US2191471A

US2191471A - Welding method

Info

Publication number: US2191471A
Application number: US133357A
Authority: US
Inventors: Robert K Hopkins
Original assignee: MW Kellogg Co
Current assignee: MW Kellogg Co
Priority date: 1936-09-16
Filing date: 1937-03-27
Publication date: 1940-02-27
Anticipated expiration: 1957-02-27

Description

R. K. HOPKINS WELDING METHOD Feba 27, 19

Original Filed Sept. 16, 1936 2 Sheets-Sheet l INVENTOR ROBERT K. HOPKINS ATTORNEY Feb. 27, 1. R. K. HOPKINS 293919471 WELDING METHOD Original Filed Sept. '16, 1936 2 Sheets-Sheet 2 lNVENTOR ROBERT K. HOPKINS Mi 15m TTERNEY 1 Patented Feb. 27, 1940 WELDING METHOD Robert K. Hopkins, New York, N. Y assignor to M. W. Kellogg 00., New York, N. Y., a cor-,

notation of Delaware Original application September 16, 1936, Serial No. 101,103, now Patent No. 2,151,914, dated March 28, 1939.

5 Claims.

This invention relates in general to electric fusion of metals and in particular to the electric fusion of metals for welding, veneering, and similar purposes.

This application is a divisional application of my application Serial No. 101,103, filed Septem ber 16, 1936 (now Patent No. 2,151,914, granted March 28, 1939), which in turn is a continuation in part of my application filed February 18, 1936, Serial No. 64,496.

In arc welding, as at present carried out, and in the veneering, or coating, of metallic surfaces by electric arc deposition of metal as disclosed in my application Serial No. 64,496, it is generally necessary to employ as an electrode a body having a special composition so that the weld metal, or the veneering'metal, may have the required composition. The electrode is made of special composition primarily in order to obtain a weld metal or a veneering metal of desired composition while compensating for losses of the constituents of the electrode in the arc and the dilution of the metal deposited from the electrode due to its penetration into the metal being welded or veneered.

In the present commercial practice, electrodes, whether bare or covered, generally include all of the constituents that they are to supply to the final deposited metal in the core or metallic portion of the electrodes even though it has long been proposed, in patents and in the literature, to include some of these constituents in the coatings of the electrodes. The proposals have not been commercially successful because of the difficulties involved in obtaining from the proposed electrodes-a deposited metal in which the constituents included in the covering are found in proper proportions throughout.

Electrode metal is usually much more costly than other metal of generally similar analysis. The high cost of electrode metal is due in part to the small quantities that are made up at one time and the difiiculties encountered in compounding some of the special analyses. The cost of electrode metal is always many times more than the total cost of its constituents in their readily available commercial forms. Furthermore, while electrode metal of a wide variety of analyses may be obtained all desirable analyses are not available. This is particularly true of chromium alloys of high chromium content and of chromium alloys of low carbon content.

I have found that it is not necessary in order to obtain the results of the prior art to incorporate all of the constituents of the finally deposited Divided and this March 2'7, 1937, Serial No. 133,357

application metal into a metallic electrode, but that all of the results of the prior art and results not obtainable by the prior art, may be obtained in a simple and.

feed to the gap between the end of the electrode w and the work, or in the vicinity of said gap, in the form of granules, pellets, wire, and the like, preferably through the electrode when it is a hollow body, the remainder of the constituents required to produce the desired deposited metal. electrode and the granules, pellets, wire and the like, are fed to the gap between the end of the electrode and the work, at such rates that at substantially each instant the proper proportion of constituents is deposited to produce the desired deposited metal, Thus, with any set of constituents it is possible simply by varying themelting rate and/or the rate of travel of the electrode and/or the rate or rates of feed of the granules, pellets, wire, etc., to obtain deposited metals of widely diflerent analysis.

The constituents of the desired deposited metal are supplied to the gap between the end of the electrode and the work, in their most economical forms. Thus, for instance, when the desired deposited metal is an iron alloy the electrode is made of mild steel or other equally cheap and readily available material. If the desired deposited metal includes chromium, the chromium con- The stituent may be supplied as granular ferrochrome or similar chromium compound or alloy; manganese may be supplied as term-manganese; nickel as a metallic nickel wire, ribbon or pellets, etc.

