GB2072072A

GB2072072A - Fusion butt-welding pipes

Info

Publication number: GB2072072A
Application number: GB8009323A
Authority: GB
Original assignee: Kobe Steel Ltd; Chiyoda Chemical Engineering and Construction Co Ltd
Current assignee: Kobe Steel Ltd; Chiyoda Chemical Engineering and Construction Co Ltd
Priority date: 1980-03-19
Filing date: 1980-03-19
Publication date: 1981-09-30
Also published as: GB2072072B

Abstract

A pipe is welded by a TIG hot wire process under conditions which satisfy the relationship <IMAGE> wherein W (mm) is the width of the welding groove, D (mm) is the diameter of the wire, F (mm/sec) is the feed rate of the wire and V (mm/sec) is the welding rate. The groove is of an "I" or "U" shape and a small width. The ratio V/W (sec) is preferably less than 42 (sec), more preferably less than 33 (sec).

Description

SPECIFICATION Method of welding a heatproof pipe This invention relates to a method of welding a heatproof pipe which provides a weld zone of the heatproof pipe with a high creep strength.

Reform tubes and cracking tubes for use in the petrochemical industry are generally made by centrifugal casting. It is, however, technically impossible to make long tubes directly by centrifugal casting.

Accordingly, several short tubes are joined together to provide a predetermined length for use in the petrochemical industry, the tubes being inherently capable of withstanding severe conditions of 800 to 100000. Of course, the joints should have a mechanical strength comparable to that of the tubes. In the past, hand welding by skilled workers was predominant since the need to prevent welding defects outweighed the need for efficiency. Hand welding was disadvantageous in that quality varied depending upon the skill and physical condition of the workers, and welding efficiency was very low. There has been a tendency to switch from hand welding to automatic welding in order to ensure consistent quality and high efficiency. In the case of the latter, welding conditions are selectable within a wide range.Extensive analysis on optimum welding conditions has indicated that creep strength varies greately depending upon welding conditions. The present invention is a result of further investigation thereon.

According to the present invention there is provided a method of welding a heatproof pipe wherein shield arc welding is carried out in the TIG manner with inert gas, through the use of hot wire under conditions which satisfy the relationship

wherein W (mm) is the width of a groove, D (mm) is the diameter of a wire, F (mm/sec) is the feed rate of the wire and V (mm/sec) is the welding rate, thereby welding the heatproof pipe in a butt joint within the groove of an "I" or "U" shape and a small width.

Although the definition of the above condition will be made more fully clear from the following description, it is of importance that the thickness of respective layers is comparatively greater in butt-welding a pipe in a multi-layer fashion. In the priorartwelding method, on the other hand, the thickness of respective layers is very small and the resulting creep strength is lower than that with the present invention and typically about 80% of that of base material in the case of MIG welding and TIG welding and about 65-70% of that of base material in the case of covered arc welding.

In the accompanying drawings: Fig. 1 is a graph of various welding conditions; Figs. 2A to 2C show schematically the deposition of metal; Fig. 3 is a graph showing the relationship between hot wire feed rate and creep rupture time; Figs. 4A to 4C are cross sectional plan views showing the growth of the deposited metal; Fig. 5 is a photomicrograph showing the cross section of the deposited metal; Fig. 6 is a graph showing the relationship between welding rate and creep rupture time; Figs. 7A and 7B are photomicrographs showing longitudinal sections of weld zones corresponding respectively to points A and B in Fig. 1; and Figs. 8A-1, 8A-2, 8B-1, 8B-2, 8C-1, 8C-2, 8D-1, 8D-2, 8E-1 and 8E-2 are photomicrographs of specimens corresponding to points A to E in Fig. 1.

In an experiment, welding beads of various thicknesses were formed in a groove prior to TIG welding using a hot wire. Fig. 1 is a graph relating welding on a substantially "I" shaped groove (groove angle = 4 ) having width W of 6 mm wide, wherein the ordinate indicates the welding rate and the abscissa indicate the feed rate of the hot wire. Straight lines showing a linear relationship indicate the thickness H in mm (calculated values) of respective passes.

The curves on the graph represent the minimum current capable of preventing inferior welding and the right upper regions with respect to the respective curves indicate that welding should be carried out with the values of current as depicted by the one more right and upper curves. The points A, B, C, D and E in Fig. 1 show respective welding conditions which are described below.

Other welding conditions were as follows: TIG electrode: 8.2 mms Hot wire: TGS-310 HO 1.2 mmo Power supply: DCSP Shield gas (Ar): Inside: 17.5 limin Outside: 12.5 limin Base material: HK-40 Inner diameter: 108 mm Outer diameter: 140 mm Current values A, B, D, E: 300 A C: 350 A Fig. 3 shows the results of two series of creep rupture tests, in one of which the stress a = 2.0 kg/mm2 and the other of which (r = 2.8 kg/mm2, on deposited metal specimens obtained under the welding conditions A, B and C in Fig. 1, with the hot wire feed rate as abscissa and the creep rupture time as ordinate.

Figures 8A-1, 8A-2, 8B-1, 8B-2, 8C-1 and 8C-2 show the cross sections of the specimens (A-1), (B-l ), (C-1), (A-2), (B-2) and (C-2) corresponding to the open circles in Fig. 3.

