Steam Engine Design: General Use
russell...@hotmail.com
in this pdf:
http://www.sunriseenergy.org/images/steam_design_details.pdf
This trail of breadcrumbs is my current and finest approach - all
things considered.
This is open for discussion and alteration.
If this survives review, i will spend the time to show more detail on
the head configurations.
BTW, congrads - we have the formation of a great crew here.
MarkNorton
https://sites.google.com/site/opensourcesteam/steam-engine/se-design-concepts
Overall, I think your exploration of the concepts are valid. Adding
a support collar to the piston rod outside of the cylinder is
generally a good idea - regardless of whether a crankshaft is used or
not. It reduces off-axis movement of the piston and thus should
reduce wear. I should point out, however, that some recommend a
vertical orientation of the engine for just this reason.
The idea of a bolt on head is a good one. There are many different
kinds of control valves and having a removable head would facilitated
experimentation with different control approaches. Some of these are
explored in,
https://sites.google.com/site/phase3project/projects/steam-engine/valves
Concerning the opposing cylinder (continuous flow) engine approach, it
may be tricky to design an identical head for both ends filling the
hole with a plug. First of all, a good seal needs to be provided for
the piston rod where it passes through. This was typically done with
a compassed packing seal historically. Maybe something that screws
into a tapped hole. However, the control mechanisms for each head
might be difficult to make uniform, especially if mechanical linkages
are used. Classically, a slide valve was used to control steam
injection and exhaust into both sides of the cylinder in such
designs. Have a look at:
https://sites.google.com/site/phase3project/projects/steam-engine/engine-designs/d-slide-engine
The piston rod seal is removed in this view for clarity.
- Mark
mekennedy1313
great for a quick and dirty valve but we need something that is long
lasting. Similarly the solenoid valve, they don't last in the high
temperature steam environment. I'd suggest a mechanically actuated
valve, taking lessons from the past perhaps a rotary valve such as the
corliss would be best. Also, what is the end use of the engine?
Stationary power generation or mobile transportation or a unit that
can be used for both?
Max K
Mark Norton
As to the end use of the engine, that is still somewhat up in the air at this time. So far, we are discussion stationary power generation for either electricity or to drive stationary equipment. Mobile transportation is a different ball game, IMO. The SACA guys have been pursuing that angle for many, many years.
Mark
russell...@hotmail.com
The drum valve is my own concept.
http://www.sunriseenergy.org/images/steam_valve_details.pdf
I have never seen an existing rotary valve- any links?
On Nov 9, 9:48 am, Mark Norton <markjnor...@gmail.com> wrote:
> Yes, I agree that bump valves are not an optimal solution. They show up in
> the OSE design because Marcin was ken to use the White Cliffs Solar Power
> Station design - which used bump valves in a converted Lister diesel
> engine. I am also favoring rotary valves in my current thinking.
>
> As to the end use of the engine, that is still somewhat up in the air at
> this time. So far, we are discussion stationary power generation for
> either electricity or to drive stationary equipment. Mobile transportation
> is a different ball game, IMO. The SACA guys have been pursuing that angle
> for many, many years.
>
> Mark
>
Mark Norton
- Mark
Max Kennedy
It can be done
MarkNorton
little sports car, it was. Mazda later abandoned its use in car
designs.
- Mark
Max Kennedy
jamie clarke
1)dyna-cam
2)wobble-plate
3)wankel
4)bourke, rotary valve,
5)opposed cylinder, sleeve valve
I was also reading about a desmodromic valve that is essentially a valve that is actuated by gears and a camshaft, its used on the ducati motor bike.
Would this be rugged enough to operate in the rigorous environment of wet steam?
http://en.wikipedia.org/wiki/Desmodromic_valve
http://akbarkaliwala.blogspot.com/
Max Kennedy
Max
solar44
Just like the definition of "easy", the definition of "everyman" needs clarification. If the goal is to "craft" an engine that can be build with "stone knives & bearskins" (pardon my gratuitous Trek reference) then we'd be better off producing synopses of the Industrial Revolution and translating them into 3rd world languages. That's not what I'm here to participate in.
If you're planning for the end of commerce, then that too isn't worth anyone's investment of time or effort. Assume machinery -- a well maintained Bridgeport (mill), and a lathe with a solid and true spindle, chuck, tailstock, and sharp tools. Also assume that anyony that has those machines also has calibrated and accurate metrology instruments; not necessarily an optical comparator, but a set of gauge blocks, dial indicators, granite, calipers, bore meter, etc... Again, these have to be considered bare minimums.
That's my $0.02 -- I have to hope that everyone agrees with me.
jamie clarke
For example, if we built a CNC milled dynacam but it only produced 100watts that would be pretty useless to everyone. Alternatively a 150ton opposed piston boiler that you can make with a pillar drill and angle grinder is probably going to get alot of people hurt if not killed.
Im liking the JP7 designs to be frank but axial engine fabrication and rotary valves got mentioned and i thought id push it a bit :)
I understand that this isnt always helpful but if we could CAM and then cast a dynacam i would be on it like sonic.
Mark Norton
"Tolerance between piston rings mounted on a piston head and the cylinder wall is an important aspect of the engine design. It needs to move freely yet provide a good seal. The better you can do, the more efficient your engine will be. If you are lathing your own piston rings you can improve the fit to the cylinder wall by ..."
In this way, a person of average skill with fair equipment will be able to build an engine. A person with better tools and more skill will be able to build a better engine from the same design. The challenge to us to enable that average person to succeed.
- Mark
Bastelmike
On Nov 10, 4:55 am, Mark Norton <markjnor...@gmail.com> wrote:
> While we are trying to create a steam engine design that can be built by
> the "average" person, there are minimum things that will need to
> stipulated. Some skill with metal working will be required - how much
> depends on the actual design. I may actually be possible to handle a range
> of skills, for example:
>
> "Tolerance between piston rings mounted on a piston head and the cylinder
> wall is an important aspect of the engine design. It needs to move freely
> yet provide a good seal. The better you can do, the more efficient your
> engine will be. If you are lathing your own piston rings you can improve
> the fit to the cylinder wall by ..."
>
> In this way, a person of average skill with fair equipment will be able to
> build an engine. A person with better tools and more skill will be able to
> build a better engine from the same design. The challenge to us to enable
> that average person to succeed.
>
> - Mark
to be a good strategy, IMO. But it should be mentioned that this "can
build" also depends on the dimensions of the design. A design with a
Ø8" piston may be manufacturable in a well eqipped home shop, while
the same design scaled to Ø30" can't be manufactured in any homeshop
due to size and weight !
Therefore we should discuss the desirable size of the design now. What
power and how many rpm are needed? Low rpms need either a gear/belt
drive or a more expensive generator for lower speed. In general
generators working with 1500 or 3000 rpm are most easily available and
less expensive.
For other purposes than electric power generation lower speeds than
these may be sufficient
Mike
jamie clarke
If we go with an incredibly simple piston, cylinder, valves and seals like the JP7 then the majority of us can make it. Changing the configuration of the pistons into a V or rotary engine will be easier for more skilled craftsmen.
I also agree with mike that a small cylinder high RPM engine would be the most useful for energy generation, this would be accessible to the most people and the easiest to build.
It might not be the cheapest or most efficient but its a great place for people to start and more experienced people like russell could attempt an axial piston type based on the JP7 components.
That way once we master one engine it will be simple to follow each others upgrades as our own skills progress.
When we talk about a device being possible for anyone to make we need to consider the X factor(most people are sat on there arse all day watching mindless crap on TV)
Designing it in such a way that the "average person"can build it seems
On Nov 10, 4:55 am, Mark Norton <markjnor...@gmail.com> wrote:
> While we are trying to create a steam engine design that can be built by
> the "average" person, there are minimum things that will need to
> stipulated. Some skill with metal working will be required - how much
> depends on the actual design. I may actually be possible to handle a range
> of skills, for example:
>
> "Tolerance between piston rings mounted on a piston head and the cylinder
> wall is an important aspect of the engine design. It needs to move freely
> yet provide a good seal. The better you can do, the more efficient your
> engine will be. If you are lathing your own piston rings you can improve
> the fit to the cylinder wall by ..."
>
> In this way, a person of average skill with fair equipment will be able to
> build an engine. A person with better tools and more skill will be able to
> build a better engine from the same design. The challenge to us to enable
> that average person to succeed.
>
> - Mark
to be a good strategy, IMO. But it should be mentioned that this "can
build" also depends on the dimensions of the design. A design with a
Ř8" piston may be manufacturable in a well eqipped home shop, while
the same design scaled to Ř30" can't be manufactured in any homeshop
solar44
>
> On Nov 10, 4:55 am, Mark Norton<markjnor...@gmail.com> wrote:
>> While we are trying to create a steam engine design that can be built by
>> the "average" person, there are minimum things that will need to
>> stipulated. Some skill with metal working will be required - how much
>> depends on the actual design. I may actually be possible to handle a range
>> of skills, for example:
>>
>> "Tolerance between piston rings mounted on a piston head and the cylinder
>> wall is an important aspect of the engine design. It needs to move freely
>> yet provide a good seal. The better you can do, the more efficient your
>> engine will be. If you are lathing your own piston rings you can improve
>> the fit to the cylinder wall by ..."
>>
>> In this way, a person of average skill with fair equipment will be able to
>> build an engine. A person with better tools and more skill will be able to
>> build a better engine from the same design. The challenge to us to enable
>> that average person to succeed.
>>
>> - Mark
> Designing it in such a way that the "average person"can build it seems
> to be a good strategy, IMO. But it should be mentioned that this "can
> build" also depends on the dimensions of the design. A design with a
> the same design scaled to �30" can't be manufactured in any homeshop
> due to size and weight !
> Therefore we should discuss the desirable size of the design now. What
> power and how many rpm are needed? Low rpms need either a gear/belt
> drive or a more expensive generator for lower speed. In general
> generators working with 1500 or 3000 rpm are most easily available and
> less expensive.
> For other purposes than electric power generation lower speeds than
> these may be sufficient
>
> Mike
We're talking about prime movers. If a plan is drawn with loose
tolerances and an "average" man produces the engine, the thing will
probably work... We have a participant that created a video channel
showing how a piston will pump almost regardless of the fit... But an
engine that loses too much of it's work to friction or blowby isn't
worth building, and represents an *ENORMOUS* waste of resources and
effort; IMHO.
Look at the crap that is coming out of India & China today. They're
built on plans >100 years old that are out there that *ANYONE* is free
to open up and produce. Some of those plans rose to a *very* high level
of sophistication. However, every one of those engine also became
*obsolete* when superior products came along. The Indian products are
made to the loosest possible specs and machined as rapidly as humanly
possible... They work, but they suck.
If it's easier for a user to retrofit a 250 W Honda generator and
fuel it on moonshine than to build a junkyard steam engine and haul half
a ton of wood a day to his boiler, which one do you think he'll do?
I dunno, perhaps I'm missing the spirit of the group. In fact I'm
confusing myself more and more each time I post. My goals for *any*
product are best quality at lowest TCO -- it's a difficult goal to
achieve. I flinch every time I hear a proposal to sacrifice quality
*BEFORE* the effect on TCO is evaluated.
