Section 3.1 The overall picture


The practical pressure range.

All steam-propelled devices have to get water from somewhere. Recently an attempt on the land speed record for steam cars was announced. The car is to driven by a turbine and is designed to achieve 200 mph. It will run for just 4 minutes before the water it carries is used completely. Steam-driven, high-speed cars do not look to be a threat to the cars using air as the working fluid. Model boats have the same problem; they need water.


Full-sized marine engines receive steam at the boiler pressure and exhaust it to a condenser at low pressure. The same water is used over and over again with a very small top-up. I do not think that many model engineers will make a success of a condensing plant because of the need to make a pump or pumps to extract air and condensate from the condenser. This means that model steam engines will exhaust to atmosphere and the water will be used and discarded. It follows then that water for the boiler must either be carried in the boiler or in on-board tanks or picked up from the lake?


Whatever decision is taken an engine that uses water economically whilst driving the boat at some desirable speed must be the design goal.


We need some idea of the practical range of pressure for miniature engines. On the face of it one might use a very small engine and very high pressures or a large engine and low pressures. In practice engines tend to be about 7/16 inch bore by 7/16 inch stroke and have two or more cylinders. With these dimensions working pressures of 50 psi above atmosphere give adequate performance. I think that there is a sound reason for this. These sizes are at the lower limit of conventional machining and to go smaller the special skills associated with making small components become essential. For instance, cutters for milling machines normally go no smaller than 1/16 inch or 1.5 mm, locating the table of the milling machine by use of graduated wheels and screws becomes relatively inaccurate and in order to go smaller special tools have to be made or purchased and a change to digital scales made.


So we need to know something about the behaviour and properties of steam in this pressure range.


Suppose that a boiler is partly filled with water and closed so that no water or steam can escape and the burner is lit. The temperature of the water will rise and at some point the pressure will rise above atmospheric and the pressure gauge will show a reading.. As the heating continues the pressure will continue to rise and so will the temperature of the boiler. The content of the boiler is now water, steam and air from the filling. If now the stop valve is opened for a short time steam and air will be blown off and then the stop valve can be closed again so that the boiler contains only water and steam. The steam is at the same temperature as the water and that temperature is called the saturation temperature.


Data is available to relate the steam pressure and the saturation temperature and is given in Table 1. Pressure gauges measure pressure using atmospheric pressure as zero. When pressures are measured relative to vacuum they are called absolute pressures. The atmospheric pressure is 14.7 psia or 15 psia for our purposes where the “a” means absolute. I shall want to look at the way that the volume of steam changes with pressure and for that I shall have to use absolute pressures.


Table 1 Saturation pressures and temperatures

Pressure psig








Pressure psia








Temp °Cc








Temp °F









Pressures of water and steam trapped in a container fall below atmospheric as the temperature falls to say room temperature. Under these circumstances the pressure when the temperature is 20°C would be only 0.34 psia. It is a pretty good vacuum.


Water boils very vigorously when steam is being generated in a boiler drum. As a result the extraction of steam to drive an engine necessarily means that the steam will contain water in the form of droplets. It is called wet steam. If, however, the steam pipe between the boiler and the engine is extended, suitably coiled, inserted into the boiler flue and exposed to the flame, the water in the steam will be evaporated and the temperature of the steam can then easily rise above the saturation temperature for the pressure and then it is called superheated steam. Necessarily superheated steam is dry. Miniature marine engines run much better on superheated steam.


Books on steam plant give the amounts of heat required to heat water to boiling at any temperature and the heat required to change the water to steam but in truth we are not really interested because it is very easy to have a gas fired burner that will generate all the steam we need and it really needs automatic control to turn it down when the working pressure is exceeded.


One piece of information that could be useful to relate steam consumption, engine dimensions and speed is the volume of steam generated from say 1 cubic inch of water at various pressures.


