CA1125383A - Welding process for production of a steel pipe - Google Patents

Abstract

WELDING PROCESS FOR PRODUCTION OF
A STEEL PIPE

ABSTRACT OF THE DISCLOSURE

Disclosed herein is an improvement of a welding process for welding a steel pipe, wherein a gas metal arc welding is performed to form the first welding layer and a submerged arc welding is performed to form the last welding layer. According to the conventional gas metal arc welding process, i.e. the MIG or CO2 welding processes, the combined use of a high welding current and a welding wire having a small diameter is known to bring about the rotation of the welding arc and the formation of an undercut along the toe of the weld metal. The purpose of the present invention is to weld a steel pipe having either a large thickness or an excellent ductility at a temperature of less than -40°C, or both. According to the present invention, the combined use of a high welding current and a wire having a small diameter is possible, by employing a gas mixture containing an inert gas as a major part thereof and CO2 as an additional part thereof as a shielding gas. In addition, by adequately selecting the wire extension, the gas metal arc welding is improved to provid the features of: deep and round penetration; stable arc formation with stiffness of the arc, and; high metal deposition rate.

Description

~ ~2~i3~33 The present invention relates to a welding process for producing a steel pipe, and more particularly, a welding process according to which a high quality steel pipe, particularly a steel pipe having a large wall thick-S ness, is welded at a high production rate.
Steel pipes with a large diameter are generally produced by a UOE process, a spiral process or a bending roller process, and these processes include a weldiny step. However, with regard to the welding step, there are presently widespread demands for a practical welding process to attain the high welding speed, and thus the high production efficiency as well as the superior quality of the steel pipes.
When a steel pipe, having a làrge diameter, for use as a pipeline for transporting petroleum or natural gas is produced by seam welding, a submerged arc welding process, hereinafter abbreviated as an SAW process, is usually adopted.

`.

.,.. ,, ,.. ~, ~

~ .; .:
:, , ,... : :

r ` ~
3 ~ 3~3~
.' . . - " ' .

, ..

It is an object of the present invention to provide a welding process for producing s~eel pipe of high quality at a high production efficiency.
The present invention involves a concept of: using a small diameter wire, i.e. a consumable, small-diameter wire, for improving the toughness o~ the weld zone of a steel pipe at low temperature; using a welding current considerably higher than that conventionally used for the small diameter wire, and; using as the shielding gas atmosphere of the GMA welding process a gas mixture in which carbon dioxide gas is added to and mixed with the -inert gas as the major part of the gas mixture~ As a result, deep penetration wi~h enough extension is achieved and the heating zone of the base metal is large in width, and thus, excellent first layer beads are stably provided.:
The GMA welding process fo~ producing the first-layer beads is combined with the saw process for the last layer, according to which the surface appearance of the beads is excellent, the shape of ~he toe of the weld is good, and a 20 smooth surface contour is easily obtained. As a result of such combination, a~ excellent quality of the weld zone of the steel pipe c~n ~e provided at high welding speed ana .
' :

.~ .

'B3 high production efficiency.
In accordance with the present invention, there is provided a welding process for the production of a steel pipe, comprising the steps of:
performing a gas metal arc ~GMA) welding under an atmosphere of a shielding gas and, thereby, forming a first welding layer on the abutted and grooved ends of a steel plate, and;
performing a subme.rged arc welding and, thereby, forming the last welding layer, the process being characterized in that;
in the gas metal arc (GMA) welding step, (1) a high electrical current (I) in ampere in the range defined by the formula:
500d ~ I ~ 500d - 150, wherein d indicates the diameter of the consumable welding wire in mm, is conducted through a small diameter wire having a diameter of from 0.8 to Z.4 mm, within the atmos-phere of a gas mixture containing mainly an inert gas and, additionally~ a carbon dioxide, wherein, the conduction of said current ~hrough said wire within said atmosphere suppresses the generation of a rotating arc, generates a stiffened arc due to the pinch force by the high current density, ensures the resistance of the arc against the : -de~lection of the arc against the magnetic blow, and enhances the penetration to the base metal, and (2), the wire extension ~R ) in mm between the end of a current contact tip and the bottom of the groove is selected ~y , ' , ~ , ,'' . "`; '' " ", ." ' ' '' ~ ':
:, . . . :. : :

5~3~3 the formula:
~ > lOd + 5, - so that the deposition rate of molten metal is essentially increased by the long wire extension. It is preferable that the arc be oscillated in the traversing direction to the weld line at a cycle rate of from 3 to 30 Hz and at a width of from 1 to 15 mm.
The present invention is illustrated in detail by comparison wi~h the prior art and in connection with the drawings, whlerein:
Figs. 1 (A) and ~B) illustrate a cross sectional view of the weld zone of a pipe as well as the penetration and ~he bead shape, when GMA welding is carried out by applying a high electrical current to a small-diameter 15 wire, (A) and (B) corresponding to the conventional process and the process according to the present invention, re-spectively;
Fig. 2 shows the relationship betwePn the wire diameter (d) in mm and the ~elding current (I) in amperes;
Fig. 3 (A) schematically illustrates a cross sectional view o~ the set-up of the conventional SAW
process;
Fig. 3 (B) is the same ~iew as in Fig. 3 (A) and illustrates an embodiment of the GMA welding process 25 according to the present invention;
Fig. 4 shows the relationship between the welding current (I) and the melting r~te of the wire per unit time, which relationship clarifies the resistance heating , " , ~ , , , , ,, , . ~ , : i :., :~ :
,- " . . , " - .