@Mike,
Speed and power are proportional, for electrical generation, the
speed of the prime has to be as stable as possible. Direct coupling the
engine to the gen is best as long as the engine can produce the rated
power at the necessary speed. If not then *any* engine that can
maintain some fixed speed can be coupled to a generator with a belt and
wheels of appropriate ratio. While belts work and are easier to cobble,
they are *always* an inferior solution.
Max Kennedy
mike stone
> I dunno, perhaps I'm missing the spirit of the group. In fact I'm
> confusing myself more and more each time I post. My goals for *any*
> product are best quality at lowest TCO -- it's a difficult goal to
> achieve. I flinch every time I hear a proposal to sacrifice quality
> *BEFORE* the effect on TCO is evaluated.
equation. You also have to consider plant costs.
Design is always a question of tradeoffs, and you can almost always
improve both quality and TCO if you're willing to spend more money on
equipment, materials, and labor. For industrial production, you can
amortize plant costs over the lifespan of your factory, but that
doesn't work for single builds or short runs.
Thing is, anyone capable of amortizing equipment purchases over a
production run can also amortize the cost of hiring an engineer with
experience in the field to do their design. Those people will have no
interest in using an Open design.
I think it's safe to say we're looking for a design that works for
single builds or short runs. That means we have to include the cost
of equipment in the TCO (and can't amortize it very far), and should
probably keep the true total cost low enough for the plans to be cost-
effective compared to hiring an engineer.
mike stone
There's a post elsewhere to the effect that we should assume a well-
stocked metal shop as the minimum plant requirements for the design.
I recall seeing 'granite surface plate and gauge blocks' on the list.
That's nice, I'll grant you, but a decent set of gauge blocks costs
about $1500.. and that's for shop grade. Amazon has a nice 81-piece
Mitutoyo grade-00 set marked down to $3200. The $350 sets give only
slightly better results than a $30 dial caliper.
Machinists have a natural desire to make parts as close to the nominal
dimensions as possible, and that drives us to buy an ever-more-
expensive collection of toys. It's easy to justify the first $100,000
of tooling and equipment as being necessary to work to an acceptable
level of precision. But as much as we hate to admit it, we can still
get good results by cutting pieces to +.002/-0 and lapping them to a
good fit.
Again, we're probably working toward a design that will be used for
single-builds or short runs. Making parts interchangeable is a way to
lower the cost of large production runs, so it probably isn't as
important here. OTOH, short runs make it cost effective to spend the
time filing/scraping/lapping parts for an ideal fit in the specific
assembly.
Paul Passarelli
Again, any machinist worth his salt will have a set... IMHO...
Biggest tolerance issue I ever had was when a customer bought a shiny new CMM and started challenging parts. Lesson: Yeah, it's possible to be "too demanding"... Had a part that was a thin-walled Aluminum tube with a diameter and roundness spec'd +/-0.001 -- The end got shoved into a hole then welded! The hole was +/- 1/16", the tolerance after welding wasn't even defined!!! The assemply was held in a rubber lined clamp. Remember idiots abound wherever one chooses to look ;^)
Taking the time to lap even a few parts, usually convinces a mechanic that instruments provide a *HUGE* ROI on his time.
We're talking about reciprocating masses, flywheels, bearings, and close tolerances. Trying to undertake such a project without the proper tools, is simply foolish. We're talking about building an engine that people will rely upon, not "Hey dudes, look what I did." +0.002/-0.0 is OK for a mounting base, but not for a bearing race, not for a valve clearance, not for the tilt on a bore, and even for some bolting faces.
--Paul
Mike Stone
> I dunno about the "0" set being merely caliper like... or saying "00" is necessary. I had 4 of the smaller sets on the shop floor and the big one in the metrology room. Don't recall a discrepancy ever showing up. And that was with basic-black granites out on the floor and the pink granites in the lab.
I was thinking of the sub-0 sets that are accurate to a few hundred thousandths.
In general, tolerances below .0001 put more demands on your thermostat than on your measuring equipment. Steel's thermal expansion is on the order of 10^-5, so a 10* C change in temperature can move you a hundred-thousandth per inch of material.
> Again, any machinist worth his salt will have a set... IMHO…
No argument there, but the cost justification is different when you can't write them off as a business expense and won't be working them to generate revenue.
> Biggest tolerance issue I ever had was when a customer bought a shiny new CMM and started challenging parts.
Oh God..
> Lesson: Yeah, it's possible to be "too demanding"... Had a part that was a thin-walled Aluminum tube with a diameter and roundness spec'd +/-0.001 -- The end got shoved into a hole then welded! The hole was +/- 1/16", the tolerance after welding wasn't even defined!!! The assemply was held in a rubber lined clamp. Remember idiots abound wherever one chooses to look ;^)
I feel your pain. ;-)
> Taking the time to lap even a few parts, usually convinces a mechanic that instruments provide a *HUGE* ROI on his time.
True for a production shop, and especially true when you have to service a production run's worth of product later. You do *not* want to do all that work again when it's time to replace a valve or something. For a single device I built for myself and will service myself, especially if I don't have to justify my salary for the time spent, it's not so bad.
> We're talking about reciprocating masses, flywheels, bearings, and close tolerances. Trying to undertake such a project without the proper tools, is simply foolish. We're talking about building an engine that people will rely upon, not "Hey dudes, look what I did."
Benchmarking exercise: make a quick list of the tools and equipment you consider minimally necessary to do the job, then go to Grizzly or your favorite supplier and price it out. That will be the capital investment barrier to entry that you feel is appropriate to the design.
Also, let's not assume that equipment equals respect. I'll put "hey dudes, look what I did" against "whatever.. I get paid the same either way" any day.
Case in point: I have a new car whose 'check engine' light kept coming on. The dealer replaced half a dozen things according to the fault codes. Finally, the light came back on before I could even get it out of the garage, and the mechanic decided it was personal. Turns out the cable harness was just slightly too short, so all the sensor connections were under tension at the point where they met the computer module. A few cycles of heating and cooling, or a few days worth of vibration were enough to produce intermittents. After rewiring nine sensor connections, they still found problems, so they traced every line in the harness. Eventually they found a line in the fuse box that was also just a millimeter or so too short. They ended up routing that one all the way to the other side of the box. At the peak of the effort, there were four mechanics tracing cables, two on the phone to the manufacturer, and one on the phone to the plant that produced the harness.
It's possible to do a professional job with limited equipment. It's possible to do a professional job *much faster* with really good equipment. It's possible to do a crap job no matter what equipment you have.
> +0.002/-0.0 is OK for a mounting base, but not for a bearing race, not for a valve clearance, not for the tilt on a bore, and even for some bolting faces.
+2/-0 *and* lapping. By itself, you're right that such a tolerance is crap. But with $25 worth of abrasives and some time, I can get matching surfaces that rival what you'd get from an experienced operator using a $25,000 machine. Yes, hand lapping is all but worthless in a production shop where a certain amount of time has to equal a certain amount of money, but it isn't so bad for one-off work.
On the flipside, a production shop can invest more time and material dialing in a process than I'd spend doing an entire one-off build the slow way. You'll make it up by getting the next fifty units faster than I could make a second one, and you'll see better interchangeability between any two parts pulled at random from the bins. Neither of those is particularly valuable to me if I only want one unit, though.
Basic production economics: the first one costs $100,000. The next hundred thousand cost $0.43 each. The question is, "do I want a hundred thousand units?"
Bastelmike
this way:
We specify the tolerances we believe to be sound for functionality and
durability! Also, we don't use designs with tolerances that you need a
swiss-built jig borer to do. Tolerances should be able to do with
normal machine stones, even if they are second-hand.
Everything else is left to the fabricator. He can obey these
tolerances, or ignore them. It's up to him. If we specify a shaft as
2.000 +0 -0.001 and he decides to use the shaft he has made with
1.990, so what? If the machine works, everyone will be happy.
If it doesn't work, its not our fault !
But if we specify lush tolerances, we will be made responsible for bad
performance or breakage.
I would definitely prefer working with tolerances than making each
part fit into its counterpart. Second method is working for single
builds, but already difficult to organize with short runs. And
producing a spare part is too problematic. Typically the counterpart
ist many miles away and weighs a ton, requiring a crane and a truck to
get it into the shop. All this action to make a small spare part,
worth a few bucks....
Mike
Max Kennedy
Mark Norton
- Mark
Paul Passarelli
A bearing its race and shaft need a "precise" fit, but whether it's 2.0000+/-0.00004, or 1.9+/-0.005 or 2.01+/-0.01 only matter if the goal is interchangeability, or we were pushing the structural strength limits of the shaft at the slightly smaller 1.9 dimension.
There have been volumes written on GDT, and for the most part that data is freely available. We all know that a bearing, shaft and mount require a set of dimensions & tolerances. Our job is to determine what dimensions works for the overall design, and the tolerances or useable limits for the context of the assembly are, and finally what precision is necessary to assure the assembly functions as intended.
It is *VERY* important to keep these three concepts in mind and I suggest we all take extra caution when bandying the terms about (myself included).
From personal experience, I've found that it takes longer to GDT a print than to create it in the first place from a blank page. It's a dark art. The upside is that when it's done properly, many of the 5-decimal place "detailed specifications" can be *ELIMINATED* producing significant cost savings!!! Believe me when dealing with parts hardened above Rc-55 the fewer "precise" details the better. As a rule of thumb, I limit the count to 1 when the material is demanding.
Mark Norton
- Mark
Ken
I’d like to toss out some observations and comments.
First of all, I think that perhaps this discussion has put the horse
before the cart; the discussion being about a steam engine design when
more properly it should probably concern the steam cycle: Generation,
Expansion, Condensation and Feed, with the engine (expander) being
just one of the four processes involved.
Steam engines don’t exist independently, they take steam from a
boiler, and it is the thermodynamic properties of the steam generated
by the boiler that both determines the potential system efficiency as
well as such factors as engine cutoff and compression ratios. A
boiler producing superheated steam at 100 psi will potentially drive a
more efficient engine than one producing saturated steam at the same
pressure. Likewise, a 1000 psi boiler will potentially drive a more
efficient engine than the 100 psi. To build an integrated steam
system, the engine and boiler must match one another and the
combination must meet the project design criteria.
Generally speaking, powertrains aren’t designed by going straight
ahead and sketching up engines. The first order of business is to
define the design goals in terms of the thermodynamic cycle to be
employed. Given such parameters as overall desired thermal
efficiency, power output, rpm or torque, single or compounded staging,
self or manual starting, fixed or variable gearing; it becomes
possible to determine the required boiler pressures and temperatures,
engine expansion and compression ratios, displacement and so on. Work
can proceed on an engine design once the engine displacement, cutoff,
torque, rpm characteristics and nature of staging are known.
In short, discussing specific engine characteristics is premature
before the Rankine cycle demands have been evaluated.
A durable, reliable and efficient engine requires some features of
high precision, others can be significantly less precise. To my mind,
the project as so far presented should be designed in such a way to
minimize the demands for precision. I tend to discriminate between
necessary and needless accuracy, take the crankshaft as an example.
