Steam engines work by using the expansion of high-pressure steam to push against moving pistons in the cylinders of reciprocating engines, or against the moving vanes of steam turbines. When the Berkeley and the Medea were built, steam piston engines, after a century of development, were universal for marine use, and they continued up through the Liberty Ships of World War II that deliberately used an old design that was easy to build..
Change was coming. In the year that Berkeley was being designed, an experimental turbine-powered vessel sneaked into the British Navy's parade for Queen Victoria's Golden Jubilee, and outran Britain's fastest destroyers. Ten years later, the Diesel engine came into marine use, at first for small vessels, so that nowadays ships are powered either by turbines (steam or gas) or by diesels. The steam reciprocating engine, nowadays, is just a historical artifact, but it represented the very best that the technology of the time could produce.
The Berkeley's steam engine and boilers are typical of almost the highest development of the marine reciprocating piston engine, the three cylindered triple-expansion engine supplied with steam from the oil-fired straight watertube boilers. The Medea's steam engine and boiler, although built six years later, are typical of the previous generation of steam plants, the two-cylindered compound double-expansion engine supplied with steam from a firetube boiler, originally burning coal but since converted to oil.
The Berkeley was originally equipped with two firetube Scotch boilers, burning coal, similar in principle to the boiler of the Medea but different in design. After two years of service, the Berkeley was converted to oil, the same fuel used by the locomotives of her owner, the Southern Pacific Railroad. After twenty-six years of service, the two Scotch boilers were replaced by the present four watertube boilers.
The steam engine plant consists of far more than the steam engine itself. The water and steam run through a continuous cycle of water boiled into steam, used in the engine, condensed back to water, and then pumped back into the boiler for reuse, a cycle first studied scientifically by the French engineer Sadi Carnot in 1824, about a century after the first steam engines had been built.
Therefore, the complete steam plant consists of a furnace in which to burn the fuel, the boiler in which the heat from the furnace is used to boil water into steam, the steam engine which turns the heat of the steam into mechanical power, the condenser which condenses the steam back into water, and the feed pump that pumps the water back into the boiler. We will look at each of these machines, but first you must learn a little about water and steam.
Carnot showed that the possible efficiency of any heat engine depends on the temperature difference between the highest and the lowest temperatures of the working fluid, the water and the steam in our ships, divided by the highest temperature. The greater the temperature range, and the hotter the highest temperature, the more power produced for the fuel burned. The highest temperature is limited by the strength of the materials from which the boiler, steam pipes, and engine are made. In the highest development of the piston steam engine, the limit was set by the lubricating oil for the valve and piston; the steam was so hot that even special thick oil ran like water, but the Berkeley never approached those temperatures. Her maximum steam temperature was about 425 °F. The lowest usable temperature is that of the surroundings. There's no point in making a lower temperature, as in a refrigerator, because it costs much more energy to "make the cold" than you could get by using that "cold." (That's also explained by the Carnot cycle, for a refrigerator is a heat engine run in reverse.) However, since a ship floats in an infinite supply of cold water, the temperature of the seawater is the lowest temperature we can use. With the temperature range that the Berkeley used, the theoretical efficiency could be no higher than about 30%. Probably, only about 10% or 15% of the energy in the fuel could be actually used in driving the ship.
Now that you have learned that the engine must work between the highest usable temperature and the lowest obtainable temperature, you need to know the properties of water and steam between those temperatures. You probably know that when water boils into steam it gets much larger, occupies much more space or volume. You probably also know that the air pressure around you at sea level is about 15 pounds per square inch. You probably all know that the boiling temperature of water into steam is 212 °F, and some of you know that if you go to higher altitudes, where the air pressure is lower, water boils at a lower temperature, making cooking slower, while if you confine the steam in a pressure cooker the temperature goes up to make cooking quicker.
There is an easy explanation for this. As water, the molecules are very close together, sticking together in fact, although they slip past each other, roll over each other, with little friction, making water a liquid. As steam, the same molecules rush about independently of each other, taking up much more space and making steam a gas, just like any other gas.
Temperature really refers to the speed with which the molecules of water are moving. At room pressure, very few of the molecules move fast enough to jump into the air against the pressure of the air and take up much more space as cold steam, or water vapor, or humidity, whatever you want to call it. That's room-temperature evaporation. As temperature increases, more and more of the molecules of water get fast enough to make the jump between the liquid and the gas against the pressure of the gas. At boiling temperature, enough of the molecules are going fast enough to push the air away and fill the space with steam. If the space is enclosed, as in a pressure cooker or a boiler, this increases the pressure. If the pressure is increased, then it is more difficult for the water molecules to jump into the gas, the steam, and take up more space, and some go slow enough to be captured again into the water.
Therefore, for every pressure there is a temperature at which water and steam can exist together. Contrariwise, for each temperature, there is a pressure at which water and steam can exist together.
The steam engine designer always works in absolute pressure, starting from zero pressure, as in the vacuum of outer space. This means that sea-level pressure is 15 psia, meaning pounds per square inch absolute. You must always remember that we exist at 15 psia. (The operating engineer considers ambient pressure to be zero. Thus his steam pressure gauges read pressure above ambient, or 15 psi less than the absolute pressure, and he measures pressure that is lower than ambient in inches of mercury of vacuum, which you will see on the gauges in the engineroom. 30 inches of vacuum is substantially equal to zero pressure.)
