Video of Ellwe Diesel? Running

[Chris W]

That's a crankcase scavenged 2 cycle similar to Fairbanks Morse. The second eccentric is likely for the jacket water pump.
The B&W Alpha shown in one of the videos is also 2 cycle but has 2 stage reciprocating blower mounted aft of the 4th cylinder for scavenging air. I've run a couple of 3 cylinder versions of that engine. 200mm bore x 320mm stroke and 400 rpm flat out.

 

[minh-thanh]

GreenTwin !​

Just my opinion,
If you really like it then design and build it
- There are 2 options:
1/ Based on that engine , design and build a similar engine and it can run with 2 stroke or 4 stroke or whatever you want.
2/ Design and build it as original.
As a Diesel Engine: Why Not ?
10 , 15 , 20 , 25 . 30 cc ...who cares ?
Making plunger with O rings : Why not ?
A friend of mine has a diesel engine with plunger and O rings and it runs pretty well
The injector? make it airtight is ok.
injector time - the time when the lobe begins to come into contact with the plunger from 22 to 45 degrees before TDC . At 45 degrees before TDC, the pressure in the cylinder is very low which is a huge advantage
Fuel ? you can run it with any fuel or fuel mixture you want
That's all you need to get the engine running, how it runs depends on how you make it and tune it...

 

[GreenTwin]

GreenTwin !​

Just my opinion,
If you really like it then design and build it
- There are 2 options:
1/ Based on that engine , design and build a similar engine and it can run with 2 stroke or 4 stroke or whatever you want.
2/ Design and build it as original.
As a Diesel Engine: Why Not ?
10 , 15 , 20 , 25 . 30 cc ...who cares ?
Making plunger with O rings : Why not ?
A friend of mine has a diesel engine with plunger and O rings and it runs pretty well
The injector? make it airtight is ok.
injector time - the time when the lobe begins to come into contact with the plunger from 22 to 45 degrees before TDC . At 45 degrees before TDC, the pressure in the cylinder is very low which is a huge advantage
Fuel ? you can run it with any fuel or fuel mixture you want
That's all you need to get the engine running, how it runs depends on how you make it and tune it...

I did a few layouts in CAD, and I would like to use a 2" bore, but that makes the flywheel over 13" diameter, which is larger than my lathe can handle.

A 1.75" bore gives a flywheel diameter of 11.7", which is a bit larger than I would prefer, but I could machine something that size probably.

A 1.6" bore gives a 10.5" diameter flywheel, and so this is probably what size I would build.

I have found that model engines with flywheels in excess of 10" (+ -) tend to be very heavy and cumbersome to handle.

I would base this engine on the original design shown in the video in the first post, as a 2-stroke diesel.

I guess I would have to build the engine, and then learn how to build the pump and injector.

I have been studying other's injector designs, and so I have some ideas.

Compression I guess would be 21:1.

I would run the engine on diesel, or perhaps kerosene.

This would be a first-time diesel for me, so much to be learned, but I can see from the work of others that it can be done.

I will start by modeling the flywheel.

I am not sure where the reeds are.
Are the under the crankcase covers?
I just don't seen a typical reed assembly anywhere on the engine.

Pat J

 

[GreenTwin]

Some photos/sections of the engine indicate a flywheel that protrudes below the bottom of the base, and others show a flywheel that does not touch the ground while the engine is on the base.

I think I will go with a flywheel that will not touch the ground with the engine on its base.

Using the smaller flywheel design, I could use a 2" bore, and a 10.5" diameter flywheel, so I guess I will use those dimensions.

Probably a 1" crankshaft.

.

 

[Steamchick]

You have worked out it is a crankcase pumped 2-stroke diesel. See the transfer passage from crankcase to port in the cylinder, shown next to the piston "gas guide" in the section drawing. That shape was an early attempt of crankcase fed 2-strokes to increase the scavenging and minimise intake loss from the cylinder. The thoughts were that intake air is directed up one side of the cylinder, pushing exhaust down the other side and out of the port. But later developments showed the transfer ports pointing diagonally up are OK, and a flat headed piston is OK. Bigger transfer ports enable more power from better cylinder filling.
Instead of making a fuel pump for the diesel imjector, you can prove the engine runs using ether loaded model "diesel" fuel. But you need a simple carb at the air intake to do that. MODEL AIRCRAFT shops stock "off the shelf" stuff Farley cheaply. When the engine works, you can concentrate on the fuel injection pump and injector.
Have fun making the model.
Also,
There was a question "where is the reed valve?" - my guess is at the air intake to the crankcase.
Also, "why the large expansion chamber at the exhaust port?" - exhausting a pipe to atmosphere creates a huge shock wave = noise. But into an expansion chamber, the shock wave rattles around, and as the gas expands it cools adiabatically and the pressure drops, slowing the shock wave. The shock wave loses energy in this expansion, so when it then makes it noise at the end of the pipe it is much quieter. So the expansion box is just a simple silencer..... nothing more. From "BANG BANG BANG" to "bum, bum, bum" via a large tin box. Not quiet, but quieter...
K2

