h a l f b a k e r y
The halfway house for at-risk ideas
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
or get an account
This post will clearly illustrate my relative ignorance of
As I understand it, a supercharger uses a mechanical
(from the engine) to blow air into the cylinders at above-
ambient pressures, increasing the power output of the
engine. A turbocharger does the same
thing, except that
the blower is powered turbinically from the exhaust
In each case, engine power is being tapped to pump
air into the cylinders, with a net gain in power.
Now imagine a conventional cylinder in an engine. The
top end (ie, above the piston) is of course where
everything happens, and is closed off by the cylinder
The bottom end (below the piston), however, is open.
What would happen if the bottom end of the cylinder
closed off, by providing a second (upside down) cylinder
head on the bottom of the engine? Ignore, for the
moment, the piston rod. In this case, obviously, the air
this bottom space would be compressed as the piston was
on its downstroke (ie, during the power phase).
With a cunning system of interconnects and valves,
therefore, it should be possible for one piston's power
stroke to provide (from its new under-cylinder)
air which could be fed into another cylinder during its
Now back to the piston rod problem. The piston rod
normally exits from beneath the piston, and moves
laterally (ie, crank-like) as well as up and down.
However, with an additional linkage, the piston rod
move only up and down, with lateral movement in a
second linked part. Thus, the piston rod could be a
cylindrical rod, moving only axially. If so, then it could
pass through a hole in the new, second (upside down)
Obviously, the power to compress the air in the under-
cylinders comes from the power stroke (ie, it's not free),
but this seems no different in theory from using a
mechanical tap or a turbine to blow the engine in the
[xenzag, May 04 2014]
[Laughs Last, May 05 2014]
a bit more complicated... uses double ended pistons [pashute, May 07 2014]
Dual sided piston motor patent
I think this patent describes your idea [pashute, May 07 2014]
Same idea, but in a circle [pashute, May 07 2014]
Wikipedia: Opposed Piston Engine
and its history, beginning at 1900 and running a car in 1904 [pashute, May 07 2014]
Scientifically presented double faced piston engine
[pashute, May 07 2014]
||The region below the piston is not just where the
piston rod is located; it also is where the piston
rod connects to the crankshaft. So, the
"crankcase" is the section of the engine containing
all the connected piston rods (and the lubricating
engine oil). All those spaces under the pistons are
connected to each other. When you have multiple
pistons, you usually have some going up their
cylinders, and an equal number going down their
cylinders. Net effect: total volume of space in the
crankcase does not change, and so can't be used
as a pump.
||There are some exceptions, such as the "Boxer"
engine, and variants thereof, as described in
another post somewhere around here not so long
ago ("Ghost Cylinder Engine" by [FlyingToaster]).
||Erm, is this not a two stroke engine?
||ie the bit under the bottom of the piston (in this case - the crank case) has a one way valve, so when the piston descends it compresses the air/fuel to go into a (in the case of a 2 stroke - its own) combustion chamber.
||//All those spaces under the pistons are connected
to each other.//
||Perhaps my explanation wasn't clear. I'm proposing
adding a second cylinder head, inverted, at the
bottom of the cylinders, with the piston rods (which
will have to be doubly-linked) passing through holes
in this second head.
||//Erm, is this not a two stroke engine? // That's
entirely possible. See first line of idea.
||Yup, you've invented the two stroke, sort of. The other advantage of
the two stroke is that the work done by the piston (the underside) is
used to both intake and compress air, then force out the exhaust
gases. You need fewer "strokes" per complete cycle. So you
effectively double the power density of the engine. Unfortunately, the
whole set up is not conducive to good lubrication. Also, in multi
cylinder engines, one piston is going up as another is coming down.
