EP1076604A1

EP1076604A1 - Welding processes with ferritic-austenitic stainless steel

Info

Publication number: EP1076604A1
Application number: EP99912747A
Authority: EP
Inventors: Sunniva Refsnes Collins; Thomas L. Dudley; Christine M. Schilt Deines; Peter Charles Williams
Original assignee: Swagelok Co
Current assignee: Swagelok Co
Priority date: 1998-05-08
Filing date: 1999-03-26
Publication date: 2001-02-21
Also published as: JP2002514513A; KR20010043411A; CN1109593C; WO1999058290A1; AU3105399A; IL136927A0; CN1302245A

Abstract

A welded article comprises first and second parts welded together such that the parts are joined by a weld, the first and second parts being formed from the same or different austenitic stainless steels, wherein the weld has a Cr-equivalent/Ni-equivalent ratio, R, of about 1.5 to 2.0, wherein Cr equivalent = Cr + 1.37 Mo + 1.45 Si + 2 Nb + 3 Ti, and Ni equivalent = Ni + 0.31 Mn + 22C + 14.2 N + Cu.

Description

WELDING PROCESSES WITH FERRUIC-AUSTENiπC STAINLESS STEEL

Related Applications This patent application is a continuation- in-part of co-pending patent application Serial No. 09/074,981 filed on May 8, 1998, for WELDING PROCESSES WITH FERRJΗC- AUSTENITIC STAINLESS STEEL, the entire disclosure of which is fully incorporated herein by reference.

Technical Field of the Invention The subject invention is directed toward the art of welding, and more particularly to improvements in welding low carbon stainless steel by controlling or selecting the chromium equivalent to nickel equivalent ratio.

Background of the Invention

AISI 316L is a low carbon austenitic stainless steel commonly used in high purity piping systems for the semiconductor, biotechnology/pharmaceutical and nuclear industries. 316 stainless steel is an austenitic chromium-nickel-molybdenum stainless and heat resisting steel. 316L stainless is a low carbon version of 316 stainless steel with superior resistance to intergranular corrosion following welding. European analogs to AISI 316L are DIN X2CrNiMo 17122 and DIN X2CrNiMo 18143.

A specification for AISI 316L stainless steel, plus a specification for the combination of AISI 316L and the two European analogs given above ("316L Type" stainless steel), are set forth in the following Table 1 :

Table 1

Component AISI 316L. wt % 316L Tvpe. wt %

C .030 max 0.030 max

Mn 2.00 max 2.00 max

P 0.045 max 0.045 max s 0.030 max 0.030 max

Si 1.00 max 1.00 max

Cr 16.00-18.00 16.00-18.50

Ni 10.00-14.00 10.00-15.00

Mo 2.00 - 3.00 2.00 - 3.00 N .10 max .10 max Type 316L stainless piping, including seamless tubes as well as welded and drawn tubes, is widely used in the above-noted industries. Usually, it is welded together in runs, as well as to various flow control devices, valves, fittings and so on. Accordingly, the weldability of 316L Type stainless steel is of critical importance to the above-noted industries.

However, a common problem encountered during welding of low carbon stainless steel such as 316L is the formation of weld slag and black spots. As used herein, "weld slag" and "black spots" are used interchangeably because although they differ in their appearance and location at a weld, they consist generally of the same chemical composition.

Welds are visually inspected and weld slag and black spots are cause for rejection of a weld. Weld slag can result in incomplete weld penetration due to interference with heat input to the weld pool. Weld slag can also produce corrosion sites, as well as oxygen free sites that promote microbial induced corrosion. Weld slag is unacceptable generally for high purity applications wherein welds are expected to be smooth, straight and flat or slightly beaded, and corrosion free.

In commercial practice, formation of weld slag and/or black spots is a vexing problem. Some heats of 316L Type stainless steel produce unacceptable amounts of weld slag on autogenous welding while others do not, even though their heat certifications indicate that all these steels are otherwise on specification. This makes it difficult or impossible to consistently and reliably produce autogenous welds free of weld slag and/or black spots.

One approach to controlling weld slag formation has been to reduce or eliminate those elements that contribute to the weld slag formation. Thus, the concentrations of silicon, calcium, titanium, zirconium and aluminum in the steels being welded are often restricted to below the maximum tolerable levels set forth in the following Table 2:

Table 2

Maximum Tolerable Concentrations of Slag-Forming Elements To Insure Elimination of Black Spots and Weld Slag

Component Maximum Tolerable Concentration, wt. %

Al 0.01

Ti 0.014

Si 0.1

Ca 0.02 Zr 0.05 Reducing the aluminum and titanium content of such alloys is also helpful in minimizing the adverse effects these elements have on alloy stability, hardness and pitting during electropolishing.

However, reducing or eliminating slag-forming elements often requires expensive refining procedures and/or starting materials. Accordingly, this approach is often too expensive to be practical from a commercial standpoint.

For a further discussion of the various aspects of welding and stainless steel is provided in my article STAINLESS STEEL FOR SEMICONDUCTOR APPLICATIONS, S. Collins, 39th Mechanical Working and Steel Processing Conference Proceedings, Iron and Steel Society, Volume XXXV, pages 607-619 (1998), the entire disclosure of which is fully incorporated herein by reference.

Accordingly, it is an object of the present invention to improve welding processes for stainless steel and especially for austenitic low carbon stainless steel to substantially reduce or eliminate weld slag while not adversely affecting other desirable properties of the stainless steel.

Summary of the Invention

In accordance with the present invention, we have discovered that acceptable welds can be formed when welding austenitic low carbon stainless steels reliably and consistently provided that the weld produced exhibits a Cr-equivalent/Ni-equivalent ratio, R, of about 1.5 to 2.0, wherein

Cr equivalent = Cr + 1.37 Mo + 1.45 Si + 2 Nb + 3 Ti, and Ni equivalent = Ni + 0.31 Mn + 22C + 14.2 N + Cu.

In particular, we have found that autogenous welding of austenitic low carbon stainless steels can be accomplished without formation of weld slag and black spots reliably and consistently provided that the weld is primarily austenitic but exhibits at least some ferritic character as reflected in the foregoing Cr-equivalent/Ni-equivalent ratio. Moreover, we have further found that these advantageous results can be achieved even if the weld contains significant concentrations of aluminum, titanium and other elements contributing to weld slag formation. Accordingly, it is possible in accordance with the present invention not only to produce high quality welds more reliably and consistently than possible in the past, but also to do so without the expensive manufacturing procedures undertaken in the past to eliminate slag forming elements from the alloys being welded. Accordingly, the present invention contemplates a new process for controlling the quality of a weld produced when welding weld-grade, on specification austenitic stainless steel parts together, the process comprising forming a weld having a Cr-equivalent/Ni-equivalent ratio, R, of about 1.5 to 2.0, wherein

More specifically, the present invention contemplates a new process for improving the autogenous welding of two parts made from the same on-specification, weld-grade austenitic stainless steel comprising determining the above Cr-equivalent/Ni-equivalent ratios of the steels forming the parts to be welded together and rejecting for use in the autogenous welding process all parts made from steels not having a Cr-equivalent/Ni-equivalent ratio of about 1.5 to 2.0.

