h a l f b a k e r y
Where life imitates science.
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
or get an account
A typical internal combustion engine converts about 1/3
it's fuel's energy to mechanical energy. Of the
half is lost as heat that goes out the exhaust, and the
other half is removed from the engine block via coolant,
and dissipated through the radiator.
The internal combustion
engine system described in this
idea uses water to recover a portion of the exhaust gas
heat, and engine block heat, and uses that recovered
to produce additional mechanical energy.
The system consists of the following main components:
The engine block, a steam condenser, a gas/liquid
separator, an exhaust gas heat recovery device (aka a
heat exchanger), a water recovery device, two
compressors, and three manifolds.
Part of the system's operation is simple, part is
First, the simpler parts:
Air enters the system through an air filter, passes
the intake manifold to the engine block to the air
manifold, then the heat recovery device, the water
recovery device, the muffler, and leaves through the
There's a smaller flow of air from the steam condenser,
through the gas/liquid separator, a compressor, and into
the heat recovery device.
Water goes from a reservoir, to a pump, through the
recovery device, to the engine block; it's injected into
engine block and converted to steam; the steam goes to
the condenser, turns to water, passes through the
gas/liquid separator, and is returned to the reservoir via
The water recovery device a simplified version of that
described in patent 4725359. Specifically, engine
gas flows across a hygroscopic membrane which is
permeable to water, but mostly impermeable to air. The
other side of the membrane is at sub-atmospheric
pressure. Water vapor from the exhaust gas condenses
the membrane, diffuses through it, evaporates on the
other side, and is removed by a compressor to the
system's condenser. The membrane is coiled in such a
that the device contains a large surface area of that
membrane in a small volume, and is shaped to minimize
the amount of exhaust gas back-pressure the device
Now, for the most complicated part of the system;
the engine block.
The engine block resembles a six cylinder Cylindrical
Energy Module engine invented by Eddie Paul, but with a
several important differences.
The cam track's curve is of the form z=L*sin(3*phi), so
there are three up strokes and three down strokes.
A derivative of the six stroke cycle invented by Leonard
Dyer is used.
Each of the two heads of the engine has the following
The air intake port has a variable geometry, allowing the
first four strokes of the combustion cycle to be similar to
an Otto cycle, or similar to an Atkinson cycle. The shape
of the intake port controls the power the engine
produces... a larger opening produces less power, a
smaller opening more. The variable geometry ports
eliminate the need for a throttle valve. Removing the
throttle valve eliminates throttling loss.
Water injection will be used to lower peak air
temperatures, prevent the formation of nitrous oxides,
and prevent knocking. The water used for this will come
directly from the water pump, bypassing the heat
device. Note that this is conventional water injection,
has nothing to do with the water injection that's used to
make this a six stroke engine.
Fuel is supplied by means of an injector. Since this is,
effect, gasoline direct injection, it should be possible to
supply a stratified fuel charge, and thus allows the
to use a lean fuel mixture for all but the heaviest engine
loads. The lean mixture reduces pollutants and
After the air exhaust port on the engine head, is a
water injector; this one does use the water from the
recovery device. The amount of water injected is
computer controlled, and is dynamically adjusted so that
the piston surface and cylinder walls are at the optimal
temperature at the end of the steam expansion stroke.
Note that, even once the engine reaches operating
temperature, and the water entering the injector is very
hot, it will still be capable of cooling off the cylinder and
piston... this is because, as the water is injected, it will
experience flash evaporation, and reach a cool enough
temperature for the portion that remains liquid to cool
cylinder and piston.
After the water injector, is the steam exhaust port, and
after the steam exhaust port is the air intake port.
Exhaust Gas Water Recovery
[goldbb, Jan 01 2011]
Cylindrical Energy Module
[goldbb, Jan 01 2011]
Conventional Water Injection
[goldbb, Jan 01 2011]
[goldbb, Jan 01 2011]
crower 6 cycle engine
[metarinka, Jan 15 2011]
Bruce Crower's Patent application
[goldbb, Jan 25 2011]
||Questions, Comments, Criticism, Buns and Bones
all gladly accepted. Of course, I'd prefer that
bones be accompanied by (valid) criticisms, but
anything's better than being Warnocked to death.
||Hmm... considerations I forgot:
||The compressor which moves steam from the
water recovery device to the steam condenser is
not needed if the condenser's pressure is
sufficiently low. However, a low pressure
condenser might be too big / heavy / costly to be
practical (especially if this engine is in a car).
||Ideally, this can be done by using high tech low
friction ceramic bearing surfaces.
||Alternatively, liquid lubricant (oil) could be
sprayed into the cylinder along with the water,
during the fifth stroke, and recovered by an
oil/water separator at the condenser's output.
