GB2195726A - Cryogenic swivel joint - Google Patents

Abstract

A swivel joint for use in a cryogenic pipeline comprises two sections, which are pivotally interconnected by means of a low friction bearing. The bearing includes at least one side bearing member 6, 14, 15 consisting of a sintered bronze metal of which the pores are filled with a polyfluoroethylene and lead containing compound. <IMAGE>

Description

SPECIFICATION Cryogenic swivel joint The invention relates to a swivel joint for use in a cryogenic pipeline.

In cryogenic pipeline systems, such as flowlines for transport of liquefied natural gas, swivel joints and other movable components are subject to extreme mechanical and thermal loads. Hence, roller or ball bearing systems, commonly available for swivel joints, require under cryogenic conditions of use frequent maintenance and replacement due to the poor tribological conditions met during operation because of the lack of a suitable fluid lubricant for the wide temperature range of ambient to cryogenic.

Thus there is need for a swivel joint which, under cryogenic conditions of use, is more reliable and less prone to wear than the currently available swivel joints.

Therefore, it is an object of the present invention to provide an improved swivel joint for cryogenic conditions of use.

The swivel joint according to the invention comprises two sections, which are pivotally interconnected by means of a low friction bearing including at least one slide bearing member consisting of a sintered bronze metal of which the pores are filled with a polyfluoroethylene and lead containing compound.

The invention will now be explained in more detail, by way of example, with reference to the accompanying drawing, which shows a longitudinal section of the upper part of a cryogenic swivel joint according to the invention.

The swivel joint shown in the drawing comprises a first, fixed, section 1 and a second, movable, section 2. The first section 1 comprises an end portion 3 on which a first tubular slide bearing member 4 is mounted. The second section 2 comprises a tubular end portion 5 comprising at the inside thereof a second tubular slide bearing member 6 which surrounds the first slide bearing member 4.

The first slide bearing member 4 is fixed to said end portion 3 of the fixed section 1 by means of a series of locking dowels 8.

The second slide bearing member 6 is glued to the inner surface of the tubular end portion 5 of the movable section 2 and consists of a sintered bronze metal of which the pores are filled with a polytetrafluoroethylene and lead containing compound.

The tubular end portion 5 of the movable section 2 consists of two parts 5A, 5B which are secured to each other and to the remaining part of the movable section 2 by means of bolts 9 and 10. Each part 5A, 5B comprises a ring-shaped stop shoulder 11, 12, respectively, which is located adjacent an end of the first slide bearing member 4. Each stop shoulder 11, 12 is provided with a thrust washer 13, 14 made of a sintered bronze metal of which the pores are filled with a polytetrafluoroethylene and lead containing compound. Said thrust washers 13, 14 are fixed to said stop shoulders 11, 12 by means of a series of locking bolts 15. A series of sealing rings 16 is arranged between the fixed and movable section to provide the necessary sealing between the interior 17 of the joint and the exterior 18 thereof.The joint may further be surrounded by a body of thermal insulation material (not shown), which isulation material may consist of an ice sheeth formed in-situ.

The present invention is at least partly premised on the insight that under cryogenic conditions of use grease-like lubricants generally fail to provide the necessary lubrication of the movable parts.

In view thereof an extensive test programme was set up during which "dry" roller bearing and low friction bearings were compared on performance and reliability under cryogenic conditions.

It was found that, contrary to expectations, roller and ball bearings, which would normally be considered first for such use, gave under similar test conditions much higher wear rates and failed more quickly than low friction bearings which were provided with a slide bearing member consisting of a sintered bronze metal of which the pores are filled with a polytetrafluoroethylene and lead containing compound. It was further found that, under cryogenic conditions of use, plain swivel bearings of the above type had a lower friction coefficient and lower wear rate than expected from ambient temperature test data.Low friction bearings made of a sintered bronze metal of which the pores are filled with a polytetrafluoroethylene and lead containing compound are known, for example from British patent specification 1,114,061, and they are marketed by the Glacier Metal Company Limited (U.K.) under the trade name GLACIER DU (registered trade mark).

Test results A suitable test rig was designed and constructed to enable a test programme to be carried out.

It was assumed that the bearing should be mounted in a centrewell bi-fluid swivel and an estimation was made of the bearing requirements for such a swivel, i.e. 2m diameter with 25 mm wide bearing track carrying an axial load of 35 tons. The estimated specific loading on the bearing of such a swivel was found to be equivalent to a load of 5000 N on a test bearing of the size used on the test rig. It was decided to carry out both continuous rotation and oscillatory rotation tests. The speed at which these tests could be carried out were limited by experimental constraints to 60 rpm for continuous rotation and 60 rpm motor speed for oscillatory motion of + 450 which is equivalent to 30 rpm mean rotational speed of the bearing.

On the basis of ambient temperature published data for GLACIER DU, it was calculated that a plain bearing constructed in the manner described in the accompanying drawing would have a useful life of 670 hours in the continuous rotation test outlined above and 1400 hours in the + 450 oscillatory test. Further, it was estimated that a friction coefficient of 0.15 would be expected in both cases.

With the test rig cooled to liquid nitrogen temperature (-196 C), the continuous rotation test was run for 50 hours. Again on the basis of ambient temperature data, it was estimated that this should have produced 13 Am of wear, but in practice none was detected. It was calculated that this test was equivalent to 1911 days (over 5 years) of operation of the postulated bi-fluid swivel assuming the latter undergoes the equivalent of 3 complete revolutions of the bearing per day. A friction coefficient of 0.13 was obtained for this continuous rotation test.