In addition, the present invention also contemplates a new article of manufacture comprising a welded article comprising first and second parts welded together such that the parts are joined by a weld, the first and second parts being formed from the same or different weld-grade austenitic stainless steels, wherein the weld has the above-noted Cr-equivalent/Ni- equivalent ratio.

Still other advantages and benefits of the invention will become apparent to those skilled in the art upon a reading and understanding of the following detailed description.

Brief Description of the Drawings

The present invention may be more readily understood by reference to the following drawings, wherein:

Fig. 1 is a schematic view illustrating one aspect of the invention in which a weld ring is used to match chemistries of two bodies being welded together; and

Fig. 2 is a graph illustrating how ferrite concentration varies as a function of the Cr- eq/Ni-eq ratio in welds produced by autogenous welding of various different 316L stainless steels; and

Fig. 3 is a compilation of photomicrographs showing how weld quality varies as a function of the Cr-eq/Ni-eq ratio of the weld in a number welded articles produced in the working examples herein; and

Figs. 4, 5, 6 and 7 are graphs illustrating how corrosion resistance varies with the Cr- eq/Ni-eq ratio and ferrite content of the weld when certain austenitic stainless steels described in the working examples are autogenously welded. Detailed Description

In analyzing the nature and occurrence of weld slag, we have discovered that acceptable welds, that is welds having no slag, are formed from various different types of weld- grade austenitic stainless steels provided that the weld is primarily austenitic but exhibits some ferritic character as reflected by the Cr-eq/Ni-eq ratio R of the weld formed being about 1.5 to 2.0. In particular, we have noticed that acceptable welds tend to be slightly magnetic, indicating some small amount of retained ferrite in the weld. This has resulted in our further discovery and understanding that the formation of weld slag bears a close relationship to the solidification mode of the weld.

There are generally understood to be four distinct solidification modes for stainless steel such as type 316L. These are austenitic, austenitic-ferritic, ferritic-austenitic and ferritic. The majority of 316L chemistries will provide welds that are of the first three stated types. The austenitic weld will solidify completely to austenite and no further high temperature transformations occur. The austenitic-ferritic weld solidifies as austenite and delta ferrite is formed from the melt retained between the austenite dendrites. In a ferritic-austenitic weld, ferrite solidifies first and austenite forms between the ferrite dendrites. The austenite phase grows as the ferrite slowly transforms into austenite, resulting in a significant decrease in the volume fraction of ferrite in the final structure. At room temperature, the weld is substantially austenite, with a small amount of retained ferrite.

The competition between ferrite- promoting elements and austenite-promoting elements can be described respectively by the chromium and nickel equivalents. Although there are several commonly used chromium and nickel equivalents equations, the equations developed by Hammar and Svensson show an excellent correlation between chemical composition and solidification mode. See O. Hammar and U. Svensson, Solidification and Casting of Metals, The Metals Society, London, 1979, pp 401-404. Thus, the preferred equations for the chromium and nickel equivalents are:

Cr eq = Cr + 1.37 Mo + 1.45 Si + 2 Nb + 3 Ti Ni eq = Ni + 0.31 Mn + 22C + 14.2 N + Cu

Using these equations, the solidification mode of welds can be predicted by the chromium equivalent to nickel equivalent ratio, herein referred to as "R", where

R = Cr eq/Ni eq. In particular, it is understood that for welds having R values less than 1.5 (R<1.5), the solidification mode is austenitic or austenitic-ferritic. Similarly, for welds whose R values are greater than 2.0 (R>2.0), the solidification mode is ferritic. For welds whose R values are between 1.5 and 2.0 (1.5<R<2.0), the solidification mode is ferritic-austenitic. Moreover, it is further understood from the literature that the above R values are approximate in nature and can vary somewhat such as, for example, about ±0.03.

Although such use of the Cr eq/Ni eq ratio to predict solidification mode has been widely reported, the relationship to acceptable welds has not been recognized. What we have found is that acceptable welds having no slag or black spots appear to have a ferritic-austenitic solidification mode, or are transitional between austenitic-ferritic and ferritic-austenitic, in which there is some retained ferrite thus making the weld slightly magnetic. In particular, we have found that austenitic welds containing about 0.3 to 5 wt.%, preferably 0.5 to 3 wt.%, ferrite consistently and reliably exhibit no black spotting or slag formation even if they contain significant slag forming elements such as Ca, Si, Al, Ti and Zr.

The absence of weld slag in autogenous welds that solidify in a ferritic-austenitic mode can be explained by the crystalline structure. Ferrite is a body centered cubic (bcc) structure while austenite is a face centered cubic (fee) structure. The primary elements of weld slag are Ca, Si, Al, Ti and Zr. All of these elements have a significantly higher solubility in bcc ferrite than in fee austenite as set forth in the following Table 3 :

Table 3

Maximum Maximum

Element Solubility in ferrite, wt % Solubility in austenite, wt %

Ca 0.024 0.016

Si 10.9 1.9

Al 30 0.95

Ti 8.7 1 Zr 11.7 1

Thus, the preferred mode for reduction or elimination of weld slag in accordance with the present invention dictates that the weld produced have a Cr eq/Ni eq ratio sufficient to insure that the weld produced contains a small but suitable amount of retained ferrite. Thus, the weld should have a Cr eq/Ni eq ratio of at least about 1.5 (e.g. 1.47 or even as low as 1.45) as this insures the weld will contain at least about 0.3 wt.% or so retained ferrite at room temperature.

In prior art processes for eliminating weld slag and black spotting, expensive refining procedures were used to remove these slag-forming elements from the weld as much as possible. The present invention departs from this approach in that it relies on the chemistry of the alloy, as defined by its Cr eq/Ni eq ratio, to achieve a small but suitable amount of ferrite in the weld. This ferrite, it is believed, acts a scavenger, taking up and holding these slag-forming elements in solid solution when the weld cools to room temperature where they are unavailable for forming slag and black spots. Accordingly, rather than eliminating slag-forming elements from the system as in prior practice, the present invention merely renders these slag-forming elements benign by locking them into the crystal structure of the alloy. This avoids the expensive refining practices used in prior practices yet still provides product alloys exhibiting no black spotting or weld slag formation on a consistent and reliable basis. Additional benefits of using ferrite-austenitic solidification mode is that the presence of a small amount of ferrite is known to reduce hot cracking and micro-cracking. The solid solution of the various slag impurities also can contribute to the hardness of the material, further improving the overall strength of the weld.