Airflow in the crankcase is tuned to optimize this.
||The other interpretation of this title, is some form of powered bowel
pressurisation mechanism. For more rip-roaring bottom-burps.
||This would be much much less efficient than a
regular supercharger or turbocharger. And
apocalyptic from a maintenance standpoint.
||split-cycle, using one physical cylinder/piston.
||//That's entirely possible. See first line of idea
||Don't worry about not having an encyclopedic knowledge of the internal combustion engine, you've missed nothing...
||I was thinking more of the line after that one.
||What is astonishing about the internal combustion
engine is how so many variants have failed to solve a
very simple problem. I am proud to have added to
||and I'm embarrassed that in my yoof I apparently permanently gave brain space to Schneurle and his twin-loop scavenger system...
||There is (currently) only one link, so it's also the first.
||I think the basic model of this idea would be a 1cyl 2-stroke that self-scavenges, with intake flapper-valves on the piston and exhaust valves on the cylinder head. The size of the crankcase determines the scavenging pressure. Oh, and a flapper valve on the crankcase.
||It's also the direct opposite of one of my recent posts, but I can see my way past that [+]
||If you think of this as a two-stroke, then there is
no aspect of supercharging because the volume of
air pushed by the back side of the cylinder is the
same or slightly less than the volume needed to fill
the piston during the intake stroke (unless you
have a stepped piston engine which is a whole
new mess of possible optimizations.)
||On a four stroke engine it could be arranged so
that two pistons are compressing air in the back
side area at the same time that one piston is in
the intake cycle. This could theoretically result in
nearly doubling the mass of air in the cylinder for
each cycle. I don't know how much additional
compression is useful, but if you were satisfied
with a boost that was significantly less than 2x
(maybe 1.2x?), it seems like a 2 cylinder boxer
engine could accomplish that with very little
modification. I assume a 4 cylinder boxer would
need to be modified to split the crankcase into
two sealed chambers.
||It seems to me that putting a joint in the piston
rod is going to really hurt performance. Any gains
due to supercharging will be lost in max RPMs.
However, if you divide up the crankcase into a
chamber for each cylinder with seals around the
crankshaft separating these chambers, then fill the
space so each rod is running in a narrow slot
(maybe even make the rods narrower), the total
displacement of each cylinder due to piston
movement will be significantly larger than the
dead space around the crank and rod, getting you
much closer to the theoretical 2x.
||The problem with what you've just said, [scad], is
that it appears to be based on a sound
understanding of the subject.
||I previously said that I figured a jointed rod would
hurt max RPMs. But then I realized that if you
have the part of the connecting rod that is
attached to the piston constrained to movement
in 1 dimension, you no longer need the wrist pin
at the piston. In addition, my understanding is
that the height of a piston is determined partially
by the need to make sure it doesn't tip as it moves
up and down. With a piston and half the rod as
one solid unit, the piston is much better
constrained against tipping, so the piston portion
might be smaller. Maybe, the reciprocating mass
could even be reduced in this arrangement. The
engine would be a bit bigger because the rotating
part of the connecting rod would probably need to
be the same length as the entire connecting rod
before to avoid putting it at a larger angle.
||[scad] you mean a "crosshead" shaft, used on larger diesels to prevent sideloading on the cylinder walls.
||// milking from one power unit to the next // I
interpret this comment to mean that you don't
think this would be useful because it's just using
power from one cylinder boost another cylinder.
If that's what you're saying then you're missing the
point of a supercharger.
||Of course the supercharger uses energy that could
be used to turn the wheels, but it forces more air
into the cylinders. That allows the engine to burn
more fuel per cycle, getting more horsepower out
of the same size engine. So why not just use a
bigger engine? Because most of the time you
don't need an engine that big. For example, the
Ford Ecoboost engines, as the name implies, are
designed to save fuel. They do that by using a
smaller engine than one might otherwise want,
but with a turbo charger to get the horsepower
needed for short periods of time. During hard
acceleration, the Ecoboost is probably not as
efficient as a larger engine putting out the same
horsepower, but most of the time, when the
turbocharger is bypassed, it will be more efficient
than the larger engine. Similarly, with [Max]'s
engine here, I would expect that during lower
power demand times the supercharger function
would be disabled by opening different values to
vent the air rather than pressurize it.
