Note: Descriptions are shown in the official language in which they were submitted.
- 20~9622
PIPE COUPLING
This invention relates to couplings or connections for
use in interconnecting lengths of pipe casing or tubing made of
steel or the like. In the following description, both terms
"coupling" and "connection" are used, usually interchangeably
without preference, it being understood that the invention
applies equally to threaded and coupled connections and to pin
and box members integral with the pipe and interconnecting
lengths of pipe.
Background, Prior Art
In some applications, steel tubing is subjected to
severe stresses, and where a series of lengths of pipe or tubing
have to be coupled together, the connection or coupling itself
must be able to bear the applied stresses. For example, for use
as oil well casing, such tubing may be used in conjunction with
steam injection into the well where temperatures of the order of
650 degrees Fahrenheit may be reached. This subjects the tubing
to compressive and tensile axial loads approaching or even
exceeding the actual yield strength of the material in the pipe
body. Thus, any connection or coupling for joining together
successive lengths of pipe must be able to withstand the axial
loading without failure and still be resistant to leakage from
internal pressures approaching the actual yield strength of the
pipe body. As the pipe is alternately heated and cooled, the
axial loading on the pipe and couplings may become alternately
- _ 20~9622
compressive and tensile, and throughout the coupling must
maintain its seal with the pipe ends in resisting internal
pressure.
Such couplings (connections) comprise a male and mating
female coupling component. The male component is a suitably
configured threaded portion at at least one end of the steel
pipe, constituting the pin member of the coupling. A mating
annular female component long enough in the axial direction to
receive the pin ends of two adjoining lengths of pipe is
internally configured and threaded at each end for mating
engagement with the pin member, thereby completing the coupling.
The annular female element is often referred to as the box
element or box member of the coupling. Equally, one end of the
pipe could be upset and internally threaded to constitute the
female component of the connection.
Conventionally, the pin member of the coupling or
connection is tapered inwardly from the proximal end of the
threaded portion to the distal end to mate with a similarly
tapered female threaded member of the coupling. The taper
facilitates entry of the pin member into the box member.
In pipe couplings, a seal is typically maintained. The
seal may be effected between the mating threaded portions of the
pin member and the box member of the coupling, but this kind of
seal is subject to ready leakage. In other couplings, some
auxiliary sealing element (e.g. an annular elastomer) is
provided. In yet other couplings of which the present invention
is a species, the axial load-bearing threaded portions of the
coupling do not themselves necessarily provide a seal; the seal
2~9622
is a separate metal-to-metal seal provided adjacent the axial
load-bearing threaded portions, in both the pin and the box
members of the coupling, or male and female elements of the
connection.
The threaded axial load-bearing portion of the coupling
(or connection) should conform to certain known design
principles. The total axial bearing surface provided by the
full-depth load-bearing threads should be at least equal to the
cross-sectional area of the pipe material. The angle of
orientation of the stab flanks of the axial load-bearing threads
should differ from the angle of orientation of the load flanks
by not less than about 15 degrees. The step height of the axial
load bearing threads should be greater than the thread radius so
as to avoid galling (metal abrasion). Tilt of the distal end of
the pin that causes the yield strength of the steel to be
exceeded is generally to be avoided. Other general principles
of thread design and coupling design will be known to those
skilled in such design work and should be applied to the design
of the connection of the present invention.
Couplings are known in which the angle of orientation
of the load flank of the axial load-bearing threaded portion is
negative.
Couplings are further known in which a metal-to-metal
seal is provided. For best results, such surfaces should have
a controlled degree of roughness, as by shot peening, grit
blasting, glass bead peening, or helical microgroove threads
having a pitch very small relative to the pitch of the load-
bearing coupling threads.
20~9622
Problems exist in known pipe couplings. Especially,
the pin end of the pipe is often subjected to rough handling with
consequent damage to coupling and sealing surfaces. A pipe
coupling should be able to function effectively if there has been
slight damage to the pin. This could be accomplished if the
sealing force were designed to be above some predetermined
minimum throughout the effective sealing area, and is well above
the minimum value at each end of the sealing area. Then, if
either end (or some point in between) of the sealing area were
slightly damaged, the seal at one end of the sealing area at
least would provide an adequate seal. But many prior couplings
having metal-to-metal sealing areas adjacent the threaded
coupling areas typically provide a minimum or even less than
minimum design sealing force at one end of the coupling and a
maximum (above design sealing force) at the other. So if that
part of the sealing area in which the sealing force is designed
to be maximum is damaged, the effective overall effective sealing
force may be only the design minimum sealing force, or
conceivably even less than the design minimum. This may be
insufficient to prevent leakage under extreme operating
conditions.
SUMMARY OF THE INVENTION
The pipe coupling or connection according to the
invention overcomes the foregoing problem by providing a metal-
20~9622
to-metal seal in which the sealing force is well above the design
minimum at each end of the sealing area and at or above the
design minimum throughout the sealing area.
A pipe coupling or connection according to the
invention is of the type comprising a female coupling component
and a male coupling component, each matingly threaded for
coupling engagement, and each provided with a sealing area
adjacent the coupling threads (load threads). The sealing area
of each coupling component is formed as a frusto-conical surface,
at least one and preferably both surfaces being slightly
imperfect. Such surface is imperfect in the sense that it has
a controlled surface roughness. In other words, it is provided
over its surface with very shallow, closely spaced fine surface
variations or irregularities, as are formed for example by shot
or glass bead peening, etching, grit blasting, or by threading
with microgrooves at a pitch very small relative to the pitch of
the coupling threads. The degree of surface roughness chosen
will depend upon the application. For example, to avoid gas
leakage, the surface irregularities or degree of roughness should
by very fine. In drilling mud, a coarser surface finish may be
satisfactory. As thus far described, the structure is
conventional.
The coupling or connection according to the invention
differs from known couplings in that it has the following
combination of characteristics, or such of them as may pertain
to the achievement of the design objective at hand:
1. The seal taper angle is of a low gradient, and
- 2059622
there is a slight mismatch between the pin seal taper and the box
seal taper, the box seal taper being slightly steeper than the
pin seal taper, so as to obtain a sealing force distributed over
the sealing area that simulates that of a shrink-fit cylindrical
seal.
2. As the coupling is made up (assembled),
interference of the pin and box sealing areas occurs at least
about as soon as, and preferably before there is any interference
between the load threads of the pin and box members in the
~icinity of the sealing areas, ("Interference in this context
means interference tending to pry apart the frusto-conical
sealing surfaces of the pin and box - i.e. interference creating
a radial force between these sealing surfaces).
3. The load flanks of the load threads are negatively
inclined.
4. The sealing areas are coated with a high-temperature
graphite particle-containing lubricant with a relatively high
content of solid graphite particles, or similar such lubricant.
5. The coupling is provided with an "insurance"
auxiliary seal operating when the pipe is in compression. (It
may also operate for tensile loads up to some threshold value).
