Bellcrank Beta Stirling Engine Project

[tvoght]

After a long time away from the shop, I'm finally ready to start a new build.


I like designing and building Stirling engines, so I've decided to build a beta Stirling. The beta type has a power piston and displacer sharing the same cylinder.


There are a variety of mechanisms that have been used to actuate the displacer
in a beta engine, and for a few reasons, I chose a bellcrank mechanism.


My first inspiration came from seeing this video on Youtube:


(riderericsson channel: "World's finest hot air fan")
<a href="https://www.youtube.com/watch?v=aEuoWWXbGrs" target="_blank">http://www.youtube.com/watch?v=aEuoWWXbGrs</a>


I don't intend to build a fan like in the video, it's the mechanism that's of interest.


I started by "prototyping" a mechanism in CAD (onshape) and animating it. I wanted to get a quick idea if I could make something work. That's in the first clip below. The cylinder has been hidden to observe the actions of the power piston and displacer, and you'll notice that none of the joints have pins in them. Pins weren't necessary to make the simulation work.


After I assured myself with the quick mockup, I started on a buildable design.
The current state of that design is shown in the second clip.


The third clip shows how an animated sectioned view can be helpful for checking interferences.


--Tim


I hope the attached animated gifs work...

[Dave Otto]

Hi Tim

That is an interesting mechanism, looking forward to seeing this one being built.
I do like hot air engines,

Dave

[tvoght]

Hi Dave!  I know you like hot air engines . You make some really beautiful fans.


 My design is not completely locked down, but there are a couple of ideas I have to expedite the build.
 First, the base plate and the standards that support the crankshaft, bellcrank and cylinder are intended to be ordered laser cut from 1/4" steel. It's something I want to try. I'll accept any accusations of laziness.
 Second, I intend to have the cooler 3d printed in aluminum. To justify the outsourcing, the cooler should be difficult or impossible to make by traditional home shop methods. That's why I have drawn it up as a pin-type heat exchanger. I think it would be an effective cooler, and something I couldn't possibly machine. I'm not committed to that particular style though. It's where I am at the moment.


Thanks for looking,
--Tim

[crueby]

That  pin type cooler would have a LOT of surface area, should work well! Does the 3d printed aluminum come out dimensionally accurate to use as is, or will it need further machining (I am assuming some other piston rides against its inner surface?)

[tvoght]

Hi Chris. The cooler will fit over a cast iron cylinder. The dimensional constraint will be the bore size where it slides over the cylinder. I am thinking (always gets me into trouble) that I can have the bore a little small and bore it to size at the mill. Ideally, it would be lapped to the cylinder for best heat transfer. What do you think? It would be my first experience with printed metal. I do know that the design I'm showing does not provide for any clamping for a boring operation, so something must be done about that.
--Tim

[crueby]

Seems like there must be some sort of goop that would be good at heat transfer that could be used to fill any small gaps?

[tvoght]

Yes the goop is used for heatsinks in electronics, often with a mica sheet in between when electrical isolation is needed. Fortunately I wouldn't need a micah insulator here. --Tim

[Dave Otto]

Seems like there must be some sort of goop that would be good at heat transfer that could be used to fill any small gaps?

That is what I was thinking, heat transfer compound used for electronics. When I assemble my Essex engine for the final time, that is what I'm planning to use on the interface of the water jacket and cold end, this is a flat to flat interface. The other option might be to do a shrink fit which would give better contact with the CI Liner.

Dave

[Jasonb]

will the 3D print go down small enough for the diameter of the pins? most have a minimum thickness though they usually flag that up as soon as you upload the file. To hold the print just make up two half round nesting jaws, wood would do.

I've just had a part done in 316L stainless and that bored out to final size OK.

Also on your laser cut standards I would allow a bit of extra metal to square up the bottom foot and bore to take your bearings.

[tvoght]

HI Jason. Pcbway specs a minimum wall thickness of 1mm, so I made the pins .04 inches diameter. Close, but still over 1mm. I did submit a file to see what the cost would be and it was accepted by the automated system. It still could be rejected by a human if I placed an order...
Thanks much for the advice on the feet of the standards.


My devious plan to submit a near-finished design for review is working! <rubbing hands gleefully>.
--Tim

[crueby]

Be ready for the price to go up after the human review, sometimes its just a bit, sometimes it double. They don't  proceed till you approve, so no risk on that.

[Jasonb]

You could also try Craftcloud, they have more options for metal printing SLM and DMLS and higher quality options though price also gets higher.

[tvoght]

Jason, thanks for the Craftcloud reference. I was not aware of them.
 With the printed part in mind, I have been refining the cooler a bit. I made the pins tapered and added fillets. This should be a better part with less tendency for pins to break.
I've designed special "outsource" versions of all the parts I want to have laser cut or 3d printed. They have extra material where I want to machine them to size. I show a pretty picture of the new cooler.


I am getting close to pulling the trigger on this design and start ordering parts. I think I already have some of the materials for starting on the cylinder, displacer, and hotcap, so getting close.


--Tim

[crueby]

To help justify the printed cooler, add up the number of pins you'd have to turn/taper, and the number of holes needed to be spot drilled, drilled, reamed, loctited if making it all from scratch! ...   

[tvoght]

To help justify the printed cooler, add up the number of pins you'd have to turn/taper, and the number of holes needed to be spot drilled, drilled, reamed, loctited if making it all from scratch! ...   

Heehee. I was just thinking of what it would take to draw this thing up without CAD...
--Tim