On the other hand, ferrite is considerably more prone to corrosion attack than austenite. Therefore, it is also desirable in accordance with the present invention to limit the ferrite concentration in the weld produced to a suitably low value to insure that welded articles produced in accordance with the present invention exhibit suitable corrosion resistance.

In this connection, Fig. 2 illustrates the relationship between percent retained ferrite and Cr eq/Ni eq ratio in some of the welds produced in the following working examples when 316L stainless steel articles were autogenously welded together. As can be seen, ferrite concentrations approaching 7-10 wt. % were achieved as the Cr eq/Ni eq ratio, R, approached 2.0. In some applications, the corrosion resistance of a weld containing 7-10 wt. % ferrite might be acceptable. On the other hand, in those applications in which the welds come into contact with corrosive materials, the ferrite concentration should be no more than about 5 wt. %, preferably no more than about 3 wt. %. Accordingly, in a preferred embodiment of the invention, the Cr eq/Ni eq ratio, R, is limited to a maximum of about 1.67, more preferably to a maximum of about 1.55, as this dictates these lower ferrite concentrations in the welds produced.

Thus, while the advantages of the present invention can be realized when the Cr eq/Ni eq ratio of the welds obtained ranges between about 1.5 to 2.0, it is desirable that the Cr eq/Ni eq ratio be maintained at the lower end of this range, such as about 1.5 to about 1.67, or more precisely 1.45 to 1.67 or even 1.47 to 1.55. These tighter ranges for the Cr eq/Ni eq ratio ensure not only that the black spots and weld slag will be eliminated but also that corrosion resistance of the weld will be enhanced.

In autogenous welding, the weld is formed solely from the parts being welded together. See, for example, United States Patent No. 5,223,686, the disclosure of which is incorporated herein by reference, in which an orbital welder is used to join adjacent sections of pipes or tubes. Other autogenous welding techniques can be used in connection with the present invention, however, such as manual welding.

Since the weld formed in autogenous welding is derived solely from the parts to be welded together, achieving the desired Cr eq/Ni eq ratio in accordance with the present invention in autogenous welding is done by selecting the parts being welded to have necessary chemistries. Where the parts to be joined are made from the same alloy heat, this selection process is easily done by insuring that this alloy heat has the desired Cr eq/Ni eq ratio. However, where the parts to be joined are formed from different heats of the same alloy or different alloys altogether, the parts should be selected to have complementary alloys, that is alloys which when melted and combined together form a molten pool having the desired Cr eq/Ni eq ratio. And if the weld to be formed will be composed more of one part than the other, this non-equality in contribution should be taken into account in selecting the alloys of the parts to be joined, all with the purpose of controlling the Cr eq/Ni eq ratio of the weld to the desired value.

Achieving a desired Cr eq/Ni eq ratio R in a particular weld produced by autogenous welding can be accomplished in accordance with the present invention by a variety of different methods. Preferably, this is done by a selection process in which candidate parts are either selected or rejected for welding based on the Cr eq/Ni eq ratios, R, of the steels forming the parts. This selection process can be done, for example, at the steel manufacturing level by rejecting from the mill all steels which will not produce a weld having the desired Cr eq/Ni eq ratio. This can also be done at the supplier level by rejecting for acquisition all parts otherwise on specification but which will not produce a weld having the desired Cr eq/Ni eq ratio. This can also be done at the fabrication level by rejecting for autogenous welding all otherwise on specification parts which will not produce a weld not having the desired Cr eq/Ni eq ratio. In all cases, certain parts and/or steels otherwise meeting the necessary product specifications, and heretofore believed to be suitable for making acceptable autogenous welds, are rejected for failing to have the requisite metallurgy.

Another way of achieving the desired Cr eq/Ni eq ratio R in a particular weld is to control the mill process used to produce the steels of the parts to be welded. In conventional practice, the compositional ranges for certain specified elements are reported by the mill to the customer for each heat of stainless steel delivered. A customer may further require that additional residual and/or trace elements be included in the report or "certification" for a variety of reasons. It is not typical, however, for the customer to participate with the mill owner in determining how particular steels will be made. Nor is it typical for the customer to order stainless steels with particular Cr eq/Ni eq ratios or to participate with the mill owner in designing manufacturing runs specifically designed to achieve particular Cr eq/Ni eq ratios.

In accordance with the present invention, the desired Cr eq/Ni eq ratio R of a weld can also be achieved by control with the mill operator of the process used to produce the steels to be welded. Standard mill processes can be used to make these steels. Examples are argon oxygen decarburization (AOD), CLU converter process (CLU), vacuum oxygen decarburization (VOD), vacuum induction melting (VIM), vacuum arc remelting (VAR), electroslag remelting (ESR) and electron beam melting (EBM). To control the Cr eq/Ni eq ratio of alloys made using these processes, conventional techniques for controlling ingredient concentrations can also be employed. For example, Co and Cu concentrations can be controlled by dilution. Dilution is typically used when an element cannot be refined out of a melt. It determines selection of scrap for the charge, and addition of other specified alloying elements to bring the residual element into an accepted range. To control the concentrations of the other ingredients in the above-noted formula for the Cr eq/Ni eq ratio, various conventional refining techniques can be used to remove these ingredients from the melt. Examples of specific refining techniques are decarburization, deoxidation, desulfurization and dephosphorization. Regardless of what techniques are used, it is important to note that elimination of a trace element from a heat, for example Ti, does not alter the weight assigned to the other trace elements in determining the Cr eq/Ni eq ratio, as described above. The value for the removed trace element in the calculation is simply zero. In any event, an effective way of achieving a desired Cr eq/Ni eq ratio in candidate stainless steel parts to be welded together in accordance with the present invention is to control the chemistry of the alloys used to form these parts at the mill, and for this purpose conventional alloy forming and processing techniques can be used.