||"crosshead shaft" Yeah that's about what I was
saying. And according to the wiki, those are
heavier and therefore only used in large slow
moving engines, so I guess I'm showing my
ignorance here. But it still seems to me that you
can stabilize the piston using less materials with a
long narrow arm than with an extended piston. I
almost wonder if they are only used in big engines
because it adds complexity. The slow part is a
result of the big engine, not the crosshead piston.
If a crosshead piston is always heavier, why
wouldn't they just make the piston in this big slow
engine a little larger to deal with the side load
rather than going to the complexity to have the
||The piston on a crosshead is heavier because it has a rod welded to it. I don't think they meant that the piston-head on a crosshead system is heavier (all other things being equal). The crank-rod (on the other end of the crosshead assembly) is the same length as it would be if there wasn't a crosshead there.
||For simplicity's sake, I'm still viewing Max's device as a single-cylinder 2-stroke, where the compressed air whooshes through the top-cylinder, scavenging the exhaust out, though he has the "undercylinder" feeding another cylinder (in a 4-stroke one would imagine), to improve pumping efficiency.
||Unless I've misunderstood the whole thing and he's actually talking about trying to increase the pressure in the cylinders in which case bollocks, where's the extra air coming from ?
||I have no idea what [Max] is talking about. He's
quite unreliable on this entire subject.
||I'm taking this as an "add idea to HB, see what happens, then stir" type event.
||// trying to increase the pressure in the cylinders
in which case bollocks, where's the extra air
coming from ? //
||The extra air comes from the fact that on a four
stroke engine, every fourth stroke is an intake
stroke on the front side, but every other stroke
on the back side can be used for compression.
That gets you close to 2:1 increase in air. Perhaps
I'm saying more than [Max] originally intended. He
||// it should be possible for one piston's power
stroke to provide (from its new under-cylinder)
compressed air which could be fed into another
cylinder during its intake stroke. //
||I interpreted this to say: // it should be possible
for one piston's power stroke [and intake stroke]
to provide (from its new under-cylinder)
compressed air which could be fed into another
cylinder during its intake stroke. //
||ahh, so two cylinders worth of air into the crankcase (or external airbox) vs one cyl's worth of 2x (minus the displacement of the shaft) out into a cylinder. Or words to that effect. A crosshead shaft protruding from a double-ended cylinder is standard in steam engines. Higher reciprocating mass and extra friction on the piston-rod(though the rings or whatever on the undercylinder head don't have to withstand any real pressure to speak of). Almost certainly better efficiency (and certainly waaay more power) than a supercharger when it's on, pumping losses when it isn't.
||Of course there's the equal possibility that he's just gone and cut'n'pasted random enginey words into the text-box.
||// The piston on a crosshead is heavier because it
has a rod welded to it // Yes, we're agreeing on
the terms. I'm saying that a piston plus crosshead
rod might weigh less than a piston design to
handle the same side loading without a crosshead
rod. I also know that this disagrees with what
Wikipedia say, but here's how I see it.
||If this is a standard piston:
||An equivalent crosshead piston might look like:
||(++ is the area with the rings) It seems like more
material would be saved by removing the skirt (----
-) around the large diameter of the piston than
would be used to add the much narrower but
longer rod (========). Now the rod will need to
handle a much larger compression load than the
skirt so maybe it would be heavier, but I wouldn't
be surprised if the real reason they aren't used in
small engines is complexity and engine size.
||If size and complexity are the real reasons, then
adding this system rather than a traditional
supercharger might actually make sense.
||Good point; wouldn't really trust WP on that one: crossheaded engines are usually large, low-speed diesels for which there's no real equivalent... and the crosshead makes for increased engine size: it might not fit into a car.