Oscillator Cylinder Pattern

[Jasonb]

You could glue the edges of two pieces of wood together with a sheet of paper sandwiched in the middle. Then cut with your holesaws and provided you lined the paper joints up as you assembled the discs the pattern would then split apart at the paper joint. For the thin discs I would just put a clamp on them while sawing as you don't have such a large glue area and the hole saws are not the most delicate cutters.

I use this method to glue blanks together before turning or machining and also to hold blanks to sacrificial boards so I have something to hold.

Before splitting apart drill right through for dowel holes then they are sure to line up. Just needs a chisel to start the splt and they all come apart easily.

[vtsteam]

Thanks Jason. Good idea.

I have another question re oscillators. For one of the size I'm making, how do you recommend doing the valve face trunnion -- flush with a conical counterbore, or with a boss and a conical counterbore, or machine a journal that fits in a bushing held by the standard? Right now I have enough meat on the valve face that I could go a fair amount deeper, depending on the style of trunnion mount.

I'm hoping to actually get some work out of this engine with steam.

[Jasonb]

On that size I  would machine that spigot/gate that you have at the same lathe setup to turn the port face.  Then when the support is machines form the matching hole for the spigot.

That way ensures that all is true. If you go for a simple threaded hole and separate pivot rod or even try to use tapers/cones there is more risk of the pivot axis not being true to the port face(s)

[vtsteam]

Thanks Jason 

I've machined all, so far, between centers, in one step, so the port face, the existing machined boss on that face (is that called a spigot? -- I've always had trouble understanding that term), and the 60 degree counterbore are all aligned the way you say.

What I don't know is how that part bears on whatever journal or standard (terms again) that affixes to the base. Is the bearing the cone shape -- like a pivot? Or is the bearing on the outside of the spigot in a conventional bearing journal? Or is it both? Plus the real thrust bearing surface must seem to be the port face itself, in order to be steam tight, right?   

So if the bearing is the cone pivot type, that must mean it is just very slightly short of full engagement, so the port face will be tight? 

[CI]

The Cretors oscillators used a spring loaded cylinder, to hold the cylinder against the port face.
I have an oscillator that uses just a bolt with jam nuts, and the problem with that arrangement is that if you operate the engine on steam, the metal expands, and the joint is too tight.
And without a spring, the port faces tend to wear, and so the bolt/nut need to be adjusted too much.
A spring will automatically adjust the joint for wear.

The spring-loaded port face takes care of a lot of problems.
Oscillators are sensitive to the tension on the spring.
Too tight and the port face drags excessively.
Too loose and there is excessive leakage.
A spring seems to let the cylinder float over a thin layer of steam/air at the port face, and thus there is not much if any friction.

Oscillators will run very well if they are set up correctly.

.

[Jasonb]

A lot will depend on how you decide to support the cylinder.

If you go with two supports that is a more rigid setup and there is less risk of the cylinder port tilting away from the port face particularly if the engine is run under load. For that setup the port face just needs a simple parallel spigot/boss and a hole in the support to suit it as per the first image. The other side of the cylinder is supported by a screw in pointed support that engages with the ctr drilled hole in the spigot on that side of the cylinder. This suits a horizontal engine more than a vertical.

The other option is to use a single support which is the port face, again the plain parallel spigot/bos scan be used but drill and tap it for a rod. This rod can then have a washer and spring slipped overit and an adjusting nut use dto pull the cylinder against the port face. As shown in teh second image. Probably less rigid but it will allow the cylinder to lift off the port face and let any condensate out rather than hydraulic lock teh cylinder. This layout is more common on vertical layout sparticularly if the cylinder is at the top as you don't want a long tall support on teh non port face side.

[vtsteam]

Thanks CI, I do know about the spring. My real question was about the nature of the trunnion bearing -- and Jason, thanks, that answers it

re. a cone bearing, I noticed that in an oscillator design by Julius (Muncatser model), the cone is 90 degrees, not 60, so I'm guessing that is preferable?

Thanks, both of you again!

[CI]

For the double cylinder support, there needs to be a bit of expansion clearance between the supports.
The outer cylinder support needs to allow the cylinder to move towards ther port face, which perhaps it does by default.
Just mentioning a few oscillator things, in case someone following is not aware.

Edit:
I am throwing out some general information about oscillators that I am aware of, in case someone comes along 5 years from now, gets inspired by this thread, and decides to build an oscillator, but is not necessarily completely familiar with their operation.
.

[Jasonb]

I don't think the angle of the screw/hole will make a lot of difference. On the couple of engines that I have made that use that type of support I have just used a 60deg centre drill, the pilot hole they form holds a bit of oil.

Muncaster does not give an angle and the sketch in his original book is lot large/clear enough to see the angle.

If you were worried about cylinder expansion making the adjusting screw go tight it would not be hard to make it a larger diameter and ream a hole in the end. You could then fit a small compression spring into the hole followed by a piece of drill rod with a suitable point. The screw on the outer part would still allow adjustment but the spring would allow for any expansion. With your size engine a 1/4" UNF thread and 1/8" or even 5/32" drill rod would work.

[vtsteam]

Great, thanks Jason.  I'll go with 60 degrees as more convenient with the tooling I have, and I was thinking that the conical pivot could be spring loaded to fit with CI's experience, as you mention. But I'll also have to think about it a little more first.

And I do imagine this as a horizontal engine.

[vtsteam]

Maybe I'm overthinking this but if say the working pressure was 50 lb/sq in (3 bar) and the valve face was 2" (5 cm) in diameter, that's about 3 sq. in (19+ sq cm) or 150 lbs (670 Newtons) of force trying to separate them.

So a spring would have to exert even more than that to keep the faces together?

[CI]

Some valve faces are open-port, and some have cast passages, where only the ports open through the valve face.
Here is an open-port design, and it works fine.
I would guess a closed design would leak a bit less.

At any rate, the force acting on the cylinder/port face would only be generated by the area of the steam ports/passages, which is a small percentage of the entire port face.
The exhaust passage/port will not generate any lifting force.

.

[CI]

This engine has cast passages, so only the ports generate pressure/force.
I see springs on either side of this engine.
I have never understood every aspect of this engine, especially that reversing vavle, but in general I understand how it functions.
I can't see where a steam and exhaust pipe would connect.
I guess this is a compound.

.

[Jasonb]

What I had to say about spring tension in my recent build thread. Basically the spring needs to apply a pressure just greater than that trying to push the port faces apart. Slow low speed display running will need a weaker spring than an engine working under load at 50psi which is actually just over 3bar not 7bar

https://www.model-engineer.co.uk/forums/topic/v-twin-kelsey/page/3/#post-828255

[Jasonb]

Pat, there are springs on both sides of that engine as it has two cylinders

The reverser is just a 4-way valve. Being a compound complicates things a bit so would need further study. Probably fed to the top and bottom of the valve as that is easier with smaller pipes or passages and HP steam then one larger pipe as exhaust from the middle of the valve.

Would make quite a nice model at about 1/3rd scale. though I expect 1:1 would be more to your liking.