Rotary Valve engine having more than two cylinders

[oliver p.Participant @oliverp-84158]

Hi Guys,

I am new to the forum and this is my first question here

So I have already built several Stirling motor models, and I thought it would be time to make something "bigger". I have no particular application for it, just for fun. What I would like to do is a steam engine model having 3 or 4 cylinders. While I was searching on the web for some inspiration, I found this plan by accident:

http://www.john-tom.com/MyPlans/Steam%20Engines/RotaryValveEngine.pdf

It looks very nice, but has two cylinders only. But I like the "Rotary Valve" it has, so I wonder whether it would be possible to extend this design to 3 or even 4 cylinders. I already started to make some simple 3D drawings to figure out how I should design the rotary valve, but so far I was not yet able to figure it out. So can anybody give some hints about this? is it possible?

For 2 cylinders, the notches in the rotary valve need to be 90 degrees apart from each other; so I assumed for 3 cylinders, 60 degrees would be appropriate, but this is not true (or I made a mistake ). I made a very simple 3D CAD drawing with all cylinders, pistons and everything, which allows me to simulate the whole thing and make an animation to figure out how the rotary valve needs to be designed. But so far I was unable to do it correctly any hints? I like the idea that there is no need for a separate valve for each cylinder but only one valve for all, and it looks a bit fancier, which I like as well.

[oliver p.Participant @oliverp-84158]

[michael howarth 1Participant @michaelhowarth1]

Hallo Oliver and welcome. I have made the two cylinder version and a nice little motor it is, running smoothly and economically from a 2.75 diameter vertical boiler. It suffers from not being a "self starter" if the crank ends up in the "wrong" place when stopped. I have found a way around that though. If you can figure out the valve events for a 3 cylinder version it would be of great interest.

Mick

[John BaronParticipant @johnbaron31275]

Hi Guys,

I've seen an 18 cylinder rotary valve engine ! It was used in aircraft some years ago. Very reliable I understand and well liked by pilots that flew aircraft with that engine. I might have a picture of it somewhere.

[duncan webster 1Participant @duncanwebster1]

To be pedantic, that's a semi rotary valve. To have 4 cylinders at 90 degrees you'd need 2 valves. The fit on the rotary valve needs to be pretty good.

Edgar Westbury did several steam engines with rotary valves, one being a radial, but I don't have any leads to drawings I fear

[oliver p.Participant @oliverp-84158]

@ Mick

cool! have you used the very plans I posted in the link? and how did you overcome the starting problems?

I think if there are 3 or more cylinders, the engine will self-start no matter in which position it was stopped previously.

@ duncan

you are right. The valve does not really rotate in this case but oscillates. I think the construction could be a bit simpler if there is a separate valve for inlet and exhaust.

If I could post a screenshot of my 3D CAD model, we could probably figure out how the valve needs to be done, but so far I didn't find out how to attach a picture to a posting here

So do you think it should be possible to have this design with 3 or more cylinders?

[michael howarth 1Participant @michaelhowarth1]

Yes, Oliver it is the same engine but I did make a few alterations / additions. Firstly I ran the crankshaft and crankshaft valve linkage in miniature bearings (arguments for and against). Secondly, I fitted Viton piston rings. There is a feed pump mounted on the RH side of the engine driven by an eccentric on the valve and a Roscoe lubricator fitted opposite the steam intake. I may also have tinkered with piston diameters but I can't remember the details.

Duncan is quite correct regarding the valve being a good fit. I lapped it in and have surprisingly minimal leakage, mainly at the feed pump end. I am working on an O ring based solution to this.

As regards the self starting problem, I have fitted a needle roller clutch (one way) bearing to the output end of the crankshaft which just requires a nudge from a servo to push the piston past TDC and away she goes.

The engine is shamefully filthy at the moment but if you would like a picture I will clean it up.

Mick

[oliver p.Participant @oliverp-84158]

Posted by mick H on 05/08/2019 09:29:53:

Yes, Oliver it is the same engine but I did make a few alterations / additions. Firstly I ran the crankshaft and crankshaft valve linkage in miniature bearings (arguments for and against). Secondly, I fitted Viton piston rings. There is a feed pump mounted on the RH side of the engine driven by an eccentric on the valve and a Roscoe lubricator fitted opposite the steam intake. I may also have tinkered with piston diameters but I can't remember the details.

Duncan is quite correct regarding the valve being a good fit. I lapped it in and have surprisingly minimal leakage, mainly at the feed pump end. I am working on an O ring based solution to this.

As regards the self starting problem, I have fitted a needle roller clutch (one way) bearing to the output end of the crankshaft which just requires a nudge from a servo to push the piston past TDC and away she goes.

The engine is shamefully filthy at the moment but if you would like a picture I will clean it up.

Mick

sounds very interesting! if you have some nice pictures that would be really interesting

[Author]

Posts