There *do* exist oils suitable for this purpose,
though they are not your typical automobile oils.
||Corrosion protection: Since the engine is burning
lean, it's inevitable that the exhaust will contain
some free oxygen, and much CO2 (some of which
will form carbonic acid later in the cycle). Since
scavenging isn't perfect, it's inevitable that some
exhaust will be in the cylinder when water is
injected for cooling. Naturally, the steam that
results will have some exhaust in it. Using
chemicals to remove that oxygen or CO2 isn't
practical. Instead, it's easier to just make all parts
which will be exposed to steam or hot water out
of plastic, stainless steel, or other corrosion
resistant material in the first place.
||Freeze protection: This is hardest...
||We could mix some sort of alcohol into the
water... as the engine runs, the alcohol will burn
off from the water, though, and will need to be
replenished. If the fuel is a suitable alcohol, then
we have a ready source of antifreeze to regularly
add to the water.
||We could make the engine sturdy enough to not
be damaged as the water freezes, and use
electrical heating elements to thaw the engine
each time we need to start it in cold weather.
||We could use gasoline as a substitute for water,
and eliminate both the water recovery device, and
the conventional water injection. Back-of-the-
envelope math suggests that the fifth-stroke
injector, used for cooling, would need to squirt
about 15x as much gasoline as the third-stroke
injector supplying gasoline for combustion; a lot,
but doable. If we *don't* do this, and use water
for cooling, the volume of water needed for
cooling is approximately twice the volume of
gasoline used for combustion.
||There're probably other solutions for the freeze
protection problem, but I can't think of any
||//Warnocked// I'm afraid if I write anything the post will fall on me and I'll be crushed.
||The exhaust-membrane sounds neat, won't that get clogged up ?
||All throughout the post you say "goes into the engine block" without bothering to say if it's going into the cylinder or around the cylinder.
||Whether the exhaust membrane will get clogged
up depends whether the exhaust contains
particulates, how many there are, and how prone
those particles are to sticking to the membrane.
Keep in mind, the exhaust is flowing across the
membrane, not being forced through it, which
should reduce the likelyhood of particles sticking.
||Ok, air going "through the intake manifold to the
engine block" is actually going into the two engine
heads; specifically the air intake port of each
||Water going "through the heat recovery device, to
the engine block" is actually going to the two
engine heads; specifically to the fifth-stroke water
injector in each head.
||Where water "is injected into the engine block"...
I meant: goes to an injector in the engine head, is
injected into each passing cylinder, moves from
each cylinder to the steam exhaust port to... but
wanted to simplify. Obviously I simplified a bit too
much. Hope this clarifies things.
[edit: removed "Will fix... tomorrow." ;)]
||Efficiency and power can be improved even
further, by using a high (20:1 or so) compression
ratio, little or no EGR, and homogeneous charge
combustion ignition (HCC).
||For the HCCI version of this idea, at low loads, the
engine would use a very lean fuel/air mix, a
minimal size air intake port, and cool the cylinder
to a relatively low temperature on the 5th stroke.
This results in the largest possible charge of air per
cycle, which results in high pressures; the high
pressure causes a high temperature, which causes
the fuel/air mix to ignite.
||As the engine load increases, the fuel/air mix
would gradually be enriched (though not
exceeding stoichiometric), the intake port kept
open past bottom dead center, and less 5th stroke
cooling would be done.
||The richer mix produces more power, the
extended intake stroke lets air get regurgitated
into the intake manifold, reducing the total air
charge and effective compression ratio (which
prevents pre-ignition and keeps the engine from
blowing up), and the reduced cooling results in
higher cylinder and piston temperatures, which
results in enough heat transfer from the cylinder
and piston to the air/fuel mix that it ignites in
spite of the lowered compression ratio.
||In other words, the driver's accelerator pedal (or
the governor, if this is in a generator instead of a
vehicle) determines the fuel/air mix, and the
engine management system simultaneously adjusts
the compression ratio and cooling to achieve the
appropriate combustion timing.
||Note that compression ignition can be used for
most of the engine's load range... since I've already
explained how we avoid blowing up for a high
load, and (since the engine already uses gasoline
direct injection) we can use a stratified charge for
extremely light loads.
||Since the HCCI version of the system is no more
mechanically complicated than the spark ignited
version, we can think a switch to it as being
basically "free." (Well, not free free; we will
probably need to add in a catalytic converter, but
at least we only need to use a 2-way cat, not a 3-
way one, since the lean fuel mix results in few
||Before anyone objects that switching the design
to HCCI would require that the engine be made
sturdier (to avoid blowing up due to the high air
pressures), consider that we already need to make
the engine quite sturdy, due to the need to avoid
blowing up due to high 5th stroke steam
||In other words, switching the design to HCCI
might not necessitate an increase in the engine's
weight... it just produces more power and higher
||Sounds similar to a crower six cycle engine. While it may be more mechanically efficient, there would be a large increase in complexity.
