Design of Reciprocating Piston Expanders
Posted by lohring
|
Design of Reciprocating Piston Expanders December 25, 2014 11:19AM |
Registered: 11 years ago Posts: 202 |
Below is an interesting description of the design and testing of two single cylinder piston expanders.
Lohring Miller
[nepis.epa.gov]
Lohring Miller
[nepis.epa.gov]
|
Re: Design of Reciprocating Piston Expanders December 26, 2014 01:17PM |
Registered: 11 years ago Posts: 202 |
I was very interested in the testing of dry piston sealing in a high pressure engine. The report points out the advantages of a cross head design. They didn't find suitable materials for long life in a high pressure and temperature cylinder. I can't find any test information on what Cyclone uses in an even higher pressure and temperature cylinder. You would think that something would show up since 1973 that's better than carbon. My favorite would be ceramics. Zicronium seems to be very promising as both a piston and cylinder material. It's strength and fracture toughness makes it similar to metals in structural applications. Equal low expansion between the piston and cylinder should allow a ringless design. The very low thermal conductivity would be a plus.
The sinitered material can be easily machined. There needs to be an allowance for shrinkage with firing and the final product would probably need to be ground for accuracy. I know how high precision dental products are manufactured. I need to do more research on bigger parts.
Lohring Miller
The sinitered material can be easily machined. There needs to be an allowance for shrinkage with firing and the final product would probably need to be ground for accuracy. I know how high precision dental products are manufactured. I need to do more research on bigger parts.
Lohring Miller
|
Re: Design of Reciprocating Piston Expanders January 13, 2015 11:53AM |
Registered: 15 years ago Posts: 180 |
Hello:
I read this report. I'm a bit puzzled by it. The pinnacle of the steam power era was the double compound unaflow with poppet valves by Skinner. Why oh why would anyone go back to trying single piston single stroke steam engines? It's as if in a hundred years someone decides they want to revive the IC engine, and they come up with a flat head single piston hit-and-miss glow plug design!
If someone wants to design a new steam engine, go back, read up on what was state of the art when abandoning steam, and pick up where they left off. Makes no sense to revive a design that was old in 1890. The skinner solved the high temp lube problem by cooling the liner with the saturated exhaust steam. clever design.
Dan
I read this report. I'm a bit puzzled by it. The pinnacle of the steam power era was the double compound unaflow with poppet valves by Skinner. Why oh why would anyone go back to trying single piston single stroke steam engines? It's as if in a hundred years someone decides they want to revive the IC engine, and they come up with a flat head single piston hit-and-miss glow plug design!
If someone wants to design a new steam engine, go back, read up on what was state of the art when abandoning steam, and pick up where they left off. Makes no sense to revive a design that was old in 1890. The skinner solved the high temp lube problem by cooling the liner with the saturated exhaust steam. clever design.
Dan
|
Re: Design of Reciprocating Piston Expanders January 13, 2015 03:33PM |
Registered: 19 years ago Posts: 1,268 |
Hello,
Abner Doble in his Ultramax design was going to cool the liner with feed water, which would provide a small bit of re-generation. I plan to use this concept in my compound engine. No oil is added to the steam for lubrication but oil is used on the cylinder wall. Very little of it should end up in the exhaust steam.
Does anyone have data as to how well this works?
Best,
Bill G.
Abner Doble in his Ultramax design was going to cool the liner with feed water, which would provide a small bit of re-generation. I plan to use this concept in my compound engine. No oil is added to the steam for lubrication but oil is used on the cylinder wall. Very little of it should end up in the exhaust steam.
Does anyone have data as to how well this works?
Best,
Bill G.
|
Re: Design of Reciprocating Piston Expanders January 13, 2015 03:35PM |
Admin Registered: 19 years ago Posts: 2,120 |
Wellllll...... I can think of plenty of reasons NOT to use a Double Acting Compound. The first is that the argument is being made from the specific to the general. Those Skinner engines did not generally run at the kinds of pressures we'd like to see in a steam automobile. The Dutcher engine used a similar concept and they found that the elevated MEP in the HP cylinder produced quite a bit of blowby past the piston rings, even after using up to 5 piston rings. Jay Carter told me he found that blowby is largely a function of MEP and residence time. If the pressure drops rapidly (as in a single expansion engine) there is much less leakage. As the RPM rises the steam has less time to play against the rings and promote leakage, which is important as DA engines have much greater reciprocating weight which tends to limit RPM.
That lower RPM translates into less power for the displacement, which isn't helping our power to weight ratio.
Reciprocating weight is a bummer; it is more difficult to achieve inertial balance, applies much more stress on the rods and cranks, leads to a heavier engine that detracts from vehicle performance.
DA engines are also taller, you basically have two cylinder heads and a crosshead with piston rod to deal with. This might not have been a big deal when cars were taller than they were wide, but it is of great importance today.
Compound engines also have more potential thermal and flow losses. That steam doesn't just reappear in the downstream cylinder, it has to be accelerated up to speed. If you figure the weight of the steam, the distance to be covered, and the amount of time the valve has to open and close you come up with a significant energy expenditure. If you're familiar with the kind of porting in high performance IC engine passages you also know about flow losses, given the density of steam versus air that is a serious matter.
The LP piston has greater friction due to the larger circumference and also loses more heat due to the greater cylinder wall surface; these cut down on the relative efficiency of the LP expander.
In a mass produced engine, compounds are going to be significantly more expensive. You are going to have more different types of unique parts and you are going to be manufacturing fewer of these parts; your tooling cost will go up for every new unique component you need but your economies of scale suffer next to the simple engine using standardized hardware.
