36x60x54 Twin Tandem Mill Engine

I have seen a tree of parts cast using the lost-PLA process, with investment casting, and with gray iron, and those parts do not have hard spots, even though they are thin (I have two of them, and have personally drilled them easily).
See video below.
So I know this method is viable if the right materials and processes are used.
This method used standard PLA filament, and not any special filament.
The holes have a taper to them intentionally, and they came out perfect, (it is a bit of a trick to have perfect cast holes).

I agree with JasonB that the bound sand approach could be tricky and/or expensive at this model engine scale.

It should be noted that the original poster is using a simplified 3D model, which may make it viable in a sand mold scenario.

The simplified design will be fine with bound sand, most commercial foundries here tend to use one of the airsetting binders for just about everything. I would think it is the same in the US. But whatever sand you use it still needs loose pieces to be able to pull it from the sand.

Hi All,
I spent the past few days drawing the HP cylinder assemblies in a simplified version from the plans, ie fasteners, material thickness, and layout. As shown before, Initially I was designing with the intention to cast, but it got pretty cumbersome and i went deep down a rabbit hole. In order to get the mechanics, I decided to fully draw this in a simplified 1/24 version that I can later go back and draw the castings. I was just making too many assumptions and design changes without knowing how the engine is going to work. I need to know the geometry is going to work on this scaled down model.

So I started piecing this engine together. I am starting to draw the LP cylinder now. Once I get that in, I can fully figure out how this was piped.

What I believe right now, Steam enters in the middle top of the HP valve body and are directed to the piston via the spool valve. The rectangular flanges on the bottom, which are connected via a rectangular duct not shown yet, connect the rod and cap side exhausts and are all ported to exit the top back round flange on the HP valve body, leading to a condenser and LP cylinder.

Mike

Looks great so far !
I think your assumptions are good so far about the steam/exhaust flow.

I am not sure I understand the flange that points upwards on the crosshead end of the valve chamber.

.

As I read it the steam enters via the central smaller round flange and the passages take it to each valve.

The exhaust from the two valves are linked together by the rectangular cross section duct below the valve chamber, that just has a connection to drain condensate.

The larger of the two flanges is the combined exhaust steam that is then piped to the inlet of the LP cylinder, hence why it is closest to the middle trunk guide to reduce the length of pipe run

The larger volume of the two "D" shaped chambers at the ends and the rectangular linking duct act as a "collector" for the steam coming out of the HP before it enters the LP

I think called a "receiver".
Some compounds have them, and some apparently do not.
I think if the phasing is correct, you don't need a receiver (perhaps on a tandem ?).

.

GreenTwin said:

I think called a "receiver".
Some compounds have them, and some apparently do not.
I think if the phasing is correct, you don't need a receiver (perhaps on a tandem ?).

.

Click to expand...

What is the use of it? An “accumulator” for steam if the LP cylinder spool valve isn’t open right when it’s discharged from the HP cylinder?

Triple expansions don’t have that do they?

Mike

If the high pressure cylinder is exhausting before the low pressure piston gets near top dead center, then the steam has to be held in a reservoir (receiver), which is just a chamber to hold the steam, until the low pressure piston gets to the right position and its valve begins to admit steam.

As I recall (not very well; check me), for a tandem, since both pistons are attached to the same piston rod, then they maintain the same relative position in the cylinder as they move, so the high pressure cylinder will always exhaust at the right time to the low pressure cylinder, and a receiver is not required. That is what I recall anyway.

For steam engines other than the tandem configuration, you may or may not need a receiver.

For a twin tandem, I guess it depends on whether you are discharging from side-to-side, or discharging on along the same piston rod. If you are discharging the high pressure steam from one side to the other, then the relative position of the crank throws becomes important.

Some steam engines had crank throws offset (such as 90 degrees) to allow starting in any position, while others such as twin engines that were used to power electrical generators often had cranks at 180 degrees so they provided smoother power to the generator, and since they did not have to reverse.

I guess theoretically you could discharge from HP1 to LP1 on the same piston rod, and then cascade across to the second HP2/LP2 on the second piston rod, but I would be a bit surprised at that arrangement.

The Titanic engines were triple-expansion, 4-cylinder, and there were two of the lowest pressure cylinders because one would have had an excessive diameter.

.

For a three cylinder compound steam engine, the idea is to create three equal forces equally spaced 120 degrees around the crankshaft as the engine rotates.
The idea is Pressure1 x Piston area 1 = Pressure 2 x piston area 2 = pressure 3 x piston area 3

Sort of a like a 3-phase motor, where a constant torque is created as the motor rotates.

Unequal forces on the crankshaft would creating twisting moments which reverse in direction, which could eventually break it.

.

GreenTwin said:

For a three cylinder compound steam engine, the idea is to create three equal forces equally spaced around the crankshaft as the engine rotates.
The idea is Pressure1 x Piston area 1 = Pressure 2 x piston area 2 = pressure 3 x piston area 3

Sort of a like a 3-phase motor, where a constant torque is created as the motor rotates.

Unequal forces on the crankshaft would creating twisting moments which reverse in direction, which could eventually break it.

.

Click to expand...

Thanks for the above engineering theory... it really got me thinking about the thermo theory behind these engines. There is a ton to learn! You are right, this engine has the HP and LP cylinders sharing a common rod.

