GB1558627A - Iron-chromium-cobalt-nickel alloys - Google Patents

Description

(54) IRON-CHROMIUM-COBALT-NICKEL ALLOYS (71) We, CABOT CORPORATION, a corporation organised and existing under the laws of the State of Delaware, United States of America, of 125 High Street, Boston, State of Massachussetts, United States of America (assignee of STEVEN JAMES MIATTHEWS), do hereby declare the invention, for which we pray that a patent may be granted to us, and the merhod by which it is to be per formed, to be particularly described in and by the following statement: This invention relates to iron-chromium-cobalt nickel welding wire and weldments made therefrom and particularly to welding wire and weldments made there from in iron-chromium-cobalt-nickel alloys which have markedly improved hot crack sensitivity.

The use of iron-chromium-cobalt-nickel alloy welding wire under specification AMS-5794A has been practiced for some time in the assembly of parts subject to high stresses up to 15000F. and moderate stresses to 2000OF. Typical examples are in aircraft applications such as tailpipes and tailcones, afterburner parts, exhaust manifolds, combustion chambers, turbine blades, buckets and nozzles. Similarly, it has been used in many applications iu the metal working industries. Unfortunately, however, weldments of such alloys tend to be hot crack sensitive and hot cracking has accordingly been a significant problem to the industry.

By practice of this invention it is hoped to provide a solution to the problem of hot crack sensitivity which has characterized weldments of this particular material.

According to the present invention there is provided an iron-chromium-cobaltnickel alloy composition for producing welding wire comprising: Fe+ impurities Balance Cr 20.0 to 22.5% Co 18.5 to 21.0% Ni 19.0 to 21.0% Mo 2.5 to 3.5% W 2.0 to 3.0 /O Ta 0.75 to 1.75% C 0.1% max Mn 1.0 to 2.0% Si 1.0% max N 0.05 to 0.2% P 0.04% max S 0.03% max Nb . less than + Ta content Preferably the alloys of the invention are made from elemental raw materials as the starting materials for the alloys rather than recovered scrap (reclaim) and by finally melting either under vacuum, partial vacuum or under an inert gas atmosphere, such as argon.

The unique character of the welding wire and weldments made according to this invention can be, perhaps, best recognized by comparative examples of welding wire and weldments made therefrom using conventional AMS-5794A weld wire and weld wire according to this invention.

EXAMPLE Five heats of alloys having the compositions set out in Table I hereinbelow were melted and formed into 1/8 inch (3.2 mm) diameter welding wire.

TABLE I CHEMICAL COMPOSITION OF WELD FILLER WIRES AMS-5794A Element Spec. Heat A Heat B Heat C Heat D Heat E Fe Bal. 30.0 29.58 31.51 30.55 29.59 Cr 20.0-22.5 21.01 21.4 20.78 21.10 21.26 Co 18.5-21.0 20.11 19.1 18.70 19.70 19.26 Ni 19.0-21.0 19.86 19.90 19.35 20.80 19.48 Mo 2.5-3.5 3.15 2.78 3.19 2.72 3.03 W 2.0-3.0 2.50 2.30 2.19 2.10 2.36 Nb - 0.98 .18 .02 .02 .05 Ta - 0.03 .80 .86 1.59 1.02 Nb + Ta .75-1.25 1.01 .98 .88 1.61 1.07 C 0.1 max. 0.05 .11 .09 .02 .10 Mn 1.0-2.0 1.54 1.38 1.31 1.28 1.40 Si 1.0 max. 0.41 .50 .41 .01 0.47 N .1-.2 0.12 .18 .16 .15 .05 P .04 max. 0.016 .026 .024 .005 .005 S .03 max. 0.018 .015 .012 .015 .005 Cu - 0.06 .08 .08 NA .02 TABLE I (Continued) AMS-5794A Element Spec. Heat A Heat B Heat C Heat D Heat E Pb - 0.001 < .001 NA NA .0005 A1 - NA* .13 .30 NA NA Zr - NA .02 .02 NA NA La - NA .02 .003 NA NA N + C - 0.17 0.29 0.25 0.17 0.15 * Not analyzed.

The relative differences among the five heats set out in Table I are set out hereinbelow in Table II.

TABLE II Welding Wire Composition Contains Heat Charge A Nb Reclaim B Ta Reclaim C Ta Reclaim D Ta Raw Material E Ta Raw Material Heats B, C, D and E also contain trace amounts of Nb but less than one-half the content of Ta.

