I have thought of using a Stirling to pump water for a hot water heat
system. The water would be the hot side at 175-200 F. and the outside
air would be the cold side. Could it be self starting? What do you
think?
LTD engine for pumping hot water. Self starting?
Moderator: stan.hornbaker
Forum rules
Be nice!
Be nice!
-
shyflowerdancingjesusgirl
- Posts: 3
- Joined: Tue Jan 21, 2003 6:58 pm
- First Name: Danny
- Last Name: Downs
Response to LTD engine for pumping hot water. Self starting?
In my opinion, a Stirling engine can be made self-starting by using 2 sets of pistons or piston/displacer combinations, placed under 90 degrees, using either a common cranq shaft or e.g. a chain to connect the shafts. Provided that the fly wheel has enough momentum this should always be self-starting. With a temperature difference of only about 100 F it should be huge if the engine is supposed to pump anything at all, though...
Response to LTD engine for pumping hot water. Self starting?
Hi Rob, how huge it would be? Any idea?
-
archimerged
- Posts: 1
- Joined: Fri Sep 22, 2006 7:47 pm
- First Name: Archimedes
- Last Name: Submerged
Response to LTD engine for pumping hot water. Self starting?
Stirling engines generally are not self starting, because the cycle involves investing some initial energy in compressing cold gas and harvesting more energy by expanding hot gas.
A nice self starting pump might be based on Minto's "wonder wheel"
which uses propane or other refrigerant in paired tanks. Heat at the bottom boils the refrigerant, which rises through a tube to the top of the wheel where it condenses slightly off center into a tank.
However, this might be a little slow pumping. You would need a
system to adjust the amount of water being pumped to the available
torque so that the wheel always turns, and if it stalls, the load
gets reduced.
Google for "propane wonder wheel" which or try e.g.,
http://www.thefarm.org/charities/i4at/lib2/rotate.htm
A nice self starting pump might be based on Minto's "wonder wheel"
which uses propane or other refrigerant in paired tanks. Heat at the bottom boils the refrigerant, which rises through a tube to the top of the wheel where it condenses slightly off center into a tank.
However, this might be a little slow pumping. You would need a
system to adjust the amount of water being pumped to the available
torque so that the wheel always turns, and if it stalls, the load
gets reduced.
Google for "propane wonder wheel" which or try e.g.,
http://www.thefarm.org/charities/i4at/lib2/rotate.htm
-
privacy.lover
- Posts: 3
- Joined: Fri Jun 01, 2007 2:29 pm
- First Name: Inventus
- Last Name: Privaticus
Response to LTD engine for pumping hot water. Self starting?
Hi,
Fist of all, I must agree with Robs assertion, that the most straightforward way of making a (conventional) Stirling engine self-starting, is to simply use more than one cylinder. The theory here is the same as for steam engines, namely that by offsetting the phase of each piston (by 90 degrees, in case of two cylinders), one of these will always be able to turn the flywheel (or whatever) when the other is at its "top" or "bottom" point (and thus cannot move the flywheel), and thereby allowing the engine to start regardless of which position in which it was stopped.
There is one snag to this though: If it so happens that your Stirling engine were to be heated very slowly (ie. if, for instance, it where to operate on solar power, where the sun will slowly raise its temperature during the early morning hours), this may well prevent it from self-starting, even if it has plenty of cylinders. The reason for this being that, unless special care has been taken to prevent it, such a slow heating process would allow the heat to disipate to the cold side of the engine. This could then prevent the necessary temperature difference, required for the engine to run, from ever occuring. However, a well insulated and otherwise properly constructed engine should prevent this from happening, at least within the relevant time-scale (in case of a solar-powered engine, this would be a couple of hours, not very difficult to achieve). Since a well insulated Stirling engine is also a more efficient one, this would be a design goal anyway (but still, something to keep in mind, at least if one isn't sure that the heat will always arrive quickly at the hot end of the engine).