The piston areas are constant, and we know pressure decreases on each piston due to lost energy from the work on the previous piston. I never thought much of the theory behind calculating the pressure reductions between the pistons or how the engineers would have came up with these piston sizes in the first place.

So the piston areas and strokes of 36x60x54 is the perfect combination to yield a balanced force distribution between the HP and LP pistons. I may have to calculate out the forces on the full size and scale down engine just to do it for fun. I am also curious if building the engine in 1/24 scale will still yield a "balanced" engine. I know this engine ran on 100psi saturated steam, so i wonder too, if i change the incoming steam pressure, will it change the balanced forces on the HP and LP cylinders? I guess i will have to dive into the steam temp/pressure tables and conservation of thermo energy!

incoming pressure is the only speed control, so i would think it shouldn't effect the balance, but I will run through the math eventually to see.

Thanks,
Mike

While I don't think the power produced scales down linearly, the forces do scale (perhaps not linearly).

It is still X psi times the high pressure piston area, and Y psi times the low pressure piston area, at least for pistons on a common piston rod.

I have seen a design for a smallish launch compound twin, and the same forumlas were used as for a ship engine.

I don't quite understand your engine enough yet to get very deep into the details.

.

Incoming pressure does not control the speed, it determines the work the engine will do. The more pounds applied to the area of the piston the more power the engine will give.

The regulator (throttle) controls the volume of steam that gets to the engine, you don't see the boiler's pressure being raised or lowered when an engines speed is being changed. The pressure seen in the valve chest will be almost identical to what is shown on the boiler. Typical regulator on a loco or traction engine just uncovers a hole so more steam can flow. Similar with governors on stationary engines they actuate a valve that opens or closes the passage so more volume can flow. Pressure stays the same

When I air run my engines the compressor is set to around 5psi and that feeds my manifold with air regulators mounted to it. I can run an engine from ticking over at maybe 40rpm upto a couple of thousand rpm all by adjusting the flow regulator while pressure stays the same.

Going back a bit and digressing slightly, I cannot see how the longitudinal webs contributed anything structurally useful to the prototype cylinder.

For a large engine, the webs would keep the cylinder walls from deflecting under load/pressure, and would allow a much lighter casting than using a thick wall.

For a model engine, I think it is more a matter of if you want to keep the original look/feel, with the understanding that I think the lagging will cover up most of the cylinder, or want to dispense with some things in order to create a faster design/engine.

.

Maybe the webs around and along the boiler were simply crinolines to support the cleading and not intended to be structural

GreenTwin said:

For a large engine, the webs would keep the cylinder walls from deflecting under load/pressure, and would allow a much lighter casting than using a thick wall.

Click to expand...

Well actually they don't, which is the point I am making. The largest stress on the cylinder under pressure is 'hoop' stress, a tangential tension in the cylinder walls: Pressure x Bore / 2 x Wall Thickness. The ring web helps with that, but the longitudinal ones do nothing. The axial stress is half the hoop stress.

Charles Lamont said:

Well actually they don't, which is the point I am making. The largest stress on the cylinder under pressure is 'hoop' stress, a tangential tension in the cylinder walls: Pressure x Bore / 2 x Wall Thickness. The ring web helps with that, but the longitudinal ones do nothing. The axial stress is half the hoop stress.

Click to expand...

That may be true about what causes the maximum stress, but webs would add a lot of rigidity along the long axis, to prevent buckling/twisting, the same way that flanges on an I-beam work.

I think you are ignoring the fact that the cylinder may support the crosshead, in which case there is more forces acting on the cylinder than just those created by internal cylinder pressure.
If the crosshead is cantilevered, then it would create a large force/moment perpendicular to the centerline of the cylinder.

If the crosshead and cylinder are supported from the bed frame, then the remaining forces may be radial.
If the wall thickness is thin enough, then the longitudinal webs would prevent the cylinder wall from deflecting, as would the ring webs.

.

GT, I don't know what buckling or twisting forces you think might exist in a steam engine cylinder, but in any case, the tubular cylinder wall will be at least an order of magnitude better than the longitudinal webs at resisting any supposed axial torsion. And while an I-beam is an efficient shape in bending, it is more or less useless in torsion. I am sticking to what I said in the first place, but as this is a digression anyway, I probably won't want to say much more about it.

I guess what I am saying is that webs are used in many places, such as cylinder heads, pistons, etc., so that the part can be cast lighter with the same strength, and perhaps reduced wall thickness.

The Stanley 20hp auto steam engines were notorious for twisting due to excessive torsion, often causing fatal engine failure.
Steam engines with their high torque can produce a huge amount of torsion.
One Stanley owner I chatted with mentioned that the folks he knew welded plates on the sides of their Stanly frames.

A steam engine can produce maximum torque at zero rpm.

As I mentioned above, if the cylinder and crosshead are bolted to a baseplate, then there is no torsion in the cylinder (assuming that the baseplate was rigid enough not to deflect under full load).
But I still say the cylinder wall could be cast significantly thinner if it had longitudinal webs.

Good discussion for sure.
All very good points.
.

Heads (end covers) and pistons are flat so need the bracing webs much like the end plates of a boiler or sides of a firebox need stays to stop them bulging out but the often thinner barrel needs no further support.