Compound engines tend to have less regular instantaneous torque production, for any given pressure there is a unique set of cutoffs for the HP and LP cylinder which cause the two to produce the same, balanced torque. As the pressure varies from this point the torque output gets "lumpier". This not only affects engine smoothness but also introduces higher stresses, especially on the lower end.
If we go back to inertia balancing; the 2 cylinder, double-acting steam engine is perhaps the worst design imaginable. The crank throws are set 90 degrees apart which introduces a strong primary shake with a primary rocking couple superimposed on it and that is topped off with a
secondary rocking couple propagating at twice the frequency as the primary forces. As rpm goes up this is hardly smooth. Jim Crank talks about the counterweighting they needed to fit to the Doble F-Series engine just to keep the windshield from shaking; but that just dampens the problem, it's inherent to the layout and is only cancelled by the addition of balance shafts or other auxiliary balance mechanisms such as dummy pistons.
Actually, if we examine the matter, we find that the Doble Ultimax engine as developed for the McCulloch-Paxton Phoenix was just about the design you are talking about and it encountered a number of difficulties.
Regards,
Ken
That lower RPM translates into less power for the displacement, which isn't helping our power to weight ratio.
Reciprocating weight is a bummer; it is more difficult to achieve inertial balance, applies much more stress on the rods and cranks, leads to a heavier engine that detracts from vehicle performance.
DA engines are also taller, you basically have two cylinder heads and a crosshead with piston rod to deal with. This might not have been a big deal when cars were taller than they were wide, but it is of great importance today.
Compound engines also have more potential thermal and flow losses. That steam doesn't just reappear in the downstream cylinder, it has to be accelerated up to speed. If you figure the weight of the steam, the distance to be covered, and the amount of time the valve has to open and close you come up with a significant energy expenditure. If you're familiar with the kind of porting in high performance IC engine passages you also know about flow losses, given the density of steam versus air that is a serious matter.
The LP piston has greater friction due to the larger circumference and also loses more heat due to the greater cylinder wall surface; these cut down on the relative efficiency of the LP expander.
In a mass produced engine, compounds are going to be significantly more expensive. You are going to have more different types of unique parts and you are going to be manufacturing fewer of these parts; your tooling cost will go up for every new unique component you need but your economies of scale suffer next to the simple engine using standardized hardware.
Compound engines tend to have less regular instantaneous torque production, for any given pressure there is a unique set of cutoffs for the HP and LP cylinder which cause the two to produce the same, balanced torque. As the pressure varies from this point the torque output gets "lumpier". This not only affects engine smoothness but also introduces higher stresses, especially on the lower end.
If we go back to inertia balancing; the 2 cylinder, double-acting steam engine is perhaps the worst design imaginable. The crank throws are set 90 degrees apart which introduces a strong primary shake with a primary rocking couple superimposed on it and that is topped off with a
secondary rocking couple propagating at twice the frequency as the primary forces. As rpm goes up this is hardly smooth. Jim Crank talks about the counterweighting they needed to fit to the Doble F-Series engine just to keep the windshield from shaking; but that just dampens the problem, it's inherent to the layout and is only cancelled by the addition of balance shafts or other auxiliary balance mechanisms such as dummy pistons.
Actually, if we examine the matter, we find that the Doble Ultimax engine as developed for the McCulloch-Paxton Phoenix was just about the design you are talking about and it encountered a number of difficulties.
Regards,
Ken
|
Re: Design of Reciprocating Piston Expanders January 13, 2015 04:34PM |
Registered: 15 years ago Posts: 180 |
|
Re: Design of Reciprocating Piston Expanders January 13, 2015 05:48PM |
Registered: 19 years ago Posts: 657 |
RESOURCE LINK
[oac.cdlib.org]
The above link is to the Bancroft library and gives one access to free online-available note and diary's of Abner Doble from his earliest experiments up to his work on the failed Ultimax.
Having, over the years, read through all of those papers 3 or more times, I can say that it is the most important information regarding light steam power available on the internet. More so for what failed and why, then what worked.
Don't just read it, STUDY IT!
Caleb Ramsby
[oac.cdlib.org]
The above link is to the Bancroft library and gives one access to free online-available note and diary's of Abner Doble from his earliest experiments up to his work on the failed Ultimax.
Having, over the years, read through all of those papers 3 or more times, I can say that it is the most important information regarding light steam power available on the internet. More so for what failed and why, then what worked.
Don't just read it, STUDY IT!
Caleb Ramsby
|
Re: Design of Reciprocating Piston Expanders January 13, 2015 06:09PM |
Admin Registered: 19 years ago Posts: 2,120 |
OK,
I see Caleb posted the Bancroft link while I was digging it up. More information can be found from the SAE report that can be found at:
Steam Automobile Pt 1Steam Automobile Pt 2
Steam Automobile Pt 3
Steam Automobile Pt 4
Ken
I see Caleb posted the Bancroft link while I was digging it up. More information can be found from the SAE report that can be found at:
Steam Automobile Pt 1Steam Automobile Pt 2
Steam Automobile Pt 3
Steam Automobile Pt 4
Ken
|
Re: Design of Reciprocating Piston Expanders January 13, 2015 06:29PM |
Admin Registered: 19 years ago Posts: 2,120 |
I probably should have mentioned that clearance volume is always an issue for uniflow engines. If you read the 1922 Stumpf book you will find he regards it as one of the greater evils and seeks to minimize it. You have to remember that Stumpf was dealing with industrial engines, often retrofitted to older steam systems. Stumpf is clear that recompressing residual steam in the cylinder at the end of the exhaust period reduces the bad effects of clearance, and the benefit extends right up to when the recompression hits the incoming steam pressure, then overcompression sets in and things get worse. Note I said reduce, compression is not a replacement for eliminating clearance.