Here is a short table of steam pressures, temperatures, and volumes.
|
Pressure, Pounds per Square Inch, Absolute |
Boiling Temperature Degrees Fahrenheit |
Volume, Cubic Feet per Pound |
|---|---|---|
|
1 |
102 |
335 |
|
2 |
126 |
174 |
|
5 |
162 |
73 |
|
10 |
193 |
38 |
|
15 |
213 |
26 |
|
20 |
228 |
20 |
|
40 |
267 |
10.4 |
|
60 |
293 |
7.1 |
|
80 |
312 |
5.4 |
|
100 |
328 |
4.4 |
|
120 |
341 |
3.7 |
|
140 |
353 |
3.2 |
|
160 |
363 |
2.8 |
|
180 |
373 |
2.5 |
|
200 |
382 |
2.3 |
When considering the volume of a pound of steam, consider that a pound of water has a volume of only 0.016 cubic feet. Therefore, even at 200 psia, when water turns into steam, its volume increases by about 14 times.
You can see that if you can use the cold seawater as the lowest temperature, say 60 °F, the pressure would be less than 1 psia, and 1 pound of steam would require more than 340 cubic feet of space. The high-pressure cylinder of the Berkeley would hold about 2 pounds of steam, so that if that were expanded all the way to less than 1 psia pressure, it would occupy about 700 cubic feet, about the volume of a bathroom. That's just too large a cylinder to be practical. Therefore, the steam was not expanded to this volume, but to the volume allowed by the largest practical cylinder.
However, that doesn't say that the lowest cylinder pressure was about 5 psia. That was the pressure at which the steam was exhausted from the cylinder, but all during the stroke the back side of the piston was under the least pressure that it was possible to obtain, that of steam at the temperature of the cooling seawater, and hence less than 1 psia. If the condenser had any significant pressure in it, that is, it was showing less than 30 inches of vacuum, the engineer investigated to see what was wrong.
The very earliest engines didn't run on any significant steam pressure, only on the difference between steam at room pressure, 212 °F, and the vacuum of steam at room temperature, 60 °F. They were very inefficient, partly because of the small temperature range, which Carnot's cycle showed could not be efficient. However, they showed that it was important to have as good a vacuum as possible as the lowest pressure in the system.
You have learned the relationship between the temperature and the pressure of steam for containers in which both steam and water exist together. This is called wet steam, because if the temperature drops at all, some of the steam condenses into a fog of water droplets. For the moment, consider that the engine is running on wet steam direct from the boiler.
Suppose that the boiler steam is at 180 psia, which is approximately the pressure in the Berkeley's boilers, and 373 °F (assuming wet steam for this discussion). This steam could be allowed to fill a cylinder, pushing its piston all the way down, and doing a lot of work. However, when it was time for the piston to come back up the cylinder, the steam would have to be allowed to escape from 180 psia and 373 °F into either the atmosphere or into the condenser. That would waste much of the energy that the boiler had worked so hard to put into the steam.
Instead of wasting that energy, only a small amount of steam is allowed into the cylinder at the top of the piston stroke. Then the supply of steam is cut off, and the pressure of the steam gradually drops as the steam pushes the piston downward, until the steam has expanded to about 5 psia and 162 °F, while the pressure against the opposite side of the piston is maintained at effective zero. Then all the energy is got from the steam that it is practically possible to get.
There are several troubles with having all the expansion in one cylinder, but the big theoretical problem that it is impossible to overcome is that the cylinder must start at 373 °F at the start and will be cooled by the steam inside to 162 °F by the end of the stroke. Then, for the next stroke, the cylinder has to be heated up again with steam at 373 °F, thus wasting a lot of steam that condenses into water as it heats up the metal of the cylinder and piston. Furthermore, that cylinder and piston must be very large and very heavy, heavy because they must be strong enough to withstand the initial pressure, but large enough to contain the fully expanded steam. This means that it will take an enormous amount of steam to heat them up for the start of each stroke.
The answer to this problem is to have multiple cylinders, each of which operates over only part of the total range of temperatures and pressures. Thus each cylinder is heated and cooled only part of the total range for each of its strokes.
The highest pressure that was used for single cylinder marine engines, (in marine engines efficiency of fuel is the paramount consideration), was about 45 psia and 275 °F. The first development was an engine in which the expansion occurred in two cylinders, the high-pressure cylinder and the low-pressure cylinder, called the compound engine. This is the type of engine installed in the Medea.
As technology advanced to make higher boiler pressures and temperatures practical, the number of expansions was increased to three, with the high-pressure, intermediate-pressure, and low-pressure cylinders, as in the Berkeley and on into the Liberty ships of World War II.
Some ships went to even higher boiler pressures and temperatures and used quadruple-expansion engines, but this development was cut off by the introduction of the turbine in place of the reciprocating piston engine. Some of the triple-expansion engines actually had four cylinders, with two low-pressure cylinders of medium size, to keep the reciprocating weights down and to reduce vibration and bearing loads.