Weld evaluation tests were conducted using the Varestraint test and a butt welded bar assembly. The Varestraint test is described in the Welding Resesrch supplement to the Welding journal. November 1966. However, each of the tests as used herein are specifically described below along with the accompanying drawings in which: Figure 1 is a bar graph showing the heat affected zone and fusion zone cracks m Varestraint Test Pads; Figure 2 is a bar graph showing total length of cracks observed in Bar Butt Weld Cross Section; Figures 3A-3E are photomicrographs of weldments using weld wire from each of heats A-E; and Figure 4 illustrates the joint design used in the Butt Weld Bar Assembly.

In the Varestraint Test the following practice was followed: Two 3/8 inch (9.5 mm) thick weld pads were deposited using each of the weld wire filler materials of heats A-E. The pads were machined fiat and subjected to standard Varestraint testing using a 25-inch (63.5 cm) radius block. The lengths of hot cracks in each Varestraint test specimen were totalled to provide a quantitative measurement by which the weldability of each weld filler material was compared.

The test specimen was a weld pad 6 inches long, 2 inches wide, 3/8-inch thick with a machined surface. The weld pad was deposited on 3/8-inch thick alloy plate by manual gas tungsten arc process, 6 layers, 13 passes per layer, each pass about 130 amps, 15 volts, 2.5 ipm.

The test parameters are set out in Table HI.

TABLE III VARESTRAINT TEST PARAMETERS Electrode: W2ThO2, 1/8 inch diameter, blunt tip Arc length cold: .094 inch Argon flow: 35 cfh Current: 250 amps Voltage: 12 - 14 volts Travel speed: 4.5 inches per minute Bend release time: 2 minutes Radius of bend: 37.5 inches Approx. strain: .92% The butt welded bar assembly practice was as follows: Welding wire from Heats A, B, C, and E were used to butt weld 3-inch (7.6 cm) diameter bars simulating an actual joint associated with the fabrication of a land base gas turbine engine. Two butt welds were made from each weld filler material. The first weld was made using low heat input (i.e., low welding current and interpass temperature). The second weld was made using higher amperage and interpass temperature. Detailed welding parameters are listed in Table IV and in Figure 4.

After welding, each butt weld was dye penetrant inspected for cracking. In addition, each joint was cross sectioned, polished, etched and examined with a 10N binocular microscope for weld metal cracking. Eight 1/2-inch '1.3 cm; diameter pills were electro-discharged machined from each weldment and machined into tensile and stress rupture test specimens. The gauge length of each specimen (approximately 1 inch) was oriented parallel to the axis of the butt welded bar and was located near the outer periphery of the joint such that the entire gauge length consisted of weld material.

TABLE IV THE BAR STOCK TAKEN FROM HEAT C Technique 1 Technique 2 Power DCSP DCSP Electrode W2ThO2 W2ThO2 Electrode Dia. (in.) 3'32 3/32 Argon Flow (cfh) 25 25 Gas Cup Size No. 8 No. 8 Voltage (volts) 12-14 12-14 Current (amps) 110-130 125-145 Interpass Temp. ( F.) 200 (max.) ~600 Number of Passes to fill Joint: 67 - (Filler A) 50 - (Filler A) 41 - (Filler B) 46 - (Filler B) 52 - (Filler C) 40 - (Filler C) 52 - (Filler (D) 41 - (Filler D) 59 - (Filler E) 44 - (Filler E) The Varestraint weldability results are presented in Table V and are summarized as a bar chart in Figure 1. The butt welded bar hot cracking results are listed in Table VI and are summarized as a bar chart in Figure 2.

