WO1998010213A1

WO1998010213A1 - Method of forming a metal-to-metal seal

Info

Publication number: WO1998010213A1
Application number: PCT/CA1997/000652
Authority: WO
Inventors: Maurice Slack; Jaroslaw Nowinka; Trent Kaiser
Original assignee: Centre For Engineering Research Inc.
Priority date: 1996-09-05
Filing date: 1997-09-05
Publication date: 1998-03-12
Also published as: AU4195497A

Abstract

Method of forming a metal-to-metal seal in high pressure applications with low contact stress. A first elongate pressure containment member (12) and a second elongate pressure containment member (16) are coupled with a thin walled extension sleeve (20) conjoined with the remote end (18) of the second member (16) positioned wholly within the first member (12) in close relation to a surface of revolution (14). The method involves placing the first member (12) and the second member (16) in an environment in which fluid pressure exceeds contact stress at some point of an operating range and tends to cause a flow of fluids from the second end (24) to the first end (22) of the sleeve (20) along an interfacial region (28) between the sleeve (20) and the surface of revolution (14). Fluid pressure and flow tends to set up a pressure gradient along the interfacial region (28) sufficient to deform the sleeve (20) and promote sealing along the interfacial region (28).

Description

Method of forming a metal -to-metal seal

FIELD OF THE INVENTION

The present invention relates to a method of forming a metal-to-metal seal in high pressure applications with low contact stress incorporating fault tolerant secondary control of leakage rates. This method has particular application for connections between tubular members, where there is relative movement between the tubular members .

BACKGROUND OF THE INVENTION

In high pressure applications, metal -to-metal contact seals are preferred over elastomeric seals. To achieve adequate sealing, contact seals typically require that the contact stress exceed the contained pressure. Efforts are typically made to minimize sealing widths. For example, in United States Patent 1,302,022 which issued to Drewry in 1919 concern was expressed of unwanted friction generated upon sliding or rotation. The sealing application of Drewry involved an inner sleeve which was received within an outer sleeve. The reference taught avoiding friction by maintaining a small amount of clearance between the sleeves except at a narrow sealing width. The Drewry reference involved a dynamic sealing application. In static sealing application the same teaching is common. An example of this is United States 3,214,198 which issued to Peuchmaur in 1965. The Peuchmaur reference discloses a seal between an outer tubular sheath and an inner tubular element. The inner tubular element has a lip in the form of a coaxial annular skirt . A ferrule is used to maintain the lip in sealing engagement with the outer tubular sheath. The lip has "at least one" annular rib which "distributes the total tightening stress transmitted by the lip to the sheath over an annular surface having a short axial length". Contact seals used in high pressure tubular fittings, typically have contact stresses much greater than the maximum pressure to be contained and distributed over very narrow widths to obtain adequate reliability. Calculations of the necessary contact stress are routinely performed using stress analysis methods such as finite element analysis.

Unfortunately, there are many applications in which it is desirable that the contact stress be minimized in order to facilitate assembly, sliding, rotation or other dynamic sealing operations. In such applications, the high contact stress of metal -to-metal contact seals make them undesirable, as it inevitably leads to higher friction forces during relative movements and greatly increased potential for galling or other damage which can initiate seal failure.

SUMMARY OF THE INVENTION

What is required is method of forming a metal -to-metal seal in high pressure applications with low contact stress.

According to the present invention there is provided a method of forming a metal-to-metal seal in high pressure applications with low contact stress. A first elongate pressure containment member is provided having a surface of revolution. A second elongate pressure containment member is provided having an exterior surface, an interior surface, a remote end, and a thin walled extension sleeve having a first end and a second end. The first end of the thin walled extension sleeve is conjoined with the remote end. The first elongate pressure containment member and the second elongate pressure containment member are coupled with the sleeve positioned wholly within the first elongate pressure containment member in close relation to the surface of revolution. The method involves placing the first elongate pressure containment member and the second elongate pressure containment member in an environment in which fluid pressure exceeds contact stress at some point of an operating range and tends to cause a flow of fluids from the second end to the first end of the sleeve along an interfacial region between the sleeve and the surface of revolution. Fluid pressure and flow tends to set up a pressure gradient along the interfacial region sufficient to deform the sleeve and promote sealing along the interfacial region.