.~..2~3 effect dependence upon the ~7ire extension;
Fig. 5 schematically illustrates a cross sectional view of the beads build up by the conventional GMA welding process together with the SAW process;
Fig. 6 schematically illustrates a cross sectional view of the weld metal build up according to the present invention, and;
Figs. 7 and 8 schematically illustrates the double groove shapes usèd in the welding for the production of pipe.

. In the SAW welding in order to ensure the dimensional accuracy required for the pipes and to simplify the welding step, the base metal (l), as seen Fig. 3A, is closely.engaged at its ends~la)and(lb)through root ~ace (3)t which is formed 15 at these end5, so that a groove without a root gap is formed between these ends. Reference numerals (ll, 13) and 14 designate the flux, the welding wireg and current contact ~ip, respec~ively. In the SAW process t from the '~

.

'~ , ' ' ' ' , . .

:.
:: ..' 2S3~3~

point of view of production ef~iciency and economy, a multi electrode system with two or three electrodes has heretofore been used and the welding is carried out in one pass. The SAW process is believed by experts in the welding art to be superior to a gas metal arc welding process and the other welding processes from the point of view of coming closest to the production of line-pipes.
This is because the penetration and the bead shape are excellent, even at a high welding speed, due to a high welding heat input, and further, the production efficiency is high because of a high metal deposition rate per unit time. The term "gas metal arc-welding" is an term used in the welding art which collectively designates: MIG welding, in which a shield gas of an inert gas with, usually, oxygen mixed in a small amount is used; carbon dioxide, arc-welding, and; consumable electrode-arc-welding, in which a gas mixture of the carbon dioxide, an inert gas and a small amount o~ oxygen is employed for the shielding gas. Gas metal arc welding is hereinafter referred to as

2~ GMA welding.
Improved toughness, for example higher than 3.5 kg-m at -60C, of the weld zone of large diameter steel pipes ~or use as pipelines has recently been required, because the pipelines are being more frequently constructed in cold climate regions than before. In order to produce a steel pipe having a large wall thickness at a high production efficiency by the SAW process with multiple electrodes, the welding is carried out at a hi~h welding heat input o~ -x ,, ,. , ; :

i;
. ~ ; :: : , .

~.Z5383 ~,~

more than 60,000 Joule/cm when the plate thickness amounts to 25 mm. As a result of such a large welding heat input, the quality of the base metal is deteriorated, for example, the lmpact property is reduced at the heat affected zone of the base metal.
In a case where no measure is employed for providing the steel material with a satisfactory, low temperature toughness and such an ordinary steel material as stipulated in API Standard 5L is welded by the SAW process, the welding heat input is limited to a low level so as to achieve the toughness at low temperature, and thus, multi~
-layer welding is necessary for the welding operation. As a consequence of limited welding heat input, the amount of deposited metal per unit time is also reduced, and therefore, the production efficiency is reduced. In addition, the weld metal of the first layers, which are formed on the inner and outer surfaces of the pipe includes the root run pass of the first layers at which the weld metal is apt to be brittle through a stress-relief annealing. ~n order to avoid the SR (stress relief) embrittlement, an additional step of back chipping is necessary, and the production efficiency is further reduced by the back chipping.
The ~MA process has heretofore heen tried as an alternative for the SAW process, in order to find a way of ~elding a steel pipe at a high production e~ficiency, guaranteeing reqired toughness at low temperature, without using a special kind of steel material~ According to the ,,.. ~, ~, , , :; . . :

.: :
:
... '; :

` ~2~i38~

ordinary GMA welding process, the welding is carried out by means of an electrode wire with a relatively small diameter, ranging ~rom 0.8 to 2.0 mm, and the disadvantages of the process are generally stated to be: (1) the amount of deposited metal per unit time, and thus the welding ef~iciency is not very high; (2) the penetration depth is shallow, and; (3) welding defects are liable to be formed due to bead shape.
The welding efficiency will now be explained with reference to Fig. 2. In the conventional MIG welding process with a small diameter wire, the welding current generally used is in the range between the lines which represent the maximum and minimum current ~~
values, respectively. Fig. 2 indicating the relationship between the welding current (I) and the wire diameter (d).
In addition, in the GMA welding process using carbon dioxide gas as the shielding gas, the welding current used is in the range between the - - - o ~ - - lines re-presen~ing the maximum and minimum current values in Fig. 2. Howevert the upper, high current part of this range used for the GMA welding with CO2 gas is adopted mainly in a case of tack welding and, therefore, the lower part below the two dots chain line shown in Fig. 2 and representing I=250~d is used for ordinary GMA welding.
The relatively small amount of metal deposition per unit time is related to the welding current ranges mentioned above.
With regard to the ~enetration depth, this depth by :, . , ' ', ' ~ ' 2S3~33 5~, the common MIG welding process presents no difficulties, because the MIG welding is carried out usually at a low ~ welding speed of less than 50 cm/min, and further, the small diameter wire used in common MIG welding exhibits a good cocentration of the arc. However, since a high welding speed of more than 50 cm/min is required for the production of a steel pipe, the penetration depth is reduced to a rather low level in connection with the current value mentioned above. The common MIG welding process with a small diameter wire, mentioned above, has the bead shape as illustrated in Fig. l(A). As seen in Fig. 1(~), the bead has a bottom part which is extremely narrow in width and elongated and a top part in a convex form, which is also narrow in width. Since the penetration width is narrow at the bottom part o~ the bead (8), slight deviation o either the seam tracking or orientation of the arc may result in an incomplete penetration at the root face (3). In addition, since the cross sectional shape o~ the bead is longitudinally elongated, such welding defects as hot cracking and lack of fusion are liable to be generated.
Another of the GMA welding process~ i.e. the ~nown MIG welding process in which a high welding curre~t is passed through a small diameter wire, will now be explained.
When the high welding current is passed through the small diameter wire, the zone o~ the base metal heated by the arc may not be enough for welding the pipe, and the shape of the deposited metal is longitudinally elongated and ~M
;~ :

; .

