BOILER FEEDERS AND SAFETY APPLIANCES
BOILER FEEDERS
      In all industrial works where a company is under contract to furnish power or light, and it is essential that there shall be no interruptions in the service, extra power units are always installed so that in case one breaks down another is at hand to take its place. In a like manner two boiler feeders are always supplied so that if one is out of order another will be ready to take its place at a moment’s notice.
      The greatest cause of danger in handling a steam boiler comes from low water, and probably the majority of boiler explosions are due to low water, either at the time of the explosion or at some previous time, which causes a weakening of the boiler plate. Since the water supply should be under constant control at all times two boiler feeders are necessary to insure safety. Both of these should be kept in proper working order at all times. Should the engineer neglect to put the extra one in repair at the earliest opportunity it should be considered a mark of incompetency, and is sufficient cause for his dismissal without further notice. No excuse should be considered where the safety of both the crew and the machine is at stake.
      It does not matter very much whether the boiler is equipped with two pumps, or two injectors, or an injector and a pump; neither does it make very much difference what make of pump or injector is used, since all standard makes are reliable and are perfectly capable of keeping the boiler supplied with water.
      To understand fully the workings of pumps and injectors it will be necessary first to consider some of the natural laws governing their action.
Atmospheric Pressure — This is the first thing to be considered in studying pumps or injectors.
      The earth is surrounded by a sea of air, and everything on the surface of the earth has to sustain a portion of the weight of the air. A column of air one inch square extending from the surface of the earth, at sea level, as high as the air reaches, weighs 14.7 pounds. If we go up on a high mountain or up in a balloon, the air presses upon us with less weight because there is not so much above us. The weight of the air does not hinder us from moving because it is equal on all sides and hence all pressures are balanced.
      A column of water one inch square and one foot high weighs 0.434 pound; consequently it would require a column of water 2.304 feet high to weigh one pound. A column of water one inch square, in order to balance atmospheric pressure would have to be 2.304×14.7=33.9 feet high.
      It is clear then that if all the air could be removed from the inside of a pipe which is closed at one end while the other end is submerged in water, that the pressure of the air on the outside would force water up in the pipe until the weight of water in the pipe exactly balances the air pressure on the surface of the water. If there is some air trapped above the water in the pipe, the water will rise only to such a height as is necessary to make the combined weight of water and the air pressure on top of the water equal the atmospheric pressure.
      The action of an ordinary suction pump is now easily explained. In Figure 18, A represents the pump cylinder, B the plunger and C the pipe extending down into the water. A check valve, D, in the plunger opens upward and another check, E, at the bottom of the pump cylinder opens the same way. A check valve or foot valve, as it is generally called, is placed at F to prevent water from running back into the well. When the pump plunger is pushed down, valve E closes, valve D opens, and the air between the plunger and E escapes to the upper side of the plunger through valve D. When the plunger moves upwards this air is carried out and the pressure inside of the pump cylinder is reduced below that of the atmosphere; consequently the water rises to a point g, where the pressure inside of the pipe exactly balances that on the outside. On the next stroke the same operation is repeated and the water rises to h. After a few more strokes water appears in the pump cylinder and then at the spout.
Figure 18 |
      Theoretically water can be raised by a pump of this kind 33.9 feet; practically, it will only rise to a height of 24 or 25 feet, owing to imperfections in the pump and the friction of valves, etc.
      While a pump of this kind is called a suction pump it does not exert any drawing force upon the water in the suction pipe. It simply removes the air from the inside of the suction pipe and atmospheric pressure does the rest. Water rises to an injector or steam siphon in the same way. For example, in the ejector, shown in section in Figure 19, steam rushing through the tubes A and B carries whatever air there may be in the suction pipe C along with it, thus forming a vacuum in the ejector and its connections. Atmospheric pressure, acting at the same time, forces water up into the ejector, where it is acted upon by the jet of steam and forced out at the delivery pipe. Water may be raised by this means about twenty feet, and forced to a height proportionate to the steam pressure. An ejector is very convenient for filling the tanks of traction engines.
Figure 19 |
The Injector — There is no machine used in steam engineering that is more difficult to understand than the injector.
      It seems incredible that a machine can be constructed which will take up a large quantity of water and then go back again into the boiler against the pressure from which it started. At first sight, it looks to be of the same nature as a perpetual motion machine, and it was considered in this light by the United States Patent Commissioner when it was submitted to him for letters patent. In fact, he refused, so it is said, to grant a patent until he had actually seen it in operation.
      The injector, by the way, is a comparatively new machine. It was invented by Mr. Henry Jacques Giffard, a Frenchman, about the year 1857, and its manufacture was begun in this country in 1860, by Wm. Sellers & Co., of Philadelphia.
      The principle of action of the injector is not easy to explain fully without the aid of some advanced mathematics; however, the following explanation will answer fairly well.
