Having opened the vent pipe valve and relieved the pressure, open the valve in the water filling tube. Close the residuum valve, then run in several gallons of water and revolve the agitator, after which draw out the remaining residuum; then again close the residuum valve and pour in water until it discharges from the overflow pipe of the drainage chamber. It is desirable in filling the generator to pour the water in rapidly enough to keep the filling pipe full of water, so that air will not pass in at the same time.

After the generator is cleaned and filled with water, fill with carbide and proceed in the same manner as when first charging.

Carbide Feed Mechanism.—Any form of carbide to water machine should be so designed that the carbide never falls directly from its holder into the water, but so that it must take a more or less circuitous path. This should be true, no matter what position the mechanism is in. One of the commonest types of forced feed machine carries the carbide in a hopper with slanting sides, this hopper having a large opening in the bottom through which the carbide passes to a revolving circular plate. As the pieces of carbide work out toward the edge of the plate under the influence of the mass behind them, they are thrown off into the water by small stationary fins or plows which are in such a position that they catch the pieces nearest the edges and force them off as the plate revolves. This arrangement, while allowing a free passage for the carbide, prevents an excess from falling should the machine stop in any position.

When, as is usually the case, the feed mechanism is actuated by the rise or fall of pressure in the generator or of the level of some part of the gas holder, it must be built in such a way that the feeding remains inoperative as long as the filling opening on the carbide holder remains open.

The feed of carbide should always be shut off and controlled so that under no condition can more gas be generated than could be cared for by the relief valve provided. It is necessary also to have the feed mechanism at least ten inches above the surface of the water so that the parts will never become clogged with damp lime dust.

Motor Feed.—The feed mechanism itself is usually operated by power secured from a slowly falling weight which, through a cable, revolves a drum. To this drum is attached suitable gearing for moving the feed parts with sufficient power and in the way desired. This part, called the motor, is controlled by two levers, one releasing a brake and allowing the motor to operate the feed, the other locking the gearing so that no more carbide will be dropped into the water. These levers are moved either by the quantity of gas in the holder or by the pressure of the gas, depending on the type of machine.

With a separate gas holder, such as used with low pressure systems, the levers are operated by the rise and fall of the bell of the holder or gasometer, alternately starting and stopping the motor as the bell falls and rises again. Medium pressure generators are provided with a diaphragm to control the feed motor.

This diaphragm is carried so that the pressure within the generator acts on one side while a spring, whose tension is under the control of the operator, acts on the other side. The diaphragm is connected to the brake and locking device on the motor in such a way that increasing the tension on the spring presses the diaphragm and moves a rod that releases the brake and starts the feed. The gas pressure, increasing with the continuation of carbide feed, acts on the other side and finally overcomes the pressure of the spring tension, moving the control rod the other way and stopping the motor and carbide feed. This spring tension is adjusted and checked with the help of a pressure gauge attached to the generating chamber.

Gravity Feed.—This type of feed differs from the foregoing in that the carbide is simply released and is allowed to fall into the water without being forced to do so. Any form of valve that is sufficiently powerful in action to close with the carbide passing through is used and is operated by the power secured from the rise and fall of the gas holder bell. When this valve is first opened the carbide runs into the water until sufficient pressure and volume of gas is generated to raise the bell. This movement operates the arm attached to the carbide shut off valve and slowly closes it. A fall of the bell occasioned by gas being withdrawn again opens the valve and more gas is generated.

Mechanical Feed.—The previously described methods of feeding carbide to the water have all been automatic in action and do not depend on the operator for their proper action.

Some types of large generating plants have a power-driven feed, the power usually being from some kind of motor other than one operated by a weight, such as a water motor, for instance. This motor is started and stopped by the operator when, in his judgment, more gas is wanted or enough has been generated. This type of machine, often called a "non-automatic generator," is suitable for large installations and is attached to a gas holder of sufficient size to hold a day's supply of acetylene. The generator can then be operated until a quantity of gas has been made that will fill the large holder, or gasometer, and then allowed to remain idle for some time.

Gas Holders.—The commonest type of gas container is that known as a gasometer. This consists of a circular tank partly filled with water, into which is lowered another circular tank, inverted, which is made enough smaller in diameter than the first one so that three-quarters of an inch is left between them. This upper and inverted portion, called the bell, receives the gas from the generator and rises or falls in the bath of water provided in the lower tank as a greater or less amount of gas is contained in it.

These holders are made large enough so that they will provide a means of caring for any after generation and so that they maintain a steady and even flow. The generator, however, must be of a capacity great enough so that the gas holder will not be drawn on for part of the supply with all torches in operation. That is, the holder must not be depended on for a reserve supply.

The bell of the holder is made so that when full of gas its lower edge is still under a depth of at least nine inches of water in the lower tank. Any further rise beyond this point should always release the gas, or at least part of it, to the escape pipe so that the gas will under no circumstances be forced into the room from, between the bell and tank. The bell is guided in its rise and fall by vertical rods so that it will not wedge at any point in its travel.

A condensing chamber to receive the water which condenses from the acetylene gas in the holder is usually placed under this part and is provided with a drain so that this water of condensation may be easily removed.

Filtering.—A small chamber containing some closely packed but porous material such as felt is placed in the pipe leading to the torch lines. As the acetylene gas passes through this filter the particles of lime dust and other impurities are extracted from it so that danger of clogging the torch openings is avoided as much as possible.

The gas is also filtered to a large extent by its passage through the water in the generating chamber, this filtering or "scrubbing" often being facilitated by the form of piping through which the gas must pass from the generating chamber into the holder. If the gas passes out of a number of small openings when going into the holder the small bubbles give a better washing than large ones would.

Piping.—Connections from generators to service pipes should preferably be made with right and left couplings or long thread nipples with lock nuts. If unions are used, they should be of a type that does not require gaskets. The piping should be carried and supported so that any moisture condensing in the lines will drain back toward the generator and where low points occur they should be drained through tees leading into drip cups which are permanently closed with screw caps or plugs. No pet cocks should be used for this purpose.

For the feed pipes to the torch lines the following pipe sizes are recommended.

3/8 inch pipe. 26 feet long. 2 cubic feet per hour. 1/2 inch pipe. 30 feet long. 4 cubic feet per hour. 3/4 inch pipe. 50 feet long. 15 cubic feet per hour. 1 inch pipe. 70 feet long. 27 cubic feet per hour. 1-1/4 inch pipe. 100 feet long. 50 cubic feet per hour. 1-1/2 inch pipe. 150 feet long. 65 cubic feet per hour. 2 inch pipe. 200 feet long. 125 cubic feet per hour. 2-1/2 inch pipe. 300 feet long. 190 cubic feet per hour. 3 inch pipe. 450 feet long. 335 cubic feet per hour.

When drainage is possible into a sewer, the generator should not be connected directly into the sewer but should first discharge into an open receptacle, which may in turn be connected to the sewer.

No valves or pet cocks should open into the generator room or any other room when it would be possible, by opening them for draining purposes, to allow any escape of gas. Any condensation must be removed without the use of valves or other working parts, being drained into closed receptacles. It should be needless to say that all the piping for gas must be perfectly tight at every point in its length.

Safety Devices.—Good generators are built in such a way that the operator must follow the proper order of operation in charging and cleaning as well as in all other necessary care. It has been mentioned that the gas pressure is released or shut off before it is possible to fill the water compartment, and this same idea is carried further in making the generator inoperative and free from gas pressure before opening the residue drain of the carbide filling opening on top of the hopper. Some machines are made so that they automatically cease to generate should there be a sudden and abnormal withdrawal of gas such as would be caused by a bad leak. This method of adding safety by automatic means and interlocking parts may be carried to any extent that seems desirable or necessary to the maker.

All generators should be provided with escape or relief pipes of large size which lead to the open air. These pipes are carried so that condensation will drain back toward the generator and after being led out of the building to a point at least twelve feet above ground, they end in a protecting hood so that no rain or solid matter can find its way into them. Any escape of gas which might ordinarily pass into the generator room is led into these escape pipes, all parts of the system being connected with the pipe so that the gas will find this way out.

Safety blow off valves are provided so that any excess gas which cannot be contained by the gas holder may be allowed to escape without causing an undue rise in pressure. This valve also allows the escape of pressure above that for which the generator was designed. Gas released in this way passes into the escape pipe just described.

Inasmuch as the pressure of the oxygen is much greater than that of the acetylene when used in the torch, it will be seen that anything that caused the torch outlet to become closed would allow the oxygen to force the acetylene back into the generator and the oxygen would follow it, making a very explosive mixture. This return of the gas is prevented by a hydraulic safety valve or back pressure valve, as it is often called.