In the highest power applications of the reciprocating piston marine engine, as in warships, the engineroom at full power was full of vapors of oil and water. Some of the bearings had to be water-cooled with hoses spraying on the rotating crankshaft. The forces required to change the direction of the pistons at each stroke limited the speed at which the engine could be run. The vibration was terrific. While oil was pumped into the steam for lubricating the valves and pistons, lubrication was still a problem and the oil had then to be removed from the condensed water before it could be returned to the boilers. The engines needed frequent maintenance.
The turbine engine had many advantages. Because it had pure rotational motion, it didn't produce vibration. It could be made with an enormous number of steam expansions, like a piston engine with many expansion cylinders. Because its wearing parts, the shaft bearings, were outside the steam chambers, their lubrication did not limit the steam temperatures that it could use. Because its vanes moved at very high speed, it took much less space for a given horsepower than the reciprocating engine with its slow-moving pistons. It cost more because it was more difficult to manufacture, but it did a much better job.
The four boilers are not the original boilers, which were two coal-burning Scotch fire-tube boilers, such as were installed in the Titanic fourteen years after the Berkeley was built. The present boilers are oil-burning Babcock and Wilcox cross-type straight-tube water-tube boilers, a typical installation for merchant ships and industrial plants of the time, probably 1926 1.
Fire-tube boilers have a large barrel containing the boiling water and steam, with tubes running from end to end through which the fire and combustion products flow to heat the water. Water-tube boilers are the opposite. Their tubes are full of water and steam and are heated by the fire on their outsides as they run through the firebox.
The Berkeley's boilers delivered steam at 165 pounds per square inch gauge pressure, or 180 psi absolute, which was superheated a further 50 degrees to 425 degrees F. 2
Across the top of each boiler is the horizontal steam drum, called that because it was about half full of water with the rest filled with steam, and from which the steam was taken. It was very important to keep the right amount of water in each boiler. Too much water, and you got water into the engine. Too little water, and the water tubes in the firebox would run dry, get overheated, and burst. You can see the glass tubes of the two water gauges for each boiler, and the three emergency water cocks for use if you can't see the glass gauges. If the water is at the correct level, you should get water out of the lower cock, steam out of the upper one, and mixed water and steam out of the middle one. You can also see the steam pressure gauges, which the fireman used as guides to regulate the amount of fuel being burned.
Looking into the firebox through its side opening, you see that the space above the fire is roofed by the sloping water tubes. The flames of the fire, and its hot gases, flow between and around the many layers of tubes that are above the fire. If you look into the side cleaning doors, you will see the many layers of tubes above the firebox. The tubes are sloped so that as the water boils into steam, the bubbles rise up the slope, making the water circulate throughout the boiler, from the drum, down the downcomers, up the sloping tubes as some of the water boils into steam, up the risers, and back to the steam drum, where the steam bubbles free out of the water and into the steam pipes and the water is returned for another circuit.
Riveted and forged into holes along the length of the steam drum are the many hollow forged steel downcomers, each with a zig-zag shape. The outer face of each downcomer has many square holes, each closed by a steel plate. Through some opened square holes you can see the ends of the water tubes, four per square hole, whose ends are expanded, steam-tight, into the inner sides of the downcomers. The water came down the hollow downcomers to enter the water tubes, which crossed the firebox.
On the other side of the firebox are risers, made just like the downcomers, from the top of which the mixture of water and steam returned, through large cross tubes, to the steam drum at the top of the boiler.
You can see the oil burners below the downcomers, giving off red light and a roaring sound, just as if oil were being burned today.
At the start I told you that whenever steam and water are together in a container, for each temperature there is one pressure. Raise the temperature, and more steam is made and the pressure goes up. Lower the pressure, say by using some steam in an engine, and some of the water boils into steam until the temperature is reduced to that appropriate to the pressure. I also told you that heat engines get more efficient if the high temperature can be raised.
That's what a superheater does. When the steam leaves the steam drum, it is no longer in contact with water, but is just a plain gas, just like air. The steam from the steam drum goes through another set of tubes that run across the top of the firebox above the water tubes. They can't be right in the fire, or, with no water in them, they would get red hot, weaken and burst. However, above the water tubes the combustion gases are just hot enough to heat the steam to a temperature that is still safe but makes the engine more efficient. That superheated steam can be expanded further in the cylinders of the engine before it starts condensing into water.
The big asbestos-covered pipe carries the steam from the boilers to the engine room. Each boiler has its own stop valve, so that any boiler can be shut down without shutting down the others. Somewhere up there, too, are the safety valves that would let steam escape up the relief pipe next to the stack if either boiler got too hot with too much pressure. The steam pipe makes the U-bends that you see above so that when it expands from room temperature as the hot steam enters it, it can flex a bit instead of trying to push the ship apart and breaking in the process.
Now we are in the engine room. The first thing the steam reaches is the throttle valve that controls the amount of steam reaching the engine from the boilers. There it is, up at the top where the steam pipe reaches the engine. The throttle valve is operated by this long rod here, from the throttle-valve lever that is at the engineer's station.
The Berkeley's engine is a three-cylinder, triple-expansion, vertical, double-acting engine. Triple-expansion means that the steam passes through the high-pressure, intermediate-pressure, and low-pressure cylinders in turn, each one larger than the one before but using steam at lower pressure and lower temperature. Vertical means that the cylinders are above the crankshaft, typical for both steam and diesel marine propeller installations.3 Double-acting means that the cylinders are closed at both top and bottom, so that the steam can push the piston both down and up, in contrast to the automobile engine in which the pistons can push only downward. These characteristics make the steam piston engine more complicated than a typical car engine.