      Let B, Figure 20, represent a cross section of a steam boiler; C and D are pipes fitted with valves which communicate with the water space and steam space respectively. We will assume a steam pressure in the boiler of 100 pounds per square inch. Now if valve C was opened, water would flow out with a velocity of a little more than 121 feet per second, a figure which can easily be verified by anyone having a knowledge of the laws of falling bodies.
Figure 20 |
      If the valve D is open, with the steam pressure as before, steam will flow out with a velocity of 2,200 feet per second. We may say roughly that steam flowing from a boiler under pressure will have a velocity of from fifteen to eighteen times that of the water. This ratio changes somewhat under various conditions as to pressure in the boiler and the pressure against which the steam escapes.
      If now, instead of allowing the steam to escape through an open pipe it were made to pass through a pipe E, having a section at F, that could be kept very cold so that the steam would be instantly condensed at that point, the resulting stream of water, while very much smaller than the steam in cross section, would still travel with practically the same velocity; and if this stream were directed back into the boiler it would have no trouble in entering therein, since it has a velocity about eighteen times as great as that of the water which opposes its entrance. Since this stream of water has such a high velocity it could easily carry with it a considerable extra load, and while its velocity would be reduced thereby, it would still have sufficient velocity to enter the boiler.
      There are really only two types of injectors, namely, the automatic and the positive. Automatic injectors have a single set of jets, or tubes, while positive injectors have two sets. If the end of the suction hose becomes uncovered and the suction breaks, the automatic injector will start again, but if the suction of the positive injector breaks it must be started again by hand. It is this property of the automatic injector that makes it the best form of injector for all road engines where the water washes back and forth in the tank a great deal.
      Figure 21 is in sectional view of a Penberthy injector, such as is used on traction engines. When steam is first admitted to the injector, it flows through the steam jet R, then down through the suction jet S, and carries with it whatever air there is in the space between jets R and S.
Figure 21 |
      This steam and air lifts the overflow valve and escapes to the atmosphere, because it has not momentum enough to enter the boiler. As soon as the air is exhausted from the inside of the injector, atmospheric pressure forces water up into the combining chamber, and condenses the jet of steam issuing from the steam jet R.
      The chamber between R and S corresponds to the cold portion F in the preceding diagram, and is maintained cold so long as fresh water enters from the tank, Whatever steam there may be in the injector is now condensed by the jet of water passing through, and consequently atmospheric pressure closes the check valve P.
      The stream flowing through the injector must be a purely liquid one, that is, it must not contain any steam or air. If it does, the resulting stream will not have enough weight combined with its velocity to overcome boiler pressure and will consequently flow out at the overflow valve.
      For the same reason, if the injector is hot, as it is if the valve in the steam pipe. leaks, it will not work because some of the suction water is changed to steam by the heat in the injector and the resulting stream will contain steam. The remedy in this case is to pour cold water on the injector until it becomes cold enough to start. An injector can not handle hot water either, because hot water will not condense all of the steam.
BOILER SAFETY APPLIANCES
      There are five safety devices usually found on every steam boiler. They are a steam gauge, a safety valve, a gauge glass, a fusible plug and try cocks. All of these devices are necessary in order to make a boiler reasonably safe, and more than this, the operator or engineer should know the construction of all of these devices intimately and know how to take proper care of them so that they may be in proper working order at all times. In addition to the safety devices above mentioned, which are to be found on every traction engine boiler, there are to be found on many stationary boilers low water alarms and automatic devices which start and stop the feed pump and thus keep the water in the boiler at a constant level. These are useful, too, and serve a good purpose, but are too complicated and expensive for traction engine purposes. After all, the best safety device that any boiler can be provided with, whether it be stationary or traction, is an intelligent, well trained man to take care of it and all of its fittings. The first named safety devices may be termed essential fittings, no matter how good the engineer is, and we will proceed to study them. The steam gauge is perhaps the most important and we will consider it first.
      The purpose of a steam gauge is to measure the outward pressure of the steam on the walls of the boiler. The steam gauge is a sort of weighing machine which weighs this pressure in pounds per square inch. If an accurate gauge is placed on a boiler, and it registers 150 pounds (per square inch), it means that on
every square inch inside of the boiler there is an outward pressure of 150 pounds due to the steam. The size of the opening through which steam flows on its way to the gauge is not a square inch in area, but this makes no difference since the gauge is made to register in pounds per square inch.