~ ~53~3 ,o narrower in width than the shape of the metal (83 formed by the conventional MIG welding process with a small diameter wire tFig. l(A))~ The weld metal of the so formed convex shaped bead may exhibit a lack of fusion at s the toe of weld and cause an undercut along the bead. It is difficult ~o utilize the narrowly protruded part of the penetration for the penetration to the root face (3) of the ends ~la, lb) of the base metal (1), because the so-called finger shaped penetration according to the MIG
welding process with a high current and a small diameter wire is so narrow and elongated that the thinly protruded part of the penetration has a deep but very narrow pene-tration. In addition, a hot cracking is liable to be formed, partly because the deposited metal (8) is longi-tudinally elongated, as seen in a cross sectional illus-tration of the metal in Fig. l(A), and partly because the cooling speed of the deposited metal is high.
In addition to the reasons mentioned abo~e because of the following reasons, it is impossible to per~orm the MIG welding by using high current for the small-diameter wire. Namely, when the welding is carried out with the use of a high current for the small diameter wire, the so-called rotation phenomena of the arc are induced, which results in spatter, shallow penetration and a non-alignment of the bead shape. Consequently, a welding current, which depends upon the wire diameter in the ranges between the two lines --~ , as stated before, is used.
Japanese Laid Open Patent ~pecification No. 51-61452 . r .:

I/

and Japanese Published Patent Specification No. 53-9571 propose a GMA welding process, according to which process a high welding current and a large diameter wire are used to enhance the quality and the production rate of thick wall pipe. In this process, the high current and the large diameter wire are combined to reduce the current density o~ the wire, and thus, the pinch force of the current, thereby ensuring the spread o the arc, improving the bead shape and achieving enough penetration due to the high currentO In the GMA welding process with a high electrical current, the current density of the large diameter wire is low, desp,ite the high electrical current, and therefore, the arc may not be stable in the groove during the welding due to low arc stiffness. For example, when the ordinary GMA welding carried out under the condition of passing 300 A through the wire of 1.2 mm in diameter is compared with the GMA welding with a high current carried out under the condition of passing 800 A, through the wire of 4.0 mm in diameter, the current densities of the former and latter weldings are 256 A/mm2 and 64 A/mm2, respectively.
The current density of the large-diameter wire is consider~
ably lower than that of the small-diameter wire. Accord-ingly, the uni-directional property of the welding arc is deficient in the latter welding and the arc stability in the groove is thus poor. This arc instability is liable to cause welding defects, such as a lack of fusion, because the arc is even more unstable at a large current, due to the magnetic blow which is induced by a direct current r B

. .
, . . . . . .. . .

`.. . i `. `
. ` ; ` . . . . .
`~ ~ . ..

~Z5383 `~

used in the GMA welding process as a power source, and;
~urther, because the generation area of the arc at the tip end of the wire has a poor stability due to the large diameter of the wire. In conclusion, the main purpose of the GMA welding process with a large diameter wire is to reduce the electrical current density, and thus the pinch force, so that the arc stiffness is low. As a result of the low pinch force, the magnetic blow is likely to occur and the arc is likely to deflect. In addition, with regard to the penetration, the obtained penetration is not very deep because of the low current density. Furthermore, the specific melting rate of wire relative to the current is inferior to that of the common GMA welding process with the large diameter wire. Since the last welding layer is formed on the widest part of the groove, such problems as lack of width or a bad shape of the bead~ and also, lack of fusion, undercut and non alignment of the reinforcement of weld result. The shortcoming of GMA welding process with the high current and the large diameter wire as compared to the SAW processJ resides in the point that such problems as mentioned above arise.
Japanese Laid Open Patent Specification No. 51-92750 proposes a w~lding process, in which intermediate welding beads (6) and (9), as illustrated in Fig. 5r are piled up by 25 the GMA welding process and beads [7~ and (10) are deposited by the SAW process on the beads (6~ and (9), respectively, to a thickness of from 2 to 7 mm. In this welding process, the intermediate beads ~6~ and (9~, i.e. the last beads by the r~

.,, . ~-. ,

3~3 .3 y~

GMA welding, have a cross sectional shape which is convex and narrow in width, as explained before in connection with the GMA welding process with a small-diameter wire, and ~urther, have a lack of ~usion at the toe of the weld.
In addition, the beads (7) and (10) spread laterally with a depth of from 2 to 7 mm~ As a result, the beads (7) and (10) exhibit very low penetration, and a lack of fusion (12) as shown in Fig. 5 is liable to occur due to the formation of thin beads (7, 10) on the beads (6) and (9), respectively.
In addition, the last layer by the SAW process has a small thickness of 2 to 7 mm, it is necessary to pile up the beads in the ~roove close to the pipe surface by the GMA
welding process in which the metal deposition rate, and thus, the welding efficiency are usually low.
The Japanese Laid Open Patent Specification No.
52-3543 discloses a welding process, in which the last welding layer is formed by the SAW process subsequently to the GMA welding process, as in the Japanese Laid Open Patent Specification No. 51-92750~ However, in the welding process of the Laid Open Specification No~ 52-3543, during the SAW process for the last layer an electrode having a rectangular cross section is positioned so that the long sides of the electrode are substantially perpendicular to the welding Iine. The rectangular electrode is used in the SAW process with a relatively low welding heat input to stably form a surface bead which is thin but wide. In the Japanese Laid Open Patent Specification Mo. 52-3543 no measure is disclosed to eliminate the lack of fusion (12]

B

.. ~..,, . . " ..~ .... . .