This is where things get tricky for an automobile. Industrial engines with good vacuum condensing could have a very small clearance volume and still fully recompress, which is wonderful. They obtained that clearance volume due to water cooling, since we can't always drive through the water our cars are limited to air cooling---between often higher ambient temperatures and much worse thermal carrying capacity we just aren't going to see the same kind of vacuum, so we need more clearance if all things are equal.
This is a bit problematic for uniflow engines as the compression starts when the exhaust port is blocked by the piston, counterflow engines can compress later and reduce clearance volume. This is a case of Stumpf being absolutely right but maybe he is being taken a bit out of context for automobile engines.
The second problem with clearance in car engines is that they run at a wide range of power levels compared to an industrial engine. Since we know about where the industrial engine will spend most of its time, we can tailor a clearance volume that gives best compression most of the time. Automobile engines are all over the place, more than likely most operation won't be THAT close to ideal recompression. We can stick a Williams valve in, design for mostly overcompression and vent off the excess. The reduction in clearance helps but the overcompression creates negative work and appears to partly offset the benefit of the higher compression.
I suspect the torque reversals that plagued the Ultimax were to a large degree a result of high uniflow compression.
Another point to consider is that maybe uniflow is a bit less effective than we would expect in a really modern engine based on Stumpf's writing, as the superheat rises the steam becomes less thermally conductive and more reluctantly gives up heat to the surroundings. This doesn't mean that uniflow doesn't prevent heat loss out the exhaust --- but maybe that the loss we will see is less than what we would expect from reading Stumpf, might be that we are again taking him out of context as he certainly wasn't contemplating running many engines at 1000 and 1000.
At no point am I suggesting uniflow isn't a valuable tool, but I am less than convinced that you can just plug in a uniflow engine and get great results. In fact, I really believe it takes a lot more work to develop a good uniflow than a good counterflow....although I do have some ideas on dealing with it.
Regards,
ken
This is where things get tricky for an automobile. Industrial engines with good vacuum condensing could have a very small clearance volume and still fully recompress, which is wonderful. They obtained that clearance volume due to water cooling, since we can't always drive through the water our cars are limited to air cooling---between often higher ambient temperatures and much worse thermal carrying capacity we just aren't going to see the same kind of vacuum, so we need more clearance if all things are equal.
This is a bit problematic for uniflow engines as the compression starts when the exhaust port is blocked by the piston, counterflow engines can compress later and reduce clearance volume. This is a case of Stumpf being absolutely right but maybe he is being taken a bit out of context for automobile engines.
The second problem with clearance in car engines is that they run at a wide range of power levels compared to an industrial engine. Since we know about where the industrial engine will spend most of its time, we can tailor a clearance volume that gives best compression most of the time. Automobile engines are all over the place, more than likely most operation won't be THAT close to ideal recompression. We can stick a Williams valve in, design for mostly overcompression and vent off the excess. The reduction in clearance helps but the overcompression creates negative work and appears to partly offset the benefit of the higher compression.
I suspect the torque reversals that plagued the Ultimax were to a large degree a result of high uniflow compression.
Another point to consider is that maybe uniflow is a bit less effective than we would expect in a really modern engine based on Stumpf's writing, as the superheat rises the steam becomes less thermally conductive and more reluctantly gives up heat to the surroundings. This doesn't mean that uniflow doesn't prevent heat loss out the exhaust --- but maybe that the loss we will see is less than what we would expect from reading Stumpf, might be that we are again taking him out of context as he certainly wasn't contemplating running many engines at 1000 and 1000.
At no point am I suggesting uniflow isn't a valuable tool, but I am less than convinced that you can just plug in a uniflow engine and get great results. In fact, I really believe it takes a lot more work to develop a good uniflow than a good counterflow....although I do have some ideas on dealing with it.
Regards,
ken
|
Re: Design of Reciprocating Piston Expanders January 13, 2015 09:22PM |
Registered: 19 years ago Posts: 657 |
To add to Kens thoughts on the uniflow, which I agree with completely.
From not only studying the extant material on engine steam usage rates but also writing a few rather complex and complete spreadsheets which analyze and "duplicate" the operation of an engine I have found that the best efficiency is obtained with a cutoff equal to 1 to 1.5 times the clearance volume with the initial pressure being 350 to 450 psi and recompression back to the initial pressure. I have my spreadsheet subtract the weight of the over-compressed steam from the steam supplied from the boiler, since it goes back into the steam chest, but I throw away the superheat that the ejected over-compressed steam has, since I see it being dispersed and absorbed by the steam chest steam and metal. My spreadsheet was written using the formulas that the late Professor Hall used to make his Perform program. Which duplicated the steam movement into and out of engine through the valves with a piston valved Walshcearts valve geared engine.
You can find his programs and papers at the following link.
[5at.co.uk]
One note on his Perform program, the "perimeter" dimension for the valve is the circumference of the piston valve minus the bridges. It is excellent and should be studied extensively by anyone serious about steam.
Take careful note of how with a very early cutoff, say 5%, the exhaust valve opens very early, BUT the actual expansion of the steam(which includes the 1.3 or there abouts factor) is such that the early exhaust closing has but a minimal effect on a real engines efficiency.