Mechanical check valves have been found unsuitable for this use and those which employ water as a seal are now required by the insurance rules. The valve itself (Figure 13) consists of a large cylinder containing water to a certain depth, which is indicated on the valve body. Two pipes come into the upper end of this cylinder and lead down into the water, one being longer than the other. The shorter pipe leads to the escape pipe mentioned above, while the longer one comes from the generator. The upper end of the cylinder has an opening to which is attached the pipe leading to the torches.



The gas coming from the generator through the longer pipe passes out of the lower end of the pipe which is under water and bubbles up through the water to the space in the top of the cylinder. From there the gas goes to the pipe leading to the torches. The shorter pipe is closed by the depth of water so that the gas does not escape to the relief pipe. As long as the gas flows in the normal direction as described there will be no escape to the air. Should the gas in the torch line return into the hydraulic valve its pressure will lower the level of water in the cylinder by forcing some of the liquid up into the two pipes. As the level of the water lowers, the shorter pipe will be uncovered first, and as this is the pipe leading to the open air the gas will be allowed to escape, while the pipe leading back to the generator is still closed by the water seal. As soon as this reverse flow ceases, the water will again resume its level and the action will continue. Because of the small amount of water blown out of the escape pipe each time the valve is called upon to perform this duty, it is necessary to see that the correct water level is always maintained.

While there are modifications of this construction, the same principle is used in all types. The pressure escape valve is often attached to this hydraulic valve body.

Construction Details.—Flexible tubing (except at torches), swing pipe joints, springs, mechanical check valves, chains, pulleys and lead or fusible piping should never be used on acetylene apparatus except where the failure of those parts will not affect the safety of the machine or permit, either directly or indirectly, the escape of gas into a room. Floats should not be used except where failure will only render the machine inoperative.

It should be said that the National Board of Fire Underwriters have established an inspection service for acetylene generators and any apparatus which bears their label, stating that that particular model and type has been passed, is safe to use. This service is for the best interests of all concerned and looks toward the prevention of accidents. Such inspection is a very important and desirable feature of any outfit and should be insisted upon.

Location of Generators.—Generators should preferably be placed outside of insured buildings and in properly constructed generator houses. The operating mechanism should have ample room to work in and there should be room enough for the attendant to reach the various parts and perform the required duties without hindrance or the need of artificial light. They should also be protected from tampering by unauthorized persons.

Generator houses should not be within five feet of any opening into, nor have any opening toward, any adjacent building, and should be kept under lock and key. The size of the house should be no greater than called for by the requirements mentioned above and it should be well ventilated.

The foundation for the generator itself should be of brick, stone, concrete or iron, if possible. If of wood, they should be extra heavy, located in a dry place and open to circulation of air. A board platform is not satisfactory, but the foundation should be of heavy planking or timber to make a firm base and so that the air can circulate around the wood.

The generator should stand level and no strain should be placed on any of the pipes or connections or any parts of the generator proper.



CHAPTER IV

WELDING INSTRUMENTS

VALVES

Tank Valves.—The acetylene tank valve is of the needle type, fitted with suitable stuffing box nuts and ending in an exposed square shank to which the special wrench may be fitted when the valve is to be opened or closed.

The valve used on Linde oxygen cylinders is also a needle type, but of slightly more complex construction. The body of the valve, which screws into the top of the cylinder, has an opening below through which the gas comes from the cylinder, and another opening on the side through which it issues to the torch line. A needle screws down from above to close this lower opening. The needle which closes the valve is not connected directly to the threaded member, but fits loosely into it. The threaded part is turned by a small hand wheel attached to the upper end. When this hand wheel is turned to the left, or up, as far as it will go, opening the valve, a rubber disc is compressed inside of the valve body and this disc serves to prevent leakage of the gas around the spindle.

The oxygen valve also includes a safety nut having a small hole through it closed by a fusible metal which melts at 250 Fahrenheit. Melting of this plug allows the gas to exert its pressure against a thin copper diaphragm, this diaphragm bursting under the gas pressure and allowing the oxygen to escape into the air.

The hand wheel and upper end of the valve mechanism are protected during shipment by a large steel cap which covers them when screwed on to the end of the cylinder. This cap should always be in place when tanks are received from the makers or returned to them.



Regulating Valves.—While the pressure in the gas containers may be anything from zero to 1,800 pounds, and will vary as the gas is withdrawn, the pressure of the gas admitted to the torch must be held steady and at a definite point. This is accomplished by various forms of automatic regulating valves, which, while they differ somewhat in details of construction, all operate on the same principle.

The regulator body (Figure 14) carries a union which attaches to the side outlet on the oxygen tank valve. The gas passes through this union, following an opening which leads to a large gauge which registers the pressure on the oxygen remaining in the tank and also to a very small opening in the end of a tube. The gas passes through this opening and into the interior of the regulator body. Inside of the body is a metal or rubber diaphragm placed so that the pressure of the incoming gas causes it to bulge slightly. Attached to the diaphragm is a sleeve or an arm tipped with a small piece of fibre, the fibre being placed so that it is directly opposite the small hole through which the gas entered the diaphragm chamber. The slight movement of the diaphragm draws the fibre tightly over the small opening through which the gas is entering, with the result that further flow is prevented.

Against the opposite side of the diaphragm is the end of a plunger. This plunger is pressed against the diaphragm by a coiled spring. The tension on the coiled spring is controlled by the operator through a threaded spindle ending in a wing or milled nut on the outside of the regulator body. Screwing in on the nut causes the tension on the spring to increase, with a consequent increase of pressure on the side of the diaphragm opposite to that on which the gas acts. Inasmuch as the gas pressure acted to close the small gas opening and the spring pressure acts in the opposite direction from the gas, it will be seen that the spring pressure tends to keep the valve open.

When the nut is turned way out there is of course, no pressure on the spring side of the diaphragm and the first gas coming through automatically closes the opening through which it entered. If now the tension on the spring be slightly increased, the valve will again open and admit gas until the pressure of gas within the regulator is just sufficient to overcome the spring pressure and again close the opening. There will then be a pressure of gas within the regulator that corresponds to the pressure placed on the spring by the operator. An opening leads from the regulator interior to the torch lines so that all gas going to the torches is drawn from the diaphragm chamber.

Any withdrawal of gas will, of course, lower the pressure of that remaining inside the regulator. The spring tension, remaining at the point determined by the operator, will overcome this lessened pressure of the gas, and the valve will again open and admit enough more gas to bring the pressure back to the starting point. This action continues as long as the spring tension remains at this point and as long as any gas is taken from the regulator. Increasing the spring tension will require a greater gas pressure to close the valve and the pressure of that in the regulator will be correspondingly higher.

When the regulator is not being used, the hand nut should be unscrewed until no tension remains on the spring, thus closing the valve. After the oxygen tank valve is open, the regulator hand nut is slowly screwed in until the spring tension is sufficient to give the required pressure in the torch lines. Another gauge is attached to the regulator so that it communicates with the interior of the diaphragm chamber, this gauge showing the gas pressure going to the torch. It is customary to incorporate a safety valve in the regulator which will blow off at a dangerous pressure.

In regulating valves and tank valves, as well as all other parts with which the oxygen comes in contact, it is not permissible to use any form of oil or grease because of danger of ignition and explosion. The mechanism of a regulator is too delicate to be handled in the ordinary shop and should any trouble or leakage develop in this part of the equipment it should be sent to a company familiar with this class of work for the necessary repairs. Gas must never be admitted to a regulator until the hand nut is all the way out, because of danger to the regulator itself and to the operator as well. A regulator can only be properly adjusted when the tank valve and torch valves are fully opened.



Acetylene regulators are used in connection with tanks of compressed gas. They are built on exactly the same lines as the oxygen regulating valve and operate in a similar way. One gauge only, the low pressure indicator, is used for acetylene regulators, although both high and low pressure may be used if desired. (See Figure 15.)

TORCHES

Flame is always produced by the combustion of a gas with oxygen and in no other way. When we burn oil or candles or anything else, the material of the fuel is first turned to a gas by the heat and is then burned by combining with the oxygen of the air. If more than a normal supply of air is forced into the flame, a greater heat and more active burning follows. If the amount of air, and consequently oxygen, is reduced, the flame becomes smaller and weaker and the combustion is less rapid. A flame may be easily extinguished by shutting off all of its air supply.

The oxygen of the combustion only forms one-fifth of the total volume of air; therefore, if we were to supply pure oxygen in place of air, and in equal volume, the action would be several times as intense. If the oxygen is mixed with the fuel gas in the proportion that burns to the very best advantage, the flame is still further strengthened and still more heat is developed because of the perfect combustion. The greater the amount of fuel gas that can be burned in a certain space and within a certain time, the more heat will be developed from that fuel.