The "compression" auxiliary seal is formed between a slightly
negatively inclined annular seat formed by a torque shoulder at
the proximal end of the box located inwardly of the sealing area,
and the terminating distal radial surface of the pin.
205~622
6. The pin seal taper may desirably be at the same
angle to the pin axis as the pin load thread taper. The box seal
taper is thus preferably formed to be slightly steeper than the
pin seal taper. The box load thread taper may be at the same
angle to the box axis as the pin load thread taper, or may be
slightly shallower, as proposed in copending application Serial
No. 2,059,594 , filed concurrently herewith.
These various characteristics of the pipe coupling of
the invention will now be discussed in more detail.
1. Pipe couplings are previously known in which the
box seal taper is steeper than the pin seal taper. It is known
that such mismatch facilitates initial contact and deflection
("tilting") of the pin nose as it engages the box sealing surface
during assembly, and tends to reduce the overall seal area,
concentrating the pressure loading of the sealing area over a
relatively small portion of the available sealing surface in the
vicinity of the distal end of the pin, as disclosed for example
in Mott U.S. Patent No. 4,736,967, granted 12 April, 1988.
Mott discloses a seal taper differential between box
and pin' but does not discuss seal taper relative to thread
taper. Nor does Mott discuss any preferred relationship between
such tapers and the grade of steel used.
Furthermore, Mott does not teach a pin/box sealing area
mismatch that provides an axial distribution of sealing force
over the sealing area simulating that which would be obtained
from shrink-fitting an internal circular cylindrical box sealing
20~9622
-
surface about a mating external circular cylindrical pin sealing
surface. Such axial distribution is characterized by a
relatively uniform sealing force (at or exceeding the design
minimum) over most of the axial length of the contacting sealing
surfaces, but having force peaks well in excess of the design
minimum at both ends of the axial extent of the contacting
sealing surfaces. These force peaks are desirable because they
provide the greatest sealing force in the areas that are most
sensitive to disturbance. At the distal end of the pin, there
is a risk of damage due to careless handling. At the proximal
end of the pin, there is the risk of seal separation due to out-
of-tolerance load thread interference in the vicinity of the
sealing area. So the availability of peak sealing forces at both
ends of the sealing area tends to offset these risks.
The seal taper of both pin and box seals must be of
relatively slight (shallow) gradient. This facilitates
maintenance of a relatively long sealing area in the axial sense
and ensures that the desired simulated shrink-fit cylindrical
sealing characteristic is obtained. Further, it ensures that
tilting of the distal end of the pin during assembly of the
coupling will not be unacceptably severe, and that there is an
adequate thickness of material at the distal end of the pin.
(Some tilting of the pin end is desirable, because it provides
the force peak at the distal end of the pin sealing area and
causes some burnishing of that part of the sealing area to
occur.)
Further, because there will be relative movement of box
and pin sealing surfaces as a result of compression/tension
2~59622
-
thermal cycling of the pipe, it is important to maintain adequate
contact between the sealing surfaces throughout the cycling.
This cannot be accomplished if the taper is too steep.
The mismatch between the box and pin tapers must be
sufficient to simulate the desired cylindrical shrink-fit seal
characteristic, but should not appreciably exceed that degree of
mismatch. Furthermore, the mismatch should not substantially
interfere with the design objective of maintaining an adequate
thickness of material at the distal end of the pin and of
avoiding undue tilting of the distal end of the pin during
assembly. This result can be achieved if the taper (in inches
per inch) mismatch is kept less than the designed gauge-point
nominal interference (in inches) between the box and pin sealing
areas, at least for most oil well pipe. Preferably the mismatch
should be about 60% of nominal interference, or at most about
70%. (For the sealing areas, the gauge point is preferably
selected to be about midway between the beginning and end of the
longer of the two sealing areas, the longer sealing area being
that of the pin for manufacturing reasons, and at a nearby mating
point of the shorter of the two sealing areas, viz, the female
sealing area). As the degree of mismatch declines in the
direction of equality of pin and box sealing surface tapers, the
force peak at the distal end of the pin sealing area drops off
relative to the force peak at the proximal end of the pin sealing
area. This result may be tolerable for some applications, but
it is preferred to have the mismatch sufficient to create a
distinct substantial force peak at the distal end of the pin
- 205~622
sealing area, since this objective tends to offset the possible
reduction in sealing efficacy of that part of the seal that may
be caused by slight damage to the distal end of the pin. Note
that interference should increase if the grade of steel increases
appreciably.
It is within the scope of the invention that only one
of the mating sealing surfaces be roughened in the manner
described, the other surface being smooth. That arrangement will
still provide a good seal. However, it is preferable that both
surfaces have a controlled surface roughness.
2. At the proximal end of the pin sealing surface, a
force peak is desired when the coupling is made up (assembled).
This objective will be defeated or impeded if there is too much
load thread interference in the vicinity of the sealing area.
We are here referring to the radial interference of box with pin
threads adjacent the seal. Desirably the bearing force when the
coupling is fully made up will tend to be concentrated at the
proximal end of the sealing area (relative to the pin), rather
than borne by the load threads in the vicinity of the sealing
area.
Furthermore, it is desired that there be some
burnishing of the sealing surfaces and mashing or compaction of
the solid particles in the sealing lubricant during assembly.
This objective cannot be achieved if interference between box and
pin load threads prevents pressure contact between box and pin
sealing surfaces during make-up.
2059622
If the simulation of the cylindrical shrink-fit seal
load force characteristic is achieved, it follows that the load
thread interference in the vicinity of the sealing area when the
coupling is fully made up is not sufficient to pry apart the
sealing surfaces of box and pin. But this design criterion is
not in and of itself sufficient to achieve the latter of the two
objectives mentioned above.
Accordingly the relative dimensions, configuration and
angles of load threads and sealing surfaces should be chosen so
that box/pin sealing surface interference occurs during make-up
at least as soon as, and preferably before, load thread
interference in the vicinity of the sealing area occurs. (Load
thread interference in the vicinity of the proximal end of the
pin load threading is less critical, since it will not usually
have any appreciable tendency to pry apart the sealing surfaces
of box and pin). The greater the lag of load thread interference
following seal surface interference during make-up, the higher
the concentration of sealing force at the proximal end of the pin
sealing surface.
3. Although the provision of negative load flank
angles on the load threads is a characteristic of some previously
known couplings, the purpose of such provision has been to
eliminate any axial force component of the tensile force acting
on the threads to tend to disengage the box and pin when the pipe
is under tension. In other words, the negative load flank angles
are there to prevent the coupling from bursting apart when
tensile stress is applied to the coupling.
2059622
._
It has not however heretofore been specifically taught
that the combination of negative load flank angles with the other
structural features mentioned above improves the seal, in that
it resists any tendency of the sealing surfaces to be pried apart
when the coupling is placed under tension.