Table 2 Volume of saturated steam in cu inches generated from 1 cu inch of water

Pressure psia








Volume cu in










These days most people fire their steam plant with gas. This is readily available in canisters intended for use by plumbers for gas torches and can be purchased in hardware shops. The gases available are propane, butane and mixtures of the two. Camping gaz contains butane and butadiene.


These canisters and all the larger bottles are really boilers. In them liquid fuel is evaporated to give gas and, by good fortune, butane gas and propane gas give useful pressures at the commonly encountered temperatures between 0°C and 25°C. Evaporation cannot take place continuously unless heat is supplied, and that heat is normally supplied by the atmosphere or the hands of the plumber. In order to make the heat transfer the temperature of the liquid must fall and, on a cold day the fall may be so great that the pressure drops below a useful value. Owners of caravans where propane is used for heating and cooking have electric gas tank heaters for very cold weather.


All this means that modellers must either address the problem of evaporation in the on-board fuel tank whether it be a purpose-made tank or a purchased canister or put up with a diminishing flame in the burner as the gas tank cools down and a model boat going slower and slower.


Text Box:  
Fig 3 The steam plant in my Windermere cruiser.
The basic parts of miniature marine steam plant.


In this boat the water is carried in the boiler. There is plenty to drive the boat weighing 27 pounds at walking pace for 25 minutes.


The gas tank is really two parallel tanks joined by pipes to give a low profile to fit into the limited head-room. The tank has two electric heating coils that are powered by a 1200 mAh NiCad, the heater being controlled by a thermistor-operated switch. This is sufficient for a 25 minute run in winter. The gas, at steady pressure, is supplied to the burner through a burner-control valve that operates in response to a thermistor in the boiler to control the boiler pressure by turning down the burner to simmer. What we have is one boiler, powered by electricity, evaporating butane/propane mix that is supplied to the burner of a steam boiler to evaporate water to drive a steam engine. Both boilers are automatically controlled.


Steam is supplied to a two-cylinder engine through a throttling valve to control speed. The exhaust steam, which is wet and oily, goes into the first separating tank where most of the oil and water are separated and retained. The exhaust steam with the remaining oil goes into second separator where more water is collected and then to a discharge pipe in the chimney and to atmosphere as steam. The exhaust pipe in the chimney is about half way up so that it induces a flow of air through the burner to aid combustion.


Other miniature steam plant will be a version of this. The major addition that is made is to carry water in tanks, use a smaller boiler and pump the water into the boiler. This requires another automatic control system to sense the water level in the boiler and operate the pump as required.


Choosing a boat

It seems to me that one should have a clear idea of what is satisfying for the ownership of a steam-powered boat.


In my view a steamboat is lots more fun that an electric boat simply because there are so many more things to be done and subsequently monitored to achieve a run of satisfying duration. I think that most would agree. A steamboat producing a plume of steam is more entertaining for the passer-by and it generates far more nostalgia and many more nostalgic tales than any electric boat. This could be another goal. Having said that, a steamboat that is unreliable and promises the onlookers more than it delivers is proportionally more disappointing. A further requirement is that it should have a reasonable turn of speed. No matter how much you might pretend that your painfully slow boat is going at “scale speed” in your heart you would really like it to be capable of going faster and so would the on-lookers. Then of course the boat must have a presence, a stylish look that marks it out from the crowd, because, in the end it is only the sight of the boat under way that is the direct reward for the effort. Figures 4 and 5 are of my two steamboats.

Text Box:  
Fig 4
Text Box:  
Fig 5

If you design your own boats then anything is possible as is shown in figures 4 and 5 but design with an eye on internal space so that the steam plant is not all cramped up or simply too big to go into the hull. I think that steamboats need to be between 48 inches and 54 inches long.


If you plan to buy a hull or a kit then choice is more limited especially size. Very few are beamy and 48 inches long. Kit producers have to live and they make kits that they can sell.


Choosing steam plant

There are three possible routes. You buy ready-made steam plant, you make your own or you use a combination of the two.


