CA1051979A - Method of arc welding with reverse side cooling for obtaining highly tough large-diameter welded steel pipes - Google Patents
Method of arc welding with reverse side cooling for obtaining highly tough large-diameter welded steel pipesInfo
- Publication number
- CA1051979A CA1051979A CA258,605A CA258605A CA1051979A CA 1051979 A CA1051979 A CA 1051979A CA 258605 A CA258605 A CA 258605A CA 1051979 A CA1051979 A CA 1051979A
- Authority
- CA
- Canada
- Prior art keywords
- cooling
- welding
- weld
- steel
- reverse side
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired
Links
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
- B23K—SOLDERING OR UNSOLDERING; WELDING; CLADDING OR PLATING BY SOLDERING OR WELDING; CUTTING BY APPLYING HEAT LOCALLY, e.g. FLAME CUTTING; WORKING BY LASER BEAM
- B23K37/00—Auxiliary devices or processes, not specially adapted to a procedure covered by only one of the preceding main groups
- B23K37/003—Cooling means
Landscapes
- Physics & Mathematics (AREA)
- Optics & Photonics (AREA)
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Arc Welding In General (AREA)
- Butt Welding And Welding Of Specific Article (AREA)
Abstract
Method of arc welding with reverse side cooling for obtaining highly tough large-diameter welded steel pipes Abstract of the Disclosure There is disclosed a novel arc welding method for manufacture of a welded steel pipe wherein the edges of a steel plate bent into a pipe are welded together on one side for forming a pipe and then the seam line of the thus formed pipe is welded by second welding at the side opposite to the side where the first welding was performed. During such second welding, the weld is cooled forcedly from the side where the first welding was performed.
Such forced cooling compensates for any physical effects accompanying the welding and imparts elevated toughness to both the weld metal and the heat affected zone.
Such forced cooling compensates for any physical effects accompanying the welding and imparts elevated toughness to both the weld metal and the heat affected zone.
Description
Back round of the Invention g This invention relates to a method of welding seam portions of a pipe fashioned from a steel plate, When welding along a seam line of a pipe fashioned from a steel plate, the heat affected zone (hereinafter abbreviated occasionally to HAZ) will become brittle due to marked formation, in the course of the welding heat cycle, of an upper bainite in the vicinity of a bond or transition zone between the weld metal and parent metal.
Such brittleness of the heat affected zone ls undesirable especially when a low temperature toughness in excess of a predetermined value is a requirement. It is believed that high toughness of the weld metal may not be realized due to formation of the proeutectoid ferrite. An accepted practice is to cool the heat affected zone during welding. In the case of submerged arc welding, hereinafter abbreviated occasionally as SAW, the cooling of the weld or weld joint may be caused by spontaneous cooling of the weld zone surrounded by the fluxes, dissipation of heat from the lower surface of the parent metal and the transfer of heat to the parent metal. Reliance on such spontaneous cooling sol~ly gives rise to retarded cooling of the weld surrounded by a thick layer of slags 9 thus the Charpy strength of the weld being lowered. Such tendency will be more noteworthy when the heat input is elevated. For this reason, a variety of cooling methods have been proposed for cooling the weld, including water cooling used in con~unction with vertical electro-slag welding. This method is however not applied to the case of~submerged arc welding. It is also known to cool the weld surface with a mist after welding and subsequent to the flux removal. With such cooling, the start time of cooling may be delayed due to hygrospic properties of the fluxes and the mounting of the flux recovery device between mb/C~ - 2 -the cooling positlon and the electrode~. Moreov~r, such cooling technique can not be u~ed for cooling the lnside weld of a large dlameter pipe which has been welded on the outside by the preceding welding operation.
In submerged arc welding, endeavors have been made to lmprove the toughness of the heat affected zone as by using fluxes of higher basicity or adding to the weld metals alloying agents such as nickel, molybdenum and tltanium. It was also proposed ~o perform a qo-called multi-layer welding at the sacrifice of welding efficiency or to add to the parent metal trace amounts of Ti, Al-B, REM-B
Ti-Ca or similar combinations of alloying agents. Especially, higher notch toughness is required of the weld metal and the heat affected zone of large diameter pipes that are designed for the frigid zone such as Arctic~ Zone where the temperatures may fall to -25 to -40 C. It is a matter of great difficulty to satisfy the above requirement by using the currently marketed steel plates and welding wires at an acceptable cost.
The present invention relates to a method of arc welding a steel pipe formed by a steel plate bent into pipe shape with abutting ends~ The method comprises welding a seam at the abutting ends by conveying the pipe through leadlng and trailing passes and forming bead layers on opposite sides of a weld ~one during respective passes, and effecting the formation of the bead layer of the weld zone during the trailing pass with a heat input of more than 36,000 J/cm while effecting forced cooling of the weld zone from the reverse side simultaneously, the forced cooling being effected to ~ cool the weld from the weld temperature to 800C within 11 seconds and from 800C to 500C at a cooling rate of 8-40C
per second.
mb/ G~ - 3 -`
According to the welding method of the pre~ent inventlon, the starting plate is bent into the ~hape of a pipe, and the longitudinal seam line of the pipe i8 welded on the inner surface of the tube. Then, the arc or sub-mer8ed arc welding is performed on the outer surface along the same seam line. During this second welding, the inside weld is sub~ected to forced cooling. Thus, the forced cooling is carried out rom the ~lde opposite to the arc used for the formation of the outside weld. Such forced cooling has been ascertained to improve the hardenability of the weld ~etal through suppression of the formation of the proeutectoid ferrite and concurrently mb/- 4 ~351'.~7'~
the tough~e~ o~' the weld throll~h ~ pression of' the form~tion of the upper 'oairlite in the heat af'f'ected ~one.
When the seam line of the thick-walled steel pipe ls formed by several welding passe3, the inventive arc wel-ding accompanied by forced cooling from the reverse side may be applied to the final welding pass.
The toughness of both the welding metal and the heat affected zone will become more pronounced by using higher cooling rates for the forced cooling through the ranges from the maximum temperature to 800C and from 800 to 500C. The cooling rate of 8 to 40C and pre-ferably of 15 to ~0C per second for the cooling range from 800 to 500C is most preferred in order to obtain high toughness for the welding metal and the heat affected zone. As cooling agent, water, liquid nitrogen or dry ice may be employed within the scope of the present invention. In water cooling, the more the flow of cooling water, the sooner the cooling. With forced circulation çooling by three nozzles with the'total flow rate of 45 liters per minute, the cooling of the welded article from 800 to 500C may be carried out at a speed three to four times faster than the case wherein no forced circulation cooling is employed. With use of liquid nitrogen or dry ice, the cooling from the maximum tempe-rature to 800C proceeds usually at a higher speed because of the precooling effects of these cooling agents.
EIowever, the cooling rate from 800 to 500C will be lower than that attained by water because the heat exchange capacity of these cooling agents is lower than . ~ ' that of water. ~
.. . , ~ . .. . . , .. . , . , , i.. ..
10519~9 The increased toughn~ss of the welding metal to be attained by the reverse side coo]ing may be -ttri~utable to the retarded precipitation of coarse ferrites from the , graill boundary with resulting refining of the ferrite ground, and to the increase in the precipitation of the lower bainite. In case of using a starting plate manufactured by controlled rolling, the weld metal of the inside weld may be increased in toughness because of the apparent disappearance of the precipitation hardening brittleness by virtue of the second welding at the outside accompanied by cooling from the inside. With use of the compensating welding wires, the to'ughness of the weld metal at the inner surface of the pipe may become lower than that measured at the intermediate zone or near the outer surface of the pipe. ~Iowever, such decrease in toughness at the inner surface may be hindered by i~crea-sing the rate of cooling. These changes in toughness in the direction of the pipe wall thickness may be compen-sated through.suitable selection of the welding wires used for the respective welding passes.
The increased toughness of the heat affected zone may be attained by promoting the cooling for the range from the maximum temperature to 800C in such a manner that the cooling through said range may be completed with-in preferably li seconds. In this case, the width of the coarse grain zone may be reduced and there is no formation of the coarsely grained preanstenitic crystals.
Furthermore, if the cooling rate for the temperature range from 800 to 500C is increased to 8C/sec or more through the reverse side cooling, a higher toughness ~051979 of ~iAZ ma~ be realized ~llrou~h suppressiorl of the upper bainite and promotion of the lower ~ainite. Gn the other hand, the weld metal and the heat affected zone will become harder for a faster cooling rate. For example, at the cooling rate where the highest toughness may be attained, the hardness will reach a value almost close to saturation. From such consideration, the optimum cooling rate for the temperature range of 800 to 500C is 15C per second or thereabouts.
According to the present invention, however, the upper limit of the cooling for the temperature range of 800 to 500C is selected toA40C per second where the tough-ness starts to be lowered under the so-called quen^hing effect.
Further objects and advantages of the present inven-tion will become more apparent from the following detailed description of the preferred embodiment and the accompany-ing drawings.
`
Brief Description of the D~awin~s Figs. l(a) and l(b) show a typical device designed for executing the inventive method, in side elevation and partial enlarged view, respectively;
Figs. 2(a~ to 2(f) show various examples of the `-reverse side cooling according to the present invention;
Figs.-3~a) is a diagram showing the cooling time required for the three test pipes to cool from the maximum temperature to 800C for the spontaneous cooling and the various inventive reverse side cooling modes;
:~05~979 Fig. ~(b~ is a ;illlilar ~iagram for the temperclture range l`rom 800 to 500C;
Fig. 4 is a diagram sho~ling the welding heat input versus cooling time for the spontaneous cooling and the various inventive reverse side cooling m~thods;
Fig. 5 shows typical cooling curves for steel test piece C by the conventional cooling and the inventive cooling;
Fig. 6(a) is a cooling time versus toughness dia-gram chart for the steel test piece A cooled by reverse side cooling through the range of the ma~imum temperature to 800C;
Fig. 6(b) is a~similar chart but for the temperature range of from 800 to 500C;
Figs. 7(a) and 7(b) are the similar diagrarn charts to Figs. 6(a) and 6(b) for the steel test piece B;
Figs. 8(a) and 8(b) are the similar diagram charts to Figs. 6(a) and 6(b) for the steel test piece C;
Figs. 9(a~ and 9(b) are the similar diagram charts to Figs. 6(a) and 6(b) for the steel test piece D;
Figs. lO~a) is a cooling time versus toughness diagram chart for HAZ of the steel test piece A for the temperature range from maximum temperature to 800C
Fig. lO(b) is a similar diagram chart but for the temperature range from 800 to 500C;
FigS. ll~a) and ll(b) are the diagram charts for the steel t st pieces B and C similar to Figs. lO(a) and lO(b);
Figs. 12(a) and 12(b) are the diagram charts for the steel test piece D similar to those shown in Figs.
