Going Forward into Development
MarkNorton
internet design plans.
I also have seen a lot of plans, drawings, etc. floating around the
internet - you have but to look. I think if you look hard into the
matter, steam engine designs fall into the following categories:
* Unproven ideas (generally never built)
* Curiosities (like Sterling Engine models)
* Good performing engines, but protected by patents (the Cyclone
Engine is a good example, though still being tested)
* Old, well proven designs most out of patent protection (age of
steam, steam car engines, etc.)
One of the main goals of this project is to develop open source plans
for a solid, reliable engine. So far, we are targeting two parametric
frameworks: a low HP engine for driving equipment and a higher HP for
electric power generation. Discussions around the requirements have
settled down and people are now exploring how to go about developing
those engines. Ken has also pointed out that the source of steam
(hence boiler design) also matters, since it impacts the operating
parameters of the engine being built.
As a side project of developing these engines (and perhaps the support
equipment like boilers, condensers, oil extractors, safety gear,
etc.), I'd like to collect information on public domain designs and
possibly even commercial offerings. After all, if someone just needs
a steam engine, the cheapest solution may be do just go a buy one,
especially if the start up cost of a machine shop might be part of the
overall costs. This documentation should include designs like the
Bourke Engine - with notes concerning it's implementation status,
design issues, etc. After all, others will come along after us and
might not have the time to wade through 100's of forum entries. BTW,
Ken, this is a problem that SACA suffers from. The Phorum is very
difficult to use as a source of information because it is completely
unorganized. Tom Kimmel said somethings similar to me, as well. I'm
looking for volunteers who might have a bit of time they could spend
gathering documentation on steam devices for reference use.
Meanwhile, there is the question of how to proceed. I am looking for
volunteers to form two project teams to further improve the
requirements and specifications of two engines designs: low power to
drive equipment, medium power to drive an electric generator. If you
are interested in participating in either of these efforts, please
respond here. I will collect the results. You can, if you like
volunteer for both projects, but I would not recommend that unless you
really have the time to devote to both. The conversations and design
results of both teams will be completely open, so you'll be able to
track what's going on in other projects, though we may need to have
some rules to avoid distracting them.
We've made a good start towards figuring out what to build, but more
work is needed. I am grateful to Ken Helmick for sharing his
experience with us. It will save us a lot of time, effort, money,
maybe even a bit of blood to leverage the work of those that have gone
down this path before us. Let's learn from that and try to add some
value.
- Mark Norton
Mike Stone
> Meanwhile, there is the question of how to proceed. I am looking for
> volunteers to form two project teams to further improve the
> requirements and specifications of two engines designs: low power to
> drive equipment, medium power to drive an electric generator. If you
> are interested in participating in either of these efforts, please
> respond here.
Since I seem to be the most vocal advocate of the low power option, I'll volunteer for that group.
Mark Norton
Mike Stone
> Thanks, Mike. Would that be the low or medium power engine project?
Low power.
Russell Philips
jamie clarke
OSE, appropedia, saca.
I can at least think about it, if i read or transcribe one a day and there is more then one of me it wont be too bad.
Ken
title, date, inventor, brief description and a classifying keyword,
with some hundreds stuck in a few files needing to be entered. All
the patents were downloaded as pdf files and are hyperlinked to the
spreadsheet. I currently sell the CD for $100 but we can work
something out instead.
Regards,
Ken
On Nov 22, 10:01 pm, jamie clarke <jamieclarke...@gmail.com> wrote:
> Someone showed a list of over 200 patents so i could try and categorize
> them or transport all that information to one wiki site.
>
> OSE, appropedia, saca.
>
> I can at least think about it, if i read or transcribe one a day and there
> is more then one of me it wont be too bad.