With seals normally used in high pressure applications, high contact stress is concentrated over a narrow sealing width. With the present method, the teaching is exactly the opposite; a comparatively low contact stress is distributed over a wide sealing width. This teaching exploits the relationship between width and maximum contact stress. As the contact width is increased, the average contact stress is decreased. This teaching allows an effective seal to be provided, with a lower average contact stress.

Formulas necessary for the calculation of necessary contact stress in high pressure applications are published by the American Petroleum Institute. These formulas teach that the contact stress must be greater than the pressure to be contained in order to avoid seal failure. In accordance with the teachings of the present method, the contact stress between the sleeve and the surface of revolution can be less than the contained pressure over the full intended pressure range. There can even be an initial gap, that is closed and sealed as a result of deformation of the sleeve by fluid pressure.

There are numerous ways of securing the sleeve to the second pressure containment member. Sleeve can be secured by welding, clamping, or with various fasteners. In view of the high pressure applications for which the present method is intended, it is preferred that the sleeve be integrally formed as part of the second elongate pressure containment member.

It is important that there be a clear di ferentiation in the thickness of the sleeve and the thickness of the second elongate pressure containment member. The sleeve cannot merely be a gradually tapered extension of the second elongate pressure containment member, and still work as intended. It is, therefore, preferred that the remote end of the second elongate pressure containment member be of substantially uniform thickness. The sleeve is also of substantially uniform thickness. The first end of the sleeve is connected to the remote end of the second elongate pressure containment member over an interval of length not greater than one thickness of the remote end of the second elongate pressure containment member.

Although beneficial effects may be obtained through the use of the method, as described above, where there is a gap, a portion of the length of the sleeve is required to compensate for the gap and this can adversely affect the sleeve's ability to seal in response to fluid pressure. Even more beneficial effects may, therefore be obtained when a narrow radially extending circumferential notch is placed near the remote end of the second elongate pressure containment member on either the exterior surface or the interior surface, depending upon where the fluid pressure is less. The notch tends to deform in response to fluid pressure acting in an axial direction and tends to force the sleeve into engagement with the surface of revolution. The notch provides for some additional flexibility which makes the sleeve more responsive to fluid pressure.

BRIEF DESCRIPTION OF THE DRAWINGS These and other features of the invention will become more apparent from the following description in which reference is made to the appended drawings, wherein:

FIGURE 1 is a side elevation view, in section, of a metal- to-metal seal constructed in accordance with the teaching of the present invention.

FIGURE 2 is a side elevation view, in section, of the metal-to-metal seal illustrated in FIGURE 1, undergoing pressure testing. FIGURE 3 is a graph setting forth seepage rates at applied pressures when the metal -to-metal seal is subjected to pressure testing, as illustrated in FIGURE 2.

FIGURE 4 is a graph setting forth friction loads at applied pressures when the metal -to-metal seal is lubricated and then subjected to pressure testing, as illustrated in FIGURE 2.

FIGURE 5 is a magnified view of the metal -to-metal seal illustrated in FIGURE 1, deformed by applied pressure.

FIGURE 6 is a graph setting forth seal contact stress distribution when the metal-to-metal seal is subjected to pressure testing, as illustrated in FIGURE 2.

FIGURE 7 is a graph setting forth average contact stress as a function of applied pressure when the metal -to-metal seal is subjected to pressure testing, as illustrated in FIGURE 2. FIGURE 8 is a side elevation view, in section, illustrating a first alternative sealing assembly constructed in accordance with the teachings of the present method.

FIGURE 9 is a side elevation view, in section, illustrating a second alternative sealing assembly constructed in accordance with the teachings of the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT The preferred method of forming a metal -to-metal seal in high pressure applications with low contact stress will now be described with reference to FIGURE 1 through 7.

Firstly, provide a first elongate pressure containment member 12 having a surface of revolution 14.