      There are two kinds of steam gauges in common use, namely, the Bourdon spring gauge and the capsular spring gauge. Both are good gauges and both have been used for a great many years, their first appearance dating from about the year 1850. Figure 22 illustrates a Bourdon spring gauge with the dial removed showing the inside mechanism. The spring consists of a brass tube of elliptical shape, closed at the upper end and connected to the steam space in the boiler at the lower end. The outer end of this spring is connected by means of a suitable link to a segment lever whose teeth mesh with a small pinion. This pinion is mounted on a spindle which carries a pointer. When the pressure of the steam is exerted on the inside of the bent spring it tends to straighten and in doing so forces the hand around the circle and over the face of a dial. This dial is graduated and marked to show the pressure in pounds (per square inch) corresponding to any position of the pointer. A small flat coil spring takes up the back lash or lost motion of the pinion and makes the pointer sensitive to any changes in the Bourdon spring. The elliptical shape of these springs makes them more sensitive than if they were round. Gauges used for low pressure work are graduated up to from thirty to forty-five pounds (per square inch) and are fitted with a light sensitive spring. Those that are designed for high pressure work, such as for traction engines, are fitted with heavier springs and are graduated up to from two hundred to two hundred fifty pounds (per square inch). Such gauges are not supposed to be accurate at low pressures below twenty-five or thirty pounds (per square inch). Above that, however, they are very nearly exact up to the limit for which they are graduated.
Figure 22 |
      Figure 23 shows a double spring gauge. These gauges are somewhat more expensive and are used largely on locomotives, being less sensitive to vibrations of the engine than single spring gauges and being also more substantial.
Figure 23 |
      If steam were admitted directly into the Bourdon spring, the gauge would not register correctly on account of the expansion due to the heat from the steam. For this reason, and because the spring would deteriorate quite rapidly under the high temperature of the steam, a siphon is placed between the boiler and the gauge. In stationary practice this siphon often consists of a single loop of pipe, but for traction engine work a brass bulb siphon, shown in section in Figure 24, is used, the loop of pipe having too much vibration for a boiler subjected to rough road conditions. The first steam that enters the siphon is condensed and a plug of water, sufficient to fill the Bourdon spring, is forced up into the gauge. Above this there is a plug of air and still below this there is steam. This mixture of air and water does not reach as high a temperature as the dry steam and does not change its temperature so quickly and is much easier on the spring. The brass siphon is arranged to drain the gauge as the steam pressure in the boiler falls.
Figure 24 |
      Care should be taken to keep an accurate steam gauge on the boiler. A gauge that needs frequent coaxing with a pitchfork handle is, needless to say, not entirely satisfactory. In general, if the gauge does not agree with the pop valve, it may be considered out of order. The pop valve is much less delicate than a gauge and while it may go wrong, it is less liable to do so than the gauge. The fact that a gauge is new is not sufficient guarantee that it is accurate. While all gauges are tested at the factory where they are made they frequently get out of adjustment before getting into the user’s hands. A new gauge came into the writer’s hands about a year ago that was out of adjustment. The hand had slipped on its pinion. By running the pressure on the standard test gauge up to one hundred pounds (per square inch) and then setting the hand on the new gauge to the same point, the correct adjustment was made. A leak in the steam connections between boiler and gauge will prevent a gauge from working correctly and perhaps ruin the gauge if it allows the water to escape from the spring. A cracked spring can not be mended. It is not likely to occur unless the gauge freezes and this very thing may occur and has often occurred while the boiler has been steamed up if the weather was very cold and a high wind was blowing. Under such conditions the gauge should be kept covered with a blanket or an old coat. The mechanism of a gauge rarely ever needs oiling, and if it is oiled, only a good grade of clock oil should be used and that sparingly. It is better not to use oil at all unless the engineer is very well acquainted with steam gauges.
      Figure 25 shows a capsular spring gauge and Figure 26 shows the construction of the capsular spring. Steam pressure from the boiler forces water between the two faces of the capsular spring and spreads them apart a distance proportional to the amount of the steam pressure. This movement is transmitted through a set of multiplying levers, a segment lever and a pinion to a spindle that carries the pointer. These gauges are strong and well made and are accurate. They are not, however, used so much on traction engines as Bourdon spring gauges.
Figure 25 |
Figure 26 |
      The fusible plug or soft plug is another important safety device that all or nearly all traction engine boilers are equipped with. It consists of a brass plug having an opening in the middle filled with tin. Figure 27 shows two styles in which these plugs are made. In the one marked A, the hole in the plug is made tapering so that when steam pressure acts on the tin filling it can not possibly be forced out by the pressure alone since the pressure acts on the large end of the tin plug. In B there is an enlargement in the middle that serves the same purpose.
Figure 27 |
      In fire box boilers the plug is screwed into the highest point of the fire box and in return flue boilers it is located in the front end, in the smoke box, just above the main flue. The tin that it is filled with melts at a temperature of about 440° F., and if the water in the boiler gets so low as to leave the top of the plug bare, the tin melts and water and steam blowout. If this happens in a fire box boiler the fire will be put out. Many return flue boilers do not have a fusible plug, but all fire box boilers do, and they are needed.