I have little experience of ready-made equipment. It is designed and presented for sale and it certainly looks attractive. Makers of steam plant and boats know that the average modeller has visions of period, slender boats moving speedily and silently across a smooth lake leaving a trail of disappearing steam. Unfortunately manufacturers know that they have to sell to people who may have little or no experience of model engineering and have few tools that might be useful during the installation of the steam plant. I am sure that the manufacturers do their best to ensure that the customer ends up with a plant that performs as well as can be expected but this may not quite be what the customer had expected. It is not easy to design and sell steam plant that will work in all the hulls that might take the eye of his customers.


Mostly people buy hulls if not kits and the makers of kits also make what they can sell. This tends to be relatively short hulls of narrow beam. So steam plant and hulls do not always match.


Those who can make steam plant will most likely build engines from recognised drawings for which castings are available. They can make tanks for gas, exhaust and on-board feed water and make a good job of the pipe-work but these days over-regulation is a deterrent to those who would otherwise make a boiler. So one ends up with a mix of commercial and home produced equipment. These people are likely to have a better appreciation of the need for space in the hull and select or build accordingly.


An overview of steam plant

I was reading a book on designing boilers for miniature steam engines. The author just threw away the statement that the overall efficiency is likely to be less than 1%. I know that the overall efficiency of locos was about 6% so I was not surprised. What he was actually saying was that of the heat in the fuel, (coal, gas, or spirit) only 1% appears as work at the crankshaft of the engine.


The great attraction for the maker of miniature steam engines is that they can be constructed with quite basic equipment and to standards that are poor by comparison with those needed to made a successful IC engine and still run well enough to drive a boat. This very fact means that the magnitude of the efficiency of say a small single-cylinder, oscillating-cylinder engine sold as a toy might may be much lower than 1%. A well-made engine to a thoughtful design may function nearer to this 1%.


This overview is aimed at running through the various factors that influence the efficiency to see what can be improved.


Burning the fuel

All our fuels are hydrocarbons. They are substantially composed of hydrogen and carbon in chemical combination. Combustion involves the chemical combination of the hydrogen with oxygen and of the carbon with oxygen. The oxygen is supplied in the air that is drawn through the burner or the grate. Unfortunately air contains about 70% nitrogen that gets in the way of the combustion of the hydrogen and the carbon. The light, active molecules of hydrogen quickly find sufficient oxygen to burn completely but the heavy molecules of carbon may not be able to find oxygen in all the nitrogen before the temperature falls below combustion temperature. Then some of the carbon burns only to carbon monoxide. In extreme cases the carbon is not burned at all and we get a smoky flame. One might say that we should supply more air and therefore more oxygen to ensure that all the carbon is burnt to carbon dioxide but then we have to heat the extra air and that heat is carried away to exhaust and lost. So a balance must be struck to get the best combustion and the lowest loss of heat to exhaust.


The water produced by the combustion of hydrogen is, of course, produced as steam. This steam contains a significant amount of heat but it is not available to us.


Fig 7

So what does this mean for miniature boilers? There is little that the model engineer can do to measure the performance of the system used for burning the fuel. We have to work by experience and common sense. Coal is the most troublesome because the control of the air supply is dependent on the skill of the fireman. Gas and pressurised spirit burners are essentially Bunsen burners with primary air induced by a jet of gas before the flame and secondary air drawn into the flue through holes of adjustable size around the flame. The burner can be adjusted once and for all by listening to the flame and from its appearance. It is hard to see what more can be done.


We shall lose heat to incomplete combustion to carbon dioxide, in the steam produced by the burning of hydrogen and to heating unused oxygen and the unavoidable nitrogen. On the face of it we might expect to lose less heat if a sufficiently high proportion of the heat produced in combustion can be passed to the steam to give a cool exhaust but no matter what we do we must have a temperature difference between the steam and the hot flue gas to cause the heat transfer. Our options are very limited.


We have to look elsewhere for possible improvement.


Heat transfer

Having produced heat by combustion as much as possible of the heat must now be transferred to the water in the boiler. We need some idea of how this heat transfer takes place. One always hears about conduction, convection and radiation as modes of heat transfer and of course these are all present in our boiler. How do they contribute to the transfer?