The load flank faces should be slightly negatively
inclined relative to the radial direction, an angle of the order
of -5 degrees (depending upon pipe diameter and thread depth)
typically being suitable. The stab flank faces of the coupling
threads are then formed at a positive angle to the radial, and,
as mentioned earlier, the sum of the stab flank angle to the
radial and the load flank angle to the radial should not be less
than about 15 degrees. So the stab flank angle could be a
minimum of about +20 degrees, assuming a load flank angle not
exceeding -5 degrees (in a negative sense).
4. As mentioned, the coupling seal designed in
accordance with the principles of this invention tends to be
optimum if a high-temperature high-solids graphite particle-
containing lubricant or equivalent is applied to the sealing
surfaces before make-up of the coupling. Lubricants employing
solid metallic particles (e.g. copper) are not as satisfactory
because it is difficult during the sliding of the box and pin
sealing surfaces during make-up to compress or break up the
metallic particles. Graphite particles are much easier to mash,
and when they do, they tend to fill the hollows in the sealing
surfaces. Even if the petroleum constituent of the lubricant is
later lost as a result of high temperatures, the graphite remains
2059622
,.3--~.
to fill the voids and hollows between pin and box sealing
surfaces.
5. Desirably, a torque shoulder is formed at the
proximal interior portion of the box against which the mating
distal end of the pin thrusts and seals when the coupling is
fully made up. The face of this torque shoulder is preferably
given a slight negative angle to impede any tendency of the
distal end of the pin to deflect or deform in an inward radial
sense, and to force it into preferred sealing engagement with the
box.
Compression of the pipe coupling forces the distal end
of the pin against the box torque shoulder, and if the facing
surfaces are reasonably smooth and well-mated, a seal is formed
that augments the seal formed by the contacting frusto-conical
sealing surfaces of the box and pin.
6. The design of load thread tapers to provide a
slight mismatch between box and pin thread pitch lines is more
fully discussed in the aforementioned copending patent
application Serial No. 2,059,594 . The pin thread taper should,
to achieve the desired increase in bearing load at the proximal
pin sealing surface, be slightly steeper than the box thread
taper.
However, the coupling design according to the invention
has significant merit in comparison with previous designs even
if there is no thread taper mismatch. The pin thread taper and
box thread taper can be identical. In that case, manufacturing
C'.
`i
-- 2053622
convenience dictates that the pin sealing surface taper should
preferably be identical with the pin thread taper, and the seal
mismatch effected by designing the box sealing surface taper to
be steeper than the pin sealing surface taper.
The coupling threads (load threads) of the male and
female components may be formed along thread pitch lines whose
surfaces of revolution form mating frusto-conical surfaces (i.e.
the thread pitch lines are tapered for easy coupling). The slope
of the frusto-conical surface along which the full depth coupling
threads (load threads) of the male (pin) coupling element are
formed should, for manufacturing convenience, preferably be equal
to the slope of the frusto-conical sealing surface of the male
coupling element. The female (box) element must of course be
internally formed to mate with the male, subject to the slight
mismatch of sealing surface slopes previously mentioned, and
subject to the possibility of a slight mismatch between the
slopes of the box and pin load threads, as described in more
detail in copending application Serial No.2,059,594 ,filed
concurrently herewith.
Note that the coupling threads of the male and female
(pin and box) members of the coupling or connection do not
necessarily provide a seal; the interference can be designed to
be relatively low. Clearance should be provided between the
crest of the pin threads and the root of the box threads, but
some interference between the root of the pin threads and crest
of the box threads, in accordance with conventional practice, may
be designed.
Alternatively, interference may result from a slight
14
'~ ~
... .
205~622
mismatch of box and pin load threads, as discussed in the
aforementioned copending application Serial No. 2,059,594
The foregoing characteristics of the
connection/coupling of the invention, afford a useful bearing
load-versus-length relationship simulating that of a shrink-fit
cylindrical seal over the length of the sealing portion that is
characterized by relatively high bearing load at either extremity
of the sealing portion of the pin and box members, and somewhat
lower bearing load (but at least as high as design minimum)
intermediate the two ends of the sealing portion, when the box
member has been fully threaded onto the pin member. What is
considered to be a sufficient axial length of the sealing area
will be dependent upon such parameters as pipe diameter, grade
of steel, wall thickness variation, etc. For most pipe suitable
for use in oil well applications, a pin sealing area of the order
of one inch in length and a box sealing area of approximately
half or more of the length of the pipe sealing area will be found
to be satisfactory.
The foregoing characteristics ensure that a very good
seal is maintained at both ends of the sealing portion if the
coupling and sealing surfaces of the pin and box members are
relatively undamaged. There is, as mentioned, a risk of some
surface damage, particularly at the distal end of the pin member.
If that damage prevents the distal end of the pin member from
engaging the mating (but slightly mismatched) sealing portion of
the box member with a sufficiently high bearing (sealing) force,
there will still be adequate bearing force at the proximal end
of the sealing area of the pin portion when the pin engages the
..
2059622
mating box member, and thus, there will be adequate bearing
pressure at least at one end of the sealing area to form an
adequate seal.
Note that a greatly mismatched taper of pin member
relative to box member sealing surfaces would give a high bearing
(sealing) load at the distal end of the pin member but a
relatively low bearing load at the proximal end of the pin
member. This is undesirable, especially because the risk of
damage is highest to the distal end of the pin. Also relatively
undesirable is the case of identical tapers of pin and box
members, affording a high bearing (sealing) load at the proximal
end of the pin member, the bearing load diminishing toward the
distal end. The invention should be implemented between these
two extremes.
For most oil well pipe, to ensure that the mismatch is
within acceptable limits, as mentioned above, the slope of the
box member sealing area over a unit length should exceed that of
the pin by less than the nominal interference between the
threaded sealing areas at or near the gauge point. This would
be a preferred characteristic for pipes of 3.5 inches to about
16 inches in diameter, and having contacting sealing areas of at
least about 3/8 inch.
A further preferred characteristic of the connection
or coupling according to the invention is that the relatively
long and low-angled sealing surfaces have sufficient interference
at the gauge point of each member to be greater than the pressure
differential for which the connection or coupling has been
designed (which according to the preferred embodiment, and in
16
20~622
-
accordance with conventional design practice, is 100~ of yield
strength of the pipe body) regardless of the bearing load at
either end of the sealing threaded portion, when either axial
tension or axial compression loads are applied to the coupling.
The preferred low-angled sealing area design, of the
pin element especially, is beneficial in that during handling,
if there is any damage, e.g. from stabbing the pin into the box,
any unwanted protrusions or other damaged parts of the distal end
of the pin may be hand dressed, i.e. filed off or ground off
without any appreciable diminution of the efficacy of the
coupling. The proximal end of the pin in engagement with the
mating coupling threads of the box can be expected to have a
sufficiently high bearing load that such load will exceed the
pressure differential for which the coupling has been designed,
even at full pipe yield pressures.
BRIEF DESCRIPTION OF THE DRAWINGS
Figure 1 is an axial partial section view of a pin
member wall constructed in accordance with the invention, at the
end of a length of pipe, as seen through the pipe wall, showing
the coupling threaded portion and adjacent sealing portion for
engaging a mating box member of the coupling.