Before you is the crankshaft with its three crank throws, one for each cylinder, equally spaced at 120 degrees of rotation. Each throw has its connecting rod going up to the crosshead guide. Because the cylinders are closed at both top and bottom, the connecting rod cannot go directly to the piston. The piston rod extends from the bottom of the cylinder, coming out through a steam-tight hole in the cylinder head, and is kept moving straight by the crosshead and its guide. The crosshead guide is that portion of the frame of the engine that is machined straight with the cylinder bore so the crosshead must move straight up and down in line with the cylinder. The connecting rod then connects this crosshead to the crankshaft.
The cylinders all have 36" stroke, but their diameters are different, as are their initial steam pressures: 22" dia @ 180 psia and 425°F; 34" dia @ 68 psia and 300°F; and 56" dia @ 27 psia and 244°F. The expansion ratio of the steam in each of the last two cylinders is easily calculated as the ratio of its volume to the volume of the one before. The expansion in the HP cylinder depends on the degree of cutoff of the high-pressure valve, which is probably about 40%, giving an expansion ratio of 2.5. The total expansion ratio of the whole engine is the product of each of the expansion ratios, or about 15 times. In going through the engine, the steam is expanded to about 15 times its initial volume, thus getting the maximum practical work out of it, according to the technology of the time.
|
Cylinder |
Initial Pressure, psia |
Initial Temp. °F |
Diameter, inches |
Capacity, cu. ft. |
Expansion Ratio |
|---|---|---|---|---|---|
|
HP |
180 |
425 |
22 |
7.92 |
2.4 |
|
IP |
68 |
300 |
34 |
18.9 |
2.4 |
|
LP |
27 |
244 |
56 |
51.3 |
2.4 |
|
Exh |
9 |
188 |
- |
- |
- |
You probably think of steam as an airy nothing, not weighing much at all. Well, it isn't. Since the capacity of the high-pressure cylinder is 7.9 cu. ft. and the cutoff is at about 40% of the stroke, with steam at boiler pressure, this cylinder will admit about 1.25 pounds of steam per stroke. At 2 strokes per revolution and 125 rpm, that is 9.4 tons of steam per hour. Nine tons of steam per hour is what it took to push the Berkeley along at 14 knots, all boiled from 9 tons of water in the boilers, used in the engine, condensed back into water in the condenser, and finally pumped back into the boilers.
The traditional way to measure the power of a reciprocating engine was to measure the pressure of the steam during each part of the stroke. For triple-expansion engines such as those on the Berkeley, this must be done for each cylinder and the results added. This measures the exact power of the steam provided in the engine, but does not account for the power that is absorbed by the engine in moving its own parts. The indicator produces a diagram that looks like a low boot or high shoe. The back of the shoe shows the increase in pressure when the steam is admitted. The level top of the diagram shows the continued pressure as the piston moves while the steam valve is still admitting steam. The curved sloping toe of the shoe shows the decrease in pressure as the steam expands after the steam supply is cut off. The tip of the toe shows the drop in pressure as the steam is exhausted, while the sole of the shoe shows the pressure during exhaust (condenser pressure or, in a multiple expansion engine, the pressure going to the next cylinder in the sequence). The area of the diagram indicates the power of each stroke of the piston.
The Berkeley's engine tested at 1,163 IHP (indicated horsepower) at 122.5 rpm. This gives 16 pounds of steam per indicated horsepower hour, which is about average for the time, design, and boiler pressure and temperature.
Each cylinder has its own steam-control valve, which admits higher-pressure steam to one end of the cylinder while allowing the lower-pressure steam to escape from the other end of the cylinder. When you looked down on the top of the engine from the main deck, you saw the three circular cylinder heads with, also, a smaller circular cover and two larger rectangular covers. The small circular cover encloses the piston-like valve of the high-pressure cylinder, while the rectangular covers enclose the flat slide valves of the other two cylinders.
Each slide valve operates inside a steam chest, a rectangular cavity which is filled with steam at high pressure. Each slide valve is like a flat, rectangular box with its open side pressed against the machined flat face of the cylinder casting. The valve is pressed against the cylinder face by the difference in pressure between the steam that surrounds it and the exhaust steam that is inside it. The cylinder face has three long, horizontal, narrow ports cast into it. The upper and lower ports connect to the top and bottom of the cylinder. The center port leads to a passage that goes out the side of the cylinder casting as the exhaust. When the valve moves down, it uncovers the upper cylinder port to let steam in to the top of the cylinder to push the piston down, while it connects the lower cylinder port to the exhaust port, to let the steam in the bottom of the cylinder out to exhaust. The top and bottom lips of the valve are designed with particular widths, so that the steam is allowed in and out at the points in the piston stroke desired by the designer.
The higher the pressure of the steam, the harder the valve is pressed against the port face and the more power it takes to move it. Therefore, when steam pressures increased, the flat slide valve was replaced by the piston valve for at least the high-pressure cylinder, as in the Berkeley. The piston valve is a circular piston with an hour-glass shape, in which the steam pressures pressing inward from each end balance each other, as does the exhaust pressure from the center outward toward each end. (Some piston valves had steam at the ends, others had the steam in the center, but they were balanced either way.)
