A Compact Pollution-Free
External Combustion Engine
with High Part-Load Efficiency
Previous Chapter | Next
Chapter
12. Expander Hot Cylinder Lubrication
The key uncertainty in the entire engine is the hot cylinder and its
lubrication, which will require experiments. Even the highest quality
oils cannot take the high time-average temperatures. They range from about
540ºC (1000ºF) on the cylinder wall, to 590ºC (1100ºF)
in the Expander inlet steam, to up to 650ºC (1200ºF) should
the lubricant contact the inside walls of the Steam Generator tubes. Therefore
the cylinders are lubricated by a combination of powdered high-temperature
lubricant, cylinder sleeves impregnated with these lubricants, and hard
surface coatings. Figure 12 shows the general arrangement.
NASA did pioneering work on high-temperature lubricants at the beginning
of the space program [16]. Lubricants
they identified, such as calcium fluoride, lead oxide, and boric oxide
are possible candidates here. There are others, which are discussed below.
The first dry lubricant to be examined will likely be graphite powder
and graphitized metals [17]. Graphite
lubricity requires water or water vapor, which is in abundance here. Fine
graphite powder would circulate through the entire steam circuit. Key
issues are degradation in the hot steam and clumping in the Feedpump.
Other powdered lubricants also look promising. These include tungsten
disulfide (WS2), selenium disulfide (SeS2) and hexagonal boron nitride
(hBN). These materials have very high temperature capability and low friction
[18]. They can be used alone or in
combination [19,20].
There are a variety of hard surface coatings that can be used in conjunction
with dry lubricants. These include nickel-boron [21],
special chrome platings [22], and titanium
carbide or tungsten carbide [23].
The above are just examples of available dry lubricants and wear-resistant
coatings. There are other options and a well-established industry available
for assistance. This is a selection task, not a research task.

|
The use of dry rather than liquid lubricants allows one to exploit
the fact that static friction is then greater than sliding. Figure
12 shows how this can be used to reduce seal wear. The primary
seal ring, on the periphery of the piston, is deliberately thin (leakage
just vents to the next stage). Forces from steam pressure keep it from
moving provided its height-to-width ratio is less than the static friction
coefficient. Therefore, the seal is stuck in position when there is a
differential pressure across the piston, as occurs during the power stroke.
The seal then essentially skims along the cylinder wall without wearing.
It resets itself only when there is no pressure difference and hence no
leakage, that is, during the exhaust stroke. A small spring provides the
resetting force.
There are a number of options for the seal itself. One has been patented
by the author [24].
Previous Chapter | Next
Chapter
|