In an oscillatory rotation (+ 45 ) cryogenic bearing test in otherwise similar conditions, the bearing was originally run for 200 hours.

The expected 16 #m of wear (from ambient temperature data) was absent, so the test was extended to 500 hours. Again, in place of the expected 24 Am loss of material, no wear was recorded. This was equivalent to 9555 days (over 26 years) on the hypothetical bi-fluid swivel. In this oscillatory motion test, a friction coefficient of 0.10 was recorded.

Thus, testing of GLACIER DU under cryogenic conditions has revealed a lower wear rate and lower friction coefficient than expected at ambient temperature.

In order to obtain some measurable wear on a GLACIER DU swivel bearing within a reasonable test duration, it was decided to increase the specific loading. In fact the specific loading was raised by a factor of 5 by skimming 80% of the surface away from a test bearing.

This was subjected to a +45 oscillatory rotation test in liquid nitrogen under a 5000N load. Motor speed was 60 rpm giving a mean bearing rotational speed of 30 rpm. This arrangement gives a Modified PV of 1.4 N/mm2 x m/s (where P = Specific bearing load, V = Rubbing velocity at mean diameter of thrust bearing, "Modified" implies that account is taken of mating surface factor, application factor and bearing size factor). Although Glacier specify 1.75 N/mm2 x m/s as the practical upper limit of PV for continuous operation, their design data stops at PV = 1.0 for thrust washers. The figures for PV = 1.4 can be obtained by extrapolation, resulting in a room temperature predicted bearing life of 90 hours and a friction coefficient of 0.11.This estimation is no more than a crude upper limit of bearing life since frictional heating -effects become important at high PV values, but it provides a point of comparison for the results of the cryogenic bearing test.

At intervals during the cryogenic bearing test, wear was monitored by measuring the wear step height with a surface roughness monitor at each of six locations around the bearing. The bearing can be considered to have reached the end of its useful life when approximately 50 ,um has been removed from its surface. This condition was reached after only 45 hours in the first test at PV = 1.4, which compared unfavourably with the rather imprecise upper estimate of lifetime from room temperature design data, i.e. 90 hours.

This was originally explained in terms of the large frictional heating in high PV conditions.

The test was then repeated a number of times with greater attention paid to effective cooling. Wear data was found to vary significantly, presumably as a result of the marginal operating conditions. The best result gave 50 ,um wear after 100 hours operation, which compared well with the ambient temperature estimate.

These high PV tests were not very meaningful in generating design data for a real cryogenic swivel, since the load/speed condition was unrealistically high. However, it did confirm the viability of this bearing material for cryogenic service at the limit of its recommended operating conditions.

Furthermore tests were carried out on the same test rig with rolling element bearings.

The fact that the rolling element bearings tested were similar in size to the dry sliding bearings investigated provides a rough assessment of the relative performance of these two different tribological approaches.

The rolling element bearing tests involved running a flat steel test washer (identical to those used in the dry sliding bearing tests) against 18 steel balls (8 mm diameter) located in a bronze cage and supported on a grooved steel ring. This arrangement could not be loaded to more than 1000N (cf 5000N used for dry sliding bearing tests) without producing a permanent indentation of the raceways. Various tests were carried out to assess the effect of various relevant parameters.

Cryogenic testing was carried out in liquid nitrogen and the results compared with a similar test in dry nitrogen gas at room temperature to illustrate any temperature effect. A room temperature test in air was also executed to demonstrate the effect of oxygen.

Oscillatory rotation testing was carried out both at + 0.5 as well as j 450 to see the effect of small angle movements, which could be important if a fretting-type mechanism operates. Finally, a test was conducted in which the bearing was coated with the dry film lubricant Molykote 340CA.

In all these tests, the total bearing motion was the same. This was achieved by testing for 10 hours at 120 rpm motor speed for the + 450 test and 200 hours at 540 rpm for the + 0.5 test. The distance moved by the bearing is the same as that achieved in 200 hours by the + 450 oscillatory rotation test on GLA CIER DU which, with five-times the applied load, produced no measurable wear. Significant raceway wear was found to occur in these rolling bearing tests. It was demonstrated that exclusion of oxygen reduced wear rate, while further improvement ensued from cryogenic operation. Wear scars in the +0.5 tests were approximately twice as deep as those in the +45 tests, but this was thought to result from wear being concentrated on a smaller area.

Claims (6)

1. A swivel joint for use in a cryogenic pipeline, the joint comprising two sections, which are pivotally interconnected by means of a low friction bearing including at least one slide bearing member consisting of a sintered bronze metal of which the pores are filled with a polyfluoroethylene and lead containing compound.

2. The joint of claim 1, comprising a low friction bearing including a first and a second slide bearing member, which members have a tubular shape, said second member being secured at the inner side of a tubular end portion of one of said sections which portion is co-axial to and surrounds said first tubular slide bearing member, said first member being secured around the other section of the joint.

3. The joint of claim 2, wherein said tubular end portion comprises a pair of ring-shaped stop shoulders, that are mounted at opposite ends of said first slide bearing member.

4. The joint of claim 3, wherein each of said stop shoulders is provided with a thrust washer which cooperates with one of said ends of said first slide bearing member, each thrust washer consisting of a sintered bronze metal of which the pores are filled with a polyfluoroethylene and lead containing compound.

5. The joint of claim 1, wherein the joint is surrounded by a body of thermal insulation material.

6. A joint as claimed in claim 1, substantially as described hereinbefore with reference to the accompanying drawing.