12 c
2018.03.30
Hi Paul,
another impressive video, making the combustion process visible ;-)
With regards to processing a homogeneous mixture, I've found a german
paper, dated 1976.
Engine was a NSU KKM 871, 2 rotors.
For preparing an ideal mixture they've used a "boiler" / drum. Fuel
was heated up to 200°C + fed into the boiler / big chamber. Mixture
temperature was 70°C out of the boiler with all fuel evenly
vaporized. Intake temperature at ports was ~ 65°C.
Compared to a conventional downdraft carburetor you can see some
advantages. » pict_1 spez. Kraftstoffverbrauch » specific fuel
consumption in g/kWh KKM » RE Idealgemisch » ideal mixture
Vergasermotoren » engines with carb Kraftstoff-Luftverhaeltnis »
Relationship of fuel + air » Lambda Teillastpunkte » part load
"points" / adjustments
Without the availability of fuel injection David Garside has found a
very good solution for the Norton motorbike: air gets into the carbs
with ~ 100°C + will be cooled down because of fuel evaporation to ~
50°C before reaching the chambers. He has used SU carburetors
(variable venturi / constant velocity carbs) - they ensure a pretty
good fuel spray.
In general there are some rules of thumb, one may pay attention to
... » duration of evaporation of a gasoline droplet » at 30°C it
takes 8 times longer than at 80°C » higher temperature is preferable
» 8 million droplets with a diameter of 10 µm will evaporate 200
times faster than 1 droplet with a diameter of 2 mm (same volume) »
the smaller the droplets the better
With fuel injection into the intake manifold or PFI ... » higher flow
velocity in the intake manifold enhances fuel preparation » lower
pressure in the intake pipe (e.g. at WOT) enhances evaporation, too
It's preferable to get fuel droplets as small as possible. That's why
the Renesis engine has primary injectors / nozzles with 12 holes,
generating a relatively fine mist of fuel.
Another solution may be a swirl injector like the one by Mitsubishi,
1996. » pict_2 + pict_3
Very small SMD (Sauter Mean Diameter)
Once the mixture is inside the engine, it's important to avoid
condensation / fall-out of the gas. Was a very big challenge for
Daimler-Benz with the DFI system. Therefore the walls should have a
temperature of at least ~ 40°C. More is better. That's one of the
reasons to "heat" the cold lobe a bit ;-) (Another one is to avoid
deformation of the housing shape)
The disadvantage with warm / hot mixture + walls is, that volumetric
efficiency will decrease. Again, finding the right balance is the
target.
Quenching ... sorry, the next part is a bit "academic" ;-)
Related sketches: t-section.pdf
At all walls of a combustion chamber you will have a wall layer - a
thin film of gas, that will not be burned. Reason: heat loss to the
walls is greater than the amount of energy produced by combustion »
extinction of the flame front.
Wall layer will be smaller / thinner with ... » higher pressure
(exponential effect - good) » richer mixture (simply more energy
available) » higher wall temperatures (limited by compulsions of
lubrication)
The thinner the "chamber" the richer the mixture has to be. But this
will - of course - increase heat losses ...
Near the "cool" walls the disintegration of molecules + atoms of fuel
will be stopped. Termination of chain reactions, free radicals will
partly be bound to the wall + converted into stable species.
Sorry, have forgotten the minimum temperature levels you need for
satisfactory chemical division.
No further chemical reaction will occur - that's the main issue /
challenge. There will be no more conversion, e.g. CO to CO2. That's
the phase, where you will loose a lot of energy / heat / power ...
A simple combustion model may make it more clearly (?) ... 1st step:
emergence of short chains (Alcanes ?)
2nd step: appearing of radicals like H, O, OH ... (Alcenes ?)
3rd step: "cold flame", generating ~ 10% of the energy / heat
4th step: "blue flame", you will "see" CO, H, H2O. Oxidation to CO
delivers ~ 30% of the heat. This phase may be (partly) not reached in
the quenching areas.