The great amount of heat contained in acetylene gas, greater than that found in any other gaseous fuel, is used by leading this gas to the oxy-acetylene torch and there combining it with just the right amount of oxygen to make a flame of the greatest power and heat than can possibly be produced by any form of combustion of fuels of this kind. The heat developed by the flame is about 6300 Fahrenheit and easily melts all the metals, as well as other solids.

Other gases have been and are now being used in the torch. None of them, however, produce the heat that acetylene does, and therefore the oxy-acetylene process has proved the most useful of all. Hydrogen was used for many years before acetylene was introduced in this field. The oxy-hydrogen flame develops a heat far below that of oxy-acetylene, namely 4500 Fahrenheit. Coal gas, benzine gas, blaugas and others have also been used in successful applications, but for the present we will deal exclusively with the acetylene fuel.

It was only with great difficulty that the obstacles in the way of successfully using acetylene were overcome by the development of practicable controlling devices and torches, as well as generators. At present the oxy-acetylene process is the most universally adaptable, and probably finds the most widely extended field of usefulness of any welding process.

The theoretical proportion of the gases for perfect combustion is two and one-half volumes of oxygen to one of acetylene. In practice this proportion is one and one-eighth or one and one-quarter volumes of oxygen to one volume of acetylene, so that the cost is considerably reduced below what it would be if the theoretical quantity were really necessary, as oxygen costs much more than acetylene in all cases.

While the heat is so intense as to fuse anything brought into the path of the flame, it is localized in the small "welding cone" at the torch tip so that the torch is not at all difficult to handle without special protection except for the eyes, as already noted. The art of successful welding may be acquired by any operator of average intelligence within a reasonable time and with some practice. One trouble met with in the adoption of this process has been that the operation looks so simple and so easy of performance that unskilled and unprepared persons have been tempted to try welding, with results that often caused condemnation of the process, when the real fault lay entirely with the operator.

The form of torch usually employed is from twelve to twenty-four inches long and is composed of a handle at one end with tubes leading from this handle to the "welding head" or torch proper. At or near one end of the handle are adjustable cocks or valves for allowing the gases to flow into the torch or to prevent them from doing so. These cocks are often used for regulating the pressure and amount of gas flowing to the welding head, but are not always constructed for this purpose and should not be so used when it is possible to secure pressure adjustment at the regulators (Figure 16).

Figure 16 shows three different sizes of torches. The number 5 torch is designed especially for jewelers' work and thin sheet steel welding. It is eleven inches in length and weighs nineteen ounces. The tips for the number 10 torch are interchangeable with the number 5. The number 10 torch is adapted for general use on light and medium heavy work. It has six tips and its length is sixteen inches, with a weight of twenty-three ounces.

The number 15 torch is designed for heavy work, being twenty-five inches in length, permitting the operator to stand away from the heat of the metal being worked. These heavy tips are in two parts, the oxygen check being renewable.



Figures 17 and 18 show two sizes of another welding torch. Still another type is shown in Figure 19 with four interchangeable tips, the function of each being as follows:

No. 1. For heavy castings. No. 2. Light castings and heavy sheet metal. No. 3. Light sheet metal. No. 4. Very light sheet metal and wire.



At the side of the shut off cock away from the torch handle the gas tubes end in standard forms of hose nozzles, to which the rubber hose from the gas supply tanks or generators can be attached. The tubes from the handle to the head may be entirely separate from each other, or one may be contained within the other. As a general rule the upper one of two separate tubes carries the oxygen, while this gas is carried in the inside tube when they are concentric with each other.

In the welding head is the mixing chamber designed to produce an intimate mixture of the two gases before they issue from the nozzle to the flame. The nozzle, or welding tip, of a suitable size are design for the work to be handled and the pressure of gases being used, is attached to the welding head and consists essentially of the passage at the outer end of which the flame appears.

The torch body and tubes are usually made of brass, although copper is sometimes used. The joint must be very strong, and are usually threaded and soldered with silver solder. The nozzle proper is made from copper, because it withstands the heat of the flame better than other less suitable metals. The torch must be built in such a way that it is not at all liable to come apart under the influence of high temperatures.

All torches are constructed in such a way that it is impossible for the gases to mix by any possible chance before they reach the head, and the amount of gas contained in the head and tip after being mixed is made as small as possible. In order to prevent the return of the flame through the acetylene tube under the influence of the high pressure oxygen some form of back flash preventer is usually incorporated in the torch at or near the point at which the acetylene enters. This preventer takes the form of some porous and heat absorbing material, such as aluminum shavings, contained in a small cavity through which the gas passes on its way to the head.

High Pressure Torches.—Torches are divided into the same classes as are the generators; that is, high pressure, medium pressure and low pressure. As mentioned before, the medium pressure is usually called the high pressure, because there are very few true high pressure systems in use, and comparatively speaking the medium pressure type is one of high pressure.



With a true high pressure torch (Figure 20) the gases are used at very nearly equal heads so that the mixing before ignition is a simple matter. This type admits the oxygen at the inner end of a straight passage leading to the tip of the nozzle. The acetylene comes into this same passage from openings at one side and near the inner end. The difference in direction of the two gases as they enter the passage assists in making a homogeneous mixture. The construction of this nozzle is perfectly simple and is easily understood. The true high pressure torch nozzle is only suited for use with compressed and dissolved acetylene, no other gas being at a sufficient pressure to make the action necessary in mixing the gases.

Medium Pressure Torches.—The medium pressure (usually called high pressure) torch (Figure 21) uses acetylene from a medium pressure generator or from tanks of compressed gas, but will not take the acetylene from low pressure generators.



The construction of the mixing chamber and nozzle is very similar to that of the high pressure torch, the gases entering in the same way and from the same positions of openings. The pressure of the acetylene is but little lower than that of the oxygen, and the two gases, meeting at right angles, form a very intimate mixture at this point of juncture. The mixture in its proportions of gases depends entirely on the sizes of the oxygen and acetylene openings into the mixing chamber and on the pressures at which the gases are admitted. There is a very slight injector action as the fast moving stream of oxygen tends to draw the acetylene from the side openings into the chamber, but the operation of the torch does not depend on this action to any extent.

Low Pressure Torches.—The low pressure torch (Figure 22) will use gas from low pressure generators from medium pressure machines or from tanks in which it has been compressed and dissolved. This type depends for a perfect mixture of gas upon the principle of the injector just as it is applied in steam boiler practice.



The oxygen enters the head at considerable pressure and passes through its tube to a small jet within the head. The opening of this jet is directly opposite the end of the opening through the nozzle which forms the mixing chamber and the path of the gases to the flame. A small distance remains between the opening from which the oxygen issues and the inner opening into the mixing passage. The stream of oxygen rushes across this space and enters the mixing chamber, being driven by its own pressure.

The acetylene enters the head in an annular space surrounding the oxygen tube. The space between oxygen jet and mixing chamber opening is at one end of this acetylene space and the stream of oxygen seizes the acetylene and under the injector action draws it into the mixing chamber, it being necessary only to have a sufficient supply of acetylene flowing into the head to allow the oxygen to draw the required proportion for a proper mixture.

The volume of gas drawn into the mixing chamber depends on the size of the injector openings and the pressure of the oxygen. In practice the oxygen pressure is not altered to produce different sized flames, but a new nozzle is substituted which is designed to give the required flame. Each nozzle carries its own injector, so that the design is always suited to the conditions. While torches are made having the injector as a permanent part of the torch body, the replaceable nozzle is more commonly used because it makes the one torch suitable for a large range of work and a large number of different sized flames. With the replaceable head a definite pressure of oxygen is required for the size being used, this pressure being the one for which the injector and corresponding mixing chamber were designed in producing the correct mixture.

Adjustable Injectors.-Another form of low pressure torch operates on the injector principle, but the injector itself is a permanent part of the torch, the nozzle only being changed for different sizes of work and flame. The injector is placed in or near the handle and its opening is the largest required by any work that can be handled by this particular torch. The opening through the tip of the injector through which the oxygen issues on its way to the mixing chamber may be wholly or partly closed by a needle valve which may be screwed into the opening or withdrawn from it, according to the operator's judgment. The needle valve ends in a milled nut outside the torch handle, this being the adjustment provided for the different nozzles.

Torch Construction.—A well designed torch is so designed that the weight distribution is best for holding it in the proper position for welding. When a torch is grasped by its handle with the gas hose attached, it should balance so that it does not feel appreciably heavier on one end than on the other.