TABLE V SUMMARY OF VARESTRAINT RESULTS Fusion Zone Cracks HAZ Cracks Test Welding No. Wire No. Total Crack Length No Total Crack Length 1 A 7 4.8 mm 8 10.9 mm 2 A 7 6.4 mm 4 6 4 mm 3 B 4 3.3 mm 0 0 4 B 9 3.7 mm 2 1.0 mm 5 C 2 1 4 mm 6 2.3 mm 6 C 6 6.0 mm 2 0.5 mm 7 D 5 1.9 mm 4 .3 mm 8 D 4 1.9 mm U U 9 E 5 1.9 mm 0 0 10 E 6 28 mm 3 0.5 mm TABLE VI SUMMARY OF BUTT WELDED HOT CRACKING BAR (Cross Section Examination) Welding Deposit Procedure Total Crack Length Wire (see Table I V) No. of Cracks (mm) A 1 9 5.5 A 2 6 2.3 B 1 12 13.4 B 2 7 6.5 C 1 8 9.4 C 2 6 2.3 D 1 0 0 D 2 1 0.3 E 1 0 0 E 2 1 0.3 The average weld mechanical properties for weld metal deposited using filler material heats A to E are reported in Table VII. Photomicrographs showing the macrostructure of the filler materials as deposited in the bar butt joints (using the low heat input welding technique) are shown in Figures 3A-3E.

TABLE VII AVERAGE MECHANICAL PROPERTIES(1) OF ALLOY WELDMENTS Test Temp. Filler 0.2% Offset Yield Ultimate(3) Ductility OF. Wire (ksi) (ksi) (% RA) RT A 78.7 (6.3) 108.2 (5.9) 21.7 (4.9) RT B 73.1(4.2) 102.9 (7.5) 22.7 (7.4) RT C 74.1 (4.4) 95.3 (6.3) 16.2 (4.1) RT D 66.3 (4.4) 101.4 (2.2) 49.9 (3.6) RT E 71.6 (5.1) 109.0 (0.6) 46.2 (4.0) RT(2) A 67.9 101.5 10.1 RT(2) B 62.8 83.6 10.4 (2) RT (2) C 65.0 93.6 10.4 RT(2) D 58.2 106.7 30.5 RT(2) E 61.0 105.7 23.7 A A 40.0 (1.) 46.2 (1.9) 30.9 (13.3) 1500 B 34.3 (5.8) 40.2 (8.5) 43.7 (16.2) 1500 C 36.9 (1.5) 42.7 (2.0 36.5 (13.0) 1500 D 36.9 (3.9) 45.8 (3.1) 28 1 (2.8) 1500 E 37.4 (1.8) 45.3 (.1.6) 44.5 (6.0) TABLE VII (Continued) Rupture Life (hours) at 15,000 psi Stress Minus One Ductility Average Life Sigma Life %RA 1500 A 714.3 631.0 22.0 (29.1) 1500 B 501.2 302.0 44.1 (19.1) 1500 C 204.2 103.3 10.5 (1.5) 1500 D 141.3 94.6 11.2 (3. 1) 1500 -E 331.1 294.0 24.1 (11.1) Notes: (1) Average of 4 tests except where noted. Tensile data were averaged arithmetrically.

Stress rupture data were averaged logarithmically.

Standard deviation values are given in parentheses.

(2) Tested after aging at 16500F/16 hours. Only two tests per heat, hence no standard deviation is given.

(3) All specimens broke in the weld metal.

The Varestraint Weldability Testing method of weld filler metal evaluation is unique and offers some distinct advantages. In this test, the autogenous Varestraint weld bead not only melts and resolidifies the weld filler material composition under evaluation but also subjects the previously deposited filler metal to a heat-affected zone environment (i.e., rapid thermal cycle plus strain). This provides meaningful information since heat-affected-zone cracking of a previous weld bead is believed to be a large contributing factor in multi-pass weld metal fissuring.

In the present examples the Varestraint weld pad specimens made with the conventional filler material (Heat A) were found to be the most aack sensitive of all the filler materials evaluated (see Figure 1). Dramatic improvements in weldability, especially in terms of Varestraint heat-affected-zone hot cracking, were seen when tantalum bearing filler materials (Heats B, C, D and E) were used to make the initial test pad. This supports applicant's discovery that tantalum substituted for niobium improves weldability in AMS-5794A alloy systems.

None of the bar weldments exhibited weld metal cracking in the final layer as determined by dve penetrant inspection. A considerable number of cracks were observed by metallographic examination beneath the final layer of welds made from Heats A, B and C, suggesting that weld metal fissuring of these alloys in multi-pass joints may be related to heat-affected-zone cracking of previously deposited weld beads.

Weld metal cracking in raw material heats (Heats D and E) were virtually nonexistent except for one small fissure in each of the joints welded using high heat inputs.

The amount of weld metal cracking in the B and C heats weld metal joints (Heats B and C in Figure 2) was high. The reasons for this is not completely understood.