Incidentally, it is important to note that in order to determine the ratio R for a particular heat, several elements must be reported that are typically not reported as part of the standard heat certification from a mill. These include percent weight Nb. Ti and Cu. These additional reporting requirements will be determined by the particular elements identified in the Cr eq and Ni eq equations used to calculate the value R. In any event, a smooth, well-formed weld produced in accordance with the present invention will be free of weld slag and black spots, will not be as susceptible to corrosion, and will not present suitable sites for microbial-induced corrosion. The formed weld will also be easier to passivate. The present invention is also applicable to non-autogenous welding in which the weld is formed from an extra material such as a weld rod or electrode (hereinafter "weld piece") in addition to the parts being welded together. In non-autogenous welding, the parts being joined can be formed from the same alloy heat. More often than not, however, they are formed from different heats of the same alloy or different alloys altogether. In these situations, it is normal practice to match the chemistries of the alloys being welded together, as closely as possible, based on heat certifications obtained from the mill. In addition, it is also normal practice to select the weld piece to have a chemistry intermediate the chemistries of the parts being welded together to achieve a weld matched as closely as possible to both parts.

In accordance with the present invention, black spots and weld slag can also be eliminated in the non-autogenous welding of austenitic stainless steel parts by selecting the welds to have ferrite contents and R valves as described above. Controlling the welds to have the desired chemistry in non-autogenous welding is done in essentially the same way as described above in the case of autogenous welding. However, in the case on non-autogenous welding the composition of the weld piece must also be taken into account in determining the chemistry of the weld ultimately produced.

For example, to facilitate non-autogenous welding of austenitic stainless steel tubing materials having different, standard chemistries, weld rings can be used to match chemistries at the weld site. More preferably, a set of weld rings can be provided that each have a different but known chemistry (chemistries can be determined by standard known techniques such as spectrochemical analysis, inert gas fusion, high temperature combustion, or wet analytical chemistry techniques). In addition, the chemistries of the base materials being welded are also determined. A weld ring is positioned between the ends of the tubes being welded, with the weld ring having a chemistry selected such that the weld pool formed from the weld ring as well as portions of the tubes which also melt will have the desired Cr eq/Ni eq ratio and therefore solidify to a ferritic-austenitic structure.

For example, assume two sections of 316L stainless steel tubing are to be butt welded together using a welding process such as with an orbital welder. Further assume that spectrochemical analysis reveals that both of the tubes have a Cr eq/Ni eq ratio of 1.4, which is below the desired range of 1.45 to 2.0 and the preferred range of 1.47-1.67. An autogenous weld under such circumstances is likely to produce slag or black spots or an otherwise unacceptable weld that will either be rejected or require rework. In accordance with the invention, a weld ring is selected that has a Cr eq/Ni eq ratio of about 1.6, and this ring is positioned axially and preferably concentrically between the tube ends being welded. This

10 procedure is illustrated in the drawing. A first tube end 10 is to be welded to a second tube end 12. Both tubes have an undesired ratio of about 1.4. A concentric weld ring 14 is positioned between the tube ends 10, 12 (in the drawing the relative axial size of the ring 14 is exaggerated somewhat for clarity). The weld ring can be formed with similar dimensions as the bodies being welded (for example, in the present example, the weld ring would be formed with similar inside and outside diameters). When the weld is formed, such as by the use of an orbital welder (not shown), the weld ring material will mix with material from each of the tube ends to produce a weld pool having a Cr eq/Ni eq ratio of about 1.5. When this weld pool solidifies, it will solidify to the ferritic-austenitic mode without the formation of slag or black spots and will further be a smooth well-formed weld.

This is but one example of the use of a weld ring to match the Cr eq/Ni eq ratio. Those skilled in the art will appreciate that the chemistry selected for the weld ring will depend on different factors such as the type of welding process used, the volume of material being melted, the type of steel and so on. But the general concept of the invention, that is to select the Cr eq/Ni eq ratios of the bodies being welded such that the Cr eq/Ni eq ratio of the weld produced is about 1.45 to 2.0, preferably 1.5 to 2.0, can be implemented simply based on the chemistry selection to produce excellent welds.

In carrying out the welding process with the weld ring, the ring 14 can initially be tack welded as at 20 to either or both of the tube ends 10, 12 prior to performing the orbital welding process. Furthermore, a welding kit 30 can be provided that has a number of weld rings 14 of various and known Cr eq/Ni eq ratios for use at the welding station. A suitable container 32 can be used to store the weld rings 14. The welder can select a weld ring that most closely will match the known chemistries of the bodies being welded to produce a weld with a ferritic- austenitic solidification mode (based on the weld having a Cr eq/Ni eq ratio in the range of 1.45 to 2.0). Preferably, the weld ring is selected so that the weld formed has a ferrite content of 3 wt. % or less. Even more preferably, the two pieces being welded together will also have ferrite contents of 3 wt. % or less.

The weld ring chemistry can further be selected to match other elements such as the sulfur content, it being known that when welding heats with different sulfur contents where the spread is more than 0.01 weight % (for example, 0.001% S welded to 0.012% S), the weld arc will tend to deflect along the surface to the low sulfur heat while penetrating more deeply into the higher sulfur heat. This arc wandering can result in uneven and incomplete weld penetration. A weld ring can be selected with the appropriate sulfur content to match the two tube materials at the weld.

11 The present invention is applicable to a wide variety of different weld-grade austenitic stainless steel alloys. In this connection, it is well known that certain grades of austenitic stainless steel cannot be welded, as a practical matter. In some instances, the corrosion resistance of the welds produced are unacceptably low. In other instances, the hardness and/or strength of the welds is inadequate. Those skilled in the art are well aware of which alloys cannot be welded, as a practical matter, and which can. The present invention is directed to an improvement in welding those austenitic stainless steels which can be acceptably welded, which are referred to herein as "weld-grade" alloys.

The present invention is particularly applicable to stainless steel alloys having the following compositions:

Table 4

ALLOY COMPOSITIONS, wt. %

Component Acceptable Preferred More Preferred

C 0.10 max 0.030 max 0.030 max

Mn 9 max 2.00 max 1.00 max

P 0.05 max 0.045 max 0.045 max s 0.04 max 0.030 max 0.005-0.012

Cr 16-25 16-18.5 16-18

Ni 8-25 10-15 10-14

Mo 7.0 max 2.00 - 3.00 2.00-3.00

N 0.7 max 0.10 max 0.10 max Nb 1.00 max 0.10 max 0.05 max

Specific alloys to which the present invention is applicable are the 300 series austenitic stainless steels, such as alloy 316, 317 and 304. Moreover, the present invention finds particular applicability to low carbon stainless steels, i.e. stainless steels containing 0.03 wt.% or less carbon, such as 316L, 317L and 304L stainless steels.