>
> On Wed, Nov 23, 2011 at 2:55 AM, Russell Philips <russellphil...@hotmail.com
>
>
>
> > wrote:
> > low power (2hp)- Hide quoted text -
>
> - Show quoted text -
Ken
There was some discussion earlier of very small powerplants. During
WW2 a number of small steam generators were built to provide power for
radios using locally obtained fuels. The engine was basically the
Stuart-Turner Sirius though I'm unsure if S-T still sells the kit. In
any case the Sirius was essentially a miniaturized version of the
1890's Westinghouse light industrial engine; an SA, 2 cylinder machine
with a single piston valve in a common cylinder head supplying both
cylinders. Even for it's day the Westinghouse was not notably
advanced, but won praise for it's cost and reliability.
http://www.prestonservices.co.uk/generators.htm (scroll down)
http://www.youtube.com/watch?v=fzXemnsajuk
Regards,
Ken
jamie clarke
Opinions?
Ken Helmick
Russell Philips
(Perhaps confirming the obvious), a 20hp boiler at 10% yields 2hp
engine output - this is the 'current projection' (requires 4m x 5m
collector if solar thermal is used for heating).
I clearly am not talking from any experience! But everywhere i read -
turbines are more efficient than pistons. If this does not apply on
small turbines - only for large - i get it. I have no data or
experience to discern. I opened to turbines (which clearly i can not
make at home) when i considered purchasing a 'common' air compressor
lower for steam use (which also is not made at home).
If a turbine (motorcycle turbines are smallest i think) is not more
efficient at these sizes (2hp) then i am happy to put this to rest-
just want to make sure.
I think you just said as much, just looking to confirm.
On Nov 23, 6:22 am, Ken Helmick <kena...@aol.com> wrote:
> Yes, build a 20 HP boiler to provide the steam! I realize that turbines sound sexy, they seem more modern, they run fast and the name sounds cool....been there and done that myself. Actually, reciprocating engines are more fun when you actually start running a powerplant, turns out the average turbine is just a box with a shaft sticking out that makes horsepower. Turbines that size are steam hogs, big appetites for the work they perform. You're going to spend about 4 times as much effort getting fuel and need to shell out a lot more cash for a boiler or solar collector. The whole notion of free energy is unrealistic, one of the first rules of economics is that almost everything has an opportunity cost...this being the value of the option you had to forego. If you have to collect wood for fuel you can't plant a field or fetch water or teach your child. If you need to buy a larger solar collector you can't spend the extra money on food, clothing and etc.
>
> Realize that turbines in this category are not typically powerplants, but powerplant auxilliaries, usually operating pumps, blowers and such like. Very often their inefficiency can be tolerated because the high BTU exhaust heat is still useful in a larger powerplant, often going to the RPSS (Reduced Pressure Steam System) for use in oil purifier heaters, distilling unit heaters, DFTs (Deaerating Feed Tanks) and so on; if there isn't enough exhaust steam to meet these demands you'd need to use a pressure reducing valve anyway, so the small turbine is unobjectionable.
>
> Even the above is a bit dated, the relative complexity, cost and reliability of small steam turbines makes it well worth using electric motors to drive auxiliaries. Don't think because steam technology has been around a long time, that it is necessarily simple and trouble free; our ancestors weren't simple minded, they just had a smaller experience base to work off of. Steam locomotives died because they were far too maintenance intensive compared to diesels; steam makes things rust and if there are leaks or water in the steam it can erode steel.
>
> -----Original Message-----
> From: jamie clarke <jamieclarke...@gmail.com>
> To: open-source-steam <open-sou...@googlegroups.com>
> Sent: Wed, Nov 23, 2011 7:38 am
> Subject: Re: Going Forward into Development
>
> Ken, Im am salivating over the 5hp turbines....
>
> Opinions?
>
Ken Helmick
|
Re: Engine Design August 13, 2011 02:00PM |
IP/Host: 108.81.224.117 Registered: 6 years ago Posts: 896 |
I know, turbines are so small for what they put out and that and no need for internal oil is most attractive. I spent several years working on this very hard, just because of those reasons. The things just eat steam when starting up, consider a water rate over 100lbs/hp/hr vs. down to maybe 15 at peak speed and they scream like a banshee too. A light weight monotube goes whoosh and there you are with an overheated dry steam generator. Even the Lungstrom counter rotating was investigated, because it looked really good, also the re-entry axial and radial impulse types. Bob Barber just shot them down one by one that day in detail and he is one turbine designer who really knows his facts. The guys at Sunstrand agreed with Bob and they had a toluene turbine powered bus running around their plant too. The head engineer said the same thing, reciprocating for a passenger car and stop fighting it!!