Secondly, provide a second elongate pressure containment member 16 having an exterior surface 17, an interior surface 19 and a remote end 18. In FIGURE 1, first elongate pressure containment member 12 and second elongate pressure containment member 16 are tubular members. This is a typical application for this method of sealing, although there are some applications in which they need not be tubular members. Second elongate pressure containment member 16 has a thin walled extension sleeve 20 having a first end 22 and a second end 24. First end 22 of thin walled extension sleeve 20 is conjoined with remote end 18 of second elongate pressure member 16. Sleeve 20 can be conjoined with second elongate member 16 by welding and forming or machining from a single piece of material. In view of the intended high pressure applications, it is preferred that sleeve 20 be integrally formed as part of second elongate pressure member 16 by machining from a single piece of material . Second elongate pressure containment member 16 is of a greater thickness than sleeve 20. Second end 24 of sleeve 20 projects past remote end 18. Depending on the requirements of the application a notch 30 may be provided near remote end 18. Notch 30 is placed so that it is exposed to the low pressure. In the intended application, as will be further described in relation to FIGURE 2, interior surface 19 is subjected to fluid pressure that is high when compared to exterior surface 17. Notch 30 is, therefore, illustrated on exterior surface 17 adjacent remote end 18.

Thirdly, couple first elongate pressure containment member 12 and second elongate pressure containment member 16 with sleeve 20 positioned in close relation to surface of revolution 14. In the illustrated embodiment, second elongate pressure containment member 16 has been telescopically inserted into first elongate pressure containment member 12. There may be an initial interfererence fit or an initial gap 26 between sleeve 20 and surface of revolution 14. In either case, a successful seal may be maintained in accordance with the teachings of the present invention, as will hereinafter be further described. Fourthly, place first elongate pressure containment number 12 and second elongate pressure containment member 16 in an environment in which fluid pressure tends to cause a flow of fluids from second end 24 to first end 22 of sleeve 20 along an interfacial region 28 between sleeve 20 and surface of revolution 14. Fluid pressure tends to set up pressure gradient along interfacial region 28 which deforms sleeve 20 sufficient to seal interfacial region 28. Contrary to API recommended practice guidelines for sealing between overlapping 'cylindrical' members, the contact stress between sleeve 20 and surface of revolution 14 may be less than the contained pressure over the full pressure range. Where a notch 30 is included, fluid pressure, acting largely in an axial direction, has the further tendency to deform notch 30 to constrict the interfacial region and control the sealing contact stress distribution.

Once the underlying principles of the described method are understood, it will be appreciated that the length and thickness of sleeve 20 required to achieve the desired result will vary with the application, as will the width and depth of the notch 30 if included. Certain of these variables can, however, be reduced to a formula: L > L. where

L. ≥ 0.2_.1, without notch and L. ≥ 0.05Λ with notch, where wavelength,

r = average sleeve radius, . = sleeve wall thickness, and v -r- Poisson's ratio

With this method of sealing there can actually be an initial gap 26, as illustrated in FIGURE 1, between first elongate pressure containment member 12 and sleeve 20. Where such a gap exists an initial flow through the gap must be provided for and small gap size may be accounted for using laminar flow assumptions in the formula:

L₂ where,

g = gap size dependent on pressure induced deformations, μ = viscosity,

P = annular pressure,

Q-_ozx - maximum allowable seapage flow rate, and

P₀ = applied differential pressure for maximum flow within operating pressure range.

Sealing can be enhanced, as it is in other sealing applications, by providing a weak solid interface compound, such as grease, an interfacial region 28. This alters the flow behaviour of the annular material from that of a fluid to a weak solid which mechanism may be described by the formula: L> > where,

g = gap size dependent on pressure induced deformations, r = allowable static shear strength of interfacial material,

P = annular pressure, and

P₀ = maximum applied differential pressure

Care must be taken in having a clear differentiation between sleeve 20 and the balance of second elongate pressure containment member 16. If the transition in thickness is gradual, there is a danger that sleeve 20 will not function as intended. Sleeve 20 has to be sufficiently thin that it will deform to seal interfacial region 28 when subjected to the pressures of the intended application. It is, therefore, to be preferred that first end 22 of sleeve 20 be connected to remote end 18 of second elongate pressure containment member 16 over an interval of length not greater than one thickness of second elongate pressure containment member 16. The physical testing of the prototype will now be described to illustrate the utility of the present invention. The seal will be referred to as MICS seal; MICS being an acronym for Minimal interfacial contact stress .

Referring to FIGURE 2, a prototype seal assembly, generally indicated by reference numeral 10, was constructed according to the teaching of the present invention. Seal assembly 10 was then subjected to physical tests designed to verify the performance of seal 10 under the application of pressure, and friction loads associated with axial sliding along the seal surfaces.