... .. r lO(a) and lO(bj;
Figs. 13(a) is a cooling tirne versus hardrless diagram chart for ~Z and weld metals of the steel test pieces A, B and C and for the temperature range from maximum temperature to 800C;
Fig. 13(b) is a similar diagram chart but for the temperature range from 800 to 500C;
Fig. 14 is a dia~ram chart showing the distribu-tion of hardness through the inside and outside weld metals for the steel test piece B;
Fig. 15(a) is diagram chart showing the distri-bution of hardness through the inside and outside weld metals for the steel~test piece B in the case of conven-tional welding with spontaneous cooling;
Fig. 15(b) is a similar chart to Fig. 15(a) but in the case of 3 nozzle water cooling with forced circu-lation;
Fig. 16(a) is a cooling time versus hardness dia-gram chart in, th~e case of the reverse side cooling for the HAZ and the weld metal for the temperature range from the maximum temperature to 800C;
Fig. l~(b) is a similar chart to Fig. 16(a) but for the temperature range from 800 to 500C;
Fig. 17(a) is a photo showing the microstructure of the coarse grain zone of the HAZ of the test piece A
obtained by conventional welding with spontaneous cooling;
Fig~ 17(b) is a similar photo but showing the same portion of the test piece A obtained by the inventive welding with reverse side cooling;
Figs. 18(a) and 18(b) are the photos similar to 11D5~'~79 Figs. 17(a) ~nd 1'1(~) but for the ,sarne portion of' th~
test piece ~;
Fig. 19(~) is a photo s~lowing the microstructure at the outer surface of the weld metal of the test piece obtained by conventional welding;
Fig. l9(b) is a similar photo to Fig. l9(1) for the same test piece B but obtained by the inventive welding with reverse side cooli~g;
~ ig. 20(a) is a diagram chart showing the coo]ing time from the maximum temperature to 800C plotted against length of the coarse grain zone of the HAZ for the test pieces A, B, C and D;
Fig. 20(b) is~a similar chart to Fig. 20(a) but -showing the cooling time from 800 to 500C;
Fig. 21 is a diagram chart showing the toughness distribution of the inside weld in the direction of the weld thickness with use of different and same welding wires for the inside weld; and Fig. ~2 is a diagram chart showing the hardness distribution through the inside and outside weld metals in the direction of thickness for the test piece C obtained by conventional SAW with use of same inside wires and the same test piece obtained by inventive SAW with use of same inside wires.
Description of the Preferred Embodiment , In Fig. l~a), a steel pipe 2 conveyed axially on transport rollers 12 has been welded previously along a longitudinal seam line from the inside and is to be .. ~ ~ . .. .. .
welded to a weldlng machlne 1 on the tr~nsport roller~ 12 along the same longitudinal seam llne but now from the outside. As the pipe 2 is being welded from the outside, it is cooled ~imultaneously by a cooling device 3 from the reverse side which is, ~n the embodiment illustrated, the inner side of the steel pipe 2.
The cooling device 3 i8 shown in the enlarged view o~ Fig~ l(b) and comprised o a boom 7 extending lengthwise of ~he transport line and fitted at the end with a cooling water tank 9 and a pump 10. The end of the boom 7 iq also provided with a support roller 11 adapted to travel along the inner surface of the steel pipe, and a cooling water nozzle o directed towards the weld along the seam line.
Cooling water may be supplied to nozzle 8 from a water pipe 5 connected to the external water source and by way of the tank 9 and the pump 10. The cooling water may be sprayed by the nozzle 8 towards the inside weld metal at a predetermined ~ flow rate by the controlled operation of the pump 10. The water pipe 5 and a power cable 4 such as wire for driving the pump 10 are passed through the boom 7 and extended to an external device including a water source and an electrical source, not ~hown. In Fig. l(b), the numeral 6 denotes a leading electrode or wire and a trailing electrode or wire of the welding machine 1 mounted in a confronting relation with the cooling nozzle 8. The numeral 13 denotes a water pan for receiving the used water which flows down and is returned to the tank 9. The boom 7 may be displaced length-wise by a support mb/~
A
~ ' ~ O ~ 37 ~
device 14 for ch~nging the po,itlon of the noz~le rela-tive to the pipe 2.
The steel pipe ~, ~hich has been welded along a longitudinal seam line on the inside may thus be welded on the outside along the same seam line as it is fed in the direction of the arrow marks in Figs. l(a) and l(b) along the transport rollers 12. Simultaneously ~, Jnslc~e with o*tei~e cooling, the cooling water is sprayed from the nozzle 8 to the inside weld directly opposite to the welding electrodes.
In Figs. l(a) and l(b), only one nozzle 8 is used for spraying cooling water to the reverse side of the outside weld. Howe~ver, two or more nozzles may be provided in tandem along the seam line, and the cooling water may also be circulated forcibly for increasing the heat exchange capacity. Alternatively, cooling water may be replaced by liquid nitrogen, or a dry ice may be mounted in direct physical contact with the inside of the pipe for cooling.
Reference is now made to Figs. 2(a) to 2(f) showing these various cooling modes that may be used with the submerged arc welding.
Fig. 2(a) shows a single nozæle 8 being used for reverse side cooling as already explained with reference to Figs. l(a) and l(b). In Fig. 2(a), the upper view shows the weld in the transverse section `~
and the lower view shows the weld in the longitudinal section. In Fig. 2(a), the three dotted lines indicate drill holes bored from the end face of the test steel piece to the bead for insertion of thermocouples used .
105197g for measurirg the temperaJur~ as will be described, In Fig. 2(b), two noz%les 8 are placed in tandem along the seam line. By using the nozzles of the same size, the flow rate of the cooling water may be doubled with consequent increase in the rate of cooling.
For example, if the flow rate is 12 liters per minute for a single nozzle, the flow rate will increase to 24 liters per minute with two nozzles shown in Fig. 2(b).
In Fig. 2(c), three nozzles 8 are placed in tandem on the seam line for obtaining the flow rate of 36 liters per minute with a still faster rate of cooling.
In Fig. 2(d), a compulsory. circulating unit 15 such as tank and pump 10 shown in Fig. l(b) is used for directing the cooling water under pressure to the three nozzles 8 for realizing a faster cooling rate by increasing the flow rate to, for instance, 45 liters per minute.
In Fi~. 2(e), there is shown a liquid nitrogen source, suc~ as y essel 16, fro~ which the liquid r.itro-gen is delivered to plural nozzles 18 mounted within a hood 17. ~he liquid nitrogen may be sprayed from the nozzles 18 to the reverse side of the outside weld metal at a rate of, for instance, 400 kg per hour.
In Fig. 2(f), an air hose 19 is d,ilated for applying a piece of dry ice 20 to the reverse side of the outer weld metal for cooling the weld.
In Figs. 2(e) and 2(f), the three dotted lines indicate the mounting positions for thermocouples as in the case of Fig. 2(a).
These various cooling modes for reverse side 105~979 cooling rnay be selectively elnployed fol cooling the weld of a steel pipe which has been welded on the inside and is beine welded on the outside. With such cooling, the formation of the proeutectoid or of the upper bainite at the ~IA% may be prevented with consequent increase in the toughness of the weld.
A series of tests on arc welding associated with reverse side cooling were conducted with test steel plates A, B, C and D, for checking the effects of vari-ous cooling rates on the toughness, hardness and micro-structure of the welding metal and the HAZ.
Four test steel pieces A, ~, C and D with varying thicknesses, grades and chemical compositions as shown in the following Ta~le 1 were employed. The test plate D was 10.3 mm thickness, the test plates ~ and ~
were prepared by controlled rolling and 20.3 and 18.3 mm thickness, respectively, while the test plate A was made of 1.4 Ni steel with thickness of 25.4 mm.
`
Tsble 1 Chemical Composition of Test Steel pieces (~adle) (%) Cr W.T C 3i Mn _ 9 Cu Ni Mo Nb V ~ol~l .~ x6S 25,4 .0~2 .25 1.17 .014 .007 _ 1.41 17 _ _ 053 x70 20,3 .090 ~1 1.45 .012 .004 _ .26 .15 .0~7 .092 .045 (` x70 18.3 .OS6 .23 1.4~ .014 .005 .19 .21 _ .039 .091 .020 _ x52 10.3 .126 15 .65 .015 .009 _ - ~ = .017 Test welding wires a, ~ used for the tests are shown in the following Table 2. Test wire a was a Ni - Mo - Ti alloy and the test wire b was free of any alloy-ing agents.
Table 2 Chemical Composition of Test Wires (%) = dia C Si Mn - S I~i Mo Ti a 4.0 .03 .17 2.03 .011 .007 .91 .52 .21 , . .
b 4.0 .07 .01 .68 .009 .017 _ _ Test fluxes X, Y used for the tests are shown in the following T~b~e 3. The test flux X was weakly basic with basicity equal to 1.2 and the test flux Y
was strongly basic with basicity equal to 1.6.
Table 3 Chemical Composition of Test Fluxes (%) .~= Si02 OaO MnO M~O ~123 CaF2 2rO2 ~ S FeO
. ?X 32 20 11 8 11 13 ~ .Oll .l63 .98 . Y 28 28 _ 19 18 ~ _ .022 .222 .22 The welding wires and fluxes and other welding conditions are shown in the following able 4 for the respective test pieces A to D.
~OSlg79 Trl~le 4 Welding Condition ~, I Le~ding rr~iling~ Speed ~eat l x . ~____ V A ¦ V ,~ ~ ~I mm/min I7put lwire , r~ ~
Inside 1 7GO 36 640 l40 I4 j 1,~50 24,400 a~a X
D _ ! _ Outsidel 940 38620 40 15 ¦ 1,000¦ 36,300 a+a Y
1 .- __ ~ , .. .. ._~ . .
Insidel 920 36 70040¦ 14 850¦ 43,100 b~a Y
C Outsid`e 960 38 700 40~ 15 800j 48,400 a~a Y
--I' . I ! ~
Insidel 940 36 70040l 14 1 ~50l 43,700 ! b+a Y
; _ _ , _ ! - _ _ Outsidq~ 960 38 700 140l15 ~ 800 ! 48,400 a+a Y
_ . ,. _ , .
Insidell,OOO 38 8004014 ¦ 7GOi 60,000n+a Y
I\ l l l . ,"~.
Outsidql,050 38 90040¦15 ¦ 700¦ 65,100a+a Y
~ _ _ lI _ .... I .1.__. _ The scarf angle for the inside and welds are 45 r~
for the respective test pieces A to D. The depth of scarf was sele.cted to one-third of thickness for each of the insid~ a~d outside welds for the respective test pieces A to D.
In Figs. 3(a) and 3(b), the cooling time intervals necessary for cooling from the maximum temperature to 800C and fro.m 800C to 500C are shown for the cases of conventional welding with natural coolin and the welding associated with the various cooling modes. It is seen from Fig. 3(a) that the cooling time necessary for cooling from the m~ximum temperature to 800C may be progressively reduced with use of one and two nozzles as compared with spontaneous cooling. With three nozzles, the cooling 105~79 time re~m~lins ,llrnost cqual to th~t o~tairled with two nozzles, and a somewhat shorter cooli~g time may be attained by use of compulsory circulation of the cooling water. It is also seen from ~'ig. 3(b) that the cooling time interval necessary for cooling through the range of 800C to 500C may be reduced progressively with increase in the num~er of the nozzles and hence in the flow rate and that use of forced circulation of cooling water results itl a cooling rate several times faster than in the case of spontaneous cooling.
It is also seen from Figs. 3(a) and ~(b) that the use of liquid nitrogen or dry ice-gives almost the same cooling rate as that~attained with forced circulation with three nozzles (flow rate, 45 liters/min.) for the temperature range from 800C to 500C. These high cool-ing effects, proper to liquid nitrogen and dry ice, may be attributed to the precooling of the weld by the liquid nitrogen prior to spraying from the nozzles or by the dry ice prio,r to~ the arc generation from the welding torch. With liquid nitrogen and dry ice, the cooling time necessary for cooling from 800C to 500C will become much longer than in the case of water cooling because of the lower heat exchange capacity of the liquid nitrogen or dry ice than that of water.