It is an object of this invention to provide a simple method for depositing, under the influence 'of the flow of electric current through a gap between the end of an electrode and the work,

metals, of diiierent composition, or metal of substantially constant composition but under dillerent conditions of deposition, from a set of constituents, whereby by varying the feed of any or all of the constituents, the required proportions of the constituents to produce the desired deposited metal is substantially constantly deposited.

It is also an object of this invention to provide a simple method for depositing metal under the influence of the flow of electric current through a gap between the end of an electrode and the ill work, in which the constituents are fed to the gap in their most economical forms at controlled rates to substantially constantly deposit the required proportions of the constituents to produce the desired metal,

It is a further object of this. invention to provide a simple method for depositing metal under the influence of the flow of electric current through a gap between the end of an electrode and the work, in which a hollow electrode of comparatively cheap and readily available material that includes a constituent, or constituents, of the desired deposited metal is used and the remainder of the contituents required to produce the desired deposited metal are supplied to the gap separately, but through or partly through the hollow electrode at such rate or rates that there is substantially constantly deposited the desired metal; the said remainder of the constituents are also preferably materials in their most economical forms.

The further objects and advantages of the invention will be better appreciated from a consideration of the following description of preferred modes and apparatus for carrying it out in practice together with the accompanying drawings, in which,

Fig. 1 is a part sectional front view of one form of an apparatus capable of carrying out the novel method of my invention,

Fig. 2 is a fragmentary view of the apparatus of Fig. 1, illustrating the manner of adding electrode sections during the metal depositing operation,

Fig. 3 is a view similar to'Fig. 1, of a modified form of apparatus, illustrating the use of a plurality of electrodes, and,

Fig. 4 is a view similar to Fig. 3, illustrating the use of a dead electrode.

The apparatus disclosed, as shown particularly in Fig. 1, includes a welding head Id of the usual construction and arrangement mounted on a suitable support, not shown, so that it can be moved through a desired path relative to work H. Work M may be mounted on an immovable support, also not shown, or it may be mounted on a support capable of movement in a predetermined path relative to welding head l0. Welding head it includes a current supply means, such as a welding current generator or a connection to a source of welding current, devices and arrangements for controlling the voltage and the amperage of the welding current, and an electrode feed motor that drives electrode feed wheel 52, together with the usual devices and arrangements for controlling the drive of the electrode ieed motor. It is to be understood that welding head it includes all of the usual arrangements and devices which are required for satisfactory operation.

Electrode feed wheel I2 is backed by an idler wheel l3 which serves to maintain electrode M in contact with feed Wheel l2. One side of the current generator, or current supply means, is connected through cable l5 to the work H, the other side being connected through cable N to a contact device i1 that passes the electric current to welding electrode I4.

While as explained hereinafter, other than hollow electrodes may be used, the apparatus of Fig. l is particularly adapted to the use of hollow electrodes. Electrode M, as shown, is made up of a plurality of sections of convenient length which are easily connected together during the operation to form an electrode I of indefinite length. continue the operation in order to replace consumed electrodes. Each length of electrode I4 is internally threaded at each end so that it may be connected to the last section in the apparatus by screwing onto threaded nipple l8. The sections of electrode M as shown, are of circular cross-section but any preferred cross-section such as, square, eliptical, and the like may be used as preferred. Also, while the sections of electrode I4 are shown as seamless, it is to be understood, that they need not have seamless walls but can be made of plate or sheet of proper thickness, shaped as desired, with the edges joined, or not as preferred. Electrode I4 is preferably bare but, if desired it may be coated externally or internally. When electrode I4 is externally coated, it is preferably provided with bare contact areas spaced along its length, and a contact device H is provided that is arranged to pass the welding current through the bare contact areas.