Each valve is moved up and down by an eccentric and eccentric rod. You can see these eccentrics in pairs beside each crankshaft throw. Each eccentric is a circular disc mounted on the crankshaft, but mounted off-center (hence the name: eccentric). Therefore, as the crankshaft rotates, the eccentric appears to move up and down. Because each eccentric is circular, it can turn within the strap that surrounds it, thus pushing and pulling the eccentric rod up and down as the crankshaft turns. This up and down motion of the eccentric rod, which works like a connecting rod, is passed to the valve rod to drive the valve up and down. The eccentric is set a little more than 90 degrees ahead of the motion of the piston, so that the valve moves down to open the top end of the cylinder to steam when the piston is at the top of its stroke, ready to apply power in the new direction. At the same time, other part of the valve opens the other end of the cylinder to exhaust, letting the used steam leave the cylinder.
Now, remember what I said about using the steam expansively and how that made the engine more efficient? The valve is so made that it closes to steam when the piston has moved only partway through its stroke. Therefore, for the rest of the stroke, while the steam keeps pushing on the piston, its pressure and temperature fall. In fact, after the expansion in the high-pressure cylinder, the steam has cooled enough, because of the work that it has done on the high-pressure piston, that part of it has condensed into water in the form of fog.
As I just said, and as you can see, each cylinder has a pair of eccentrics, not just one, to operate its valve. One eccentric is positioned on the crankshaft for ahead rotation, the other for astern rotation. The two eccentric rods are connected to the opposite ends of the curved link that you see above you, and the block that runs in the slot of the link is the lower end of the valve operating rod. When that link is moved to one side, only one eccentric rod moves the valve, say in the timing required for going ahead, while when the link is moved to the other side, the other eccentric rod moves the valve, for movement in the opposite direction.
This link, called the Stephenson link because it was first designed in his locomotive design office, has another advantage. As the link is moved a little way from one end of its travel, both eccentrics contribute to the motion of the valve. This changes the cutoff, the position of the piston at which the valve stops supplying steam to the cylinder. This changes the amount that the steam will expand in the cylinder for the rest of the stroke. Changing this isn't particularly important for ships, which usually run at full design speed for most of their voyages, but for locomotives, which require enormous pulling force to start a train or to pull it up a grade, but which require much less force to keep the train rolling on the level, the variable cutoff allowed the use of long cuttoff and much steam, although inefficiently, for starting the train, but short cutoff, using less steam but using it more efficiently, for just rolling along. If the same pressure drop were obtained by partially closing the throttle valve, most of the energy already put into the steam would be lost and the engine would operate very inefficiently.
The link is curved to a circle with the same radius as the length of the eccentric rods, so that moving the link does not, of itself, change the position of the valve.
As I said, high-pressure steam from the boilers goes only to the valve of the high-pressure cylinder. That's fine when the engine is running; the steam has to wait in the steam pipe until the high-pressure valve opens to one end or the other of the cylinder. However, it is different when the engine is stopped and you want to start it. Remember what I said about using the steam expansively; the valve may be closed for both ends of the cylinder. Then you can't start the engine because no cylinder can receive steam. To start the engine, you may have to admit steam to the other cylinders, just until the engine starts moving. Well, one of the valves will be open to steam, if it could get steam. Up at the side of the cylinders are small steam pipes running from the throttle valve through small valves marked Bypass Valves to the valve chests of the intermediate- and low-pressure cylinders.
When the engineer needs to start the engine, and it won't start just by opening the throttle valve, he can open either of these bypass valves to let a little steam into the valve chests of the other cylinders just to get the engine moving. Once it is turning, he then closes the bypass valves to stop wasting high-pressure steam in the low-pressure cylinders.
The Stephenson links of the valve gears of all the cylinders are all shifted together, by the link rods, cranks, and wayshaft that connect them. For the Berkeley, this would have to be done for every crossing of San Francisco Bay. Working this by hand would be hard work, and slow. Therefore, this shaft is rotated by the power reverse cylinder. The steam to this cylinder is controlled by a small valve worked by the engineer through the lever that has detent stops marked Ahead and Astern for the power reverse valve positions that make the power reverse piston go up or down.
Also up alongside the cylinders you will see the other small valves marked Cylinder Drains, which are worked by long shafts from the engineer's position. Consider starting the engine from cold. When steam enters the cold engine, much more of the steam will condense until the valve chests, valves, cylinders and pistons heat up to the temperature of the steam. That will make a lot of water in the cylinders. If, when the piston approached the end of its stroke, the space remaining was filled with water instead of steam, the piston would hit the water just as hard if it had hit the cylinder head directly. That would cause great damage to the engine. Therefore, each cylinder is fitted with these drain cocks at each end, to let the water out as it forms, until the engine gets up to working temperature.
When you see movies of steam locomotives starting out, you often see bursts of steam blowing sideways from the cylinders. It looks spectacular, and sometimes is done just for show, but its real purpose is to blow the condensed water out of the cylinders until they get to operating temperature, just like the cylinder drains on the Berkeley's engine.
