COMPRESSED AIR PRODUCTION.

Scientific American Supplement, No. 799, April 25, 1891 by Various, is part of the HackerNoon Books Series. You can jump to any chapter in this book here. COMPRESSED AIR PRODUCTION.

COMPRESSED AIR PRODUCTION.

By WM. L. SAUNDERS, C.E., of New York.

I cannot but realize as I stand before you that I would be very much more at home were I in your midst. I feel but little older and so very much less wise than when I sat in the class room an undergraduate of the University of Pennsylvania, that I trust I may expect you to give me this afternoon, not only your attention, but your sympathy.

The present situation is not without suggestions of my own experience. I recall a lecture in the ordinary course, given by our professor of mining, whose struggles with the English language were quite as conspicuous as were our efforts to tell what he was driving at. He was describing an ordinary windlass hoist used at the shaft of a mine. He said "There is a windlass at de top of de shaft around which is coiled a rope, on de two ends of which is fastened two er—er—pans, one of which is a bucket and de oder a platform." I mention this because I shall ask you to attribute my shortcomings in this lecture, not so much to my lack of familiarity with my native tongue, as to—well, because I was not educated at Cornell University.

We all know what free air is. You who are privileged to live upon these beautiful hills overlooking Ithaca and the lake, doubtless know more about free air than we do who are choked in the dusty confines of New York City. Compressed air is simply air under pressure. That pressure may be an active one, as in the case of the piston of an air compressor; or passive, as with the walls of a receiver or transmission pipe. It is usual to define compressed air as air increased in density by pressure, but we know that we may produce compressed air by heat alone. A simple illustration of this is the pressure which will blow a cork from an empty bottle when that bottle has been placed near the fire. Here we have pressure, or compressed air, in the bottle produced by heat alone.

Having defined compressed air, we must next define heat; for in dealing with compressed air, we are brought face to face with the complex laws of Thermodynamics. We cannot produce compressed air without also producing heat, and we cannot use compressed air as a power without producing cold. Based on the material theory of heat, we would say that when we take a certain volume of free air and compress it into a smaller space, we get an increase in temperature because we have the heat of one volume occupying less space, but no one at this date accepts the material theory of heat. Your distinguished director, Professor Thurston, in discussing "Steam and its Rivals," in the Forum, said: "The science of Thermodynamics teaches that heat and mechanical energy are only different phases of the same thing, the one being the motion of molecules, and the other that of masses." This is the accepted theory of heat. In other words, we do not believe that there is any such thing as heat, but that what we call heat is only the sensible effect of motion. In the cylinder of an air compressor the energy of the piston is converted into molecular motion in the air and the result, or the equivalent, is heat. A higher temperature means an increased speed of vibration, and a lower temperature means that this speed of vibration is reduced. If I hold an open cylinder in my left hand and a piston in my right, and place the piston within the cylinder, I here have a confined volume of air at the temperature and the pressure of this room. These particles of air are in motion and produce heat and pressure in proportion to that motion. Now if I press the piston to a point in the center of the cylinder, that is, to one-half the stroke, I here decrease the distance between the cylinder head and the piston just one-half, hence each molecule of air strikes twice as many blows upon the piston and head in traveling the same distance and the pressure is doubled. We have also produced about 116 degrees of heat, because we have expended a certain amount of work upon the air; the air has done no work in return, but we have increased the energy of molecular vibration in the air and the result is heat.

But what of this heat? What harm does it do? If I instantly release the piston which I hold at one-half stroke it will return to its original position, less only a little friction. I have, therefore, recovered all, or nearly all, the power spent in compressing the air. I have simply pressed a spring, and have let it recover. We see what a perfect spring compressed air is. We see the possibility of expending one horse power of energy upon air and getting almost exactly one horse power in return. Such would be the case provided we used the compressed air power immediately and at the point where the compression takes place. This is never done, but the heat which has been boxed up[1] in the air is lost by radiation, and we have lost power. Let us see to what extent this takes place.

Thirteen cubic feet of free air at normal temperature and barometric pressure weigh about one pound. We have seen that 116 degrees of heat have been liberated at half stroke. The gauge pressure at this point reaches 24 pounds. According to Mariotte's law, "The temperature remaining constant, the volume varies inversely as the pressure," we should have 15 pounds gauge pressure. The difference, 9 pounds, represents the effect of the heat of compression in increasing the relative volume of the air.

The specific heat of air under constant pressure being 0.238, we have 0.238 × 116 = 27.6 heat units produced by compressing one pound or thirteen cubic feet of free air into one-half its volume. 27.6 × 772 (Joule's equivalent) = 21,307 foot pounds. We know that 33,000 foot pounds is one horse power, and we see how easily about two-thirds of a horse power in heat units may be produced and lost in compressing one pound of air. I would mention here that exactly this same loss is suffered when compressed air does work in an engine and is expanded down to its original pressure. In other words, the heat of compression and the cold of expansion are in degree equal.

Experiments made by M. Regnault and others on the influence of heat on pressures and volumes of gases have enabled us to fix the absolute zero of temperature as -461 degrees Fahrenheit. This point, 461 degrees below zero, is the theoretical point at which a volume of air is reduced to nothing. The volume of air at different temperatures is in proportion to the absolute temperature, and on this basis Box gives us the following table:

TABLE l.—OF THE VOLUME AND WEIGHT OF DRY AIR AT DIFFERENT TEMPERATURES UNDER A CONSTANT ATMOSPHERIC PRESSURE OF 29.92 INCHES OF MERCURY IN THE BAROMETER (ONE ATMOSPHERE), THE VOLUME AT 32° FAHRENHEIT BEING 1.