Above welding head |8 and in line with the path of electrode I4, is supported hopper l9, from which material is passed through the hollow electrode M to the gap between electrode l4 and work Hopper |9 includes a cylindrical body 2|! and a conical bottom 2| that is provided with an orifice 22. Orifice 22 may be used to meter the material on its way to electrode H. In such case, orifice 22 is preferably adjustable to provide for variation in the rate of feed of the material. The material, which may be made up or one or more component materials, is passed to hopper l9 by means of one or-more conveyor belts 23.

Conveyor belts 23 may be used as the metering means instead of or in conjunction with orifice 22. In such case, while conveyor belts 23 may be of any usual construction, they should be provided with means for moving them at a range of predetermined rates and also with means, such as knives 24, for maintaining a definite depth of material upon them. When the material is metered by belts 23, orifice 22 may be dispensed with, or made of such size that it will not impede the fiow of material out of hopper l9. To prevent large-sized particles of the material from stopping the flow through hopper I9, a screen 25 of suitable mesh is provided adjacent the top of hopper i9.

In order to shut off the flow of material out of hopper I9, as for instance, at the end of the operation or during an interruption thereof, a valve 26 is provided which is adapted to seat on orifice 22; Valve 26 is mounted at one end of an articulatable bar 21. 'The other end of bar 21 is connected to the armature of an electro-magnet 28 and to a spring 29. The arrangement is such, that the pull of spring 29 tends to seat valve 26 and electro-magnet 28, when its circuit is closed, unseats valve 26. The circuit of the electro-magnet 28 is opened and closed by manipulation of a button switch 30 located on the front of welding head ID. The arrangement may also be such that the circuit of magnet 28 is automatically opened and closed with the opening and closing of the welding circuit.

The material passes from orifice 22 into a cylindrical communicating member '3|. Member 3| has its top united to the conical portion 2| of hopper I9 and surrounds the lower end thereof. Between the ends of member 3| is a diaphragm plate 32 that is provided with a central orifice 33. Windows or openings are provided in the walls of member 3| above plate 32 In this manner, there is no need to. dis.

Eli)

terial.

to prevent changes in pressure in the space between the top of member 3| and plate 32 which pressure changes would efiect the flow of the material. The top of plate 32 is provided with a conical recess to facilitate the passage of material to and through orifice 33. Orifice 33 is larger than orifice 22 and is, of course, also larger than an orifice equivalent to belts 23, when belts 23 are used as the metering means, so that it will have no efiect on the metering oi the ma- Orifice 33 may be closed to shut oil the flow of material to the lower end of member 8| by a valve 34 provided for this purpose.

Valve 35 is mounted at one end of an articulatable bar 35. The other end of bar 35 is connected to a spring 36 and to the armature of an electro-magnet 37. Spring 36 exerts a pull tending to remove valve 33 from contact with orifice 33 and electro-magnet 37, when its circuit is closed, exerts a pull that overcomes the pull of spring 36 and seats valve 36 on orifice 33. The circuit of electro-magnet 3'! includes a switch 38 that is mounted on a support member 39 adjacent the path of the electrode at a predetermined distance from feed wheel l2. Switch 38 may be of any preferred construction but should be such that it includes a spring pressed member that is adapted to bear against electrode I l to open and maintain the circuit open when the end of electrode I4 is above it and to close the circuit and maintain it closed when electrode Hal is out of contact with said spring pressed member. A support member 30 is fixed a short distance above support member 39 and has mounted on it a switch 4| similar to switch 38. Switch 4! opens and closes the circuit of an alarm means such as hell 32 that is intended to warn the operator of the machine, of the approaching necessity of adding a section to electrode l3.

To the bottom of cylindrical communication member 3| is hinged a funnel-like member 43 Whose spout 416 is adapted to extend into the upper end of electrode [3 to pass the material from communicating member 3! to the hollow center of electrode I 6. Spout M is made of such a length that it will extend to a point approximately level with or preferably a short distance below the spring pressed member of switch 38.

During the operation of the apparatus, whenever the end of a section of electrode Hi passes switch 3!, bell 32 will sound to warn the operator and call his attention to the fact that a new section must be added to electrode 13.