The low-pressure steam escapes from the low-pressure cylinder through the condenser trunk into the condenser, which is a large chamber that is cast as part of the frame of the engine. You can see the large rectangular condenser trunk extending downward from near the low-pressure cylinder. Compare this with the small steam pipe (much smaller than the insulation that encloses it), and you will have an idea of how much the steam expands when going through the engine. It expands about 13 times.
The condenser is the reverse of the boiler. Remember that the boiler takes water and passes it through tubes that are heated by the fire until it turns into steam. The condenser is a similar chamber through which a nest of tubes pass. However, these tubes convey the cooling water, while the exhaust steam fills the chamber. The condenser consists of three chambers, the center of which, and much the largest, is the vacuum chamber into which the exhaust steam flows. At each end of the condenser is a separate water chamber, closed by the cover that you see. These two water chambers are connected by many tubes that go straight through the main condenser chamber. Cold seawater is sucked in from overside by a circulating pump, passed into one water chamber, through the tubes, where the water picks up heat by condensing the steam that surrounds the tubes, is collected in the other water chamber, and then returns overside through the cooling water outlet.
Remember the discussion about steam at the beginning? Steam can exist at room temperature if the pressure is low enough, in a partial vacuum. If the engine is able to use very low-pressure, low-temperature steam, it will be more efficient. The condenser exposes the used steam to tubes cooled by cold seawater. Therefore, the pressure can be very low and the engine most efficient. However, the condenser will eventually fill up with water that is condensed from the steam (and with the small amount of air that was originally dissolved in the boiler feed water). Therefore, the condenser is emptied by the Condensate Pump. This is also called the Air Pump, because it removes both the condensed water and the air that accumulates with it. Because the water has a much smaller volume than the steam from the boiler, the condensate pump can be much smaller than the engine itself.
The condensate pump is a vertical pump with a single steam cylinder and two water cylinders. One water cylinder is driven directly by the steam cylinder, the other is driven by a rocking beam pivoted so that as one water piston descends, the other ascends. The steam valve is actuated by the motion of the rocking beam, so that when the piston moves down the valve is also moved down, ready to admit steam to the bottom of the cylinder to drive the piston up again.
You can guess how much water comes from the condenser by the sizes of the pipes that connect the condenser, the condensate pump, and the hot well.
The condenser must be continually cooled by a flow of cold seawater through its tubes. This water is pumped through the condenser and back overside by this centrifugal circulating pump. This circulates a lot of water, as you can guess by the sizes of the pipes that connect it to the condenser and overside. To cool 9 tons of steam per hour requires about 90 tons of seawater per hour. A centrifugal pump works something like a propeller, in which a rotating, bladed wheel spins the water in a circle, and hence increases its pressure so that it moves through the pipes. Centrifugal pumps are best at moving much water at low pressure, as used here for the condenser, and work at relatively high speed. This one is driven by a single-cylinder steam engine whose valve is operated by a single eccentric (because it is never reversed). There is also a second circulating pump of the cross-compound type.
A cross-compound pump consists of two pumps in one frame. Each pump has a steam cylinder that directly drives its pump cylinder. The reason that there are two pumps built into one frame is that pump #1 drives the valve for pump #2, and vice versa. You can see the valve linkage above the piston rods. Therefore, when pump #1 makes a stroke to end A, it shifts the valve that causes the steam to drive pump #2 to end B. As pump #2 reaches end B, it shifts the valve for pump #1, making it return to end B. The valve linkages are set up so that pump #1 always causes pump #2 to go to the same end as pump #1 is, while pump #2 always causes pump #1 to go to the opposite end as pump #2 is.
You will see many pumps of this design in the Berkeley's engine room. Cross-compound pumps are not particularly efficient, because they admit steam to the cylinder for the full length of the stroke, instead of using the steam expansively, but they are convenient when small amounts of fluid must be pumped.
Cross-compound pumps are used for bilge water removal, fresh water service, lubricating oil, fuel oil (in the boiler room), and the fire pump (in the engine room casing on the main deck).
The condensed water from the condenser goes from the condensate pump into these rectangular tanks, named the hot well. That is, they act as the well that supplies the boiler with water, and the water is warm, not quite cold. When in use, the tanks of the hot well were filled with loofas, a kind of vegetable sponge that is the skeleton of a particularly fibrous vegetable squash. Remember, to lubricate the engine's valves and pistons, oil was pumped into the steam just as it entered the engine. That oil comes out with the condensed water, but it should not be pumped back into the boilers. The loofa sponges absorbed the oil and were wrung out and replaced as they filled with oil.
The boiler water is used over and over again, for two reasons. At sea, there is no natural source of fresh water, and salt water ruins boilers. 4 Even where fresh water was available on shore, as for steam locomotives, the railroad system engineer had to be careful to site his water tanks where "boiler quality" water, free of sediment and dissolved minerals, was available. Once used, the water is of boiler quality, being distilled water. So even though the Berkeley had sources of fresh water on each side of the Bay, she used the water from the hot well over and over again, with only enough new water to replace that which was lost by leakage to the atmosphere.
