CHAPTER VII

PART ONE.--STEEL WELDING

(=83=) The term “steel,” as used in the following pages, unless otherwise specified, will be the term applied to wrought-iron and low-carbon steels. High-carbon and alloyed steels are to be considered only in advanced work and will therefore not be deemed a topic of interest to the beginner in laying his foundation.

(=84=) The welding of steel is much more difficult than cast iron on account of the many points which must be observed. In cast iron the metal is brought to a molten state and may be worked in that condition for some time without any apparent change in the characteristics of the metal. A flux is used to break up the oxide or scale and the metal will flow very easily. The flux is necessary because the oxide has a higher melting-point than the iron itself. When working on steel, it will be observed that just the reverse is true, that its oxide has a lower melting-point than the steel and consequently no flux or cleaning powder is necessary when working upon it.

(=85=) A large quantity of steel kept in a molten condition by the flame acting upon it is very easily influenced. The same area is not kept in a molten condition as with cast iron. The heat does not hold to the vicinity of the weld nearly so much as in cast iron because of the greater conductivity of the metal. If the flame is removed, the molten metal will set almost immediately. This means that the flame must be in contact with the metal at all times. It must be a strictly neutral flame or else one of the two gases will be introduced into the weld and its strength will be materially affected. The size of this flame must be such that too great an area will not be covered, yet enough must be covered to keep the metal along the line of the weld in a molten condition. If a carbonizing flame is used, one which has an excess of acetylene, such as was shown in Fig. 23, much carbon will be taken up by the metal, producing a brittle weld. If the flame is oxidizing, that is, contains an excess of oxygen which is noticed by the shortening of the flame and an accompanying hissing sound, Fig. 25, the metal will burn and a white foam will appear on the weld like a milky white glue. This tends to weaken the weld. This same effect will be in evidence if too large a tip is used. On the other hand if the tip is too small not enough heat is obtained and the oxides and other impurities which may be present will not be allowed to float to the surface but will be trapped in the weld.

(=86=) The filler-rod used on steel should be as near the same grade, if not better than the metal to be welded and should be very low in its carbon content. A pure grade of soft iron wire or mild steel will make a very good filler-rod for ordinary purposes. The size of this filler-rod is very important, for it should fuse at the same time as the metal being worked upon, and unless it does this the weld will not be satisfactory. If the filler-rod is too large it will not be at the fusion point when the work is, and will not fuse with it. If the rod is brought to a melting-point the work will have too much heat and will burn. On the other hand, if the filler-rod is too small, it will burn up before the work is at the fusion point, or in other words, the work will still be too cold when the rod is melted.

(=87=) There are many different methods of executing a steel weld, and it has been noted that very few experienced welders handle their steel in the same manner. Most of these methods are very difficult to learn and can be perfected only after years of practice. However, a simple method which will produce results is thought the most advisable for the beginner. A careful examination and study of this point has brought out the following method, which is very easily picked up and which dispenses with most of the torch movements that are generally advocated by the old time welders.

[Illustration: FIG. 57.--Preparing and Heating Steel before Welding.]

(=88=) When welding two pieces of steel bars, the cross-section of which will measure one-half inch by three inches, they are beveled off and prepared in the manner illustrated in Fig. 57, either by means of a chisel, file, or by the use of a grinding wheel. About an eighth of an inch of the original stock is left on the bottom side and the angle formed from these two places when brought together, should be 90 degrees. When the pieces have been prepared and placed in the position shown in the illustration, the neutral flame is then brought down at right angles to the plane of the metal, so that the end of the cone will just lick the surface. It is moved up and down upon each side of the part to be welded until each piece is brought to a red heat, for a distance of at least one inch back. The position of the torch during this operation can be seen in Fig. 57. From this time on, the operator should work as rapidly as possible, for the quicker the fusion of the metal is brought about, the less oxide or scale will appear and a better weld will result. The description of this process may take some length but the actual fusion not nearly so long.

[Illustration: FIG. 58.--In Welding Steel, the Beginner Should Fuse His Pieces together along the Bottom with the Torch Flame, Adding no New Metal. The Metal on Both Sides of the Torch Flame is Melted together until a Small Pool of Molten Metal Appears, then the Torch is Twisted Smartly away, as Shown by the Arrow, and the Metal Allowed to “Set” for an Instant before Proceeding along the Line of Weld.]

