Rotary Valve Engine

[PaulR]

Following on from this thread https://www.modelenginemaker.com/index.php/topic,12120.0.html, I've drawn up fledgling plans for my next nutty project. This will be a single acting engine with a rotary valve controlling admission at the back and once again made from bar stock with just a lathe (so kept as simple as possible). The original 'scraps and solder' engine uses a side link set 90 degrees off the crank pin to control the valve but I figured this could be done using a second crank disk (or crank arm) and a straight side-link like a pair of coupled loco wheels, so long as the flats on the valve shaft are positioned perpendicularly to the valve crank pin (not sure I've explained that well but the 'plan' might help!). Air would exhaust via a hole on one of the flats on the valve and through a longitudinal hole.

Two things are bothering me slightly about this. Firstly, would the mechanism tend to lock up or would the flywheel inertia get it past the difficult bits? Secondly would it work ok without a crosshead guide? I've seen a few very small engines with this sort of arrangement and no guide which seem to work fine.

Comments and suggestions would be very welcome before I start creating scrap 

PS Sorry about the unconventional and very rough CAD, I haven't got time to learn to do it properly at the minute!

[crueby]

A flywheel ought to get it past the dead spots till the next power stroke.   

[redhouseluv]

I must've missed the previous thread on this engine design, just read through it. I don't know if the mechanism will work, but it certainly looks like a interesting one.

I saw a mentioned making it with an eccentric first then substituting with the rotary valve once working, is that what you intend to do or are you going straight in.

N.B. your CAD is still considerably better than mine! 

[PaulR]

I saw a mentioned making it with an eccentric first then substituting with the rotary valve once working, is that what you intend to do or are you going straight in.

I'm going for the rotary valve straight off. I've made a couple of engines with that type of valve in the past and they mostly worked ok. I was thinking about making this a two cylinder jobbie (probably straightforward) or a double acting single but that made my head hurt so I'll stick with the simplest arrangement first.

[PaulR]

A flywheel ought to get it past the dead spots till the next power stroke.   

Thanks Chris. I was also thinking an o-ring on the piston might help given that it's single acting, I need to read up on how to use them and whether the drag will offset any benefit.

[crueby]

A flywheel ought to get it past the dead spots till the next power stroke.   

Thanks Chris. I was also thinking an o-ring on the piston might help given that it's single acting, I need to read up on how to use them and whether the drag will offset any benefit.

I use viton rings on all mine, a bit of trial and error to get the depth of the groove just right, but they seal really well with vety little drag, given a drop of oil. Would be worth the experiment. For a single cylinder single ac5ing, you might need a bit more flywheel.

[Jasonb]

If you can Simulate the valve action correctly in your CAD then it should be OK, I've only used that type of valve with eccentrics.

Regarding the cross head, why not do away with that piston rod and clevis and run the conrod straight to the piston with a wrist pin much like an IC engine. Having a longer piston will also help keep things lined up, it does not matter if the piston comes out beyond the end of the cylinder.

On such a small bore I would just run the piston without rings and cut a couple of oil grooves. The drag from a ring is not going to help get you over centre.

[PaulR]

I use viton rings on all mine, a bit of trial and error to get the depth of the groove just right, but they seal really well with vety little drag, given a drop of oil. Would be worth the experiment. For a single cylinder single ac5ing, you might need a bit more flywheel.

[PaulR]

Regarding the cross head, why not do away with that piston rod and clevis and run the conrod straight to the piston with a wrist pin much like an IC engine. Having a longer piston will also help keep things lined up, it does not matter if the piston comes out beyond the end of the cylinder.

Thanks Jason. That was the original plan but it has a knock on effect - either the crank has to be further away, the bore/cylinder bigger or the conrod smaller. If it was a 'bucket' type piston I'd probably have to make it longer anyway. I'll have a ponder while I'm denied workshop time due to decorating 

[Zephyrin]

the choice of the o'ring is easily made with a complete set of 1/10 mm drills or equivalent, (which you probably have, only up to 10mm in my case): by placing the ring on the drill shank and and by inserting it in the bore, the finding the optimum depth of the groove to be machined is easy.

[Jasonb]

Cylinder can stay the same, I have assumed 10mm bore as the 8mm top left is smaller than the view from the other end.

Position of crank relative to cylinder can stay the same.

Longer piston which is what does the guiding. As you have done away with the piston rod the conrod needs to be longer. The longer conrod also means it does not reach such a steep angle which also helps reduce up/down thrust. Simple slot in the piston will do.

18mm long piston with wrist pin 3mm from end. 53mm conrod ctrs. Piston to end cover has 1mm clearance. Though you don't really need a 3mm long spigot to locate the cover, 1mm would do so the piston can be 20mm long so even better guiding.

[PaulR]

Cylinder can stay the same, I have assumed 10mm bore as the 8mm top left is smaller than the view from the other end.

Position of crank relative to cylinder can stay the same.

Longer piston which is what does the guiding. As you have done away with the piston rod the conrod needs to be longer. The longer conrod also means it does not reach such a steep angle which also helps reduce up/down thrust. Simple slot in the piston will do.

18mm long piston with wrist pin 3mm from end. 53mm conrod ctrs. Piston to end cover has 1mm clearance. Though you don't really need a 3mm long spigot to locate the cover, 1mm would do so the piston can be 20mm long so even better guiding.

Thanks for this, all of which makes perfect sense. I suppose the increased mass of the the longer piston and the friction it creates would require more pressure but I guess the values concerned are very small. And yes, a 3mm spigot on that cover is a bit much!

[PaulR]

the choice of the o'ring is easily made with a complete set of 1/10 mm drills or equivalent, (which you probably have, only up to 10mm in my case): by placing the ring on the drill shank and and by inserting it in the bore, the finding the optimum depth of the groove to be machined is easy.

Nice, I hadn't thought of doing that.

PS When I first read your reply I took is as "a complete set of 0.1mm drills" and nearly fell off my chair 

[Jasonb]

If you want to reduce piston weight then you can drill up from the slot and friction could be reduced by cutting a waist around the piston. However a good fitting piston and very light oil should not present any significant friction, certainly less than any ring or packing. see first image

Your relatively small bore and long stroke are going to limit what you can do without risking the conrod hitting the end of the cylinder.

The middle picture shows one of my single acting engines with a long piston which runs fine with no seals from a slow tickover upto speed on 5psi. so friction is not an issue.

The third shows another single acting engine, although the 12mm dia piston is grooved for a ring I run it without, again good from tick over to 2500rpm

[PaulR]

If you want to reduce piston weight then you can drill up from the slot and friction could be reduced by cutting a waist around the piston. However a good fitting piston and very light oil should not present any significant friction, certainly less than any ring or packing. see first image

Your relatively small bore and long stroke are going to limit what you can do without risking the conrod hitting the end of the cylinder.

The middle picture shows one of my single acting engines with a long piston which runs fine with no seals from a slow tickover upto speed on 5psi. so friction is not an issue.

The third shows another single acting engine, although the 12mm dia piston is grooved for a ring I run it without, again good from tick over to 2500rpm

Thanks again, more great ideas for me to think on. I might well shorten the stroke and lengthen the piston as you suggest.