What is your opinion on semi-rotary Corliss style valves?

Hey, what is everyones honest opinion on corliss style semi-rotary valves? I found a sentence on Kimmel's modern steam website about "rotary" valves leaking when cold and seizing when hot, but I thought he might just be talking about full rotation rotary valves. Also, does everyones opinion on them change based on if they are used at low ~125 psi steam pressures and temperatures versus more modern high pressure/temperature steam?

https://youtu.be/OMqqOln4s0Y

I designed and built this engine some time ago, and I've recently got back into steam and was looking at easy fabrication, and I believe this design I made would actually work very well and be easy to build since it is almost entirely made of imprecise holes in a block of metal with little actual milling.

It uses a variable cam operated semi-rotary valve for steam flow. It would probably be more efficient if the intake and exhaust were separated to their own valves.

Let me know what you think about my engine and semi-rotary valves.

I've no idea how good they are technically but I can say from hearing an huge mill engine running that they are very irritatingly noisy as they flip at each stroke - very unlike the typical quiet running which we expect.

Mike

Come and visit our
The New England Wireless and Steam Museum
I think it’s Oct 4th this year all engines running.
[www.newsm.org]
you can judge for your self.
Rolly

Thank you for your interest, and stimulating thoughts regarding steam engine configurations. I'm not an expert here, just gonna try to stimulate some interest in others and hopefully some comments from the "real steam people" who have massive amounts of experience compared to the rest of us!

My impression is that the Corliss valve has some serious advantages--including simplicity of fabrication and operation. Addionally, the "trip" mechanism which is associated with this valve type adds to the advantage of the Corliss type--but it appears this advantage was only applicable to relatively low engine speeds (as compared to the speeds being attained by "modern" steam engines).

I invite you (and other Corliss enthusiasts) to relate their experiences in converting this concept to a more "modern" engine, with the higher rpm's that are implied.

Ain't steam fun?

B.

I spend a lot of time reading old engineering texts and trade journals.

Corliss engines worked well in the days before commercial electricity driving mills and machine shops. The typical application involved widely varying loads which were accomodated by the trip mechanism adjusting intake valve cut off. The seperate intake and exhause ports were an advantage.

They typically operated with 100-150 psi saturated steam, often condensing. I never read about a Corliss engine using high pressure or superheated steam.

They were low speed engines. They generally used pretty large cylinders. I expect there would be problems trying to scale one down.

Kerry

Bill Hinote Wrote:
-------------------------------------------------------
> Addionally, the "trip"
> mechanism which is associated with this valve type
> adds to the advantage of the Corliss type--but it
> appears this advantage was only applicable to
> relatively low engine speeds (as compared to the
> speeds being attained by "modern" steam engines).

I have no intention of using the silly and complicated corliss valvegear, only the valves themselves. My engine shall be cam driven with selectable cams, as in the video I liked above. This should remove any low speed limitation. I was simply wondering if there were issues with the valves, things like wear, warping, superheat issues, etc. For some reason, I seem to have the most difficult time designing easy to build poppet valve engines.

There are two major problems associated with rotary valves:

1. You need a close fit between the valve and the valve bore to prevent leakage. This is OK with saturated steam but when you get into high degrees of superheat the differential thermal expansion will tend to cause the valves to stick in the bores. Graeme Vagg told me his first rotary valve engine ran beautifully on air but as soon as he hooked it up to his PS-77 automotive boiler it ran for about 30 seconds before seizing up violently.

Even assuming tight clearances there is a steam engine axiom that states: "There are two types of valves, poppets and those which leak">

2. Rotary valves work better in saturated steam systems. Given the tight clearances between valve and valve bore mentioned above, it is apparent that some kind of lubrication is needed to reduce friction so as to minimize both loss of power and wear. Saturated steam engines tend to inject oil into the steam itself, thereby lubricating the valve surfaces and controlling friction. This approach is not satisfactory for very high superheat, the oil decomposes into solids that coat the surfaces. These solids not only have insufficient lubricity but can also sometimes be down right abrasive.

A superb example of this was oil carryover into the boiler from Doble steam automobile engines. The boiler temperatures caused the oil to decompose, coating the boiler tube and blocking flow (when it didn't cause burnout due to the insulating properties which prevented the steam flow from carrying away excess external heat). Various cleaning methods were employed including burning the carbon out by lighting the end and supplying air so that it would burn down the tube much like tobacco in a cigarette while the German railways sand blasted the tube interior.

