Non-Reciprocating Expanders

Non-Reciprocating Expanders
This page describes a number of technologies that have been experimented with for light steam power applications over the last century and more. For the most part they have not been successful. Even so, many people new to steam seem to have an instinctive need to try to reinvent these technologies. There is nothing wrong with experiementing with unconventional engine designs, and if that interests you, learning about previous designs will help speed your work. If you'd prefer to not reinvent an existing technology, knowing what has been done before will help with that, too.

Organic Rankine Cycles
Most of the reported current research using these expanders is for systems using organic working fluids, such as refrigerants. These fluids have molecules much larger than that of water and leak through small gaps much more slowly. Many of them will also work with a soluble lubricant that flows around the system and greatly helps with sealing. No similar additive has been found practical for steam.

The likely reason for so much research using these kinds of expanders is that they are commercially available. It is far easier for a university researcher to purchase an off-the-shelf air conditioning compressor and optimize a cycle around it than to design and fabricate a reciprocating engine.

Turbines
Steam turbines become more efficient as they become larger. They are also much more compact than reciprocating engines of similar power, and as power levels go up, they become relatively less expensive to construct. On the other hand, this scaling phenomena works against them as they are reduced in size. Use examples of Lear turbine and Cummins ORC waste heat system. For smaller systems, the Ljungstrom radial outflow turbine has some potential. The discs can be machined on CNC mills relatively easily. Because they contain multiple stages and counterrotating discs, rotational speeds are considerably lower than those of axial turbines.

Rotary Engines
Rotary engines seem to hold a special place in the human psyche. There have been literally thousands of rotary steam engines patented in the United States. Almost none of them have been sold commercially, and of the few that have, none were produced for very long. The only successful internal combustion rotary engine, the Wankel, has been converted to steam a number of times over the years with no real success. Quote from old engineering text that rotary engines are useless. No doubt another 2000 rotary engines have been patented since then.

A common justification for rotary engines is that there is "energy wasted" by reciprocating pistons in a conventional engine. The thought seems to be that reversing the direction of a piston twice in each rotation causes significant energy to be lost. Rest assured, though, that the kinetic energy of a piston is conserved. As it decelerates its energy is transfered to the crankshaft, after which it is transfered back to the piston when accelerating in the other direction.. Think of a pendulum swinging. It reverses twice on each swing, exchanging kinetic for potential energy and vice-versa, yet would swing forever if air resistance and bearing friction were eliminated.

Should anyone wish to pursue a rotary engine design, there are two main opportunities: seals and expansion.

Sealing
As the rotor (or rotors) turns it needs to seal the pressurized steam within a chamber. The geometry of the chamber causes the pressure to create a torque on the rotor, and as the rotor turns it extracts work from the steam.

This chamber is formed by edges and surfaces of the rotor and housing almost, but not quite, touching. The parts move past each other at high speed, so if they touched anywhere they would quickly wear. On the other hand, steam is a very thin gas and will rapidly leak through a very small (e.g., less than 0.001"/25 microns) gap.  As it leaks past the moving parts the engine efficiency rapidly drops.

A further engineering challenge is that different areas of the engine will see different temperatures, ranging from almost the steam supply temperature at admission to approaching ambient temperature on the outer casing. This temperature differential leads to considerably uneven thermal expansion of the parts, making it even more difficult to control clearances.

It may be noted that the first iteration of the BMW Turbosteamer car, which uses exhaust heat with a Rankine cycle to power the alternator, used a converted rotary vane pump as the expander. The second iteration of the Turbosteamer uses a reciprocating piston engine. Even a building full of German mechanical engineers (in white coats, no doubt) couldn't make a rotary expander work.

The racing community has been experimenting with high-tech materials for apex seals in Wankel engines. Some of these may hold promise as seals in rotary steam engines.

Expansion
The amount of work that can be extracted from a quantity of steam is greater if the steam is allowed to expand while the work is being done. Think about a bicycle pump. Push down on the handle and you do work on the compressed air. Release the handle and the compressed air can do work pushing the handle back up as it expands.

Now if you compress the air and let it escape to the atmosphere, the energy it contained is lost without doing any work. This is a problem with many rotary engine designs due to their geometry. The steam doesn't expand appreciably in the engine, so it is exhausted while still at high pressure.

A similar problem can also occur at steam admission. If steam is admitted to a fairly large chamber, a considerable amount of steam must flow into the engine before any work is done. The energy in this steam isn't fully used. (This is similar to having a large clearance volume in a reciprocating engine and the inefficiency it leads to there.)

Screw Expanders
These are usually screw air compressors run in reverse. The name for one common type is the Lysholm. With steam they have the same sealing problems as the rotary engines. There are large lengths of contact at screw to screw and screw to casing junctions that must be sealed to obtain reasonable efficiency. There have been some attempts reported of injecting liquid water or oil to improve sealing with limited success.

Scroll Expanders
Some refrigeration compressors in refigerators and air conditioning systems use two plates with deep grooves cut according to an Archimedes scroll that orbit against each other. As they do the scrolls slide past each other they create contained volumes of fluid. The orbiting motion causes the contained volumes to contract (when used as a compressor) or expand (when used as an expander).

Again, sealing is the major problem with steam. An experiemental study found efficiencies of xx-xx for steam compared to xx-xx for organic [ref]