In Fig. 4~ the cooling time necessary for cooling from 800C to 500C is shown for varying heat inputs ~or the steel piece A for conve~tional welding and the inve~tive weld~ng with the various cooling modes.
The straight solid line in Fig. 4 represents the calcu- .
lated value for the conventional SAW with natural cooling.
_ .. . ., . .. . ... , y Thc mcuuur~d V~ ICS ~or th~ cas~ for conven~l(>nal we~dlng an(l tl~ inv~ntl~e cooling ~lth wa~er coolinK with one noz%]e are shown in Fig. 4 in thc vlcinlty of 65,0~0 J/cm, while the m~asured values for the other cooling mode~ are shown in Fig. 4 for the heat input of 65,000 J/cm. In Fig. 4, thc blanl~ed marks represent the measured values at the intermedlate portion of the test piece and the black marks represent the values measured at the outer surface of the test piece. It is seen from Fig. 4 that a rate of cooling almost three times faster than that with conve~tional welding may be obtained with the inventive welding with one nozzle. With increase in the number of nozæles and hence in the flow rate of cooling water, a faster rate of cooling may be obtained and, with forced circulation of cooling water with three nozzles (flow rate:
45 lit./min.), the mean cooling rate may be elevated to 23 C/sec. As àlso seen from Fig. 4, the rate of cooling at the intermediate zone is faster than that at the upper ` surface of the test piece as a result of the reverse side cooling.
Fig. 5 shows typical cooling curves for the test pieces C which are subjected to the conventional cooling with natural cooling and the inventive welding with reverse side cooling. As seen from Fig. 5, the maximum cooling capacity may be obeained with forced circulation of cooling wnter with three nozzles (flow rate, 45 liters per minute).
The eest steel piece A was sub~ected to a submerged arc welding pass on the inside under the conditions shown in Table 4 and then to a submerged arc welding on the outside while the inside of the pipe was cooled with water, nitrogen and dry lce according to the above-mentioned six cooling modes. Figs. 6(a) and 6(b) show the measured values of the ~b/~ - lô -' 1051g79 weld ~ou~ n~ p]~tt~ ~gtlinst coo]lng tirnc neces~r~ for coollng from th~ mnximl~m temperaLIJre ~o ~C and from 800C to 500oc, respectively. The toughness of tll~ weld was measured three places, namely close to the outer surface of the weld, the lntermediate portion and close to the inncr surface of thc weld. The same wires a for the leading and trailing wires and the flux Y were used for forming the inside and outside weld metals of the weld.
As seen from Fig. 6(a), the shorter cooling time for the specified temperature range is effective to improve ehe toughness of the weld metal. As also seen from Fig.
6(b), the shorter cooling time for the specified temperature ranBe imparts higher toughness to the weld metal. The weld metal of the test piece A has presumably elevated hardenability and a substantially constant value of vE - 30C
= 127 to 136 ft - lb may be obtained at a mean coo~ing rate of 15C per second (cooling time interval, 20 seconds) for the temperature range of 800C to 500C.
The test steel piece B was also subjected to a submerged arc welding pass under the conditions shown in Table 4. The pipe thus formed was then seam welded at the outside by submerged arc welding while the inside weld was cooled in the above-mentioned manner.
Figs. 7(a) and 7(b) show the measured values of toughness of the weld metal against the cooling time intervals necessary for cooling through the specified ranges of temperature, In Figs. 7(a) and 7(b), the wires a and b and the flux Y were used for the inside welding and the same wires a with the flux Y were used for the outside welding. The weld toughness was measured at three positions, namely close to the outer surface of the weld, the intermediate portion ~b/ ~ - 19 -an(] closc to Lh~ ln~ r slJrf.lcc of ~hc wel(l.
As seell from Fi~s. 7~a) and 7(b), thc shorter the cooling timc lnterva]s for the ~pecified ranges of tempera-ture, the higher the toughness of the weld metal. E~pecially, the cooling time for thc temperature range from 800C to 500 C shown in Fig. 7(b) decidedly affects the toughness of the weld metal. In the present case, the compensating wires (leading wire a and trailing wire b) were used for inside welding to reduce the quenching effect that might be exerted on the inside weld at the time of the outside welding. Wllen the test piece B manufactured by controlled rolling is sub~ected to the conventional welding, the inside weld will be lowered in toughness compared to outside weld or the transition zone between tbe outside and inside welds because of precipitation hardening caused by the precipita-tion of Nb and V from the parent metal and of Ti from the welding wires. In the case of the inventive welding with reverse side cooling, there is not sufficient time and ~ temperature -for these elements to precipitate and hence the zone of precipitation brittleness can be minimized.
Thus, with the welding method of this invention, the need for using compensating wires may be eliminated because of the suppression of the precipitation hardening of the weld metal applied in the preceding pass. The toughness of the weld metal may be further improved by using the wires with high hardenability such as wires a for the leading and trailing wires.
In Figs. 7(a) and 7(b), the wires a and b and the flux Y were used for the inside weld, while the wires a and the flux Y were used for the outside weld.
mb ~ 20 -.
: :
105197~t rc ~ f ; i rn i l .l r t~ t~ c ~ (l f ~r t 1 t ~ ~: l p l cl ~ ~ C ~ r~ wll ,in F~ J ~ 3 ( ~ ; c~ n fro~ tllese Figurcs, t1le test result.s for tllc stee~ ple(e C are subsLanti~lly the same ac; tllose for the steel pieces ~ and R.
In Figs. 6 to 9 inclllsive, the blanked circles represent the data as measured towards tlle outer surface, whereas the semi-blanked circles and the black circ].es rcpresent tlle test data as measured with the thermocouples ln at the intcrmediate ~ones and towards the inner surface, respectively.
When a thin steel piece of a lower grade is formed into a pipe by SA~ with use of a conventional wire a, it may frequently occur that the weld metal obtained has not sufficient toughness. For this reason, it WAS
proposed in the past to use a Ni - Mo - Ti wire and a flux added with ~InO and ZrO2 such as flux X shown in Table 3. Figs. 9(a) and 9(b~ show the toughness against cooling time of the weld joint obtained by using the low-grade thin-walled test steel piece D and a combination of the conventional fluxes and wires.
The steel piece D was formed into a pipe shape by submerged arc welding on the inside by using the same wires' à for the leading and trailing wires and the f],ux X.
The tube thus formed was then subjected to the submerged arc welding on the outside by using the same wires a~a and the flux Y while cooling the inside of the tube simultaneously, The toughness of weld metal was measured ne'a-r t~e outex and inner surfaces and the intermediate portion with respect to the cooling timc for the two ranges of cooling temperature.
mb/~,~ - 21 -~05~979 Whllc no flxcd relation may lc obffervccl to exi~t ln Fig. 9(a) between the touf~hncss of the weld met~ and the cooling time above 800C it may bc apparcnt from Fig. 9(b) that the shorter cooling time for thc range of 800C to 500 C ls highly effective to improve the toughness of the we]d. It may be seen from Fig. 9(b) that the toughness may be improved by about twice in terms of vF-30C and by about 40C in terms of vTrs as compared wi~h the cflse of using spontancous cooling.
In Figs. lO(a) and ]O(b) the toughness of HAZ of a steel A pipe obtained by submerged arc welding under the welding conditions shown in Tflble 4 is plotted against cooling time. It may be seen from these Figures that the UAZ toughness may be improved with shorter cooling time for the both temperature ranges and in the case of reverse side cooling from 800C to 500C within 14 seconds the toughness may be improved by about twice in terms of vE-30C and by about 15C in terms of vTrs as compa~re~d with the case of spontaneous cooling.
mb/p~ - 22 -~051~37~
It is beli~ved that pro!lotc-l cooling through the ter~pe-rature above 800C resl)lt;s in reduced s~ 7,C3 of preaus-tenitic crystal grairs arld prevention of the enfeebled grai~ boundary. It is also p~esumed that promoted cooling through 800C to 5Q0C favors the formation of a highly tenacious lower ~ainite while preventing the for-mation of the upper balnite which may be detrimental from the viewpoint of toughness.
Figs. ll(a) and ll(b) show t~e similar results obtained with steel types B and C. With the type B, with faster cooling rates through the both temperature rar.ges, the HAZ will become increasingly tough. Above all, the ccoling time through the range of 800C to 500C
decidedly affects the toughness cf HAZ. When t~e ~iA7, has been cooled in 12.5 seconds through said temperature range, the measured v~lues of toughness may be improved by about 1.8 times in terms of vE-30C and by about 33C-in terms of vTrs as compared with the case of spon-taneous cooling. With type C, the cooling time through 800C to 500C decidedly affects the toughness of the resulting ~IAZ. For example, when the latter has been cooled in 17 seconds through said range, the values of toughness may be improved by about 1.5 times in terms of vE-30C and by about 38C in terms of vTrs as compared with the case of spontaneous cooling. The similar results may be observed in Figs. 12(a) and 12(b) for the steel type D, While there is observed no fixed relation between the toughness and cooling time for the temperature range above 80QC, it may be seen from Fig. ll(b) that shorter cooling time required for cooling through 800C
..... .... .. . . . . . . .. . . . .. . . . . .. .. . .
to 500(- r~sll]t~ ln a ~nurkc~lly l~ rov~d ~ollg~lnc~ ~n terme of vTrs.
Flgs. 13(a) an~ 13(h) S~IOW thc effects of the inventive rever6~ side coollng on the maximum h~rdness of the ueld metal and I~A% of test pipes formed of the steel pieces A, B and C. The test pipes were prepared by welding along a longitudinal ~oint line on the inside and then welding along the same joint line on the outside with simultaneousi forced cooling from the side of the inner weld, as described in the foregoing. With these steel types, there is again no definite relation between the hardness and the cooling time through the temperature range above 800C. However, as shown in Fig. 13(b), with shorter cooling time for the temperature range through 800C to 500C, the hardness of the weld ~one may be increased progressively until substantially constant values of Hv.
max. 238 for weld metal and Hv. max. 260 for llAZ are attained for the steel types A and-B. With the steel types B and C, ~ as`the cooling rate increases for the range of 800C to 500C, the maximum hàrdness of the weld metal tends to decrease at the outset and then starts to increase at a slow rate while HAZ tends to increase gradually. With steel type B, the hardness of HAZ reaches the Hv. max 258 at the mean cooling rate of 24C per second for the range of 800C
to 500C. The carbon equivalent Ceq of the steel type C
is 0,368 which is lower than that of steel type B (0.385~, and thus the hardenability of the steel C is lower than that of the steel B, This accounts for a somewhat lower value of the maximum hardness of mb/p~ - 24 -105~7~
the 1~7, ~1n~ T1e~d rnetal of ~ne stccl t~pe C tha~ th-t of the steel type B.
With tne stee~ types ~ and C obtained by control~ed rolling, the weld metal will be lowered in hardnes3 with shorter reverse side cooling time for the temperature range from 800C to 500C, as discussed in the foregoing.
Fig. 14 shows the hardness distribution in the direction Or plate thickness of the inside and outside weld metals for the steel type B weld joint. In general, ~ith the conventional S A W followed by spontaneous cooling, designated as Conv. SAW in the drawing, the inner and outer su~;faces and the portions ad,jacent thereto are softened by heating to approximately the fusing points and the elements such as Nb, Ti and V are turned into solid solution. On the other hand, the intermediate portion of the weld metal spaced about 2 m~ from the inner and outer surfaces are heated to a range from 600~C to Ac point and hardened by precipitation of Ni, V and Ti in the form of carbides and nitrides. However, when the steel pipe welded 011 the outside is welded on the inside with simultaneous reverse side cooling designated as C. C. SAW in the drawing, the range of temperature and time necessary for precipitation of these elements will be limited thus the precipitation hardening may be suppressed with consequently lowered hardness in the inr;er zone.