The present invention finds particular applicability to the alloys described above which also contain more than insignificant amounts of slag forming elements, i.e. Al, Ti, Si, Ca and Zr. As previously indicated, prior art approaches to eliminating weld slag and black spotting have centered around keeping slag-forming elements below certain maximum tolerable concentration levels as set forth in Table 2. These approaches are very expensive, since severe refining procedures and/or expensive starting materials must be used. In accordance with the present invention, the alloys being welded can include one or more of these slag-forming elements in concentrations greater than the above maximum tolerable levels. Therefore, the use

12 of expensive raw materials and operating procedures characteristic of prior art approaches for avoiding weld slag and black spotting can be totally avoided in accordance with the present invention.

This is more fully illustrated in the following Table 5, which shows the concentration levels of slag forming elements which can be exceeded in alloys processed by the present invention without formation of weld slag or black spots. Specifically, the column headed "Prior Art Maximums" in this table shows the levels of slag forming elements regarded as maximums in the prior art for avoiding black spots and weld slag, as set forth in the above Table 2, and indicates that these concentration levels can be exceeded in the steels being welded in accordance with the present invention without forming black spots and weld slag. Similarly, the columns headed "More Contaminated" and "Highly Contaminated" indicate still higher concentrations of slag forming elements that can be tolerated in the steels being welded in accordance with the present invention without formation of black spots and weld slag.

Table 5

ACCEPTABLE CONCENTRATION LEVELS OF SLAG FORMING ELE

Component Prior Art More Contaminated Highly

Maximums Contaminated

Al >0.010 >0.020 >0.60

Ti >0.014 >0.020 >0.7

Si >0.1 >0.75 >1.5

Ca >0.02 >0.02 >0.024 Zr >0.05 >0.05 >0.15

Thus, Table 5 shows that alloys containing more than 0.1 wt.% silicon, for example, can be welded without formation of black spots or weld slag in accordance with the present invention, even though 0.1 wt.% is regarded in the prior art as the maximum tolerable concentration of this element for producing welds free of black spots and weld slag. Similarly, Table 5 also shows that more contaminated alloys can be welded without formation of black spots and weld slag even though they contain much higher levels of slag forming elements, for example, more than 0.75 wt.% or even more than 1.5 wt.% silicon. It will therefore be appreciated that the expensive refining and raw material selection procedures used in prior art processes for avoiding black spotting and weld slag formation can be totally eliminated in accordance with the present invention.

13 In accordance with preferred embodiments of the invention as indicated above, the Cr eq/Ni eq ratios of the alloys selected for welding are controlled to within fairly tight ranges, e.g. 1.45-1.55, 1.5-1.67, etc. In practice, it may be difficult to accomplish this control by adjusting the chemistries of these alloys at the mill using conventional alloy forming and processing techniques. Accordingly, the present invention, in still another embodiment, provides a new, simplified process for adjusting the chemistries of candidate austenitic stainless steels during manufacture to achieve these narrow Cr eq/Ni eq ratios.

In this connection, copper is considered an undesirable trace element for stainless steel chemistries, because there is no practical way of refining copper out of iron-containing alloys. Accordingly, the concentration of copper is minimized in manufacture of most stainless steels, with copper being present typically at a background levels of only about 0.10 wt % or less, and rarely does the copper content exceed 0.50 wt %.

In accordance with this aspect of the present invention, however, copper is intentionally added to candidate alloys to reduce the Cr eq/Ni eq ratio to within the desired range of 1.45-1.67. In particular, the copper content of candidate alloys is increased in accordance with this aspect of the present invention to amounts above the typical background levels of 0.10 wt %, more preferably to levels above 0.25 wt %, even more preferably to levels above 0.35 wt %. Indeed, in some instances, the copper content of candidate alloys can be as high as 0.50 wt % or even higher, thereby permitting the Cr eq/Ni eq ratio to be lowered very easily.

Adding copper to candidate alloy heats in accordance with this aspect of the present invention can be done in any conventional manner. For example, copper can be added to the alloy heat at the ladle metallurgy station after the heat has been otherwise fully compounded but before it is cast. Alternatively, copper can be added to the heat during alloy manufacture. For example, copper can be one of the original ingredients in the batch subjected to initial melting in the electric arc furnace, or copper can be added along with other element additions during Argon-Oxygen Decarburization (AOD) or other conventional processing in later stages of the alloy manufacturing operation. In any event, intentionally adding copper is an easy way to lower the Cr eq/Ni eq ratios of candidate low carbon austenitic stainless steels during alloy manufacture so as to bring this ratio at or near the desired range of about 1.45 -1.67. Furthermore, copper has an added benefit in that it helps reduce corrosion sites and also reduces microbial induced corrosion (MIC).

In accordance with still another aspect of the invention, the desired tighter range on the Cr eq/Ni eq ratio can be achieved using an AOD/VAR. In a vacuum arc remelting (VAR)

14 process, a cast steel electrode having the desired chemistry for the final product is drip melted into a water cooled copper mold. This remelt is performed under very low pressure conditions, typically not exceeding 0.1 Torr. The VAR process is used to remove dissolved gasses in the heat, typically oxygen and hydrogen. The VAR process also removes nitrogen, and thus presents an opportunity to further adjust the Cr eq/Ni eq ratio. The VAR process also removes manganese. Lowering the nitrogen and/or manganese content in the chemistry will increase the Cr eq/Ni eq ratio. In a typical VAR process, a melt can lose about 50% of the nitrogen and 10- 20% of the manganese in the electrode in a near total vacuum, and about 10-20% of the nitrogen and no manganese in a partial vacuum.

The amount of nitrogen that is removed during a VAR remelt thus depends in part on how low the vacuum is pulled. Since the chemistry of the heat is known prior to the VAR process, the amount of nitrogen and manganese to be removed can be controlled by controlling the vacuum pulled in the VAR system. Thus, if the Cr eq/Ni eq ratio is low (below 1.45, for example), the VAR process can be used to adjust the ratio up into the desired range.

A combination of the AOD process and the VAR process can also be used to tightly control the final Cr eq/Ni eq ratio of the heat. As noted herein above, copper can be added to the heat, such as at the ladle metallurgy station, to lower the Cr eq/Ni eq ratio. For example, copper may be added to bring the ratio down to just below 1.45, such as about 1.43. The VAR process can then be used to raise the ratio to the desired range , for example, 1.45-1.55 by the removal of nitrogen and manganese. It should also be noted that if the vacuum pulled in the system is relatively fixed, the amount of nitrogen and manganese removed during the AOD process can be accurately predicted. Accordingly, nitrogen, for example, can be added at the ladle metallurgy station during the AOD process so that after the VAR process the target equivalent ratio is achieved.