The dimensions for the main and rod bearings are necessarily accurate,
the precise contours on the counterweight are immaterial so long as
they contain enough mass which can be removed to facilitate balancing.
Just how this accuracy is achieved can make a great difference in the
skills and equipment necessary to build a successful engine. If
babbited bearings are used, the crank journals will need to be
machined to just above size, ground to almost the right size and
polished to size. If off the shelf ball or roller bearings are
employed, the need for accuracy is drastically reduced, the necessary
accuracy having been built into the bearing by the manufacturer.
The stated goals for this powerplant are relatively undemanding, and
workable engines can be designed by referencing a number of period
works. A good example would be the ICS correspondence course “Steam
Engine Design”
http://www.amazon.com/Engine-Design-International-Correspondence-Schools/dp/0917914104
which provides rule of thumb formulas pertaining to the design of a
classic bottle engine as well as drawings with relative proportions, a
capable mechanic using both the calculations and the drawings can
produce functional working drawings of a single cylinder slide valve
engine suitable for use under modest steam conditions. Beyond that
one needs to work wood to make patterns to sand cast the components,
such casting work being readily available from local foundries; and
naturally comes the machining and assembly. Some changes to the
design incorporating modern piston rings, corrugated ribbon packing,
roller bearings and so on seem reasonable.
Further changes might include the use of weldments in place of some
castings, more modern fasteners, shafting and so on. For a simple
engine developing modest power from low condition steam, the answers
are all there without need to seriously reinventing the wheel.
Some serious consideration should be given to:
The boiler design and construction, starting with something as basic
as fire tube versus water tube and going on to once through or
recirculation and even then forced or natural recirc.
Feed systems and control, how you propose to get water into the boiler
and regulate the amount fed at any time.
Burners are a topic that can be surprisingly tricky, for solid fuels
the grate area can vary wildly based on just the fuel used. Provision
for supplying secondary and primary air can be critical. Burner firing
rate must be controlled along with feed rate in order to provide for
steam delivery on demand.
Presumably condensers will be employed so as to reduce the amount of
make up feed water, minimize boiler contamination and reduce the costs
of feed water treatment; whether these condensers are atmospheric or
subatmospheric, air or water cooled can be issues of discussion.
In short, a steam engine is part of a steam system and the system is
no better than the least capable component.
Regards,
ken
Mike Stone
> I’d like to toss out some observations and comments.
Ken, thank you for the information (BTW - in reference to your credentials, allow me to say, "wow).
I admit that I've had some concerns about keeping the steam up to temperature, but didn't have a coherent process framework like the one you laid out.
I happen to be interested in the lower power device, so let me toss out a few thoughts and questions related to that steam cycle:
- I favor a low pressure design for safety, and because low pressure is probably easier to design for (the more energy there is in a system, the more care you have to put into keeping it where you want it). What range of pressures would you consider safe for amateurs?
- I lean toward superheated steam. From what I can tell, trying to keep steam in the cylinder just at the point of saturation is more hassle than it's worth. I understand that letting steam leave the cylinder dry allows energy to escape as waste, but I'd rather eat that loss than deal with the pressure losses associated with condensation in the cylinder. What range of temperatures do you consider reasonable for the supply steam, and (since I'd like to minimize the loss as much as is reasonable) what temperature margin above saturation do you think is achievable for the exhaust steam?
- Low pressure pretty much implies single expansion. As I understand it, expanding to well below atmospheric pressure (3-5 psi absolute) is a good idea.. you convert much more of the energy in the steam into work (piston motion). Is that reasonable, or am I talking banana fritters?
- As I understand it, double-acting cylinders solve enough problems to be worth the additional complexity. They keep the forces moving the piston inside the cylinder, rather than making the piston push the flywheel half the time and the flywheel push the piston the other half. Steam expanding on one side drives the exhaust phase on the other side, and steam going through recompression cushions the piston as it prepares to change direction. Again, am I talking sense, or in toes-in-the-nose territory?
- Expanding to sub-atmospheric pressure implies a condenser. I should think something like a lab aspirator would be good enough, combining the venturi effect with condensation. I see that happening with reserve water, but do you think an injector would be worth the effort?
- At the boiler side, this is admittedly a bit fancy, but a downdraft gasifier isn't terribly hard to make. All it wants is a good shake every once in a while to keep voids from forming in the combustion layer, and it reduces solid fuel to a flammable gas. It provides a natural path for gradually heating the feed water, and being able to burn a clean gas at the final heating stage would probably make the boiler itself simpler. Are you aware of other designs using the same idea, and what's wrong with them that I haven't considered?
- Finally, this may be a bit to implementation-specific, but I happen to like Corliss valves. They strike me as being easy to design and build, easy to adjust, and simple enough to be robust. Is there another valve system you like better, and if so, why?
Jason Learned
A better design would be to have the riser insulated until it gets to the top then have the gasses flow over and down to an exhaust from the bottom and then piped back up again for draw. The counter flow would allow the steam generation to take place at the top of the riser in a coil while water is pre-heating at the bottom end before the gasses reach the exhaust stack. Maybe this is not a great design but it seems like a good way of having multiple fuels (albeit liquids and gasses would need some inserts) and rocket stoves burn clean, plus a hopper with pelleted fuel could be set to a given feed rate. And I like the gravity feed of this design. Does this seem like a good boiler? I think Ken would be a better judge.
Jason
jamie clarke
This is what the original reason was for having a steam engine. A 30m^2 solar concentrator can produce 21kW of steam i think.
A gasifier would probably be useful to your average off-gridder but the fuel type would be dependent on the size of the grate as Ken has said. Steam injection can create even more syngas so its a perfect candidate for a combined cycle aswell.
That being said I think it would be a good idea to develop for the concentrator and then design a gasifier either to run a smaller engine or to match the solar concentrator.
Most of you may have different design goals in mind though so what do you think?
Jason Learned
jamie clarke
Hopefully everyone can get some use out of it then.
Thanks for your comment, glad you like the technology.
Robert Baruch
by J. Whitham from 1889, on Google Books (http://books.google.com/
books?id=yh9MAAAAMAAJ&dq=%22steam%20engine%20design
%22&pg=PP1#v=onepage&q&f=false)
I can't really tell when the ICS book was actually written --
sometimes you get a republished book that ends up having a publishing
date later than the original publishing date. I think Google Books
shows 1907, but there's no access :(
--Rob
Robert Baruch
"Steam Engine Design" appears to be part of a larger group of works,
the entirety of which is here:
http://books.google.com/books?id=Nh9WAAAAMAAJ&dq=steam%20engine%20design&pg=PP9#v=onepage&q&f=false
Ken
Hi Guys,
Thanks for the warm welcome.
Fair number of issues bought up, I'll try to cover them as best as
possible, if I fall short just whack me upside the head and hand out a
reminder.
Safety should always be of paramount interest, I had a great-great-
grandfather die in a boiler explosion at the Union Flour Mill in
Detroit circa 1870 and it would be nice to make that the limit of
family steam fatalities. Common perception is that steam plant safety
is very closely related to pressure, but the correlation is really
fairly poor. A very extreme illustration will point out the true
dangers in steam systems, suppose we have one container of just 1
cubic inch and inside is superheated steam at 3000 psi and we also
have a second container of 50 cubic feet containing no steam but
pressurized to 50 psi and filled with water at saturation
temperature. If we magically make the container around the 1 cubic
inch disappear, we get a small pop. If we magically make the 50 cubic
foot container disappear, the pressurized water will drop to
atmospheric pressure, and being at saturation temperature we will find
that part of the water will flash to steam until the latent heat of
vaporization absorbs the excess energy. The vaporizing water will now
occupy far more volume and expand at high velocity, the sheer mass of
the steam causing a large energy release. Like I said, an
exaggeration of cartoon like proportions, but it illustrates the point
that it is the amount of energy stored and not the pressure that
presents a safety hazard. Safety lies in controlling the available
energy much more than in limiting pressure, I find that people tend to
be surprised to find out what the steam tables actually say:
Steam at 100 psi, 1000 F has an enthalpy of 1531.75 BTU/lb while
steam at 1000 psi and 1000 F has an enthalpy of 1505.44 BTU/lb....the
higher pressure steam is carrying less energy per pound. To be fair
the 100 psi steam has a density of 0.132570 lbs/ cubic foot while the
1000 psi has a density of 1.22278 lbs/ cubic foot; so on a per pound
basis the higher pressure is actually slightly less energetic while on
a volume basis it is about 10 times more energetic. We can readily
see that the weight of steam and water in the boiler is more of a
safety issue than the pressure.
Superheat is a given for efficient steam plant operation. This is not
to say that saturated steam engines have not been successful, the vast
majority of light steam engines in history have used saturated steam,
but the amount of energy (oil, coal, plutonium, wood, solar collector
square footage) required to develop each horsepower drops as the
temperature rises. My own experience is that 'free' energy can be very
capital or labor intensive and efficiency matters. If we consider the
problems the specific heat of vaporization presents to the steam
cycle, this makes eminent sense. At boiling temperature, significant
energy (heat) is added to change the phase from liquid to gas without
adding a single degree of temperature; since we can only extract work
from the steam by expansion, it follows that the latent heat ties up
energy in phase change that we can't recover as work. Any energy
(heat) added to raise the temperature above saturation yields a
percentage that is potentially recoverable as work and the overall
potential efficiency rises.
Superheat is wasteful if complete expansion isn't utilized, an engine
at 100 psi running at 1000 F and exhausting at atmospheric
temperatures will exhaust at horrendous temperatures; an engine using
the same steam with a cutoff at 6% of the stroke will be exhausting
with less than 50 degrees of superheat, if the steam temperature is
900 F the exhaust will be right at the saturation line. The desired
level of performance dictates the steam conditions, which in turn
dictate engine design.
Subatmospheric condensing (below atmospheric pressure) provides two
avenues to increase efficiency. First, the cutoff can be shorter and
consequently more expansion can be wrung out of the same steam
(supposing enough superheat is added), the added expansion raises
efficiency. Ideally, recompressing remnant steam in the cylinder to
the admission pressure boosts efficiency (compressing above this
reverses the gains) and furthermore, reducing the clearance volume at
TDC also boosts efficiency. Vaccuum condensing allows full
recompression and yet allows the use of minimal clearance volume to do
so. The pressure to which you can exhaust is limited by the condenser
capability, as the steam exhausts the engine and is condensed, the
steam undergoes a drastic drop in volume and produces a partial
vacuum. The limiting factor is the cooling medium employed and
effectiveness of the heat transfer; for a pressure of 3 to 5 psia we'd
need to maintain condenser temperatures in the 150-175 degree range,
this is challenging in an automobile due to space limitations but for
a small stationary unit seems achievable.
Double acting cylinders versus single acting (DA vs. SA) is one of
those ongoing steam nut arguments. Those in love with classics will
favor DA design, and often tend to proclaim it generally superior. I'm
far less sold, it tends to depend on what you are trying to achieve.
DA engines get twice as many power pulses, so that's a plus. SA
engines can be made to run faster, can accept valves in head (more
later), need no packing or crossheads. The lack of crossheads and
packing simplifies SA construction but then leaves the problem of
moisture blowby to the crankcase which DA tends to avoid. My personal
position (which many may argue) is that there isn't a right answer
here, but that the decision depends more on other choices made in the
engine design.