On the other hand, with increase in the rate of reverse side cooling, the outside weld is hardened, because the same wires a are used for the outside we.ding.
Therefore, with the weld joint obtained by the inventive welding, the region of maximum hardness will be shifted 1(~51979 from the con~ rable regior~ of the weld o~tained the con-ventional welding which is not accompanied by rever~e side cooling. The E~Z hardness is increased with an elevated rate of reverse ~ide cooling for the respective steel types as discussed in the foregoing, with the region of maximum hardness being situated at a mid zone which is spaced 0.5 to 1 mm apart from the fusion line, as indicated in Figs. 15(a) and 15(b). It is the cool-ing time for the range of 800C to 500C, rather than that for the range above 800C, that affects markedly on the hardening of the weld. It is believed that the cooling time for the temperature range above 800C affects the grain sizes of the austenite crystal grains, whereas the cooling time for the range between 800C and 500C
favors the precipitation of the lower bainite while suppressing the formation of the upper bainite.
In ~igs. 16(a) and 16(b), the test results on the effects of reverse side cooling on the hardness of the HAZ and weld meta~l for the low grade steel ~ are illust-rated. It is seen from these Figures that the hardness ~-of HAZ and weld metal is markedly affected by the cooling rates for the specified ranges of temperature and that the hardening may be more pronounced with faster rates of cooling. As sllown in said Figures, with reverse side ;
cooling in 15 seconds through the range of 800~C to 500C, the hardness of the weld metal amounts to Hv. max. 242, while that of IIAZ equals to Hv. max. 193.
Figs. 17(a), 17(b) and Figs. 18(a), 18(b) show the photos (magnification ratio : 400) of the microstructures .
of the HAZ for the steel types A and B, respectively, '. - .. ~ , :
10519'79 which are obtair~ed by con~entior,a] wclding arld tha~
obtained by the inventive welding with use of forced circulation with three nozzles. As seen from these Figures, the sizes of the preaustenite crystal~ of the coarsely grained HAZ have been reduced for both the types A and B as a result of the reverse side water cooling. It is also seen that the HAZ is ~ade highly tough on account of the predominant formation of the u~cr bainite and the corresponding suppression of the upper bainite.
Figs. l9(a), l9(b) are the photos taken by an electron microscope (magnification ~atio : 3000) showing the microstructures of the outside weld metals of a steel B tube obtained by conventional welding and a tube of the same steel type obtained by the inventive welding.
As seen from these Figures, by the reverse side cooling, the formation of coarse ferrites in the weld zone has been suppressed and a highly refined ferrite structure may be observed along with a small amount of bainite.
The width and siæe of the coarsely grained HAZ can be reduced by reverse side cooling. Figs. 20(a), 20(b) show the widths and lengths of the coarse grain zone of the HAZ for the respective steel types for the various rates of reverse side cooling. As seen from these Figures, the widths and lengths of the coarse grain zone of HAZ
for the respective steel types may be reduced by using ~ -~
shorter cooling rates for the temperature ranges above 800C and from 800C to 500C. With use of liquid nitrogen and dry ice, a faster cooling rate may be attained for the temperature range above 800C o~ account of pre- ``
.~ .
cooling e~r~c~ n(l hencc tllc 81~.e8 of ~hc corlr~e ~,r~in zone mny bc re~u~ed evcn lf a ~]owcr coollng rate ~hou]d be used for the temper~turc range bctwecn 800"C and 500C.
Thus, for elevating the toughness of IIAZ, lt is necessary to cool promptly through the above-mcntioned two temperature ranges. Prompt coollng through the range above 800C is effective to reduce the width of the coarse grain zone and to prevent the preaustenite crystal grains from becoming coarse. On the other hand, prompt cooling through the range of from 800C to 500C is effective to prevent the precipitation of the upper bainite and to favor the formation of the lower balnite. The above applies to the weld metal as well. Thus, prompt cooling through the two temperature ranges is effective to make the ferrite ground of the molten metal more refined and to facilitate the formation of the lower bainite with resulting increase in the toughness of the weld metal.
In case the parent metal is prepared by controlled rolling, the outside weld of a pipe which has been welded previously on the inside is known to become extremely brittle by precipitation hardening when allowed to cool. So far, the compensating wires were used for welding the outside of a steel pipe welded previously on the inside. In case of steel pipe that is welded at the inside and outside by using an arc welding accompanied by reverse side cooling, the inside weld tends to be lowered in hardness as compared with the outside weld. In order to make the hardness and toughness of the weld more mb/J C> - 28 -~0~ f~
uni~orl~l, te~its have ~)een co;iducte~ t,y u~sirlg the sarrlr: wire _ and dif~erent wires a, b ~or the lea~in~ arld trail~ng wires. The test results ale sllotln in Fi~s. 21 and 22.
In Fig. 21, the toughness of the weld is plotted again~t the distance from outside for the cases in which the steel pipes (types ~ and C) were subjected to the two side welding followed by spontaneous cooling (Conv. Weld) and to the inventive welding (C. C. SAW) with use of two nozzles (flow rate, 24 liters per minute). As seen from Fig. 21, with use of the leading wire b and ~*~}n ~
wire a, the inside surface zone of the weld metal obtained l~0rG
by the inventive welding is ~e~ff tenacious than the intermediate and outs,ide surface zones, whereas the intermediate zone of the weld metal obtained by the inventive welding with use of the same wire a for both the leadilg and trailing wires i3 almost as tenacious as the remaining zone. ~ig. 22 shows the hardness of the weld plotted against distance from outside surface of the weld, with the p~rtition line between the oulside and in-side weld being disposed at about 13 n~ from the ourside surface. The test piece was steel C and the dotted and solid lines represent the curves obtained with conventional welding and with the inventive welding with two nozzles, respectively. It may be presumed that the inside weld obtained by usint~ the com~ensating wires twires a ~nd b) is lowered in hardness because its harde-ning process is affected adversely by the reverse side cooling and also the precipitation hardening is prevented thereby from occurring. As seen from Fig. 2~, with use of the same wires a for both the leading and trailing -.. . . . .
1~51979 wires, uniform distribution of hardness may be attained ln the direction of the plate thickness. It 1~ belleved that precipitation hardening i~ the vlclnity of the in~ide surface of the weld is ~uppressed by reverse slde cooling while the inside surface is slightly hardened by the coollng thus imparting sufficient toughness to the zone close to ehe inside surface. Thus, with the inventive welding with reverse side cooling, if the same wires a are used for the leading and trailing wires on the occasion of the in~ide welding, the weld metal may be made uniformly tough along its thickness. In this way, uniform hardness may also be attained along the plate thickness because of suppression of the precipitation hardening which might otherwise occur at near the inside surface.
In the foregoing, the inside weld is formed previously on a tube and the outside of the tube is welded simultaneously with cooling at the inside. It is however possible to weld at the outside of the pipe and then to weld at the inside thereof with cooling simultaneously at -~
the outside. Also, when the inside and outside welding is -performed in plural passes, the reverse side cooling may be performed in the last pass. Namely, one side welding is effected on a leading or forward pass and reverse side welding with concurrent cooling is effected in a trailing or reverse pass.
Although the present invention has been described for the case of submerged arc welding with two electrodes, it can be applied to the case of gas shield welding. A
single electrode or three or more electrodes may also be ~0 employed. However, since the submerged arc welding is usually associated with the lower rate of cooling, and the use of plural electrodes is usually associated with higher heat input, the present invention may be mb/ -?
applied advantageou~ly to the ca~e o~ one-pas~ submerged arc welding of a large-diameter thick-wfllled steel tubin~
on the inside and outside with use of two or more elect-rodes.
t : '
Such brittleness of the heat affected zone ls undesirable especially when a low temperature toughness in excess of a predetermined value is a requirement. It is believed that high toughness of the weld metal may not be realized due to formation of the proeutectoid ferrite. An accepted practice is to cool the heat affected zone during welding. In the case of submerged arc welding, hereinafter abbreviated occasionally as SAW, the cooling of the weld or weld joint may be caused by spontaneous cooling of the weld zone surrounded by the fluxes, dissipation of heat from the lower surface of the parent metal and the transfer of heat to the parent metal. Reliance on such spontaneous cooling sol~ly gives rise to retarded cooling of the weld surrounded by a thick layer of slags 9 thus the Charpy strength of the weld being lowered. Such tendency will be more noteworthy when the heat input is elevated. For this reason, a variety of cooling methods have been proposed for cooling the weld, including water cooling used in con~unction with vertical electro-slag welding. This method is however not applied to the case of~submerged arc welding. It is also known to cool the weld surface with a mist after welding and subsequent to the flux removal. With such cooling, the start time of cooling may be delayed due to hygrospic properties of the fluxes and the mounting of the flux recovery device between mb/C~ - 2 -the cooling positlon and the electrode~. Moreov~r, such cooling technique can not be u~ed for cooling the lnside weld of a large dlameter pipe which has been welded on the outside by the preceding welding operation.
In submerged arc welding, endeavors have been made to lmprove the toughness of the heat affected zone as by using fluxes of higher basicity or adding to the weld metals alloying agents such as nickel, molybdenum and tltanium. It was also proposed ~o perform a qo-called multi-layer welding at the sacrifice of welding efficiency or to add to the parent metal trace amounts of Ti, Al-B, REM-B
Ti-Ca or similar combinations of alloying agents. Especially, higher notch toughness is required of the weld metal and the heat affected zone of large diameter pipes that are designed for the frigid zone such as Arctic~ Zone where the temperatures may fall to -25 to -40 C. It is a matter of great difficulty to satisfy the above requirement by using the currently marketed steel plates and welding wires at an acceptable cost.
The present invention relates to a method of arc welding a steel pipe formed by a steel plate bent into pipe shape with abutting ends~ The method comprises welding a seam at the abutting ends by conveying the pipe through leadlng and trailing passes and forming bead layers on opposite sides of a weld ~one during respective passes, and effecting the formation of the bead layer of the weld zone during the trailing pass with a heat input of more than 36,000 J/cm while effecting forced cooling of the weld zone from the reverse side simultaneously, the forced cooling being effected to ~ cool the weld from the weld temperature to 800C within 11 seconds and from 800C to 500C at a cooling rate of 8-40C
per second.
mb/ G~ - 3 -`
According to the welding method of the pre~ent inventlon, the starting plate is bent into the ~hape of a pipe, and the longitudinal seam line of the pipe i8 welded on the inner surface of the tube. Then, the arc or sub-mer8ed arc welding is performed on the outer surface along the same seam line. During this second welding, the inside weld is sub~ected to forced cooling. Thus, the forced cooling is carried out rom the ~lde opposite to the arc used for the formation of the outside weld. Such forced cooling has been ascertained to improve the hardenability of the weld ~etal through suppression of the formation of the proeutectoid ferrite and concurrently mb/- 4 ~351'.~7'~
the tough~e~ o~' the weld throll~h ~ pression of' the form~tion of the upper 'oairlite in the heat af'f'ected ~one.