On the Barber-Nichols web site, there is a splendid set of curves for every type of turbine and steam-vapor expander known and it is deadly accurate. Take a look at the speeds one has to run his expander and the efficiencies and you will see what Bob was talking about. Then there is that 1/2-V ratio for the impulse type and the 1-1 velocity ratio for the reaction type you have to deal with to get any efficiency. Even in our race car, the leisurely acceleration from rest was more like an old clapped out Buick with DynaFlush until it got to about 40,000 rpm then all hell broke loose and it really started to go.
Then came the three Lysholms in series and one bought for study too and then the Wankel with several purchased again for study, all of which had so many problems I reluctantly had to admit defeat and go back to the piston expander.
Jamie Clarke
Jamie Clarke
If you want better efficiency it's all about the hot end. Steam can't go as high as you need it to but a good steam system is the basis for any form of larger scale power generation.
Nuclear reactors boil water and they are only cost competitive per watt because they work I'm the giga watt regime.
With solar I want to see a 3000 degree helium cycle cooled by a team turbine in the future.
Closed helium loops r the key to space flight and energy.
Sent from my iPhone
Paul Passarelli
We have a "Chymist" working with us! Excellent! Just remember when @Jamie puts the negative number in his energy computations, it's OK! So our small (2kW) system will have ΔG = -1719.7 kcal/hr Based on the conversion factor: 1 kcal = 0.001163 kWh
And if I recall correctly a Big Mac had 451 kcal or maybe it was 541 it's been >30 years since I took that particular chemistry class ;^)
--Paul
jamie clarke
I prefer the term alchemist :)
really enjoying reading into the sceptical chymist
jamie clarke
Russell Philips
Perhaps selecting an existing expired patent is the quickest path
through the mine field to our design.
Perhaps applying our design requirements would narrow the patented but
expired designs.
What would be the top 3-5 candidates?
Ken
First thing to remember is that most patents are sheer garbage, it
isn't like anyone has to actually build and test an idea before
applying for a patent and I would estimate that less than 1/3 of steam
engine patents ever saw construction.
The most successful traditional engines didn't suffer from patents at
all, the basic designs arose among various builders and desirable
features were copied until a general consensus emerged...this all
occuring from 1850 to 1890. Of course, the end of that development
process was about 120 years ago, the consensus designs were the most
economical for the technology of that day, skills and materials that
were common then are harder to find today whereas we have ready access
to different resources.
Assuming that a self starting engine is not a high priority, we aren't
necessarily limited to the 2 cylinder double acting engines that were
common at that time. Ditching the DA design eliminates the crosshead,
piston rod and stuffing box and permits more care free high speed
operation. The disadvantage of SA engines is water contamination of
the crankcase. By running a small steam line through the case, the
oil can be kept hot enough to drive of condensate, however. Antique
engines also ran on heavily babbited bearings, these needed to be
taken up frequently as they wore, more modern shell type antifriction
metal bearings work better but are very sensitive to oil
contamination. On the other hand, high quality drawn cup needle
bearings are readily available today; the lubrication needs are
minimal as evidenced by 2 stroke IC engines lubricated with just some
oil mixed in with gasoline. Likewise, you might simply go to somewhat
over size sealed ball bearings (oversized to obtain longer life) and
skip the lubrication altogether. Traditional side valves wear and
don't do well when superheat is applied, piston valves do better but
the shared admission and exhaust port is wasteful. Today we can buy
cheap poppet valves that are hard, strong, durable and temperature
resistant whereas back in those days poppets tended to need frequent
regrinding; so what was accepted practice in those days might not be
the best decision today.
