Towards a two-stroke turbodiesel aero engine

The saga continues. But it doesn't look good in a real sense ; I mean I wanted this really to be up-scaleable, not a joke in a shed.

yet
1) We began to test ignition. We have 5 tiny injection holes in the injector, arranged very precisely at 72 deg, pointing down at 20 deg. We aim them perfectly into 5 small combustion chambers on the piston top (all of which we've worked out and made. The foto shows the neat jig which ensured we were doing right). This improves combustion enormously.
2) Commercial road engines have had megabucks spent to optimise low-temperature burning to reduce NO2, but that is against efficiency. We don't have that problem, and are putting extra air into the exhaust port to set fire to carbon (seen as soot) and get more out of the Turbo-charger. (So immediately an electronic problem occurs which we haven't tracked yet - scream -I work at it more slowly now, so as not to go completely mad).
3) We are not getting the torque we want. We can't understand this but are working on it. Obviously, it's "not enough air". You only have a quarter of a rev in non-motor bike two stroke to get the air in. You need force. Big time.

Malc
 

Attachments

  • injector jig 09-33-59.jpg
    injector jig 09-33-59.jpg
    33.3 KB · Views: 1
  • Fluted piston top.jpg
    Fluted piston top.jpg
    40.8 KB · Views: 2
Last edited:
i think your a god'n sir. (Removed disguised attempt at swearing as per our T&C's) all the haters that dont want it to work. it will work. what about a supercharger????
 
Last edited by a moderator:
Well, thank you.
Thermodynamically, it is a great mystery that with a centrifugal supercharger (which is in place) the air density (that I need) is proportional to air pressure, but air pressure is determined by air flow. Not the other way round.
This is strange. You'd think you'd get pressure and flow would follow - but not so.

(I am working with a bypass of air to increase flow - hoping to get more pressure)
 
Good luck you crazy engine building person you. I wouldn't know where to start with all these shenanigans. But I do know that the world would be a duller place without tinkerers and underfunded dreamers.

Take frank whittle for instance
 
Thanks.

Some diesel experts said I'd be lucky even to get it running. It's been as hard as I expected. Things that seemed impossible were relatively easy, things I assumed would work were not.
I have had a journey with numerous very steep learning curves, ranging from elementary thermodynamics to electronics. Many parts had to be machined and I was very fortunate there.
Throw in odd failures, hard to trace, and it's more maddening than rewarding.
And it still isn't much good!!!!

(Before a mountain walk - several Cairngorm hills - about 10 years ago, I said, "Well, we can always turn back" My daughter said "We are the two people in the world who never turn back").

At present, I can't understand why my supercharger bypass (air going round in a circle) does not improve pressure.
 
Maybe it's because you are effectively equalising the pressure that the supercharger creates, pressure going in is the same as going out...

does this engien have a turbo also?
 
Well, fair comment. But it is a recognised way of making a "big" compressor avoid getting stuffed when it can't provide forward flow. In a nutshell, I thought if a portion of the air is fed back to the sealed intake plenum (wait for clarification!), it would add to intake energy and get an extra push, adding to net pressure at compressor outlet. I think this is an OK idea - I thought it out but it is written in to "How to Deal with Surge," a complex thesis entirely beyond me written in English by a Dutch guy. (Surge proves to be a well known Russian criminal known to enjoy damaging turbocompressors). My bypass dimensions are guess-work (but small compared to main airflow) so safe.

There is a turbocharger
which does add net pressure to a plenum from which the supercharger gets its air. The bypass feeds into this, streamlined, then expanding. So it's turbocharger, then compressor. At start up the compressor easily sucks air through the turbocharger.

I was puzzled as to why Airflow determines Pressure in a centrifugal compressor. I would have thought Pressure would create the Flow. But the answer is obvious, now I've been told:
A rotating wheel makes stationary air move incredibly fast (invariably > Mach 0.3). The air is both spinning and moving sideways. This involves adding lots of energy to it. It's kinetic, moving, energy. But we want pressure, not movement (Well, a bit of movement).
The "nozzle" in the casting/machining of the compressor feeds into the widening spiral diffuser which slows it down. Where does all that energy go? Into pressure. So Flow=speed>slowing>Pressure. (The volume flow doesn't get back to "base", ie what was being sucked in).