The head and nozzle may be placed so that the flame issues in a line at right angles with the torch body, or they may be attached at an angle convenient for the work to be done. The head set at an angle of from 120 to 170 degrees with the body is usually preferred for general work in welding, while the cutting torch usually has its head at right angles to the body.

Removable nozzles have various size openings through them and the different sizes are designated by numbers from 1 up. The same number does not always indicate the same size opening in torches of different makes, nor does it indicate a nozzle of the same capacity.

The design of the nozzle, the mixing chamber, the injector, when one is used, and the size of the gas openings must be such that all these things are suited to each other if a proper mixture of gas is to be secured. Parts that are not made to work together are unsafe if used because of the danger of a flash back of the flame into the mixing chamber and gas tubes. It is well known that flame travels through any inflammable gas at a certain definite rate of speed, depending on the degree of inflammability of the gas. The easier and quicker the gas burns, the faster will the flame travel through it.

If the gas in the nozzle and mixing chamber stood still, the flame would immediately travel back into these parts and produce an explosion of more or less violence. The speed with which the gases issue from the nozzle prevent this from happening because the flame travels back through the gas at the same speed at which the gas issues from the torch tip. Should the velocity of the gas be greater than the speed of flame propagation through it, it will be impossible to keep the flame at the tip, the tendency being for a space of unburned gas to appear between tip and flame. On the other hand, should the speed of the flame exceed the velocity with which the gas comes from the torch there will result a flash back and explosion.

Care of Torches.—An oxy-acetylene torch is a very delicate and sensitive device, much more so that appears on the surface. It must be given equally as good care and attention as any other high-priced piece of machinery if it is to be maintained in good condition for use.

It requires cleaning of the nozzles at regular intervals if used regularly. This cleaning is accomplished with a piece of copper or brass wire run through the opening, and never with any metal such as steel or iron that is harder than the nozzle itself, because of the danger of changing the size of the openings. The torch head and nozzle can often be cleaned by allowing the oxygen to blow through at high pressure without the use of any tools.

In using a torch a deposit of carbon will gradually form inside of the head, and this deposit will be more rapid if the operator lights the stream of acetylene before turning any oxygen into the torch. This deposit may be removed by running kerosene through the nozzle while it is removed from the torch, setting fire to the kerosene and allowing oxygen to flow through while the oil is burning.

Should a torch become clogged in the head or tubes, it may usually be cleaned by removing the oxygen hose from the handle end, closing the acetylene cock on the torch, placing the end of the oxygen hose over the opening in the nozzle and turning on the oxygen under pressure to blow the obstruction back through the passage that it has entered. By opening the acetylene cock and closing the oxygen cock at the handle, the acetylene passages may then be cleaned in the same way. Under no conditions should a torch be taken apart any more than to remove the changeable nozzle, except in the hands of those experienced in this work.

Nozzle Sizes.—The size of opening through the nozzle is determined according to the thickness and kind of metal being handled. The following sizes are recommended for steel:

Davis-Bournonville. Oxweld Low Thickness of Metal (Medium Pressure.) Pressure 1/32 Tip No. 1 Head No. 2 1/16 2 5/64 3 3/32 3 4 3/8 4 5 3/16 5 6 1/4 6 7 5/16 7 3/8 8 8 1/2 9 10 5/8 10 12 3/4 11 15 Very heavy 12 15

Cutting Torches.—Steel may be cut with a jet of oxygen at a rate of speed greater than in any other practicable way under usual conditions. The action consists of burning away a thin section of the metal by allowing a stream of oxygen to flow onto it while the gas is at high pressure and the metal at a white heat.



The cutting torch (Figure 23) has the same characteristics as the welding torch, but has an additional nozzle or means for temporarily using the welding opening for the high pressure oxygen. The oxygen issues from the opening while cutting at a pressure of from ten to 100 pounds to the square inch.

The work is first heated to a white heat by adjusting the torch for a welding flame. As soon as the metal reaches this temperature, the high pressure oxygen is turned on to the white-hot portion of the steel. When the jet of gas strikes the metal it cuts straight through, leaving a very narrow slot and removing but little metal. Thicknesses of steel up to ten inches can be economically handled in this way.

The oxygen nozzle is usually arranged so that it is surrounded by a number of small jets for the heating flame. It will be seen that this arrangement makes the heating flame always precede the oxygen jet, no matter in which direction the torch is moved.

The torch is held firmly, either by hand or with the help of special mechanism for guiding it in the desired path, and is steadily advanced in the direction it is desired to extend the cut, the rate of advance being from three inches to two feet per minute through metal from nine inches down to one-quarter of an inch in thickness.

The following data on cutting is given by the Davis-Bournonville Company:

Cubic Feet Cost of Thickness of Gas Inches Gases of Cutting Heating per Foot Oxygen Cut per per Foot Steel Oxygen Oxygen of Cut Acetylene Min. of Cut 1/4 10 lbs. 4 lbs. .40 .086 24 $ .013 1/2 20 4 .91 .150 15 .029 3/4 30 4 1.16 .150 15 .036 1 30 4 1.45 .172 12 .045 1 1/2 30 5 2.40 .380 12 .076 2 40 5 2.96 .380 12 .093 4 50 5 9.70 .800 7 .299 6 70 6 21.09 1.50 4 .648 9 100 6 43.20 2.00 3 1.311

Acetylene-Air Torch.—A form of torch which burns the acetylene after mixing it with atmospheric air at normal pressure rather than with the oxygen under higher pressures has been found useful in certain pre-heating, brazing and similar operations. This torch (Figure 24) is attached by a rubber gas hose to any compressed acetylene tank and is regulated as to flame size and temperature by opening or closing the tank valve more or less.

After attaching the torch to the tank, the gas is turned on very slowly and is lighted at the torch tip. The adjustment should cause the presence of a greenish-white cone of flame surrounded by a larger body of burning gas, the cone starting at the mouth of the torch.



By opening the tank valve more, a longer and hotter flame is produced, the length being regulated by the tank valve also. This torch will give sufficient heat to melt steel, although not under conditions suited to welding. Because of the excess of acetylene always present there is no danger of oxidizing the metal being heated.

The only care required by this torch is to keep the small air passages at the nozzle clean and free from carbon deposits. The flame should be extinguished when not in use rather than turned low, because this low flame rapidly deposits large quantities of soot in the burner.



CHAPTER V

OXY-ACETYLENE WELDING PRACTICE

PREPARATION OF WORK

Preheating.—The practice of heating the metal around the weld before applying the torch flame is a desirable one for two reasons. First, it makes the whole process more economical; second, it avoids the danger of breakage through expansion and contraction of the work as it is heated and as it cools.

When it is desired to join two surfaces by welding them, it is, of course, necessary to raise the metal from the temperature of the surrounding air to its melting point, involving an increase in temperature of from one thousand to nearly three thousand degrees. To obtain this entire increase of temperature with the torch flame is very wasteful of fuel and of the operator's time. The total amount of heat necessary to put into metal is increased by the conductivity of that metal because the heat applied at the weld is carried to other parts of the piece being handled until the whole mass is considerably raised in temperature. To secure this widely distributed increase the various methods of preheating are adopted.

As to the second reason for preliminary heating. It is understood that the metal added to the joint is molten at the time it flows into place. All the metals used in welding contract as they cool and occupy a much smaller space than when molten. If additional metal is run between two adjoining surfaces which are parts of a surrounding body of cool metal, this added metal will cool while the surfaces themselves are held stationary in the position they originally occupied. The inevitable result is that the metal added will crack under the strain, or, if the weld is exceptionally strong, the main body of the work will he broken by the force of contraction. To overcome these difficulties is the second and most important reason for preheating and also for slow cooling following the completion of the weld.

There are many ways of securing this preheating. The work may be brought to a red heat in the forge if it is cast iron or steel; it may he heated in special ovens built for the purpose; it may be placed in a bed of charcoal while suitably supported; it may be heated by gas or gasoline preheating torches, and with very small work the outer flame of the welding torch automatically provides means to this end.

The temperature of the parts heated should be gradually raised in all cases, giving the entire mass of metal a chance to expand equally and to adjust itself to the strains imposed by the preheating. After the region around the weld has been brought to a proper temperature the opening to be filled is exposed so that the torch flame can reach it, while the remaining surfaces are still protected from cold air currents and from cooling through natural radiation.

One of the commonest methods and one of the best for handling work of rather large size is to place the piece to be welded on a bed of fire brick and build a loose wall around it with other fire brick placed in rows, one on top of the other, with air spaces left between adjacent bricks in each row. The space between the brick retaining wall and the work is filled with charcoal, which is lighted from below. The top opening of the temporary oven is then covered with asbestos and the fire kept up until the work has been uniformly raised in temperature to the desired point.