[Illustration: FIG. 59.--Method of Adding “Filler-rod” in Welding Steel. Note that the Rod is Worked behind the Flame.]

(=89=) When the red-hot stage is reached, the neutral flame is brought down to the very lowest part of the “V” at the side nearest the operator and held there until the metal has melted and is about to collapse. The flame is then quickly twisted away for just a second to let the metal set. Perhaps this operation will fuse about one-half inch or less along the bottom of the “V.” This same operation is repeated along the line of weld until the whole piece is fused along the bottom. It will be noted that no filler-rod has as yet been used. After the last portion has been fused, the flame is brought back to the starting-point and played not only on the bottom, which has already been fused, but on the sides of the “V” as well, bringing an area of about one inch in diameter to a molten condition. The tip of the welding torch is held in a vertical position all this time to introduce as much heat into the weld as possible. During this operation the filler-rod, which should measure three-sixteenths or one-quarter inch in diameter, is picked up by the operator’s free hand and its end brought near the heat of the flame so that it may be warmed and will not chill the metal when introduced into the weld. When the melted metal is running freely, the tip of the welding torch is slowly inclined in the direction of the part to be welded and is advanced along the “V-ed” out portion at this angle as rapidly as the metal can be made to melt. This position is shown in Fig. 59. It will be noted that as the flame advances along the line of the weld the molten metal will mount up behind it of its own accord, providing the metal is in a molten condition, when the flame passes over it. During this period the filler-rod is stirred into the molten metal in a circular movement which should be in back of the torch as much as possible. This means that the torch comes in contact with the filler-rod but very little and the rod is melted, not by the flame, but by the molten metal of the piece being welded. It will be noticed at times, when too much metal has been welded and the torch is not advancing rapidly enough, that some of the molten metal will run ahead of the flame, into that part of the “V” yet to be fused, and to the unwary student this will be looked upon as a safe place to add his filler-rod. However, when the piece is broken and the cross-section of the weld examined, it will be found that in this part of the weld, the metal has only been laid on and not fused. The beginner should watch this operation and see to it that this molten metal is not permitted to run ahead of his torch, an act which he can overcome by the proper manipulation of his filler-rod, which really governs all the melted metal behind the flame. If not enough metal has been added to fill in the “V” to the proper thickness, this operation can be repeated until enough metal has been added. By practicing this method the student can be taught to execute a very successful weld and reinforce it all in one operation without any chance of burning his filler-rod or lapping his metal. More practice is required to successfully weld steel than most other metals and the beginner should not be discouraged if it takes him some time to conquer this metal. It should be forcibly impressed on the student that the metal must be in a molten condition before the filler-rod is added, or else it will stick and prevent his working readily and in addition will produce a very faulty weld. Fusion is the thing to bear in mind for without it success cannot be expected.

[Illustration: FIG. 60.--This Method of Adding the “Filler-rod” when Welding is not Recommended for the Beginner.]

(=90=) While outside appearances should not be considered as a prime requisite, when beginning it is always well to add more metal than is really necessary in order to reinforce the weld as much as possible. It cannot be expected, however, that a steel with the same cross-section as the original will possess the same properties and be as strong, for a weld is only a casting unless treated otherwise and the steel or wrought iron used in the specimens is of rolled stock. If too much metal has been added and dressing down is necessary, the student will find that by using a slightly oxidizing flame the surplus metal can be burnt away very rapidly and a very good-looking job can be executed much more rapidly than if a neutral flame were used. It is well to remember, however, that this is used only in dressing off pieces and in places where the strength of the weld is not to be jeopardized.

(=91=) When advancing in steel work, it will be noticed that the same provision for contraction and expansion is not considered in as great proportions as on cast iron, and the reason is quite evident. In cast iron we find the metal is very brittle and will not give without breaking, whereas on steel it is more ductile and will twist and bend before breaking. This does not mean, however, that the important points of expansion and contraction are to be neglected in steel work, for they are very important, as we shall see later on.