WORKING EXAMPLES

To further illustrate the present invention, the following working examples are provided:

15 Examples 1 to 7

2" seamless tubes having an outside diameter of 2.0 inches and an inside diameter of 1.87 inches were fabricated from seven different heats of 316L stainless steel. The compositions of these steels, their Cr eq Ni eq ratios, and their ferrite contents as measured by a Fischer ferritescope are set forth in the following Table 6:

Table 6

4 6 2 3 5 7 1

Cr 16.88 17.03 16.92 17.19 17.4 17.54 16.8

Mo 2.09 2.11 2.06 2.05 2.11 2.26 2.1

Si 0.41 0.41 0.34 0.36 0.5 0.49 0.4

Nb 0.015 0.016 0.014 0.025 0.01 0.012 0.01 1

Ti 0.01 0.01 0.01 0.01 0.01 0.01 0.01

Ni 12.92 13.05 12.69 12.38 12.37 12.42 10.22

Mn 1.25 1.31 1.16 1.81 1.07 1.09 1.82

C 0.022 0.024 0.024 0.014 0.018 0.022 0.012

N 0.04 0.041 0.044 0.051 0.018 0.011 0.052

Cu 0.41 0.26 0.28 0.23 0.22 0.23 0.3

Ni eq 14.77 14.83 14.48 14.20 13.57 13.63 12.09

Cr eq 20.42 20.60 20.31 20.62 21.09 21.43 20.33

Creq/Nieq 1.38 1.39 1.40 1.45 1.55 1.57 1.68 % Ferrite 0 0.15 0.27 0.82 2.07 3.04 3.91

Each tube was subdivided into sections, and two sections of each tube were autogenously welded together using a Hobart CT 150 DC autogenous welder. The welds were performed in a glove box with a shielding gas of 96% argon - 4% hydrogen, and a weld current of 47 amps and a weld speed of 6.3 in./min. The electrode was a standard 3/32 inch thoriated tungsten electrode. Thereafter, the welds formed were visually inspected for black spots and weld slag. In addition, photomicrographs were taken of each weld, these photomicrographs being set forth in Fig. 3. It was found that high quality welds free of black spots and slag were produced in each instance in which the Cr eq/Ni eq ratio of the steel being welded was 1.45 or above. On the other hand, welds produced from steels having Cr eq/Ni eq ratios below 1.45 exhibited significant black spotting and slag formation.

16 These results are visually illustrated in Fig. 3 where it can be seen that significant black spotting and slag formation occurred in the welds produced in Examples 4. 6 and 2 in which the steels had Cr eq Ni eq ratios of 1.38, 1.39 and 1.40, respectively, while essentially no black spotting or slag formation occurred in the welds produced in Examples 3, 5, 7 and 1 in which the steels had Cr eq Ni eq ratios of 1.45 or above. This shows the importance of maintaining the Cr eq/Ni eq ratio of a weld at or above about 1.45 in order to inhibit or prevent formation of black spots and weld slag when austenitic stainless steel parts are welded together in accordance with the present invention. Examples 8 to 28

Stainless steel tubes having OD's ranging from ! to 2.0 in. in diameter were produced from 21 additional heats of 316L stainless steel. The compositions of these steels are given in the following Table 7:

Table 7

Cr Mo Si Nb jj Ni

8 17.4400 2.6100 0.5100 0.0012 0.0006 14.1400

9 17.3700 2.6300 0.5300 0.0051 0.0005 14.1000

10 17.4400 2.6300 0.5100 0.0007 0.0009 14.0400

11 17.3700 2.6100 0.6400 0.0015 0.0007 14.0100

12 16.6800 2.1400 0.3100 0.0060 0.0008 12.2000

13 16.4600 2.3400 0.5600 0.0014 0.0007 12.6100

14 16.2900 2.3400 0.5200 0.0028 0.0005 12.4200

15 16.9700 2.3300 0.5500 0.0024 0.0005 12.4800

16 17.2000 2.2900 0.0100 0.0005 0.0013 13.3100

17 17.0500 2.3400 0.4900 0.0026 0.0005 12.6700

18 16.6200 2.3400 0.5500 0.0066 0.0011 12.3400

19 17.9300 2.2500 0.2000 0.0530 0.0100 13.2800

20 16.5800 2.3200 0.7400 0.0140 0.0018 13.0700

21 17.5300 2.3900 0.5300 0.0170 0.0005 12.4600

22 17.4100 2.2000 0.4700 0.0061 0.0005 12.3300

23 16.8500 2.1300 0.3900 0.0010 0.0036 10.9300

24 17.4500 2.1700 0.5400 0.0016 0.0006 12.5100

25 16.8400 2.1300 0.3200 0.0160 0.0005 10.1500

26 17.5400 2.0800 0.4400 0.0013 0.0005 12.1000

27 17.4600 2.1100 0.5100 0.0018 0.0009 12.1600 28 16.5300 2.6300 0.3800 0.0061 0.0005 10.1600

17 Mn C Cu P S

8 1.6600 0.0320 0.0280 0.3500 0.0190 0.0070

9 1.6900 0.0260 0.0380 0.2800 0.0230 0.0060

10 1.6700 0.0270 0.0350 0.2800 0.0220 0.0070

11 1.6500 0.0210 0.0320 0.2400 0.0250 0.0070

12 1.1400 0.0200 0.0210 0.3000 0.0300 0.0070

13 1.3500 0.0300 0.0078 0.0500 0.0120 0.0080

14 1.3600 0.0280 0.0078 0.0500 0.0110 0.0110

15 1.3900 0.0430 0.0100 0.0500 0.0120 0.0070

16 0.0200 0.0080 0.0030 0.0100 0.0040 0.0030

17 1.3200 0.0320 0.0063 0.0500 0.0100 0.0070

18 1.3700 0.0310 0.0092 0.0500 0.0160 0.0080

19 0.2600 0.0210 0.0069 0.0500 0.0190 0.0010

20 0.1500 0.0140 0.0080 0.0500 0.0170 0.0010

21 1.1100 0.0290 0.0230 0.2600 0.0170 0.0060

22 1.0500 0.0200 0.0210 0.2200 0.0210 0.0070

23 1.5400 0.0240 0.0660 0.2200 0.0170 0.0070

24 0.2600 0.0220 0.0098 0.0500 0.0150 0.0080

25 1.4100 0.0140 0.0750 0.3000 0.0250 0.0160

26 0.2900 0.0150 0.0100 0.0500 0.0220 0.0110

27 0.2500 0.0140 0.0078 0.0500 0.0250 0.0090 28 1.8200 0.0220 0.0540 0.0250 0.0120 0.0110

Each of these stainless steel tubes was subdivided into sections. Two sections of each tube were then autogenously welded together using an orbital welding system. The welds were performed in a glove box with a shielding gas of 100% argon. The weld current ranged from 20 to 47 amps, and the weld speed of ranged from 0.5 to 8.3 inches/minute. The electrode was a standard 3/32 inch thoriated tungsten electrode. For each heat, a portion of an unwelded section of tube plus the weld were subjected to corrosion resistance testing. In one test, the pitting potential of the tested metal was determined by ASTM G-61. In this test method, an external power supply is used to gradually raise the electrical potential of the material being tested in a given solution, while electrical current is measured, until pitting corrosion occurs. The potential at which the current rapidly increases due to pitting is defined as the pitting potential. A higher pitting potential signifies a higher resistance to pitting corrosion.