  Temperature   Volume in     Weight of a
  in degrees.   cubic feet.   cubic foot in lb.
       32          1.000         0.0807
       42          1.020         0.0791
       52          1.041         0.0776
       62          1.061         0.0761
       72          1.082         0.0747
       82          1.102         0.0733
       92          1.122         0.0720
      102          1.143         0.0707
      112          1.163         0.0694
      122          1.184         0.0682
      132          1.204         0.0671
      142          1.224         0.0660
      152          1.245         0.0649
      162          1.265         0.0638
      172          1.285         0.0628
      182          1.306         0.0618
      192          1.326         0.0609
      202          1.347         0.0600
      212          1.367         0.0591
      230          1.404         0.0575
      250          1.444         0.0559
      275          1.495         0.0540
      300          1.546         0.0522
      325          1.597         0.0506
      350          1.648         0.0490
      375          1.689         0.0477
      400          1.750         0.0461
      450          1.852         0.0436
      500          1.954         0.0413
      550          2.056         0.0384
      600          2.15[2]       0.0376
      650          2.260         0.0357
      700          2.362         0.0338
      750          2.464         0.0328
      800          2.566         0.0315
      850          2.668         0.0303
      900          2.770         0.0292
      950          2.872         0.0281
    1,000          2.974         0.0268
    1,100          3.177         0.0254
    1,200          3.381         0.0239
    1,300          3.585         0.0225
    1,400          3.789         0.0213
    1,500          3.993         0.0202
    1,600          4.197         0.0192
    1,700          4.401         0.0183
    1,800          4.605         0.0175
    1,900          4.809         0.0168
    2,000          5.012         0.0161
    2,100          5.216         0.0155
    2,200          5.420         0.0149
    2,300          5.624         0.0142
    2,400          5.828         0.0138
    2,500          6.032         0.0133
    2,600          6.236         0.0130
    2,700          6.440         0.0125
    2,800          6.644         0.0121
    2,900          6.847         0.0118
    3,000          7.051         0.0114
    3,100          7.255         0.0111
    3,200          7.459         0.0108

The effect of this heat of compression in increasing the volume, and the heat produced at different stages of compression, are shown by the following table:

TABLE 2.—HEAT PRODUCED BY COMPRESSION OF AIR.

  --------+-----------------------+----------+------------+-------------
          |      Pressure.        |          |            |
  Atmo-   +-----------+-----------+  Volume  |Temperature |    Total
  spheres.|Pounds per |Pounds per | in Cubic | of the Air | Increase of
          |Square Inch|Square Inch|   Feet.  | throughout | Temperature.
          | above a   |above the  |          |the Process.|   Degrees.
          | Vacuum.   |Atmosphere |          |  Degrees.  |
          |           |(Gauge     |          |            |
          |           |Pressure). |          |            |
  --------+-----------+-----------+----------+------------+-------------
    1.00  |   14.70   |    0.00   |  1.0000  |    60.0    |    00.0
    1.10  |   16.17   |    1.47   |  0.9346  |    74.6    |    14.6
    1.25  |   18.37   |    3.67   |  0.8536  |    94.8    |    34.8
    1.50  |   22.05   |    7.35   |  0.7501  |   124.9    |    64.9
    1.75  |   25.81   |   11.11   |  0.6724  |   151.6    |    91.6
    2.00  |   29.40   |   14.70   |  0.6117  |   175.8    |   115.8
    2.50  |   36.70   |   22.00   |  0.5221  |   218.3    |   158.3
    3.00  |   44.10   |   29.40   |  0.4588  |   255.1    |   195.1
    3.50  |   51.40   |   36.70   |  0.4113  |   287.8    |   227.8
    4.00  |   58.80   |   44.10   |  0.3741  |   317.4    |   257.4
    5.00  |   73.50   |   58.80   |  0.3194  |   369.4    |   309.4
    6.00  |   88.20   |   73.50   |  0.2806  |   414.5    |   354.5
    7.00  |  102.90   |   88.20   |  0.2516  |   454.5    |   394.5
    8.00  |  117.60   |  102.90   |  0.2288  |   490.6    |   430.6
    9.00  |  132.30   |  117.60   |  0.2105  |   523.7    |   463.4
   10.00  |  147.00   |  132.30   |  0.1953  |   554.0    |   494.0
   15.00  |  220.50   |  205.80   |  0.1465  |   681.0    |   621.0
   20.00  |  294.00   |  279.30   |  0.1195  |   781.0    |   721.0
   25.00  |  367.50   |  352.80   |  0.1020  |   864.0    |   804.0
  --------+-----------+-----------+----------+------------+-------------

A cubic foot of free air at a pressure of one atmosphere (equal to 14.7 pounds above a vacuum) at a temperature of 60 degrees, when compressed to twenty-five atmospheres, will register 367.5 pounds above a vacuum (352.8 pounds gauge pressure), will occupy a volume of 0.1020 cubic foot, will have a temperature of 864 degrees, and the total increase of temperature is 804 degrees.

The thermal results of air compression and expansion are shown by the accompanying diagram.

The horizontal and vertical lines are the measures of volumes, pressures and temperatures. The figures at the top indicate pressures in atmospheres above a vacuum, the corresponding figures at the bottom denote pressures by the gauge. At the right are volumes from one to one-tenth. At the left are degrees of temperatures from zero to 1,000 Fahrenheit. The two curves which begin at the upper left hand corner and extend to the lower right are the lines of compression or expansion.

The upper one being the Adiabatic curve, or that which represents the pressure at any point on the stroke with the heat developed by compression remaining in the air; the lower is the Isothermal, or the pressure curve uninfluenced by heat. The three curves which begin at the lower left hand corner and rise to the right are heat curves and represent the increase of temperature corresponding with different pressures and volumes, assuming in one case that the temperature of the air before admission to the compressor is zero, in another sixty degrees, and in another one hundred degrees.

Beginning with the adiabatic curve, we find that for one volume of air when compressed without cooling the curve intersects the first vertical line at a point between 0.6 and 0.7 volume, the gauge pressure being 14.7 pounds. If we assume that this air was admitted to the compressor at a temperature of zero, it will reach about 100 degrees when the gauge pressure is 14.7 pounds. We find this by following down the first line intersected by the adiabatic curve to the point where the zero heat curve intersects this same line, the reading being given in figures to the left immediately opposite. If the air had been admitted to the compressor at 60 degrees, it would register about 176 degrees at 14.7 pounds gauge pressure. If the air were 100 degrees before compression, it would go up to about 230 degrees at this pressure. Following this adiabatic curve until it intersects line No. 5, representing a pressure of five atmospheres above a vacuum (58.8 lb. gauge pressure), we see that the total increase of temperature on the zero heat curve is about 270 degrees, for the 60 degree curve it is about 370 degrees, and for the 100 degree curve it is about 435 degrees. The diagram shows that when a volume of air is compressed adiabatically to 21 atmospheres (294 lb. gauge pressure), it will occupy a volume a little more than one-tenth; the total increase of temperature with an initial temperature of zero is about 650 degrees; with 60 degrees initial temperature it is 800 degrees, and with 100 degrees initial it is 900 degrees. It will be observed that the zero heat curve is flatter than the others, indicating that when free air is admitted to a compressor cold, the relative increase of temperature is less than when the air is hot. This points to the importance of low initial temperature.