When the seam line of the thick-walled steel pipe ls formed by several welding passe3, the inventive arc wel-ding accompanied by forced cooling from the reverse side may be applied to the final welding pass.
The toughness of both the welding metal and the heat affected zone will become more pronounced by using higher cooling rates for the forced cooling through the ranges from the maximum temperature to 800C and from 800 to 500C. The cooling rate of 8 to 40C and pre-ferably of 15 to ~0C per second for the cooling range from 800 to 500C is most preferred in order to obtain high toughness for the welding metal and the heat affected zone. As cooling agent, water, liquid nitrogen or dry ice may be employed within the scope of the present invention. In water cooling, the more the flow of cooling water, the sooner the cooling. With forced circulation çooling by three nozzles with the'total flow rate of 45 liters per minute, the cooling of the welded article from 800 to 500C may be carried out at a speed three to four times faster than the case wherein no forced circulation cooling is employed. With use of liquid nitrogen or dry ice, the cooling from the maximum tempe-rature to 800C proceeds usually at a higher speed because of the precooling effects of these cooling agents.
EIowever, the cooling rate from 800 to 500C will be lower than that attained by water because the heat exchange capacity of these cooling agents is lower than . ~ ' that of water. ~
.. . , ~ . .. . . , .. . , . , , i.. ..
10519~9 The increased toughn~ss of the welding metal to be attained by the reverse side coo]ing may be -ttri~utable to the retarded precipitation of coarse ferrites from the , graill boundary with resulting refining of the ferrite ground, and to the increase in the precipitation of the lower bainite. In case of using a starting plate manufactured by controlled rolling, the weld metal of the inside weld may be increased in toughness because of the apparent disappearance of the precipitation hardening brittleness by virtue of the second welding at the outside accompanied by cooling from the inside. With use of the compensating welding wires, the to'ughness of the weld metal at the inner surface of the pipe may become lower than that measured at the intermediate zone or near the outer surface of the pipe. ~Iowever, such decrease in toughness at the inner surface may be hindered by i~crea-sing the rate of cooling. These changes in toughness in the direction of the pipe wall thickness may be compen-sated through.suitable selection of the welding wires used for the respective welding passes.
The increased toughness of the heat affected zone may be attained by promoting the cooling for the range from the maximum temperature to 800C in such a manner that the cooling through said range may be completed with-in preferably li seconds. In this case, the width of the coarse grain zone may be reduced and there is no formation of the coarsely grained preanstenitic crystals.
Furthermore, if the cooling rate for the temperature range from 800 to 500C is increased to 8C/sec or more through the reverse side cooling, a higher toughness ~051979 of ~iAZ ma~ be realized ~llrou~h suppressiorl of the upper bainite and promotion of the lower ~ainite. Gn the other hand, the weld metal and the heat affected zone will become harder for a faster cooling rate. For example, at the cooling rate where the highest toughness may be attained, the hardness will reach a value almost close to saturation. From such consideration, the optimum cooling rate for the temperature range of 800 to 500C is 15C per second or thereabouts.
According to the present invention, however, the upper limit of the cooling for the temperature range of 800 to 500C is selected toA40C per second where the tough-ness starts to be lowered under the so-called quen^hing effect.
Further objects and advantages of the present inven-tion will become more apparent from the following detailed description of the preferred embodiment and the accompany-ing drawings.
`
Brief Description of the D~awin~s Figs. l(a) and l(b) show a typical device designed for executing the inventive method, in side elevation and partial enlarged view, respectively;
Figs. 2(a~ to 2(f) show various examples of the `-reverse side cooling according to the present invention;
Figs.-3~a) is a diagram showing the cooling time required for the three test pipes to cool from the maximum temperature to 800C for the spontaneous cooling and the various inventive reverse side cooling modes;
:~05~979 Fig. ~(b~ is a ;illlilar ~iagram for the temperclture range l`rom 800 to 500C;
Fig. 4 is a diagram sho~ling the welding heat input versus cooling time for the spontaneous cooling and the various inventive reverse side cooling m~thods;
Fig. 5 shows typical cooling curves for steel test piece C by the conventional cooling and the inventive cooling;
Fig. 6(a) is a cooling time versus toughness dia-gram chart for the steel test piece A cooled by reverse side cooling through the range of the ma~imum temperature to 800C;
Fig. 6(b) is a~similar chart but for the temperature range of from 800 to 500C;
Figs. 7(a) and 7(b) are the similar diagrarn charts to Figs. 6(a) and 6(b) for the steel test piece B;
Figs. 8(a) and 8(b) are the similar diagram charts to Figs. 6(a) and 6(b) for the steel test piece C;
Figs. 9(a~ and 9(b) are the similar diagram charts to Figs. 6(a) and 6(b) for the steel test piece D;
Figs. lO~a) is a cooling time versus toughness diagram chart for HAZ of the steel test piece A for the temperature range from maximum temperature to 800C
Fig. lO(b) is a similar diagram chart but for the temperature range from 800 to 500C;
FigS. ll~a) and ll(b) are the diagram charts for the steel t st pieces B and C similar to Figs. lO(a) and lO(b);
Figs. 12(a) and 12(b) are the diagram charts for the steel test piece D similar to those shown in Figs.
... .. r lO(a) and lO(bj;
Figs. 13(a) is a cooling tirne versus hardrless diagram chart for ~Z and weld metals of the steel test pieces A, B and C and for the temperature range from maximum temperature to 800C;
Fig. 13(b) is a similar diagram chart but for the temperature range from 800 to 500C;
Fig. 14 is a dia~ram chart showing the distribu-tion of hardness through the inside and outside weld metals for the steel test piece B;
Fig. 15(a) is diagram chart showing the distri-bution of hardness through the inside and outside weld metals for the steel~test piece B in the case of conven-tional welding with spontaneous cooling;
Fig. 15(b) is a similar chart to Fig. 15(a) but in the case of 3 nozzle water cooling with forced circu-lation;
Fig. 16(a) is a cooling time versus hardness dia-gram chart in, th~e case of the reverse side cooling for the HAZ and the weld metal for the temperature range from the maximum temperature to 800C;
Fig. l~(b) is a similar chart to Fig. 16(a) but for the temperature range from 800 to 500C;
Fig. 17(a) is a photo showing the microstructure of the coarse grain zone of the HAZ of the test piece A
obtained by conventional welding with spontaneous cooling;
Fig~ 17(b) is a similar photo but showing the same portion of the test piece A obtained by the inventive welding with reverse side cooling;
Figs. 18(a) and 18(b) are the photos similar to 11D5~'~79 Figs. 17(a) ~nd 1'1(~) but for the ,sarne portion of' th~
test piece ~;
Fig. 19(~) is a photo s~lowing the microstructure at the outer surface of the weld metal of the test piece obtained by conventional welding;
Fig. l9(b) is a similar photo to Fig. l9(1) for the same test piece B but obtained by the inventive welding with reverse side cooli~g;
~ ig. 20(a) is a diagram chart showing the coo]ing time from the maximum temperature to 800C plotted against length of the coarse grain zone of the HAZ for the test pieces A, B, C and D;
Fig. 20(b) is~a similar chart to Fig. 20(a) but -showing the cooling time from 800 to 500C;
Fig. 21 is a diagram chart showing the toughness distribution of the inside weld in the direction of the weld thickness with use of different and same welding wires for the inside weld; and Fig. ~2 is a diagram chart showing the hardness distribution through the inside and outside weld metals in the direction of thickness for the test piece C obtained by conventional SAW with use of same inside wires and the same test piece obtained by inventive SAW with use of same inside wires.
Description of the Preferred Embodiment , In Fig. l~a), a steel pipe 2 conveyed axially on transport rollers 12 has been welded previously along a longitudinal seam line from the inside and is to be .. ~ ~ . .. .. .
welded to a weldlng machlne 1 on the tr~nsport roller~ 12 along the same longitudinal seam llne but now from the outside. As the pipe 2 is being welded from the outside, it is cooled ~imultaneously by a cooling device 3 from the reverse side which is, ~n the embodiment illustrated, the inner side of the steel pipe 2.
The cooling device 3 i8 shown in the enlarged view o~ Fig~ l(b) and comprised o a boom 7 extending lengthwise of ~he transport line and fitted at the end with a cooling water tank 9 and a pump 10. The end of the boom 7 iq also provided with a support roller 11 adapted to travel along the inner surface of the steel pipe, and a cooling water nozzle o directed towards the weld along the seam line.
Cooling water may be supplied to nozzle 8 from a water pipe 5 connected to the external water source and by way of the tank 9 and the pump 10. The cooling water may be sprayed by the nozzle 8 towards the inside weld metal at a predetermined ~ flow rate by the controlled operation of the pump 10. The water pipe 5 and a power cable 4 such as wire for driving the pump 10 are passed through the boom 7 and extended to an external device including a water source and an electrical source, not ~hown. In Fig. l(b), the numeral 6 denotes a leading electrode or wire and a trailing electrode or wire of the welding machine 1 mounted in a confronting relation with the cooling nozzle 8. The numeral 13 denotes a water pan for receiving the used water which flows down and is returned to the tank 9. The boom 7 may be displaced length-wise by a support mb/~
A
~ ' ~ O ~ 37 ~
device 14 for ch~nging the po,itlon of the noz~le rela-tive to the pipe 2.
The steel pipe ~, ~hich has been welded along a longitudinal seam line on the inside may thus be welded on the outside along the same seam line as it is fed in the direction of the arrow marks in Figs. l(a) and l(b) along the transport rollers 12. Simultaneously ~, Jnslc~e with o*tei~e cooling, the cooling water is sprayed from the nozzle 8 to the inside weld directly opposite to the welding electrodes.
In Figs. l(a) and l(b), only one nozzle 8 is used for spraying cooling water to the reverse side of the outside weld. Howe~ver, two or more nozzles may be provided in tandem along the seam line, and the cooling water may also be circulated forcibly for increasing the heat exchange capacity. Alternatively, cooling water may be replaced by liquid nitrogen, or a dry ice may be mounted in direct physical contact with the inside of the pipe for cooling.
Reference is now made to Figs. 2(a) to 2(f) showing these various cooling modes that may be used with the submerged arc welding.
Fig. 2(a) shows a single nozæle 8 being used for reverse side cooling as already explained with reference to Figs. l(a) and l(b). In Fig. 2(a), the upper view shows the weld in the transverse section `~
and the lower view shows the weld in the longitudinal section. In Fig. 2(a), the three dotted lines indicate drill holes bored from the end face of the test steel piece to the bead for insertion of thermocouples used .
105197g for measurirg the temperaJur~ as will be described, In Fig. 2(b), two noz%les 8 are placed in tandem along the seam line. By using the nozzles of the same size, the flow rate of the cooling water may be doubled with consequent increase in the rate of cooling.
For example, if the flow rate is 12 liters per minute for a single nozzle, the flow rate will increase to 24 liters per minute with two nozzles shown in Fig. 2(b).
In Fig. 2(c), three nozzles 8 are placed in tandem on the seam line for obtaining the flow rate of 36 liters per minute with a still faster rate of cooling.
In Fig. 2(d), a compulsory. circulating unit 15 such as tank and pump 10 shown in Fig. l(b) is used for directing the cooling water under pressure to the three nozzles 8 for realizing a faster cooling rate by increasing the flow rate to, for instance, 45 liters per minute.