      In case a plug melts out anyone can fill it by melting a little tin in a suitable iron dish and pouring the hole in the plug full. If the plug is stood up on an iron plate it will prevent the tin from running through. After the plug is filled the tin should be tamped in with a hammer and punch. In filling be sure there is no moisture in the plug. If there is, the hot metal will turn it quickly into steam and here will be a little explosion and some one is apt to get burned with the hot tin. The plug should be filled at the beginning of every season. If left in too long it becomes crystallized and does not melt readily. It is a good plan to take the plug out every time the boiler is cleaned and see that the top is not covered with scale. A little scale on the top can easily prevent the steam from blowing out even if the tin has melted. It is also a good plan to coat the threads of the plug with graphite so that it will unscrew easily next time. Oil put on the threads will burn, forming a deposit of carbon that will make it stick and consequently oil should not be used. In concluding this bit of advice in regard to fusible plugs it may be well to add that an iron plug such as a spike is a very poor substitute for tin and is not to be recommended although some fellows who claim to be engineers use it occasionally. Babbitt is not good either although it is better than the spike. The reason babbitt is not very good is that it has a rather uncertain melting point, depending upon its composition, and may be too high. Pure tin is by all odds the best and every engine should be provided with a bar to be kept in the tool box for emergency.
      The next safety device we will consider is the safety valve or spring pop valve, a sectional cut of one type of which appears in Figure 28. It is made of brass throughout except the springs and the handle. The lower end G screws into the steam space in the boiler and admits steam to the lower side of the main valve A. Rod B rests on the top of this valve and is held down by means of the cap H and main spring S. In order for valve A to rise it must compress the spring S. A lock nut holds the top of this spring in place and if it is screwed down it puts more load on the spring and of course more load on the top of the main valve. A full turn of this lock nut, by the way, is equivalent to adding about thirty pounds pressure on the top of the valve.
Figure 28 |
      It doesn’t pay, therefore, to use a monkey wrench very freely on this lock nut unless you want to carry a tremendous pressure on the boiler. On the top of valve A there is another valve C, called an auxiliary valve. This valve is held to its seat by an auxiliary spring E. It will be noticed that this valve and spring are attached firmly to the stem of the main valve and must move with it. The purpose of this auxiliary valve will presently be described.
      All pop valves are provided with what is called a pop chamber, into which the steam first expands after it passes the main valve. This is shown at M, in the figure. When the pressure in the boiler is less than the compressive force on spring S, the main valve remains seated, but when it raises to a point just a trifle above the load on the spring, the valve rises and steam flows out around the valve seat V, and up into the pop chamber M, underneath the valve C. In expanding, the steam acquires considerable velocity, which is changed to pressure when stopped by the valve C. The force that now opens the main valve is the steam pressure acting on the lower side of A plus the pressure on C in the pop chamber. This total pressure is more than sufficient to open the main valve and it pops wide open. It would remain open until the steam in the boiler had fallen a considerable amount below the popping off point if there was not some provision made to relieve the pressure in the pop chamber. This is accomplished in this machine by making the compressive force on spring E much less than on spring S. This allows the valve C to lift and let the steam escape from the pop chamber. The load on spring E can be regulated by means of the nut D. If this is made heavy the pressure in the boiler will fall a considerable amount before the main valve returns to its seat. If made light, on the other hand, there will be only a slight fall in pressure. It is set correctly when it leaves the factory and needs no further attention unless the pressure at which the main valve works is changed a great deal. In that case it may be necessary to make some adjustment.
      In other types of pop valves there are different methods used to accomplish the same object that the auxiliary valve does in the pop valve above described. These devices are known as regulators and provide means for relieving the pressure in the pop chamber at varying rates. In almost all pop valves, except the one described, this regulator must be adjusted whenever the load on the main valve is changed very much, otherwise the pressure in the boiler will be either reduced by too small an amount, or else too much pressure will be lost every time the pop valve acts. In general, the regulator should be set to reduce the pressure in the boiler about three pounds (per square inch).
      The pressure at which the pop valve is set on a new engine is what the manufacturer considers the safe working pressure for his boiler. While the boiler will undoubtedly be safe with somewhat higher pressures when new it is not good sense to screw down the pop and increase the pressure. As the boiler grows older it is not able to stand such high pressures as when new and the “pop” should be set lower. It may be set down as a general rule, though not applicable in every case, that the engineer who has a hankering to screw down the pop valve is a fellow who needs pretty close watching. It might be safer to let him haul water.
Information Sources
- The Thresher’s Guide, Volume I, 1910 pages 32-43
Being a reprint from the Threshers’ School Of Modern Methods of the American Thresherman
Published By The American Thresherman, Madison, Wisconsin
Copyright 1910 By The American Thresherman
Prepared By Professor P.S. Rose, North Dakota Agricultural College
This book courtesy of Brian Szafranski