For simplicity sake, we can classify Corliss valves as a subgroup of
rotary valves...they rotate in one direction then back again rather
than continuously in the same direction. Rotary valves have done well
in the past....in saturated steam engines...they tend to fail when
superheat is applied, for a couple of very good reasons. The higher
temperatures superheat brings about tends to cause uneven thermal
expansion, causing the valve to bind in the bore. The lack of
moisture in the steam makes it easier for uneven heating to occur and
exacerbate the problem. The second problem is lubrication, wet steam
tends to lubricate the valve to some degree, superheated steam is not
only dry...but it tends to decompose lubricating oils. Rotating
valves are generally problematic if we want to use good levels of
superheat. I should mention that the traditional slide valve is
equally problematic because the steam in the valve chest tends to
place great side loads on the valve which causes rapid wear due to the
problems with oil breakdown and dry steam having no entrained water.
Piston valves are far more forgiving in superheated applications which
is why we see them in the later steam locomotives, Doble steam cars,
later industrial engines and so on. Someone once commented that there
are two types of steam engine valves....those that leak and poppet
valves. The standard IC engine poppet would have a lot to commend
it...you can buy them off the shelf in any size from those used in
lawn increments to those used in giant ships diesel engines. They
need little lubrication as the valve stem travels with almost no side
loads. The poppet can be placed in the cylinder head, as in an IC
engine, and open downwards, contributing almost no additional
clearance volume to lower efficiency. Poppets have been used in DA
engines in a number of ways such as placing them in the side walls or
by employing L head configurations, but generally they would be
considered more suitable for SA applications in the sizes we are
discussing...although one of our SACA members has built steam launch
DA uniflow engines with poppets.
Boilers can be a whole can of worms, we can look at fire tubes (a drum
contains the water/steam mix and is perforated by tubes through which
the combustion gasses pass) and fire tubes (water flows through tubes
and heat applied externally generates steam). Fire tubes are easy to
control, have significant reserve steam to keep the engine running
even if the fire dies, take time to bring to pressure and
temperature. Water tubes are more difficult to control, have less
reserve, come to temperature and pressure quickly. There are
innumeral variations on the themes, most water tube boilers have drums
which store water and add reserve and simplify control issues.
Monotube boilers are also called 'once through' tubular boilers, the
water is fed into one end and steam emits from the far end, they fire
up very fast (see Doble) are difficult to control and have minimal
reserve. LaMonts use drums and a pump to circulate the water through
the coils so rapidly that only about 1 part in 5 actually boils, the
remaining water keeping the tubes cool enough to avoid damage under
very high firing rates. Some naturally circulated boilers do much the
same but take advantage of the differences in density between steam,
hot and cold water to create a circulation pattern.
There are a few general rules to boilers, you can only fire them so
hard before the metal overheats and breaks down. The firing rate is
generally dependant on the cooling provided by the water in the
boiler. Stagnant water forms steam bubbles and even regions of steam
that cling to the boiler tube walls and insulate the wall from the
water inside, leading to rapid overheating. Rapidly moving water
scours the steam from the wall and delays burnout, the greater
proportion of water to steam circulating through the boiler also makes
a difference. As both the portion of water and the circulation
velocity rise, the tendency to burnout drops and the safe firing rate
increases. It is generally harder to circulate water in a fire tube
boiler.
The pressure a container can withstand is tied both to the wall
thickness and the diameter, as the wall thickness rises, the allowable
pressure also rises. By the same token, as the diameter decreases,
the pressure which can be safely applied goes up. It's easy to see
how this works in favor of water tube boilers, as they have no need
for a larger pressure vessel.
Another point to keep in mind is that the volume to surface area ratio
changes as the tube diameter changes. A 2 inch tube has twice the
area of a 1 inch tube of identical length, but it has 4 times the
volume...so the area to volume ratio of the 2 inch tube is 1/2 that of
the 1 inch. Tube volume determines how much water must be boiled, but
the tube surface area determines how much heat can be applied to the
water, so the smaller tube is more capable at applying heat to the
water. Of course, smaller tube has less flow rating, this can be
addressed by using multiple tubes operating in parallel, but also
opens the door to other problems. If one tube in a parallel
arrangement starts to boil first, the vapor bubble creates flow
resistance which causes water to move down the non boiling
tubes...which keeps them cool and starves the boiling tube, causing it
to overheat and burn through. Such issues are addressed in a variety
of ways such as by adding enough flow resistance to each to so that
the boiling resistance in any one tube is small enough in relation to
the total so that no flow starvation occurs. By keeping each tube
length identical and exposing each to the same heat levels, the
tendency to boil early is minimized. If at some point just before
boiling the tubes all enter a header and the contents are all mixed to
obtain an equal temperature and then once again redistributed to
parallel flows we can reduce the tendency to premature boiling in any
individual flow path.
By now you may have noticed that I didn't specify an ideal boiler.
Experience indicates that such declarations are usually
counterproductive. It may be helpful, however, to consider that some
of the stuff shown on the internet may appear to lack thorough
consideration of just these basic factors let alone such issues as
active versus passive superheat control. Probably too early to worry
about specificity in a boiler until such time as the desired operating
parameters are laid down and then some kind of hard numbers can be
applied to a boiler design.
The second reference Robert pointed out is the book I mentioned.
Pages 58-62 in particular give enough data to build a workable bottle
engine, surely not the pinnacle of heat engine technology, but
workable enough on steam up to about 175 psi and very mild superheat.
Regards,
Ken
jamie clarke
Bastelmike
my belief is that a turbine would compete very well in efficiency and
power, provided the turbine is designed well.
As for safety, I don't see much difference, the turbine having a bit
more potential dangers. If the fast moving parts of a turbine
disintegrate and the case is not very strong built, you might get a
disaster. All the potential dangers with hot steam under pressure are
the same; both designs need some type of boiler.
A turbine design is much more difficult to manufacture than a piston
engine. Rotor, Blades, the case. The rotating parts have to get
balanced.
I can't imagine that "anyone" can built a working and safe steam
turbine.
Mike
On Nov 13, 7:37 am, jamie clarke <jamieclarke...@gmail.com> wrote:
> How would a small turbine compare ken in terms of ease, safety, efficiency
> and power? Im assuming they used turbines of the dwight d eisenhower, im
> guessin you would have more problems with the gearing but simplifies the
Ken
Turbines generally don't scale well. For one thing, you have a
clearance between the tubine rotor and the casing, the clearance is
fairly small on a large turbine and it becomes more progressively
difficult to close the clearance up as the turbine scales down. At
the same time, the tip speed of a turbine has to be the same no matter
what the size is in order to maintain comparable efficiency. Let's
say the spouting velocity from the turbine nozzle is 4,000 ft/second
at our operating pressure, then an 'ideal' turbine would have a tip
velocity of 1/2 that speed...the goal of a turbine is to rob the steam
of all velocity and transfer it to the rotor...if the steam left the
turbine motionless we'd have a theoretically perfect machine. Of
course, any real world turbine is going to fall short for a variety of
reasons, but we can see that the rpm is going to climb incredibly as
the turbine diameter falls because good nozzles will produce much the
same steam velocity in small or large scale. Consider a small turbine
of 1 inch diameter, the circumference is 3.14159 inches and the
desired tip speed is about 2000 ft-min, which translates to 24,000
inches per second; we are looking at something on the order of maybe
7,500 revolutions per SECOND for an efficient turbine, or about
450,000 relatively unobtainable rpm. Admittedly this is all
criminally oversimplified but probably close enough to make the point;
small turbines are wasteful. Actually, if I were going to try such a
thing, I'd probably look at simply trolling the junkyards til I found
a radial inflow turbine out of an automotive turbocharger; the SACA
Phorum has a member who built a car which years ago set the steam LSR
and he used such turbines built by Barber-Colman, as I recall the
expander shape being closer to the automotive turbos compressor wheel
than the turbine due to the differing properties of steam. This
turbine came from the old Lear bus program when they closed down and
Bill Lear cleaned house. The consensus is a bit fuzzy but generally
it is felt that the break even point between reciprocating engines and
turbines lies somewhere between 100 and 1000 horsepower.
Regards,
Ken
jamie clarke
Dentist drills rotate at 1, million rpm and they operate on compressed air so i can imagine tiny gas turbines replacing batteries or something in the future.
Im thinking of something like a turbo charger from an old sports car, obvioulsy it would be difficult if not impossible for me to manufacture and balance everything properly but if its better and turbo's are abundant then i say why not?
Mike Stone
> Fair number of issues bought up, I'll try to cover them as best as
> possible, if I fall short just whack me upside the head and hand out a
> reminder.
Whew.. lot of good information there. I've read through it about five times now. That's what happens when you ask questions of someone who knows more than you: you learn things.
Cherry-picking bits and pieces: it seems clear that poppet valves win, safety is best served by keeping the total energy within reasonable bounds, not just the pressure, and that the DA/SA question will be an interesting discussion for the future.
Not to get too deep into boiler design before it's time, but am I correct in thinking that water pipe would make it easier to keep the total (potentially dangerous) energy low? It certainly seems to have advantages WRT the size of the pipes, and the fast circulation of the LaMonts design sounded attractive.
Given the desirability of fast water flow through both the feed system and the condenser, as well as pressure for lubrication and pumping water from the condenser back into the feed system, I think it might be a good warm-up exercise to design a few fractional-horsepower engines to do those tasks. Even if it ends up being easier to pull all that power from the main shaft in the final design, I know I could use practice stumbling through the design process.
Next question, since this seems to be one of the foundation questions when defining the Rankine cycle: is there some generally preferred ratio of stroke length to piston diameter? The stroke is what converts heat energy to work energy, so the general rule seems to be "more is better." 1:1 offers the minimum surface are for heat loss through radation, 2:1 offers the maximum volume relative to surface area, and I'm sure we could find arguments in favor of just about any other number if we looked hard enough.
Bastelmike
a longer stroke will mean lower rpm to keep piston speed within
acceptable und usual limits. Because I assume the rpm of the 15hp
engine to be considerably under generators 1500 or 3000 rpm, it would
mean the need for a larger gear drive. The cylinder will be longer,
machining it will become more difficult.
So a longer stroke won't come without a price to pay.
Mike
Mike Stone
> a longer stroke will mean lower rpm to keep piston speed within
> acceptable und usual limits. Because I assume the rpm of the 15hp
> engine to be considerably under generators 1500 or 3000 rpm, it would
> mean the need for a larger gear drive.
RPM, stroke length, and area of the piston head are all related. Assuming 200 psi and 1500 rpm, you can make a 15 HP engine with roughly a 3" bore and a 3" stroke, or with a 2" bore and a 6" stroke. Given the same amount of clearance in both cases, the clearance would be a smaller fraction of the 6" stroke.
I can't imagine umpty-odd generations of steam geeks not A) settling on a ratio that works best, or B) arguing about nonstop.
> The cylinder will be longer, machining it will become more difficult.