In Fi~. 2(e), there is shown a liquid nitrogen source, suc~ as y essel 16, fro~ which the liquid r.itro-gen is delivered to plural nozzles 18 mounted within a hood 17. ~he liquid nitrogen may be sprayed from the nozzles 18 to the reverse side of the outside weld metal at a rate of, for instance, 400 kg per hour.
In Fig. 2(f), an air hose 19 is d,ilated for applying a piece of dry ice 20 to the reverse side of the outer weld metal for cooling the weld.
In Figs. 2(e) and 2(f), the three dotted lines indicate the mounting positions for thermocouples as in the case of Fig. 2(a).
These various cooling modes for reverse side 105~979 cooling rnay be selectively elnployed fol cooling the weld of a steel pipe which has been welded on the inside and is beine welded on the outside. With such cooling, the formation of the proeutectoid or of the upper bainite at the ~IA% may be prevented with consequent increase in the toughness of the weld.
A series of tests on arc welding associated with reverse side cooling were conducted with test steel plates A, B, C and D, for checking the effects of vari-ous cooling rates on the toughness, hardness and micro-structure of the welding metal and the HAZ.
Four test steel pieces A, ~, C and D with varying thicknesses, grades and chemical compositions as shown in the following Ta~le 1 were employed. The test plate D was 10.3 mm thickness, the test plates ~ and ~
were prepared by controlled rolling and 20.3 and 18.3 mm thickness, respectively, while the test plate A was made of 1.4 Ni steel with thickness of 25.4 mm.
`
Tsble 1 Chemical Composition of Test Steel pieces (~adle) (%) Cr W.T C 3i Mn _ 9 Cu Ni Mo Nb V ~ol~l .~ x6S 25,4 .0~2 .25 1.17 .014 .007 _ 1.41 17 _ _ 053 x70 20,3 .090 ~1 1.45 .012 .004 _ .26 .15 .0~7 .092 .045 (` x70 18.3 .OS6 .23 1.4~ .014 .005 .19 .21 _ .039 .091 .020 _ x52 10.3 .126 15 .65 .015 .009 _ - ~ = .017 Test welding wires a, ~ used for the tests are shown in the following Table 2. Test wire a was a Ni - Mo - Ti alloy and the test wire b was free of any alloy-ing agents.
Table 2 Chemical Composition of Test Wires (%) = dia C Si Mn - S I~i Mo Ti a 4.0 .03 .17 2.03 .011 .007 .91 .52 .21 , . .
b 4.0 .07 .01 .68 .009 .017 _ _ Test fluxes X, Y used for the tests are shown in the following T~b~e 3. The test flux X was weakly basic with basicity equal to 1.2 and the test flux Y
was strongly basic with basicity equal to 1.6.
Table 3 Chemical Composition of Test Fluxes (%) .~= Si02 OaO MnO M~O ~123 CaF2 2rO2 ~ S FeO
. ?X 32 20 11 8 11 13 ~ .Oll .l63 .98 . Y 28 28 _ 19 18 ~ _ .022 .222 .22 The welding wires and fluxes and other welding conditions are shown in the following able 4 for the respective test pieces A to D.
~OSlg79 Trl~le 4 Welding Condition ~, I Le~ding rr~iling~ Speed ~eat l x . ~____ V A ¦ V ,~ ~ ~I mm/min I7put lwire , r~ ~
Inside 1 7GO 36 640 l40 I4 j 1,~50 24,400 a~a X
D _ ! _ Outsidel 940 38620 40 15 ¦ 1,000¦ 36,300 a+a Y
1 .- __ ~ , .. .. ._~ . .
Insidel 920 36 70040¦ 14 850¦ 43,100 b~a Y
C Outsid`e 960 38 700 40~ 15 800j 48,400 a~a Y
--I' . I ! ~
Insidel 940 36 70040l 14 1 ~50l 43,700 ! b+a Y
; _ _ , _ ! - _ _ Outsidq~ 960 38 700 140l15 ~ 800 ! 48,400 a+a Y
_ . ,. _ , .
Insidell,OOO 38 8004014 ¦ 7GOi 60,000n+a Y
I\ l l l . ,"~.
Outsidql,050 38 90040¦15 ¦ 700¦ 65,100a+a Y
~ _ _ lI _ .... I .1.__. _ The scarf angle for the inside and welds are 45 r~
for the respective test pieces A to D. The depth of scarf was sele.cted to one-third of thickness for each of the insid~ a~d outside welds for the respective test pieces A to D.
In Figs. 3(a) and 3(b), the cooling time intervals necessary for cooling from the maximum temperature to 800C and fro.m 800C to 500C are shown for the cases of conventional welding with natural coolin and the welding associated with the various cooling modes. It is seen from Fig. 3(a) that the cooling time necessary for cooling from the m~ximum temperature to 800C may be progressively reduced with use of one and two nozzles as compared with spontaneous cooling. With three nozzles, the cooling 105~79 time re~m~lins ,llrnost cqual to th~t o~tairled with two nozzles, and a somewhat shorter cooli~g time may be attained by use of compulsory circulation of the cooling water. It is also seen from ~'ig. 3(b) that the cooling time interval necessary for cooling through the range of 800C to 500C may be reduced progressively with increase in the num~er of the nozzles and hence in the flow rate and that use of forced circulation of cooling water results itl a cooling rate several times faster than in the case of spontaneous cooling.
It is also seen from Figs. 3(a) and ~(b) that the use of liquid nitrogen or dry ice-gives almost the same cooling rate as that~attained with forced circulation with three nozzles (flow rate, 45 liters/min.) for the temperature range from 800C to 500C. These high cool-ing effects, proper to liquid nitrogen and dry ice, may be attributed to the precooling of the weld by the liquid nitrogen prior to spraying from the nozzles or by the dry ice prio,r to~ the arc generation from the welding torch. With liquid nitrogen and dry ice, the cooling time necessary for cooling from 800C to 500C will become much longer than in the case of water cooling because of the lower heat exchange capacity of the liquid nitrogen or dry ice than that of water.
In Fig. 4~ the cooling time necessary for cooling from 800C to 500C is shown for varying heat inputs ~or the steel piece A for conve~tional welding and the inve~tive weld~ng with the various cooling modes.
The straight solid line in Fig. 4 represents the calcu- .
lated value for the conventional SAW with natural cooling.
_ .. . ., . .. . ... , y Thc mcuuur~d V~ ICS ~or th~ cas~ for conven~l(>nal we~dlng an(l tl~ inv~ntl~e cooling ~lth wa~er coolinK with one noz%]e are shown in Fig. 4 in thc vlcinlty of 65,0~0 J/cm, while the m~asured values for the other cooling mode~ are shown in Fig. 4 for the heat input of 65,000 J/cm. In Fig. 4, thc blanl~ed marks represent the measured values at the intermedlate portion of the test piece and the black marks represent the values measured at the outer surface of the test piece. It is seen from Fig. 4 that a rate of cooling almost three times faster than that with conve~tional welding may be obtained with the inventive welding with one nozzle. With increase in the number of nozæles and hence in the flow rate of cooling water, a faster rate of cooling may be obtained and, with forced circulation of cooling water with three nozzles (flow rate:
45 lit./min.), the mean cooling rate may be elevated to 23 C/sec. As àlso seen from Fig. 4, the rate of cooling at the intermediate zone is faster than that at the upper ` surface of the test piece as a result of the reverse side cooling.
Fig. 5 shows typical cooling curves for the test pieces C which are subjected to the conventional cooling with natural cooling and the inventive welding with reverse side cooling. As seen from Fig. 5, the maximum cooling capacity may be obeained with forced circulation of cooling wnter with three nozzles (flow rate, 45 liters per minute).
The eest steel piece A was sub~ected to a submerged arc welding pass on the inside under the conditions shown in Table 4 and then to a submerged arc welding on the outside while the inside of the pipe was cooled with water, nitrogen and dry lce according to the above-mentioned six cooling modes. Figs. 6(a) and 6(b) show the measured values of the ~b/~ - lô -' 1051g79 weld ~ou~ n~ p]~tt~ ~gtlinst coo]lng tirnc neces~r~ for coollng from th~ mnximl~m temperaLIJre ~o ~C and from 800C to 500oc, respectively. The toughness of tll~ weld was measured three places, namely close to the outer surface of the weld, the lntermediate portion and close to the inncr surface of thc weld. The same wires a for the leading and trailing wires and the flux Y were used for forming the inside and outside weld metals of the weld.
As seen from Fig. 6(a), the shorter cooling time for the specified temperature range is effective to improve ehe toughness of the weld metal. As also seen from Fig.
6(b), the shorter cooling time for the specified temperature ranBe imparts higher toughness to the weld metal. The weld metal of the test piece A has presumably elevated hardenability and a substantially constant value of vE - 30C
= 127 to 136 ft - lb may be obtained at a mean coo~ing rate of 15C per second (cooling time interval, 20 seconds) for the temperature range of 800C to 500C.
The test steel piece B was also subjected to a submerged arc welding pass under the conditions shown in Table 4. The pipe thus formed was then seam welded at the outside by submerged arc welding while the inside weld was cooled in the above-mentioned manner.
Figs. 7(a) and 7(b) show the measured values of toughness of the weld metal against the cooling time intervals necessary for cooling through the specified ranges of temperature, In Figs. 7(a) and 7(b), the wires a and b and the flux Y were used for the inside welding and the same wires a with the flux Y were used for the outside welding. The weld toughness was measured at three positions, namely close to the outer surface of the weld, the intermediate portion ~b/ ~ - 19 -an(] closc to Lh~ ln~ r slJrf.lcc of ~hc wel(l.
As seell from Fi~s. 7~a) and 7(b), thc shorter the cooling timc lnterva]s for the ~pecified ranges of tempera-ture, the higher the toughness of the weld metal. E~pecially, the cooling time for thc temperature range from 800C to 500 C shown in Fig. 7(b) decidedly affects the toughness of the weld metal. In the present case, the compensating wires (leading wire a and trailing wire b) were used for inside welding to reduce the quenching effect that might be exerted on the inside weld at the time of the outside welding. Wllen the test piece B manufactured by controlled rolling is sub~ected to the conventional welding, the inside weld will be lowered in toughness compared to outside weld or the transition zone between tbe outside and inside welds because of precipitation hardening caused by the precipita-tion of Nb and V from the parent metal and of Ti from the welding wires. In the case of the inventive welding with reverse side cooling, there is not sufficient time and ~ temperature -for these elements to precipitate and hence the zone of precipitation brittleness can be minimized.
Thus, with the welding method of this invention, the need for using compensating wires may be eliminated because of the suppression of the precipitation hardening of the weld metal applied in the preceding pass. The toughness of the weld metal may be further improved by using the wires with high hardenability such as wires a for the leading and trailing wires.
In Figs. 7(a) and 7(b), the wires a and b and the flux Y were used for the inside weld, while the wires a and the flux Y were used for the outside weld.
mb ~ 20 -.
: :
105197~t rc ~ f ; i rn i l .l r t~ t~ c ~ (l f ~r t 1 t ~ ~: l p l cl ~ ~ C ~ r~ wll ,in F~ J ~ 3 ( ~ ; c~ n fro~ tllese Figurcs, t1le test result.s for tllc stee~ ple(e C are subsLanti~lly the same ac; tllose for the steel pieces ~ and R.
In Figs. 6 to 9 inclllsive, the blanked circles represent the data as measured towards tlle outer surface, whereas the semi-blanked circles and the black circ].es rcpresent tlle test data as measured with the thermocouples ln at the intcrmediate ~ones and towards the inner surface, respectively.