The thickness per layer was small (H = 0.5) as depicted in Fig. 1 with the cross sections of the weld zones shown in Fig. 2A at the respective points (A-l ) and (A-2). In Figs. 2Ato 2C, the lines in the respective layers show the direction of growth of columnar crystals, open arrows show movement of heat and the size and number of the arrows show the amount of heat released. When the thicknesses were small and H = 0.5, the columnar crystals were primarily grown along the thickness as indicated in Fig. 2A and the creep rupture time was relatively short as indicated by (A-1) and (A-2) in Fig. 3. Where the hot wire feed rate was increased with 300 A of current as indi The drawings originally filed were informal and the print here reproduced is taken from a later filed formal copy.

cated by B in Fig. 1, the amount of heat release was increased in parallel with the thickness as indicated in Fig. 2B and the creep rupture time was substantially a maximum as indicated in Fig. 3. The deposited metal developed under the welding condition as defined by the point C in Fig. 1 had the cross section shown in Fig. 2C wherein the columnar crystals were remarkably grown in parallel with the thickness and a tendency to gradually shorten the creep rupture time as indicated in Fig. 3.

On the basis of those results, a lower limit of the hot wire feed rate when the creep rupture time was longer than that with C-1 and C-2 in Fig. 3 was approximately 290 cmlmin for the two series in Fig.

3. Taking this valve on the Fig. 1, and evaluating its intersections with the welding rate used at points A, B and C (about 350 mm/min), gave a thickness of approximately 1.6 mm. Thus, the thickness per layer should be at least 1.6 mm. The amount of the wire -- welded per unit time is generally defined by

wherein W (mm) is the width of the groove, D (mm) is the diameter of the hot wire and V (mmlsec) is the welding rate. The amount of the deposited metal is then W x H x V. As stated above, the minimum thickness H is 1.6 mm when W = 6 mm.Because (HIW) ~ (1.6/6) the amount of the wire welded and the deposited metal are correlated as follows:

The above formula can be rewritten:

This is the minimum requirementforthe present invention.

A cross section which seemed to be a counter surface of the columnar crystals can be seen in Figs.

8C-1 and 8C-2, corresponding to (C-i) and (C-2) in Fig. 3. As HIW increased as indicated in Fig. 2C, the counter surface atthe growth end of the columnar crystals exhibited anisotropy and a high possibility of creep rupture thereat.

Fig. 7A shows a longitudinal cross section of a weld zone corresponding to the point A in Fig. 1 and Fig. 7B corresponds to the point (B) in Fig. 1. Fig. 5 showing a photomicrograph of the central portion of the deposited metal in an enlarged scale corresponding to the point C in Fig. 1. Whereas no growth of the columnar crystals parallel to the thickness can be observed in Figs. 7A and 7B, a well defined counter surface can be recognized at the central portion in a vertical direction in Fig. 5.

Analysis of the welding conditions for avoiding the formation of the counter surface reveals that the creep rupture time could be prevented keep from becoming shorter even when H1W was the same and the welding rate was sufficiently low. Fig. 6 illustrates the results of creep rupture tests on the deposited metal specimens made under the conditions E, D and C of Fig. 1 just as the two series in Fig. 3. In each case, as the welding rate decreased the creep rupture time became longer and typically started increasing at 250 mmlsec and became substantially longer at 200 mmlsec.Generalizing the relationship between the welding V (mmlsec) and the groove width W (mm) in terms of VIW, VIW was about 42 (sec) and about 33 at 250 mmlsec and 200 mmlsec respectively. Therefore, VIW is preferably less than 42 and more preferably less than 33.

Figs. 4a, 4b and 4c are explanation diagrams for the above described conclusion at a low welding rate a, a middle welding rate 6 and a high welding rate c, wherein welding advances downwardly in the drawings and the solid lines represent the direction of growth of the columnar crystals and the dotted lines represent isothermal lines. When the welding rate was low, the isothermal lines traversed the centre of the groove width and the columnar crystals were grown along the welding advance direction. Accordingly, in the case where the columnar crystals were observed to face within a longitudinal cross section as indicated in Fig. 2C, the direction of growth of the columnar crystals viewed from a plan view was as depicted in Fig. 4a and it was possible to prolong the creep rupture time as at the points (E-i) and (E-2) in Fig. 6. Figs. 8D-1, 8D-2, 8E-1 and 8E-2 are photomicrographs of the specimens corresponding to points (D-1), (D-2), (E-i) and (E-2) in Fig. 6 wherein the counter surface as in Figs 8C-1 and 8C-2 disappeared.

Claims

1. A method of welding a heatproof pipe wherein shield arc welding is carried out in the TIG manner with inert gas, through the use of hot wire under conditions which satisfy the relationship

wherein W (mm) is the width of a groove, D (mm) is the diameter of a wire, F (mmlsec) is the feed rate of the wire and V (mmlsec) is the welding rate, thereby welding the heatproof pipe in a butt joint within the groove of an "I" or "U" shape and a small width.

2. A method according to Claim 1, wherein the ratio (VAT) (sec) of the welding rate V (mmlsec) to the width W (mm) of the groove is less than 42 (sec).

3. A method according to Claim 1 wherein the ratio (VIW) (sec) of the welding rate V (mmlsec) to the width W (mm) of the groove is less than 33 (sec).

4. A method of welding a heatproof pipe as claimed in Claim 1, substantially as herein described,