PART TWO.--STEEL WELDING

(=92=) It is still supposed that the beginner knows very little about the various kinds of metals, or methods of distinguishing between them. This is of great importance and should at once be overcome, as he will not at all times have someone over him to diagnose his case and tell him the proper procedure. For instance, were he to be given a piece of cast steel to weld, thinking that it was cast iron, he would use a cast-iron filler-rod in executing his weld. The results of such a weld would not be very favorable, and the same would hold true if a steel filler-rod were used on cast iron. An occasional glance at the table in paragraph 67 will acquaint him with the various tests to make when deciding upon the nature of the piece to be worked upon. The tests should be applied in every doubtful instance. When working on cast steel, a student may think that he must have a cast-steel filler-rod, but this is an exception to the general rule and he can use the same style filler-rod as he would employ on ordinary steel work. It might be mentioned here that when working on alloyed and high-carbon steels, the filler-rod generally contains some of the alloy or carbon which will tend to replace that destroyed by the action of the flame in the original metal.

(=93=) In welding cast steel the same procedure takes place as previously described for steel, and it should present no real difficulties after that process is understood. There may be more sand, oxide and other impurities present on account of the nature of the metal, but these can all be worked out if plenty of heat is applied. At times, when working in steel, it will be found that there may be a small hole develop in the center of the weld and as the torch is worked into this hole it is found that it goes down a short distance and seemingly refuses to be worked out. This is what most welders call a “crater,” and is caused by the metal at the bottom not being hot enough for the surrounding melted metal to fuse it. When found they should be removed before adding any more metal. By playing the torch flame around and around it, so that the heat may be transmitted to the bottom of the “crater” and it brought to the melting-point like the surrounding metal and suddenly jerking the torch away, it will disappear. “Craters” are generally formed during the first part of the weld, especially if the “V” is narrow, and they are hard to handle when deep. Under no circumstances should the filler-rod be melted into them in trying to make them disappear, as this will only mean covering them over.

(=94=) Some welders find that hard spots develop in their welds which they have difficulty in overcoming, and it is a very serious handicap when the weld is to be machined, for ofttimes it will break very expensive tools and leave a portion of a drill or die broken off in the metal. It is probably safe to say that the principal cause of hard spots in steel welds is due to lack of heat. This, if given careful thought and consideration, will be brought home forcibly to the welder as he proceeds in his work, for the lapping of metals, trapping of oxides, “craters,” too rapid cooling, etc., may all be directly attributed to a lack of sufficient heat. If the metal is in a molten state, all impurities will be brought to the surface, for they are bound to be displaced by the weight of the metal, the same as corks in a barrel will float to the top if water is introduced. The water in this case has a greater specific gravity than the corks.

[Illustration: FIG. 61.--The Open Ends on long Steel Welds will Overlap as the Welding Progresses if Improperly Started.]

[Illustration: FIG. 62.--Showing how Open Ends of Steel pieces are Spread Slightly to Overcome Lapping of Ends in Making Weld.]

(=95=) In welding on sheet iron and steel, many operators will find that they have more difficulty in executing a successful weld than on slightly heavier work. This is no doubt due to the thin nature of the work and the ease with which it may be burned or carbonized if the operator is not alert. When working on such material a very small filler-rod is used if thought necessary but this rod must be as free from impurities as possible. When working on a long seam such as may be encountered on a steel tank, it will be noticed that in welding from one end along the seam that the metal ahead of the flame will tend to overlap as shown in Fig. 61. This may be overcome by tacking (that is, fusing the metals together), at various points before starting the weld, or the parts ahead of the torch can be separated as is shown in Fig. 62 and held this way by using a wedge. This is moved along as the weld advances and permits the edges to close together. Another method used by manufacturers who make a specialty of this work is to construct a jig which clamps the ends rigidly and they are welded while in this position. This phenomenon in steel welding will appear rather strange to the welder who has had some experience on thin cast-iron work, such as oven doors and the like. In these he found that as his weld advanced, the welded portion before him would separate, and when he had welded about four inches or so it would be necessary for him to jump his flame back to the beginning of his weld and heat up that portion, in order to close up the cracks before him previous to his continuing the work. This is illustrated in Fig. 63. This may be explained by the fact that steel is a very ductile metal and when it is fused, the expansion is taken care of internally by the two edges combining. Then, in cooling, the metal contracts, an action much more rapid in steel than in cast iron, and draws the edges of the steel plates past each other so that they overlap. In cast iron, which is rigid, the edges are expanded by the fusion of the metal and this space is then filled up with new metal, holding the edges apart. As the weld progresses the metal ahead of the torch is pushing out, and that behind is cooling off, which acts as a lever on each side to open up the unwelded ends.

