Q)

~ 300 - ....... 40 a;

AS-t:> ~~
a; a.
a. E
0
E u
o 30
U 200

150
""" ~ 25

1 2 5 10 20 50 100 200 500 1000

Life. 1000 cycles

Fig. 3-Surface condition effect on the fatigue properties of a high-strength steel (Ref. 7).

1100 ---
o Small inclusions

"'-- • Large inclusion~

~
1000 e------------
M 140
0-
::;:
I-...
~ 900

.g'"
E 800 1--
"""- ~ ~ I---. 1

~ ~.
$
~

700 100

Number of cycles to failure

Fig. 4-Fatigue life of AISI-SAE 4340 steel showing effect of inclusion

size (Ref. 9).

50 -.:.
I: ~ ~t. i I
Eo
E 40
f' .. ~

-, ~ t
~ ~ ,"-- i·
~!+
_1-

~ 30
.~ 30 cpm ,-:" "
,
-'Q
,..:;
::
a.
E
«
20 • Air
m
• 3 0 Sail Water + :I Cathodic Protection
6

-~
u-

rn
6 3% Salt
6 cpm
Water
I, I II .... ~
~ 10

I Ilf~: ~~II~ f~\:~I+' ~~I;lodiC TtTT

~.~
Ci5
o
lQ4 10'
Number of Cycles to Failure

Fig. 5-Fatigue life effect of cycle rate, corrosion pro-

tection, and nonprotected on as-rolled SM50A
(0.17 C-1.35 Mn-0.35 Si) steel (Ref. ~).

447
 10.000..------;=-------------------,

10.000

.....
... 50.000
...
til
It:
l-
en 40,000
Q ~---B
til
::i
:: 30.000 F
4

AS - IN AIR
20.000 CD- IN ACID BRINE
EF - IN ACID BRINE WITH INHIBITOR
10.000 L..;---.L;---...l...;---...l.-,---l.-:-----l.-;------JI..;--~
102 10 1 10' 10 1 10' 10' 10' 10'
NUMBER OF CYCLES TO FAILURE

Fig. 6-Effect of corrosion and corrosion inhibitor on fatigue

strength of high-strength sucker rod material (Refs.
7 and 10).

Fatigue strength, S

V)~ / Gerber's parabola

~-~ / Modified Goodman line
en
Tensile
strength, Su

Mean stress, Sm

Fig. 7-Generalized theory responses for the effect of mean

stress on maximum allowable stress in fatigue (Ref. 7).

448
 55,000 ::, ~::t .. ,: :.: •.

«
CI)

N
Z

c
.
e
c:
Ll-
e
«
w
c:
~
CI)
In
..J
..J.
c:
c..
c..
II
-c..
C I)

CI).
CI)
w
c:
t-
CI)
W
..J
In
«
~
..J
..J
«

~h :~:,i !j''-..~ j,., . ., I':, ,"" ." I

., ' ':' . -'---1---'-"-'-I',.:c'-+':'rl--'-~+·+4--· --1-- ----'+'-".-I-.-!-'----'-1
10,000 .... ;u.;;..;..;
t;..;! .L-....L:.:. :.~'. .;~.:l.:.i. .;.'. l.' ...;.i:'..;..:L:;"':'..L:!";"~•...J•.:..''.:..:."~'-...:.J.' ...;.".:...."L......L..;..',...t.:.'!...J'IJ..',,;,,:
:..J'...;''..;..i '.:...I'...J: i.l..f:......:..L..l.";.;.".L.:..:....;.1:..;.:!
1.;.;i ~
I..;..' ..L..::~:
L1.Ji:..:.,: i:.;.:...:'.L.:..c:.:.!:I.;..:.;.:.:..L...:.....l--l

-0.4 -0.2 ,0 0.2 0.4 0.6 0.8 1.0

MPRL/PPRL
Fig. 8-Modified allowable stress curve for the three recommended classes of used API Grade C sucker rods.
449
 «
(I) 45,000 1--:~i-+----'- ...• i . { - : Pt-+·
--+--+--+- i -±·.,.,.:r·.f-.-+~JI!/,--+-+---j:f-V++-+:-'-I
N
. i> . \ . I . I:· \ i ' : .,.:: X l J I ...
1 : :_.-1f--..'....I-'--J.--'-.:l-·I ... . I ! . .i :: I :~ " : 1 1 V ; .
~~ --t-- I--+-.
Z 1--';...:.1
. 1
1
1.'.' : ,.-+--+---+-,--+--4.'-.. ·1 : ' 1 · I"
! . : ::. I,. .