= = =
With regard to the comments from Archimedes:
I beg to differ about his conclusions. Apprantly he suffers from a few misunderstandings with regard to the operational principles behind Stirling engines (and/or about thermofluid theory in general). It would appear that he is, among other things, confusing some details about how Stirling engines work, with how some internal combustion engines work, specifically those operating according to the Otto and Diesel cycles (aka. ignition and compression engines).
The piston of a Stirling engine is moved by the working-gas being repeatedly compressed and expanded, that much is true. But NOT like the way you see it in the aforementioned internal combustion engines, which DO have to "overcome" their compression strokes (as well as the intake and exhaust strokes, for four-stroke engines). Rather, the "compression" of the working-gas in a Stirling engine, which occurs when the gas is cooled, aids the piston-movement by providing an under-pressure (ie. "sucking" the piston in the right direction), just like the "expansion", which occurs when the gas is heated, aids the piston movement by providing an over-pressure (ie. "pushing" the piston, also in the "right" direction since this occurs when the piston is moving the other way).
This is in fact one of the reasons why the Stirling cycle is more efficent than the Otto and Diesel cycles, as the engine "harvests" energy on every stroke, rather than every other (for two-stroke engines) or every fourth (for four-stroke engines) as is the case for the Otto and Diesel engines (which then have to use part of this energy to keep running through the remaining cycles, where notably the compression-cycle needs a significant "investment" of energy).
As such, the only reason why most (single-cylinder) Stirling engines have a flywheel at all, is the same as why a steam-engine has one, namely to get past the "dead" parts of their strokes (ie. when the piston is at the extreme ends of its motion). Conversely (single-cylinder) internal combustion engines need their flywheels to keep turning, not only through these "dead parts" of their power-strokes, but also through the remaining stoke(s) (one for two-stroke and three for four-stroke engines). The extreme case being a four-stroke, single-cylinder Diesel engine, which not only have to produce enough power, during its single power-stroke, to overcome the dead parts of this stroke, but also through the complete intake and exhaust strokes, and most notably its compression-stroke (which will typically be VERY hard to overcome, since a Diesel engine needs a very high compression-ratio in order to work at all, and an even higher one to work efficiently). This is why you will find a HUGE flywheel on such engines (which can sometimes be found in small fishing boats - You will know them by their distinct, slow "tuck-tuck-tuck" sound). Such engines often incoorporate a pressure-release valve, which needs to be engaged before the electrical starter can turn the engine around (or, indeed, if you want to have any luck starting it by cranking it manually!)
Just like any other engine-type (Rankine/steam, Otto/ignition or Diesel/compression), the Stirling engine have no need for a flywheel when equipped with more than one cylinder (or rather, more than one piston, since you could make an engine where two pistons share a single cylinder). In real-life engines (of any kind), a small, light-weight "flywheel" will be present more often than not, for various reasons, notably other than keeping the engine going (thus making the term "flywheel" a bit of a misnomer, although the name has stuck).
Oh, and never mind about the "wonder-wheel", this is NOT an efficent engine-design, at least not in its current form (few people have made them work at all, and those who have seldom achieve more than one revolution every other minute at best, in most cases they achieve only a few revolutions per hour!) Personally I don't think this concept will ever prove to be good for anything, an opinion supported by the fact that its use is mainly advocated by crackpots and people who generally fail to understand the underlying principles.
Finally, I'm not at all sure that you would necessarily need to "adjust the amount of water pumped to match the available torque", or rather: This would occur naturally, given how pumps work (ie. if/when the torque is lowered, the speed of the engine, and thus the pump, would be reduced which would also reduce the amount of water pumped). What you might like to be able to adjust (only, in most situations, including yours, you can't) is the "head" (the water-pressure against which the pump has to operate), which would allow you to keep the pump working at a certain speed and thus keep the pumped volume per hour constant (in you case, this would require you to bypass part of the water-circulation pipes, in order to lower the pressure required to keep a certain flow-rate, which would unfortunatly cause the affeced parts of your home to be without heat).
(I'm sorry to "break your bubble" Archimedes, but you really should make sure to know your thermofluid theory, before posting such comments!)