[Illustration: FIG. 63.--This Illustration Shows how the Open Ends of Thin Cast-iron Pieces Spread apart as the Weld Progresses. To Close the Edges together, Jump the Torch Flame from _B_ to _A_; as _A_ heats up, _B_ Cools and the Lever-like Action Closes the Opening.]

(=96=) To weld a broken automobile frame successfully the body of the car should be raised if necessary, to keep it from burning and all pipes, wires and gasoline leads protected with a covering of asbestos paper. Plenty of room should be allowed, so that the welder may have easy access to the break, and the frame should be jacked up on both sides of the break until the frame is in proper alignment. Then weld the crack from the outside, working across the top, then down the side and across the bottom, reinforcing a little if necessary on all sides but the bottom. Then repeat this operation on the inside, reinforcing at all points. Then take a strip of steel about one-eighth or one-quarter inch thick and six or eight inches long and as wide as the bottom of the frame. This piece should be welded securely to the bottom of the frame with the former break in the middle of the strip. A cut representing this job is shown in Fig. 64. By this method the frame can be made stronger than originally.

[Illustration: FIG. 64.--A Good Method of Reinforcing a Weld on an Automobile Frame is Here Shown. The Patch as Pictured Here is only “Tacked On.” It Should be Welded Securely to the Bottom of the Frame on all Four of its Edges.]

PART THREE.--STEEL WELDING

(=97=) Aside from the difficulties already mentioned in steel welding, there are many others. A few of these will be taken up in order to let the beginner know how to approach the various problems which may confront him. But in no wise is this to be considered to be a treatise on advanced work. Ofttimes the question arises, Can springs be successfully welded? Now, while springs have been welded, and they have been tested out thoroughly, yet the practice of spring welding with the oxy-acetylene flame is not to be recommended. There are those who will weld leaf springs, such as are found on automobiles, and will apply rapid blows with the hammer, while their weld is still in a heated condition and then plunge the spring in water or oil to harden it and the weld. A close observer will readily see why this procedure is not correct. Springs of this nature are made up of metal which takes a uniform hardening, and were it not so they could not be considered springs. Now, if there is a fracture and a foreign metal, which under no circumstances can be expected to take the same hardening as the rest of the spring, is introduced into the weld, it can easily be seen why a fusion of this kind is not to be relied upon. If it were possible to diagnose or take an analysis of the metal in the spring and use a filler-rod which, after being acted upon by the flame, would come out the same as the metal in the spring, then some success might be expected, but until that time, welding of springs will not be encouraged. Unless perchance the break is of such a nature that it can be reinforced readily and is in such a position that a resilient quality is not necessary.

[Illustration: FIG. 65.--Building Up Worn Shafts.]

(=98=) Work on crank-shafts often causes perplexity on the part of the beginner, for he usually hears this matter discussed pro and con. Crank-shafts of four inches in diameter can be successfully welded with the oxy-acetylene flame, and even larger, if correct methods are employed. There are many points which the welder considers before deciding whether a weld of this nature is advisable. Of course the usability of the piece after it has been welded is the main issue when executing any kind of a repair job. Now, a crank-shaft will generally break in either of two ways; by some external force, such as a connecting rod breaking loose, or by crystallization, which is usually due to fatigue. Now, in the latter case, ofttimes the shaft will break in the cheek of the “offset,” and possibly no part of the shaft is thrown out of alignment. When such is the case, welding is usually recommended and the shaft may be brought back to a useful state in very quick order. However, in the former case, the shaft is apt to be sprung, and while it could be welded, the machine work necessary to restore it to normal requires much time, and it has been known, where after spending a matter of days in trying to get proper alignments, work was scrapped as useless. So it is entirely up to the welder in work of this kind to determine whether the job is worth while or not. There are certain parts of a crank-shaft upon which welding work can be done with a marked degree of success, such as building up worn bearings and the like. In doing work of this kind it is recommended that the welder fuse his metal in a line parallel to the center line of the bearing, seeing to it that he has a perfect fusion between the surface of the bearing and the metal he is fusing and adding plenty of metal, to insure enough being used, so that no low spots will show up when it is machined. It is considered that by adding the metal as suggested the welder will hold his heat much better than if he attempted revolving the shaft continually. Fig. 65 will show the method here outlined in a very clear way.