We have now seen that the economical production of compressed air depends upon the following conditions:

(1) A low initial temperature.

(2) Thorough cooling during compression.

It has been demonstrated by experiments made in France that the power required to compress moist air is less than that for dry air. A table showing the power required to compress moist and dry air has been prepared from the data of M. Mallard and shows that for five atmospheres the work expended in compressing one pound of dry air is 58,500 foot pounds, while that for moist air is 52,500 foot pounds. In expansion also moisture in the air adds to the economy, but in both cases the saving of power is not great enough to compensate for the many disadvantages due to the presence of water. Mr. Norman Selfe, of the Engineering Association of N.S.W., has compiled a table which shows some important theoretical conditions involved in producing compressed air.

So much for the theory of compression. We now come to the practical production of compressed air.

The first record that we have of the use of an air compressor is at Ramsgate Harbor, Kent, in the year 1788. Smeaton invented this "pump" for use in a diving apparatus. In 1851, William Cubitt, at Rochester Bridge, and a little later an engineer, Brunel, at Saltash, used compressed air for bridge work. But the first notable application of compressed air is due to Professor Colladon, of Geneva, whose plans were adopted at the Mont Cenis tunnel. M. Sommeiller developed the Colladon idea and constructed the compressed air plant illustrated in Fig. 2.

The Sommeiller compressor was operated as a ram, utilizing a natural head of water to force air at 80 pounds pressure into a receiver. The column of water contained in the long pipe on the side of the hill was started and stopped automatically, by valves controlled by engines. The weight and momentum of the water forced a volume of air with such shock against a discharge valve that it was opened and the air was discharged into the tank; the valve was then closed, the water checked; a portion of it was allowed to discharge and the space was filled with air, which was in turn forced into the tank. The efficiency of this compressor was about 50 per cent.

At the St. Gothard tunnel, begun in 1872, Prof. Colladon first introduced the injection of water in the form of spray into the compressor cylinder to absorb the heat of compression.

Fig. 3 illustrates the air cylinder of the Dubois-Francois type of compressor, which was the best in use about the year 1876. This compressor was exhibited at the Centennial Exposition and was adopted by Mr. Sutro in the construction of the Sutro tunnel. A characteristic feature seems to be to get as much water into the cylinder as possible. The water which flooded the bottom of the cylinder arose from the voluminous injection; this water was pushed into the end of the cylinder and some of it escaped with the air through the discharge valve.

An improved pattern of this compressor is shown in Fig. 4.

These illustrations are interesting from an historical point of view, as indicating the line of thought which early designers of air-compressing machinery followed. As the necessity for compressed air power grew, inventors turned their attention to the construction of air-compressing engines that would combine efficiency with light weight and economy of space and cost. The trade demanded compressors at inaccessible localities, and in many cases it was preferred to sacrifice isothermal results to simplicity of construction and low cost.

It is evident that an air compressor which has the steam cylinder and the air cylinder on a single straight rod will apply the power in the most direct manner, and will involve the simplest mechanics in the construction of its parts. It is evident, however, that this straight line, or direct construction, results in an engine which has the greatest power at a time when there is no work to perform. At the beginning of the stroke steam at the boiler pressure is admitted behind the piston, and, as the air piston at that time is also at the initial point in the stroke, it has only free air against it. The two pistons move simultaneously, and the resistance in the air cylinder rapidly increases as the air is compressed. To get economical results it is, of course, necessary to cut off in the steam cylinder, so that at the end of the stroke, when the steam pressure is low, as indicated by the dotted line (Fig. 5), the air pressure is high, as similarly indicated. The early direct-acting compressor used steam at full pressure throughout the stroke. The Westinghouse pump, applied to locomotives, is built on this principle, and those who have observed it work have perhaps noticed that its speed of stroke is not uniform, but that it moves rapidly at the beginning, gradually reducing its speed, and seems to labor, until the direction of stroke is reversed. This construction is admitted to be wasteful, but in some cases, notably that of the Westinghouse pump, economy in steam consumption is sacrificed to lightness and economy of space.

Many efforts were made to equalize the power and resistance by constructing the air compressor on the crank shaft principle, putting the cranks at various angles, and by angular positions of steam and air cylinders. Several types are shown in Fig. 6.

Angular positions of the cylinder involve expensive construction and unsteadiness. Experience has conclusively proved that it does not pay to build air compressors with vertical cylinders, and moreover we have found out that there is nothing in the apparent difficulty in equalizing the strains in a direct-acting engine. It is simply necessary to add enough weight to the moving parts, that is, to the piston, piston rod, fly wheel, etc., to cut off early in the stroke and secure rotative speed with the most economical results and with the cheapest construction. It is obvious that the theoretically perfect air compressor is a direct-acting one with a conical air cylinder, the base of the cone being nearest the steam cylinder. This, from a practical point of view, is impossible. Mr. Hill, in referring to the fallacious tendencies of pneumatic engineers to equalize power and resistance in air compressors, says: "The ingenuity of mechanics has been taxed and a great variety of devices have been employed. It is usual to build on the pattern of presses which do their work in a few inches of the end of the stroke and employ heavy fly wheels, extra strong connections, and prodigious bed plates. Counterpoise weights are also attached to such machines; the steam is allowed to follow full stroke, steam cylinders are placed at awkward angles to the air-compressing cylinders and the motion conveyed through yokes, toggles, levers; and many joints and other devices are used, many of which are entire failures, while some are used with questionable engineering skill and very poor results."

Fig. 7 illustrates the theory of Duplex Air Compressors. The hydraulic piston or plunger compressor is largely used in Germany and elsewhere on the Continent of Europe, but the duplex may be said to be the standard type of European compressor at the present time. It is also largely used in this country. Fig. 7 shows the four cylinders of a duplex compressor in two positions of the stroke. It will be observed that each steam cylinder has an air cylinder connected directly to the tail rod of its piston, so that it is a direct-acting machine, except in that the strains are transmitted through a single fly wheel, which is attached to a crank shaft connecting the engines. In other words, a duplex air compressor would be identical with a duplex steam engine were it not for the fact that air cylinders are connected to the steam piston rods. The result is, as shown in Fig. 7, that, at that point of the stroke indicated in the top section, the upper right hand steam cylinder, having steam at full pressure behind its piston, is doing work through the angle of the crank shaft upon the air in the lower left hand cylinder. At this point of the stroke the opposite steam cylinder has a reduced steam pressure and is doing little or no work, because the opposite air cylinder is beginning its stroke. Referring now to the lower section, it will be seen that the conditions are reversed. One crank has turned the center, and that piston which in the upper section was doing the greatest work is now doing little or nothing, while the labor of the engine has been transferred to those cylinders which a moment before had been doing no work.