That condensed water has to be pumped back into the boiler against the pressure of the steam in the boiler. There are two boiler pumps, because the boiler fires must be put out and the ship must stop if no boiler pump is working. The main boiler-feed pump is a single-cylinder, direct drive pump with the steam cylinder above the water cylinder, with the water valve chambers prominently in view. The valves inside are just flappers that fall over a grating. As the pump piston drives, the water is pumped out through the grating, lifting the flapper. When the pump piston goes in the opposite direction, the flapper falls suddenly onto the grating, preventing the water from returning. This sudden closing causes strong pressure pulses, water hammering, in the discharge pipe. On the discharge side there is a tall copper bell, that is kept full of air to act as a spring that smooths out the bumps in the water pressure as the water valves open and close.
The auxiliary boiler-feed pump is of the cross-compound design.
Connected to each end of the main engine's crankshaft are the propeller shafts to the propellers at each end of the ship. Each propeller will be pushing or pulling the ship along. That means that there must be a connection between each propeller and the hull of the ship, to transmit the force that moves the ship. Since the shaft must be rotating for the propellers to develop thrust, this connection must be a thrust bearing of some type. Since the engine's crankshaft should not be designed to take end thrust, at each end of the engineroom, where the shaft leaves that compartment, there is a thrust block to take the thrust, in whichever direction, of the propeller to which it is connected. Each thrust block consists of three collars rigidly mounted on the shaft, and on each side of each collar are two thrust rings that are fixed to the hull. These are all enclosed and run in a bath of oil, with oil cups to supply fresh oil, and water pipes for cooling water. As the shaft turns and pushes in one direction, its collars push up against the rings, which absorb the thrust while allowing the shaft to turn.
Each thrust block has three collars because only a single collar would scrape against its ring under the full thrust of the propeller. This is because, although the collars are continually supplied with oil, the oil is not forced between the collars and the rings. There is sufficient frictional loss that the thrust bearings must be cooled by water, delivered and returned through the pipes connected to each bearing. This was the original type of thrust block, and as ships increased in size and power, they became increasingly unreliable, liable to run hot and scrape metal to metal. This difficulty was corrected just about the time of Berkeley's design by the Kingsbury thrust bearing.
The Kingsbury bearing used only one collar. However, the ring was divided into six or eight segments, called slippers, which were mounted on a fixed ring. Each slipper presented a flat face against the shaft's collar, and a slightly curved face against the fixed ring, so it could rock a little in the direction that allowed the leading edge to lift away from the shaft's collar just a small amount. As the shaft turned, it picked up oil from the bath below. That oil was squeezed between the collar and the slipper, lifting the edge of the slipper so that the slipper was gliding on a wedge of oil under pressure. Of course, the oil was squeezed out at the inner and outer edges of the slipper, but with oil of the proper consistency, and enough rotational speed of the shaft, that movement was so slow that the slipper still floated for its entire length on the wedge of oil. There was no metal-to-metal contact, just metal to oil to metal again. Isn't that a wonderful idea? As it happens, that mechanism, squeezing the oil into a wedge inside the bearing, is the same mechanism that allows the cylindrical bearings of your car's engine to run for thousands of miles with little wear. Kingsbury was inventive enough to work out how to apply the same principle to a flat thrust bearing.
The engine is served by a lubricating pump that supplies oil through pipes to the engine's steam supply and to its major bearings. The pump is a horizontal cross-compound pump, and you can see the oil piping on many parts of the engine.
The fire and bilge pumps are also cross-compound pumps, with rather large water cylinders to supply large quantities of water at little pressure without using excessive steam.
The cabin is supplied with warm air driven by this large fan, rather a new idea at the time. Because this fan would be run most of the time on cold San Francisco Bay, its engine was designed to be efficient. It is a compound, or double-expansion, engine, with two cylinders controlled by a single valve driven by one eccentric.
The Berkeley was supplied with electricity from two steam-powered generators. The original installation didn't work very well, and was replaced after a few years. That is probably why the main generator, in the engine room, is driven by a steam turbine instead of a steam reciprocating engine as is all the other auxiliary equipment. The steam turbine gives a smooth, vibration-free rotation with very little to go wrong with it, and packs much power into small space and weight. This one on the Berkeley is the first sign of times to come, when nearly all the horsepower produced by steam engines comes from turbines instead of reciprocating engines.
General Electric was taking no chances with the future when it put the nameplate on the generator's turbine in 1907. The nameplate says that the turbine is licensed for all uses except as a prime mover for marine or aviation uses. If you wanted to power your plane with a steam turbine like that one, you would have to pay a higher license fee. Well, GE didn't start making aircraft turbines until more than 35 years later, and then they were gas turbines, not steam ones.
The auxiliary generator is inside the engine-room casing on the main deck, driven by a single-cylinder steam engine.
Now that you understand how the triple-expansion engine works, you can understand the significant advances that it embodies
The first useful steam engine, built by Newcomen in 1712, pumped water from coal mines. It used so much coal for its power that it was used only at coal mines. Its boiler was separate, but all other functions occurred inside the single cylinder. This was a vertical, open-topped cylinder whose piston was pushed downward by atmospheric pressure when the steam inside it was condensed by a water jet. The piston pulled down a chain attached to one end of a pivoted beam, whose other end lifted the pump pistons. The system was balanced so that the pump pistons fell of their own weight, pulling the piston to full stroke. The lower end of the cylinder had three cocks: for steam from the boiler, for water from an overhead tank, and to the drain.