5th step: the main heat / energy producing step. Most of the time
this will not be reached in the spheres / sectors of quenching:
conversion of CO to CO2. This step generates ~ 60% of the heat /
usable energy. + only just now you can measure the big / huge rise in
temperature.
Sorry, my English isn't good enough to explain everything correct. We
would need a chemist now ;-)
Another challenge will be to initiate combustion In the mixture + not
at the edge. Flame front is "creeping" along the sliding surface +
will be cooled down unnecessarily. Possible countermeasures ? Perhaps
laser or some kind of "jet" ignition ... ?
After all, picture_4 + picture_5 show the influence of heat loss,
crevice volume + leakage. The last point is the most important ...
But now it's really time to "shut my mouth". So much blah-blah ...
;-)
Best regards
Frank Herfert
http://www.enginelabs.com/engine-tech/video-the-secrets-of-the-rotary-engine-seen-from-the-inside/
; an interesting video.
Murry Rozansky
I don't know how applicable my observation is to the full scale
automotive version but I did notice, in the slow motion acetylene
fueled take at the end, that the squish zone was where the combustion
seemed to start, i.e. the trailing end of the combustion chamber.
The resulting squish flow seemed to spray the flaming result of the
combustion in the squish area into the main charge area ahead of the
squish flow area in the rotors direction of rotation.
I don't see where the squish flow tends to blow the flame out, as
posited here before, but rather that it seems to act like a "jet"
ignition projecting flame into the main body of the charge.
It appears that the combustion at the trailing end of the combustion
chamber actually completes while the flame is still vigorous in the
leading end of the combustion chamber.
This may be apples to oranges. I don't know how this applies to a
full scale version.
Mark LaPierre
Did you read the U of Vienna paper?
Paul Lamar
I did read the paper that you attached. My first comment is that the
software used was optimized for use in a cylindrical SI piston engine
and had to be modified for an approximation of the conditions inherent
in a Wankel rotary engine. The test engine was in built in cooperation
with Austro Engine GmbH. The limits of efficiency of the Austro
fuel/air cooled rotor engine has been well discussed in this email
forum. I am not surprised to see the large Implementation and
Combustion losses indicated on Figure 37 between the test engine and the
SI engine at WOT.
Only one point of ignition, i.e. one spark plug fired one time, was
considered. The video link showing the operation of the scale model
Wankel engine used a glow plug as the ignition source. In the scale
model it appeared that the ignition event began very close to the
trailing edge of the combustion chamber and proceeded to flow in the
direction of rotor rotation. It looked like a flame thrower squirting
flaming gasses in the direction of rotor rotation igniting the main mass
of fuel forward.
The text explained how they experimented with moving the spark ignition
timing earlier in the rotation cycle to negligible effect. There was no
comment of moving the ignition timing past TDC to any degree.
On the Lemans engine an additional spark plug was added. Was that plug
located, or should I say used, at the trailing edge of the combustion
chamber perhaps to the same effect as seen in the scale model Wankel in
the video?
If a transparent wall was installed and a video made of the combustion
process then there would be visual evidence to support their
mathematical modeling. One should also consider the validity of the
assumptions made in the mathematical modeling.
In the losses calculations the analysis engine was compared to a
theoretically perfect engine having no leakage. That's a pretty tough
comparison. Any real world engine will look bad in that comparison.
Mark LaPierre
------------------------------------------------------------------
Please add your questions or comments at the end of the thread. Not
the beginning.
I am going to enforce that rule. If people don't do that it will not be
forwarded. My ISP changed the rules.
The Rotary Engine free News Letter. Powered by Linux. If you want off
the list PLEASE let me know and I will remove you.
ACRE NL web site.
http://www.rotaryeng.net
You Tube
http://tinyurl.com/beqqxas or PaulLamar2
Copyright 1998-2018 All world wide rights reserved.