[Illustration: FIG. 66.--Shaft Broken at End of Square Shank, its Weakest Point.]

[Illustration: FIG. 67.--Broken Part of Shaft Removed and New Piece Added, thereby Moving the Weld away from the Weak Part.]

(=99=) When working on shafts the welder will encounter such articles as automobile propeller shafts and rear axles, which generally break adjoining the square ends. He will no doubt wonder whether it is advisable to weld this square end back on, or whether to try and build up the shaft the desired length. Undoubtedly the point of fracture is the weakest portion of the entire shaft, else it would not break there. The execution of a weld at this point where no additional metal can be added or any means of reinforcing used is not to be recommended. Fig. 66 will show the problem which confronts the welder, and Fig. 67 the suggested means of overcoming the difficulty. By removing about four inches from the broken end of the shaft and adding a new piece, about ten inches long, of the same diameter, the weld will be removed from the weak point; a heavier weld can be made, and the end can be machined off to the desired size. This procedure is recommended on all jobs of like nature.

(=100=) Occasionally case-hardened ring gears are brought to the welder to have teeth built up or new ones added, and although the welder must realize that the hardening is destroyed by the action of the flame, yet he does not understand why it is necessary to reharden the gear. A little thought on this subject will make him appreciate the fact that if he destroys certain properties in metal which have been introduced for a reason, these must be replaced if he would bring the job back to normal. It would be like heating up a tempered lathe tool, or cold chisel for that matter, and trying to use it before it had been retempered. Therefore if hardening or temper is destroyed by the flame it must be restored.

[Illustration: FIG. 68.--When Welding a Small Section to a Larger One, the Flame of the Torch is Directed toward the Heavier of the Two.]

(=101=) If a weld were to break, it would be necessary for the welder to remove all metal added in the first weld before attempting to reweld. This is true of his own work as well as that of others which he may be called upon to do. For no matter how good the surface may appear, without a solid foundation no weld is of any value, and unless he clears out all of the old metal he cannot be sure of the work. This will apply not only to steel work, but to all metals, and it is a point which should be borne in mind.

(=102=) At times there are jobs come up in which one piece of work is to be fused to another which is much larger, and will absorb much more heat during the weld. When handling such work, it will be necessary to play the torch upon the larger piece most of the time, as shown in Fig. 68, in order to bring both pieces to a fusion point at the same time and keep them in that condition.

(=103=) Once in a while it will be necessary for a welder to fuse cast iron to steel or vice versa, and the question will arise as to which filler-rod he will use. It has been found that a cast-iron filler-rod can be used with success and of course when using a cast-iron filler-rod, a cast-iron flux will be necessary. Work of this nature is not very frequent.

PART FOUR.--STEEL WELDING

(=104=) When steel is in a melted condition, it seems to be in a very susceptible state. It appears to absorb gases, and with constant working an oxidation is in evidence which materially effects the strength of the metal. When working on vanadium and other alloyed steels, if kept in a molten condition too long, many of their principal characteristics are destroyed. For this reason it is advisable to execute steel welds just as rapidly as possible. While this is true of most work, it is especially to be emphasized on steel. To assist the welder in executing welds on large steel castings, the pieces are generally preheated, so that the work will take less time, be more successful, and save both oxygen and acetylene. When working on preheated jobs, in order to get the desired angle on the filler-rod so the welder may use it without discomfort, a light heat is played on the filler-rod, a matter of six or eight inches from the end being fused and then bent to an angle of 90 degrees, so that the operator may hold the rod at some distance from his work and still introduce it in the manner he desires. Some operators weld their cast-iron filler-rods together, to get the desired angle as shown in Fig. 69, but this is not as common as the steel method, probably because cast iron will not bend and it requires some time to weld the rods together in this manner.