Referring to FIGURE 2, seal assembly 10 consists of first pressure containment member 12 and second pressure containment member 16. Pressure containment members 12 and 16 were formed from bars and machined long internal bore cavities. They were then telescopically connected to form a fully enclosed pressure containment unit upon assembly. Second pressure containment member 16 was machined to incorporate a 12 inch long thin walled sleeve 20 at one of its ends, remote end 18. A radial clearance gap 26 of 0.0004 inches was provided between an outside surface of sleeve 20 and inside surface of revolution 14 of first pressure containment member 12; both of which surfaces constitute the seal area. Two holes 32 and 34 were drilled through first pressure containment member 12 and second pressure containment member 16 in order to accommodate fixtures for hydraulic lines supplying pressure fluid and connecting pressure measuring instrumentation.

Tests were conducted on both a "dry" specimen and also on a specimen having seal surfaces which were lubricated with grease. First pressure containment member 12 was then lowered onto second pressure containment member 16 to a predetermined depth. The unit was placed vertically between a top compression plate 36 and a bottom compression plate 38 of a testing machine (not shown) . A fluid containment vessel 40 was attached to bottom compression plate 38 to collect leaked fluid. The specimen was filled with water pumped into inside cavity 42 through the lower of the access holes, access hole 32. Internal pressure was applied first by adding extra pressure fluid. At each given threshold pressure the inflow line was shut off. The specimen was then compressed by the testing machine, with inside pressure building up. After the peak value was reached, the testing machine stroke was reversed, and the specimen was allowed to expand back until it reached the original threshold pressure. More water was pumped inside the specimen to bring up the pressure to the next threshold value, at which point the load cycle was repeated. During testing, no leak was detected from the lubricated specimen at any pressure during static pressure increases, nor during load cycles with sliding up to a pressure of approximately 5,500 psi . Above this pressure, seepage was observed during compression and expansion sliding driven by the testing machine, of magnitude less than 0.15 bbl/day. Seepage rates detected from the unlubricated test under static conditions, shown in FIGURE 3, exhibit an approximately linear dependence on the pressure above the activation pressure of approximately 2000 psi, with estimated leakage magnitude approximately 1 bbl/day at 8,000 psi. which is considered to be within acceptable limits.

FIGURE 4, represents friction loads obtained during the compression of the lubricated specimen. The abscissa represents the inside pressure; the ordinate corresponds to the difference between the total sliding load and end pressure load, and is therefore a measure of friction load. Based on the above values, the friction coefficient determining friction along lubricated surfaces was estimated at 0.088. Test data obtained during specimen expansion showed a relatively smaller difference between total load and the product of inside pressure times cross-section area, indicating smaller friction loads, especially under higher pressures . Friction coefficient for dry specimen was estimated in a similar manner and found equal to approximately 0.33, with friction loads up to 30,000 lbf.

The tests verified the ability of the MICS seal to ensure pressure integrity between two tubular members subjected to internal pressures up to 9,000 psi and relative axial movement of the seal surfaces . Seepage rates obtained from the lubricated specimen during load cycles under high pressures and from the dry specimen, were very low. Friction loads associated with axial compression or expansion of the assembly under pressure were small, confirming that metal -to-metal sealing in combination with the further tendency of the largely axial fluid pressure load between the pressured surfaces and the unpressured notch cavity to constrict the interfacial region and control the sealing contact stress distribution was achieved with minimized contact stress yielding acceptably low overall friction sliding loads for full scale applications. A seal geometry incorporating notch 30 but otherwise similar to that used for the physical tests, as shown in FIGURE 2, was analyzed using the finite element analysis (FEA) method. The mesh for the model constructed provided an initial notch of 0.00025 inches. The analysis considered the tendency of pressure to penetrate the contact region so that pressure was applied to all inside surfaces up to the seal location. FIGURE 5, shows the deformed mesh in the seal region after seal contact has been activated by flow or stress gradients in the grease along the interfacial region. The deformations are exaggerated to better visualize the ability of the gap to control the contact region geometry and hence contact stress distribution under pressure loading.