I shouldn't think there would be much difference for ratios below about 5:1. Beyond that, I lean toward using schedule 40 or 80 pipe since it's already rated for pressure and tends to have good dimensional uniformity over moderate lengths. That reduces the problem to 'cut to length' and maybe some honing to make sure the inside surface is smooth.
Ken
good power to weight ratios...which tend to be somewhat contradictory
goals. High efficiency engines have short cutoff for full expansion,
which translates into low MEP (mean effective cylinder pressure),
which means reduced power output. The way to bring the power up is to
boost rpm, which is obvious from IC engine technology. Here is where
the differences in steam engines kick in. Short cutoff means the
admission valve is open for a very short portion of the stroke, and
thus the admission time period drops as rpm climbs. So far this is
obvious, the problem is that it seems to be generally assumed that
steam velocity is instantaneous whereas it takes time for steam to
build up the velocity needed to clear the valve; the short admission
period means that as rpm climbs we can start to choke the steam
admission to the cylinder. Further complicating the issue is
compression, if we choose to recompress the remnant steam mass to near
admission pressure to improve thermal efficiency (it does this by
partly offsetting the negative consequences of clearance volume) we
reduce the initial acceleration because the differential pressure
across the valve is very low at TDC...which means that short cutoff
high speed engines tend to experience fall off in torque as rpm
climbs. This is one of those "do as I say and not as I do"
situations, one of the features I want to test out in my engine is a
new steam admission method to permit high rpm and short cutoff without
inducing the choking or port blocking phenomenon. Since I'm still
slowly fabricating and testing is a ways off, I'm very reticent to go
into any greater detail and this is actually a bit more than I've
mentioned on the SACA Phorum.
Most engines with comparable MEP and loads have similar piston speeds,
meaning short stroke engines run at higher rpm than long stroke to
produce the same power and by extension the longer stroke engine has a
longer admission valve opening period for the same cutoff. If a short
cutoff, high expansion engine is desired it might be a bit more
advantageous to go with longer strokes so as to minimize port blockage
effects; I'd tend to think the longer cutoff engine is less sensitive
to shorter stroke and higher rpm. Remember that beyond the desired
cutoff, the type of engine also makes a difference, a DA engine has a
piston rod and cross head which adds quite a bit of reciprocating
weight that the SA engine doesn't have to deal with; a longer stroke,
lower rpm engine is going to take less of a pounding. It is also going
to be important to take the engine mission into account, if the driven
load requires relatively low rpm and high torque, the long stroke
engine is preferable to adding gearing.
Personally, I think the route to go in cylinder fabrication is to
either cast or machine a cylinder bore out of a convenient material,
purchase a centrifugally cast moly-iron sleeve, thermal fit the sleeve
to the cylinder (cutting in uniflow ports if desired) and then do a
final bore and hone on the sleeve. Some offshore companies make
sleeves of relatively questionable metallurgy, one domestic
manufacturer of which I am familiar charges a bit extra but the sleeve
quality is equal to production grade: http://www.lasleeve.com/all-purpose-sleeves.
Another simple approach is to go down to your local hydraulics
supplier and purchase hydraulic cylinder stock...a metal tube bored,
honed and polished to diameter.
An automotive turbocharger should work to make power, but probably the
using the compressor disc as a radial inflow will be more effective
than the actual turbine disc....that would be worth experimenting
with. The very high turbine speeds will require some kind of large
mechanical step down ratio to mate the turbine to a load, this might
be problematic. The turbine will probably be something on the order
of 1/4 as efficient as a decent reciprocator, so a much larger boiler
will be needed burning a commensurately larger amount of fuel. The
feed pump will be larger to pump the additional water, and will absorb
that much more power from the output, further reducing overall
efficiency. The project is do-able but I'm not sure it looks at all
practical. As a side note, hobbyists build high rpm turbines and
balance them all the time. One typical method is to clamp a hard and
polished dowel pin to a table top such that the dowel is very
horizontal. Hang the rotor from the dowel (the dowel should be a very
loose fit in the rotor ID) and tap the table repeatedly with a small
rubber mallet, or attach a mechanical vibrator. The heavy side of the
rotor will eventually migrate to the bottom and stay there. Lighten
this position very slightly and repeat until the rotor tends to stay
in whatever position it was first hung. This should provide a
satisfactory static balance...static balances being suitable for disc
shaped objects...longer objects such as drive shafts and multi throw
crankshafts need dynamic balance and this requires specialized
hardware such as I use at my place of employment.
Regards,
Ken
solar44
Sorry, but the 3x3 and the 2X6 are not equivalent. I have a quick-n-dirty spread sheet I used to compute that with all other things being equal, a 3x3 makes 15 HP @ 3260 RPM whereas a 2x6 makes the same 15 HP at 1223RPM.
The reason there is and never will be a best it that there is no such thing.
BTW; Schedule 40/80 pipe is "round" only in the sense that it is not "square" and visa-versa. However there is a common commodity known as DOM (Drawn Over Mandrel) tubing that is nearly round enough, such that only a honing is usually enough to make a highly serviceable cylinder.
Here are the cell Formulas & Headings for columns A-R
| Power HP | Press PSI | Len ft | Area in^2 | N RPM | K Cyl | Const lbf * ft * mn-1 | T hi F | T lo F | P NH3 Hi | P NH3 lo | delta P | Cutoff % | Stroke in | Bore in | Displace in^3 | Displace cc |
| =PRODUCT(B2:F2)/G2 | =M2*N2 | =O2/12 | =PI() * ((P2)/2)^2 | #N/A | #N/A | =33000 | #N/A | #N/A | #N/A | =630.22 | =286.57 | =K2-L2 | #N/A | #N/A | #N/A | =O2*D2*F2 | =Q2*2.54^3 |
Inputs are columns E, F, N,O,P Columns I&J, K&L temps & pressures for R-717 from the ASHRAE book, column H is blank.
jamie clarke
Solar you obviously know your shit.
Whats your opinion of the design we should choose for a 30m^2 solar concentrator putting out steam at a minimum boiler pressure of 200psi with an end goal of generating electricity at 3000rpm? Can you describe the quality of steam this would produce or what would be needed to get the required steam quality for generating electric?(super heat, steam pressure)
How many cylinders to get enough displacement? Judging by your previous statement 3x3 would be the stroke and bore to achieve the rpm without gearing.
If anyone else has an idea for an entire system please focus on this as the design requirements.
Ken maybe schnuerle porting could resolve your valve/choking problem if you sized the ports properly.
http://en.wikipedia.org/wiki/Schnuerle_porting
solar44
Hmmm, 30 m^2 of collector... that's a good start! Although the PV people like to cite 1000W/m^2 as their "standard sunlight conditions" I like to work at 800 because it's more realistic when accounting for clouds, haze, dust, etc... That said we can assume ~24kW(thermal) as our gross energy.
Now, with 200 PSI as a criteria, we have an implied saturated boiler temperature of just above 380F or 195C, 466K or 840R -- the latter two being very important when considering radiation losses from the solar collectors, and when we perform our sanity check against the Carnot Cycle.
The next issue we must consider is the climate of the site. Because that will determine our condenser temperature. Of course we might conclude that the value of the waste heat is greater than the value of the mechanical work that our engine will extract. In which case, we'll probably be running the computations with significantly different back-pressure on the exhaust.
The single cylinder 3x3 is certainly large enough to absorb all the energy that the above collector could deliver with a significant FOS, so we're A-OK in that regard. If I were designing a S&T engine in this range I might be talking about a 3600 RPM 3 cylinder 52.4mm bore 46mm stroke radial. But I will not be held to those specs just yet as they are back of the envelope.
Even without the heat-sink info, there are some rules we can apply to out design. In order to secure super-heated steam we have a few options. 1) we can burn fuel to supply topping heat to the steam. 2) we can divide the solar array into a boiling portion and a superheating section (this is how the S&T system works). 3) we can use a closed loop of Heat Transfer Fluid and a heat exchanger (like the big SEGS installations) 4) we can isentropically expand saturated wet steam (sorry Ken, I hope you weren't drinking anything when you read the last one.)
Dinner bell... Later guys. :^)
.
jamie clarke
jamie clarke
Erik will know best about dividing the concentrator into superheat and boiler but i think that would be straightforward enough for him to program.
What fluid would the heat exchange process involve? Im thinking helium in a turbine again uh oh.
I dont get how the isentropic expansion would heat something up?
jamie clarke
There isn't as much of traffic pollution in the locations we are visting and researching.
Mike Stone
> Sorry, but the 3x3 and the 2X6 are not equivalent. I have a quick-n-dirty spread sheet I used to compute that with all other things being equal, a 3x3 makes 15 HP @ 3260 RPM whereas a 2x6 makes the same 15 HP at 1223RPM.
Hmm.. check my math here:
IHP = plan/33000
p: pressure (psi)
l: stroke length (feet)
a: area of the piston (square inches)
n: number of strokes per minute
Now, I admit that I rounded the figures last time.. the actual bore for a 1:1 ratio is 2.932".
2.932 / 12 = .244
l = .244
2.932 / 2 = 1.466 (radius)
1.465^2 = 2.149 (r^2)
2.149 * 3.14159 = 6.751 (pi r^2)
a = 6.751
plugging those in with p=200 and n=1500:
IHP = 200 * .244 * 6.751 * 1500 / 33000
IHP = 14.975
For the 3:1 ratio the bore should be 2.033".
2.033 * 3 = 6.099
6.099 / 12 = .508
l = .508
2.033 / 2 = 1.017 (radius)
1.017^2 = 1.034 (r^2)
1.034 * 3.14159 = 3.248 (pi r^2)
a = 3.248
IHP = 200 * .508 * 3.248 * 1500 / 33000
IHP = 14.999
More generally, if we replace IHP with 15 in the base equation:
15 = 200 * la * 1500 / 33000
la = 15 * 33000 / 200 * 1500
la = 1.65
Any combination of stroke length (in feet) and piston area (in square inches) that multiplies to 1.65 should work.
Solar, I noticed that your spreadsheet multiplies the change in pressure by the cutoff percentage to calculate p. That may be where we're getting different results. Either that, or I'm missing something fundamental.
Paul Passarelli
Yes, the pressure (P) in the equation would only be the steam pressure if the valve was open for the entire downward stroke. Then the device would be a "hydraulic motor", not a "heat engine". In other words, the "thermodynamic efficiency" would approach 0%, (not actually nil, but the analysis would have to include the boiler-cylinder system) but the beast would still produce work from the mass flow (aka m-dot).
The numbers I copied from the sheet actually had the relic R-717 numbers for 180F boiler and 120F condenser but I was only trying to highlight the difference attributable to bore & stroke.
"Cutoff" is what distinguishes a heat engine. Because only after cutoff does the "heat" or "enthalpy" of the vapor actually "change" to mechanical work.
BTW, the equation I used for MEP from cutoff is grossly oversimplified, so It's not something I'd suggest you rely on.
Mike Stone
> Yes, the pressure (P) in the equation would only be the steam pressure if the valve was open for the entire downward stroke. Then the device would be a "hydraulic motor", not a "heat engine". In other words, the "thermodynamic efficiency" would approach 0%, (not actually nil, but the analysis would have to include the boiler-cylinder system) but the beast would still produce work from the mass flow (aka m-dot).