In another test, the critical pitting temperature was determined by ASTM G-150. In this test method, the temperature at which current density increase rapidly beyond a set limit at a set electrical potential is determined. An NaCl solution is used, and the electrical potential is held constant in the passive region. Starting at a temperature of 0 °C, the temperature is raised slowly at a rate of 1 °C per minute until pitting occurs.

The Cr eq/Ni eq ratios of the various steels used in these experiments, the ferrite content of these steels and the results obtained are set forth in the following Table 8. In

18 addition, the relationship between Cr eq/Ni eq ratio and ferrite content in these steels is graphically illustrated in Fig. 2.

Table 8

Cr eq./ AR Welded AR Welded % Ferrite

Ni eq. Cr eq. Ni eq. CPT avε. CPT ave. Ep av EP av max.

8 16.1 1 21.78 1.35 23.1 20.6 689 439 0.21

9 16.02 21.78 1.36 25.4 23.1 1101 601 0.23

10 15.93 21.81 1.37 25.7 20.1 1108 637 0.29

11 15.68 21.91 1.40 23.1 18.2 641 523 0.3

12 13.59 20.09 1.48 13.0 10.2 341 330 OJ

13 13.85 20.51 1.48 15.3 10.6 400 252 1.21

14 13.62 20.28 1.49 15.8 13.1 663 326 2.23

15 14.05 20.99 1.49 20.0 13.9 520 417 1.36

16 13.54 20.36 1.50 42.3 22.3 963 753 1.62

17 13.92 21.00 1.51 29.5 12.4 749 356 1.81

18 13.63 20.67 1.52 22.5 12.3 507 244 2.38

19 13.97 21.45 1.54 34.6 23.9 1143 882 2.39

20 13.59 20.90 1.54 21.8 16.4 487 453 1.73

21 14.03 21.63 1.54 33.2 22.6 1140 522 2.24

22 13.61 21.14 1.55 23.9 15.9 586 418 3.93

23 13.09 20.37 1.56 17.1 13.2 394 351 3.23

24 13.26 21.24 1.60 20.8 20.4 669 542 3.87

25 12.26 20.27 1.65 18.7 4.1 593 235 4.53

26 12.71 21.05 1.66 16.5 15.5 656 507 5.15

27 12.71 21.12 1.66 19.6 16.3 681 574 3.58 28 12.00 20.72 1.73 20.4 7.0 661 257 6.4

AR = As received

From the above, it can be seen that the corrosion resistance of all welds, as measured by both pitting potential and critical pitting temperature, was good. However, those welds whose Cr eq/Ni eq ratios were 1.55 or below and which also had ferrite contents of about 3 wt. % or less, had even better corrosion resistance as measured by their critical pitting temperatures. This is more clearly shown in Figs. 4, 5, 6 and 7 which graphically illustrate the critical pitting temperatures of the welds produced as a function of Cr eq/Ni eq ratio and also of ferrite content. As can be seen from Figs. 4 and 5, corrosion resistance as measured by critical pitting temperature decreases only slightly as the Cr eq/Ni eq ratio increases from below 1.45 to about 1.55. However, as the Cr eq/Ni eq ratio increases above about 1.55, the corrosion resistance of the welds decreases much more rapidly. Similar results are shown in Figs. 6 and 7, which show that corrosion resistance begins a much more rapid decrease as ferrite content of the weld exceeds about 3 wt. %. These results shows that welds having Cr eq/Ni eq ratios between about 1.45 and 1.55 are characterized not only as being free of black spots and weld

19 slag but also as having superior corrosion resistance, especially if these welds also have a ferrite content of 3 wt.% or less.

While the invention has been shown and described with respect to specific embodiments thereof, this is for the purpose of illustration rather than limitation, and other variations and modifications of the specific embodiments herein shown and described will be apparent to those skilled in the art within the intended spirit and scope of the invention as set forth in the appended claims.

20

Claims

WE CLAIM:

1. A welded article comprising first and second parts welded together such that the parts are joined by a weld, the first and second parts being formed from the same or different austenitic stainless steels, wherein the weld has a Cr-equivalent/Ni-equivalent ratio, R, of about 1.5 to 2.0, wherein

2. The welded article of claim 1, wherein one or both of the first and second parts contain at least one of the following elements in an amount greater than the following minimums

Component Minimum Concentration, wt. %

Al 0.010

Ti 0.010

Si 0.75

Ca 0.02 Zr. 0.05

3. The welded article of claim 1, wherein the first and second parts each have a Cr-equivalent/Ni-equivalent ratio of about 1.5 to 2.0.

21

4. The welded article of claim 1, wherein the first part and the second part are formed from the same or different alloys having the following composition:

Component Concentration. wt.%

C 0.10 max

Mn 9 max

P 0.05 max

S 0.04 max

Cr 16-25

Ni 8-25

Mo 7 max

N 0.7 max

5. The welded article of claim 1 , wherein the first and second parts are formed from a 300 series stainless steel.

6. The welded article of claim 1 , wherein the first and second parts are formed from 316L stainless steel.

7. The welded article of claim 1, wherein the weld is formed by autogenous welding.

8. The welded article of claim 1, wherein the article is a pipe or tube system for supplying high purity fluid.

9. The welded article of claim 1, wherein the first part, the second part or both are tubes.

10 The welded article of claim 1, wherein the Cr-equivalent/Ni-equivalent ratio is about 1.5 to 1.67,

11. The welded article of claim 1 , wherein the Cr-equivalent/Ni-equivalent ratio is about 1.45 to 1.55.

22

12. The welded article of claim 1, wherein the weld contains 0.3 to 5 wt.% ferrite.

13. A process for controlling the quality of a weld produced when welding on specification austenitic stainless steel parts together, comprising selecting the weld formed to weld the parts together to have a Cr-equivalent/Ni-equivalent ratio, R, of about 1.5 to 2.0, wherein

14. The process of claim 13, wherein the weld is formed by melting together portions of the first part and the second part as well as a separate weld piece different from the first and second parts to form a molten pool which solidifies to form the weld, and wherein the first part, the second part and the weld piece are each selected so that the weld will have a Cr-equivalent/Ni-equivalent ratio of about 1.5 to 2.0.