Counterflow piston valved engines are not nearly as outdated or technically ignorant as many suppose.
Another massive factor in regards to Stumpfs work is the sheer size of the engines. At working temperature cast iron has a k factor of around 25 btu sq ft deg F per ft per hr. With a 24" stroke that equates to a k factor of 12.5 for the length of the stroke, with a 4" stroke it equates to a k factor of 75 for the length of the stroke. That is a difference of 600% !!! Stumpfs engines were "small" if the stroke was 24" . For most of the "modern" automotive sized uniflow engines 4" is a BIG stroke. Read through Stumpfs book(s) and pay attention to the stroke dimensions of them. A very conservative average is 36" . As for "modern" high speed, high pressure, high temperature light steam engines the average is closer to a 3" stroke. Using the later mentioned figures, the k factor for Stumpfs engines is 8.33 and for the "modern" uniflow it is 100! That is a difference of 1,200% .
If one figures the distance between where the live steam is cutoff and where the exhaust starts. If the inlet were cutoff at 5% (which I believe to be a VERY short cutoff and of questionable value in the real world) that would be 1.8" of stroke for the 36" Stumpf uniflow, the exhaust starts at 10%, so that is 3.6", in total that is 5.4". Subtract that from the 36" stroke and we have 30.6" of "temperature gradient" cylinder wall, which gives a k factor of 9.87. For the 3" stroke engine, the cutoff is at .15" and the exhaust ports at .3" long, for .45" which subtracted from the 3" stroke give us 2.55", which gives us a k factor of 117.6. The difference is now 1,192%. Although the percentage is less, the k factor of the small engine is just terrific!
To put this in perspective, sandstone has a k factor of 1.1 btu sq ft deg F per ft per hr. So, using the last figures, Stumpfs engine is equivalent to 1.34" of sandstone and the smaller engine to .11" of sandstone. Imagine placing a torch on one side of the stone and your hand on the other, the thick stone would hold off the heat for a good long while, the thin stone would get hot quick.
To get back to the beginning of this little rant, the effective superheat when operating at a THROTTLED pressure of 350 to 450 psi is much greater then when running at 1,000 psi inlet pressure. The lower the pressure the more btu of superheat the steam holds for a given temperature. Holding the superheat down in the 850 F range makes a lot more sense to me when the huge k factor from hot to cold end is considered.
Keep firmly in mind that Stumpf used LOW pressure and superheat, yet he still had serious issues with carbonized oils because of how hot the compressed steam got. With a small engine, as has been pointed out by Ken before, the surface area to volume ratio is so great and the residence time so low, that these issues don't seem to appear. I don't recall Ted Pritchard mentioning any issues with carbonized oil in his uniflow and he was rather frank about failure, but he is dead now.
Instead of thinking of the counterflow piston valved engine as a outdated relic, try to see how naturally versatile it is as compared to a uniflow or a compound engine.
I would also like to note that the D type slide valved engine can be close to as good as the piston valve engine, as long as the pressure is kept low, superheat relatively low and copious quantities of oil supplied. From the old locomotive statistics I found studies of the actual pull required for various valves as testes as the engine ran down the track! I was shocked at how low the coefficient of drag was for the D type valves. I recall .026 as being the number, I would have to dig through my notes to find it exactly. This varied with the supply of oil and quality of manufacture. I think that with direct oiling of the valve they would hold up very well.
Likewise the Corliss valve, the warping that they are known for with superheat is attributed to the irregular cross section of the valves on the massive engines that employed them. An engineers study of them that I came across found that very irregular valve section caused warping with SATURATED steam of low pressure. However, with a consisted cross section of the valve, superheat of 700 F and higher didn't warp them. The irregular sections I am referring to were stiffening ribs and knobs. As far as clearance volume is concerned it is going to be difficult to get much better then the Corliss valves. The Corliss, like ALL valves has its own construction, design and layout issues.
There is no magic button that can be pushed that will eject a "perfect" engine. I think that closest would be to duplicate the Bryan engines design.
Caleb Ramsby
From not only studying the extant material on engine steam usage rates but also writing a few rather complex and complete spreadsheets which analyze and "duplicate" the operation of an engine I have found that the best efficiency is obtained with a cutoff equal to 1 to 1.5 times the clearance volume with the initial pressure being 350 to 450 psi and recompression back to the initial pressure. I have my spreadsheet subtract the weight of the over-compressed steam from the steam supplied from the boiler, since it goes back into the steam chest, but I throw away the superheat that the ejected over-compressed steam has, since I see it being dispersed and absorbed by the steam chest steam and metal. My spreadsheet was written using the formulas that the late Professor Hall used to make his Perform program. Which duplicated the steam movement into and out of engine through the valves with a piston valved Walshcearts valve geared engine.
You can find his programs and papers at the following link.
[5at.co.uk]
One note on his Perform program, the "perimeter" dimension for the valve is the circumference of the piston valve minus the bridges. It is excellent and should be studied extensively by anyone serious about steam.
Take careful note of how with a very early cutoff, say 5%, the exhaust valve opens very early, BUT the actual expansion of the steam(which includes the 1.3 or there abouts factor) is such that the early exhaust closing has but a minimal effect on a real engines efficiency.
Counterflow piston valved engines are not nearly as outdated or technically ignorant as many suppose.