Here's the operating sequence. Open the steam cock to let steam into the cylinder, so that the weight of the pump could pull the piston to full stroke. Open the drain so the steam could blow out the water from the previous stroke. Close the drain and steam cocks. Open the water cock to let water into the cylinder (remember, this steam is at ambient pressure; the water tank was on the engine-house roof). The water spray condensed the steam, making a vacuum, so that the atmospheric pressure above the piston pushed it down, thus lifting the pump pistons and pumping out the water. Then close the water cock and open the steam cock again so the piston could be pulled upward again by the weight of the pump pistons. In the early engines, the valve operation was by hand, but very soon the valves were operated automatically by rods from the main beam.
The Newcomen engine wasted steam for two reasons.
Fifty years after Newcomen's invention, James Watt perfected it. Watt was an instrument mechanic associated with Glasgow University, where Joseph Black had discovered latent heat, the additional heat required to be added to water to make steam, or to be taken from steam to condense it into water. Watt was asked to repair a model of a Newcomen engine. In doing so, in 1765, he realized that much steam would be saved if the cylinder was always kept hot and the steam was condensed in a separate condenser that was always kept cold. The separate condenser made the steam engine economical for many other purposes than pumping coal mines. In 1782, Watt put a head on the open cylinder, so that steam pressure could push the piston down as well as push it up, in what is called the double-acting cylinder. Watt's engines allowed the use of higher pressure steam whose valve was closed before the piston reached the end of the stroke, thus using the steam expansively for the remainder of the stroke, but Watt was very conservative, fearing explosions, and never progressed to high-pressure steam, relying on low-pressure steam and the vacuum produced by the condenser.
Oliver Evans, Richard Trevithick, George Stephenson, and others, developed the high-pressure steam engine that did not use a condenser, but discharged the steam at, or above, atmospheric pressure. If the boiler pressure was more than several times that of the atmosphere, the engine was smaller and lighter than a condensing engine, and not much less economical. This is the engine that powers steam locomotives.
High-pressure steam required better materials and construction. Once these became available, it was possible to supply high-pressure steam to condensing engines, thus saving steam by closing the steam valve with the piston near the start of its stroke and using the expansive power of the steam for the rest of the stroke. The higher the initial pressure and temperature, the earlier in the stroke the steam valve could be closed. A large, single-cylinder marine engine is shown in the first picture.
With steam of higher pressure and temperature at the beginning of the stroke, but of the same low pressure and temperature as early engines at the end of the stroke, the cylinder and piston had to be reheated at the start of each stroke, wasting steam heat as the Newcomen engine had done.
John Elder, in 1854, divided the steam expansion into two cylinders, so that each cycled through a smaller range of temperatures and pressures. This was called the compound engine, as used in the Medea, and was suitable for steam pressures of about 100 psia. The compound engine was sufficiently economical to enable steamships to sail almost anywhere in the world.
Improvements in materials and designs, even in lubrication, allowed still higher pressures and temperatures, and this allowed a further division of the steam cycle into three cylinders, the triple-expansion engine. The British Navy used the first triple-expansion engine, in a cruiser in 1887, and that type of engine was used in the American cruiser Olympia, built six years before the Berkeley, also at the Union Iron Works in San Francisco. Compared to the Berkeley's single engine of 1163 horsepower, the Olympia had two engines of almost 9,000 horsepower each (6,750 designed). The triple-expansion engine worked best at pressures of 155-185 psia, so that the Berkeley was at the upper range of desirable steam pressures for the triple-expansion design.
The next logical step was to divide the steam cycle among four cylinders, the quadruple-expansion engine. This design used steam pressures of 240 psia or higher, and was the final development of marine reciprocating steam engines. Such an engine is that in the picture, for the steamship Inchmona, using steam at 270 psia
The next step was the turbine engine, which could use much higher steam pressures and temperatures. The standard USN steam plant of WW II used steam at 600 psia and 850 °F.
1. Some say the boilers were replaced in 1924.
2. Some say that the new boilers were run at the 200 psig for which they were rated.
3. Paddlewheel engines were often two-cylinder horizontal, for sternwheelers and for Mississippi sidewheelers with separate engines for each independent sidewheel, or inclined, for sidewheelers with a single paddleshaft, or single-cylinder with the cylinder at the bottom, for sidewheelers with walking beam engines.
4. In the early days of steam at sea, they did use seawater in the boilers, but the salt built up inside as the water boiled off. This required both frequent blowing down of the boiler to remove the over-salted water, and frequent shutdown for boiler cleaning. It also limited steam pressures and temperatures to inefficient levels. In those days, the condenser did not keep the steam separated from the cooling water, but just injected the cooling seawater into the condenser chamber.
This picture of a "walking beam" single-cylinder paddlewheel engine was taken while the Southern Pacific ferry Ukia was being rebuilt into the Southern Pacific ferry Eureka, which is now at the San Francisco Maritime Museum. Note the large size of the cylinder compared to that of the workmen.
This is the quadruple-expansion five-cylinder engine of the express cargo vessel Inchmona, built about 1904. You can estimate the size of the engine from the handwheel by which the engineer changed the gear from ahead to astern.
Return to: John Forester's Home Page Up: Maritime