15. The process of claim 13, wherein the weld is formed by autogenous welding.

16. The process of claim 13, wherein the first part and the second part each has a Cr-equivalent/Ni-equivalent ratio of about 1.5 to 2.0.

23

17. The process of claim 16, wherein the first part and the second part are formed from the same or different alloys having the following composition:

Component Concentration. wt.%

C 0.10 max

Mn 9 max

P 0.05 max

S 0.04 max

Cr 16-25

Ni 8-25

Mo 7 max

N OJ max

18. The process of claim 13, wherein the first part and the second part are made from different heats of 316L stainless steel.

19. The process of claim 13, wherein the parts being welded are tubes.

20. The process of claim 13, wherein the weld contains 0.5 to 3.0 wt.% ferrite.

21. The process of claim 13, wherein the Cr-equivalent/Ni-equivalent ratio is about 1.45 to 1.55.

22. The process of claim 13, wherein the Cr-equivalent Ni-equivalent ratio is about 1.5 to 1.67.

23. A process for improving the autogenous welding of two parts made from the same on-specification austenitic stainless steel comprising, determining the Cr-equivalent/Ni-equivalent ratios, R, of the steels forming the parts to be welded together, wherein

Cr equivalent = Cr + 1.37 Mo + 1.45 Si + 2 Nb + 3 Ti, and Ni equivalent = Ni + 0.31 Mn + 22C + 14.2 N + Cu, and rejecting for use in the autogenous welding process all parts made from steels not having a Cr-equivalent Ni-equivalent ratio of about 1.5 to 2.0.

24

24. The process of claim 23, wherein parts not having a Cr-equivalent/Ni- equivalent ratio of about 1.5 to 2.0 are rejected by acquiring from the supplier of such parts only those parts made from steels having a Cr-equivalent Ni-equivalent ratio of about 1.5 to

2.

25. The process of claim 16, wherein parts not made from steels having a Cr- equivalent/Ni-equivalent ratio of about 1.5 to 2 are rejected by examining all on- specification candidate parts for the autogenous welding process and rejecting those on specification parts not made from steels having a Cr-equivalent'Ni-equivalent ratio of about 1.5 to 2.

26. Piping for use in transporting high purity fluids, wherein the piping is formed from an austenitic stainless steel having a Cr-equivalent/Ni-equivalent ratio, R, of about 1.5 to 2.0, wherein

27. The piping of claim 26, wherein the stainless steel is 316L stainless steel.

28. The piping of claim 27, wherein the piping is composed of a seamless tube.

29. The piping of claim 27, wherein the piping is composed of a welded and drawn tube.

30. The piping of claim 26, wherein the stainless steel contains 0.5 to 3.0 wt.% ferrite.

31. The piping of claim 26, wherein the Cr-equivalent/Ni-equivalent ratio is about 1.45 to 1.55.

32. The process of claim 13, wherein the Cr-equivalent/Ni-equivalent ratio is about 1.5 to 1.67.

25

33. A method for improving welding of an austenitic stainless steel, comprising adding copper to the stainless steel during manufacture in a quantity to produce a heat having a Cr-equivalent to Ni-equivalent ratio (R) in the range of about 1.45 to 1.55.

34. The method of claim 33, wherein the step of adding copper is performed at the ladle metallurgy station.

35. The method of claim 33, wherein copper is added to the stainless steel during manufacture to lower the Cr-equivalent to Ni-equivalent ratio (R) of the steel and thereafter the steel is subjected to vacuum arc remelting to remove nitrogen and manganese from the steel to thereby increase the Cr-equivalent to Ni-equivalent ratio (R) of the steel to 1.45 to

1.55.

36. A method for improving autogenous welds in austenitic stainless steel, comprising: a) determining respective chemistry of each of a number of austenitic stainless steel heats; b) determining the Cr-equivalent to Ni-equivalent ratio R for each of said heats; c) selecting for autogenous welding at least one heat having a ratio R in the range of about 1.5 to 2.0.

37. A method for improving autogenous welds in austenitic stainless steel, comprising: a) specifying a respective chemistry for a number of austenitic stainless steel alloy heats; b) selecting from said number of heats at least one heat having a Cr-equivalent to Ni-equivalent ratio R in the range of about 1.5 to 2.0; and c) producing an autogenous weld in a body made from the selected heat.

26

38. A method for improving autogenous welds in austenitic stainless steel, comprising: a) specifying respective chemistry of each of a number of austenitic stainless steel heats; b) specifying the Cr-equivalent to Ni-equivalent ratio R for each of said heats; and c) selecting for autogenous welding at least one heat with a ratio R such that an autogenous weld exhibits a ferritic-austenitic solidification mode.

39. A method for improving a welding process for austenitic stainless steel, comprising the steps of: a) determining a Cr-equivalent to Ni-equivalent ratio for a first body being welded; b) determining a Cr-equivalent to Ni-equivalent ratio for a second body being welded; c) selecting a weld ring and positioning the weld ring at the weld site between the first and second bodies, with the weld ring being selected to have a Cr-equivalent to Ni- equivalent ratio that when welded with the first and second bodies produces a weld having a ferritic-austenitic solidification mode.

40. A kit for use in a welding operation for welding two bodies of austenitic stainless steel together, the kit comprising: a) a plurality of weld elements, each weld element being of a suitable size and shape to be positioned at a weld site between the two bodies; b) each weld element having a known Cr eq/Ni eq ratio; c) the kit containing a plurality of said weld elements having different said ratios whereby a welder can select a weld element that when welded to the two bodies forms a weld having a ferritic-austenitic solidification mode.

27

41. A process for controlling the quality of a weld produced when welding on specification austenitic stainless steel parts together, comprising forming the weld to have sufficient ferrite content so that the weld, as a whole, when cooled to room temperature can hold in solid solution at least one of the following slag forming elements in an amount greater than the following minimum concentration levels : Component Minimum Concentration, wt. %

Al 0.010

Ti 0.010

Si 0.75

Ca 0.02

Zr. 0.05 whereby formation of black spots or weld slag from such slag forming elements is substantially prevented, the ferrite content of the weld also being no more than about 5 wt. %.

28