When much work of the same general character and size is to be handled, a permanent oven may be constructed of fire brick, leaving a large opening through the top and also through one side. Charcoal may be used in this form of oven as with the temporary arrangement, or the heat may be secured from any form of burner or torch giving a large volume of flame. In any method employing flame to do the heating, the work itself must be protected from the direct blast of the fire. Baffles of brick or metal should be placed between the mouth of the torch and the nearest surface of the work so that the flame will be deflected to either side and around the piece being heated.

The heat should be applied to bring the point of welding to the highest temperature desired and, except in the smallest work, the heat should gradually shade off from this point to the other parts of the piece. In the case of cast iron and steel the temperature at the point to be welded should be great enough to produce a dull red heat. This will make the whole operation much easier, because there will be no surrounding cool metal to reduce the temperature of the molten material from the welding rod below the point at which it will join the work. From this red heat the mass of metal should grow cooler as the distance from the weld becomes greater, so that no great strain is placed upon any one part. With work of a very irregular shape it is always best to heat the entire piece so that the strains will be so evenly distributed that they can cause no distortion or breakage under any conditions.

The melting point of the work which is being preheated should be kept in mind and care exercised not to approach it too closely. Special care is necessary with aluminum in this respect, because of its low melting temperature and the sudden weakening and flowing without warning. Workmen have carelessly overheated aluminum castings and, upon uncovering the piece to make the weld, have been astonished to find that it had disappeared. Six hundred degrees is about the safe limit for this metal. It is possible to gauge the exact temperature of the work with a pyrometer, but when this instrument cannot be procured, it might be well to secure a number of "temperature cones" from a chemical or laboratory supply house. These cones are made from material that will soften at a certain heat and in form they are long and pointed. Placed in position on the part being heated, the point may be watched, and when it bends over it is sure that the metal itself has reached a temperature considerably in excess of the temperature at which that particular cone was designed to soften.

The object in preheating the metal around the weld is to cause it to expand sufficiently to open the crack a distance equal to the contraction when cooling from the melting point. In the case of a crack running from the edge of a piece into the body or of a crack wholly within the body, it is usually satisfactory to heat the metal at each end of the opening. This will cause the whole length of the crack to open sufficiently to receive the molten material from the rod.

The judgment of the operator will be called upon to decide just where a piece of metal should be heated to open the weld properly. It is often possible to apply the preheating flame to a point some distance from the point of work if the parts are so connected that the expansion of the heated part will serve to draw the edges of the weld apart. Whatever part of the work is heated to cause expansion and separation, this part must remain hot during the entire time of welding and must then cool slowly at the same time as the metal in the weld cools.



An example of heating points away from the crack might be found in welding a lattice work with one of the bars cracked through (Figure 25). If the strips parallel and near to the broken bar are heated gradually, the work will be so expanded that the edges of the break are drawn apart and the weld can be successfully made. In this case, the parallel bars next to the broken one would be heated highest, the next row not quite so hot and so on for some distance away. If only the one row were heated, the strains set up in the next ones would be sufficient to cause a new break to appear.



If welding is to be done near the central portion of a large piece, the strains will be brought to bear on the parts farthest away from the center. Should a fly wheel spoke be broken and made ready to weld, the greatest strain will come on the rim of the wheel. In cases like this it is often desirable to cut through at the point of greatest strain with a saw or cutting torch, allowing free movement while the weld is made at the original break (Figure 26). After the inside weld is completed, the cut may be welded without danger, for the reason that it will always be at some point at which severe strains cannot be set up by the contraction of the cooling metal.



In materials that will spring to some extent without breakage, that is, in parts that are not brittle, it may be possible to force the work out of shape with jacks or wedges (Figure 27) in the same way that it would be distorted by heating and expanding some portion of it as described. A careful examination will show whether this method can be followed in such a way as to force the edges of the break to separate. If the plan seems feasible, the wedges may be put in place and allowed to remain while the weld is completed. As soon as the work is finished the wedges should be removed so that the natural contraction can take place without damage.

It should always be remembered that it is not so much the expansion of the work when heated as it is the contraction caused by cooling that will do the damage. A weld may be made that, to all appearances, is perfect and it may be perfect when completed; but if provision has not been made to allow for the contraction that is certain to follow, there will be a breakage at some point. It is not possible to weld the simplest shapes, other than straight bars, without considering this difficulty and making provision to take care of it.

The exact method to employ in preheating will always call for good judgment on the part of the workman, and he should remember that the success or failure of his work will depend fully as much on proper preparation as on correct handling of the weld itself. It should be remembered that the outer flame of the oxy-acetylene torch may be depended on for a certain amount of preheating, as this flame gives a very large volume of heat, but a heat that is not so intense nor so localized as the welding flame itself. The heat of this part of the flame should be fully utilized during the operation of melting the metal and it should be so directed, when possible, that it will bring the parts next to be joined to as high a temperature as possible.

When the work has been brought to the desired temperature, all parts except the break and the surface immediately surrounding it on both sides should be covered with heavy sheet asbestos. This protecting cover should remain in place throughout the operation and should only be moved a distance sufficient to allow the torch flame to travel in the path of the weld. The use of asbestos in this way serves a twofold purpose. It retains the heat in the work and prevents the breakage that would follow if a draught of air were to strike the heated metal, and it also prevents such a radiation of heat through the surrounding air as would make it almost impossible for the operator to perform his work, especially in the case of large and heavy castings when the amount of heat utilized is large.

Cleaning and Champfering.—A perfect weld can never be made unless the surfaces to be joined have been properly prepared to receive the new metal.

All spoiled, burned, corroded and rough particles must positively be removed with chisel and hammer and with a free application of emery cloth and wire brush. The metal exposed to the welding flame should be perfectly clean and bright all over, or else the additional material will not unite, but will only stick at best.



Following the cleaning it is always necessary to bevel, or champfer, the edges except in the thinnest sheet metal. To make a weld that will hold, the metal must be made into one piece, without holes or unfilled portions at any point, and must be solid from inside to outside. This can only be accomplished by starting the addition of metal at one point and gradually building it up until the outside, or top, is reached. With comparatively thin plates the molten metal may be started from the side farthest from the operator and brought through, but with thicker sections the addition is started in the middle and brought flush with one side and then with the other.

It will readily be seen that the molten material cannot be depended upon to flow between the tightly closed surfaces of a crack in a way that can be at all sure to make a true weld. It will be necessary for the operator to reach to the farthest side with the flame and welding rod, and to start the new surfaces there. To allow this, the edges that are to be joined are beveled from one side to the other (Figure 28), so that when placed together in approximately the position they are to occupy they will leave a grooved channel between them with its sides at an angle with each other sufficient in size to allow access to every point of each surface.



With work less than one-fourth inch thick, this angle should be forty-five degrees on each piece (Figure 29), so that when they are placed together the extreme edges will meet at the bottom of a groove whose sides are square, or at right angles, to each other. This beveling should be done so that only a thin edge is left where the two parts come together, just enough points in contact to make the alignment easy to hold. With work of a thickness greater than a quarter of an inch, the angle of bevel on each piece may be sixty degrees (Figure 30), so that when placed together the angle included between the sloping sides will also be sixty degrees. If the plate is less than one-eighth of an inch thick the beveling is not necessary, as the edges may be melted all the way through without danger of leaving blowholes at any point.



This beveling may be done in any convenient way. A chisel is usually most satisfactory and also quickest. Small sections may be handled by filing, while metal that is too hard to cut in either of these ways may be shaped on the emery wheel. It is not necessary that the edges be perfectly finished and absolutely smooth, but they should be of regular outline and should always taper off to a thin edge so that when the flame is first applied it can be seen issuing from the far side of the crack. If the work is quite thick and is of a shape that will allow it to be turned over, the bevel may be brought from both sides (Figure 31), so that there will be two grooves, one on each surface of the work. After completing the weld on one side, the piece is reversed and finished on the other side. Figure 32 shows the proper beveling for welding pipe. Figure 33 shows how sheet metal may be flanged for welding.

Welding should not be attempted with the edges separated in place of beveled, because it will be found impossible to build up a solid web of new metal from one side clear through to the other by this method. The flame cannot reach the surfaces to make them molten while receiving new material from the rod, and if the flame does not reach them it will only serve to cause a few drops of the metal to join and will surely cause a weak and defective weld.



Supporting Work.—During the operation of welding it is necessary that the work be well supported in the position it should occupy. This may be done with fire brick placed under the pieces in the correct position, or, better still, with some form of clamp. The edges of the crack should touch each other at the point where welding is to start and from there should gradually separate at the rate of about one-fourth inch to the foot. This is done so that the cooling of the molten metal as it is added will draw the edges together by its contraction.