= = =
Back to the main point:
Since you want to pump water, rather than "rotational movement", I would suggest that you skip this step altogether, and consider a "Stirling-pump". This is to say, make a Stirling engine that operates directly on/with the water you want to pump, as opposed to first using the hot water to power a conventional Stirling engine, and then have its flywheel drive a pump, which is taking the long way around, to say it mildly (ie. it will only serve to lessen the efficiency of the overall system).
As such there are several known designs for engines (or rather, pumps) working on the Stirling principle, but using a fluid medium instead of mechanical pistons and displacers. These usually still have some moving parts, even if the fluid (ie. water) is responsible for doing most of the work. However, I'm rather sure you could make a Stirling pump where ONLY the fluid is moving (well, possibly a few valves as well, although it should be possible, if not necessarily practical, to eliminate these as well).
It is likely that the efficiency of such a Stirling Pump will be quite low, especially if LTD (even if it should, theoretically, be a simpler and thus more efficient way of doing things). Low efficiency means a large engine/pump in relation to its power output. Since I must assume that your hot-water heating system will present the pump with a rather large head ("back-pressure"), you will likely need a reasonably powerful, and thus somewhat large, Stirling engine/pump in order to circulate the water at a proper speed. However, since you seem to have a reasonable temperature-difference (barring my lack of intuition for the Farenheit scale - I'm one of those freaks born with ten fingers rather than twelve or twenty, feet that aren't one foot long, and in a strange place where water boils at 100 degrees and freeze at zero), I should think it wouldn't need to be so big as to make it impractical, in fact I'm guessing it would be handy enough to be placed in your boiler room, unless it happens to be closet-sized and already crammed full of pipes and stuff?
(And btw.: Right choice there - Water heating systems being MUCH better than the air-based equivalent, and in so many ways as to make it a given choice - Don't let anyone tell you differently. Indeed, in my entire country there is but a handful of homes still using air-based systems, and mostly their owners are rewarded for this show of "traditionalism" with a bunch of allergies and steep heating-bills!)
Let me know if you are still seriously considering a Stirling solution for your circulation-pump. I could then try to find time to finalise one of my more promising Stirling-pump designs and mail you some drawings...
Regards,
Inventus.
Fist of all, I must agree with Robs assertion, that the most straightforward way of making a (conventional) Stirling engine self-starting, is to simply use more than one cylinder. The theory here is the same as for steam engines, namely that by offsetting the phase of each piston (by 90 degrees, in case of two cylinders), one of these will always be able to turn the flywheel (or whatever) when the other is at its "top" or "bottom" point (and thus cannot move the flywheel), and thereby allowing the engine to start regardless of which position in which it was stopped.
There is one snag to this though: If it so happens that your Stirling engine were to be heated very slowly (ie. if, for instance, it where to operate on solar power, where the sun will slowly raise its temperature during the early morning hours), this may well prevent it from self-starting, even if it has plenty of cylinders. The reason for this being that, unless special care has been taken to prevent it, such a slow heating process would allow the heat to disipate to the cold side of the engine. This could then prevent the necessary temperature difference, required for the engine to run, from ever occuring. However, a well insulated and otherwise properly constructed engine should prevent this from happening, at least within the relevant time-scale (in case of a solar-powered engine, this would be a couple of hours, not very difficult to achieve). Since a well insulated Stirling engine is also a more efficient one, this would be a design goal anyway (but still, something to keep in mind, at least if one isn't sure that the heat will always arrive quickly at the hot end of the engine).
= = =
With regard to the comments from Archimedes:
I beg to differ about his conclusions. Apprantly he suffers from a few misunderstandings with regard to the operational principles behind Stirling engines (and/or about thermofluid theory in general). It would appear that he is, among other things, confusing some details about how Stirling engines work, with how some internal combustion engines work, specifically those operating according to the Otto and Diesel cycles (aka. ignition and compression engines).