There are some advantages in this duplex construction, and some disadvantages. The crank shafts being set quartering, as is the usual construction, the engine may be run at low speed without getting on the center. Each half being complete in itself, it is possible to detach the one when only half the capacity is required. The power and resistance being equalized through opposite cylinders, large fly wheels are not necessary. Strange to say, the American practice seems to be to attach enormous fly wheels to duplex air compressors. It is difficult to justify this apparently useless expense in view of the facts shown in Fig. 7. A fly wheel does not furnish power, nor does it add to the economy of an engine except in so far as it enables it to cut off early in the stroke, and to equalize the power and resistance. In other words, a fly wheel is not a source of power, and in many cases it is only a means by which we accomplish rotative speed. It takes power to move matter, and assuming that other conditions are equal, every engine that carries a fly wheel that is larger than is necessary consumes a certain number of foot pounds in turning so much metal around through space. Were it possible to cut off at the same point and rotate as positively without a fly wheel, it would be done away with entirely. Some straight line air compressors are so constructed that the momentum of the piston and other moving parts is nearly sufficient to equalize the strains without a fly wheel; but the fly wheel is there because it insures a definite length of stroke, and because it enables us to operate eccentrics and to regulate the speed of the engine uniformly.

Objections to the duplex construction are: The strains are indirect, angular and intermittent. It is necessary therefore to largely increase the strength of parts; to add a crank shaft of large diameter with enormous bearings, and to build expensive and very secure foundations. Should the foundations settle at any point, excessive strains will be brought upon the bearings, resulting in friction and liability to breakage. A steam engine meets with a resistance on its crank shaft that is uniform throughout the stroke; while an air compressor is subject to a heavy maximum strain at the end of the stroke, hence the importance of direct straight line connection between power and resistance.

The friction loss on a duplex compressor seldom gets lower than 15 per cent., while straight line compressors show as low a loss as 5 per cent. Fig. 8 illustrates the Rand Duplex Air Compressor, a machine largely used in America, especially in the Lake Superior iron mines. Fig. 9 illustrates a Duplex Compound Condensing Corliss Air Compressor built by the Ingersoll-Sergeant Drill Company. This is a compressor made of the best type of Corliss engine, with air cylinders connected to the tail rods of the steam cylinders. One of these machines, of about 400 horse power capacity, is now at work furnishing compressed air power for the Brightwood Street Railway in Washington, D.C. Fig. 10 illustrates the Norwalk direct-acting straight line air compressor, with compound air cylinder. The chief purpose of compounding is to reduce the maximum strain. This construction also adds to isothermal economy. The large cylinder to the left determines the capacity of the compressor, the air being compressed first to a low pressure (ordinarily about 30 pounds per square inch), afterward passing through an intercooler, by which its temperature is reduced, and then it is compressed still higher, even to 5,000 pounds per square inch if desired. The terminal strain, which is so severe in air compressors, is here considerably reduced, as in this case it is only equal to the area of the initial air piston multiplied by its low air pressure.

Economical results are attained with this compressor at low cost of construction. The fly wheels are small, and the bearings narrow, because the maximum strain is less, and the momentum of the piston and other moving parts is such that most of the high initial steam power is taken up in starting these parts and is afterward given out at the end of the stroke, when the steam pressure is low. The strains are direct, and expensive foundations are not required. Fig. 11 illustrates the Ingersoll-Sergeant Compound Straight Line Air Compressor. This differs from the one just described chiefly in that it is single-acting, while the other is double-acting.

By single-acting is meant that the air cylinders compress their respective volumes of air once every revolution. The air is admitted to the large cylinder through the piston, is compressed to about 30 pounds, and on the return stroke the pressure is raised to almost any point required, and in proportion to the diameter of the smaller cylinder. Though single-acting, the capacity of one of these compressors is about equal to that of the double-acting machine of the same cost of construction. The initial air cylinder is made large enough to correspond with the capacity of the smaller double-acting cylinder. The strains are equalized because the area of the large cylinder multiplied by its low pressure is exactly equal to that of the small cylinder multiplied by its high pressure. The maximum strains are reduced considerably below those which exist in compressors that do not compound the air.

The advantage of the single-acting air cylinder over the double is that it compresses a volume of free air only once every revolution, hence there is a better chance to cool the air during compression. The cylinders have time to impart to the water jackets the heat produced by compression and are kept cooler. The large air head of the initial cylinder is jacketed, also adding to isothermal economy.

Fig. 12 illustrates the Ingersoll-Sergeant Piston Inlet Cold Air Compressor. This a straight line direct-acting engine, with steam and air pistons connected to a single rod through a crosshead which connects with two fly wheels. The strains are direct and the power and resistance are equalized by the inertia of the crosshead, piston, rods, and fly wheels. The Meyer's adjustable cut-off is used on the steam cylinder. The air cylinder is provided with a tail rod tube through which all the air is admitted into the cylinder.

Fig. 13 illustrates an unloading device and regulator applied to the Ingersoll-Sergeant compressor.

The purpose of this unloading device is to maintain a uniform air pressure in the receiver and a uniform speed of engine, notwithstanding the consumption of the air, and to do this without waste of power or attention on the part of the engineer. A weighted valve of safety valve pattern is attached to the air cylinder, and is connected with the air receiver, and with a discharge valve on each end of the air cylinder, also with a balanced throttle valve in the steam pipe. When the pressure of the air gets above the desired point in the receiver, the valve is lifted and the air is exhausted from behind the discharge valves, thus letting the compressed air at full receiver pressure into the cylinder at both ends, and balancing the engine. At the same instant the compressed air is exhausted from the little piston connected with the balanced steam valve and the steam is automatically throttled, so that only enough steam is admitted to keep the engine turning around, or to overcome the friction, no work being done.

