Single cylinder horizontal enigne as test bed for multi cilinder engine design

[xander janssen]

Hi all,

After building several stirling engines and one Jan Ridders engine, my ultimate dream is to design and build a multi (3 or 4) cylinder engine like the old Holt75, Fairbanks Morse etc engines.

I would like to see how low the RPM's at which it is still running reliably can get, preferably below 500 RPM. In order to investigate the feasiblilty, I start with a single cylinder engine on which I can easily test various key numbers e.g. bore, stroke, head and valve design, flywheel inertia, ignition etc.

I already got some books (incl Doug Kelley) and drawings on a lot of other engines on which I base my key numbers to start with e.g. 25 mm bore, 25 mm stroke and flywheel inertia.

Once this single cylinder is running, the various parameters will be changed to see how slow I can get it running. Once that is determined, the idea is to build a "modular" vertical engine, starting with a single cylinder with the same parameters and then extend that to a 2, 3 and finally maybe even a 4 cylinder.

 

[xander janssen]

Nice seal between piston and cylinder., needed to run the engine without rings.

Only have to part it off.

 

[xander janssen]

Piston is now parted off from the stock. When I place the cylinder vertical on a piece of rubber and place the piston in the cylinder, the piston remains elevated on the air trapped below the piston. Once the cylinder is lifted/tilted from the rubber, the air can escape and the piston falls through the cylinder under its own weight.

Next will be the conrod

 

[minh-thanh]

"" Single cylinder horizontal enigne as test bed for multi cilinder engine design ""

 

[Vietti]

I too have pursued making a hit and miss run slow and coast long. My current thoughts are:

1. Use 0 ring not CI rings

2. Put needle bearings on the crank mains and rod. Since you make a built up crank this should be easy.

3. Leave piston long above the wrist pin so CR can be lowered in steps. Actually I tried it and a higher CR did not work as well as I thought it might. I thought if it hits harder it should coast longer, maybe not true. 4:1 seems about right. I tried over 6:1 and gradually decreased it. I think the extra energy generated on firing is compensated by the extra energy used on the compression stroke, don't know.

4. Heavy and or large diameter flywheels help, though if you overdo it, the engine has to hit more than once to come up to governed speed.

I'll be curious to see how your tightly fitted piston with no rings works out. Smaller bore engines do it all the time though. I just finished a R&V that runs under 500 rpm and will coast for 6 seconds.

I put up a YouTube video long ago of a Red Wing with an oscillating mag that reliably coasts 4 seconds with CI rings.

Hope others will jump in here with ideas of making slow running, long coasting hit and miss engines.

 

[xander janssen]

I too have pursued making a hit and miss run slow and coast long. My current thoughts are:

Thanks for your information. I will change the CR by reducing the height of the head in steps since the piston is already finished and never thought of changing CR by removing material from the piston

I do not want to build this engine as a hit-and-miss, since I want to see how low RPM I can go in continous running. This because once the parameters for slow running are determined I want to extend this to a slow running multi cylinder which will never run in hit-and-miss.

If I can get it running as slow as this beautiful engine of Find Hansen, I would be overly excited

 

[minh-thanh]

If I can get it running as slow as this beautiful engine of Find Hansen, I would be overly excited

You can !
Step by step ,.. you will get there.

 

[Steamchick]

...reminded me of a 1950 BSA B31 I owned... 7:1 compression, but worn so was more like 6:1... on tick-over I could manually retard the ignition to just after TDC... and it ticked-over slower than that with its massive flywheels.
K2

 

[Vietti]

When you get to the cams consider very little or no overlap if all you want is a slow running engine.

 

[xander janssen]

Piston insert finished

 

[xander janssen]

Conrod bearings, main bearings, crankshaft, piston/cylinder are all sufficiently square and parallel for a smooth motion without binding. Thank god

single cylinder engine: piston, conrod, crankshaft and main bearings all in line and square.

 

[xander janssen]

Having an engine to run that slow, that must be really rewarding

A nice slow running single cylinder, check from 2:00 onwards

 

[kf2qd]

Slow running will require a heavy flywheel.

 

[xander janssen]

Cooling sleeve with a tight slide-fit on the cylinder lining. Hence the short tapered lead-in to ease installation and scratches due to the fitting test.

 

[xander janssen]

Drawback of not designing upfront and making drawings is that I have to cope with several non-optimal choices in the beginning that need a work around along the route.

Some frame bolts were in the way and cannot be removed without compromising the squareness of the frame. Hence the cooling fins needed an ugly cutout, not visible in this picture



So far I learned a lot to be usefull when I start designing the multi-cylinder engine.

 

[xander janssen]

First time stainless steel (AISI 303) in a precision part.

The stem is diameter 3 with +0 -0.01 mm tolerance.

 

[GreenTwin]

Here is a Fairbanks at a local engine show.

I have another photo of my dad running a big 3-cylinder Fairbanks that he helped restore.

I have to find the photo of the 3-cylinder.

.

 

[GreenTwin]

My dad running the 3-cylinder.
It was a very smooth engine.
I can't remember the maximum rpm, but it ran pretty slow at idle, but did not run cleanly at idle, since it was basically designed to operate continuously under full load (it powered a generator for a small city, of of several).

.

 

[xander janssen]

My dad running the 3-cylinder.
It was a very smooth engine.
I can't remember the maximum rpm, but it ran pretty slow at idle, but did not run cleanly at idle, since it was basically designed to operate continuously under full load (it powered a generator for a small city, of of several).

.
View attachment 138317

This exactly the type/appearance of the multi cylinder I would like to build.

 

[xander janssen]

Inlet valve cage assembly ready.

Need to do some more deburring