The piston of a Stirling engine is moved by the working-gas being repeatedly compressed and expanded, that much is true. But NOT like the way you see it in the aforementioned internal combustion engines, which DO have to "overcome" their compression strokes (as well as the intake and exhaust strokes, for four-stroke engines). Rather, the "compression" of the working-gas in a Stirling engine, which occurs when the gas is cooled, aids the piston-movement by providing an under-pressure (ie. "sucking" the piston in the right direction), just like the "expansion", which occurs when the gas is heated, aids the piston movement by providing an over-pressure (ie. "pushing" the piston, also in the "right" direction since this occurs when the piston is moving the other way).
This is in fact one of the reasons why the Stirling cycle is more efficent than the Otto and Diesel cycles, as the engine "harvests" energy on every stroke, rather than every other (for two-stroke engines) or every fourth (for four-stroke engines) as is the case for the Otto and Diesel engines (which then have to use part of this energy to keep running through the remaining cycles, where notably the compression-cycle needs a significant "investment" of energy).
As such, the only reason why most (single-cylinder) Stirling engines have a flywheel at all, is the same as why a steam-engine has one, namely to get past the "dead" parts of their strokes (ie. when the piston is at the extreme ends of its motion). Conversely (single-cylinder) internal combustion engines need their flywheels to keep turning, not only through these "dead parts" of their power-strokes, but also through the remaining stoke(s) (one for two-stroke and three for four-stroke engines). The extreme case being a four-stroke, single-cylinder Diesel engine, which not only have to produce enough power, during its single power-stroke, to overcome the dead parts of this stroke, but also through the complete intake and exhaust strokes, and most notably its compression-stroke (which will typically be VERY hard to overcome, since a Diesel engine needs a very high compression-ratio in order to work at all, and an even higher one to work efficiently). This is why you will find a HUGE flywheel on such engines (which can sometimes be found in small fishing boats - You will know them by their distinct, slow "tuck-tuck-tuck" sound). Such engines often incoorporate a pressure-release valve, which needs to be engaged before the electrical starter can turn the engine around (or, indeed, if you want to have any luck starting it by cranking it manually!)
Just like any other engine-type (Rankine/steam, Otto/ignition or Diesel/compression), the Stirling engine have no need for a flywheel when equipped with more than one cylinder (or rather, more than one piston, since you could make an engine where two pistons share a single cylinder). In real-life engines (of any kind), a small, light-weight "flywheel" will be present more often than not, for various reasons, notably other than keeping the engine going (thus making the term "flywheel" a bit of a misnomer, although the name has stuck).
Oh, and never mind about the "wonder-wheel", this is NOT an efficent engine-design, at least not in its current form (few people have made them work at all, and those who have seldom achieve more than one revolution every other minute at best, in most cases they achieve only a few revolutions per hour!) Personally I don't think this concept will ever prove to be good for anything, an opinion supported by the fact that its use is mainly advocated by crackpots and people who generally fail to understand the underlying principles.
Finally, I'm not at all sure that you would necessarily need to "adjust the amount of water pumped to match the available torque", or rather: This would occur naturally, given how pumps work (ie. if/when the torque is lowered, the speed of the engine, and thus the pump, would be reduced which would also reduce the amount of water pumped). What you might like to be able to adjust (only, in most situations, including yours, you can't) is the "head" (the water-pressure against which the pump has to operate), which would allow you to keep the pump working at a certain speed and thus keep the pumped volume per hour constant (in you case, this would require you to bypass part of the water-circulation pipes, in order to lower the pressure required to keep a certain flow-rate, which would unfortunatly cause the affeced parts of your home to be without heat).
(I'm sorry to "break your bubble" Archimedes, but you really should make sure to know your thermofluid theory, before posting such comments!)
= = =
Back to the main point:
Since you want to pump water, rather than "rotational movement", I would suggest that you skip this step altogether, and consider a "Stirling-pump". This is to say, make a Stirling engine that operates directly on/with the water you want to pump, as opposed to first using the hot water to power a conventional Stirling engine, and then have its flywheel drive a pump, which is taking the long way around, to say it mildly (ie. it will only serve to lessen the efficiency of the overall system).