When a thin steel piece of a lower grade is formed into a pipe by SA~ with use of a conventional wire a, it may frequently occur that the weld metal obtained has not sufficient toughness. For this reason, it WAS
proposed in the past to use a Ni - Mo - Ti wire and a flux added with ~InO and ZrO2 such as flux X shown in Table 3. Figs. 9(a) and 9(b~ show the toughness against cooling time of the weld joint obtained by using the low-grade thin-walled test steel piece D and a combination of the conventional fluxes and wires.
The steel piece D was formed into a pipe shape by submerged arc welding on the inside by using the same wires' à for the leading and trailing wires and the f],ux X.
The tube thus formed was then subjected to the submerged arc welding on the outside by using the same wires a~a and the flux Y while cooling the inside of the tube simultaneously, The toughness of weld metal was measured ne'a-r t~e outex and inner surfaces and the intermediate portion with respect to the cooling timc for the two ranges of cooling temperature.
mb/~,~ - 21 -~05~979 Whllc no flxcd relation may lc obffervccl to exi~t ln Fig. 9(a) between the touf~hncss of the weld met~ and the cooling time above 800C it may bc apparcnt from Fig. 9(b) that the shorter cooling time for thc range of 800C to 500 C ls highly effective to improve the toughness of the we]d. It may be seen from Fig. 9(b) that the toughness may be improved by about twice in terms of vF-30C and by about 40C in terms of vTrs as compared wi~h the cflse of using spontancous cooling.
In Figs. lO(a) and ]O(b) the toughness of HAZ of a steel A pipe obtained by submerged arc welding under the welding conditions shown in Tflble 4 is plotted against cooling time. It may be seen from these Figures that the UAZ toughness may be improved with shorter cooling time for the both temperature ranges and in the case of reverse side cooling from 800C to 500C within 14 seconds the toughness may be improved by about twice in terms of vE-30C and by about 15C in terms of vTrs as compa~re~d with the case of spontaneous cooling.
mb/p~ - 22 -~051~37~
It is beli~ved that pro!lotc-l cooling through the ter~pe-rature above 800C resl)lt;s in reduced s~ 7,C3 of preaus-tenitic crystal grairs arld prevention of the enfeebled grai~ boundary. It is also p~esumed that promoted cooling through 800C to 5Q0C favors the formation of a highly tenacious lower ~ainite while preventing the for-mation of the upper balnite which may be detrimental from the viewpoint of toughness.
Figs. ll(a) and ll(b) show t~e similar results obtained with steel types B and C. With the type B, with faster cooling rates through the both temperature rar.ges, the HAZ will become increasingly tough. Above all, the ccoling time through the range of 800C to 500C
decidedly affects the toughness cf HAZ. When t~e ~iA7, has been cooled in 12.5 seconds through said temperature range, the measured v~lues of toughness may be improved by about 1.8 times in terms of vE-30C and by about 33C-in terms of vTrs as compared with the case of spon-taneous cooling. With type C, the cooling time through 800C to 500C decidedly affects the toughness of the resulting ~IAZ. For example, when the latter has been cooled in 17 seconds through said range, the values of toughness may be improved by about 1.5 times in terms of vE-30C and by about 38C in terms of vTrs as compared with the case of spontaneous cooling. The similar results may be observed in Figs. 12(a) and 12(b) for the steel type D, While there is observed no fixed relation between the toughness and cooling time for the temperature range above 80QC, it may be seen from Fig. ll(b) that shorter cooling time required for cooling through 800C
..... .... .. . . . . . . .. . . . .. . . . . .. .. . .
to 500(- r~sll]t~ ln a ~nurkc~lly l~ rov~d ~ollg~lnc~ ~n terme of vTrs.
Flgs. 13(a) an~ 13(h) S~IOW thc effects of the inventive rever6~ side coollng on the maximum h~rdness of the ueld metal and I~A% of test pipes formed of the steel pieces A, B and C. The test pipes were prepared by welding along a longitudinal ~oint line on the inside and then welding along the same joint line on the outside with simultaneousi forced cooling from the side of the inner weld, as described in the foregoing. With these steel types, there is again no definite relation between the hardness and the cooling time through the temperature range above 800C. However, as shown in Fig. 13(b), with shorter cooling time for the temperature range through 800C to 500C, the hardness of the weld ~one may be increased progressively until substantially constant values of Hv.
max. 238 for weld metal and Hv. max. 260 for llAZ are attained for the steel types A and-B. With the steel types B and C, ~ as`the cooling rate increases for the range of 800C to 500C, the maximum hàrdness of the weld metal tends to decrease at the outset and then starts to increase at a slow rate while HAZ tends to increase gradually. With steel type B, the hardness of HAZ reaches the Hv. max 258 at the mean cooling rate of 24C per second for the range of 800C
to 500C. The carbon equivalent Ceq of the steel type C
is 0,368 which is lower than that of steel type B (0.385~, and thus the hardenability of the steel C is lower than that of the steel B, This accounts for a somewhat lower value of the maximum hardness of mb/p~ - 24 -105~7~
the 1~7, ~1n~ T1e~d rnetal of ~ne stccl t~pe C tha~ th-t of the steel type B.
With tne stee~ types ~ and C obtained by control~ed rolling, the weld metal will be lowered in hardnes3 with shorter reverse side cooling time for the temperature range from 800C to 500C, as discussed in the foregoing.
Fig. 14 shows the hardness distribution in the direction Or plate thickness of the inside and outside weld metals for the steel type B weld joint. In general, ~ith the conventional S A W followed by spontaneous cooling, designated as Conv. SAW in the drawing, the inner and outer su~;faces and the portions ad,jacent thereto are softened by heating to approximately the fusing points and the elements such as Nb, Ti and V are turned into solid solution. On the other hand, the intermediate portion of the weld metal spaced about 2 m~ from the inner and outer surfaces are heated to a range from 600~C to Ac point and hardened by precipitation of Ni, V and Ti in the form of carbides and nitrides. However, when the steel pipe welded 011 the outside is welded on the inside with simultaneous reverse side cooling designated as C. C. SAW in the drawing, the range of temperature and time necessary for precipitation of these elements will be limited thus the precipitation hardening may be suppressed with consequently lowered hardness in the inr;er zone.
On the other hand, with increase in the rate of reverse side cooling, the outside weld is hardened, because the same wires a are used for the outside we.ding.
Therefore, with the weld joint obtained by the inventive welding, the region of maximum hardness will be shifted 1(~51979 from the con~ rable regior~ of the weld o~tained the con-ventional welding which is not accompanied by rever~e side cooling. The E~Z hardness is increased with an elevated rate of reverse ~ide cooling for the respective steel types as discussed in the foregoing, with the region of maximum hardness being situated at a mid zone which is spaced 0.5 to 1 mm apart from the fusion line, as indicated in Figs. 15(a) and 15(b). It is the cool-ing time for the range of 800C to 500C, rather than that for the range above 800C, that affects markedly on the hardening of the weld. It is believed that the cooling time for the temperature range above 800C affects the grain sizes of the austenite crystal grains, whereas the cooling time for the range between 800C and 500C
favors the precipitation of the lower bainite while suppressing the formation of the upper bainite.
In ~igs. 16(a) and 16(b), the test results on the effects of reverse side cooling on the hardness of the HAZ and weld meta~l for the low grade steel ~ are illust-rated. It is seen from these Figures that the hardness ~-of HAZ and weld metal is markedly affected by the cooling rates for the specified ranges of temperature and that the hardening may be more pronounced with faster rates of cooling. As sllown in said Figures, with reverse side ;
cooling in 15 seconds through the range of 800~C to 500C, the hardness of the weld metal amounts to Hv. max. 242, while that of IIAZ equals to Hv. max. 193.
Figs. 17(a), 17(b) and Figs. 18(a), 18(b) show the photos (magnification ratio : 400) of the microstructures .
of the HAZ for the steel types A and B, respectively, '. - .. ~ , :
10519'79 which are obtair~ed by con~entior,a] wclding arld tha~
obtained by the inventive welding with use of forced circulation with three nozzles. As seen from these Figures, the sizes of the preaustenite crystal~ of the coarsely grained HAZ have been reduced for both the types A and B as a result of the reverse side water cooling. It is also seen that the HAZ is ~ade highly tough on account of the predominant formation of the u~cr bainite and the corresponding suppression of the upper bainite.
Figs. l9(a), l9(b) are the photos taken by an electron microscope (magnification ~atio : 3000) showing the microstructures of the outside weld metals of a steel B tube obtained by conventional welding and a tube of the same steel type obtained by the inventive welding.
As seen from these Figures, by the reverse side cooling, the formation of coarse ferrites in the weld zone has been suppressed and a highly refined ferrite structure may be observed along with a small amount of bainite.
The width and siæe of the coarsely grained HAZ can be reduced by reverse side cooling. Figs. 20(a), 20(b) show the widths and lengths of the coarse grain zone of the HAZ for the respective steel types for the various rates of reverse side cooling. As seen from these Figures, the widths and lengths of the coarse grain zone of HAZ
for the respective steel types may be reduced by using ~ -~
shorter cooling rates for the temperature ranges above 800C and from 800C to 500C. With use of liquid nitrogen and dry ice, a faster cooling rate may be attained for the temperature range above 800C o~ account of pre- ``
.~ .
cooling e~r~c~ n(l hencc tllc 81~.e8 of ~hc corlr~e ~,r~in zone mny bc re~u~ed evcn lf a ~]owcr coollng rate ~hou]d be used for the temper~turc range bctwecn 800"C and 500C.
Thus, for elevating the toughness of IIAZ, lt is necessary to cool promptly through the above-mcntioned two temperature ranges. Prompt coollng through the range above 800C is effective to reduce the width of the coarse grain zone and to prevent the preaustenite crystal grains from becoming coarse. On the other hand, prompt cooling through the range of from 800C to 500C is effective to prevent the precipitation of the upper bainite and to favor the formation of the lower balnite. The above applies to the weld metal as well. Thus, prompt cooling through the two temperature ranges is effective to make the ferrite ground of the molten metal more refined and to facilitate the formation of the lower bainite with resulting increase in the toughness of the weld metal.
In case the parent metal is prepared by controlled rolling, the outside weld of a pipe which has been welded previously on the inside is known to become extremely brittle by precipitation hardening when allowed to cool. So far, the compensating wires were used for welding the outside of a steel pipe welded previously on the inside. In case of steel pipe that is welded at the inside and outside by using an arc welding accompanied by reverse side cooling, the inside weld tends to be lowered in hardness as compared with the outside weld. In order to make the hardness and toughness of the weld more mb/J C> - 28 -~0~ f~
uni~orl~l, te~its have ~)een co;iducte~ t,y u~sirlg the sarrlr: wire _ and dif~erent wires a, b ~or the lea~in~ arld trail~ng wires. The test results ale sllotln in Fi~s. 21 and 22.