It is very likely that the burner will produce its flame as lots of small flames of gas flowing through a piece of ceramic like the brilliants of a gas fire. The ceramic certainly seems to be white-hot and radiates heat. In vertical boilers the ceramic may be quite large and be directly under a tube plate with water on it’s inside surface. Both the ceramic and the lower tube plate radiate but as the ceramic is much hotter than the tube plate the net flow of heat is from the ceramic to the tube plate and then on by conduction through the copper to the water. In the burner shown above the ceramic is quite small and facing into the lower flue of a horizontal boiler. It is hard to see on what surface the radiant heat from the ceramic actually falls. Some modellers burn the gas in lots of small flames produced by using a perforated steel plate instead of the ceramic. It seems likely that radiation plays a significant part in heat transfer in vertical boilers but not in horizontal boilers.


In both cases a large proportion of the heat is in the combustion gases and has to be transferred by a combination of convection and conduction.


Convection is the transfer of heat by movement. In our case the movement is that of the combustion gas flowing through the vertical tubes of a vertical boiler or the horizontal flues of a horizontal boiler.


Ultimately the heat must go from the combustion or flue gas to the water by conduction through first a thin layer of slow-moving gas that forms on the solid surfaces then through copper and then into the water through another thin layer of slow-moving water attached to the inside surfaces. These two slow-moving layers are called boundary layers and have a complicated behaviour. They limit the rate of heat transfer. Their effect can be reduced a little if the flue gas swirls about and the water is agitated. We must leave the water to generate convection currents as it boils but we can arrange our tubes and cross tubes to give the maximum turbulence in the flow. This is most easily done in the horizontal flue where the cross tubes can be arranged to agitate the flow as much as possible.


The result of all this is that the flue gases cool as they progress through the boiler. It is a matter of design to get the gas to move quickly but not so quickly that it passes through the boiler before a practical amount of heat has been extracted from it.


So the heat transfer is mainly by convection and conduction with a contribution by radiation. It is all very complicated and will inevitably involve trial. There is not much that can be done to alter the modes of heat transfer in any useful way. It is really very basic.


If heat transfer by radiation takes place between surfaces and conduction takes place through surfaces the total area of the heat transfer surface is a dominant factor. The capacity of the boiler to evaporate water to produce steam is determined by this area. If the boiler is not to be fed with water by pumping it must hold enough water to give a satisfying running time. Big boilers have big surfaces for heat transfer and it is relatively easy to have sufficient area of heating surface and the temperature of the exhaust will be quite low. However, if the boiler is pumped because of scale appearance, it will be quite small compared with the un-pumped boiler and may well be short of heating surface, be too small for the desired burners, and it may lead to having to accept a high exhaust temperature.



The boiler will obviously get hot. It will operate at the temperature of the steam, at more than 140°C or 280°F. It can be insulated to reduce the direct heat loss to the atmosphere. If the appearance of the boiler is not important it can be heavily insulated with say a balsa wood jacket but, if it is, say, to be used in a Windermere cruiser, wooden slats must be used and this is not quite so effective. We can limit this loss but in truth it is not something that affects the performance of the engine. The loss can be met by burning more gas. It is simply wasteful.



The steam

Assuming that sufficient heat can be transferred to the water in the boiler, steam can be generated at a useful pressure. It is possible to follow the process. Heat is required first to bring the water to the boil. This can be reduced by filling the boiler with hot water but this only conserves gas. However gas must be burnt continually to convert this hot water into steam to drive the engine. The heat required to convert water into steam is about four times that required to heat water from freezing to boiling point.


In a boiler the water is converted continuously to steam and drawn off at a steady rate to feed the engine. The process involves boiling which is a vigorous process and the “steam” that leaves the boiler drum is a mixture of steam and water droplets. The steam is wet. Some boilers will be fitted with a drying loop (Some call this a superheater but I find it hard to believe that a loop placed at the outlet from the boiler to the chimney will produce any superheated steam.) in which the steam flowing to the engine is heated to convert the unwanted water droplets to steam. When this is working properly steam that is said to be dry is fed to the engine. Of course the boiler really can superheat the steam sufficiently to soften plumber’s solder if a properly designed loop is fitted.