When electrode I l passes switch 38, electromagnet 31 will pull valve 33 into orifice 33 and thus shut ofi the fiow of material to funnel member 63. When the end of electrode M passes the end of spout 63, the operator will grasp spout A3 and swing it out of position as shown in Fig. 2 and pass a new section of electrode Id over spout M. Spout 43 is then brought back to the normal position and the new section connected to the last section of electrode I 3 by screwing the sections together on the threaded nipple Hi. When the new section is swung back to the normal position and into contact with switches 38 and GI, the circuits controlled by switches 38 and M will again be opened, bell 52 will stop sounding and valve 34 will be pulled off orifice 33 to reestablish the flow of the material. It is to be noted that orifice 33 is large enough so that it will quickly allow the material accumulated during the shut-01f period of valve 34 to pass the funnel member 43. Since the operation just described is an extremely simple one, it can be performed so quickly that there is in fact no substantial interruption in the flow of the material to the bottom of electrode l4. And thus the overall quantity of the material will be supplied to the deposited metal thereby maintaining the analysis uniform. I

The apparatus shown in Fig. 3 includes the elements of the apparatus of Fig. 1, above described, and also electrodes 45 and t6, together with their feed means and current supply means. It is to be understood that while two

electrodes

45 and 46 are mentioned only one or more than two may be used. Electrode G5 is driven toward work II by electrode feed wheel 41 that is backed up by idler wheel 38. The motor for driving feed wheel 3?, as well as the control arrangements therefore, not shown, are located in back of the panel of welding head It. The current supply means for electrode 85, not shown, which may be a welding current generator or a connection to an outside source of welding current, together with the usual control arrangements, are located in back of the panel of welding head Ill. A cable 49 connects the current supply means to contact means 50 that is adapted to pass the welding current to electrode 35. The other side of the current supply means for electrode 35 may be connected to the work by a separate cable, not shown, or it may be connected by cable l5. Electrode 6B is driven towards the work ll by electrode feed wheel 5i that is backed up by idler wheel 52. The motor for driving feed wheel 5! as well as the control arrangements therefor, not shown, are located in back of the panel of welding head It. The current supply means for electrode 36, not shown, which may be a welding current generator or a connection to an outside source of welding current, together with the usual control arrangements are also located in back of the panel of welding head ID. A cable 53 connects the current supply means to contact means 53 that is adapted to pass welding current to electrode 56. The other side of the current supply means for electrode d5 may be connected to the work by a separate cable, not shown, or it may be connectedby cable l5,

As above stated, electrodes ll, 35, and 66 may be separately supplied with welding current or they may all be supplied from a single source. It is also within the scope of my invention to feed all of these electrodes from a single feed motor, by connecting all three of feed wheels i2, 61, and 5| through a suitable train of shafts and gears to the drive shaft of a single feed motor. The single feed motor may be controlled in accordance to the average characteristics of all three arcs or any one of the arcs may be used to efiect the control. Also, the current and voltage control may also be efiected in accordance with the average characteristics of all three arcs or any one of them.

Welding electrodes

14, 45, and 86 may be arranged on a line as shown or they may be arranged in any preferred manner, however, it is important they be so closely spaced that all three of them deposit their metal in a single pool. When so spaced the amount of overroll, or metal overlying the metal of workpiece II and not completely welded thereto particularly at the edges of the deposited metal is reduced. By properly adjusting the spacing between the electrodes and by choosing the proper conditions of the deposition at each of the electrodes, as indicated in Fig. 3 of the drawings the full area of the workpiece covered by the molten pool is fused to a substantially uniform depth. Thus, overroll is completely eliminated and a coating or veneer is produced that is of relatively uniform thickness and is united throughout to the metal of the workpiece.

Electrodes

45 and 46 may be of any preferred cross-section, they may be solid or hollow and they may be coated or bare. Electrodes a and 46 may be of the same analysis or of different analysis as required to produce the desired results.