When the compressor is unloaded, it is evident that the function of the air piston is merely to force the compressed air through the discharge valves and passages from one end to the other until more compressed air is required, this being indicated by a fall in the receiver pressure. The weighted valve now closes and the small connecting pipes are instantly filled with compressed air; the steam valve automatically opens, and the compression goes on in the regular way. Another function of this device is to prevent the compressor from stopping or getting on the center. Direct-acting compressors are liable to center when doing work at slow speed.

Fig. 15 illustrates the Ingersoll-Sergeant Air Cylinder and Piston.

Fig. 16 shows the piston inlet valve, situated at G in Fig. 15. Two of these valves are placed in each piston of a double-acting air cylinder, the piston being hollow and the free air being admitted through a tail-rod pipe, letter E, Fig. 15. JJ are water jacket passages for cooling the air during compression. Owing to the absence of inlet valves, large water jackets are provided, not only around the cylinder itself, but through the heads. As the heat of compression is greater near the end of the stroke, the advantage of a cool head is manifest. H H are the discharge valves through which the compressed air is forced.

The most interesting feature of this cylinder is the piston inlet valve. It is evident that this valve being attached to the piston needs no springs or other connections, but is opened and closed exactly at the right time by its natural inertia. With only about ¼ of an inch throw of valve a large area is opened, through which the free air is drawn. The valve is made of a single piece of composition metal and is practically indestructible. Its construction is such that it fills the clearance spaces to a greater extent than is usual in air compressors. A singular feature is that indicator cards taken on these cylinders show a free air line in some cases a little above the atmospheric line. Poppet valve compressors almost invariably show a slight vacuum, due to several causes, mainly the duty performed in compressing the springs of the valves, but the vacuum is also influenced by insufficiency of valve area, hot air cylinders, etc. This cylinder gives its full volume of air, and apparently a little more at times, because the air is admitted by a concentrated inlet in which free air is always moving in one direction. After it has been started, the speed of the compressor is such that the air attains a momentum due to its velocity and density; this serves a useful purpose in piling up the free air in the cylinder before the inlet valve closes on the return stroke.

Taken from a 16x18 Sergeant piston inlet air compressor,
meyer's cut-off at 3/10. Steam at 58 lb.; air pressure,
77 lb.; total engine friction, 5 per cent.

Fig. 17 illustrates a combined steam and air indicator card taken from one of these cylinders. It will be observed that with steam and air cylinders equal in diameter and stroke, an air pressure of 77 pounds is reached with a steam pressure of only 58 pounds. The reason for this is plainly shown in the cards, their areas being nearly equal. What is made up in the air card by high pressure is represented in the steam card by greater volume. The indicated efficiency deduced from these cards is 95 per cent., that is, the area of the air card divided by the area of the steam card, representing the resistance divided by the power, results in 95 per cent. While several cards have been taken on the cylinders showing a loss by friction of only 5 per cent., yet on the average the best practice shows a loss of 6 per cent. or an efficiency of 94 per cent. This result indicates an almost perfect proportion between power and resistance, and good workmanship in air-compressing machinery. It is difficult to conceive an engine of this size being worked with a less expenditure for friction than 5 or 6 per cent. Were it possible to retain the heat which is in the air, and which is represented by the space between the dotted isothermal curve and the actual curve, we might attain high efficiency in using compressed air power, but it is evident that the power represented by the area of this space will be lost by radiation of heat before it is used in an engine situated several hundred feet away.

These indicator cards show at a glance that heat is responsible for the important air losses, and that so far as the design of the compressing engine is concerned, we have attained a point very near perfection. All the devices, past, present and future, on which inventors spend so much time, and in the development of which capitalists are innocently inveigled, aim to save this six per cent. loss! We hear a good deal about "Centrifugal Air Compressors," "Rotaries," "Plunger Pumps," etc., designs involving expensive complications without any heat advantage, and which seem to be based upon the "iridescent dream" of a large loss in the present method of compressing air. Here we have a simple engine, compact and complete in itself, capable of high speed without injury, constructed on the basis of our best steam engine practice, which produces compressed air power at a loss of only six per cent.

Clearance is not taken into consideration in the foregoing figures, but clearance is very much more of a bete noir in theory than in practice. The early designers, as shown in the "Dubois-Francois" illustrations, Figs. 3 and 4, regarded clearance loss as a very serious matter. Even at the present time some air compressor manufacturers admit water through the inlet valves into the air cylinder, not so much for the purpose of cooling as to fill up the clearance space. A long stroke involving expensive construction is usually justified by the claim that a large saving is effected by reduced clearance loss. Let us see what the effect of this clearance is. Assuming that we have an air compressor which shows an isothermal pressure line, there would be some loss of power due to clearance space, because we would have a certain volume of air upon which work was done and heat produced, that heat having been absorbed and the air being retained in the cylinder and not serving any useful purpose. But let us assume that we have a compressor which shows an adiabatic pressure line. We now have the air in the clearance space acting precisely as a spring, compressed at each stroke, retaining its heat of compression, and giving it out against the air piston at the point when the stroke is reversed. There is no loss of power in such a case as this, but, on the contrary, the air spring is useful in overcoming the inertia of the piston and moving parts. The best air compressors give a result about midway between the isothermal and the adiabatic, and the net loss of power directly due to clearance is so small as to be practically unworthy of consideration.

It must not be inferred from the preceding remarks that the designer of an air compressor may neglect the question of clearance. On the contrary, it is a very important consideration. If we assume a large clearance space in the end of an air cylinder of a compressor which is furnishing air at a high pressure, we may readily conceive that space to be so large, and that pressure so high, that the entire volume of the cylinder would be filled by the air from the clearance space alone, and the compressor would take in no free air and would, of course, produce no compressed air.

Loss in capacity of air compressors by clearance is in direct proportion to the pressure.

Owing to the loss of capacity by clearance space at high pressures, it is important that compound air cylinders should be used for furnishing air at high pressure. With compound air cylinders the air is compressed to alternate stages of pressure in the different cylinders, and the clearance loss is thus reduced because of the reduced density of the air in the clearance spaces. In ordinary practice air compressors deliver the air at less than 100 pounds pressure, so that with a properly designed air cylinder the clearance space is so small that the capacity of the compressor is not materially affected.

Two systems are in use by which the heat of compression is absorbed, and the difference between one and the other is so distinct that air compressors are usually divided into two classes (1) wet compressors, (2) dry compressors.

A wet compressor is that which introduces water directly into the air cylinder during compression.

A dry compressor is that which introduces no water into the air during compression.