`
. . . . ..

53~3~
- 13a -shown in Fig~ 5. The GMA welding process of the Japanese Laid Open Patent Specification No. 52-3543 is carried out ~at a low electrical current density as in the Japanese Laid Open Patent Specification No. 51-61452 and Japanese Published Patent Specification No. ~3-9571, which involve the difficulties explained before.

According to the present invention, a small-diameter wire having a diameter ranging from 0.8 to 2.4 mm, preferably from 1.2 to 1.6 mm, is fed at a high speea, for example from 25 to 35 m/min with regard toJfox example, a 1.2 mm diameter wire; a shielding gas atmosphere contains an inert gas as a major part thereof and a carbon dioxide gas as an additional part thereo~, and; a welding current is adjusted to a hiyh level in such a manner as to realize a 5 spraying arc with high current density by the formula:
500d ~ I (A) 2 500d - 150 ... (1) The inert gas is at least one member selected from the group consisting of argon and helium, but argon is commonly used. The welding current according to the ~ormula ~1) is reduced to a current density ranging from 400 to 530 A/mm2 with regard to, for example, a 1.2 mm diameter wire. Such a high current or current density has been deemed to be unsuitable for welding, as stated above. When a large current is used in combination wi~h the small diametex ' .
' ' ' " ' . .

. . ' ~ . ; ' 253~3 wire in the conventional MIG welding process, the wire becomes plastic due to the resistance heating in the wire extension and the form of the wire tip end is changed into a long conical shape. Consequently, as is known by experts in the art, the arc is changed to a rotating spray arc, with the result that spatter at an appreciable level, shallow penetration and non-alignment of the bead shape occur. However, according to the present invention, since carbon dioxide gas is added to the shielding gas so as to form a shielding gas mixture, the conical part of the wire tip end is shortened and the rotating spray arc is sup-pressed. It is accordingly possible to take advantage of the so-called stiffened arc thinly squeezed by the pinch force. Namely, deep penetration is provided due to the fact that the arc is thinly stiffened, and the resistance of the arc against the magnetic blow and deflection is enhanced due to the fact that the arc is stiffened by the pinch force. Furthermore, the amount of metal deposition per unit time, and thus the welding efficiency are increased due to the high welding current. The shielding gas may contain a small amount of oxygen without any effect on the ad~antages mentioned above~ -The welding current (I) in amperes is selected/
depending upon the wire diameter (d) in mm, within the range defined by the formula:
500d ~ I > 500d 150, for the following reasons. ~amely, the wire extension is long according to the present invention, enough pinch ... :

.

force is exerted on the arc column, and the wire is oscil-lated according to a preferable feature of the present -invention. The relationship 500d ~ I must be satisfied in order to avoid abnormal vibration of the wire tip end of the wire and the arc. Namely, lf ~he welding current exceeds 500d, the resistance heating becomes excessive and brings about abnormal vibration. In other words, if the welding current (I) exceeds the upper limit and becomes excessively high, the wire unstably melts due to the rotating of the arc and the molten pool becomes unstable due to excessively large pressure of the arc. When the welding current (I) is lower than the lower limit defined by the formula:
500d - 150 the electrical current value reguired for exhibiting enough pinch force is not provided, and the high efficiency required for the welding for the production of pipe is not achieved. In other words, below the lower limit the melting rate of the wire is too low to achieve a high welding efficiency.
The diameter of the electrode wire is from 0.8 to 2.4 mm for the following reasons. When the diameter of the welding wire is greater than 2.4 mm, the electrical resistance heat generated in the wire tip extension is considerably lower than that of the wire with a diametex of less than 2.4 mm. The deposition rate is desirably increased over the rate proportional to the welding current when the wire diameter is 2.4 mm or less according to the :

, present invention, as explained later. However, such desirable increase is not achieved with a wire diameter above 2.4 mm. In addition, although the pinch force is desirably increased in such a manner as to stabilize the arc according to the present invention, such desirable increase is not achieved with a wire diameter above 2~4 mm.
On the other hand, when the welding wire diameter is less than 0.8 mm, the maximum allowable electrical current is too low to achieve a high deposition rate of the wire.
Furthermore, when a wire of less than 0.8 mm is fed at a high speed, such trouble as the buckling of the wire is liable to take place, and therefore, it is questionable whether the welding can be stably carried out on an industxi-al scale. The wire diameter from 1.2 to 1.6 mm is prefer-able.
The advantages of a high current density can befully utilized without inducing rotation of the arc, due to the fact that the shielding gas mixture is used for the welding atmosphere and, further, the current range denoted as D in Fig. 2, which was previously not adopted, is used.
According to the present invention, the distance between the front end of the current contact tip and the bottom of the groove, i.e. the wire extension in mm, is specified by the relationship:
~ ~lOd ~ 5 ................ 12), wherein the value d designates the diameter of the wire in mm. A significance of welding with a small diameter wire is that the rate of metal deposition per unit time, .. . . . .

s ..
. ~ . . .