Figure 2 is an axial partial section view of a wall
portion of the mating female or box member of a coupling
according to the invention showing the coupling threaded portion
and sealing threaded portion for receiving the pin member of
Figure 1.
17
- 2059622
-
Figure 3 is a partial view in axial section of a
modified alternative construction of the wall portion of the box
member modified to permit chaser manufacture of the box threads,
and otherwise conforming to the box structure of Figure 2.
Figure 4 is a detailed enlarged partial axial section
view of the pin thread of Figure 1, taken at the gauge point.
Figure 5 is a detailed enlarged partial axial section
view of the box thread taken at the gauge point of the box of
Figure 2.
Figure 6 is a detailed enlarged partial axial section
view of the coupling thread of Figure 4 shown engaging the
coupling thread of Figure 5, under load, assuming equal taper of
pin and box load thread (i.e. having the same thread pitch line).
Figure 7 is a greatly enlarged detailed partial axial
section view of a portion of a microgroove sealing area suitable
for use as the sealing area of either the pin member or box
member of Figure 1.
Figures 8A through 12B are a series of graphs and
schematic views of the relative degrees of taper of pin and box
sealing areas of the pin and box members of the coupling
according to a preferred embodiment of the invention (Figures 9A
through 9D), such preferred embodiment being presented in
Figures 1 to 6, when the pin member has fully engaged the box
member, as compared with alternative configurations whose
characteristics are shown in the remaining ones of these figures.
The dimensions, slopes and mutual spacing of the pin and box
sealing areas have been exaggerated for ease of comprehension.
Figure 8A is a graph showing a desirable shrink-fit
18
2~59622
circular cylindrical seal bearing load vs. distance
characteristic. The orientation of Figure 8A through 12B is
the same as that of Figures 1 and 2, with the distal end of
the pin to the right, and the proximal end of the pin to the
left.
Figure 8B schematically represents the box and pin
sealing area relationship giving rise to the characteristic
depicted in Figure 8A.
Figure 9A is a graph showing a representative seal
bearing load vs. distance characteristic for the seal of a
coupling or connection designed according to the principles
of the invention, where thread interference in the vicinity
of the sealing area begins about simultaneously with seal
surface interference, as the coupling is being made up, and
where box thread taper is the same as pin thread taper.
Figure 9B schematically represents ~he slightly
mismatched gently sloped tapers of box and pin sealing
surfaces according to the invention, giving rise to the
graphs of Figures 9A, 9C and 9D.
Figure 9C is a graph showing a representative seal
bearing load vs. distance characteristic for the seal of
Figure 9B, but where thread interference in the vicinity of
the sealing area lags the occurrence of sealing area
interference as the coupling is being made up.
Figure 9D is a graph showing a representative seal
bearing load vs. distance characteristic for the seal of
Figure 9B as modified in the same way as for Figure 9C, but
--19--
2oS~622
with pin thread taper being slightly steeper than box thread
taper.
Figure lOA is a graph showing a representative
seal bearing load vs. distance characteristic for a coupling
box seal/pin seal relationship in which both box and pin
sealing
-19A-
20~622
surfaces have the same gently sloped taper.
Figure 10B schematically represents the box and pin
sealing area relationship giving rise to the characteristic
depicted in Figure 10A.
Figure llA is a graph showing a representative seal
bearing load vs. distance characteristic for a coupling box
seal/pin seal relationship in which both box and pin sealing
surfaces have the same steeply sloped taper.
Figure llB schematically represents the box and pin
sealing area relationship giving rise to the characteristic
depicted in Figure llA.
Figure 12A is a graph showing a representative seal
bearing load vs. distance characteristic for a coupling box
seal/pin seal relationship in which the box and pin seal are
greatly mismatched.
Figure 12B schematically represents the box and pin
sealing area relationship giving rise to the characteristic
depicted in Figure 12A.
DETAILED DESCRIPTION OF THE INVENTION
The end of a steel pipe, tube or casing 11 is formed
to provide a pin generally indicated as 12. Pin 12 has a
threaded portion 19 beginning at a chamfered starting thread 15
located at a position short of the distal end 18 of the pipe 11
and extending axially therefrom to terminate in a vanish point
13. The thread pitch line of threaded portion 19 of the pin 12
is sloped inwardly from its proximal end at vanish point 13
20S3622
toward its distal end. The pin 12 terminates in a frusto-conical
sealing area 21 provided with a controlled surface finish to
provide a limited degree of roughness, e.g. helical microgrooves
formed by way of threading, as more particularly illustrated in
Figure 7. The angle of slope of the sealing surface 21 along the
frusto-conical surface is equal to that of the thread pitch line
of the threaded portion 19.
It will be noted that in Figure 1, the depth of the
roots, and the height of the crests of the threads of threaded
portion 19 relative to the roots of the threads, of pin 12
increases from the vanish point 13 to a maximum about midway
along the axial length of coupling portion 19, well before
reaching the starting thread 15. As illustrated, seven of the
threads are perfect threads.
It can be seen from referring to Figure 4 that the
load flank face 25 of the pin threads is negatively inclined at
a very slight angle, shown as approximately -3 degrees, to the
radial plane 27. This load face angle orientation is maintained
throughout the entire threaded portion 19 of pin 12. The angle
of inclination of the load flank face chosen will vary with
choice of thread height, but can generally be expected not to
exceed (in a negative sense) -10 degrees even for very shallow
threads.
It can also be seen from Figure 4 that the stab flank
face 29 of the pin threads has a positive angular orientation
relative to the radial plane 27. According to preferred design
practice, the difference in angle between the orientation of the
load flank face 25 and the stab flank face 29 should not be less
2~622
_ ..
than about 15 degrees. So if, for example, the load flank face
25 is at a negative angle of -3 degrees, then the angle of
orientation of stab flank face 29 should not be less than about
18 degrees. The stab flank face angle or orientation is also
maintained uniformly throughout the length of the threaded
portion 19 of the pin member 12.
Referring to Figure 2, the box 31 of which half of a
complete wall length (in the axial direction) is illustrated in
Figure 2, is internally configured and threaded to mate with the
pin 12 of Figure 1. The other half of box 31 (not illustrated)
is similarly internally configured and threaded to receive the
pin of the next length of pipe. In an integral connection, the
pin could be formed as illustrated in Figure 1, the female end
as illustrated in Figure 2 (or Figure 3, as an alternative to
Figure 2).
Specifically, the female coupling element 31 is
provided beginning at its distal end 33 with a threaded portion
generally indicated as 35 extending into the interior of box
member 31 as far as a terminating thread 38. Further inwardly
from thread 38 is a gap functioning as a single-point threading
tool relief groove, generally indicated as 39, terminating in a
shoulder 41 which defines the outermost limit of an interior
frusto-conically shaped, microgroove sealing surface generally
indicated as 43, which terminates in a limit or torque shoulder
45 forming a negatively inclined annular seat 46. The negative
inclination of seat 46 tends to prevent the pin end 18 from
climbing over the shoulder 45 when excess torque or high axial
loading is applied to the coupling.