L ".
.

.
.7'1.!"i
:
If. - .

~ ~
C
o
a:
LL
o
«
w
a:
~
m 35,000~+- -:~-
(I)
.
' J l " i~ _.L __L_ i J
I : l/I
I
--f ~ II
;. J. ~y !/: , . . .
: ,
:':n? .:'
-oJ
-oJ" i I ~-Ji/ i i ~~iI'-L1./ : ;. I: '.1.·' . '.
-i-j·--+-+-+-+-~-:;~ .. ~-r-I-'-r-c>'/i~~-r: : r
1

a:
! :. V·:· Y r-+ . .i
Q.
Q. ··1 ·1· : I~" . .. .:1.
II .! . :>",,".vI~1
..~
/, - I
,- . ·1·I! I.·· : I·'
30,000 ' I ." ... , ., :
(I)
Q.
(I)"
";:10;.1
';"':l" :.
JJ" . :,.,.. '~ ~ ~.iCl'''vr--
. . ~~~f-~i:- -r -1-
1 I·
__ , i' 1 .
.T
f-~t
(I)
~ ".:',' i/ :'. ~!7: :.:!., I I T .·.1
W

...
a: ;<-- '.: :':1 ·· •. · l / : i .. :~ . i. '1'. :1
(I) :I:
W
-oJ .'!" ;'1 i
m
«
I'.:.
:T!~H~
::·!:::I~:
f--+-'. -~
I··~

I JiI"., .:
II'
i
I
I:
~ .~
i
j;'"
l'.:.. '::'! ~.,.. ,_c!--i- L- .-i-.~-+--+'-'..j-.:..:-.. ........j.:c-l
. -j-'
j' .

§ I.:~._>l :. !...,,~il
~.~ I ..til'
,
t~., ~-l-:~'_c
·1·· ' !
~l---i:--'-+--+--1~t-+--+--l
i
!-:--..
-oJ
-oJ 20 000:' .>: :'! i""" : " ; '. : , : .i ' i:
« , 1:1: I Y
'-:l~j,,£;..I-
1'1"
- --i-' '1
I.
--+...:.. : ' ! ... i l : : :
~i- -I ' I
. i
:..:.:"+'..:..:.+--+-.:..:.+.:.:..+-1
:~ j'. . I· : " I i.':

:. I ! i I' .
: 1. . I I--t,,· --t- ~+'-r~- '4- ~~+-
15,000 --.J..:1--- --+-
I
~-t-,.- .I . ' - ~.
i I. i
-J-'H-
Ii
--L-
.!
-1! . .
:.. iI·.
.....I
" '1 ·'1 I i :.. I I
! : , I. .!.: i·
~I'-'- --+- c--..l.....- c-~ '-",'--' --";-: 1-- ...L_-r--r- --!-_. '--:- ----t-- '--r-'- f - ' ; - _ .

10000
, I';!'
!;.
d:i1
, .. , .
! I.
."
I.

I. .
· 1 . '
I ...
: i'
I : !

.. i .'.
1
-:-I----+--'-,- --ti
i
.
.1':
.:
·t-
::
~I;;-- --';-,'
:j.
II

--T
!
i"':"'" .
.:
H- \

--0.4 --0.2 0 0.2 0.4 0.6 0.8 1.0

MINIMUM POLISHED ROD LOAD/PEAK POLISHED ROD LOAD SM/SA
Fig. 9-Modlfled allowable stress curve for the three recommended classes of used API Grade 0 sucker rods.
450