h a l f b a k e r y
Crust or bust.
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
or get an account
The Atkinson cycle engine was developed over a century
ago. The aim was to maximize the power stroke of the
cycle to extract more work from combustion. The original
design used a complex series of linkages to modify the path
of the piston so that the compression stroke was relatively
compared to the power stroke.
Modern applications of this engine look very conventional
however. No linkages or additional complexity. Instead,
they modify the timing of the intake valve. The point is to
open the intake valve as the piston starts to move down,
drawing in fresh air. Conventional engines close the valve
as the piston reaches the bottom of the stroke, then
compress all this air for later combustion. The Atkinson
modification found in many engines such as the one in the
Toyota Prius, closes the intake valve late. Now, the air
enters the cylinder before a portion is pumped back out,
effectively lessening the intake volume and increasing the
power:intake ratio to achieve the Atkinson effect.
Now, I think this is backwards. The same effect, a smaller
intake charge, can be achieved by closing the intake valve
much earlier. A conventional Otto cycle engine closes the
intake valve just after BDC, around 50 degrees ABDC or so.
How about 50 degrees or more before? What effect would
this have? Well, the intake valve closes while the piston is
still descending. The piston continues to descend and now
pulls a partial vacuum in the cylinder... The crank keeps
rotating, the piston rises and eliminates the partial vacuum
before continuing to create compression and the rest of the
Now, given the perhaps unmatched maturity of the spark-
ignited gasoline engine, this is definitely a dreadful idea. I
can't, however, figure out why. I see several advantages:
1. The early intake valve closing achieves the reduced-
charge Atkinson effect, WITHOUT the pumping losses
associated with intake and partial exhaust (de-intake?)
through the intake valve. Why move the air in, if you're only gonna move it right back out again?
2. As the piston approaches the bottom of its stroke, it has
to be decelerated, then it hangs around doing very little
before it has to be accelerated back up the cylinder. By
closing the intake valve early, the partial vacuum* applies
a force which decelerates the piston then accelerates it
again in the opposite direction. It acts as a spring.
You can demonstrate this with one of the large numbers of
syringes that I'm sure everyone has on their work desk...
pull out the plunger a little, block off the end... keep
pulling and let go.. The atmosphere pushes it back.
Newcomen wasn't a fool. So instead of wasting crank-
derived energy pushing some of the intake charge back
out, you save energy by moving less air.
While piston deceleration force is delivered to the crank, I
reckon about this much:
Piston+wrist pin+1/2 rod = 0.64 kg
Bore&Stroke 85 mm x 90 mm
Force = 0.64x15,000ms/s = 9600 N (for 5000 rpm)
Work Done 432 J/rev.
Piston area 0.0054 m2
Pulling 20% max vacuum, average 10% = 10133 N/m2
Mean Pressure force = 55 N
Work Done = 2.5 J/rev
So about 0.5 % of the piston acceleration forces never
move through the piston assembly. Hmm, nice but not a
3. The low pressure component of the cycle could increase
fuel vaporization. Gasoline already boils close to engine
operating temps, lower the pressure for a little while and
this should speed up vaporization.
4. The intake cam duration is shorter. The intake valve
spring is compressed for less time and the cam is
consequently is doing less work.
Like I say, this is almost certainly a bad idea, but I don't
*Or pressure differential between the crankcase and
cylinder... if you're gonna be picky about considering the
Early closing of intake valve, reduced pumping losses
[bs0u0155, Jun 02 2016]
Variable Valve Timing
[bs0u0155, Jun 02 2016]
[bs0u0155, Mar 03 2021]
||um... this might be a stupid question I'm not sure, don't you need the oxygen to make the gasoline ignite? By firing after creating a partial vacuum you'd be asphyxiating the gas... wouldn't you?
||//don't you need the oxygen to make the gasoline
||Absolutely you do. I'm not pulling much of a vacuum, 80%
of a normal intake charge is retained. The key to an
Atkinson cycle is that the gasses from the burned fuel and
air are much more than the gasses that went in.