As such there are several known designs for engines (or rather, pumps) working on the Stirling principle, but using a fluid medium instead of mechanical pistons and displacers. These usually still have some moving parts, even if the fluid (ie. water) is responsible for doing most of the work. However, I'm rather sure you could make a Stirling pump where ONLY the fluid is moving (well, possibly a few valves as well, although it should be possible, if not necessarily practical, to eliminate these as well).
It is likely that the efficiency of such a Stirling Pump will be quite low, especially if LTD (even if it should, theoretically, be a simpler and thus more efficient way of doing things). Low efficiency means a large engine/pump in relation to its power output. Since I must assume that your hot-water heating system will present the pump with a rather large head ("back-pressure"), you will likely need a reasonably powerful, and thus somewhat large, Stirling engine/pump in order to circulate the water at a proper speed. However, since you seem to have a reasonable temperature-difference (barring my lack of intuition for the Farenheit scale - I'm one of those freaks born with ten fingers rather than twelve or twenty, feet that aren't one foot long, and in a strange place where water boils at 100 degrees and freeze at zero), I should think it wouldn't need to be so big as to make it impractical, in fact I'm guessing it would be handy enough to be placed in your boiler room, unless it happens to be closet-sized and already crammed full of pipes and stuff?
(And btw.: Right choice there - Water heating systems being MUCH better than the air-based equivalent, and in so many ways as to make it a given choice - Don't let anyone tell you differently. Indeed, in my entire country there is but a handful of homes still using air-based systems, and mostly their owners are rewarded for this show of "traditionalism" with a bunch of allergies and steep heating-bills!)
Let me know if you are still seriously considering a Stirling solution for your circulation-pump. I could then try to find time to finalise one of my more promising Stirling-pump designs and mail you some drawings...
Regards,
Inventus.
Response to LTD engine for pumping hot water. Self starting?
Dear Inventus,
your comments are quite educated and I wish to receive the plans of the engine/pump that you previously offered.
Kind regards,
Edward
your comments are quite educated and I wish to receive the plans of the engine/pump that you previously offered.
Kind regards,
Edward
-
stan.hornbaker
- Posts: 532
- Joined: Thu Jun 20, 2002 6:01 pm
- First Name: William S.
- Last Name: Hornbaker
Response to LTD engine for pumping hot water. Self starting?
Colin D. West published "Liquid Piston Stirling Engines" in ~1978, reprinted by "Stirling Machine World" in ~198? and is no longer available at a reasonable price.
Inventus Privaticus ckaims to have "a more promising design" perhaps he can provide the design for you.
Generally LTD Stirling engines of any design are Low Power machines and useful as study aids to learn basic principles. Fluid engines are slow due to the inertia of the mass of water to be moved back and forth. As for self-starting, that too is not a feature of any Stirling engine I have encountered to date.
Inventus Privaticus ckaims to have "a more promising design" perhaps he can provide the design for you.
Generally LTD Stirling engines of any design are Low Power machines and useful as study aids to learn basic principles. Fluid engines are slow due to the inertia of the mass of water to be moved back and forth. As for self-starting, that too is not a feature of any Stirling engine I have encountered to date.
Response to LTD engine for pumping hot water. Self starting?
For the individuals posting the comments that Stirlings can't be self-starting:
http://www.youtube.com/profile_videos?user=mdevink&p=r
mdevink's ringbom engines are self starting. Watch the last 3 videos of this group: Crazy Engine, etc. Beautiful.
His website:
http://www.homepages.hetnet.nl/~gvink01/index.html
And sorry, no, I don't speak the Dutch language...
http://www.youtube.com/profile_videos?user=mdevink&p=r
mdevink's ringbom engines are self starting. Watch the last 3 videos of this group: Crazy Engine, etc. Beautiful.
His website:
http://www.homepages.hetnet.nl/~gvink01/index.html
And sorry, no, I don't speak the Dutch language...