The apparatus shown in Fig. 4 includes the elements of the apparatus in Fig. 1, above described, and also, provides means forfeeding, what may best be called a dead electrode 58, to work ll. Dead electrode 55 is fed to work II by a feed wheel 57 that is backed up by an idler wheel 58. Dead electrode 56 is driven by a motor, not shown,.located in back of the panel of welding head Ill". The motor just mentioned may be controlled in any preferred manner to feed dead electrode 56 to work I I at any desired rate. Dead electrode 56 may be solid or hollow and of any preferred cross-section and may be located in any position relative to electrode I, however, it

is important that it be located sufllciently close' to electrode ll that it will be fused by the heat generated by the flow of welding current through the gap between electrode l4 and work H. Dead electrode 56 may be of any preferred analysis and while only one is shown any number desired may be used.

The feed motor for dead electrode 56 may be dispensed with if desired and feed wheel 51 driven from the motor that drives feed wheel l2. If this is done dead electrode 58 may be fed at same rate as electrode H or by the use of any of the well known variable speed drive arrangements it may be fed at any preferred rate. The apparatus of Fig. 4 illustrates the use of a dead electrode. however, it is within the scope of this invention to combine the apparatus of Figs. 3 and 4 so that live electrodes, 45 and 46, as well as dead electrodes 56 may be used with the hollow electrode of Fig. 1.

The novel method of this invention is of general application and can be applied with success to the coating or veneering of metallic articles, as described in detail in my copending application Serial No. 64,496, as well as to the joining or seam welding of metallic articles by the usual arc welding methods as well as by the method disclosed in the patent to B. S. Robinofl et al., No. 1,782,316. The novel method is not limited to any particular metal or alloy but in view of the wide use of steel and corrosion resistant alloys in the fabrication of metallic articles, it will be disclosed in connection with the coating or veneering of steel with corrosion resistant alloys.

The article to be coated or veneered, work II, is substantially horizontally positioned beneath welding head ID with the point, where it is desired to start the first band of veneer 59, beneath the path of electrode [4. Electrode I4 is then fed to work II and a gap starter such as a ball of steel wool, a sliver of graphite. an iron nail, etc., interposed between and in contact with electrode l4 and work II. A light frame 60 is placed around the area of work II to be covered by the first band of veneer 59. Frame 60 is then filled with flux 62 to form a blanket of substan-- tial thickness on work I I over and adjacent the situs of band 59.

Flux 62 may be any fusible compound, or mixture of fusible compounds, that do not produce an objectional amount of gas under the influence of the discharge of the welding current through the gap between the end of electrode I4 and work II and do not substantially add any undesirable constituents to or remove any desirable constituents from the deposited molten metal. silicates or components of silicates in general are suitab 6 particularly if they are dried or calcined or sintered prior to use. However, prefused silicates either neutral, basic or acid are sometimes preferred because of their pronounced non-gassing character. Manganese silicate, iron silicate, calcium silicate, aluminum silicate, glass and the like, both alone and in mixture have been found satisfactory. While flux 62 should be easily fusible its melting point should be high enough to assure that when molten it does not flow too readily but remains on, and in the immediate vicinity of, the molten metal to form a heat retaining and gas excluding blanket for the molten metal.

Flux I2 is preferably used in the granular form and while great variation is possible in the size of the flux particles, it is at present preferred to employ a flux the particles of which vary from powder size to A of an inch and more. The blanket of flux 82 is preferably of a depth sufficient tofsubmerge the arc", i. e. prevent the visual and audible manifestations of an arc from being apparent, and present a substantially quiet top layer of flux. It is my present belief that when an arc is submerged it is not extinguished and that the welding heat is generated by the arc discharge rather than by the passage of the welding current through a pool of molten flux. However, as this application is a disclosure of fact, it is to be understood that I am not to be bound by any theory or hypothesis as to what takes place beneath the surface of the flux blanket 62. Flux blankets 82 varying in depth from one inch or less to six inches and more have been found satisfactory.

To obtain band 59 with substantially straight edges strips 6| may be used. Strips 6| are located near the desired position of the edges of band 59 and enough space is allowed so that the molten metal does not contact strips 6|. Strips 6| restrain the flow of the flux and thus restrain the molten metal.