Another massive factor in regards to Stumpfs work is the sheer size of the engines. At working temperature cast iron has a k factor of around 25 btu sq ft deg F per ft per hr. With a 24" stroke that equates to a k factor of 12.5 for the length of the stroke, with a 4" stroke it equates to a k factor of 75 for the length of the stroke. That is a difference of 600% !!! Stumpfs engines were "small" if the stroke was 24" . For most of the "modern" automotive sized uniflow engines 4" is a BIG stroke. Read through Stumpfs book(s) and pay attention to the stroke dimensions of them. A very conservative average is 36" . As for "modern" high speed, high pressure, high temperature light steam engines the average is closer to a 3" stroke. Using the later mentioned figures, the k factor for Stumpfs engines is 8.33 and for the "modern" uniflow it is 100! That is a difference of 1,200% .
If one figures the distance between where the live steam is cutoff and where the exhaust starts. If the inlet were cutoff at 5% (which I believe to be a VERY short cutoff and of questionable value in the real world) that would be 1.8" of stroke for the 36" Stumpf uniflow, the exhaust starts at 10%, so that is 3.6", in total that is 5.4". Subtract that from the 36" stroke and we have 30.6" of "temperature gradient" cylinder wall, which gives a k factor of 9.87. For the 3" stroke engine, the cutoff is at .15" and the exhaust ports at .3" long, for .45" which subtracted from the 3" stroke give us 2.55", which gives us a k factor of 117.6. The difference is now 1,192%. Although the percentage is less, the k factor of the small engine is just terrific!
To put this in perspective, sandstone has a k factor of 1.1 btu sq ft deg F per ft per hr. So, using the last figures, Stumpfs engine is equivalent to 1.34" of sandstone and the smaller engine to .11" of sandstone. Imagine placing a torch on one side of the stone and your hand on the other, the thick stone would hold off the heat for a good long while, the thin stone would get hot quick.
To get back to the beginning of this little rant, the effective superheat when operating at a THROTTLED pressure of 350 to 450 psi is much greater then when running at 1,000 psi inlet pressure. The lower the pressure the more btu of superheat the steam holds for a given temperature. Holding the superheat down in the 850 F range makes a lot more sense to me when the huge k factor from hot to cold end is considered.
Keep firmly in mind that Stumpf used LOW pressure and superheat, yet he still had serious issues with carbonized oils because of how hot the compressed steam got. With a small engine, as has been pointed out by Ken before, the surface area to volume ratio is so great and the residence time so low, that these issues don't seem to appear. I don't recall Ted Pritchard mentioning any issues with carbonized oil in his uniflow and he was rather frank about failure, but he is dead now.
Instead of thinking of the counterflow piston valved engine as a outdated relic, try to see how naturally versatile it is as compared to a uniflow or a compound engine.
I would also like to note that the D type slide valved engine can be close to as good as the piston valve engine, as long as the pressure is kept low, superheat relatively low and copious quantities of oil supplied. From the old locomotive statistics I found studies of the actual pull required for various valves as testes as the engine ran down the track! I was shocked at how low the coefficient of drag was for the D type valves. I recall .026 as being the number, I would have to dig through my notes to find it exactly. This varied with the supply of oil and quality of manufacture. I think that with direct oiling of the valve they would hold up very well.
Likewise the Corliss valve, the warping that they are known for with superheat is attributed to the irregular cross section of the valves on the massive engines that employed them. An engineers study of them that I came across found that very irregular valve section caused warping with SATURATED steam of low pressure. However, with a consisted cross section of the valve, superheat of 700 F and higher didn't warp them. The irregular sections I am referring to were stiffening ribs and knobs. As far as clearance volume is concerned it is going to be difficult to get much better then the Corliss valves. The Corliss, like ALL valves has its own construction, design and layout issues.
There is no magic button that can be pushed that will eject a "perfect" engine. I think that closest would be to duplicate the Bryan engines design.
Caleb Ramsby
|
Re: Design of Reciprocating Piston Expanders January 13, 2015 10:15PM |
Registered: 19 years ago Posts: 657 |
Seeing as how this is a thread about engine design. I thought I would mention the effect of the bore and stroke ratio on the clearance volume and surface area at cutoff, then I remembered that I had written a spreadsheet to study this effect. I couldn't find it so I whipped up another one, it is a bit simple mind you.
Later on I may add the effect on the k factor to it and a few other things, but for now it is what it is.
Caleb Ramsby
Later on I may add the effect on the k factor to it and a few other things, but for now it is what it is.
Caleb Ramsby
|
Re: Design of Reciprocating Piston Expanders January 13, 2015 10:26PM |
Registered: 15 years ago Posts: 180 |
I think there are two camps in this hobby: those who want to drive a Locomobile-like car around the neighborhood (i would like that too), and those who want to continue in the footsteps of Doble.
Nothing wrong with a low pressure, D or piston valved, walshaerts gear engine. It's a neat thing to watch run. It's just the ticket if you're building a steam buggy or a steam bicycle. But I don't see how it fits in a modern steam car. You're trying to get the lowest water consumption possible so that it is (hopefully) fully condensing under most circumstances. With low pressure steam the condenser is just way too big.
I'm open to changing my mind though...
Dan
Hey! how about converting an Amish Wagon to steam! that would be neat. just the right look for an old time steam carriage.
Nothing wrong with a low pressure, D or piston valved, walshaerts gear engine. It's a neat thing to watch run. It's just the ticket if you're building a steam buggy or a steam bicycle. But I don't see how it fits in a modern steam car. You're trying to get the lowest water consumption possible so that it is (hopefully) fully condensing under most circumstances. With low pressure steam the condenser is just way too big.