Regards,
Ken
On Nov 25, 6:19 am, Russell Philips <russellphil...@hotmail.com>
wrote:
> Perhaps selecting an existing expiredpatentis the quickest path
Mike Stone
> Ditching the DA design eliminates the crosshead,
> piston rod and stuffing box and permits more care free high speed
> operation.
That would mean putting a rotary joint in the piston and running the connecting rod directly from the piston to the crankshaft, correct?
Obviously it's a worthwhile idea -- pretty much every IC engine in the world works that way -- but it could make the piston harder to build. In an IC engine, you want a rib inside the piston that carries force from the crown to the wrist pin, then you want a slot in the rib to make room for the end of the conrod. That's a moderately complicated shape at the bottom of a hole, so it presents some challenges for fabrication.
The piston-rod/stuffing-box/crosshead design gives you more reciprocating mass, but it's stupid-simple to build.
I'm looking at the lower-power design, which will have smaller parts. That means mass penalties probably won't be as bad for me, but I'll be trying to work in smaller spaces. Let's talk through the relative problems and work out a rough idea of the scale where a more advanced piston design starts to pay off.
First question: is it easier to design a piston for a steam engine than for an IC engine?
Second question: is there a general rule of thumb for the amount of energy you want in a reciprocating system? Energy increases with the square of speed, but mass increases with the cube of size.. a 1" cube of steel moving 2.8 ft/sec has roughly the same energy as a 2" cube of steel moving 1 ft/sec.
Follow-up point: as I understand it, you design the recompression stage to balance the energy in the reciprocatng parts. Ideally, the force needed to recompress the gas in the cylinder will be just enough to bring the piston to a stop at TDC. As long as the pressure necessary to stop the piston is lower than your supply pressure, it's all good, and you see a modest increase in engine efficiency.
Ken Helmick
From: Mike Stone <mi...@yawp.com>
To: open-source-steam <open-sou...@googlegroups.com>
Sent: Mon, Nov 28, 2011 2:49 pm
Subject: Re: Going Forward into Development
Mike Stone
> Essentially you're building the same component whether you go with a DA or SA, so the complexity isn't any different except now you don't need to worry about the stuffing box aligning perfectly with the piston rod so as to avoid wear.
Got it. As a lathe guy, I tend to forget that keeping things concentric requires a large, heavy, and expensive tool. Everyone feel free to call me on that if I start getting snooty about the cost of tools I don't happen to have in my own shop.
> Anyhow, nothing says an SA piston is necessarily complex. A once piece piston is typically cast with a core molded in a core box to give the added internal geometry.
That's the method I assumed would be most common. Casting is the king of making the same shape twice, but does require a foundry. Steel casting is serious business, and probably not for amateurs.
> Two piece pistons have long been built which consisted of a simple inverted cup with a simple yoke attached to the top of the cup. A simple mill and lathe job.
Cool.. an alternate design that can be made with other tools answers every concern I have.
> I don't know of any rules of thumb for reciprocating energy except that you want the lightest mass possible consonant with adequate strength.
Darn. Now I'll have to find another way to justify my 1/4" bore, 1/2" stroke, 45,000 rpm engine. ;-)
> Well, the old time steam books considered recompression a loss and they only tolerated it because they figured it dampened the piston at TDC. As it turns out, recompression causes a loss of peak power output, but it also improves overall thermal efficiency as your recompression pressure approaches the admission pressure.
That's the theory I was using.. you want conditions inside the cylinder just before intake as close as possible to the conditions of the supply steam. Steam can do a certain amount of work for every degree of temperature it loses, and a sudden drop in temperature as it enters the cylinder wastes just that much heat. If the cylinder pressure is lower than the supply pressure, you get an expansion which robs you of heat again.
> Clearance volume robs efficiency. If we assume isentropic expansion and compression we realize that upon release, the steam entropy drops drastically although it had been constant during the expansion part of the stroke. On the upstroke that low entropy steam is caught in the clearance volume and mixes with higher entropy admission steam....reducing overall entropy. Buy recompressing we tend to offset some of this thermal loss, the power output drops because we used work to do the compression but that is offset by the drop in steam consumption.