Wet compressors may be subdivided into two classes.

(1) Those which inject water in the form of a spray into the cylinder during compression.

(2) Those which use a water piston for forcing the air into confinement.

The injection of water into the cylinder is usually known as the Colladon idea. Compressors built on this system have shown the highest isothermal results, that is, by means of a finely divided spray of cold water the heat of compression has been absorbed to a point where the compressed air has been discharged at a temperature nearly equal to that at which it was admitted to the cylinder. The advantages of water injection during compression are as follows:

(1) Low temperature of air during compression.

(2) Increased volume of air per stroke, due to filling of clearance spaces with water and to a cold air cylinder.

(3) Low temperature of air immediately after compression, thus condensing moisture in the air receiver.

(4) Low temperature of cylinder and valves, thus maintaining packing, etc.

(5) Economical results, due to compression of moist air (see table 3).

TABLE 3.—SHOWING THE RELATIVE QUANTITY OF WORK REQUIRED TO COMPRESS A GIVEN VOLUME AND WEIGHT OF AIR, BOTH DRY AND MOIST—ALSO RELATIVE VOLUMES WITH AND WITHOUT INCREASE OF TEMPERATURE FROM COMPRESSION.

_______________________________________________________________________________________
  |                |                                  |
  |Compression at  |Compression                       |
  |a Constant      |with                              |
  |Temperature.    |Increase of                       |
  |Mariotte's Law. |Temperature.                      |
__|________________|__________________________________|________________________________
  |     |     |    |     |     |    |     |     |     |    |     |     |   |     |
 1|0.1  |     |    |     |     |    | 20  | 68  |1.0  |    |     | 68  |   |     |
 2|0.5  | 7199|1468|0.612| 7932|1618| 85.5|186  |1.222| 733|0.092|111  |3.0|23500|22500
 3|0.333|11356|2316|0.459|13360|2725|130.4|267  |1.375|2004|0.150|135.5|4.0|37000|35000
 4|0.25 |14260|2909|0.374|17737|3618|165.6|330  |1.495|3477|0.196|153.5|4.8|48500|45000
 5|0.200|16580|3383|0.320|21209|4326|195.3|384  |1.595|4629|0.213|167  |5.4|58500|52500
 6|0.167|18475|3768|0.281|24310|4959|220.5|429  |1.681|5835|0.240|179  |6.0|67000|60000
 7|0.143|20038|4087|0.252|27048|5517|243.2|470  |1.758|7040|0.260|190  |6.4|75000|66000
 8|0.125|21422|4370|0.229|29518|6021|263.6|506.5|1.828|8096|0.274|     |   |     |
 9|0.111|     |    |0.210|     |    |282  |539.6|1.891|    |     |     |   |     |
10|0.100|     |    |0.195|     |    |299  |570.2|1.950|    |     |     |   |     |
_______________________________________________________________________________________
  |     |     |    |     |     |    |     |     |     |    |     |     |   |     |
 1|  2  |  3  |  4 |  5  |  6  |  7 |  8  |  9  | 10  | 11 |  12 |  13 | 14|  15 |  16
__|_____|_____|____|_____|_____|____|_____|_____|_____|____|_____|_____|___|_____|_____


Column  Heading
1       Tension in Atmospheres.
2       Volume.
3       Work of Compression. Cubic Meters in Kilogram-meters.
4       Work of Compression. Cubic Feet in Foot Pounds.
5       Volume.
6       Work of Compression. (Dry.) Cubic Meters in Kilogram-meters.
7       Work of Compression. (Dry.) Cubic Feet in Foot Pounds. Deduced from 6.
8       Temperatures. (Dry.) Cent.
9       Temperatures. (Dry.) Fah.
10      Ratio of Greater to Less Temperature. Absolute.
11      Loss of Work in Compressing one Cubic Meter in Kilogram-meters.
        By Increase of Temperature alone.
12      Percentage of Work of Compression Converted into Heat and Lost.
        By Increase of Temperature alone.
13      Final Temperature if Water is used in Compression. Fah.
14      Percentage of Water to Air Required.
15      Foot Pounds to Compress One Pound Air. Dry.
16      Foot Pounds to Compress One Pound Air. With sufficient Moisture.

The first advantage is by far the most important one, and is really the only excuse for water injection in air compressors. We have seen (table 3) that the percentage of work of compression which is converted into heat and loss when no cooling system is used is as follows:

Compressing to 2 atmospheres loss  9.2 per cent.
   "         " 3     "        "   15.0  "   "
   "         " 4     "        "   19.6  "   "
   "         " 5     "        "   21.3  "   "
   "         " 6     "        "   24.0  "   "
   "         " 7     "        "   26.0  "   "
   "         " 8     "        "   27.4  "   "

We see that in compressing air to five atmospheres, which is the usual practice, the heat loss is 21.3 per cent., so that if we keep down the temperature of the air during compression to the isothermal line, we save this loss. The best practice in America has brought this heat loss down to 3.6 per cent. (old Ingersoll Injection Air Compressor), while in Europe the heat loss has been reduced to 1.6 per cent. Steam-driven air compressors are usually run at a piston speed of about 350 feet per minute, or from 60-80 revolutions per minute of compressors of average sizes, say 18" diameter of cylinder. Sixty revolutions per minute is equal to 120 strokes, or two strokes per second. An air cylinder 18" in diameter filled with free air once every half second, and at each stroke compressing the air to 60 pounds, and thereby producing 309 degrees of heat, is thus, by means of water injection, cooled to an extent hardly possible with mere surface contact. The specific heat of water being about four times that of air, it readily takes up the heat of compression.

A properly designed spray system must not be confused with the numerous devices applied to air cylinders, by means of which water is introduced. In some cases the water is merely drawn in through the inlet valves. In others it passes through the center of the piston and rod, coming in contact with the interior walls of the air cylinder between the packing rings. Introducing water into the air cylinder in any other way, except in the form of a spray, has but little effect in cooling the air during compression. On the contrary, it is a most fallacious system, because it introduces all the disadvantages of water injection without its isothermal influence. Water, by mere surface contact with air, takes up but little heat, while the air, having a chance to increase its temperature, absorbs water through the affinity of air for moisture, and thus carries over a volume of saturated hot air into the receiver and pipes, which on cooling, as it always does in transit to the mine, deposits its moisture and gives trouble through water and freezing. It is, therefore, of much importance to bear in mind that unless water can be introduced during compression to such an extent as to keep down the temperature of the air in the cylinder, it had better not be introduced at all.