2o59622
Although the threaded portion 35 of box 31 and the
sealing surface 43 of box member 31 are both tapered so as to
receive in coupling and sealing engagement the mating pin 12 of
Figure 1, nevertheless the degree of taper of the interior
sealing surface 43 of box 31 is deliberately chosen to be
slightly steeper than the degree of taper of the mating sealing
surface 21 of pin 12. The reason for this is to provide a
preferred bearing load-versus-length relationship, as discussed
above and to be discussed in greater detail below with reference
to Figures 9A and 9B.
The threads 35 (load threads) of box 31, shown in
enlarged profile in Figure 5, are angled to mate exactly with the
threads of pin 12. Further, the thread pitch line of threads 35
is at least approximately that of threads 19. In other words,
the surfaces of revolution of the thread pitch lines for the
coupling threads of the male and female coupling components are
mating or nearly mating frusto-conical surfaces. (They may be
slightly mismatched, as described in the aforementioned copending
application Serial No.2,059,594 ). The box thread is deeper
than the pin thread so as to afford the necessary thread
clearance. Their respective dimensions are better understood by
referring to Figure 6.
It can be seen from Figure 6 that a representative load
flank face 25 of a representative pin thread 26 is in flush
engagement with a mating load flank face 51 of a thread 52 of box
31. Between the crest 28 of thread 26 and root 53 of the mating
box member threaded portion 35 is a clearance gap 61. Further,
there is a clearance gap 63 between stab flank face 29 of thread
~,
205~622
,
26 and mating stab flank face 55 of mating threaded portion 35
of box 31. On either side of thread 26, it can be seen that the
crests 57 of the box threads 35 are in flush engagement with the
root faces 24 of pin coupling threads 19.
In other words, there is interference and flush
engagement between the crests of the box thread and roots of the
pin thread, and compressive interference between the load flank
faces of box and pin threads (when the threads are fully engaged
and normal tensile stress occurs in the coupling). In the same
condition, there is clearance between the pin thread crests and
box thread roots and clearance between the stab flanks of the box
and pin threads. Note, however, that if the coupling were not
stressed in tension but instead were stressed in compression, the
stab flank faces rather than the load flank faces of the pin and
box threads would become in load-bearing contact with one
another.
Figure 3 illustrates an alternative structure for the
interior of the box member. In Figure 2, the coupling threaded
portion was shown to terminate in a final thread crest 38,
followed by a thread relief groove 39, followed by a curved
shoulder 41, and sealing surface 43. By contrast, the relief
groove 39 is omitted in the Figure 3 alternative embodiment, and
instead, there is a thread run-off area 73 intermediate the end
of the coupling threaded portion 35 and the sealing surface 44,
merging with surface 44 via a curved shoulder transition portion
75. The sealing surface 44 of Figure 3 is substantially
identical to the sealing surface 43 of Figure 2, with the
qualification that the total axial distance occupied by the
24
20!~96~2
sealing surface 44 of Figure 3 is somewhat shorter than the total
axial distance occupied by the sealing surface 43 of Figure 2.
While the Figure 3 embodiment has less total sealing
area than the Figure 2 embodiment, nevertheless the Figure 3
embodiment is easier to manufacture using a chaser technique,
using the same tool bit (requiring no withdrawal of one tool bit
and insertion of a separate tool bit). The sealing surface 44
can, using the chaser technique, be machined first as a helical
microgroove surface, immediately followed by the machining of the
threaded portion 35, without withdrawing the tool. The Figure
2 embodiment does not admit of this possibility, but would
require three separate tool bits to cut the threads, the thread
relief groove, and the sealing area respectively, assuming that
microgrooves are formed on the sealing area.
In this specification, reference will occasionally be
made to the gauge point of the load threads and of the sealing
surfaces of both pin and box. This is the point at which nominal
design values are selected for whatever parameters pertain to
such point. For example, the nominal interference value designed
for the coupling is selected relative to the gauge points of the
box and pin - the sealing surface gauge points for seal
interference, the load thread gauge points for thread
interference.
The selection of the gauge point is arbitrary to some
extent, but ordinarily conveniently chosen as some intermediate
point rather than a terminating point (of sealing surface, or
threading, as the case may be). Suitably selected gauge points
are shown schematically in Figures 1 and 2. Pin thread gauge
2059622
point 20 is selected to be at or near the mid-point of the range
of perfect threads on the pin. Box gauge point 36 is selected
to be approximately aligned with pin gauge point 20 when the
coupling is made up. Pin seal gauge point 16 is selected to be
in the vicinity of the mid-point (axially) of the effective pin
sealing surface. Box seal gauge point 42 is selected to be
approximately aligned with pin seal gauge point 16 when the
coupling is made up.
In use, the pin 12 of Figure 1 is stabbed into the
opening generally indicated as 47 of the box 31. Pin 12 is
thrust in sufficiently far that contact is made between the
starting thread 15 and a contacting thread surface of the
threaded portion 35 of box 31, following which engagement of the
threaded portions 19, 35 of pin 12 and box 31 respectively
begins. The box 31 is then rotated relative to pin member 12 or
vice versa so as to screw the pin member 12 into the box member
31. Rotation of the box member 31 relative to pin member 12
continues until the limit of the threaded portions is reached and
the coupling threaded portion 19 of pin member 12 fully engages
the mating coupling threaded portion 35 of box member 31.
Rotation is effectively terminated when distal end 18 of pin 12
comes into pressure contact with annular seat 46 of box 31. This
contact, assuming that the distal end 18 of pin 12 matingly seats
against torque shoulder 45 in annular seat 46, will tend to form
an effective auxiliary seal when the coupling is under
compression.
Before this point is reached, the sealing portion 21
of pin 12 will have commenced engagement with the mating (but
- 26 -
C;' "`
Sf
2059622
slightly differently tapered, as mentioned above and discussed
in further detail below) interior sealing surface 43 (or 44) of
box 31. If the sealing areas are surface-roughened by
microgroove threading, it is apparent that the pitch of the
microgroove threads on sealing surfaces 21, 43 of pin 12 and box
31 respectively must be very much smaller than the pitch of the
threaded coupling portions 19, 35. It follows that the
microgrooves on pin 12 will skip relative to the microgrooves of
box 31, as the box 31 is screwed onto pin 12. This action
generally will not damage the sealing surfaces 21, 43
appreciably, but will tend to smooth out any surface
irregularities and will also, if a sealing compound has been
applied to the sealing surfaces, tend to spread the sealing
compound over the sealing surfaces and cause entrapment of the
sealing compound by depressions in the mating sealing surfaces
21, 43 of the pin 12 and box 31 respectively so as to facilitate
formation of a large effective sealing area as between the
microgroove sealing surface 21 and microgroove sealing surface
43. The entrapment of sealing lubricant will serve to protect
against wear and will reduce any propensity of the sealing areas
to gall destructively.