However, it may be related to the fact that Heats B and C contained carbon levels near the maximum amount (.1%) for AMS-5794A welding wire. Carbon levels near the low end of the specification range are considered preferable for maximum hot cracking resistance.

Hot crack resistant Heat E contained a high level of carbon (.1%) but also contained a low level of nitrogen (.05%). Based upon these results, it appears that resistance to weld metal fissuring in the tantalum-bearing alloys of this invention is further enhanced by maintaining a carbon plus nitrogen level of less than 0.2%.

The low level of cracking experienced in both the Varestraint tests and the butt welded bar assemblies suggest that raw material, tantalum-bearing iron chromium cobait nickel alloys (Heats D and E) are the most crack resistant alloys of the five materials studied. The reason for their superior weldability, at first glimpse, is believed due to (1) substitution of tantalum for niobium, (2) the use af elemental raw materials in the melting charge and (3) maintainance of a low level of carbon and nitrogen.

Heats D and E were processed in vacuum melting equipment. Because of the raw material usage, the cost per pound of this product would be high. However, the increased cost should be offset by improved weld performance, specifically reduction in rework due to weld metal fissuring.

The room and 1500 F. tensile properties of all five alloy weldments were found to be comparable exceut for one notable difference. The room temperature ductility of the raw material, tantalum-bearing weld filler materials (Heats D and E) were two to three times greater than Heats A, B or C, even after aging at 16500F. for 16 hours.

Highly ductile weld filler materials are an asset in welding heavy section joints since the solidified weld zone is more "forgiving" and can better tolerate residual welding stresses without cracking the base material.

The average stress rupture lives of weldments made using Heats D and E (141 hours and 331 hours, respectively) were lower than similar weldments made with conventional, niobium bearing filler wire (714 hours). However, these values are in excess of the 100-hour minimum required in a commercial specification for alloy investment castings of this type.

Dramatic improvements in hot crack sensitivity of AMS-5794A weld metal can be readily achieved inside the scope of the present material specification (AMS5794A) by (1) substituting tantalum in lieu of niobium and (2) using elemental raw materials in the melting operation.

A weld filler material of this nature not only offers better weldability but also improves room temperature ductility of the weldment in both the as-welded and aged condition.

Accordingly, the preferred analysis for a weld wire according to this invention is: Element Aim (Weight %) Fe 30.0 Cr 21.0 Co 20.0 Ni 20.0 Mo 3.0 W 2.3 Mn 1.5 Ta 1.0 N 0.15 C 0.050 max.

Si 0.500 max.

P 0.010 max.

S 0.015 max.

Cu 0.050 max.

Pb 0.001 max.

Claims (9)

Impurities Balance WHAT WE CLAIM IS:-

1. An iron-chromium-cobalt-nickel alloy composition for producing welding wire comprising: FeRimpurities Balance Cr 20.0% -22.5% Co 18.5% -21.0% Ni 19.0% -21.0% Mo

2.5% - 3.5% W 2.0% - '3.0% Ta 0.75%- 1.75% C 0.1% max.

Mn 1.0% - 2.0% Si 1.0% max.

N 0.05%-0.2% P 0.04% max.

S 0.03% max.

Nb less than i Ta content 2. An alloy composition as claimed in claim 1 which is made by melting elemental metals to form the composition.

3. An alloy composition as claimed in claim 2 wherein the elemental metals are melted in a vacuum.

4. An alloy composition as claimed in claim 2 wherein the elemental metals are melted under an inert gas atmosphere.

5. An alloy composition as claimed in any one of claims 1 to 4 wherein the carbon plus nitrogen level is held below 0.2%.

6. An alloy composition as claimed in any one of claims 1 to 5 consisting of: Fe 30.0 Cr 21.0 Co 20.0 Ni 20.0 Mo 3.0 w 2.3 Mn 1.5 Ta 1.0 N 0.15 C 0.050 max.

Si 0.500 max.

P 0.010 max.

S 0.015 max.

Cu 0.050 max.

Pb 0.001 max.

Impurities balance.

7. An alloy as claimed in claim 1 and substantially as hereinbefore described with reference to the Examples.

8. A welding wire made from an alloy as claimed in any one of claims 1 to 7.

9. A weldment between metal members to be connected made from an alloy as claimed in any one of claims 1 to 7.