Looks like I oversimplified: the full name for 'p' in my version would be 'mean effective pressure'.. Ps * (1 + ln N) / N, where Ps is the supply pressure and N is the expansion ratio (ignoring all the complications of clearance and back-pressure).
For 50% cutoff, the mean pressure would be about 85% of delta-P. Flipping that over, to get 200psi mean pressure, you'd want about 118% of that, or a delta-P of about 236psi.
BTW - I made a copy of your spreadsheet and tried the (2.932", 2.932"), (6.099", 2.033") figures.. they match pretty well at around 8.85 HP. That extra .07" from rounding things up to (3", 3") really makes a difference.
Thing is, multiplying delta-P by the cutoff percentage gives you the final pressure but not the mean pressure.
Try this in the B2 cell:
=M2*(1+LN(1/N2))/(1/N2)
Mike Stone
> BTW, the equation I used for MEP from cutoff is grossly oversimplified, so It's not something I'd suggest you rely on.
Note to self: coffee first, 'Send' button second. Works *much* better that way.
Looks like we just got caught on a difference in MEP calculations, made worse by a big-ass rounding error.
Ken
discussion?
I wrote a spreadsheet to calculate steam engine performance and
efficiency a few years back, I set it to calculate a 3 x 3 engine
achieving 15 hp at 3260 revs. I assumed counterflow operation, a
cutoff of 25%, 5% clearance volume, release at 98%, recompression at
10%, 85% mechanical efficiency, exhaust to atmosphere. It then took
208 psi admission pressure at 500 F (115 degrees superheat) to make an
estimated 15 BHP.
Total MEP was 101 psi comprised of 117 expansion psi and 16
compression. The water rate was 17.12 lbs/hp-hr which comes out to a
thermal efficiency of about 11.7%. At release steam was 44.8 psi and
274 F. Steam recompression was to 62 psi and 483 degrees, when mixed
with the incoming steam on the next stroke the final steam temperature
was 512 F.
According to the T-V graph accompanying the output, the expansion
curve intercepted the saturation curve somewhere around 62% of stroke,
so the exhaust would be wet and the overall efficiency may be somewhat
below that calculated due to the problems of initial condensation
caused by passing cold, wet exhaust through the cylinder head of a
counterflow engine.
The spreadsheet uses the IAASP steam table formulations and calculates
temperature and pressure every 0.1% of stroke for both isentropic
expansion and compression and the MEP is basically taken as the areas
under the curve.
To illustrate the changes temperature makes, when running at the same
conditions but with 800 F steam the horspower drops slightly to 14.9
and the water rate changes to 12.7 lbs/hp-hr with a calculated
efficiency of 14.1 %, the release temperature is now a toasty 411
degrees F and carrying well over 100 F superheat, indicating we could
have improved efficiency at the loss of horsepower by shortening the
cutoff. Changing the temperature to a cool 385 F, just a fraction
over saturated, the horsepower rises to 15.75, the water rate is 19.4
lbs/hp-hr and the efficiency about 11%, undoubtedly optimistic due to
initial condensation.
Regards,
Ken
Mark Norton
- Mark
Robert Baruch
Engine Design book (on Google Books). I set up some of the equations
in the first section (on piston area) in Mathematica, then annotated
by "using my words".
The result is in this PDF: http://halfbakedmaker.org/steamengine1.pdf
Important note: the function Log[x] in Mathematica is the natural
logarithm.
At the end of the paper, I plugged in some example numbers for a 5 HP
engine. I used some numbers that may or may not make sense, because I
have no experience whatsoever, and I have no idea what's considered
normal.
Anyway, please take a look at it, and if you find any errors in my
understanding of Whitham, please let me know.
--Rob
Jason Learned
Robert Baruch
Jason Learned
Ken
Just getting home from vacation so kind of pressed for time, please
excuse any excess brevity.
It isn't really your math that worries me but rather the original
premises used by the publisher upon which your formulas are based. My
take is that the equations might promote difficulty because they don't
really take steam energy into account, but are instead based purely on
pressure. The problem is that pressure isn't the only variable
affecting steam expansion. Suppose we have a solar collector
providing saturated steam, there really isn't any way to judge just
how much energy is in a given cubic inch of solar collector output.
Saturated steam is rated by the percentage of quality, if it is pure
steam the steam quality is 100%, if it is pure water the quality is
0%, and so on for any intermediate proportions. One engine could be
running with steam at 100% quality, the other at 50% and yet the
equations treat them as identical as though the the very large amount
of energy in the form of the heat of vaporization isn't present. The
matter just gets worse if we add superheat to the equation, at the
same pressure we can have saturated steam of 50% quality and
superheated steam with 500 degrees F superheat; without taking
enthalpy into account the calculations aren't going to be very
representative of the actual vapors. At the time that work was written
superheat was still pretty 'out there', most boiler pressures weren't
all that far apart and the boilers tended to be large kettles heated
by tons of fuel, all of which made for relatively certain steam
conditions...and I question whether the current climate is likely to
make it feasible to assume that kind of uniformity in steam condition.
Regards,
Ken
mekennedy1313
engines could a large one be adapted as a steam valve. Don't really
know how an injector works but from what I see online it is a highly
reliable solenoid valve. Since they are designed for thousands of rpm
they should be reliable with a possible exception of high temperatures.
Ken
inch. The Sarco Spirax website steam tables give the density of steam
at 150 psig and 500 F as .299614 lbs/cubic foot...dividing by 1728
makes it about .000173 lbs/ cubic inch. The gasoline would seem to be
about 150 times as dense as the steam, so it would take a much bigger
fuel injector to deliver comparable flow, not sure you could even find
a direct port injection unit that size. Some folks on the SACA site
have discussed this recently, but instead of fuel injectors they were
looking at solenoid intake and exhaust valves being developed for IC
use. The problems start to come at short cutoff, the cycle time can
get very brief compared to IC engines which keep the valves open for
almost an entire stroke, and the larger valves have greater mass
making the whole evolution more difficult. The IC people have come a
long way since I first heard a steam nut talk about this some 12, 13
years ago so I would expect they will reach the magic number in the
bye and bye. One thing to keep in mind is that steam pressure tries
to push an admission poppet valve open (unless you use a reverse draft
angle on the poppet head or employ a balance piston); this force may
require a somewhat stronger valve steam and spring, both of which may
put a bit of additional load on a solenoid valve. Overall it isn't a
bad idea, though, and way back when some large, slow indiustrial steam
engines actually used solenoid valves and electric variable cutoff
valve gear.
Regards,
Ken
Max Kennedy
It can be done
Ken Helmick
Mode of operation is an issue, injectors work because the fuel is highly pressurized relative to the cylinder, thus permitting high mass flow. The last thing you want in a steam engine admission valve is a significant pressure drop across the valve, the drop in pressure is called a 'throttling' loss...and one reason that steam engines often change power settings by varying cutoff rather than by throttle. So even a large injector may be on the small side because it is meant to work with much more pressure across the valve.
Robert Baruch
were. I found another Google Book at http://books.google.com/books?id=n6dKAAAAMAAJ,
Creighton's The Steam Engine and Other Heat-Engines, which also gives
a treatment of steam quality and mentions superheating. It's less of a
design book than Steam Engine Design is.
Metacomment: the reason I'm looking for older works on the subject is
that although their understanding was less than complete, their
understanding was also simpler. I'm hoping that by understanding the
earlier explanations (as long as they aren't totally wrong), I can
bootstrap myself into understanding the later explanations. Also, in
the end, I'm trying to build a steam engine that does approximately
what it needs to do without exploding. I don't really need to figure
out why it's 10% less efficient than expected. That's what fudge
factors are for :)
--Rob
Paul Passarelli
We all "know" that liquids are incompressible... However as a general rule of thumb, we should think of liquids as being "precompressed" fluids (liquids or gasses) at about 2GPa. In other words highly resistant to more compression.
That incompressibility is why the Diesel can apply a "signal" to the injector that is reliable & accurate. The closest thing to the injector that may be being envisioned is the "bump valve" that uses the piston to open the port causing admission. Someday when we have indestructible materials, that solution would be just fine, but the wear suffered by steel and or ceramics means trouble.
Robert Baruch
diagram from the Creighton book showing the flow of energy in a steam
engine. You can label all the flows with their BTUs and temperatures
for a quick, easy-to-read overview of a given engine. You can cross
out the pieces you don't have, like the Return from Jacket and the
Fuel Pump.
Illustrator format: http://halfbakedmaker.org/heat-diagram.ai
PDF: http://halfbakedmaker.org/heat-diagram.pdf
Ken
Hi Rob,
Mark was nice enough to put up a page for me to post to, thanks very
much, Mark! I decided to test it out (not realizing how frustrating
that would be) by sticking up some screen shots of my steam cycle
calculation spreadsheets. There are 4 sheets, three of them for the 3
x 3 engine of 15 hp discussed previously with saturated, 115 F and 415
F superheat, along with calculations for my automotive test cylinder.
The first portion of each image shows the data input boxes to the left
and the calculation results to the right; immediately followed by
Pressure-Volume and Temperature-Volume diagrams.
https://sites.google.com/site/opensourcesteam/ken-s-page/steam-cycle-calculator
This spreadsheet is probably massive overkill for what you want to do
as it takes up a lot of time analyzing the curves in fine detail and
examines the consequences of recompressing the remnant steam mass and
mixing that with the incoming steam. Recalling how the spreadsheet
works, however, I think I can figure out a very streamlined method you
could use. The basic formulas would be from early 20th century thermo
texts, but there are some modern spreadsheet addins that can supply
data more accurately and rapidly than the old steam tables. By using
the 'Goal Seeking' function in Excel it should be possible to use
these formulations to derive the pressure and enthalpy at release
based on relative cylinder volumes and assuming isentropic expansion.
From there the whole thing sort of writes itself. Might take me a
couple days to dig the basic data out and make sure I'm not talking
through my hat...but the current spreadsheet uses the same basic
approach and its results seem to jive with other sources.
Regards,
Ken On Nov 16, 10:41 am, Robert Baruch <robert.c.bar...@gmail.com>
wrote:
Ken
to delve back into my steam technology patents database; suddenly
remembered the Fickett engine:
http://www.google.com/patents?id=vDJkAAAAEBAJ&zoom=4&pg=PA1#v=onepage&q&f=false
First off, I'd probably never consider the hydraulic valve operators
used on this engine, there are reasons we have camshafts and rocker
arms. I'd also move the spring out of contact with the steam, it
takes some very tricky and costly springs to withstand superheat and
still stay springy.
The Fickett engine has some very nice features worth studying:
1. It's a uniflow. Uniflow engines tend to be more efficient than
counterflow because the cold exhaust steam is ejected through the belt
in the middle of the cylinder rather than passing through the head and
cooling that
off just before the next load of steam passes through.
Worth noting...pure uniflow engines compress the remnant
steam mass and negate some efficiency loss due to clearance volume.
Downside is that compression reduces power while improving efficiency
and
limits low throttle pressure operation...admission,
throttle and exhaust pressures must be carefully chosen or auxiliary
clearance volumes or auxiliary exhaust valves considered.