Care must be used to see that the work is supported so that it will maintain the same relative position between the parts as must be present when the work is finished. In this connection it must be remembered that the expansion of the metal when heated may be great enough to cause serious distortion and to provide against this is one of the difficulties to be overcome.

Perfect alignment should be secured between the separate parts that are to be joined and the two edges must be held up so that they will be in the same plane while welding is carried out. If, by any chance, one drops below the other while molten metal is being added, the whole job may have to be undone and done over again. One precaution that is necessary is that of making sure that the clamping or supporting does not in itself pull the work out of shape while melted.

TORCH PRACTICE



The weld is made by bringing the tip of the welding flame to the edges of the metals to be joined. The torch should be held in the right hand and moved slowly along the crack with a rotating motion, traveling in small circles (Figure 34), so that the Welding flame touches first on one side of the crack and then on the other. On large work the motion may be simply back and forth across the crack, advancing regularly as the metal unites. It is usually best to weld toward the operator rather than from him, although this rule is governed by circumstances. The head of the torch should be inclined at an angle of about 60 degrees to the surface of the work. The torch handle should extend in the same line with the break (Figure 35) and not across it, except when welding very light plates.



If the metal is 1/16 inch or less in thickness it is only necessary to circle along the crack, the metal itself furnishing enough material to complete the weld without additions. Heat both sides evenly until they flow together.

Material thicker than the above requires the addition of more metal of the same or different kind from the welding rod, this rod being held by the left hand. The proper size rod for cast iron is one having a diameter equal to the thickness of metal being welded up to a one-half inch rod, which is the largest used. For steel the rod should be one-half the thickness of the metal being joined up to one-fourth inch rod. As a general rule, better results will be obtained by the use of smaller rods, the very small sizes being twisted together to furnish enough material while retaining the free melting qualities.



The tip of the rod must at all times be held in contact with the pieces being welded and the flame must be so directed that the two sides of the crack and the end of the rod are melted at the same time (Figure 36). Before anything is added from the rod, the sides of the crack are melted down sufficiently to fill the bottom of the groove and join the two sides. Afterward, as metal comes from the rod in filling the crack, the flame is circled along the joint being made, the rod always following the flame.



Figure 37 illustrates the welding of pieces of unequal thickness.

Figure 38 illustrates welding at an angle.

The molten metal may be directed as to where it should go by the tip of the welding flame, which has considerable force, but care must be taken not to blow melted metal on to cooler surfaces which it cannot join. If, while welding, a spot appears which does not unite with the weld, it may be handled by heating all around it to a white heat and then immediately welding the bad place.



Never stop in the middle of a weld, as it is extremely difficult to continue smoothly when resuming work.

The Flame.—The welding flame must have exactly the right proportions of each gas. If there is too much oxygen, the metal will be burned or oxidized; the presence of too much acetylene carbonizes the metal; that is to say, it adds carbon and makes the work harder. Just the right mixture will neither burn nor carbonize and is said to be a "neutral" flame. The neutral flame, if of the correct size for the work, reduces the metal to a melted condition, not too fluid, and for a width about the same as the thickness of the metal being welded.

When ready to light the torch, after attaching the right tip or head as directed in accordance with the thickness of metal to be handled, it will be necessary to regulate the pressure of gases to secure the neutral flame.

The oxygen will have a pressure of from 2 to 20 pounds, according to the nozzle used. The acetylene will have much less. Even with the compressed gas, the pressure should never exceed 10 pounds for the largest work, and it will usually be from 4 to 6. In low pressure systems, the acetylene will be received at generator pressure. It should first be seen that the hand-screws on the regulators are turned way out so that the springs are free from any tension. It will do no harm if these screws are turned back until they come out of the threads. This must be done with both oxygen and acetylene regulators.

Next, open the valve from the generator, or on the acetylene tank, and carefully note whether there is any odor of escaping gas. Any leakage of this gas must be stopped before going on with the work.

The hand wheel controlling the oxygen cylinder valve should now be turned very slowly to the left as far as it will go, which opens the valve, and it should be borne in mind the pressure that is being released. Turn in the hand screw on the oxygen regulator until the small pressure gauge shows a reading according to the requirements of the nozzle being used. This oxygen regulator adjustment should be made with the cock on the torch open, and after the regulator is thus adjusted the torch cock may be closed.

Open the acetylene cock on the torch and screw in on the acetylene regulator hand-screw until gas commences to come through the torch. Light this flow of acetylene and adjust the regulator screw to the pressure desired, or, if there is no gauge, so that there is a good full flame. With the pressure of acetylene controlled by the type of generator it will only be necessary to open the torch cock.

With the acetylene burning, slowly open the oxygen cock on the torch and allow this gas to join the flame. The flame will turn intensely bright and then blue white. There will be an outer flame from four to eight inches long and from one to three inches thick. Inside of this flame will be two more rather distinctly defined flames. The inner one at the torch tip is very small, and the intermediate one is long and pointed. The oxygen should be turned on until the two inner flames unite into one blue-white cone from one-fourth to one-half inch long and one-eighth to one-fourth inch in diameter. If this single, clearly defined cone does not appear when the oxygen torch cock has been fully opened, turn off some of the acetylene until it does appear.

If too much oxygen is added to the flame, there will still be the central blue-white cone, but it will be smaller and more or less ragged around the edges (Figure 39). When there is just enough oxygen to make the single cone, and when, by turning on more acetylene or by turning off oxygen, two cones are caused to appear, the flame is neutral (Figure 40), and the small blue-white cone is called the welding flame.



While welding, test the correctness of the flame adjustment occasionally by turning on more acetylene or by turning off some oxygen until two flames or cones appear. Then regulate as before to secure the single distinct cone. Too much oxygen is not usually so harmful as too much acetylene, except with aluminum. (See Figure 41.) An excessive amount of sparks coming from the weld denotes that there is too much oxygen in the flame. Should the opening in the tip become partly clogged, it will be difficult to secure a neutral flame and the tip should be cleaned with a brass or copper wire—never with iron or steel tools or wire of any kind. While the torch is doing its work, the tip may become excessively hot due to the heat radiated from the molten metal. The tip may be cooled by turning off the acetylene and dipping in water with a slight flow of oxygen through the nozzle to prevent water finding its way into the mixing chamber.

The regulators for cutting are similar to those for welding, except that higher pressures may be handled, and they are fitted with gauges reading up to 200 or 250 pounds pressure.

In welding metals which conduct the heat very rapidly it is necessary to use a much larger nozzle and flame than for metals which have not this property. This peculiarity is found to the greatest extent in copper, aluminum and brass.

Should a hole be blown through the work, it may be closed by withdrawing the flame for a few seconds and then commencing to build additional metal around the edges, working all the way around and finally closing the small opening left at the center with a drop or two from the welding rod.

WELDING VARIOUS METALS

Because of the varying melting points, rates of expansion and contraction, and other peculiarities of different metals, it is necessary to give detailed consideration to the most important ones.

Characteristics of Metals.—The welder should thoroughly understand the peculiarities of the various metals with which he has to deal. The metals and their alloys are described under this heading in the first chapter of this book and a tabulated list of the most important points relating to each metal will be found at the end of the present chapter. All this information should be noted by the operator of a welding installation before commencing actual work.

Because of the nature of welding, the melting point of a metal is of great importance. A metal melting at a low temperature should have more careful treatment to avoid undesired flow than one which melts at a temperature which is relatively high. When two dissimilar metals are to be joined, the one which melts at the higher temperature must be acted upon by the flame first and when it is in a molten condition the heat contained in it will in many cases be sufficient to cause fusion of the lower melting metal and allow them to unite without playing the flame on the lower metal to any great extent.

The heat conductivity bears a very important relation to welding, inasmuch as a metal with a high rate of conductance requires more protection from cooling air currents and heat radiation than one not having this quality to such a marked extent. A metal which conducts heat rapidly will require a larger volume of flame, a larger nozzle, than otherwise, this being necessary to supply the additional heat taken away from the welding point by this conductance.

The relative rates of expansion of the various metals under heat should be understood in order that parts made from such material may have proper preparation to compensate for this expansion and contraction. Parts made from metals having widely varying rates of expansion must have special treatment to allow for this quality, otherwise breakage is sure to occur.

Cast Iron.—All spoiled metal should he cut away and if the work is more than one-eighth inch in thickness the sides of the crack should be beveled to a 45 degree angle, leaving a number of points touching at the bottom of the bevel so that the work may be joined in its original relation.