I'm open to changing my mind though...
Dan
Hey! how about converting an Amish Wagon to steam! that would be neat. just the right look for an old time steam carriage.
|
Re: Design of Reciprocating Piston Expanders January 13, 2015 10:48PM |
Registered: 19 years ago Posts: 657 |
Dan,
If you are putting me in the Locomobile camp, think again. I have absolutely ZERO interest in making one of those and puttering around. I don't intend to knock them, but that just isn't my interest.
What I have been working on for the past few years is an extremely high performance motorcycle design, where condensation is a HUGE issue. My research has proven to me that the water rate of uniflow and compound engines over the range of road usage is at least equal to if not higher then that of a properly designed counterflow piston valved engine with a low clearance volume and tight piston valve rings. Running 5% cutoff 400 psi and 850 F with recompression back slightly higher then inlet pressure "should" give around 8 to 9 lbs hp hr. That would be for cruising and very moderate acceleration. Hard core acceleration would be with much longer cutoffs, drawing on the boilers reserve and using a pressurized condenser to deal with the excessive steam consumption. I would also like to refrain from speaking of my design too much, not for "privacy" issues, it's just that the speaking of it detracts from the designing of it!
Contrary to how I am sure I often come across, I have no interest in "converting" peoples opinions. I just like to like to put resources out there for people and share with them what I have discovered in my research.
I don't mean to be an a**hole here Dan but, how do you plan on "following in the footsteps of Doble" if you have not as of yet spent the hundreds of hours studying all of his papers that I linked too? Much less spent hours upon hours running Professor Halls Perform program to see just what a counterflow engine is capable of.
Caleb Ramsby
PS. Time for me to get back to work and stop this "debating" nonsense.
If you are putting me in the Locomobile camp, think again. I have absolutely ZERO interest in making one of those and puttering around. I don't intend to knock them, but that just isn't my interest.
What I have been working on for the past few years is an extremely high performance motorcycle design, where condensation is a HUGE issue. My research has proven to me that the water rate of uniflow and compound engines over the range of road usage is at least equal to if not higher then that of a properly designed counterflow piston valved engine with a low clearance volume and tight piston valve rings. Running 5% cutoff 400 psi and 850 F with recompression back slightly higher then inlet pressure "should" give around 8 to 9 lbs hp hr. That would be for cruising and very moderate acceleration. Hard core acceleration would be with much longer cutoffs, drawing on the boilers reserve and using a pressurized condenser to deal with the excessive steam consumption. I would also like to refrain from speaking of my design too much, not for "privacy" issues, it's just that the speaking of it detracts from the designing of it!
Contrary to how I am sure I often come across, I have no interest in "converting" peoples opinions. I just like to like to put resources out there for people and share with them what I have discovered in my research.
I don't mean to be an a**hole here Dan but, how do you plan on "following in the footsteps of Doble" if you have not as of yet spent the hundreds of hours studying all of his papers that I linked too? Much less spent hours upon hours running Professor Halls Perform program to see just what a counterflow engine is capable of.
Caleb Ramsby
PS. Time for me to get back to work and stop this "debating" nonsense.
|
Re: Design of Reciprocating Piston Expanders January 13, 2015 11:17PM |
Registered: 15 years ago Posts: 180 |
No no, not trying to debate. I do keep an open mind. What you say is interesting.
I think there is a place for steam in some kind of "supercar". The paxton engine had a starting torque of 900 ft-lbs. The LS7 engine can't touch that. If you've got enough steam that is. But if it can be done with a counter flow engine, why not? After all a steam plant is more than just the expander.
I'm willing to take a look at your numers.
Dan.
I think there is a place for steam in some kind of "supercar". The paxton engine had a starting torque of 900 ft-lbs. The LS7 engine can't touch that. If you've got enough steam that is. But if it can be done with a counter flow engine, why not? After all a steam plant is more than just the expander.
I'm willing to take a look at your numers.
Dan.
|
Re: Design of Reciprocating Piston Expanders January 13, 2015 11:28PM |
Registered: 15 years ago Posts: 180 |
I understand and acknowledge that i do not have the experience with steam that many have on this site. I do however have several old engineering textbooks on steam, and I am just expressing what I have read. I am of German descent and sometimes my "helpful remarks" sound more like "condescending criticism". Sorry :-(
Dan
Dan
|
Re: Design of Reciprocating Piston Expanders January 13, 2015 11:58PM |
Admin Registered: 19 years ago Posts: 2,120 |
Hi Dan,
Having worked on the development of the LS-7 and LS-9, I have to stand up for them. The problem a steam engine is going to have isn't torque, it's power. When you have an LS-9 peaking out with a bit over 600 ft-lbs at 6600 RPM, you have a hideous amount of power. About 650 HP. Maybe not as much torque as that steamer, but with all those revs you can afford to gear down massively. Call it about 2.29 to 1 in first gear for the ZR-1 and about 2.73 for the rear end (you can get higher and lower). So we're north of 3500 ft-lbs at the wheels if we do things right....and the engine is still running over 1,000 rpm. Realistically, the torque is so great that you will need launch control to limit output and control wheel slip anyhow...even so the new Z06 is pulling 0-60 in about 3 seconds.