2. The overhead poppet valve. They did things right, the poppet
doesn't significantly add to the clearance volume. The poppet lifts
to open rather than drops, thus the spring does not have to be massive
to offset admission
steam pressure. The lifting poppet also relieves any
overcompression back to the steam chest, thereby reducing efficiency
losses due to over compression AND functioning as a safety valve in
case of condensate
buildup in the cylinder. The force needed to open the poppet
against steam pressure is minimized because the cylinder compression
counters the closing forces provided by the steam chest pressure.
3. Steam chest. The designer made the steam chest wrap around the
upper part of the cylinder, thereby adding a free steam blanket to the
upper cylinder and reduing the likelihood of initial condensation.
Not covered are the crank bearings. One of the problems with SA
engines is that blowby past the piston rings can mix with the oil in
the crankcase and form an emulsion that looks and feels about like
mayonaisse to the obvious detriment of lubrication. A good grade of
synthetic oil helps overcome this problem. Some engines, such as
those of Jay Carter and Peter Barrett employed centrifugal purifies to
both remove water from the lubricating oil but to also remove cylinder
lubricating oil from the feed water...that oil can decompose in the
boiler tubes and the resulting insulating layer can lead to
overheating and tube burnout. Other designers have simply routed a
small steam line through the crankcase and added just enough heat to
the oil to vaporize the water away.
Having mentioned Jay Carter, might as well bring up his valve
briefly. Back in the 70s Jay and his son (inventor of the
Cartercopter) built one of the most successful steam cars ever, the
engine used a bump valve for very short cutoff and high efficiency.
Jay had longevity problems with the valve but through material choice
and careful weight reduction managed to come up with a valve that
lasted well...
I bring this up because bump valve engines tend to have optimum
running speeds, an engine tuned to run at the right speed might be a
nice choice for a generator...relatively simple, efficient and stable.
http://www.google.com/patents?id=kBY7AAAAEBAJ&zoom=4&pg=PA1#v=onepage&q&f=false
Regards,
Ken
Robert Baruch
That's awesome, and very nice of you. I suppose we could always try
Google Docs and sharing. Whatever I can't get through Excel, I can
most likely put together using Mathematica...
--Rob
Ken
Glad to help. I was using a shareware Excel spreadsheet called X-
steam, but they don't seem to be around any more. I still have all
the embedded Excel add-ins, however, and it's easy enough to send the
spreadsheet as an attachment...maybe just post it to the site Mark was
kind enough to provide. For the basic power and efficiency formulas
I'm referencing "Elementary Thermodynamics", Virgil Moring Faires, ME
& MS, MacMillan, Fifth Printing, 1943.
My approach is to assume isentropic expansion, because that gives us a
known figure at the end of cutoff and at the beginning of release. It
also calculates density for temperature and pressure, and we know both
of these variables at admission and can easily calculate density at
release by calculating the change in cylinder volume. The X-steam
software calculates a number of variables including density as a
function of pressure and entropy. Since we already know the density
and entropy at release, we can use the Goal Seek function in Excel to
find the pressure which causes the density to equal the calculated
value after expansion...I do this with a button and a macro. Between
this sleight of hand and the X-steam functions we end up with
beginning and ending pressure, enthalpy, volume, entropy, specific
volume and so on. Just need to plug the data into standard textbook
formulas for incomplete expansion to derive the figures for an ideal
engine output and efficiency.
I did a rough alpha test on lunch break, the answers are similar to
the calculator screen shots I posted earlier, but not identical.
Steam expansion follows an almost polynomial curve and I've determined
the polynomial coefficient changes throughout the curve, so it's
unlikely that any rule of thumb calculation is going to develop the
same equivalent MEP. The screen shots break the curve into 1000
expansion and compression points and integrate the results to find
MEP, so that's probably a bit more accurate than the lunch time
results...but let's not discuss the huge difference in complexity or
file size.
I need to do a bit more work to sort out the efficiency calculations,
arrange everything into easily read format and add some more coherent
labels and explanations so that my work can be double checked without
excess trauma,
Regards,
Ken
Ken
I finished a spreadsheet up last night, seems to be working OK. The
efficiency and power numbers are high for a number of reasons besides
just not accounting for cylinder heat losses and friction. The
approach doesn't really simulate the polytropic nature of expansion
and this will lead to overly optimistic MEP. The spreadsheet doesn't
figure in polytropic compression of the steam remnant mass. It also
doesn't take into account the mixing of high entropy admission and low
entropy exhaust steam in the clearance volume. For uniflow and other
high compression engines, a more elaborate sheet is called for.
I don't see a way to post a file on Google, the Excel spreadsheet
won't work if uploaded into Googledocs due to the steam function add-
ins. I posted a copy to the Steam Automobile Club of America website,
should be downloadable from there:
http://steamautomobile.com/phorum5214/read.php?2,18956
The original version is a tiny bit too big to meet the SACA upload
criteria, so tried a different format. I hope this one has all the
add-ins included. Let me know if there are any difficulties and we
can arrange to e-mail directly.
Also feel free to comment on content, methods of calculation, desired
features and so on.
Regards,
Ken
Ken
E-bay. I've seen his trike run, the engine did OK.
http://www.ebay.com/itm/250929205669?ssPageName=STRK:MEWAX:IT&_trksid=p3984.m1438.l2649
He used a 2 cylinder air compressor (you can get them for a couple of
hundred bucks at Harbor Freight) and fitted new cylinder heads and a
cam.
Not a super high performance unit but more than capable of operating
from a solar collector and capable of driving a modest gen set.
Regards,
ken
Chuck
The steam effort over at Open Source Ecology got me started thinking
about this sort of thing. At OSE, they have been using quite a few
hydraulically-powered tools, and that biased me towards investigating
a direct steam-piston-to-hydraulic-piston design. I just put up a page
here: http://opensourceecology.org/wiki/Steam_Engine_Design/Free_Piston_Steam_Hydraulic
.
I made a simulation model of this engine using the electronics design
tool SPICE.
Whether or not the free piston design is of interest to this group,
you may find the SPICE simulation approach helpful.
Cheers,
Chuck
Max Kennedy
Ken
piston, they are also very heavy and operate at minimal velocity. A
reciprocating engine, even one of modest speed, is much faster and the
weight of a ram would be prohibitive. It would make much more sense
to just convert an economical off-the-shelf air compressor than
attempt to fit a variety of disparate elements; I think it would be
faster, cheaper and more certain. A compressor would give you the
entire engine from the motor mount all the way up to the top of the
cylinder block deck...just need to fit a cylinder head. If using
poppet valves, off-the-shelf valves, spings, keepers and bushing could
be obtained cheaply from Briggs and Stratton or other makers. The
valve gear (steam lingo for valve timing mechanism) can be based on
eccentrics and therefore made from off-the-shelf round stock with no
special cam grinding. This sort of thing has been done many times
with old refrigerator and A/C compressors, small engines and the
like. There are enough Third World manufacturers producing
compressors that the cost of a finished cast iron machine is less than
you could have the parts cast in a job shop, let alone the
accompanying machining, bearings, rings, nuts, bolts fittings and so
on.
Wally Woods Rules for Commerical Artists:
1 Never draw what you can copy
2. Never copy what you can trace
3. Never trace what you can cut out and paste down
Regards,
Ken
2.
Regards,
Ken
On Nov 21, 9:18 am, Max Kennedy <mekennedy1...@gmail.com> wrote:
> Hydraulic pistions are designed for high pressures and good seals, would
> they be a candidate for modification into steam pistons? Different seals
> would obviously be needed.
>
> Max
>
>
>
>
>
> On Mon, Nov 21, 2011 at 7:01 AM, Chuck <cfh...@gmail.com> wrote:
> > Hi folks,
>
> > The steam effort over at Open Source Ecology got me started thinking
> > about this sort of thing. At OSE, they have been using quite a few
> > hydraulically-powered tools, and that biased me towards investigating
> > a direct steam-piston-to-hydraulic-piston design. I just put up a page
> > here:
> >http://opensourceecology.org/wiki/Steam_Engine_Design/Free_Piston_Ste...
> > .
>
> > I made a simulation model of this engine using the electronics design
> > tool SPICE.
>
> > Whether or not the free piston design is of interest to this group,
> > you may find the SPICE simulation approach helpful.
>
> > Cheers,
> > Chuck
>
> --
> It can be done- Hide quoted text -
>
> - Show quoted text -
Russell Philips
motor mount all the way up to the top of the cylinder block
deck...just need to fit a cylinder head.
Would a generic air compressor lower handle:
'pressure issues' ?
'heat issues' ?
This would be a big step forward if there are no issues? (anything -
anyone?)
jamie clarke
Most air compressors can handle 200psi, what psi are we thinking?
Ken
what saturation temperature. If you're really serious about solar
production, you are going to have a rough time hitting the kinds of
temperatures that people in SACA, or running Live Steam locomotives,
would consider trivial. Say you figure you can generate steam at
about 350 F and still have enough BTU left over to provide some
superheat, say a couple hundred degrees, that puts you in the ballpark
of 150 psi.
This is a fairly good ballpark, you can generate some reasonable level
of power for given displacement without too horrendously bad
efficiency. Just don't expect to be challenging a diesel engine, or a
lawn mower, for conversion efficiency.
At 150 psi the engine is fairly simple. Not a lot of reason for
complexities like compound operation (though a 2 stage compressor has
compounded cylinders), but a 1 or 2 cylinder single stage compressor
would do your needs with adequate valving. At these pressures the
cutoff need not be that severe, so valve gear issues are relatively
easily managed. This pressure, by the way, puts you pretty well in
the middle of the pack for historical light steam power. Any number
of old elevators, winches, pumps, launches, machine shops and so on
were right in this area. Now, if I were to be burning gasoline...or
using scads of power like an automobile...I'd be running up a 1000 psi
and 1000 F...or higher in an effort to push thermal efficiency towards
20%; just as an illustration of what steam conditions you commit to
when trying to become competitive with IC.
Regards,
Ken
On Nov 22, 12:15 pm, jamie clarke <jamieclarke...@gmail.com> wrote:
> Wish i knew the answer, ive seen several versions of this but can't tell
> you about heating issues.
>
> Most air compressors can handle 200psi, what psi are we thinking?
>
> On Mon, Nov 21, 2011 at 9:08 PM, Russell Philips <russellphil...@hotmail.com
> > > > - Show quoted text -- Hide quoted text -
Mark Norton
- Mark
Chuck
combustion engines to reduce heat transfer to piston crown and
cylinder head. I wonder whether it is worthwhile in a steam engine.
Especially in a lower-rpm engine, I would think there are
thermodynamic benefits.
Insulating the cylinder walls, too, would be wonderful but I think
that in the "ceramic engine" craze of the 80's they never found the
right materials for that.
Speaking of cylinder walls, though...some of the nickel composite
platings like Ni-P-BN(h) look interesting.