This steam has absorbed a substantial quantity of heat.


When the steam leaves the engine it is wet steam again at a lower pressure and it contains a very large proportion of the heat that was given to it during its production in the boiler. That heat is discharged to the atmosphere through exhaust traps. Only a very small part of the heat supplied to the steam is actually converted to work to drive the engine.


Boiler pressure and exhaust pressure

I have already discussed these pressures but I include it again as part of this overview.


Steam plant for power generation and for marine propulsion exhaust to condensers. In these condensers the exhaust steam is cooled causing it to condense to water with an associated pressure drop to only 2 or 3 psia, ie 13 or 12 psia below atmospheric pressure. Three-cylinder compound engines and steam turbines were required to take advantage of the condenser. Steam locos never used condensers and always exhausted to atmospheric pressure. There were good practical reasons for this the main one being the complete absence of a method of cooling any condenser that might have been fitted.


It is possible to make a miniature condenser to go with a miniature engine but the problem is the design and manufacture of a pump to extract water and air from the condenser. Most miniature engines exhaust to atmosphere.


This settles the exhaust pressure.


The boiler pressure is not so easily settled. One factor is the physical size of the engine. Any miniature steam engine is likely to be strong enough to run on pressures up to 100psi or even higher. The power output of the engine will go up with the boiler pressure. For tugs, straight running boats etc a high output may be desirable but for scale or “scalish” boats the speed must be limited to what looks right for the hull. Then we run into a problem. For a boiler pressure of 100 psi the size of a two-cylinder engine required to drive a quite large scale type boat would be quite small, so small in fact that it becomes difficult to make. For instance narrow ports become difficult to make for want of suitable tools and, as the smallest milling cutter is normally 1/16² with 3/2² being the next size available it is much easier to make a larger engine, eg 3/8² x 3/8² or 7/16² x 7/16² and use a lower boiler pressure. Screw threads also become a problem. 10 BA starts to look quite big when the bore is 1/4².


So realistically the range of boiler pressures is say 20psi to 50 psi and we can calculate the highest possible overall thermal efficiency. They turn out as follows:-

   20 psig….5.5%;      30 psig….7%;       40 psig….8.3 %;        50 psig…..9.2 %


We cannot get anywhere near to these figures but the figures do tell us that we would stand a better chance of getting up to 1% if we used 50 psig rather than 20 psig. This means that we have conflicting requirements for good practice. We would like to run at 50 psig or higher but, we also want an engine that is relatively easily made with readily available tools.


Water in the cylinders

The big problem with miniature steam and probably full size as well is water in the cylinders. If the engine has slide valves or piston valves and the width of the valve faces exceeds that of the ports water can be trapped in the cylinders. The slide valve is pressed on to its ports by the boiler pressure so it will not lift, the piston valve cannot lift. So very high pressures can be developed as the piston attempts to compress the water. In miniature engines the commonly used solution to this is to use valves with almost no overlap on the ports.


Economical use of steam

There are two ways of achieving this economical use of steam. One is to superheat the steam and the other is to use the steam expansively in the engine. The first affects the design of the boiler, the second the design of the engine.


In an engine that has no valve overlap in order to avoid water locking steam is admitted for all the working stroke and is rejected at atmospheric pressure throughout the exhaust stroke. This is very wasteful. If the supply of steam were to be cut off at about half stroke the steam would go on expanding doing useful work.  It is a much more efficient way to use the steam. The calculated efficiencies suppose that the steam is expanded from the boiler pressure to the exhaust pressure doing work all the way. If our engine is to edge towards the 1% we must think of ways to let the steam expand in the cylinders.


We need to look much more carefully into the design of a miniature steam engine.


In order to superheat the steam we must look at the design of the boiler.