Maybe you could get a steam engine to do that, but the only way you'd ever fit it into a comparable car would be to use a boiler with a lot of reserve and have one or two good, hot starts before you draw the boiler down. That won't help your time at the Nurburgring. And that's the problem with a steam sports car. An IC engine does everything in the cylinder, the engine is a fairly self contained unit. Since no one has managed to build a really good compression fired automotive boiler, we are going to have a relatively bulky burner, then a boiler to capture that heat to turn it into steam...and that takes real surface area, even the incredibly compact SES boiler with 160 HP isn't as powerful per square foot of engine compartment occupied as a high performance IC engine. Then we get to an expander into whatever room is left to develop the power. Toss in the problems of trying to condense all that steam and the relatively lower thermal efficiency and there just isn't any real practical way to challenge a modern sports car that can run lap after lap at speeds that only dedicated race cars could manage not so long ago.
Truthfully, I don't see the sports car being exactly a good fit for a steam power plant in any case. We're talking about machines where companies put in exhaust bypasses so that they will roar under certain conditions; the steam plant's inherent characteristics are almost antithetical to a high performance sports car. To me, the touring sedan makes a much, much better fit. Quiet, smooth, deep thrust...but not too pushy or flashy. It seems a natural.
Regards,
Ken
Having worked on the development of the LS-7 and LS-9, I have to stand up for them. The problem a steam engine is going to have isn't torque, it's power. When you have an LS-9 peaking out with a bit over 600 ft-lbs at 6600 RPM, you have a hideous amount of power. About 650 HP. Maybe not as much torque as that steamer, but with all those revs you can afford to gear down massively. Call it about 2.29 to 1 in first gear for the ZR-1 and about 2.73 for the rear end (you can get higher and lower). So we're north of 3500 ft-lbs at the wheels if we do things right....and the engine is still running over 1,000 rpm. Realistically, the torque is so great that you will need launch control to limit output and control wheel slip anyhow...even so the new Z06 is pulling 0-60 in about 3 seconds.
Maybe you could get a steam engine to do that, but the only way you'd ever fit it into a comparable car would be to use a boiler with a lot of reserve and have one or two good, hot starts before you draw the boiler down. That won't help your time at the Nurburgring. And that's the problem with a steam sports car. An IC engine does everything in the cylinder, the engine is a fairly self contained unit. Since no one has managed to build a really good compression fired automotive boiler, we are going to have a relatively bulky burner, then a boiler to capture that heat to turn it into steam...and that takes real surface area, even the incredibly compact SES boiler with 160 HP isn't as powerful per square foot of engine compartment occupied as a high performance IC engine. Then we get to an expander into whatever room is left to develop the power. Toss in the problems of trying to condense all that steam and the relatively lower thermal efficiency and there just isn't any real practical way to challenge a modern sports car that can run lap after lap at speeds that only dedicated race cars could manage not so long ago.
Truthfully, I don't see the sports car being exactly a good fit for a steam power plant in any case. We're talking about machines where companies put in exhaust bypasses so that they will roar under certain conditions; the steam plant's inherent characteristics are almost antithetical to a high performance sports car. To me, the touring sedan makes a much, much better fit. Quiet, smooth, deep thrust...but not too pushy or flashy. It seems a natural.
Regards,
Ken
|
Re: Design of Reciprocating Piston Expanders January 14, 2015 01:31AM |
Registered: 15 years ago Posts: 180 |
|
Re: Design of Reciprocating Piston Expanders January 14, 2015 01:59AM |
Registered: 19 years ago Posts: 657 |
Hey Dan,
The big Stanley engines made well over one ton of torque, not as efficiently as is possible, but still.
The engine is certainly only one component of a steam power plant. Ken mentioned the reserve of the boiler. That is a BIG part of the powerplant, in my opinion second only to the fires density, the engine coming third and the condenser last. To study the reserve I wrote a rather massive spreadsheet that replicates the vehicle, engine, boiler, pumps ect. to determine the effect that the water storage, burner output, engine size, gearing, cutoff boiler pressure, water pump size, superheater diameter and length, ect. on the dynamic ability of a vehicle. Doing so took 300 hours, I have spent many times that using said program and can say without question that a steam powerplant without a significant reserve capacity is junk. I can also say that the best 1/4 time will be had if the engine is proportioned to break the tires traction at 1/3 of the boilers pressure and at the longest cutoff for the engine run. For a one mile run it is best to hook up in shorter cutoff when the draw down peaks.
Everything is so D*#& complicated and connected that it is an absolute nightmare attempting to determine how one slight change of one component effects all of the others. A case in point is that when I use the above mentioned spreadsheet, I cross reference it with the one of my boiler design spreadsheets, say if the pressure loss is too great in the superheater and I make it bigger in diameter, then I go to the boiler spreadsheet and change it to see how much hotter the superheater would run with the low fps steam flow through it. It is the same thing with the engines gearing, gearing it higher will help acceleration, but then I got to cross reference it with the valve flow spreadsheet to make sure that it isn't choking itself to death, if it is then it needs larger valves, then it is to the drafting board to layout new valves and ports then back to the valve flow spreadsheet to adjust the parameters such as the larger clearance volume, then go around and around many, many times.
I should note that I am rather far removed from being a highly experienced steam guy, I have spent a lot of time devouring information, yes, but there is still a lot to learn.
One thing I do know for sure is that the 8 to 9 lbs per hp hr that I stated in my earlier post, regarding my calculations as to the counterflow engine, it would be lucky to be a real world 12 to 13 lbs per hp hr. The uniflow engines are the same way, there is a substantial drop from the theoretical efficiency to the dyno efficiency to the ROAD efficiency.
I will link yet again to this:
[www.steamcar.net]
A few quotes from the above link about the SES project.
"The SES system was designed to the practical limit of both parameters, settling eventually on 1000 psi and 1000 ºF to the expander inlet at all times."