As mentionedpreviously, a high-temperaturehigh-solids
graphite particle containing sealing lubricant is preferably
used. As the coupling is made up, the relative sliding action
under pressure of the pin and box sealing surfaces tends to mash
the graphite particles and force them to occupy the hollows and
voids in and between the engaging sealing surfaces, promoting
effective sealing.
2059622
According to the invention, the thread configuration,
sealing surface configuration and designed interference are
selected so that when the pin sealing surface 21 first makes
interfering engagement with the box sealing surface 43, as the
coupling is being made up, there has not yet been any undue
interference between pin threads 19 and box threads 35 in the
vicinity of the sealing surfaces 21, 43. (Such undue thread
interference would undesirably prevent the desirable relative
sliding action under pressure between the sealing surfaces 21,
43).
There need never be any interference between the pin
threads 19 and box threads 35 in the vicinity of the sealing
surfaces 21, 43, or at all, so far as creating an effective seal
is concerned.
The profile of a microgroove surface suitable for use
as the controlled roughened surfaces of sealing areas 21, 43 of
pin member 12 and box member 31 respectively is shown (in
profile) in the partial section view of Figure 7. The pattern
is an undulate (wave) pattern with sharp narrow crests and
relatively wide shallow concavely curved troughs. The pitch of
the helical microgrooves 21, 43 on pin member 12 and box member
31 respectively is very small relative to the pitch of the
coupling threaded portions 19, 35 of pin member 12 and box member
31. For example, it may be of the order of .01 inches per
revolution, as compared with a pitch of .200 inches per
revolution for the load threads, in the case of about 5 to 10-
inch diameter pipe. The depth of the microgrooves (the distance
between the peak of the crest and the root of the trough) will
28
- 20S9622
vary with the pitch and type of cutting tool chosen. The depth
of surface irregularities may be expected to be anywhere from
about 30 to 250 microns, depending upon application.
It is not essential that the surfaces of sealing areas
be formed as helical microgrooves. Any similar undulate, mottled
or roughened surface would suffice. The surface may be formed
by any suitable technique such as acid etching, ball or glass
peening, or grit blasting. What is required is a fine, shallow
series of surface variations or irregularities that create very
shallow hills and valleys, relatively finely formed on the
frusto-conical surfaces. This kind of formation traps sealing
compound and typically results in the prevention of galling of
metal, as "hill summits" or crests yield as the pin is coupled
to the box, thereby improving the seal by reducing the depth of
the microgrooves or other depressions. The controlled surface
roughening characteristics may depend upon application and
environment. Gas-tight seals require a shallower depth of
surface depressions than liquid-tight seals. It is obvious that
the surface roughening should not create an axially extending
trough (through which leakage could occur) from one end of the
sealing area to the other.
Figures 8A through 12B illustrate graphically the
effect of mismatching the tapers of the sealing portions of the
pin and box respectively, by comparing alternative selections of
relative box and pin taper. The figures with suffix letter B
depict schematically in axial section, and with exaggeration of
slope angles and dimensions, the opposing effective sealing areas
of the pin and box, showing the complete axial length of mating
29
- 20~9622
sealing (e.g. microgroove) portions. Vertically aligned with
each of these schematic depictions is at least one graph, the
abscissa of which is the distance along the sealing area of the
pin and box while the ordinate is the bearing load on the sealing
surfaces when the pin has been fully screwed into the box. (At
that point, the pin distal face 18 makes contact with annular
seat 46 of the box 31.) The bearing load constitutes the sealing
force, and is a measure of the efficacy of the seal.
If the tapers of the box sealing area and the pin
sealing area are chosen to be identical, and zero, so that
circular cylindrical sealing surfaces are presented, then the
plot of bearing load against axial distance is that appearing in
curve A of Figure 8A. The interference as a consequence of the
shrink-fit of box on pin is designed to be sufficient to provide
a minimum bearing load M throughout the axial length of the
mating sealing areas. The minimum value M is appreciably
exceeded at the two ends. The Figure 8A design is impractical,
because engagement can be effected only by shrink fitting. But
the characteristic curve A is a model to be emulated because of
the load peaks at both ends of the sealing area and the uniform
loading therebetween.
According to the invention, the slopes of the tapered
pin and box sealing surfaces are selected to be gently angled and
slightly mismatched, as appears schematically in Figure 9B. In
that case, the bearing load varies with axial distance over the
sealing area according to curve Bl in the graph of Figure 9A.
The bearing load is at a design minimum M at an intermediate
point along the sealing threaded portions of the pin and box and
205~622
-
rises to a significantly higher value at both ends of the sealing
threaded portion of the pin and box. This curve Bl is the best
available simulation of curve A of Figure 8A obtainable from any
of the designs of Figures 9B, lOB, llB, 12B.
It has been assumed in depicting the graph and physical
arrangement of box and pin of Figures 9A, 9B that there is some
thread interference in the vicinity of the sealing surfaces when
the coupling is made up. If thread interference is deliberately
designed to lag the occurrence of sealing surface interference
by a considerable distance as the coupling is being made up, a
superior result is obtained, viz. that of Figure 9C. In that
case, curve B2 is essentially similar to curve Bl over most of
the axial distance along the sealing area, but curve B2 rises to
a significantly higher bearing force value than does curve Bl in
the vicinity of the proximal end of the sealing area relative to
the pin. This extra measure of pro~;m~l-end sealing force tends
to cause some burnishing of the sealing surfaces where that force
is present and facilitates mashing of sealing compound also.
It is not necessary, even when the coupling is
completely made up, that there be any thread interference in the
vicinity of the sealing area. In such case, proximal pin sealing
surface bearing load considerably exceeds the bearing load that
would result if adjacent thread interference were to lag sealing
surface interference by only a slight delay during make-up
("delay" of course being used in a relative box/pin rotational
movement sense, not in an absolute time sense).
If there is a slight mismatch between pin load thread and
- 31 -
-~, ~ ....
-- 2059622
box thread taper, the pin ~hread taper being slightly steeper
(more inclined to the axial) than the box thread taper, as taught
in the aforementioned copending patent application Serial No
2,059,594 such that the thread taper mismatch facilitates the
avoidance of thread interference in the vicinity of the sealing
area, then the result is a further increase in the value of the
bearing force (sealing force) at the proximal end of the sealing
area relative to the pin. The result is graphically depicted in
Figure 9D.
If the tapers of the pin and box sealing surfaces are
chosen to be equal but gently angled (Figure lOB) then the
bearing load against distance plot along the sealing area would
appear as curve C in the graph of Figure lOA. Bearing load would
be highest at the proximal end of the effective pin sealing
surface and lowest (not much higher than design minimum N) at the
distal end of the pin sealing sur~ace. While the design of Figure
lOB has some value, it affords uncomfortably little margin at the
distal pin end to accommodate variations in tilt of the pin
sealing area that result from manufacturing tolerances. There is
also a risk that the pin distal end bearing load may fall below
design value M.