~.~2~3~3 hereinafter, referred to as the deposition rate, deviates from the proportional relationship with the welding current -and increases greater than the value defined by this relation-ship when the welding current is increased. This phenomenon is understood to result from the resistance heating in the wire tip extension due to the welding current. In Fig. 4, the effect of the wire extension on the deposition rate of the small diameter wire is sche-matically illustrated by the solid line and the one dot chain line, indicating a smaller and larger wire extension, respectively. However, with regard to the large diameter wire, the heating effect at the wire tip extension is so small that the relationship between the deposition rate and the welding current slightly exceeds the dotted straight line, as seen from the three dot chain line. In order to ensure a high deposition rate for high efficiency welding with the large diameter wire, the current must, therefore, be increased, for example, up to point A in Fig. 4, with the result being that instability of the arc, such as a magnetic blow, is caused. In addition, the low current density, as stated before in connection with the Japanese Laid Open ~atent Specification No. 51-61452, is a reason for arc instability. On the other hand, with regard to the small diameter wire, even at a usual extension length (~) the resistance heating effect can be observed, as can be understood from the solid line and point B, although the electrical current value is still high. However, when the extension length is based on the formula of ~ ~ 10d ~ 5 , -- ~2S383 according to the present invention, the electrical current is reduced, as indicated by the one dot chain line and the point C in Fig. 4. Such a low electrical current as indicated by the point C is more advantageous for the stabilization of the arc and the bead shape than that indicated by the point B. In addition, as stated above, the current density and thus the pinch force, of the small diameter wire at the point C is far greater than that of the large diameter wire at the point A. Accordingly, a deep penetration effect is exhibited even when oscillation at high frequency is applied to the welding arc by oscil-lating the electrode wire at high frequency. For the reasons stated above, the wire extension * is specified to be larger than 10d + 5, wherein the value "d" is the diameter of the wire in mm. The wire extension should desirably be limited to below an excessive value which would bring about the excessive softening of the wire and an abnormal arc oscillation due to the oscillation at high frequency which is occasionally carried outO The maximum wire extensions (L) at 1.2 mm and 2.4 mm of the wire aiameter (d) are 30 mm and 50 mm, respectively.
In the conventional MIG welding process with a small diameter wire, wherein an argon gas with a mixture of oxygen in an amount of from 1 to 2% is used as the shielding gas, the welding current which is used for stable welding at a given wire extension has been experi-mentally determined to be not exceeding the maximum value defined by the two dot chaln line denoted as A/R in Fig. 4O

.:
:

This line represents the maximum melting rate at which the MIG-welding arc exhibits a stable axial spray transfer, and over which the rotating spray is generated. However, since the shielding gas atmosphere of mixed gases, mentioned above, is used in accordance with the present invention, the rotating-spray arc is suppressed even at the electrical current value exceeding the critical value of the A/R
line, and the high welding current can, therefore, be used, while a stable arc is generated. The shielding gas contains from 5 to 30~, preferably from 10 to 20%, by volume of the carbon dioxide gas, and the inert gas is essentially in balance and may contain a minor amount of oxygen. The welding voltage may be selected within an optional range, for example the range of from 30 to 40V, but is deriably a high voltage, for example ranging from 37 to 40V, in order to increase the width of the bottom of the penetration at the root face 3 (Fig. 1(A) ~ UP to an extent allowable in the weldingO
According to a preferable embodiment of the present invention, the current contact tip is oscillated, thereby oscillating the arc at a hi~h fre~uency in a direction transverse to the weld line. As a result, a flat bead and an excellent wide penetration can be provided. Namely, the heated zone of the base metal by the arc is extended and the penetration shape is widened by oscillating the arc at a high frequency in order to attain excellent fusion at the root face and excellent toe of the weld.
That is, as seen in Fig. 3~B), the current contact tip (16) , i , i;38~

is oscillated at a high frequency of three times or more per second (Hz), between the positions indicated by the ~ solid and dotted lines, so tha-t the arc is oscillated in the direction perpendicular to -the weld line. The width oE the penetration is then spxead, while the depth of the papillary penetration due to the high current density of the arc is decreased. Despite of this decrease, when a high current density arc is used, the depth of the pene-tration is sufficient for fusing the root face (3) of the pipe, which is produced by welding without forming a root gap between the ends of the base metal. In addition, the penetration is caused to spread in the width direction thereof. ~ccordingly, the so obtained penetration shape is suitable for the penetration shape at the root face (3).
In addition, since the oscillation at high frequency increases both the heated zone of the base metal (l) by the arc and the dispersion zone of the molten metal supplied from the wire (13), and further, since the arc force being applied to the molten metal is dispersed, a superior toe of weld is formedO In the welding of thick plate at high welding speed and low welding heat input, hot cracking of the beads is generally caused, due to the rapid cooling.
However, such problem is solved by the improved bead shape and penetration. In order that no zi~zag shape of the penetration be caused at a high welding speed, and also, in order that the arc force to be smoothly applied to the wide zone of the molten metal in the groove, the oscillation frequency is re~uired to be not less than three Hz. The . .
.. . . : , ,~ ............ , ~ ~

,, ~
..