||The 6 cycle engine offers similar performance gains without a net gain in complexity, weight or size.
||It is indeed a form a six stroke cycle, but with the
following important differences from Bruce
||First, a CEM engine block has far, far, fewer
moving parts than a conventional engine. There
are no crankshafts, no connecting rods, no
camshaft, no valves, no timing belt. This means
less weight and less maintenance. A CEM engine
is also more compact that a conventional engine
with the same displacement.
||This means that if we're comparing engine blocks
alone, and not other parts of the system, mine
wins on weight, cost, and maintenance.
||Bruce Crower's six stroke engine will require a
steam manifold which connects to the head of
each cylinder. I can't visualize the fresh air,
exhaust air, and exhaust steam manifolds in such a
way that it doesn't resemble a mutant squid.
||A CEM engine block has only two heads, so there
would only be three pipes on each end, for a total
of six for the entire engine... two fresh air, two
exhaust air, two exhaust steam.
||Even in his patent application (US20070022977),
Bruce Crower's engine does not have an exhaust
gas water recovery device.
||Since a six stroke engine will inevitably lose a little
bit of steam with every single cycle (steam will
remain in the combustion chamber past the end of
the steam exhaust stroke, and mix with the air
during the air intake stroke), the water will
inevitably need to be replenished.
||Because my system includes an exhaust gas water
recovery device, should never need additional
||By preheating the water using the exhaust gas
heat recovery device, before injecting it, the
thermal efficiency and power are much greater
than they are in Bruce Crower's six stroke engine.
||In fact, if we were to pretend that the heat
recovery device can be 100% efficient, or close to
it, and assume the same engine displacement,
number of working chambers, and number of
combustion cycles per second: my engine will
produce as much more mechanical power than a
Crower Six Stroke engine, than a Crower Six Stroke
engine produces compared a four stroke engine.
||Of course, I'll freely admit that the heat recovery
device is neither 100% efficient, nor weightless,
nor tiny, but considering how much this one
simple no-moving-parts component can boost
power and efficiency, I'd posit that it's a
worthwhile component to include in the system,
in spite of it's weight and size.
||Plus, it will lower the exhaust gas temperature...
probably enough that no additional cooling is
needed to pass the exhaust to the water recover
device. And since the water recovery device is
key to preventing this system from needing
annoying regular water replenishment, well,
||Since my engine is a CEM engine, one revolution
of it's shaft will result in one complete six stroke
cycle for all twelve of it's working chambers. For a
Crower Six Stroke engine to produce those same
twelve combustion strokes and twelve steam
expansion strokes, it's crankshaft needs to make
three complete rotations, and thus a much higher
||Thus, my engine will run at a much lower speed
while producing more power.
||A six stroke six cylinder CEM is naturally
dynamically balanced, and (if the two engine
heads are 60, 180, or 300 degrees apart) produces
very even torque. Furthermore, the spinning
engine block will have enough angular mass to act
as a flywheel, so no explicit flywheel will be
||This helps ensure low vibration, quite running,
smooth power, and high maximum RPM.
||What kind of cylinder layout would be needed to
dynamically balance a Crower Six Stroke engine,
and how big a flywheel would it need?
||While the Crower six stroke engine could use
variable valve timing for increased fuel economy,
there is no mention of this in his patent
application. Presumably, engine power is
modulated purely through a throttle valve.
||By varying the geometry of the intake port, my
engine's power can be controlled just as fuel-
efficiently as BMW's Valvetronic system, or Nissan's
Variable Valve Event and Lift system.
||This idea is not completely original -- several of
Mazda's Wankel engines have variable geometry
intakes, though I believe that they also use
||It's also not exclusively beneficial to six stroke CEM
engines ... any CEM engine could benefit from
using variable port geometry instead of a throttle
in order to control power.
||Lastly, I'd be surprised if Bruce Crower has even
considered combining the six stroke cycle with
||Ok, this one's the real last difference: where I
specify exhaust gas heat recovery happening in a
separate component (that component being some
type of heat exchanger), Bruce Crower's patent
application suggests routing the exhaust gas
through passages in the engine head!
||While heating the engine head would result in
slightly more heat available for the production of
steam, much of that heat would just end up in the
air of the engine bay, not in the water.
||Furthermore, this would surely result in high
exhaust gas back pressure.