 

[Roger B]

The intake reed valve is visible on the left hand side of the crankcase of this Petter Atomic diesel.

 

[GreenTwin]

Very nice section Roger, thank you for that.

Pat J

.

 

[minh-thanh]

Hi Pat J !
Seems like 2 stroke engine is better choice than 4 stroke
no gears , no valves , no cams .. that saves quite a lot of time
I'm very interested about it . Please update if you can

 

[Nerd1000]

How did the oiling on these things work? I assume it is total loss like almost every crankcase scavenging 2-stroke but as a diesel there is no carb to introduce oil premixed with the fuel. Could be that there is a small oil injection pump somewhere on the engine that pumps to the main bearings? Or maybe even just a drip oiler into the intake?

 

[Roger B]

My 20cc two stroke diesel was loosely based on the Petter Atomic with some Field Marshall ideas for the crankcase. This ran but not very well. I think it actually used the fuel that was atomized by blowing past the piston rather than the fuel that was injected. I confirmed this by converting it to a four stroke. It now runs on the injected fuel but a lot of unburnt fuel/clag builds up in the crankcase. I am still working on improved injector designs.



Some thoughts/comments on two stroke diesels.

The Ellwe engine is described as having a pre chamber. This design tended to improve running with low grade fuels but the additional heat losses made the engine harder to start. The pictures show a starting paper holder (T handle sticking out of the cylinder head). A roll of blotting paper impregnated with saltpeter (potassium nitrate) was inserted in this and lit before attempting to start the engine (also used on the Field Marshall amongst others).

A two stroke doesn’t start compression until the exhaust ports have closed so the compression ratio needs to be calculated from this point.

Lubrication on the Field Marshall was pumped to various points via a metering system. The oil that collected in the bottom of the crankcase was returned to the reservoir by the crankcase compression and a one way valve. I installed this on my two stroke diesel but let what collected go to a separate vessel.

Two strokes have less parts than four stokes but all the ‘difficult’ stuff is incorporated in the cylinder. A change to the port geometry/timing may require a complete new cylinder and jacket rather than just a new camshaft on a four stroke.

 

[GreenTwin]

The black box on the front of the Ellwe engine in the first post is an oiler.

I am not sure about oiling the small end rod bearing.

.

 

[GreenTwin]

Hi Pat J !
Seems like 2 stroke engine is better choice than 4 stroke
no gears , no valves , no cams .. that saves quite a lot of time
I'm very interested about it . Please update if you can

I have to agree, the 2-stroke is a much simplier design, or so it would seem.

Perhaps a bit more complex from a casting/cores standpoint.

I have not had time to start on the flywheel; lots of work deadlines this week.

.

 

[GreenTwin]

So far, I have mainly worked on steam engine designs, which really don't have a compression ratio like an internal combustion engine.

So I will have to plow some new knowledge ground here.

Tonight, I am looking at a 10.6" diameter flywheel, 1.6" bore, 2" stroke, overall engine height of about 14".

A 2" stroke with a 1.6" bore gives a swept volume of 4.02 cubic inches.

The clearance volume is the combustion chamber volume, plus any volume added by a head gasket.

For simplicity, lets assume a flat-topped piston for now.

To give a 21:1 compression ratio, the clearance volume would need to be 1/21 of 4.02 cubic inches, which is 0.19142 cubic inches.

So the clearance from the top of the piston at TDC (top-dead-center) to the top of the combustion chamber (assuming for now a simple round flat combustion chamber), would be 0.09523 inches.

You folks check my math on these calculations; thanks.

So the clearance would be roughly 0.1", which I think is achievable.