53~33 oscillation frequency is required to be not more than thirty Hz, in order -that the molten metal at the front end of the wire will be stably transferred to the base metal.
The oscillation frequency is preferably from 5 to 20 Hz.
The oscillation width is from l to 15 mm for the following reasons. ~t less than l mm, the penetration cannot ef- -fectively be widened, while above 15 mm the prepared edge is excessively melted so that welding defects are liable to be formed on the toe of the first welding layers. The effects of the oscillation imparted to the current contact tip in the traversin~ drection to the weld line are, equa]ly obtained by the oscillation wherein the end of the current contact tip draws, the trajectory of a circle or psuedo circle, as one cycle of the oscillation, while the tip is advancing along the weld line. The trajectory of the oscillation at a high frequency is not limited specifically as long as the tra~ectory has a component traversing the weld line.
The SR embrittlement can be effectively prevented, according ~o another preferable embodiment of the present invention, by defining the length of the root face and the angle of the groove at the root. The 5R brittleness, which results from an appreciable dilution of the weld metal in the first layers by the base metal, can be prevented effectively, when the GM~ welding is performed under the shielding atmosphere oE the gas mixture mentioned above, so as to form the first welding layers on the double ~roove, which has, according to another preferable embodiment, a ... . - :. .
... . .
, : . : .

Z53~33 thickness of the root face ranging from 3 to 8 mm and a groove angle at groove bottom ranging from 45 to 160.
Generally speaking, the precipitation hardening elements enter into the weld metal of the first layers due to the dilution by the base metal and cause the SR brittleness.
When the SAW process is applied to the welding of the first layers, the dissolved oxygen and phosphorous, which promote the SR embrittlement, tend to be rich in these layers. When the GMA welding is applied to the welding of the first ~ayers, the dissolved oxygen and phosphorous are reduced and the SR embrittlement can advantageously be reduced.
In general, the thickness of the root face exceeds 8 mm, the bead is formed in a pear shape as seen in Fig. 1 (A). However, with such bead shape hot cracking may occasionally be generated. On the other hand, when the root face is less than 3 mm, such difficulties as burn--through may result, even though tack welding has been carried out.
A small groove angle is advantageous for the welding efficiency, but with a groove angle of less than 45, the bead becomes pear shaped and a welding crack may, therefore, be generated, when the first layer is formed. On the other hand, with a groove angle over 160 r the penetration amount is increased but the welding efficiency is con-siderably decreased due to the increase in the cross sectional area of the groove. It is desirable to adopt the groove shape as shown in Fig. 8 to decrease the groove ~2~33 area to as small as possible from the point of view of welding efficiency by the SAW process. It is obvious from ~ Fig. 8 that the groove angle at the surface welding layer by the SAW process may be less than 45, but the groove angle at the bottom should be ~5 or higher.
Finally, the welding of the last layer is explained.
Although the GMA welding process according to the present invention, as stated above, stably provides the bead of the first layer is free from welding cracks and which has a deep penetration, the bead shape presents a problem because the groove part of the pipe is widened at a position close to the surface of the pipe when the GMA welding is applied to the welding of the last layer. Namely, beads which are convex and narrow in width are liable to be obtained, because in general, the GMA welding process exhibits inferior toe of the weld than the SAW process, and also due to the melting ability of the high current density arc of the GMA welding process, which ability is high in the depth-direction but low in the width-direction of the arc.
TIG welding is not as practical for the welding of the last layer as the SAW process, because there is not enough reinforcement of the weld for a thick wall pipe by TIG welding, and further, the weldi~g properties of TIG
welding at high speed are inferior to those of SAW process.
In summary, the features of the SAW process as compared to those of GMA welding resides in the facts that: a wide bead having a shallow penetration can easily .;

: . ,,, . ... . ~ , -::

~2~ 83 be formed due to the spread of the arc, which has a low curren-t density, and; according to the characteristics of the slag shield type welding, the toe of the weld is e~cellent and the smooth bead can easily be obtained.
For the purpose oE reducin~ the cooling rate by increasing the welding heat input through the welding of the last layer, within the limit to ensure the toughness, the SAW process, which enables welding at the welding heat input ranging from 25 to 45 KJ/cm, must be applied to the last layer. ~lso, for the purpose of mitigating an unde-sirable increase of the hardness in the weld zone, which increase results from the low welding heat input of each welding layer and, thus, from a high cooling rate according to the GMA welding of the present invention, the SAW
process must be applied for the last welding layer, thereby annealing the previous layers.
~ s explained above, the weld zone of a steel pipe, particularly a thick wall steel pipe, with an improved quality can be provided, according to the present invention, by using GMA welding with high melting rate, and preferably high frequency oscillation, for the welding of the first layers, and by using the SAW process for the welding of the last layers. The present invention is, therefore, characterized by the selection of the welding processes for the first and last layers of a steel pipe, particularlly a thick wall steel pipe. Namely, in Fig. 6, the intermedia-te layers 6 and 9 may be formed by any welding process, such as the ordinary GMA process, which presents no welding ... . . ..