Cheers,
Chuck
Jamie Clarke
Sent from my iPhone
Ken
time, money and are talking about a powerplant consuming enough energy
to make for a reasonable ROI (Return On Investment, note that
engineers and accountants often work hand in hand). On the other
hand, the old timers understood the need to preserve heat, so they
invented lagging. Early lagging was little more than wooden strips
held to the cylinder, barrel fashion, with brass or german iron
bands. Then came the asbestos years, fortunately I missed those when
in the navy. There are all kinds of lagging solutions out there, for
different budgets, purposes and so on. Last tour I did with the navy
reserve was in the Lagging Shop at the San Diego shipyard.
Legal requirements have no pat answer, they vary widely from state to
state. In Michigan automobiles are exempted. The USCG inspects
vessel boilers, and if not for hire and below a certain size they take
a pass. First thing to do is simply check the Yellow pages and find
the local state certified boiler inspector and have a discussion as to
what criteria he uses and what you need to do to pass; buddy of mine
with a live steam locomotive fretted up a storm then after talking
with the guy could only comment on how easy he was to work with. I'm
automatically suspicious of trying to adapt other pressure vessels for
steam use, so you're on your own with SCUBA tanks. As I stated in an
earlier post, the key to safe operation is minimum energy storage, a
pound of hot pressurized water being more deadly than an ounce of hot
pressurized steam. Hartley O Baker built automotive boilers with a
large diameter pipe bent into a helical coil which surrounded the
boiler, made a decent storage drum and since the needed wall thickness
goes up with diameter, was cheaper and lighter than a drum; just a
thought.
Regards,
Ken
jamie clarke
:)
To keep things simple:
Are the most dangerous pipes the biggest ones?
jamie clarke
Whats a safe wall thickness if your likely to make a mistake and get burnt bad?
Does anyone reckon this could this be adapted to superheat helium? It would be more leaky, thats one issue i can think off.
Thinking outloud, correct me if im wrong:
So the point is not to make a pressure vessel that can store tons of pressure like a scuba tank would
But to allow the fastest flow through to reach superheat so you need the least amount of water in there at one time.
Helices are good for reducing drag and increasing area. Drag rises to the 8th power of the diameter i think.
Ahh, pictures speak a thousand words.
Im assume you need a pretty nice pipe bender to get one of those boilers, cant beat a good old helix though :)
Ken Helmick
From: jamie clarke <jamiecl...@gmail.com>
To: open-source-steam <open-sou...@googlegroups.com>
Sent: Wed, Nov 23, 2011 1:18 pm
Subject: Re: Steam Engine Design: General Use
Jamie Clarke
Sent from my iPhone
Ken Helmick
Russell Philips
had a round hole for an intake valve, how big would it need to be? And
would the exhaust valve hole need to be bigger?
Any rough estimates?
gosh that just sounds bad... Russell you're such a valve hole -ha!
On Nov 23, 3:56 pm, Ken Helmick <kena...@aol.com> wrote:
> Hi Jamie,
>
> The video isn't an instantaneous thing...first I have to cross the state and do so when Tom is actually doing some bending...then I'd need to figure out how the hell to get a YouTube account....then I'd need to successfully obtain an account and upload video without becoming so frustrated that I end up in a tower with a rifle. I'm a simple kinda guy with limited skill sets.... like hunting, fly fishing, tool and die making and vector analysis in multiple planes; social media tends to exceed my tolerance thresholds....
>
> Yeah, the Goldilocks analogy is right. Notice that increasing the circulation ratio decreases the likelihood of burnout at a given firing rate but increases the total stored energy in the system, TANSTAAFL, There Ain't No Such Thing As A Free Lunch. Anyhow, we're not talking about the level of performance of even a 100 year old steam car; stationary outfits can give up some compactness and performance in exchange for simplicity and reliability. Part of the reason I have been typing these small essays is because this is the sort of background that was fairly common a century ago, and much less so today; no need to invent mistakes that great grandpa would have avoided on sheer instinct.
>
> Regards,
>
> Ken
>
> -----Original Message-----
> From: Jamie Clarke <jamieclarke...@gmail.com>
> To: open-source-steam <open-sou...@googlegroups.com>
> Sent: Wed, Nov 23, 2011 3:25 pm
> Subject: Re: Steam Engine Design: General Use
>
> Ok so it's a goldy locks kind of thing.
>
> Not too fast, not too slow with the flow and Too much steam is safer then too much hot water.
>
> Do wish I was a better listener now, it's got to be good to ask though, where else would I get an answer like that.
>
> Got to see this pipe bender! What's your YouTube account?
>
> Sent from my iPhone
>
> On Wed, Nov 23, 2011 at 5:59 PM, jamie clarke <jamieclarke...@gmail.com> wrote:
>
> Cheers ken, excuse my ignorance. Learning on the job
>
> :)
>
> To keep things simple:
>
> Are the most dangerous pipes the biggest ones?
>
> http://www.virtualsteamcarmuseum.org/images/vscmimages/Stewart,%20H%2...
>
> Regards,
>
> Ken
Mike Stone
Ken, what's the steam fraternity's opinion of pot metal for castings?
For those getting used to the jargon, 'pot metal' is a loosely-defined family of alloys made mostly of zinc with some aluminum and other metals thrown in. It's popular for castings because it melts around 750* F, as opposed to 1200* F for aluminum and 2100* F for cast iron. It takes less equipment and skill to cast pot metal simply because there's less energy in the molten metal.
Better yet, the stuff is *everywhere*. A good portion of the castings you'll find in any scrapyard will be pot metal, so it's an easily scavenged and recyled material.
Of course those benefits are worthless if the material can't do the job.
Mike Stone
While we're on the subject, I'll float the idea of lost foam casting:
http://www.buildyouridea.com/foundry/lost_foam_howto/lost_foam_howto.html
It's a variant of lost wax casting, using styrofoam rather than wax. You make the shape you want out of foam, give it a thin coat of plaster to get a good surface, bury it in sand, then pour the molten metal right into the plastic. The heat of the metal vaporizes the foam and the metal flows into the void left behind.
It can't begin to compete with pattern casting at the industrial scale, but it offers some advantages for short runs and one-off work. You don't have to plan a pattern in a way that allows you to lift the pieces out of the sand, and you don't have to ram the actual mold so you can lift out the pattern but still assemble the mold for casting. Both of those are high skills and I admire the people who do them well, but there is a bit of a learning curve.
Ken
Hi Mike,
I've never heard of anyone using pot metal in a steam engine. Pot
metals are generally considered 'junk' as far as any engineering
standards are concerned, there is no way to discern strength,
resiliency, cyclical fatigue or much of anything else...every pot
varies immensely from another. By contrast, if you have something
cast out of AlMag or Tenzalloy, you know what you're getting. Also,
hard to envision pot metal in 1000 F steam car cylinder heads :-).
Then there is the entire issue of corrosion, they make sacrificial
anodes out of zinc, bolting it up to another metal is just asking for
the zinc to corrode away. I have seen tough zincs, but generally they
are on the expensive side and used in precision die casting because
they go superplastic when the melt is injected under high pressure.
For those unsure about why you'd want to cast parts, or what patterns
are, I slipped some photos of castings and patterns of my engine
project at:
https://sites.google.com/site/opensourcesteam/ken-s-engine-patterns-and-castings
I hate to disagree with you about lost foam, but General Motors uses
lost foam to cast some cylinder heads in mass production that just
can't be made otherwise..so it is a process to take seriously in mass
production. Rather than make fuel injector rails from all kinds of
pieces of tubing and connectors, then soldering all that stuff
together and then assembling that to the head; it is cast in one
piece integral to the cylinder head itself using lost foam. Nice
thing is that you can cast the impossible, like negative draft
angles. The foam patterns are made in injection molds, and the
various parts are glued together to make parts with features that
can't come from any open and shut casting process. Then coat with
refractory and sand cast.
As you can see from the photos in the link above, I've had some
experience with patterns and casting (also fabricating injection
molds, which skills transfer nicely). The advantage is that you can
make very complex shapes and cast them with a reasonable degree of
faithfulness depending on various foundry limitations. It's very hard
to cut foam to any close design...unless you pay more and get a higher
density foam with a fine grain structure. We have some guys at work
who do that, use machine tools to cut foam accurately for protoype
lost foam castings. That foam is nothing like the stuff your new tv
is packed in. The results are very impressive. Only downside is that
if you want to make 2, 3, 4 or more you have to make a new foam
pattern each time. With conventional pattern work the patterns are
more time consuming but given careful handling hundreds or thousands
can be built...more if you go to metal. As they say, you pays your
money and you takes yer chances!
Regards,
Ken
Mike Stone
> I've never heard of anyone using pot metal in a steam engine. Pot
> metals are generally considered 'junk' as far as any engineering
> standards are concerned,
About what I figured. Double check me on this though: what do you guys (GM) use for, say, the body of a carburetor? I'm sure it's spec'd to a fare-thee-well, but I don't know the exact metal.
> Also, hard to envision pot metal in 1000 F steam car cylinder heads :-).
That was my big question. Steam engines get HOT.
> https://sites.google.com/site/opensourcesteam/ken-s-engine-patterns-and-castings
Very nice. Do you do your own ramming and pouring?
> I hate to disagree with you about lost foam, but General Motors uses
> lost foam to cast some cylinder heads in mass production that just
> can't be made otherwise..so it is a process to take seriously in mass
> production.
I didn't know that. I knew that some manufacturers had done research into reprap-style "extrude a zillion layers of foam then see if it's close enough to tolerances to be useful" stuff, but didn't know it had gotten mature enough for production. Cool.
> Nice thing is that you can cast the impossible, like negative draft
> angles.
Pockets, openings between surfaces, voids that would take complicated cores. Obviously I'm a fan.
> As you can see from the photos in the link above, I've had some
> experience with patterns and casting (also fabricating injection
> molds, which skills transfer nicely).
Again I say: 'wow'. Injection molds are just about the bitchiest branch of machining I've ever heard of.. thermal expansion issues, seating and matching issues, surface finish issues, etc, etc. Never done it myself, probably wouldn't last a day at it if I tried, and the two men I knew who could talk casually about it were both master machinists.
> With conventional pattern work the patterns are
> more time consuming but given careful handling hundreds or thousands
> can be built...more if you go to metal. As they say, you pays your
> money and you takes yer chances!
I grew up in Davenport.. John Deere, Alcoa, and Litton territory (not to mention the Rock Island Arsenal). Our next-door neighbor was a pattern maker, so I grew up with a deep respect for guys who make a living whittling to close tolerances. ;-)
Russell Philips
Be sure to scroll down to version 1.5
Max Kennedy
Mike Stone
> If I interpret the drawing right then in v1.5 you have the bearings, blue boxes with black dots, riding on the cylinder shoulder. in that configuration how do you compress the stuffing in the stuffing box to achieve a seal?
I think it's a drawing of the cylinder head itself. The cylinder would sit below the drawing, held in place by the bolts (vertical black bars at either end).
A question of my own: It looks like the pink area is the supply steam and the red area is the exhaust port. Figure 1.5 seems to show a direct connection between the two, so I'm probably reading that part wrong. What are the intake and exhaust paths?
jamie clarke
:)
Max Kennedy
jamie clarke
Im also assuming the elongated valve section is a ring that is added to keep the valve sealed shut and avoid expansion and contraction problems.
Even if im wrong i think im getting the jist of it now.