After the blanket of flux 62 is positioned the various control arrangements are adjusted and set to maintain the predetermined voltage and amperage, which in turn maintains a predetermined melting rate of electrode H and rate of travel along work II, as well as the predetermined rate of feed of the material from hopper l8 to and through hollow electrode II. The welding current circuit is then closed. With the submerged arc extremely high energy values may be used, thus 75 k. w. and more may satisfactorily be used with electrodes of about diameter. The'initial flow of current takes place through the arc starter which upon being consumed leaves a gap between the end of electrode I4 and work II. The current flow through the gap melts the surface of work II, the end of electrode I4 and some of flux 62. Since the fluxes mentioned specifically above, and practically all known useful fluxes, are second degree conductors and hence have a substantial conductivity when molten, it may be that the molten flux forms all or a part of the current path between the end of electrode l4 andwork II. If such is the case, the gap will be filled with molten flux and the welding heat will result from the passage of the welding current through the molten flux. However, if such is not the case, and the probabilities are that it is not, the molten flux will form a bubble in which an arc is present that plays between the end of electrode It and work ll. As the passage of current continues the metal fused from electrode l4 and from work H, as well as the metal that results from the material passing to the gap between the end of the electrode M and work I i coalesces into a unitary molten pool.

After the initial passage of current work H or welding head ill, or both, are moved to cause the molten metal to deposit on work II in the form of a wide band 59. When the work H and electrode J5 move relative to each other the flux melted by the welding-heat remains on and covers the molten metal. Because of the heat retaining capacity of flux 02 as well as the welding heat generated at the gap, the molten metal spreads out into a wide band and thoroughly fuses work ii. The operation is then allowed to continue until the full length or width of work M is covered by band 59. It is of course to be understood that during the band depositing operation sections of electrode M are added as required in the manner previously described. The remainder of the bands 59 required to completely cover the surface of work H are then deposited. The procedure is the same except that only one strip 68 is used, on the side opposite to the already deposited band or bands 59, and the new bands 59 are so located that their contiguous edges overlap so as to assure a continuous and properly fused surface as well as proper fusion into work H.

The band forming operation with the apparatus of Figs. 3 and 4 is substantially identical with that just described and a detailed description of the operation with each of these apparatus is not thought necessary.

The penetration of the veneer 59 into work H, or as may be otherwise stated, the amount of metal or work II that finds its way into veneer as, is an important factor in obtaining a veneer 59 of desired analysis. Penetration is effected by many factors among which may be mentioned, (1) rate of travel of the electrode relative to the work, (2) the current density and total energy dissipation at the gap between the electrode and the work, (3) the temperature of the work and (a) the position of the work with respect to the horizontal. In general a high rate of travel results in a higher penetration than a low rate. High current densities give higher penetration than lower current-densities; the same effect is noticed if the current density per unit of metal area is maintained constant and the area of the electrode is changed, an electrode of small area will give a greater penetration than an electrode of larger area. For a given set of conditions the higher the temperature of the work, the greater the penetration obtained, thus the penetration may be increased by preheating the work. Also, for a given set of conditions the penetration may be varied by tilting the work relative to, the horizontal. If the work is tilted so that the molten metal tends to flow down hill in back of the electrode the penetration is increased, if the work is tilted so that the molten metal tends to flow down hill ahead of the electrode the penetration is decreased. By controlling these factors almost any penetration desired may be obtained. In practice, penetrations ranging from 5 to have been obtained, i. e., veneers have been deposited that extended from 95 to 5% below the original surface of the work.

Because of the number of factors involved and the variations of each factor possible, no attempt will be made to set forth the specific manner for obtaining any particular penetration. It is believed that the above explanation will enable any skilled worker in the art to attain any desirable penetration in the range set forth.