problems, provided that the welding of the first layers is carried out by the GMA welding with high melting, and preferably with high frequency oscillation, according to the present invention. However, it is desirable, from the point of view of welding ef~iciency and quality, to apply the GMA welding process according to the present invention to the welding of the intermediate layers. When the ordinary GMA welding process is applied to the intermediate layers, the beads are liable to be a convex shape with a small width. The SAW process, which is carried out after this GMA welding process, is liable to cause a lack of fusion 12 (Fig. 5) at the toe of the weld bead formed by the ordinary GMA process. However, such weldin~ deffect can be prevented by selecting the bead thickness P of the SAW weld to be at least 8 mm, desirably more than 10 mm~
When the bead thickness P is 7 mm or less, a lack of fusion may result. Insufficient toughness vE-60=3.5 Kg.m of the weld metal and heat affected zone of the last layers cannot be brought about due to an excessive welding heat input, when the head thickness P is 8 mm or more.
The GMA and SAW processes according to the present invention can broadly be applied to production of large ;
diameter steel pipes which are required to possess an excellent toughness at low temperature. It is, however, desirable to apply these processes to the production of a thick wall steel pipes having a thickness of not less than one half inch (12.7 mm~, particularly one inch t25.4 mm~
or more. These welding processes can be adopted in the ,~

3~

production line of a UOE process, a spiral process and a bending roller process, which are not explained in detail herein becuase they are well known. The materials of the steel pipe may be ordinary, for example as stipulated in the API Standard 5L, or may be special. Using the process of the present invention, it is possible, in the pipes for a pipeline and a pipe structure constructed at sea, to reduce the stress concentration at the toe of weld and to enhance the fatigue resistance.

The present explanation will now be further explained in detail by way of the following Examples.
Exam ~e_ A steel pipe having an outer diameter 104 cm (41 inches), a wall thickness of 25 mm and a designation of X-65 was produced and tested. The composition of the steel plate was as shown in Table I.

Table I

.
Steel c h e m i c a 1 c o m p o s i t i o n C o n t e n t s (wt %) ;
.. . .. ___. .. .. .. .... , Grade CSi Mn P S Ni V ~b _ _ _ _ X - 650.060.20 1.35 0.001 0.004 0.25 0.14 0.03 .. __ _ _.__ _, , ,, ,,~ , , __ __ The groove shape shown in Fig. 7 and Table II was formed at the abutted ends of a steel plate. After the tack welding, the GMA welding with hiyh freguency oscillation ; ~ ., : ,-: " .

` ~L253~3~

and SAW process according to the present invention, as well as the MIG welding with a small diameter wire and small current, were carried out under the conditions specified in Table II.

. : , . , . : ~

~2,53f~3 ~ 28 -_ _ L~ o o N I I r~ I Ul rl o ~ o ~" "~ b. N W W f~ O N r; r;
.~ O ~ a r~ r~ N

r~ ~ ~D ~ ~
15 ~ ~,N U~ ~DN Ul ~¦ .

@ N N O ~ U'l W 0 ~3 t'l r~ w Ih 1~ ~ ~
~ '~ ' , r 3 rl w t 1 N ~ dP ~ W 1`7 ~
E~ ~ ¦¦ I b . :~: N r~ r~; ~ ~ ~.

~ ~ rJ ~ ~ r~ ~ ~ ~
~ ~ ¦ ~ d ~ 3 ~

~53~33 The composition of the welding wire used was as shown in Table III, below.

Table III

. C~u~al Co osition W 1 r e C o n t e n t s (wt ~) C Si Mn P S Cu __ _ ~ _ Small diameter wire 0.11 0.68 1.34 0.009 0.005 0.26 1.2 mm in diameter Wire for SAW 0.12 0,05 1.9 0.008 0.004 0.30

4.0 mm in diameter The flux used in the experiment of the present Example during the SAW process of the surface layer was a basic, melt type, with SiO2-CaO~TiO2~CaF2 as major com-ponents. The results of the welding are illustrated in the lower six lines of Table II. As seen from Table II, the welding efficiency is considerably improved by the present invention over that of the conventional MIG welding with a small diameter wire. The GMA welding process according to the present~invention would be more effieient than the SAW process, if the two processes were compared to one another on the presumption that -their welding heat inputs are equal in the GMA and SAW processes. The pene-tration according to the present invention is high. The root face of 5 mm thickness could, therefore, be used .
according to the present lnvention, while the root face of '' ,, . ~ ; . .

, '. : ' :, ` ' `: ' ,. . ' ' :
, ' ` ~ ` ' :. ' ; '' ' ':

~12~3~3 only 3 mm thickness could be used according to the con-ventional GMA welding process with a small diameter wire, because of low penetration and weld crack.
The low welding efficiency in the ordinary GMA process is associated with the thin root face and, thus, the large cross sectional area of the grooves.
In addition, the obtained beads by the present invention had an excellent bead shape and enough width as compared to those of the conventional GMA welding process.
This is because of the fact that the SAW process was used for the last welding layers.
In conclusion, by carrying out the process according to the present invention, a sound weld metal without cracks in the first layers and a homogeneous, smooth surface of beads are provided, and the weld æone of the pipe with improved toughness is produced at a high welding efficiency.
Example 2 A steel pipe having an outer diameter of 155 cm (61 inches) and a wall thickness of 34 mm was produced from a 2.5~ Ni- Mn- Nb- V-system, low alloyed steel by the -~
proce~s according to the present invention and the con-ventional, tandem type SAW process, and then tested. The groove shape shown in Fig. 7 and specified in Table IV was ~ormed on the abutted ends of the steel plate. After the tack welding, the two welding processes mentioned above were carried out under the conditions specified in Table IV.