[Illustration: FIG. 69.--Kinks for Handling “Filler-rod” on Large Work to Remove Welder’s Hand away from Heat of Flame.

(_a_) shows how the steel “Filler-rod” is heated by the torch flame about 6 inches from the end and bent to the angle desired.

(_b_) illustrates how cast “Filler-rods” are handled. Since they will not bend, they are welded in the T shape shown. First one side is used in fusing, and then the other.]

(=105=) In some parts of the country boiler flues are acted upon and eaten away by the impure water used, and when high prices prevail, retipping is generally resorted to. A simple method in which they can be satisfactorily and cheaply done is as follows: Cut off the poor end until solid metal is reached, with a pipe cutter, which will tend to “V” the work as it cuts and at the same time will squeeze the edge of the pipe in. After cutting, this end of the flue is placed on the horn of an anvil and the burr on the inside, which has been made by the cutter, is flattened out. It is very important that the flue be of the same size throughout in order to permit its being cleaned. It is then placed in “V” blocks or a trough, made of angle iron, such as shown in Fig. 70, and the new end which has been prepared in much the same way is placed in the position shown in _A_ in the same figure. The piece is tacked on at two or more spots and then laid aside until the whole set of flues has been prepared in this manner. Then they are replaced in the trough and welded, one after another, being turned at one end by a helper, thus allowing the welder to do continuous work. Care must be taken at all times that perfect fusion takes place between the flue proper and the piece being added, yet at no time should the metal be allowed to run on the inside of the pipe. More metal can be added than is really necessary and can later be dressed down on a grinding wheel to the desired size, which must be such that replacement of the flue can be made. Various-sized pipes can be welded in much the same way where no reducers are obtainable, the only change being that there must be a step made in the trough which will permit the various-sized pipes being lined up correctly. Jigs for the speeding up of manufactured articles which are to be welded are always being brought out by the ingenious workman and are to be encouraged whenever possible.

[Illustration: FIG. 70.--Showing a Simple Way to “Line-up” Flues when Retipping. _B_ Represents the old Flue, and _A_ the New Piece to be Added.]

[Illustration:

(_Courtesy of the Oxweld Acetylene Co._)

FIG. 71.--Welded Cracks between Staybolts.]

(=106=) In the repair of boilers many a feasible job has been given up as impossible by the unthinking welder. Cracks have been found in fire-box sheets around the staybolts which, as soon as they are touched with the flame, seem to run and keep running. They really discourage those who are not familiar with this class of work. Many such welds have been executed and are apparently all right until tested, when they give way and make the job worse than it was previously. The trouble is in these instances that the welder has made no provision for contraction and while the job might appear to be successful, yet the internal strains exerted will not show themselves at the test. Many boiler shops have found that the flat patch is not to be relied upon and when a crack is found between two stay-bolt holes, such as shown in Fig. 72, a round hole is cut as shown by the dotted line. A circular plate is then cut slightly larger than this hole and after being brought to a red heat, it is bellied by the use of a hammer or a set of dies, so that it assumes the shape of a saucer and is called by many a “dished” patch. Some idea may be had of such a patch from Fig. 73.

[Illustration: FIG. 72.--A Crack between the Staybolts in a Boiler should be Cut Out as Shown by the Dotted Line, to Prepare it for a “Dished” Patch.]

[Illustration: FIG. 73.--A “Dished” Patch.]

(=107=) The patch is placed in the sheet with the concave side toward the operator and should be securely welded in place, adding as little metal for reinforcement as possible, but seeing to it that a perfect fusion is made between the patch and the sheet all the way through. As soon as the weld is complete the torch is played upon the high part of the patch, which is protruding, and as the weld cools off, sharp quick blows can be applied to the center of the patch, which should be kept in a heated condition until it is nearly flat. This will take care of any contraction that might set up and is a very good way of handling patches which do not exceed six or eight inches in diameter.

[Illustration: FIG. 74.--A “Corrugated” Patch.]