In veneering steel with corrosion resistant alloys ordinary mild steel pipe with a carbon content in the neighborhood of 0.03% or less is a preferred material for electrode I4. The chromium required is preferably supplied by using ferro-chrome as the material passed to the gap between electrode Ml and work ll through electrode l4. Ferro-chrome of commerce usually contains, in round numbers 70% chromium and 30% iron and can be obtained with carbon contents ranging from 0.06% to 6.0%. When nickel in comparatively small quantities is required in the alloy it may also be supplied as material passed through hollow electrode M. The nickel may be metallic nickel ingranular or powder form or it may be some suitable nickel compound or alloy in the granular or powder form.

It should be evident to any one skilled in the art that a wide range of chromium-iron and chromium-iron-nickel alloy veneers are possible with the materials above mentioned simply by varying the penetration, the electrode feed, and the feed of the material passed through the electrode. In depositing any particular veneer, as for instance a 16% chrome veneer with a maximum of 0.12 carbon a penetration which allows the desired results to be obtained is first determined. With the work ll of 0.25 carbon steel a 40% penetration is satisfactory. -This penetration will cause work I i to supply 40 parts,

based on as the total parts, of the iron to I the final alloy'and .10 part of carbon. If a ferrochrome of 0.06 carbon is chosen the 16 parts of chromium required will besupplied by 22.8 parts of ferro-chrome. This amount of ferro-chrome will supply 6.8 parts of iron to the final alloy and 0.013 partof carbon. If a mild steel of .02 carbon is used as electrode M, the remainmg 37.2 parts of iron required will be supplied by an equal number of parts of electrode M. The electrode 16 will in this case add .006 part of carbon to the alloy. Thus, the finalveneer will contain the 16% chromium required and will have a carbon content of 0.119%. This analysis will be obtained by so adjusting and controlling the operation that 37.2 lbs. of electrode l4, and 22.8lbs. of ferro-chrome are alloyed with 40 lbs. of the metal of work I l fused by the current discharge. It is to be noted here that there is a loss of constituents during the operation due to the absorption by the flux and other causes so that in order to obtain the results indicated it will be necessary to increase the quantities above the values given to take care of the losses. It is sometimes desirable to add some of the alloying constituents through flux 62 by incorporating them in flux 62. In such casesthe quantity of the materials fed through hopper l9 will be modified accordingly.

With the materials specifically disclosed above a wide change in analysis of alloy is possible. Thus, the carbon content of the alloy may be reduced materially by using lower penetrations. The chromium content may be varied by changing the relative rate of feed of the ferro-chrome and melting rate of the electrode. By choosing the proper rates of feed and the proper penetration, chromium alloys may be deposited with .06 carbon and less and also with 26% chromium and more.

In carrying out the novel method, hollow electrode It may in general, be made up preponderantly of the metal that forms the base of the alloy of the veneer and the material fed through the hollow electrode ll will be made up of materials that predominate in the alloying ocnstituents of the veneer. However, cases will arise when this will be departed from, thus in depositing a high nickel, nickel-chrome-iron alloy, the penetration chosen may be such that the base metal will supply the major proportion oi the iron required. In such case it would be preferable to make the hollow electrode of pure nickel or some suitable nickel alloy. Thus, in interpreting this disclosure it is to be remembered that it contemplates the use of the materials for electrodes and feed material through the electrode ll best'suited to give the desired results.

In using the apparatus of Figs. 3 and 4 the method above described is also carried out, however, with these apparatus further variations of the method are possible. Thus, if it is desired to deposit an alloy having alloying constituents in minor proportions as, for instance, a 4-6 chrome steel, a portion of the steel required may be supplied through electrodes 45 and/or 56 and/or dead electrode 56. These apparatus are also suited to the cases where cheap metal is available having a composition not very much different from the desired alloy. In the later cases all or some of the electrodes may be made of the available material and the rest of the constituents required for the desired alloy fed through the hollow electrode.

I claim:

1. A method of depositing metal upon a plane surface of a workpiece under the influence of heat generated by the passage of electric current across a gap between a hollow consumable metal electrode and the workpiece, said gap maintained beneath a protective blanket of flux, including the steps of feeding metallic material throughsaid hollow metal electrode to said gap, and feeding a metallic member spaced from said hollow metal electrode and disconnected from the electric current circuit thereof to the vicinity oi said gap independently of said hollow metal electrode to be fused by the heat generated at said gap, the metallic material, the metallic member, the material of the workpiece, and the material of the hollow electrode supplying constituents to produce the desired deposited metal.