2S3~3~

r ~ u~

~ N ~r ll~ ~ .

o 1- u~ o o ,~ o U~ D O ~ O I
. N O O~ .-1 ~ O ~ O CO
i~

~ ~ ~ r u 1~1 ~

~ , ~

~----38~

The compositions of the wires used were as shown in Table V.

Table V
. ..

W i r e _ Chemical ~osition _ C o n t e n t s (wt %) _ C Si Mn P 5 Ni Mb Small diameter wire 0.04 0.39 0.85 0.003 0.004 3.5 0.30 0.01 2.0 mm in diameter .

Wire for SAW 0.07 0.14 1.50 0.004 0.004 3.5 0.30 0.01 4.0 mm in diameter .. _ The flux used in the present Example during the SAW
process was a basic, melt type, with Sio2-CAo-Tio2-CAF2 as the major components As is clear from Table IV, the welding efficiency is improved by the process according to the present invention -over the conventional, tandem type SAW process. The advantages of the process according to the present invention reside in that: firstly, the root face can be thick due to a deep penetration welding, with the result that the~
cross sectional area of the grooves can be reduced; secondly, the melting rate of wire is considerably high, although the welding heat input is low~ with the result that the number of welding passes can be reduced, and; thirdly, SR
embrittlement is not likely to occur, due to the low welding heat input, although the dilution of the weld metal in the first layers by the base metal is large. -:: ~ ,: :

~L~2538~

Back chipping is not necessary because of the non occurrence of the SR embrittlement.
Cracks in the first layer were not observed after the completion of welding, although the thickness of root face was as large as 6 mm. The bead surface turned out to be excellent after the completion of welding, since the SAW process was jointly used with the GMA process for the last layers.
Example 3 The steel pipe of Example 1 was produced, except that the high frequency oscillation was not performed durlng the GMA welding. The conditions for producing the pipe were as shown in Table VI. The production conditions not specified in Table VI were the same as in Example 1.

:. . . :
.: , . ~ . ., , .

~L~25383 Y S

1~ N 11Vl ~a ~
~ ~ ' ~r .3N O ~ ) O O_~_I '.D U7 trl ~
_ N ~ ~ U~ ~ _( N

O j~ N ~ ~ O O O~0 ~ r'l 11'~ N r I ~
.a ~ ~ ~ dP N ~ Y N

,: , . , ' ;' ', ~ : . ~ j ' ' ~;2S38~

As is clear from Table VI, the obtained welding results in the present Example are substantially the same as those in Example 1.

- : .:. .:. :,, .

., . :

Claims (9)

What we claim is:

1. A welding process for the production of a steel pipe, comprising the steps of:
performing a gas metal arc welding under an atmosphere of a shielding gas and, thereby, forming a first welding layer on the abutted and grooved ends of a steel plate, and;
performing a submerged arc welding and, thereby, forming the last welding layer, the process being characterized in that in said gas metal arc welding step, (1) a high electrical current (I) in amperes in the range defined by the formula:
500d ? I ? 500d - 150 , wherein d indicates the diameter of the welding wire in mm, is conducted through a small diameter wire having a diameter of from 0.8 to 2 4 mm, within said atmosphere of a gas mixture containing mainly an inert gas and addition-ally a carbon dioxide, wherein the conduction of said current through said wire within said atmosphere suppresses the generation of a rotating arc, generates an arc stiffened due to the pinch force by the high current density enough to ensure the resistance of the arc against the deflection of the arc against the magnetic blow, and to enhance the penetration to the base metal, and (2) the wire extension ( ? ) in mm between the end of a current contact tip and bottom of the groove is selected by the formula:
?>10d + 5 , so that the deposition rate of molten metal is essentially increased.

2. A process according to Claim 1, wherein said arc is oscillated in a direction transverse to the weld line at a cycle rate of from 3 to 30 Hz and at a width of from 1 to 15 mm.

3. A process according to Claim 2, wherein said oscillation rate is from 5 to 20 Hz.

4. A process according to Claim 1 or 2, wherein the wire diameter is from 1.2 to 1.6 mm.

5. A process according to Claim 1 or 2, wherein said gas mixture contains from 10 to 20% of carbon dioxide and the inert gas essentially in balance.

6. A process according to Claim 1 or 2, wherein said steel pipe has a wall thickness of not less than approximately 12.7 mm.

7. A process according to Claim 1 or 2, wherein said steel pipe has a wall thickness of not less than approximately 25.4 mm.

8. A process according to Claim 1 or 2, wherein said groove is formed on each side of a steel plate, and each said groove has an angle of from 45 to 160° and the root face formed by said abutted ends is from 3 to 8 mm long.

9. A process according to Claim 1, 2 or 3, wherein an intermediate layer or layers between said first and last welding layers are formed by a gas metal arc welding step, wherein (1) a high electrical current (I) in amperes in the range defined by the formula:
500d ? I ? 500d - 150 , wherein d indicates the diameter of the welding wire, is conducted through a small diameter wire having a diameter of from 0.8 to 2.4 mm, within said atmosphere of a gas mixture containing mainly an inert gas and additionally a carbon dioxide, wherein the conduction of said current through said wire within said atmosphere suppresses the generation of a rotating arc, generates an arc stiffened due to the pinch force by the high current density enough to ensure the resistance of the arc against the deflection of the arc against the magnetic blow, and to enhance the penetration to the base metal, and (2) the wire extension (?) in mm is selected by the formula:
? > 10d + 5 , so that the melting rate of the wire is essentially increased by the long wire extension.