The term "tilt" used in the jargon of pipe connections
refers to the change in pin seal taper occurring at the distal
end of the pin upon assembly of the pin into the box. Because of
the taper of the pin, the pin wall tends to become thin at the
distal end, and of course the steel is elastic, so the distal
portion of the pin wall deforms ;strains) as the pin is threaded
into tight engagement with the box. The degree of tilt will
~,.- ~ ,,
~- ^-~,
2~9622
._
depend upon pipe diameter, interference, starting taper, seal
area length, etc. If, because of manufacturing tolerances of the
design of the tapers selected, the tilt of the pin is greater
than expected, the sealing force (bearing force) at the distal
end of the pin may fall below design minimum value M if the
Figure lOB configuration is selected, but this is not likely if
the Figure 9B configuration is selected.
Some tilt of the distal end of the pin is desirable
during make-up of the coupling, since that tends to facilitate
mashing of the sealing lubricant solid particles and burnishing
of the sealing surfaces. But too great a tilt, which can be
caused by too great a mismatch between box and pin sealing
surfaces, can cause undesirable strain of the distal end of the
pin beyond the yield strength of the steel, in turn causing
permanent deformation and possible loss of distal end bearing
load.
Note that because both box and pin seal surface tapers
are relatively shallow, the interference between box and pin
sealing surfaces does not unacceptably increase as the coupling
is made up. As the pin advances into the box, although the force
between the pin and the box due to interference increases, that
force is borne by an axially longer sealing area as the pin
advances. Consequently, the point-to-point forces do not become
unacceptably large. Eventually as the pin advances towards its
final point of contact against the torque shoulder of the box,
there is some tilting of the distal end of the pin, which of
course relieves the stress on the tilted portion.
The choices of interference, degree of tilt, and box
2059622
-
and pin tapers, will vary according to the designer's preference,
and due regard will be paid to the grade of steel employed. As
the grade increases, the designer may wish to increase the angle
of the box taper relative to that of the pin. This will tend to
ensure that the distal end of the pin maintains a tight contact
with the torque shoulder of the box and will not disengage as the
coupling if finally made up. Higher grades of steel have the
capacity to absorb more energy from increased interference than
lower grades, and thus can tilt further than would be the case
for lower grade. Typically the box taper chosen will be from
about 3 to 4 for grades of steel of yield strengths from 40,000
to 80,000 psi, and the taper angle will be from 4 to 5 for
grades of steel having yield strengths from about 80,000 to
150,000 psi. The 150,000 psi material will require more
interference than say 50,000 psi material. The increased
interference prevents the distal end of the pin from moving out
of engagement with the box. It is important that internal
pressure of gas (say) within the pipe coupling should not
penetrate between the sealing surfaces to attempt to pry them
apart. The coupling design according to the present invention
is relatively free from such risk because of the significant
amount of tilting of the distal end of the pin that occurs upon
make up.
Figures llA and llB reveal that the design
of Figure lOB can become totally unacceptable if the degree of
taper of the mating sealing surfaces is sufficiently large.
Where the sealing surfaces are steeply inclined, the bearing load
diminishes to zero well before the distal end of the pin sealing
34
- 20S~622
surface is reached, and is above design minimum M only over an
unacceptably short sealing area. The falling of curve D below
the abscissa indicates absence of surface contact between the box
and pin sealing areas. (In fact, the bearing load cannot fall
below zero, so the portion of the curve below the abscissa is
imaginary) .
If the tapers of pin and box sealing surfaces are
greatly mismatched, the sealing portions of the pin and box tend
to lose surface contact and bearing force at the proximal end of
the sealing portion of the pin, corresponding to the outermost
end of the sealing area of the box. The highest bearing force
in the case of too much mismatch is to be found at the distal end
of the pin, where adequate contact between the pin member and box
member threaded portions can occur. This is manifest in the
graph of Figure 12A, in which curve El reflects the increase in
bearing load from its minimum near the proximal end of the pin
sealing surface (undesirably, below design load M) to a maximum
at the distal pin end, relative to the greatly mismatched taper
design of Figure 12B. Curve E2 illustrates a mismatch
sufficiently severe that contact between the pin and box sealing
areas is lost at the proximal end of the pin. (Again, the
portion of the curve below the abscissa is imaginary).
By contrast to Figure 12B, the tapers of both pin and
box sealing areas are relatively gentle in the design of Figure
9B, albeit slightly mismatched. If however the sealing surfaces
were sharply sloped, as in Figure llB, the bearing load at the
distal end of the pin sealing surface would tend to fall below
acceptable design minimum value M. So the invention should be
2059~22
-
practised under the condition that design minimum bearing
(sealing) force M is exceeded throughout the axial length of the
mating sealing surfaces. Beyond one extreme of the preferred
range, force M would be just slightly exceeded at the distal end
of the pin. That would offer little or no improvement over the
Figure lOB configuration. Just beyond the other extreme of the
range, as one approaches the Figure 8B design, it becomes
impossible to fit the pin into the box without severe galling.
The invention is optimally practised about half-way between these
extremes.
In all the graphs of Figures 8A through 12A, the
premise has been assumed that at any point along the effective
sealing area, the bearing load should always be equal to or above
some predetermined minimum acceptable design bearing load M, if
at all possible, so that no matter where one looks at bearing
load, one will find that predetermined minimum equalled or
exceeded. This minimum is equalled or exceeded at all points
along each of the curves A, Bl, B2, B3 and C. The criterion
cannot be obtained in the too sharply sloped configuration of
Figure llB, and it may not be obtained in the Figure 12B
configuration. The minimum value M is desirably appreciably
exceeded at both ends of the sealing area, as illustrated in
curves A, Bl, B2 and B3, since leaks tend to develop at the ends.
The configuration is which this desideratum is optimally obtained
is the slightly mismatched taper configuration of Figure 9B, with
further improvement available if thread interference lags seal
interference during make-up, and if slight thread taper mismatch
is selected as previously discussed.
36
2059622
,
In all of Figures 9B through 12B, it is assumed that
the degree of taper on the internal sealing surface of the box
is equal to or exceeds the degree of taper on the sealing portion
of the pin. A mismatch in the other sense, in which the taper
in the box is less pronounced than the taper on the pin, would
be quite unsatisfactory, because then the pin member would have
difficulty adequately penetrating the box member, and undesirably
high interference would result.
Note that if the distal end of the pin member is
damaged in handling, such that an insufficient bearing load is
present at that end, there will nevertheless be an adequately
high proximal pin end bearing load, if the slightly mismatched
taper arrangement according to the invention is selected. The
invention, however, gives the additional advantage that if the
distal end of the threaded portion of the pin is only slightly
damaged, then hand dressed, an adequately tight seal is created
at the distal end as well as the proximal end, whilst in between,
the sealing force is everywhere higher than the minimum bearing
force for which the seal has been designed.
The Figure 9B selection of box and pin sealing area
tapers tends to be optimally resistant to surge stresses, and
tends to afford optimal opportunity for hand-dressing repair of
damaged sealing area surfaces.
