h a l f b a k e r y
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
or get an account
It's pretty easy to convert a two-stroke engine into a
"uniflow" steam engine: you just need some sort of valve
in place of the spark plug that allows steam into the
cylinder at top-dead-center. The steam expands, pushes
the piston downward, and exits out the exhaust port.
This valve can
be something as simple as a "bash valve,"
which is basically just a one-way check valve that is
opened when the piston slams into it.
My idea is to add a bash valve onto a two-stroke motor,
that it can be powered either by steam, internal
combustion, or some combination of both. If you start
a water-cooled engine, you won't need a large boiler,
the water coming out of the engine will be near boiling
anyway. You could even add a water jacket around the
exhaust pipe and get even more heat out. Then you'll
need to heat it up a bit more in the boiler so that it has
more pressure than the combustion gasses at TDC.
You would have to start this engine like a normal two-
stroke until it warms up and the boiler reaches a certain
pressure. Then you could either switch to combined
combustion/steam power, or shut off the gasoline to the
carb and go to steam completely. If you're doing this
need a separate oil injector rather than running premix.
And you'll want to run it on gasoline only for a minute or
two before you turn it off, so that the piston rings don't
When running on both steam and internal combustion,
there will be an upper limit to the amount of steam you
can inject per revolution before it "quenches" the
combustion in the cylinder and you get less power.
you will get slightly more power than a regular two-
slightly better efficiency, much better emissions (the
condensing steam will trap soot and unburned oil/fuel),
and a slightly heavier, costlier, and more complex
Most two-stroke engines fire the spark plug about 15 to
degrees before TDC, and if the bash valve opens at, say,
10 deg. before TDC, I don't think it would interfere with
combustion. A more sophisticated steam inlet valve
allow for more fine-tuning in this area. If you can time it
correctly, then the incoming steam would also pressurize
the fuel/air charge, giving it a higher effective
compression ratio without causing pre-ignition. And if
steam was hotter than the ignition temperature for
gasoline (about 500F, IIRC), you might even be able to
ditch the spark plug and just use the steam to ignite the
Of course this idea would probably work with a four-
engine as well. I'm just thinking of what I know best.
I'm sure there's some issues with the idea, and I haven't
done the math to see if it would be worth it, but I
I'd put it out there. One issue I can think of right now is
that combustion pressures can reach as high as 2,300 psi
in some engines (not sure what it is for a typical two-
stroke), so you would need a strong boiler (maybe made
from stainless-steel tubes) and a high-pressure water
pump (maybe from a pressure washer) to overcome the
Uniflow Steam Engine
without the two-stroke part [discontinuuity, Jan 04 2013]
video of a steam-powered moped
with its stock two stroke engine converted into a uniflow steam engine [discontinuuity, Jan 04 2013]
Bash Valve description
[discontinuuity, Jan 04 2013]
more on bash valves
including info on converting two stroke engines to steam [discontinuuity, Jan 04 2013]
my illustration of the whole assembly
[discontinuuity, Jan 04 2013]
displacement steam oiler
no moving parts! [discontinuuity, Jan 04 2013]
[ytk, Jan 04 2013]
cools the intake charge to prevent detonation in turbocharged or high compression engines [discontinuuity, Jan 04 2013]
steam diesel hybrid locomotive
Sort of baked, although alternating and not at the same time. Also I don't know if this was 4 stroke or 2 stroke. [discontinuuity, Jan 06 2013]
(?) Turbo two stroke diesel, not even slightly baked, vaguely preheated...
[normzone, Jan 08 2013]
Four-stroke steam engine
Quite long-winded, but a good demonstration on how superheated and pressurized water can flash into steam after being injected into a hot cylinder. [discontinuuity, Jan 08 2013]
||Sounds a bit like the Crower six-stroke engine (link).
||Don't you hate that when you have a great idea, just
to find out that someone like Crower built it already!
I like the way your mind works. I have heard a bit
about Large ocean liners reusing exhaust etc.. But
never thought you could "burn" your own coolant!
bun for you.
||Yes, I was thinking a bit of the Crower six-stroke
when I came up with this. The problem with the six-
stroke engine is it has 3 times as many strokes as it
||But seriously, I think that a two-stroke engine would
benefit from steam injection, and it has the benefit
of fewer "coasting" strokes.
||Do they inject water vapor into internal combustion
engines? I've seen some YouTube videos but I don't
believe everything I see
||Yes, lots of turbocharged cars have water injection
(in the intake manifold). This is meant to cool down
the intake charge and prevent detonation. As far as
I know, the Crower engine is the only internal
combustion engine in which water is injected
directly into the cylinder.
||Why not just use port timing? Suzuki has used
rotary-valve 2 strokes in the 70s. Use a
reed-valve on start up, then when switching to
steam power, physically block the reeds. Then use
rotary valve for steam injection in the head?