For now, I will use a 1" crankshaft.
The crankshaft scales at 0.73", but the last thing I want to flex is the crankshaft, so we will try a 1" shaft, and see if that will fit.

The rim depth scales at 1.43", but I am going to stretch that to 2", to give some extra mass.

Hub diameter will be 2.25" (scaled is 1.9").

Rim thickness will be 1.68", measured radially.

I will stretch the base down, and leave a 0.5" clearance between the ground and the bottom of the flywheel.
This appears to be how the ELLWE in the first post is designed (a taller base than the sectional sketch).

The crank pin would be about 0.82" diameter.

I guess the crankshaft could be a simple design, with no counterweights, and perhaps the flywheel could be used for balance, with addition or subtraction of weight in a critical spot.

Six spoke flywheel.

.

 

[minh-thanh]

I am still working on improved injector designs.

Hi Roger B !
Just curious....have you checked the air tightness of the injectors?
When I assemble the injector I usually test it with compressed air - about 1 -> 2 bar and dip the nozzle in the oil - if no air comes out it's fine
For me, the injector needs to be sealed because it will ensure the pressure in the pump system....
With the needle , you don't need to make it too small . , based on your design your needle is smaller than mine - and my injector works fine with my 10 cc engine , and i'm really satisfied with it , so there's no need to make it too small.
Just my opinion

 

[GreenTwin]

Here is the 2D drawing I came up with for the flywheel.

1.25" diameter crankshaft

 

[GreenTwin]

For the injector, I will use Find Hansen's design, which is just like a standard IC engine valve, but much smaller.

The flat head of the valve faces the combustion chamber, and has a spring on it to keep it closed.

The pressure of the oil forces the valve open slightly, and the compression inside the cylinder forces the valve to seat and seal completely.

See 0:42




.

 

[Roger B]

The effective swept volume of a two stoke is from when the ports close, typically 75% of the actual stroke. This will reduce your swept volume to ~3 cubic inches and your clearance volume to ~0.15 cubic inches.

The piston area is around 2 square inches so if you allow 0.04 inch piston clearance that is a volume of 0.08 cubic inches leaving 0.07 cubic inches for the combustion chamber. A 0.25 inch diameter 0.25 inch deep chamber has a volume of 0.012 cubic inches so the piston clearance will have to be reduced. I run about 0.02 inches which would work in this case.

My engine is 1 inch bore 1.6 inch stroke. The crankshaft journals are 0.5 inch (12mm). Compression pressure on the two stroke was 450 psi the four stroke is nearer 600 psi. Similar values for your 1.6 inch bore would give a compression load on the crankshaft of 900 lbs.

I have used both Find Hansen style poppet injectors and more conventional needle style. They work, however getting good atomization to avoid unburnt fuel blowing past the piston and out of the exhaust is difficult.

 

[Roger B]

Hi Roger B !
Just curious....have you checked the air tightness of the injectors?
When I assemble the injector I usually test it with compressed air - about 1 -> 2 bar and dip the nozzle in the oil - if no air comes out it's fine
For me, the injector needs to be sealed because it will ensure the pressure in the pump system....
With the needle , you don't need to make it too small . , based on your design your needle is smaller than mine - and my injector works fine with my 10 cc engine , and i'm really satisfied with it , so there's no need to make it too small.
Just my opinion

Hi Minh,

I test the injectors using fuel with a test pump. As you say if they don't leak the engine will usually run. I am trying to improve the atomistion to get better and cleaner running. This is the spray pattern from the latest needle type injector used in the test run above which is part atomised and part jet.

 

[minh-thanh]

Hi Roger B !
There's a difference in my engine's fuel tank and yours
My engine's fuel tank is only one or two centimeters higher than the pump, - I think if it's lower it's fine - because the pump is fine if I leave it on the next day.
But I don't understand why your fuel tank has to be so high, and it could be the problem with the pump, ball valve and not your injectors., or something is not right,

 

[minh-thanh]

Hi Roger B !

or something is not right,

Maybe I'm wrong to say that, I'm sorry .

More....
My 10 cc engine runs fine with a 2.4mm diameter pump cylinder so I don't think you need to worry too much about the pump cylinder diameter, somewhere between 2 -> 2.5mm is fine.
With the IN ball valve, the spring only works to hold the ball in the closed position with a very very light force
I know what I'm saying is a bit different than you calculated but it's fine with my engine