In Fig. 21, the toughness of the weld is plotted again~t the distance from outside for the cases in which the steel pipes (types ~ and C) were subjected to the two side welding followed by spontaneous cooling (Conv. Weld) and to the inventive welding (C. C. SAW) with use of two nozzles (flow rate, 24 liters per minute). As seen from Fig. 21, with use of the leading wire b and ~*~}n ~
wire a, the inside surface zone of the weld metal obtained l~0rG
by the inventive welding is ~e~ff tenacious than the intermediate and outs,ide surface zones, whereas the intermediate zone of the weld metal obtained by the inventive welding with use of the same wire a for both the leadilg and trailing wires i3 almost as tenacious as the remaining zone. ~ig. 22 shows the hardness of the weld plotted against distance from outside surface of the weld, with the p~rtition line between the oulside and in-side weld being disposed at about 13 n~ from the ourside surface. The test piece was steel C and the dotted and solid lines represent the curves obtained with conventional welding and with the inventive welding with two nozzles, respectively. It may be presumed that the inside weld obtained by usint~ the com~ensating wires twires a ~nd b) is lowered in hardness because its harde-ning process is affected adversely by the reverse side cooling and also the precipitation hardening is prevented thereby from occurring. As seen from Fig. 2~, with use of the same wires a for both the leading and trailing -.. . . . .
1~51979 wires, uniform distribution of hardness may be attained ln the direction of the plate thickness. It 1~ belleved that precipitation hardening i~ the vlclnity of the in~ide surface of the weld is ~uppressed by reverse slde cooling while the inside surface is slightly hardened by the coollng thus imparting sufficient toughness to the zone close to ehe inside surface. Thus, with the inventive welding with reverse side cooling, if the same wires a are used for the leading and trailing wires on the occasion of the in~ide welding, the weld metal may be made uniformly tough along its thickness. In this way, uniform hardness may also be attained along the plate thickness because of suppression of the precipitation hardening which might otherwise occur at near the inside surface.
In the foregoing, the inside weld is formed previously on a tube and the outside of the tube is welded simultaneously with cooling at the inside. It is however possible to weld at the outside of the pipe and then to weld at the inside thereof with cooling simultaneously at -~
the outside. Also, when the inside and outside welding is -performed in plural passes, the reverse side cooling may be performed in the last pass. Namely, one side welding is effected on a leading or forward pass and reverse side welding with concurrent cooling is effected in a trailing or reverse pass.
Although the present invention has been described for the case of submerged arc welding with two electrodes, it can be applied to the case of gas shield welding. A
single electrode or three or more electrodes may also be ~0 employed. However, since the submerged arc welding is usually associated with the lower rate of cooling, and the use of plural electrodes is usually associated with higher heat input, the present invention may be mb/ -?
applied advantageou~ly to the ca~e o~ one-pas~ submerged arc welding of a large-diameter thick-wfllled steel tubin~
on the inside and outside with use of two or more elect-rodes.
t : '
Claims (6)
OR PRIVILEGE IS CLAIMED ARE DEFINED AS FOLLOWS:
1. A method of arc welding a steel pipe formed by a steel plate bent into pipe shape with abutting ends, said method comprising welding a seam at the abutting ends by conveying the pipe through leading and trailing passes and forming bead layers on opposite sides of a weld zone during respective passes, and effecting the formation of the bead layer of the weld zone during the trailing pass with a heat input of more than 36,000 J/cm while effecting forced cooling of the weld zone from the reverse side simultaneously, the forced cooling being effected to cool the weld from the weld temperature to 800°C within 11 seconds and from 800 C to 500°C at a cooling rate of 8-40°C per seconds.
2. A method as claimed in claim 1 wherein the bead layer of the weld zone in the forward welding pass is effected by submerged arc welding using front and rear Ni-Mo-Ti electrode wires.
3. The method as claimed in claim 1 wherein the forced cooling is performed at a rate of 15 to 30°C per second until the temperature of the weld is cooled from 800°C to 500°C.
4. The method as claimed in claim 1 wherein water is used as cooling medium.
5. The method as claimed in claim 1 wherein liquid nitrogen is used as cooling medium.
6. The method as claimed in claim 1 wherein the dry ice is used as cooling medium.
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP9575575A JPS5220345A (en) | 1975-08-08 | 1975-08-08 | Process for welding steel pipes |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CA1051979A true CA1051979A (en) | 1979-04-03 |
Family
ID=14146302
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CA258,605A Expired CA1051979A (en) | 1975-08-08 | 1976-08-06 | Method of arc welding with reverse side cooling for obtaining highly tough large-diameter welded steel pipes |
Country Status (6)
| Country | Link |
|---|---|
| JP (1) | JPS5220345A (en) |
| CA (1) | CA1051979A (en) |
| DE (1) | DE2635743C2 (en) |
| FR (1) | FR2320159A1 (en) |
| GB (1) | GB1552660A (en) |
| IT (1) | IT1067834B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10695876B2 (en) | 2013-05-23 | 2020-06-30 | Crc-Evans Pipeline International, Inc. | Self-powered welding systems and methods |
| US10828715B2 (en) | 2014-08-29 | 2020-11-10 | Crc-Evans Pipeline International, Inc. | System for welding |
| US11175099B2 (en) | 2013-05-23 | 2021-11-16 | Crc-Evans Pipeline International, Inc. | Systems and methods for use in welding pipe segments of a pipeline |
| US11458571B2 (en) | 2016-07-01 | 2022-10-04 | Crc-Evans Pipeline International, Inc. | Systems and methods for use in welding pipe segments of a pipeline |
| US11767934B2 (en) | 2013-05-23 | 2023-09-26 | Crc-Evans Pipeline International, Inc. | Internally welded pipes |
Families Citing this family (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS52110244A (en) * | 1976-03-12 | 1977-09-16 | Hitachi Ltd | Method and apparatus for welding |
| JPS53115387U (en) * | 1977-04-07 | 1978-09-13 | ||
| WO1995023669A1 (en) * | 1994-03-01 | 1995-09-08 | Siemens Aktiengesellschaft | Process for welding metal workpieces and device for implementing it |
| DE19608108C2 (en) * | 1996-03-02 | 1999-04-15 | Audi Ag | Cooler |
| DK1531959T3 (en) | 2002-07-17 | 2008-06-16 | Shell Int Research | Method of joining extensible tubes |
| ATE435085T1 (en) * | 2002-07-25 | 2009-07-15 | Shell Int Research | FORGING WELDING OF PIPES |
| US7282663B2 (en) | 2002-07-29 | 2007-10-16 | Shell Oil Company | Forge welding process |
| US7774917B2 (en) | 2003-07-17 | 2010-08-17 | Tubefuse Applications B.V. | Forge welding tubulars |
| DE102006033992A1 (en) * | 2006-01-23 | 2007-08-02 | Schmidt + Clemens Gmbh + Co. Kg | Welding process for e.g. welding stainless steel pipes comprises placing two pipes in a position necessary for forming the peripheral joint, moving a cooling body into the pipes and further processing |
| GB201203030D0 (en) | 2012-02-22 | 2012-04-04 | Tubefuse Applic B V | Forge welding of tubular articles |
| EP3389914A1 (en) * | 2015-12-18 | 2018-10-24 | Autotech Engineering, A.I.E. | Reinforcing structural components |
| CN116213887B (en) * | 2023-05-08 | 2023-08-11 | 中南大学 | Argon arc welding repair method for 1.2738 steel plastic mold |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE1007448B (en) * | 1954-11-17 | 1957-05-02 | Philips Nv | Process for inert gas arc welding in an atmosphere containing carbon dioxide with a melting wire electrode |
| DE1115381B (en) * | 1959-07-01 | 1961-10-19 | Philips Nv | Use of a welding electrode made of a steel core with a basic, low-hydrogen envelope containing ZrO and a quantity of silicate |
-
1975
- 1975-08-08 JP JP9575575A patent/JPS5220345A/en active Granted
-
1976
- 1976-08-05 GB GB3260576A patent/GB1552660A/en not_active Expired
- 1976-08-06 IT IT2612476A patent/IT1067834B/en active
- 1976-08-06 CA CA258,605A patent/CA1051979A/en not_active Expired
- 1976-08-06 FR FR7624126A patent/FR2320159A1/en active Granted
- 1976-08-09 DE DE19762635743 patent/DE2635743C2/en not_active Expired
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10695876B2 (en) | 2013-05-23 | 2020-06-30 | Crc-Evans Pipeline International, Inc. | Self-powered welding systems and methods |
| US11175099B2 (en) | 2013-05-23 | 2021-11-16 | Crc-Evans Pipeline International, Inc. | Systems and methods for use in welding pipe segments of a pipeline |
| US11767934B2 (en) | 2013-05-23 | 2023-09-26 | Crc-Evans Pipeline International, Inc. | Internally welded pipes |
| US10828715B2 (en) | 2014-08-29 | 2020-11-10 | Crc-Evans Pipeline International, Inc. | System for welding |
| US11458571B2 (en) | 2016-07-01 | 2022-10-04 | Crc-Evans Pipeline International, Inc. | Systems and methods for use in welding pipe segments of a pipeline |
Also Published As
| Publication number | Publication date |
|---|---|
| FR2320159A1 (en) | 1977-03-04 |
| DE2635743A1 (en) | 1977-02-17 |
| IT1067834B (en) | 1985-03-21 |
| GB1552660A (en) | 1979-09-19 |
| DE2635743C2 (en) | 1983-12-08 |
| FR2320159B1 (en) | 1980-04-04 |
| JPS5311496B2 (en) | 1978-04-21 |
| JPS5220345A (en) | 1977-02-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CA1051979A (en) | Method of arc welding with reverse side cooling for obtaining highly tough large-diameter welded steel pipes | |
| US4152568A (en) | Method of arc welding with reverse side cooling for obtaining highly tough large-diameter welded steel pipes | |
| KR101999820B1 (en) | A laser welding method for making a sheet metal semi-finished product made of hardenable steel and comprising an aluminum or aluminum-silicon based coating | |
| KR101860128B1 (en) | Method for laser welding one or more workpieces made of hardenable steel in a butt joint | |
| EP0812647B1 (en) | Weld wire | |
| JP4857015B2 (en) | Gas shielded arc welding flux cored wire and welding method | |
| JP5019781B2 (en) | MIG arc welding method using gas shielded arc welding flux cored wire | |
| US10933488B2 (en) | Method of resistance spot welding | |
| US4258242A (en) | Welding process for production of a steel pipe | |
| JP4811166B2 (en) | Manufacturing method of super high strength welded steel pipe exceeding tensile strength 800MPa | |
| CN102203303B (en) | Method for manufacturing steel plate and steel pipe for ultrahigh-strength line pipe | |
| JP4837789B2 (en) | Steel sheet for ultra-high strength line pipe and method for manufacturing steel pipe | |
| EP0899050A1 (en) | Bonding method of dual phase stainless steel | |
| KR102121188B1 (en) | Gas shield arc welding method and manufacturing method of welding structure | |
| US4817859A (en) | Method of joining nodular cast iron to steel by means of fusion welding | |
| JP2007516351A (en) | Manufacturing method of stainless steel pipe used for piping system | |
| CN114905149B (en) | Laser powder filling welding and heat treatment method for coated steel | |
| JP7047387B2 (en) | Manufacturing methods for steel sheets, butt welded members, hot pressed molded products, steel pipes, hollow hardened products, and steel plates. | |
| JPH05185236A (en) | Method of bonding to manganese steel part and other carbon steel part, and aggregate obtained by the method | |
| Okaguchi et al. | Development and mechanical properties of X120 linepipe | |
| US3660629A (en) | Method and device for welding high tensile strength steel of higher strength | |
| CN108127216A (en) | A kind of low-alloy bainite steel manual argon arc welding connects technique | |
| JPH08309428A (en) | Production of welded steel tube | |
| JP2002219576A (en) | Spot welding method for high strength steel sheet excellent in fatigue strength property of welding part | |
| SU1825682A1 (en) | Process of arc welding and surfacing |