But then again, I don't know jack about engines.
Also: is there much fact behind the idea of
injecting steam into the combustion chamber
during two-cycle operation could really increasing
power or efficiency? Water-injection is common
practice in high performance turbo applications to
try and reduce the temperature of your intake
charge in a bid to effect its density. Why not just
make a V-twin powerplant and keep one cylinder a
conventional two-stroke and the other a
steam/compression style cylinder and blend their
||I was also thinking about blocking off the transfer
ports between the crankcase and cylinder during
the steam cycle. You could do this with a rotating
cylinder sleeve, maybe even diverting the exhaust
into a condenser rather than a conventional
exhaust pipe. You could even use the sleeve valve
to open and close the steam inlet, although it
might leak at the 2,300 psi pressure I was thinking
||Rotary valves might work for the steam inlet, but
they couldn't be driven off the crankshaft like in a
two stroke, since you would want the steam to
come in through the head. You could have a separate shaft above the head for this, similar to a
camshaft, but driven at the same speed as the
crank. If not bash valves, I was thinking of using
high-pressure diesel injectors for the steam, as
long as they would hold up to the high
||Steam injection during combustion would
definitely add some power, since the expanding
steam would impart more force to the piston and
crankshaft. The only question is how much more
power, and at what point would there be diminishing returns as the steam began to put out
the fire in the cylinder.
||If both the steam and fuel were directly injected
into the cylinder, you could alternate between
combustion and steam power, so as to avoid any
"quenching" of the combustion by the steam. The
combustion cycles would also keep the cylinder
walls and piston hot so that any water droplets in
the steam would flash into vapor.
||A twin-cylinder engine would work with steam in
one cylinder and combustion in the other, but you
would still be wasting much of the combustion
cylinder's heat. Not to mention the uneven power
outputs would make vibration and smoothness an
issue. But it would definitely be an easier way to
do things, and you could still use some of the
waste heat from the combustion engine to pre-
heat the boiler water. If I remember correctly,
BMW was recently testing something along those
||or you could make the combustion-stroke and steam-stroke interchangeable: one or the other. If there's enough of a head of steam that a steam-stroke would be equivalent to a combustion stroke in power, then the steam stroke is used until the pressure backs off a bit.
||Yes, that's what I was thinking. One combustion
stroke to heat up the cylinder, and then a steam
stroke. Or maybe several combustion strokes
followed by several steam strokes.
||One advantage to using steam is that the engine
will potentially last longer, since pressure is
applied to the piston more gradually than the
"bang" in a combustion engine.
||It also seems that something like this has been
tried, with a hybrid steam/diesel locomotive.
Check out the link. Although in this case the
engine switched between combustion and steam
based on its speed, in order to take advantage of
the high starting torque of the steam. This might
have applications for my engine, in that you could
eliminate the clutch and low gears and start the
vehicle on steam. Of course, you would need to
wait for the boiler to heat up first.
||bellauk65 read it again,this time a little slower, let it
sink in a bit... Its actually a good idea, good enough
that its (partially) baked by crower
||bellauk65, you could inject superheated water and
have it flash into steam inside the cylinder, but any
large temperature difference between the water and
the cylinder would cause metal fatigue over time.
||Can you tell me more about this nozzle?
||//metal fatigue over time// don't use metal. try
porcelain or some high heat plastic.
||Porcelain would be even more fragile than metal,
I don't know of any plastic that will stay intact at
temperatures of up to 400C like aluminum will.
||There are some ceramic materials that would work
for the cylinder liner and piston, but from what I
understand they are quite expensive or difficult to
||It would be possible to inject steam (or
superheated water) at roughly the same
temperature as the combustion chamber while still transferring heat from the combustion chamber to
the steam. Since the steam would be undergoing
expansion (and also a phase change), it would be
absorbing energy even while everything remains
mostly the same temperature, and thus avoid
thermal stress and metal fatigue from rapid
heating cycles. Check out the "four-stroke steam
engine" link for (lots) more info on this subject.