Regards,

Ken

frustrated Wrote:
-------------------------------------------------------
> There are two major problems associated with
> rotary valves:
>
> 1. You need a close fit between the valve and the
> valve bore to prevent leakage. This is OK with
> saturated steam but when you get into high degrees
> of superheat the differential thermal expansion
> will tend to cause the valves to stick in the
> bores. Graeme Vagg told me his first rotary valve
> engine ran beautifully on air but as soon as he
> hooked it up to his PS-77 automotive boiler it ran
> for about 30 seconds before seizing up violently.
>
> Even assuming tight clearances there is a steam
> engine axiom that states: "There are two types of
> valves, poppets and those which leak">
>
> 2. Rotary valves work better in saturated steam
> systems. Given the tight clearances between valve
> and valve bore mentioned above, it is apparent
> that some kind of lubrication is needed to reduce
> friction so as to minimize both loss of power and
> wear. Saturated steam engines tend to inject oil
> into the steam itself, thereby lubricating the
> valve surfaces and controlling friction. This
> approach is not satisfactory for very high
> superheat, the oil decomposes into solids that
> coat the surfaces. These solids not only have
> insufficient lubricity but can also sometimes be
> down right abrasive.
>
> A superb example of this was oil carryover into
> the boiler from Doble steam automobile engines.
> The boiler temperatures caused the oil to
> decompose, coating the boiler tube and blocking
> flow (when it didn't cause burnout due to the
> insulating properties which prevented the steam
> flow from carrying away excess external heat).
> Various cleaning methods were employed including
> burning the carbon out by lighting the end and
> supplying air so that it would burn down the tube
> much like tobacco in a cigarette while the German
> railways sand blasted the tube interior.
>
> Regards,
>
> Ken

Ah, this is exactly what I was looking for, thank you.

Off topic for the thread, but I don't feel like making another topic so soon.

What do you think about this boiler design? similar to an ofeldt, but with coils circling around the central tank. I plan on using some sch 80 black pipe and threaded end caps for serviceability, with 1/2in copper tubing coiled around the outside held in place with wire ties or similar (maybe sheet metal coil frame?). Surface area is ~8 square feet, which should be good for a horsepower or so. Are threaded fittings ok at 100-150psi? Also, how to best attach the copper tubing to the side of the steel pipe without welding? I was thinking about picking up a wire feed mig welder but I would rather silver solder with soldering experience than weld with no experience.

The Corliss valve is not a rotary valve as some may think. It’s more like a D valve that sits in half a sphere and rotates back and forth to open and close. It wears in like any D valve. Not like a piston valve there wears out. It needs lubrication like any other steam engine. The steam pressure holes it on the seat like any D valve or grate valve
These engines like the Green and many others have a separate valve for intake and exhaust to achieve better expansion events.
Rolly

I'm assuming each coaxial coil is connected to the central drum at both top and bottom rather than as a single coil stack. That would basically be a Bosolver boiler. These have an advantage over an Ofeldt in that you don't have the same large gaps for combustion gasses to pass through without giving up heat to the tubes.

The inner tube should be wound most tightly and each successive tube less tightly --- because all the tubes should be the same length. The goal is to get about the same heat transfer rate in each tube. If the tubes get longer then the amount of heat absorbed will be uneven, one tube will eventually fill with steam and restrict further water flow through it. At the same time, the other two tubes will tend to receive more flow which will in turn keep them cooler. The steam filled tube may eventually overheat to the point that the tube burns out.

frustrated Wrote:
-------------------------------------------------------
> I'm assuming each coaxial coil is connected to the
> central drum at both top and bottom rather than as
> a single coil stack. That would basically be a
> Bosolver boiler. These have an advantage over an
> Ofeldt in that you don't have the same large gaps
> for combustion gasses to pass through without
> giving up heat to the tubes.
>
> The inner tube should be wound most tightly and
> each successive tube less tightly --- because all
> the tubes should be the same length. The goal is
> to get about the same heat transfer rate in each
> tube. If the tubes get longer then the amount of
> heat absorbed will be uneven, one tube will
> eventually fill with steam and restrict further
> water flow through it. At the same time, the
> other two tubes will tend to receive more flow
> which will in turn keep them cooler. The steam
> filled tube may eventually overheat to the point
> that the tube burns out.