As mentioned, with reference to a sealing area gauge
point chosen intermediate the ends of the contacting sealing
areas, the degree of mismatch should be kept less than the gauge-
point nominal interference between the box and pin sealing areas,
in order to obtain a curve B1, B2 or B3 bearing load vs. distance
37
2059622
characteristic rather than a cur~e El characteristic The
nominal interference will he relal-ed to the yield strength of the
steel A mismatch of the order o~ SO or 60~ of nomlnal
interference is likely to be optlmal
Due regard must be palcl ro permltted tolerances in the
chosen design of coupling according to the invention Tolerances
should be chosen for both sealirg surfaces and load threads that
tend to minimize risk of gallinc, of the sealing surfaces during
make-up of the coupling. On the olher hand, tolerances should not
be chosen that would make possib3e a reduction of bearing load
throughout the effective sealing area below design minimum. The
design minimum normally should be at least egual to the expected
pressure differential at the yield strength of the selected
steel. More tolerance is permitted for higher grade steel than
for lower grade steel.
EX~'LE 1
Pin and box members according to the foregoing description were
prepared for use in couplings for 7-inch pipe having wall
thickness linear density ratings of 23 and 26 pounds per foot.
such a coupling is intended for use with well casings where steam
injection within the casing is rec.uired. Depending upon the
length of pipe and the expected p-essures, a 55,000 psi minimum
yield strength or 80,000 psi minimum yield strength steel may be
selected. Temperatures up to 65G degrees Fahrenheit must be
withstood, and axial tensile and compressive loads are expected
to occur which apprcach or even exceed the actual yield strength
38
.~
205~622
, .
of the material in the pipe body. The coupling was designed to
withstand this axial loading without failure whilst maintaining
adequate resistance to leakage from internal pressures ranging
up to actual yield strength of the pipe wall.
The coupling was prepared with approximately twelve
complete turns of threads tapered at .104 inches per inch for the
pin and box, and having a pitch of .200 inches per revolution.
Of the twelve threads on the pin, seven were perfect threads, and
the other five were partial threads diminishing to the vanish
point 13 as illustrated in Figure 1. The flank face orientation
for the pin threading was the same as that for the box threading,
namely -3 degrees for the load flank and +18 degrees for the stab
flank.
For the sealing surfaces, the microgrooves were formed
by a 3/64 inch radius turning tool fed at an axial feed rate
selected within the range of about .008 inches to .015 inches per
revolution. If the coupling will be used in a gaseous
environment, such as a heavy oil steam environment, a feed rate
nearer the lower value (.008 inches/revolution) is preferred.
For leak resistance in a conventional oil environment, a feed
rate nearer the higher value (.015 inches/revolution) is
preferred. The total sealing length of the pin member was
selected to be .900 inches; the sealing portion in the box member
would be slightly smaller, depending upon whether the Figure 2
or Figure 3 embodiment is chosen. The microgrooves at the .008
inches/revolution pitch are about .0002 inch in depth and at .015
inches/revolution pitch are about .0006 inch deep. The crests
tend to be flattened upon tightening the coupling, perhaps
39
2059622
removing at least about 20~ from the trough depth, and more in
the vicinity of the area of highest bearing pressure.
The gauge points for the pin and box were selected
as follows:
The pin and box thread gauge position was selected
to be at a point axially where the threads in the made-up
(assembled) position of pin and box members were directly
coincident and spaced trom the last full-depth thread (after
which only partial flank depth occurs, diminishing towards the
vanish point) . This axial position was also selected so as to
afford greater than zero interference at the ends of the
threaded portions in the vicinity of the outer end of the box,
less as one progresses inwardly (because of the load thread
taper mismatch between box and pin).
The pin seal gauge point 16 was arbitrarily chosen
to be 3/8 (.375) inch from the distal end 18 of the pin 12.
The box seal gauge point 42 was also arbitrarily selected to
be .375 inch from the box seat 46 (see Figure 2). The
calculated minimum interference load M was then based upon the
nominal interference (.019 inch) at a location 37s inch from
the distal end of the pin which coincides exactly with gauge
point of the box seal. Selecting the gauge point is somewhat
arbitrary on the basis that to stop leakage, the interference
at any point along the active seal must be equal to or greater
than the minimum seal interference (0.012" in this case).
Since the coupling has unequal seal tapers for the pin and the
box, in this case it is more convenient to locate the gauge
points at the same axial location from the torgue shoulder.
20~622
_
The pin sealing area taper was .104 inches per inch on
diameter, the same as that of the thread pitch line of the
threads, whilst the box sealing area taper was .1146 inches per
inch on diameter. This is a mismatch of .0106 inches per inch
on diameter, or less than the minimum gauge point sealing area
interference of .012 inches on diameter. The tolerance of the pin
thread and seal at the gauge point was + .004 inches on diameter,
and that of the box was + .003 inches on diameter.
In the Example 1 configuration, there was at least some
thread interference in the vicinity of the seal area;
consequently the sealing force vs. distance characteristic
resembled that of Figure 9A.
EXAMPLE 2
The parameters, dimensions etc. were the same as for
Example 1, except that the box seal taper was .110 inches per
inch on diameter, that the box thread taper was selected to be
.095 inches per inch, and the box seal gauge point 42 was
selected to be .500 inch from the box seat 46.
In the Example 2 configuration, there was no thread
interference in the vicinity of the sealing area even when the
coupling was made up. And there was slight mismatch between box
and pin thread taper, the pin taper being slightly more inclined
to the axis than the box taper, with the result that the sealing
force vs. distance characteristic resembled that of Figure 9D.
Terminology
The scope of the invention is as presented in the
41
2059622
appended claims.
In the appended claims:
1. The term "connection" includes a coupling.
2. The phrase "relatively shallow" with reference to
the slopes of the frusto-conical sealing surfaces of the box and
pin implies that:
(i) the taper is not so great as to give a bearing-
load-vs.-axial-distance characteristic similar to that of Figure
llA;
(ii) the taper is not so great as to create a
significant risk of loss of seal due to thermal cycling of the
coupling (i.e., alternate stressing of the coupling in tension
and compression);
(iii) the taper is not so great as to reduce distal-end
pin wall thickness unacceptably; and
(iv) the taper is nevertheless sufficient to avoid
galling of the sealing surfaces during assembly of the coupling.
3. The term "slightly less" with reference to the
slope of the frusto-conical sealing surface of the pin relative
to that of the box implies that:
(i) the mismatch is sufficient to avoid a bearing-
load-vs.-axial-distance characteristic similar to that of Figure
lOA;
(ii) the mismatch is not so great as to generate an
effective contacting sealing area between the box and the pin
that is unduly short in the axial direction;
(iii) the mismatch is not so great as to give a bearing-
load-vs.-axial-distance characteristic similar to that of Figure
42
20~622
12A; and
(iv) the mismatch is not so great as to cause undue
tilt of the pin during assembly.
43
