Crank operated lever controls rotary valve and fries brain!

[PaulR]

Someone please point me in the right direction with this, it's making my head hurt!

This render shows a version of the 1950 Popular Science magazine 'scraps and solder' rotary valve engine, the valve being operated by the side crank. The original plan has the inlet and exhaust ports (ie holes through the valve shaft) 115 degrees apart whereas the plan for this one shows 105 degrees. I'm trying to work out how things would change if I modified the dimensions further (eg cylinder/stroke length, valve housing size etc) but I can't seem to figure it out.

As I see it, the valve rod's crank pin is doing what an eccentric would but, because the pin connecting the valve rod to the lever isn't constrained horizontally, it seems to be introducing an order of complexity that my brain can't handle! I suppose the empirical way would be make some temporary parts and spot the points on the valve shaft through the cylinder or in/out pipes but I'd rather try to work it out.

[crueby]

Interesting  design!




It does work just like an eccentric would if the center of the eccentric was in the same place as the valve pin. It does look a whole lot different though! Its a common arrangement on locomotives,  with the valve pin on the arm off the crank pin.


To play with offset lengths and positions, you could design it as an eccentric and then swap it for a pin and arm later, if that keeps the brain temperatures  down! Some old hand cranked tools put a spiral arm out to the handle just for looks, but they behaved the same as if the arm was straight.

[Charles Lamont]

The return crank acts, as you say, exactly as an eccentric would. At the valve end, the rocking of the lever means that end of the linkage has some up and down movement too, but not enough to have a great deal of effect on the behaviour of the valve. We would need to see more of the design of the rotary valve before commenting, and a whole lot of dimensions before advising on any calculations.

[Jasonb]

Easiest thing is to draw it in cad, section and then you can watch what is happening as you make any changes.

I have done a couple of rocking valve engines but used a pocket in the valve rather than just holes but both were worked out in CAD.

As said the arrangement at the crankshaft end is no different to an eccentric.

Vertical movement at the rod end has little effect unless you make the lever very short

You need to watch any images taken from the likes of Grabcad, it is not unknown for errors to creap in

[Jasonb]

Original on pages 196 & 197

https://books.google.co.uk/books?id=yywDAAAAMBAJ&printsec=frontcover&rview=1&source=gbs_ge_summary_r&redir_esc=y#v=onepage&q&f=false

[PaulR]

Thanks for the input. Here's the thread on HMEM where this particular example came from, the topic includes the original plans and a couple of builds both of which use an eccentric (which I'd prefer not to do) - hope it's ok to post that here?

https://www.homemodelenginemachinist.com/threads/scraps-and-solder-become-an-engine.26180/#post-408806


Looking at it again, I think valve pin (b in my diagram) is describing a circle based on its distance from the crank pin a (ie with radius a-b). Seems I just need to work out the angle subtended by the arc that link pin c travels which is in turn related to the same radius and the length of the connection between c and the centre of the valve.

Or maybe not! 

[PaulR]

Easiest thing is to draw it in cad, section and then you can watch what is happening as you make any changes.

I have done a couple of rocking valve engines but used a pocket in the valve rather than just holes but both were worked out in CAD.

As said the arrangement at the crankshaft end is no different to an eccentric.

Vertical movement at the rod end has little effect unless you make the lever very short

You need to watch any images taken from the likes of Grabcad, it is not unknown for errors to creap in

Thanks for this - nice models! I'm just relearning some 2D CAD and had a fiddle with Tinkercad to make some quick mock-ups, I'll look into 3D once my brain cools a bit!

[PaulR]

I made this cardboard thingy to no particular dimensions just to see what happens to the link between the crank-pin and the valve-pin and what that does to the valve link/lever at the right-hand end. I think the positions shown are correct but I suppose it could behave differently in practice!

Whether or not its it's pure coincidence or I suppose more likely related to the different radii upon which the pins are set on the crank disk, it appears to move the lever a little over 90 degrees, so not unlike the two lots of plans (105 and 115 degrees).

PS Hopefully the GIF will animate properly in your browser.

[crueby]

Gif works fine here!


If the arm is slightly longer or the valve crank closer to the crankshaft, the action would be centered better, and it wouldnt risk jamming when the crankshaft end is all the way out.


Gotta love Cardboard Aided Design!

[PaulR]

If the arm is slightly longer or the valve crank closer to the crankshaft, the action would be centered better, and it wouldnt risk jamming when the crankshaft end is all the way out.

Thanks for the help... I'll make some adjustment, see what happens.
Cardboard Aided Design, I like it! 

[PaulR]

The recycled wood version made it a lot easier to work out what does and doesn't affect the valve lever's arc of travel and what that'll do to the port positions 

[PaulR]

Well I was completely wrong about the path taken by the valve's crank pin - it's more like a pointy sausage (!!) than a circle as this image shows (the piston is replaced by a slider which is functionally the same as far as the mechanism is concerned). The attached short video shows the motion from a 3D perspective. Now I can fiddle with the components to get satisfactory proportions for the real world

[Jasonb]

The others that I have seen like it all move in a circle so would have expected yours to do the same.

The pivot at the crank end of the rod is at a set radius from the axis of thhecrankshaft so should remain like that for the full rotation of the crankshaft and therefore describe a circle.

[crueby]

Um, thats not right. The arm off the crank pin should be fixed to the crank pin, not revolving opposite it.

[Jasonb]

Looks like you are allowing one of the joints to ratate, they should all be one rigid assembly.

<a href="https://www.youtube.com/watch?v=2FfDrhXgnoY" target="_blank">http://www.youtube.com/watch?v=2FfDrhXgnoY</a>