Would that matter as much if each coil has its own separate connection into the central tank, instead of a common header? Also, since I'm planning on using wood or wood pellets for firing the heat load shouldn't be too hard.

It really doesn't matter if the connection is to a tank or a header, if the flow is allowed to divide among various parallel paths it will vary in inverse proportion to the resistance of each path.

I should really spin your question on its head. "Would it be that difficult to use the same length tube for each coil if it reduces the potential for problems?"

Ken

That is a fair point. I was mostly just curious. Now, what are good ways to attach the tubing to the boiler without welding for <150psi use?

If your going to silver solder the best way is to weld in a nipple and drill the nipple for the OD of the tube and silver the tube into the nipple. Then you’re only heating the end of the nipple to do the silver joint.
Rolly

The Corliss valve is a slide valve is Roll stated. The Corliss engine has separate inlet and exhaust valves. The uniflow engine was compared to the Corliss engine in the uniflow book by Stumpf.

I have looked at a Corliss type duel slid valve. It works like a Corliss only having two rotating slid plates that alternate opening and closing functionality.

The Corliss valve gear provides the advantages of quick opening and closing and thus less flow restriction. A cam could also provide the same. Separate inlet and exhaust Corliss engine provided close results in the uniflow comparison in Stumph's The Uniflow Engine. The Corliss did not have as high a compression. I wonder how it would of compared with equal compression.

With the Corliss valve gear I think RPM is too limited. I do not know how fast a maglatch valve can go. But suspect they should compare to a cam as springs provide the cam tracking force.

There are many Corliss engine vids on youtube. They also retained prefferance over uniflow in high power output installations.

As exhaust valves they are a challenge to achieving a low clearance volume. My idea is a variable clearance engine with clearance losses being eliminated by compression. At max output the engine size is dependent on clearance. Zoro clearance giving the max power for a given engine displacement. other factors being equal and non zero clearance elimitation by compression.

Andy

Andy
I like the grate valve as apposed to the Corliss valve.
Very little movement and you can achieve a large opening.
The Green engine uses grate valves for the intake.
Rolly
[www.newsm.org]

I understand your preference Rolly.

Using separate inlet and exhaust valves I think a sliding exhaust valve would not be a good choice. The inlet has steam pressure holding it sealed. The same would be a problem for exhaust unless it were inside the chamber. If I remember right pressure was used to hold the seal on Corliss engines. With high pressure the required force would be excessive. It's not exactly the pressure but the variance. With compression the inlet valve could have low pressure differential when transitioning. The high transition speed ment little change in the pressure during valve movement.

I first thought of using a poppet valve exhaust.

A rotating valve simular to your drawing is something I have considered. A small rotation would open or close the valve,

One consideration with the grated type valve you favor is the opening area with the valve in the head. You have a limited area the opening can occupy. With a single opening you get the max flow area. Dividing it up as you suggest would result in half the flow are.

Generally steam engine valves are located in passageways leading to the cylander. The above applies when the valves are directly on the expansion chamber.

There are many things one must think about. Cylander condensation is an important design issue. One reason I favor a floating slide type of inlet valve.

We can, borrowing from IC engines, we can get more flow area with a spherical or wedge head design.

Andy

My preference to the grate valve is only in relationship to the large old engine valves. Much easer to maintain and yes all valves held against there seat by the steam pressure require lost power to move them. If you know how to set the crank angle and valve, those big old engines start right up on there own as soon as the steam valve is opened.

The Doble inside admission type piston valves takes almost nothing to move. Only the loss of the packing gland on the valve rod. The D valves on the Stanley takes almost three horse power to move.
Rolly

The Corliss valves transition very quickly on all Coriless engines I have seen. With compression near steam chest pressure the friction would be small compared to slide valves on most old engines having little compression.]

Unfortunately with separate inlet and exhaust valves that only works for inlet.

Andy

All files from this thread

File Name	File Size		Posted by	Date
boiler 1.png	222.1 KB	open | download	zimirken	07/30/2015	Read message
Corliss&RWM.jpg	28.6 KB	open | download	Rolly	07/30/2015	Read message
P1010003.JPG	199.5 KB	open | download	Rolly	07/30/2015	Read message
540 X 456.jpg	77.2 KB	open | download	Rolly	07/30/2015	Read message
Grate Valve.jpg	72.8 KB	open | download	Rolly	08/01/2015	Read message