Normally, the extra gas in the same space creates excess
pressure.... but once the piston gets to the bottom of the
stroke, the pressure can't be translated into useful
engine power, instead, it just waits until the exhaust
valve opens and whooshes out with real vigor as waste
heat, pressure and sound. That waste heat and pressure
may be used to power the turbine of a turbocharger in
other engines. The Atkinson cycle reduces the waste by
restricting the air (and fuel, since the engine is designed
to maintain a constant ratio of air:fuel for optimal
combustion) that goes into the cylinder. This means that
the smaller air&fuel combustion can expand more
completely. This engine has a cooler, quieter exhaust,
uses less fuel per unit of output power, and allows you to
bore people in pub conversations. The principle downside
is that the engine is totally gutless at low revs. That
"waste" pressure of the standard, Otto cycle engine,
makes it tourqy and punchy. Toyota didn't care too
much with the Prius, since the electric motor does a lot
of the low speed work and, how can I say this... the Prius
customers had pre-managed expectations when it came
||A few manufacturers are now selling non-hybrid Atkinson
cycle engines, which are using fairly fancy variable valve
timing to switch the engine in and out of Atkinson mode
depending on load.
||I suppose the best way to understand it is to think of a full charge, un throttled which is what we are working on from a thermodynamic perspective. Your design would achieve the same effect as a throttle does, reducing charge by restriction: the charge is still compressed at the same ratio as it is discharged. Instead you must reduce the efficiency of the compressor on the intake stroke (and increase the compression) WITHOUT forcing the engine to suck a vacuum which is where the benefit is gained.
||//Your design would achieve the same effect as a
throttle does, reducing charge by restriction://
||I was thinking about this. A long time ago, while observing
the long tortuous route the intake for my crappy-but-
loveable 1998 Nissan Micra took on it's pilgrimage around
the empty engine bay, I wondered if restriction was
essentially designed in. One snazzy cone filter later and I
had a noisier car with slightly worse fuel economy. So,
||You are entirely correct about the throttle of course.
What I've invented is an engine running at part throttle.
The throttle in this case is the intake valve rather than
the butterfly valve in the throttle body. Now, if you could
vary the valve timing well enough, you could gain some
control over the process... What I've invented there is
variable valve timing for the intake, like BMW's VANOS.
And that's already been invented.
||If you had REALLY good control of inlet valve timing and
lift, you wouldn't need the throttle body at all.
Unfortunately BMW already invented that too, it's called
Valvetronic. There are others too.
||//Instead you must reduce the efficiency of the
compressor on the intake stroke (and increase the
compression) WITHOUT forcing the engine to suck a
vacuum which is where the benefit is gained.//
||Well, no. I dispute that pulling a vacuum is energetically
worse than pumping air in then out again. Closing the
inlet valve early does work quite well at reducing
pumping losses. See intro of <link>. So, why is it a
dreadful idea? Well, closing the inlet super early gives you
a really short duration. That means you either have to
have a really aggressive and inefficient cam profile, OR
you go with much lower lift, which makes the engine poor
at high rpm. The Atkinson cycle allows a lazy long
duration cam profile and you actually need less lift for
the same total airflow.
||Cool, I learned something.
||So, at the end of the day, is there any difference between what a throttled engine achieves vs. an Atkinson ?
||(but yes, this idea is pretty much obvious - if you want to restrict the aircharge in an Atkinson, why not do it on the intake stroke, to avoid pumping air around needlessly [edit: hmm, okay the intake valves don't have to work as frantically])
||One advantage of your idea compared to throttling is
that with throttling, the piston is pulling a partial vacuum
for the entire stroke. Energy is lost as heat as air is
forced through the throttle. With your idea, there is less
resistance for most of the stroke. When the valve closes,
it closes all the way, so air isn't being forced through a
restricted opening, and the partial vacuum that develops
is actually slightly useful in slowing down the piston.
||The valve issues do sound like one possible reason not to
do this though.
||In the "Variable Valve Timing" link there is a section on
"Early intake valve closing (EIVC)". It sounds like it
describes this idea exactly. It reduces pumping losses by
40% and increases fuel economy by 7%. The downside
they mention is that it increases hydrocarbon emissions
because it reduced the cylinder temperature.
||That seems strange because I always had the impression
that other design trade-offs where made to prevent the
cylinder from getting too hot, so something that reduces
the temperature seems like a good thing.
||so has anyone thought about a slurry of industrial diamonds and gallium lining a plenum at they ylinder walls to be ultra warmth conductive to reduce emissions? 1 part per hundred is like an could have fiscal value
||I always thought this was a jolly good idea of mine. It turns out
VW have termed it the "Budack" cycle <link>. They've got
around the problem of the lumpy cam using those fancy
hydraulic cam-follower fulcrums.
||I recently saw (umm... something?) about the Budack cycle.
A quick Google brings up articles from as far back as 2017,
so you still beat them. Damn professionals, reading the
||//Damn professionals, reading the halfbakery again...//
||I could be quickly silenced should a shiny new/well restored
old VW camper were to be offered...
||Pulling a partial vacuum in the engine I would think would be
bad for the oiling system somehow. I suspect you'd get a lot
of cylinder wall um... what's the word... (tarnishing?) with
this arrangement. I could be