Finally I get it! Aaagh....:blink:

Racing car diffuser: simpler version of these goings on. No energy added. Volume flow is same, and pressure drops.

But I am giving it a rest for a while.
 
Last edited:
Just re-capping. I took a Petter AC1 single cylinder engine base and designed a completely different head to operate the rig as a Two-stroke. Indeed it looked a bit like a VW Golf set-up, but it was designed entirely from first principles.

The reason for the two-stroke idea is three-fold: 1) Two bangs per two cycles therefore smoother than a four-stroke - aircraft propellers hate vibration; (2) Slightly less "power" per bang, again promoting smoothness; (3) two-strokes are very efficient at a particular narrow rev range, and aircraft mostly fly at Cruise or (occasionally) at Full Power.

The air from the blower drops straight down but, at entry, is angled by the back of the valve head to promote loop scavenging.

The Combustion Chamber for this diesel is tent shaped - quasi flat bottom (=piston top) (don't panic), two opposing walls made of valve faces and blank surface.

The injector (Delphi; ex-Clio) is at the apex of the tent, 7 mm off true cylinder centre. After much interesting analysis, I understand the injector, ie inductive & impedance properties, current change during action and voltage kick-back. All this is OK.
There are 6 holes (170 mu dia) at 72 deg intervals, squirting down at 18 degrees from horizontal. The pressure is about 1000 bar. The fuel is about 80 deg C (from pressure applied to its viscosity) on entry to the cylinder.

The best previous piston top had 6 flutes which "reflected' the jet (see earlier post & foto).
The current top is, I think OK (or we wouldn't have done it) and has a ring machined as a groove just where the jets would impinge. Its centre is centred on the axis of the injector tip. This ring, in section, has a vertical outer wall 2 mm deep but a there is a slope up and in towards the flat of the piston. There are 2 minor supplementary rings concentric with this, of smaller dia.
We had found that at these delivery pressures, the rings (like the flutes) provide walls which very finely atomise the fuel.

Usually, with these injectors, the fuel jets catch fire before they hit the piston (classically the chamber wall; in our case, the wall of the groove)

Next (the famous) Bump Space (BS) at TDC. I didn't know what this was - just hadn't heard the phrase. But anyway, Petter have their acceptable range.
Owing to a miscalculation, I have a BS a tiny bit less than Petters limit (40 mu less as I recall). Since I know there is totally free movement (ie piston does not strike exh valve face), I was happy.

But could this affect anything? I don't believe so. Of course the CR goes up a fraction - but squish air movement? As the piston approaches TDC the BS is millimetres and the piston is slowing to walking space. You don't expect that, do you? But squish effects must have occurred much before TDC.

We believed the air flow in our chamber would be like a horizontal rolling wave around a horizontal axis, compared to the horizontal rotation around a vertical axis in a toroidal chamber, ie it should be fine (& there's that VW four-stroke head!).

We put the design on the Lotus Engine Simulation software and it met expectations big time which was encouraging. But that assumes perfect combustion so the design is as of this date, still unproven.

Problem we had: Difficulty starting. We blow in warm air (multiple gloplug system) and there's a gloplug in the combustion chamber, tho not ideally situated because of lack of room. The starter turns it fast. Brown smoke appeared in puffs but there's no "catching on".
Sometimes we thought the firing throws the starter pinion off but then doesn't continue on its own so the pinion re-engages unpleasantly.

It's as though the scavenging is failing and the cylinder fills up with burnt gases so cannot combust. Could this be a total design flaw? Remember, this is my design, not a modified existing item.

We alter valve timing and injection timing but its much the same story. In general, inlet opens 20 deg after exh and closes 10 deg after exh closes. We have altered injection from between 18 and 5 deg BTDC.
We tend to go EVO at 95 ATDC, or recently 120 ATDC. Compression is high so we lose a little by closing IV late (eg 250 ATDC, ie 70 ABDC.) but it should not affect things.

We think of every possibility and still it wouldn't start. Could the exhaust be blocked (remember there is the turbine there)? I blow into it, no probs, except dirty face. Blow air in; air flows thro' exh valves (set open) and out of the inlet. Seems OK.