FIGURES 6 and 7, provide evaluation of the contact pressure. As shown, a wide seal contact width is achieved so that the low contact stress condition required for sliding is achieved. Also as shown in FIGURE 7, because no initial interference was used an activation pressure of 1366 psi is required. Used with the flow equations, this value allows appropriate selection of design parameters governing either the seepage flow or compound seal requirements. While FIGURE 2 illustrates one form of sealing assembly there are a variety of alternative sealing assemblies that can be constructed using the teachings of the present method. In each case high pressure acts upon sleeve 20. In each case fluid attempts to migrate from an area of high pressure at send end 24 of sleeve 20 toward first end 22. Referring to FIGURE 8, there is illustrated a manner in which sleeve 20 can be adapted to seal on an exterior surface of a tubular member with pressure coming from two directions. Referring to FIGURE 9, there is illustrated a manner in which two of seal assemblies 10 can be combined., to meet the needs of an application in which pressure is on both sides. Axis of symmetry 51 can be on either side of sleeve 20, as illustrated in FIGURE 1.

It will be apparent to one skilled in the art that the present method teaches an effective method of forming a metal - to-metal seal in high pressure applications with low contact stress. It will also be apparent to one skilled in the art that modifications may be made to the illustrated embodiment without department from the spirit and scope of the invention as hereinafter defined in the Claims.

Claims

THE EMBODIMENTS OF THE INVENTION IN WHICH AN EXCLUSIVE PROPERTY OR PRIVILEGE IS CLAIMED ARE DEFINED AS FOLLOWS:

1. A method of forming a metal-to-metal seal in high pressure applications with low contact stress in which a first elongate pressure containment member (12) is provided having a surface of revolution (14) , a second elongate pressure containment member (16) is provided having an exterior surface (17), an interior surface (19) , a remote end (18) , and a thin walled extension sleeve (20) having a first end (22) and a second end (24) , the first end (22) of the thin walled extension sleeve (20) being conjoined with the remote end (18) ; the first elongate pressure containment member (12) and the second elongate pressure containment member (16) are coupled with the sleeve (20) positioned wholly within the first elongate pressure containment member (12) in close relation to the surface of revolution (14), the method comprising the step of: placing the first elongate pressure containment member (12) and the second elongate pressure containment member (16) in an environment in which fluid pressure exceeds contact stress at some point of an operating range and tends to cause a flow of fluids from the second end (24) to the first end (22) of the sleeve (20) along an interfacial region (28) between the sleeve (20) and the surface of revolution (14) , such fluid pressure and flow tending to set up a pressure gradient along the interfacial region (28) sufficient to deform the sleeve and promote sealing along the interfacial region (28) .

2. The method as defined in Claim 1, the sleeve (20) being integrally formed as part of the second elongate pressure containment member (16) .

3. The method as defined in Claim 1, a narrow radially extending circumferential notch (30) being placed near the remote end (18) of the second elongate pressure containment member (16) on one of the exterior surface (17) and interior surface (19) where the fluid pressure is less than on the other of the exterior surface (17) and the interior surface (19) , such that the notch deforms in response to fluid pressure acting in an axial direction, tending to force the sleeve (20) into engagement with the surface of revolution (14) .

4. The method as defined in Claim 1, having remote end (18) of the second elongate pressure containment member (16) of substantially uniform thickness, the sleeve (20) of substantially uniform thickness, and the first end (22) of the sleeve (20) connected to the remote end (18) of the second elongate pressure containment member (16) over an interval of length not greater than one thickness of the remote end (18) of the second elongate pressure containment member (16) .

5. The method as defined in Claim 1, having the thickness and the length of the sleeve (20) determined in accordance with the following formula:

L > L. where

L. ≥ 0.2.1, without notch and __, > 0.05.1 with notch, where λ = — = characteristic shell wavelength,

r = average sleeve radius, / = sleeve wall thickness, and v - Poisson's ratio

6. The method as defined in Claim 1, having an initial gap between the first pressure containment member (12) and the second pressure containment member (16) , the thickness and length of the sleeve (20) necessary to compensate for such initial gap being determined in accordance with the following formula : ^Lι ^{> where}»

g = gap size dependent on pressure induced deformations, μ ~ viscosity,

P = annular pressure,

Q_ntl. = maximum allowable seapage flow rate, and

7. The method as defined in Claim 1, having an initial gap between the first pressure containment member (12) and the second pressure containment member (16) with a weak solid interfacial compound disposed in the interfacial region (28) , the width and length of the sleeve (20) necessary to compensate for such initial gap being determined in accordance with the following formula:

where,

g = gap size dependent on pressure induced deformations, τ = allowable static shear strength of interfacial material, P = annular pressure, and P₀ = maximum applied differential pressure.