The entire piece should be preheated in a bricked-up oven or with charcoal placed on the forge, when size does not warrant building a temporary oven. The entire piece should be slowly heated and the portion immediately surrounding the weld should be brought to a dull red. Care should be used that the heat does not warp the metal through application to one part more than the others. After welding, the work should be slowly cooled by covering with ashes, slaked lime, asbestos fibre or some other non-conductor of heat. These precautions are absolutely essential in the case of cast iron.

A neutral flame, from a nozzle proportioned to the thickness of the work, should be held with the point of the blue-white cone about one-eighth inch from the surface of the iron.

A cast iron rod of correct diameter, usually made with an excess of silicon, is used by keeping its end in contact with the molten metal and flowing it into the puddle formed at the point of fusion. Metal should be added so that the weld stands about one-eighth inch above the surrounding surface of the work.

Various forms of flux may be used and they are applied by dipping the end of the welding rod into the powder at intervals. These powders may contain borax or salt, and to prevent a hard, brittle weld, graphite or ferro-silicon may be added. Flux should be added only after the iron is molten and as little as possible should be used. No flux should be used just before completion of the work.

The welding flame should be played on the work around the crack and gradually brought to bear on the work. The bottom of the bevel should be joined first and it will be noted that the cast iron tends to run toward the flame, but does not stick together easily. A hard and porous weld should be carefully guarded against, as described above, and upon completion of the work the welded surface should be scraped with a file, while still red hot, in order to remove the surface scale.

Malleable Iron.—This material should be beveled in the same way that cast iron is handled, and preheating and slow cooling are equally desirable. The flame used is the same as for cast iron and so is the flux. The welding rod may be of cast iron, although better results are secured with Norway iron wire or else a mild steel wire wrapped with a coil of copper wire.

It will be understood that malleable iron turns to ordinary cast iron when melted and cooled. Welds in malleable iron are usually far from satisfactory and a better joint is secured by brazing the edges together with bronze. The edges to be joined are brought to a heat just a little below the point at which they will flow and the opening is then quickly-filled from a rod of Tobin bronze or manganese bronze, a brass or bronze flux being used in this work.

Wrought Iron or Semi-Steel.—This metal should be beveled and heated in the same way as described for cast iron. The flame should be neutral, of the same size as for steel, and used with the tip of the blue-white cone just touching the work. The welding rod should be of mild steel, or, if wrought iron is to be welded to steel, a cast iron rod may be used. A cast iron flux is well suited for this work. It should be noted that wrought iron turns to ordinary cast iron if kept heated for any length of time.

Steel.—Steel should be beveled if more than one-eighth inch in thickness. It requires only a local preheating around the point to be welded. The welding flame should be absolutely neutral, without excess of either gas. If the metal is one-sixteenth inch or less in thickness, the tip of the blue-white cone must be held a short distance from the surface of the work; in all other cases the tip of this cone is touched to the metal being welded.

The welding rod may be of mild, low carbon steel or of Norway iron. Nickel steel rods may be used for parts requiring great strength, but vanadium alloys are very difficult to handle. A very satisfactory rod is made by twisting together two wires of the required material. The rod must be kept constantly in contact with the work and should not be added until the edges are thoroughly melted. The flux may or may not be used. If one is wanted, it may be made from three parts iron filings, six parts borax and one part sal ammoniac.

It will be noticed that the steel runs from the flame, but tends to hold together. Should foaming commence in the molten metal, it shows an excess of oxygen and that the metal is being burned.

High carbon steels are very difficult to handle. It is claimed that a drop or two of copper added to the weld will assist the flow, but will also harden the work. An excess of oxygen reduces the amount of carbon and softens the steel, while an excess of acetylene increases the proportion of carbon and hardens the metal. High speed steels may sometimes be welded if first coated with semi-steel before welding.

Aluminum.—This is the most difficult of the commonly found metals to weld. This is caused by its high rate of expansion and contraction and its liability to melt and fall away from under the flame. The aluminum seems to melt on the inside first, and, without previous warning, a portion of the work will simply vanish from in front of the operator's eyes. The metal tends to run from the flame and separate at the same time. To keep the metal in shape and free from oxide, it is worked or puddled while in a plastic condition by an iron rod which has been flattened at one end. Several of these rods should be at hand and may be kept in a jar of salt water while not being used. These rods must not become coated with aluminum and they must not get red hot while in the weld.

The surfaces to be joined, together with the adjacent parts, should be cleaned thoroughly and then washed with a 25 per cent solution of nitric acid in hot water, used on a swab. The parts should then be rinsed in clean water and dried with sawdust. It is also well to make temporary fire clay moulds back of the parts to be heated, so that the metal may be flowed into place and allowed to cool without danger of breakage.

Aluminum must invariably be preheated to about 600 degrees, and the whole piece being handled should be well covered with sheet asbestos to prevent excessive heat radiation.

The flame is formed with an excess of acetylene such that the second cone extends about an inch, or slightly more, beyond the small blue-white point. The torch should be held so that the end of this second cone is in contact with the work, the small cone ordinarily used being kept an inch or an inch and a half from the surface of the work.

Welding rods of special aluminum are used and must be handled with their end submerged in the molten metal of the weld at all times.

When aluminum is melted it forms alumina, an oxide of the metal. This alumina surrounds small masses of the metal, and as it does not melt at temperatures below 5000 degrees (while aluminum melts at about 1200), it prevents a weld from being made. The formation of this oxide is retarded and the oxide itself is dissolved by a suitable flux, which usually contains phosphorus to break down the alumina.

Copper.—The whole piece should be preheated and kept well covered while welding. The flame must be much larger than for the same thickness of steel and neutral in character. A slight excess of acetylene would be preferable to an excess of oxygen, and in all cases the molten metal should be kept enveloped with the flame. The welding rod is of copper which contains phosphorus; and a flux, also containing phosphorus, should be spread for about an inch each side of the joint. These assist in preventing oxidation, which is sure to occur with heated copper.

Copper breaks very easily at a heat slightly under the welding temperature and after cooling it is simply cast copper in all cases.

Brass and Bronze.—It is necessary to preheat these metals, although not to a very high temperature. They must be kept well covered at all times to prevent undue radiation. The flame should be produced with a nozzle one size larger than for the same thickness of steel and the small blue-white cone should be held from one-fourth to one-half inch above the surface of the work. The flame should be neutral in character.

A rod or wire of soft brass containing a large percentage of zinc is suitable for adding to brass, while copper requires the use of copper or manganese bronze rods. Special flux or borax may be used to assist the flow.

The emission of white smoke indicates that the zinc contained in these alloys is being burned away and the heat should immediately be turned away or reduced. The fumes from brass and bronze welding are very poisonous and should not be breathed.

RESTORATION OF STEEL

The result of the high heat to which the steel has been subjected is that it is weakened and of a different character than before welding. The operator may avoid this as much as possible by first playing the outer flame of the torch all over the surfaces of the work just completed until these faces are all of uniform color, after which the metal should be well covered with asbestos and allowed to cool without being disturbed. If a temporary heating oven has been employed, the work and oven should be allowed to cool together while protected with the sheet asbestos. If the outside air strikes the freshly welded work, even for a moment, the result will be breakage.

A weld in steel will always leave the metal with a coarse grain and with all the characteristics of rather low grade cast steel. As previously mentioned in another chapter, the larger the grain size in steel the weaker the metal will be, and it is the purpose of the good workman to avoid, as far as possible, this weakening.

The structure of the metal in one piece of steel will differ according to the heat that it has under gone. The parts of the work that have been at the melting point will, therefore, have the largest grain size and the least strength. Those parts that have not suffered any great rise in temperature will be practically unaffected, and all the parts between these two extremes will be weaker or stronger according to their distance from the weld itself. To restore the steel so that it will have the best grain size, the operator may resort to either of two methods: (1) The grain may be improved by forging. That means that the metal added to the weld and the surfaces that have been at the welding heat are hammered much as a blacksmith would hammer his finished work to give it greater strength. The hammering should continue from the time the metal first starts to cool until it has reached the temperature at which the grain size is best for strength. This temperature will vary somewhat with the composition of the metal being handled, but in a general way, it may be stated that the hammering should continue without intermission from the time the flame is removed from the weld until the steel just begins to show attraction for a magnet presented to it. This temperature of magnetic attraction will always be low enough and the hammering should be immediately discontinued at this point. (2) A method that is more satisfactory, although harder to apply, is that of reheating the steel to a certain temperature throughout its whole mass where the heat has had any effect, and then allowing slow and even cooling from this temperature. The grain size is affected by the temperature at which the reheating is stopped, and not by the cooling, yet the cooling should be slow enough to avoid strains caused by uneven contraction.