I think of it as 'circulating energy'.. energy in the machine that doesn't do any work, has to be spent at startup, is lost at shutdown, but just sort of sloshes back and forth during operation. As long as it amortizes to "engineer's zero" (any number so small that, if it were smaller, I wouldn't care) over a reasonable run and doesn't force me to increase the mass of the machine just so it can play with itself, I'll generally file it under "stuff to worry about later."
In this case, the alternatives seem to be passing energy back and forth between the piston and the recompression steam or spending it rattling the joints between the piston and the flywheel.
> If you REALLY want to understand the concepts behind good reciprocating steam plants, I suggest you go to google books and download the 1922 edition of Johann Stumpfs "The Unaflow Engine". Gotta warn you, it's not for the faint of heart.
I'll have to check that out. Thanks.
My current go-to reference is Ewing's 1894 "The Steam-Engine and Other Heat-Engines". I like the fact that it covers the history of steam engines as well as giving the math that makes everything work. I'm less fond of his tendency to define symbols about five pages after he starts using them in calculations though.
Ken
proposal, they work better once something is on the table. Given the
clip of Tom Kimmel posted earlier, I sketched up an engine that
encompasses at least some of the qualities he was mentioning....and he
was pretty much spot on.
https://sites.google.com/site/opensourcesteam/ken-s-page/uniflow
Ken
Mike Stone
> It's almost impossible to get a committee to formulate an original
> proposal, they work better once something is on the table. Given the
> clip of Tom Kimmel posted earlier, I sketched up an engine that
> encompasses at least some of the qualities he was mentioning....and he
> was pretty much spot on.
It's not really my project since I'm looking at the low-power engine, but I'll chime in with some praise, observations, and a couple of questions:
- Construction looks nice and simple. I like the weldment at the outer edge of exhaust ring. The valve assembly looks like good design practice distilled to its basics. The piston design is clever.. I wouldn't have thought to break a complex product down into that set of simple shapes. All in all, it looks like a very buildable design.
- The crown of the piston looks a bit thin. The steam will load it like a simply-supported beam with the supports at both ends, so a certain amount of flex will be inevitable. It might be worth putting a collar around the bolt to provide a direct line of compression from the center of the crown to the wrist pin.
- I'm a little unsure about the diagonal bar in the timing advance mechanism. It has the beauty of simplicity, but it applies torque in a way that could get ugly. There's a long lever arm between point that receives torque from the rod on the left side and the point that passes torque along to the rod on the right side. In addition, the sharp corners at the ends of the bar will act as stress-raisers. Those aren't fatal flaws, but it's probably an area that wants some care when you're setting dimensions.
- Your follow-up post already mentioned making the top half of the piston a bit smaller so it has room to expand yet stay clear of the cylinder wall. I like that. It might give the piston some tendency to wobble on the return stroke, especially with the offset wrist pin. Again though, a bit of care with the dimensions at the bottom of the piston skirt should keep that in check.
- I admit that the offset write pin escaped my attention until you mentioned it. I don't know enough to challenge the advantages you listed, so I'll just sort of nod and smile.. they sound good, and make sense.
Ken
Like I said before, it was just to get some ideas on the table. That
basic design could easily operate counterflow by eliminating the
uniflow exhaust ports and adding the second exhaust valve. OR, one
could add the second exhaust valve to the to the uniflow design
(called an auxiliary or aux exhaust), depend on the uniflow ports to
get rid of much of the steam and then use the aux exhaust to delay
compression onset to permit reduced clearance volume. A lot is going
to depend on boiler temperature and pressure as well as the
condenser. Steam generally provides the vacuum in a condenser, steam
condenses and occupies much less volume, nothing can fill the volume,
so you have a vacuum. Problem is, it's hard to get all the air out of
the feed water and it tends to accumulate in the condenser, so an
exhaust pump is needed. Anyhow, the initial steam condition and
cutoff determines exhaust pressure and the condenser pressure and
compression ratio determines recompression...so you need to tailor the
engine to whatever you are planning to do.