If too little water is introduced into an air cylinder during compression, the result is warm, moist air, and if too much water is used, it results in a surplus of power required to move a body of water which renders no useful service. The following table deduced from Zahner's formula gives the quantity of water which should be injected per cubic foot of air compressed in order to keep the temperature down to 104 degrees Fah.

  _________________________________________________________________________
                 |                 |                  |
                 |                 |Weight of water   |Weight of water
                 |                 |to be injected at |to be injected at
                 |Heat units devel-|68° Fah. to keep  |68° Fah. to keep
  Compression    |oped in 1 lb.    |the temperature at|the temperature at
  by atmosphere  |free air by      |104° Fah. in lbs. |104° Fah. in lbs. of
  above a volume.|compression.     |of water and per  |water for 1 cubic
                 |                 |lb. of free air.  |foot of free air.
  _______________|_________________|__________________|____________________
                 |                 |                  |
         2       |        3.702    |        0.734     |        0.056
         3       |        5.867    |        1.664     |        0.089
         4       |        7.406    |        1.469     |        0.113
         5       |        8.598    |        1.701     |        0.131
         6       |        9.570    |        1.891     |        0.145
         7       |       10.398    |        2.063     |        0.158
         8       |       11.109    |        2.204     |        0.167
         9       |       11.740    |        2.329     |        0.179
        10       |       12.301    |        2.440     |        0.188
        11       |       12.813    |        2.542     |        0.195
        12       |       13.278    |        2.634     |        0.202
        13       |       13.706    |        2.719     |        0.209
        14       |       14.102    |        2.798     |        0.215
        15       |       14.471    |        2.871     |        0.223
  _______________|_________________|__________________|____________________

Objections to water injection are as follows:

(1) Impurities in the water, which, through both mechanical and chemical action, destroy exposed metallic surfaces.

(2) Wear of cylinder, piston and other parts, due directly to the fact that water is a bad lubricant, and as the density of water is greater than that of oil, the latter floats on the water and has no chance to lubricate the moving parts.

(3) Wet air arising from insufficient quantity of water and from inefficient means of ejection.

(4) Mechanical complications connected with the water pump, and the difficulties in the way of proportioning the volume of water and its temperature to the volume, temperature and pressure of the air.

(5) Loss of power required to overcome the inertia of the water.

(6) Limitations to the speed of the compressor, because of the liability to break the cylinder head joint by water confined in the clearance spaces.

(7) Absorption of air by water.

Before the introduction of condensing air receivers, wet air resulting in freezing was considered the most serious obstacle to water injection; but this difficulty no longer exists, as experience has conclusively demonstrated that a large part of the moisture in compressed air may be abstracted in the air receiver. Even in the so-called dry compressors a great deal of moisture is carried over with the compressed air, because the atmosphere is never free from moisture. This subject will be referred to more fully when treating of the transmission of compressed air.

By far the most serious obstacle to water injection, and that which condemns the wet compressor, is the influence of the injected water upon the air cylinder and parts. Even when pure water is used, the cylinders wear to such an extent as to produce leakage and to require reboring. The limitation to the speed of a compressor is also an important objection. The claim made by some that the injected water does not fill the clearance spaces, but is aerated, does not hold good, except with an inefficient injection system. The writer has increased the speed of an air compressor (cylinders 12 in. and 12 in. by 18 in., injection air cylinder) ten revolutions per minute by placing his fingers over the orifice of the suction pipe of the water pump. The boiler pressure remained the same, the cut-off was not changed and the air pressure was uniform, hence this increase of speed arose from the fact that the water was restricted and the clearance spaces were filled with compressed air, which served as a cushion or spring. While the volume of compressed air furnished by this compressor would be somewhat reduced by the restriction of the water, yet the increase in speed which was obtained without any increase of power fully compensated for the clearance loss.

Mr. John Darlington, of England, gives the following particulars of a modern air compressor of European type:

"Engine, two vertical cylinders, steam jacketed, with Meyer's expansion gear. Cylinders, 16.9 inches diameter, stroke 39.4 inches; compressor, two cylinders, diameter of piston, 23.0 inches; stroke 39.4 inches; revolutions per minute, 30 to 40; piston speed 39 to 52 inches per second, capacity of cylinder per revolution, 20 cubic feet: diameter of valves, viz., four inlet and four outlet, 5½ inches; weight of each inlet valve, 8 lb.; outlet, 10 lb.; pressure of air, 4 to 5 atmospheres. The diagrams taken of the engine and compressor show that the work expended in compressing one cubic meter of air to 4.21 effective atmospheres was 38,128 lb. According to Boyle and Mariotte's law it would be 37,534 lb., the difference being 594 lb., or a loss of 1.6 per cent. Or if compressed without abstraction of heat, the work expended would in that case have been 48,158. The volume of air compressed per revolution was 0.5654 cubic meter. For obtaining this measure of compressed air, the work expended was 21,557 pounds. The work done in the steam cylinders, from indicator diagrams, is shown to have been 25,205 pounds, the useful effect being 85½ per cent. of the power expended. The temperature of air on entering the cylinder was 50 degrees Fah., on leaving 62 degrees Fah., or an increase of 12 degrees Fah. Without the water jacket and water injection for cooling the temperature it would have been 302 degrees Fah. The water injected into the cylinders per revolution was 0.81 gallon."

We have in the foregoing a remarkable isothermal result. The heat of compression is so thoroughly absorbed that the thermal loss is only 1.6 per cent.; but the loss by friction of the engine is 14.5 per cent., and the net economy of the whole system is no greater than that of the best American dry compressor, which loses about one-half the theoretical loss due to heat of compression, but which makes up the difference by a low friction loss.

The wet compressor of the second class is the water piston compressor, Fig. 18.

The illustration shows the general type of this compressor, though it has been subject to much modification in different places. In America, a plunger is used instead of a piston, and as it always moves in water the result is more satisfactory. The piston, or plunger, moves horizontally in the lower part of a U shaped cylinder. Water at all times surrounds the piston, and fills alternately the upper chambers. The free air is admitted through a valve on the side of each column and is discharged through the top. The movement of the piston causes the water to rise on one side and fall on the other. As the water falls the space is occupied by free air, which is compressed when the motion of the piston is reversed, and the water column raised. The discharge valve is so proportioned that some of the water is carried out after the air has been discharged. Hence there are no clearance losses.