After the weld has been completed the steel must be allowed to cool until below 1200 Fahrenheit. The next step is to heat the work slowly until all those parts to be restored have reached a temperature at which the magnet just ceases to be attracted. While the very best temperature will vary according to the nature and hardness of the steel being handled, it will be safe to carry the heating to the point indicated by the magnet in the absence of suitable means of measuring accurately these high temperatures. In using a magnet for testing, it will be most satisfactory if it is an electromagnet and not of the permanent type. The electric current may be secured from any small battery and will be the means of making sure of the test. The permanent magnet will quickly lose its power of attraction under the combined action of the heat and the jarring to which it will be subjected.

In reheating the work it is necessary to make sure that no part reaches a temperature above that desired for best grain size and also to see that all parts are brought to this temperature. Here enters the greatest difficulty in restoring the metal. The heating may be done so slowly that no part of the work on the outside reaches too high a temperature and then keeps the outside at this heat until the entire mass is at the same temperature. A less desirable way is to heat the outside higher than this temperature and allow the conductivity of the metal to distribute the excess to the inside.

The most satisfactory method, where it can be employed, is to make use of a bath of some molten metal or some chemical mixture that can be kept at the exact heat necessary by means of gas fires that admit of close regulation. The temperature of these baths may be maintained at a constant point by watching a pyrometer, and the finished work may be allowed to remain in the bath until all parts have reached the desired temperature.

WELDING INFORMATION

The following tables include much of the information that the operator must use continually to handle the various metals successfully. The temperature scales are given for convenience only. The composition of various alloys will give an idea of the difficulties to be contended with by consulting the information on welding various metals. The remaining tables are of self-evident value in this work.

TEMPERATURE SCALES Centigrade Fahrenheit Centigrade Fahrenheit 200 392 1000 1832 225 437 1050 1922 250 482 1100 2012 275 527 1150 2102 300 572 1200 2192 325 617 1250 2282 350 662 1300 2372 375 707 1350 2462 400 752 1400 2552 425 797 1450 2642 450 842 1500 2732 475 887 1550 2822 500 932 1600 2912 525 977 1650 3002 550 1022 1700 3092 575 1067 1750 3182 600 1112 1800 3272 625 1157 1850 3362 650 1202 1900 3452 675 1247 2000 3632 700 1292 2050 3722 725 1337 2100 3812 750 1382 2150 3902 775 1427 2200 3992 800 1472 2250 4082 825 1517 2300 4172 850 1562 2350 4262 875 1607 2400 4352 900 1652 2450 4442 925 1697 2500 4532 950 1742 2550 4622 975 1787 2600 4712

METAL ALLOYS (Society of Automobile Engineers)

Babbitt— Tin........................... 84.00% Antimony...................... 9.00% Copper........................ 7.00%

Brass, White— Copper........................ 3.00% to 6.00% Tin (minimum) ................ 65.00% Zinc.......................... 28.00% to 30.00%

Brass, Red Cast— Copper........................ 85.00% Tin........................... 5.00% Lead.......................... 5.00% Zinc.......................... 5.00%

Brass, Yellow— Copper........................ 62.00% to 65.00% Lead.......................... 2.00% to 4.00% Zinc.......................... 36.00% to 31.00%

Bronze, Hard— Copper........................ 87.00% to 88.00% Tin........................... 9.50% to 10.50% Zinc.......................... 1.50% to 2.50%

Bronze, Phosphor— Copper........................ 80.00% Tin........................... 10.00% Lead.......................... 10.00% Phosphorus.................... .50% to .25%

Bronze, Manganese— Copper (approximate) ......... 60.00% Zinc (approximate) ........... 40.00% Manganese (variable) ......... small

Bronze, Gear— Copper........................ 88.00% to 89.00% Tin........................... 11.00% to 12.00%

Aluminum Alloys— Aluminum Copper Zinc Manganese No. 1.. 90.00% 8.5-7.0% No. 2.. 80.00% 2.0-3.0% 15% Not over 0.40% No. 3.. 65.00% 35.0%

Cast Iron— Gray Iron Malleable Total carbon........3.0 to 3.5% Combined carbon.....0.4 to 0.7% Manganese...........0.4 to 0.7% 0.3 to 0.7% Phosphorus..........0.6 to 1.0% Not over 0.2% Sulphur...........Not over 0.1% Not over 0.6% Silicon............1.75 to 2.25% Not over 1.0%

Carbon Steel (10 Point)— Carbon........................ .05% to .15% Manganese..................... .30% to .60% Phosphorus (maximum).......... .045% Sulphur (maximum)............. .05% (20 Point)— Carbon........................ .15% to .25% Manganese..................... .30% to .60% Phosphorus (maximum).......... .045% Sulphur (maximum)............. .05% (35 Point)— Manganese..................... .50% to .80% Carbon........................ .30% to .40% Phosphorus (maximum).......... .05% Sulphur (maximum)............. .05% (95 Point)— Carbon........................ .90% to 1.05% Manganese..................... .25% to .50% Phosphorus (maximum).......... .04% Sulphur (maximum)............. .05%

HEATING POWER OF FUEL GASES

(In B.T.U. per Cubic Foot.) Acetylene....... 1498.99 Ethylene....... 1562.9 Hydrogen........ 291.96 Methane........ 953.6 Alcohol......... 1501.76

MELTING POINTS OF METALS Platinum....................3200 Iron, wrought...............2900 malleable.................2500 cast......................2400 pure......................2760 Steel, mild.................2700 Medium....................2600 Hard......................2500 Copper......................1950 Brass.......................1800 Silver......................1750 Bronze......................1700 Aluminum....................1175 Antimony....................1150 Zinc........................ 800 Lead........................ 620 Babbitt..................500-700 Solder...................500-575 Tin......................... 450

NOTE.—These melting points are for average compositions and conditions. The exact proportion of elements entering into the metals affects their melting points one way or the other in practice.

TENSILE STRENGTH OF METALS

Alloy steels can be made with tensile strengths as high as 300,000 pounds per square inch. Some carbon steels are given below according to "points":

Pounds per Square Inch Steel, 10 point................ 50,000 to 65,000 20 point..................... 60,000 to 80,000 40 point..................... 70,000 to 100,000 60 point..................... 90,000 to 120,000 Iron, Cast..................... 13,000 to 30,000 Wrought...................... 40,000 to 60,000 Malleable.................... 25,000 to 45,000 Copper......................... 24,000 to 50,000 Bronze......................... 30,000 to 60,000 Brass, Cast.................... 12,000 to 18,000 Rolled....................... 30,000 to 40,000 Wire......................... 60,000 to 75,000 Aluminum....................... 12,000 to 23,000 Zinc........................... 5,000 to 15,000 Tin............................ 3,000 to 5,000 Lead........................... 1,500 to 2,500

CONDUCTIVITY OF METALS

(Based on the Value of Silver as 100)

Heat Electricity Silver....................100 100 Copper.................... 74 99 Aluminum.................. 38 63 Brass..................... 23 22 Zinc...................... 19 29 Tin....................... 14 15 Wrought Iron.............. 12 16 Steel..................... 11.5 12 Cast Iron................. 11 12 Bronze.................... 9 7 Lead...................... 8 9

WEIGHT OF METALS

(Per Cubic Inch) Pounds Pounds Lead............ .410 Wrought Iron..... .278 Copper.......... .320 Tin.............. .263 Bronze.......... .313 Cast Iron........ .260 Brass........... .300 Zinc............. .258 Steel........... .283 Aluminum......... .093

EXPANSION OF METALS

(Measured in Thousandths of an Inch per Foot of Length When Raised 1000 Degrees in Temperature) Inch Inch Lead............ .188 Brass............ .115 Zinc............ .168 Copper........... .106 Aluminum........ .148 Steel............ .083 Silver.......... .129 Wrought Iron..... .078 Bronze.......... .118 Cast Iron........ .068



CHAPTER VI

ELECTRIC WELDING

RESISTANCE METHOD

Two distinct forms of electric welding apparatus are in use, one producing heat by the resistance of the metal being treated to the passage of electric current, the other using the heat of the electric arc.

The resistance process is of the greatest use in manufacturing lines where there is a large quantity of one kind of work to do, many thousand pieces of one kind, for instance. The arc method may be applied in practically any case where any other form of weld may be made. The resistance process will be described first.

It is a well known fact that a poor conductor of electricity will offer so much resistance to the flow of electricity that it will heat. Copper is a good conductor, and a bar of iron, a comparatively poor conductor, when placed between heavy copper conductors of a welder, becomes heated in attempting to carry the large volume of current. The degree of heat depends on the amount of current and the resistance of the conductor.