The crown of the piston is close to what you would see on a comparable
automotive engine, so depending on pressure it should be ok. On the
other hand, they cast the unit as one piece and may be getting better
stress distribution. I have similar design in my own engine and use a
more elaborate inner yoke with a steeple configuration to absorb
pressure better and a support around the screw that rises all the way
up to the piston to support the crown. That's effectively equivalent
to what you proposed and obviously I have no problems with the
idea...consider it to be conservative engineering, actually.
The diagonal bar should be strong enough, the valve loads shouldn't be
that large. I would build the unit up rather than making it in
pieces, so that is why I didn't include a radius, if forming the part
from a whole, I would most definitely radius the corners. Actually,
the part could be round instead of square, since the hole is not
coaxial with the shaft, it will work just about as well and you could
ream the hole to size and stuff an oil lite bushing in there.
Truthfully, for a real working engine, I think the Rite's governor is
a better choice than any conventional valve gear...it will control
cutoff automatically to hold desired speed to a close range and by
once properly adjusted tends to shoot for longest expansion for any
power demand. It's dirt simple and reliable. In a machine shop it
will automatically throttle up as you take a bigger cut on a piece of
machinery, and do the same to farm equipment as you toss in bigger
loads. Admittedly, the Rite's governor is not a 'fun' thing...you
can't speed the engine up and slow it down to get that nice 'aaahhhh'
feeling, just sits there and lets the operator concentrate on
something other than the engine.
The piston shouldn't wobble even with clearance at the top, it is a
VERY long piston, compare it to pictures of modern automotive
practice. I would cut a series of tiny 'vees' along the slightly
recessed area in order to promote turbulence in the steam flow,
retarding the flow along the piston, they call this a labyrinth seal.
Piston ring blowby is one of the bigger losses in higher pressure
engines, Stanley steamers had horrific blowby. From what I can find,
blowby is caused by high MEP and residence time. Stanleys had longish
cutoff, so the pressure was at admission pressure for much of the
stroke, They also tended to run at low rpm, so that high mean
pressure had plenty of time to work past the rings. If you go to
shorter cutoff, the MEP is reduced and the penetrating power drops.
Speed up the rpm, and there is less time for penetration to happen.
Then, if you obstruct the flow, you might have it well under control.
Note that shorter cutoff means less power unless you increase engine
displacement and/or jump up the rpm...and higher rpm reduces residence
time. If you wanted to go for REALLY high temperatures, I might
consider adding a cooling jacket to the upper inch of the cylinder and
pumping boiler feed water through that in an effor to reduce wall
temperatures even further. This is one of those steam engine
dilemmas. Higher temperature boosts efficiency but causes lubrication
to fail. Cooling the wall is a bit counterproductive, but if we use
feed water at least the energy is going back to the boiler. Also,
superheated steam is a relatively poor heat conductor and we shouldn't
be affecting the temperature in any steam except that along the wall.
This is also why I try to keep the piston and rings out of contact up
here. Truthfully, the piston changes might be all that is needed for
really high temperature, the whole subject is one that likely needs
more experimentation.
>shrug< Hopefully it provides a starting point and people take the relative simplicity and ease of construction to heart. Generally speaking, it is very easy to design very awesome looking, complex devices. It is the extremely simple machines that take skill, time and effort. About the best praise a designer can hear is: "Any idiot could have come up with that!", which is the exact opposite of the truth.
Regards,
Ken
Mike Stone
> Like I said before, it was just to get some ideas on the table.
Indeed. It's a good, solid, simple design. My comments probably jumped ahead of the process a bit, since they all boil down to dimension and tolerance issues. One of my habits, which I can't quite bring myself to call 'bad', is to overengineer the hell out of an initial design, then trim away material once I know there's clearly too much.
So.. let me back up a step and ask the correct question: roughly what dimensions did you have in mind? More generally, what range of PLAN values would work well for this design?
Ken Helmick
From: Mike Stone <mi...@yawp.com>
To: open-source-steam <open-sou...@googlegroups.com>
Subject: Re: Going Forward into Development
Jamie Clarke
Sent from my iPhone