(=108=) A “corrugated” patch has been brought out more recently than the “dished” patch, and as its name would indicate, it has corrugations around at least three of its sides. While a “dished” patch is limited in its scope and cannot be applied to square holes unless the square holes be cut round, the “corrugated” patch knows absolutely no limits as to size or shape. While its preparation is probably more difficult, yet its purpose is the same, that is, to take care of the contraction which takes place in sheets of metal where heat has been introduced. To prepare a “corrugated” patch, a piece of metal which is somewhat larger than the hole is taken and the corrugation is made by placing two rods on one side and somewhat separated and between them on the other side another rod. With this section of the patch heated to a red heat, a drop hammer is played upon it and a corrugation effected. Or an easier method is by the use of specially prepared dies, which will turn out a patch in quick order. It must be remembered that while the patch shown in Fig. 74 is only for a very simple job, which is rectangular in shape, yet “L” shaped patches can be prepared and handled in the same manner. When the corrugation has been introduced into the patch, the latter is cut so that it will fit the hole, and it is tacked in position with the bellied sides out. The method used in applying a patch of this kind is to weld the uncorrugated side, then start up the corrugated side and weld for two or three inches, then play the torch upon the corrugation, adjoining the part welded, and slightly hammer to assist in the expansion of the same; then return to the weld, continuing it until the corrugation can again be played upon. By doing this, when finished the patch will be flat and no signs of the corrugations will be shown. While many patches of this nature are in use giving the very best service, the welder who looks upon the finished job cannot tell how it has been accomplished.

(=109=) While the methods here given seem only to apply to boiler work, they are not so restricted and can be applied to tanks and various vessels with success. However, when welding on tanks which have contained inflammable gases or gasoline the welder is cautioned to take every measure to safeguard himself, and while it is known that much work is being done on such jobs, it is not recommended and in fact quite the contrary. It is true that there are such methods as filling the containers with water; cleansing with live steam, and so forth, but the cautious man will refrain from working on these vessels even though such measures have been taken. Gasoline has a faculty of penetrating the pores of metallic surfaces, and although these vessels have been emptied and have remained so a matter of a year, the gasoline is still present to some extent, as is evidenced by the fact that as soon as heat is applied and the molecules of the metal are expanded, the gas is released in sufficient quantities to cause an explosion. This is not in one instance only, but in many, so it has been thought best to discourage any welding work on vessels which have contained gasoline at any time.

[Illustration: FIG. 75.--Working a Vertical Weld on Steel, from the Top Down.]

(=110=) While it is possible to weld cast iron on the vertical, by the use of carbon blocks and so forth, the same kind of work can be accomplished on steel with much ease, without the use of any blocks, or materials other than the filler-rod and the welding torch. There are two methods of handling vertical welds; welding from the top down, or starting from the bottom and working up. The former seems to be condemned by those who have never tried it, on account of the carelessness which is apt to be used on work of this kind. However, for the beginner, it is thought advisable to teach this method, as there are many places where it can be used advantageously. The metal at the top of the seam, such as a broken automobile frame, or the like, is brought to a molten state and held there, not only by the velocity of the flame, but also by the filler-rod, as is shown in Fig. 75. With the choosing of a tip of the correct size, the melted metal can be held under control with much ease, after a little practice, and it is allowed to descend as soon as the metal below it is in the proper shape for fusion. The filler-rod is added continually, for it is never lifted out of the molten metal, merely stirred a little from side to side as it descends. None of the melted metal is allowed to precede the flame, and at all times the operator can see whether the edges to be fused are at the right heat. As soon as the bottom is reached, the weld can again be gone over if it is not thought strong enough, and reinforced as much as desired. As soon as the operator is familiar with this method, he will find that much more speed can be developed, less filler-rod lost and less lapping done than by building up from the bottom.

(=111=) In welding over head there is a tendency on the part of most welders to avoid the use of enough heat to bring their metal to a molten state, for fear that it will drop upon them. It must be remembered that lack of heat means poor welds and that the metal must be in a molten condition whenever the weld is to be made. As soon as a little practice is given to this kind of work, the welder will see that the melted metal can assume some proportions without dropping off, despite its weight. It has probably been noticed that on “sweating” water tanks drops of water accumulate on the bottom of the tank and do not fall off. It is the same sort of problem in the case of melted steel. The adhesion of the molecules and the surface tension are the forces that keep the metal from dropping.