This hydraulic compressor seems to have a certain charm about it, which has resulted in its adoption in Germany, France and Belgium, and by one of the largest mines in the United States. Its advantages are purely theoretical, and without certain adjuncts which have been in some cases applied to it, even the theory is a very bad one.

The chief claim for this water piston compressor is that its piston is also its cooling device, and that the heat of compression is absorbed by the water. So much confidence seems to be placed in the isothermal features of this machine that usually no water jacket or spray pump is applied. Mr. Darlington, who is one of the stanch defenders of this class of compressors, has found it necessary to introduce "spray jets of water immediately under the outlet valves," the object of which is to absorb a larger amount of heat than would otherwise be effected by the simple contact of the air with the water-compressing column. Without such spray connections, it is safe to say that this compressor has scarcely any cooling advantages at all, so far as air cooling is concerned. Water is not a good conductor of heat. In this case only one side of a large body of air is exposed to a water surface, and as water is a bad conductor, the result is that a thin film of water gets hot in the early stage of the stroke and little or no cooling takes place thereafter. The compressed air is doubtless cooled before it gets even as far as the receiver, because so much water is tumbled over into the pipes with it, but to produce economical results the cooling should take place during compression.

Water and cast iron have about the same relative capacity for heat at equal volumes. In this water piston compressor we have only one cooling surface, which soon gets hot, while with a dry compressor, with water jacketed cylinders and heads, there are several cold metallic surfaces exposed on one side to the heat of compression, and on the other to a moving body of cold water.

But the water piston fraternity promptly brings forward the question of speed. They say that, admitting that the cooling surfaces are equal, we have in one case more time to absorb the heat than in the other. This is true, and here we come to an important class division in air compressing machinery—high speed and short stroke as against slow speed and long stroke. Hydraulic piston compressors are subject to the laws that govern piston pumps, and are, therefore, limited to a piston speed of about 100 feet per minute. It is quite out of the question to run them at much higher speed than this without shock to the engine and fluctuations of air pressure due to agitation of the water piston. The quantity of heat produced, that is, the degree of temperature reached, depends entirely upon the conditions in the air itself, as to density, temperature and moisture, and is entirely independent of speed. We have seen that it is possible to lose 21.3 per cent. of work when compressing air to five atmospheres without any cooling arrangements. With the best compressors of the dry system one-half of this loss is saved by water jacket absorption, so that we are left with about 11 per cent., which the slow moving compressor seeks to erase. We are quite safe in saying that the element of time alone in the stroke of an air compressor could not possibly effect a saving of more than half of this, or 5½ per cent. Now, in order to get this 5½ per cent. saving, we reduce the speed of an air-compressing engine from 350 feet per minute to 100 feet per minute. We must, therefore, in one case have a piston area three and one-half times that of the other in order to get the same capacity of air, and in doing this we build an engine of enormous proportions with heavy moving parts. We load it down with a large mass of water, which it must move back and forth during its work, and thus we produce a percentage of friction loss alone equal to twice or even three times the 5½ per cent. heat loss which is responsible for all this expense in first cost and in maintenance, but which really is not saved after all unless water injection in the form of spray also forms a part of the system.

It is obvious that cost of construction and maintenance have much to do with the commercial value of an air compressor. The hydraulic piston machine not only costs a great deal more in proportion to the power it produces, but it costs more to maintain it, and it costs more to run it. It is not an uncommon thing to hear engineers speak of the hydraulic piston compressor as the "most economical" machine for the purpose, but that it is so "expensive" and takes up so much room, and requires such expensive foundations that, unless persons are "willing to spend so much money," they had better take the next best thing, a high speed machine. We hear of "magnificent air-compressing engines, the largest in the country," and pilgrimages are made to see these artificial wonders when, not unlike the old pyramids, they represent a pile of inert matter—a monument to moneyed kings.

The hydraulic piston compressor has one solitary advantage, and that is, it has no dead spaces. It was conceived at a time when dead spaces were very serious conditions—were positive specters! Valves and other mechanism connected with the cylinder of an air compressor were once of such crude construction that it was impossible to reduce the clearance spaces to a reasonable point, and, furthermore, the valves were heavy and so complicated that anything like a high speed would either break them or wear them out rapidly, or derange them so that leakages would occur. But we have now reduced inlet and discharge valves and all other moving parts connected with an air cylinder to a point of extreme simplicity. Clearance space is in some cases destroyed altogether by what is, as it were, an elastic air head which is brought into direct contact with the piston. All this reduces clearance to so small a point that it has no influence of any consequence. The moving parts are made extremely simple, even arriving at a point where inlet valves are opened and closed by their natural inertia. Mr. Sturgeon, of England, has applied a most ingenious and successful inlet valve, which is opened and closed by the friction of the air piston rod through the gland. We have, therefore, reached a point at which high speed is made possible.

Long-stroke air compressors are evidently objectionable on the basis of greater expense of construction. All the parts must be larger and heavier. The fly wheels are increased enormously in diameter and weight, and the strength of bearings must be enlarged in proportion. It is difficult to equalize power and resistance in air compressors with long strokes. The speed will be jerky, and when slow, the fly wheel rather retards than assists in the work of compression. This action tends to derange the parts and makes large bearings a necessity. The piston in a long-stroke compressor travels through considerable space before the pressure reaches a point where the discharge valve opens, and after reaching that point it has to go on still further against a prolonged uniform resistance. This makes rotative speed difficult. During the early part of the stroke, the energy of the steam piston must be stored up in the moving parts, to be given out when the steam pressure has been reduced through an early cut-off. With a short stroke and a large diameter of steam cylinder we are able to get steam economy or early cut-off and expansion without the complications of compounding.

I use material terms because they add to simplicity of expression and notwithstanding the fact that heat is vibration.

[Transcribers note: last digit illegible]

About HackerNoon Book Series: We bring you the most important technical, scientific, and insightful public domain books.

This book is part of the public domain. Various (2004). Scientific American Supplement, No. 799, April 25, 1891. Urbana, Illinois: Project Gutenberg. Retrieved https://www.gutenberg.org/cache/epub/11649/pg11649-images.html

This eBook is for the use of anyone anywhere at no cost and with almost no restrictions whatsoever. You may copy it, give it away or re-use it under the terms of the Project Gutenberg License included with this eBook or online at www.gutenberg.org, located at https://www.gutenberg.org/policy/license.html.