'2. The method of depositing alloy metal of a predetermined analysis which comprises; discharging electric current through a gap beneath the surface of a protective blanket of flux, between the end of a hollow metal electrode, made of one or more of the constituents of the desired alloy, and a metal body; supplying metal, made of one or more of the constituents of the desired alloy, in particle form through said electrode at a rate independent of the rate of feed of the electrode to be fused at the gap with the metal of the electrode, the metal of the electrode, the metal in particle form together with the metal of said body containing all of the constituents of the desired alloy; said electrode conducting substantially all of the metal in particle form supplied through it to the gap; and adjusting awaan the fusion rate of the electrode and the rate of supply of the metal in particle form to fuse at the gap predetermined proportions of the constituents of the desired metal; the gap being maintained beneath the surface of the flux blanket throughout the operation.

3. The method of depositing alloy metal of a predetermined analysis which comprises; discharging electric current through a gap beneath the surface of a protective blanket of flux, between the end of a hollow metal electrode of substantially unbroken periphery and made of one or more of the. constituents of the desired alloy, and a metal body; supplying metal, made of one or more of the constituents of the desired alloy, in particle form through said electrode at a rate independent of the rate of feed of the electrode to be fused at the gap with the metal of the electrode, the metal of the electrode, the metal in particle form together with the metal of said body containing all 'of the constituents of the desired alloy; said electrode conducting substantially all of the metal in particle form supplied through it to the gap; and adjusting the fusion rate of the electrode and the rate of supply of the metal in particle form to fuse at the gap predetermined proportions of the constituents of the desired metal; the gap being maintained beneath the surface of the flux blanket throughout the operation.

4. The method of depositing alloy metal of a predetermined analysis which comprises; discharging electric current through a gap beneath the surface of a protective blanket of flux, between the end of a hollow metal electrode, made of one or more of the constituents of the desired alloy, and a metal body; supplying metal, made of one or more of the constituents of the desired alloy, in particle form through said electrode at a rate independent of the rate of feed of the electrode to be fused at the gap with the metal of the electrode; the metal of the electrode, the metal in particle form together with the metal of said body containing all of the constituents of the desired alloy; said electrode conducting substantially all of the metal in particle form supplied through it to the gap; and adjusting the fusion rate of the electrode and the rate of supply of the metal in particle form to fuse at the gap predetermined proportions of the constituents of the desired metal; the relation between the cross-sectional area of the metal of the hollow electrode and the void of the hollow electrode and the fusion rate of the hollow electrode and the rate of supply of the metal in particle form being such that for each unit of length of electrode burned oil the volume of the void passing to the gap is substantially greater than the volume of the metal in particle form whereby metal in particle form does not seal the end of the electrode and accumulate therein; the gap being maintained beneath the surface of the flux blanket throughout the operation.

5. The method of depositing alloy metal of a predetermined analysis which comprises; discharging electric current through a. gap beneath the surface of a protective blanket of flux, between the end of a hollow metal electrode, made of one or more of the constituents of the desired alloy, and a metal body; supplying metal, made of one or more of the constituents of the desired alloy, in particle form through said electrode at a rate independent of the rate of feed of the electrode to be fused at the gap with the metal of the electrode, discharging electric current manner through a second gap beneath the surface of the protective blanket of flux, between the end of a second metal electrode, made of one or more of the constituents of the desired alloy, and the metal body; the electrodes being spaced apart and the gaps being so positioned that the metal fused by the electrodes forms a common molten pool; the metal of the electrodes, the metal in particle form together with the metal of said 10 body containing all of the constituents of the desired alloy; said hollow electrode conducting substantially all of the metal in particle form supplied through it to the gap; and adjusting the fusion rates of each of the electrodes and the rate of supply of the metal in particle form.

to fuse at the gaps predetermined proportions.

of the constituents of the desired metal; the gaps being maintained beneath the'surface of the flux blanket throughout the operation.