Then I decided to take the exh manifold off and peer into the exh port (where I can see into the cylinder during "EV open". Will I see a flame momentarily?

My God, It fires up beautifully straight away. After a year of struggling. Not even smoky. Within minutes the problem is clear. Not our engine, its the turbocharger!! Its leaking engine oil (which supply its bearings) onto the exh turbine, which has formed a tarry deposit on the blades. The turbine spins OK but has been restricting, but not blocking, passage of gases between the blades.

We will rebuild it and look forward to telling you about the future of the Light Aircraft Industry!!!!
 
Last edited:
It's been a while, good to hear from you again! Well done and keep us posted on the latest developments. Glad you ironed out the turbo glitch! I bet that came as a relief!
 
seems complicated for such an investment, of time and money!

but(i hope we shall see)!

something of interest , have you ever studied any of the General motors Detroit Diesel two stroke engine, very popular in USA and in military equipment!

1st ones were N/A with a scavenger pump, then on to turbocharged into the scavenger pump, final versions were turbo with a bypass valve in the scavenge pump!

they were very powerful for there displacement size.

drawback was emissions and fuel consumption!

personally i liked them, unusual smooth sound, along with smooth running, you should hear a V12 detroit 2 stroke at full song, marvelous!

all 12 cylinders fire in one revolution of crank, can we say whine!

and they were some of the 1st modular engines, many interchangable parts, 2cyl. pistons fit from 2-3-4-6 engines in-line, and also in V-type engines, engines from 2 cyl. up to V16 cylinders, small parts fit all engines,.

anyway there design principles are very interesting!
 
also let me add, back a couple yrs, i met a person from UK, and he showed me a 2 stroke diesel engine made in UK, for light aircraft!

it was named GEMINI,6 piston 3 cylinder opposed, much like the german Jumo engine of WW2, engineering design principles!

it was at an air show in USA, and it started and run fine!

i was quite impressed, but have never heard any more about the development of the product!
 

Attachments

  • merc cruiser, and gemini diesel, diesel hawk, super boat, tracto 015.jpg
    merc cruiser, and gemini diesel, diesel hawk, super boat, tracto 015.jpg
    63.4 KB · Views: 2
Well, sincere thanks for that.
I didn't know this for certain but our plan is exactly that. Wilksch, with whom I was involved, has some engines flying but they suffer from weight because of a high cylinder block. The Jumo engine was very efficient but very complex. It had two cranks, needing heavy gearing, and slightly asymmetric stroke positioning to get away from classical two-stroke symmetric valve timing. Our engine has completely asymmetric timing. I was aware of such an engine about 15 years ago but was never clear if it made it to any trials. Was that the Gemini, I wonder?

We are proceeding steadily but I am ageing and that might be a problem.

I have had to master aspects of thermodynamics and electrical engineering to be confident of what we are doing. It would be too easy to say "We are doing what all the car manufacturers are beginning to do. ie, a small engine with high air compression feed." (An engine just burns air for its power). The OEMs can move a million times faster than us!

If you've read my piece, you'll see the engine is "more" compact because it doesn't need a long piston as a sleeve
valve.

I'd be happy to hear anything you'd like to add.
There is more detail (including uncertainties on malcolmcochran.wordpress. com)

Malc
 
Last edited:
Re post 92.
Doesn't help.
Of course I know about other two stroke diesels. The first diesels were two-strokes. Of interest, maybe, what won WW1? Not the usual answer. It was Henry Ricardo's tank which simply made trench warfare out of date at a stroke. Powered by his two-stroke. Do you know the Ricardo company? Still going (very) strong.

Recently, I was talking to an elderly world class scientist (in fact my mentor) and he said (about something else) "Advances are made when two two proven ideas are combined." That is what we have done.
I still can't believe it hasn't been done before but I cannot find any reference to our system.

You'd have to read all I've written to get the idea. But a quick way might be to look up SAFRAN's news. They've built a big diesel but I'm not clear if its two or four stroke.
 
Similar threads

Similar threads


Please watch this on my YouTube channel & Subscribe.


Back
Top