"Interestingly, temperature had little effect on performance in the range above some 750 ºF."
"Certainly, SES could have saved itself lots of headaches had it known this from the beginning, because lowering the temperature while receiving the same performance would have eliminated much of that expensive research into exotic lubricants and wear properties. Lower temperature would have also made the use of the finicky throttle valve easier, a valve that was hoped would not be necessary but in practice the fully automatic control was too difficult to achieve without it. It also turned out that, contrary to the data of specific steam consumption obtained on a dynamometer, overall efficiency did not suffer all that much in the car, even improved for some operating regimes, at a lower than 1000 psi inlet pressure."
I think that what is happening is that with the very hot steam and the smaller engines the greater heat transfer from the greater temperature differential disallows the theoretical efficiency that 1,000 F steam "should" give.
Also, Skinner didn't employ high temperatures or high pressures, as Ken stated, they used VERY low condenser pressures.
Ken,
One solution for "hard core acceleration" I have come up with is a good oll' bypass valve for the condenser, which lets the engine "bark" with the best of them! Just strap in a larger water tank. Also, where in the heck can someone in the U.S.A. maintain 200 mph for more then 2 miles on the open road? How about 150 mph on the open road, 2 what 3 miles at most. 100 mph, well on the interstate for a while or in the middle of the desert.
The above sounds sarcastic, but seriously, really seriously, what are your, or anyone else's who is reading this, estimate as to the greatest distance that one could travel in America on a regular basis at 100 mph, 150 mph and 200 mph? The answers to those questions dictate to a massive extent the ratio or maximum firing rate to the reserve capacity of the boiler.
My work with the above mentioned spreadsheet has shown me that outputs well above 1,000 hp are possible with enough reserve and a moderate firing rate, that would be with a relatively large engine, boiler pressure of 2,000 psi and tire spinning torque at 500 psi and 70% cutoff, that would give 1/4 mile times and speeds that would have to be seen to be believed and would be better then any car licensed to drive on the road. That is no joke or exaggeration.
At sustained high outputs condensing becomes a big issue, the best solution I have found with my feed water heater and condenser spreadsheets is to abandon the "fan" because at those speeds it is called a "propeller" and use an Exair or similar air accelerator with tight ducting for the condenser(s) and an engine exhaust powered air pump to power the air accelerators.
As for the boiler, yet again the spreadsheets, I have written dozens of them over the years, with various tube layouts and configurations. I firmly believe that it is safe to say that an overall heat transfer coefficient of 75,000 btu sq ft per hour is more then possible with a forced re-circulation type boiler. That includes the superheater, evaporation and economizer sections of the boiler. If anyone reading this gets around to writing their own boiler analysis spreadsheet, I have found that the minimum step rate for each pass of tubes is 100. That is 100 calculations for the heat transfer per bank of tubes. Any less and the output from the spreadsheet is deranged. I calculate the temp rise of the tubes and influence of the flow rates on the heat transfer of the superheater and the economizer, when I was investigating finned tubing I even had it calculate the fin temp rise. After a lot of work I don't see that fins, from an overall standpoint, are of a great advantage. At very high outputs the fin rise makes then uneconomical and bulky. The heat transfer to boiling water is extremely complex so I use Wolverine tubes db3cal spreadsheet to cross reference my boiler designs. I will attach their program to this post since they supply it for free to anyone. I also use free source X Steam for Excel for the steam properties, I will also attach it. I used those two programs and old engineering books as the source material for the spreadsheets. . . that and a LOT of time.
I wish I could supply my more complex spreadsheets to people, but they are similar to someones cluttered desk, that person knows where everything is, but nobody else does. . .
*($% the db3cal program is too large to attach here, so here is the link to their website to download it [www.wlv.com]
Caleb Ramsby
Edited 1 time(s). Last edit at 01/14/2015 02:05AM by Caleb Ramsby.
The big Stanley engines made well over one ton of torque, not as efficiently as is possible, but still.
The engine is certainly only one component of a steam power plant. Ken mentioned the reserve of the boiler. That is a BIG part of the powerplant, in my opinion second only to the fires density, the engine coming third and the condenser last. To study the reserve I wrote a rather massive spreadsheet that replicates the vehicle, engine, boiler, pumps ect. to determine the effect that the water storage, burner output, engine size, gearing, cutoff boiler pressure, water pump size, superheater diameter and length, ect. on the dynamic ability of a vehicle. Doing so took 300 hours, I have spent many times that using said program and can say without question that a steam powerplant without a significant reserve capacity is junk. I can also say that the best 1/4 time will be had if the engine is proportioned to break the tires traction at 1/3 of the boilers pressure and at the longest cutoff for the engine run. For a one mile run it is best to hook up in shorter cutoff when the draw down peaks.
Everything is so D*#& complicated and connected that it is an absolute nightmare attempting to determine how one slight change of one component effects all of the others. A case in point is that when I use the above mentioned spreadsheet, I cross reference it with the one of my boiler design spreadsheets, say if the pressure loss is too great in the superheater and I make it bigger in diameter, then I go to the boiler spreadsheet and change it to see how much hotter the superheater would run with the low fps steam flow through it. It is the same thing with the engines gearing, gearing it higher will help acceleration, but then I got to cross reference it with the valve flow spreadsheet to make sure that it isn't choking itself to death, if it is then it needs larger valves, then it is to the drafting board to layout new valves and ports then back to the valve flow spreadsheet to adjust the parameters such as the larger clearance volume, then go around and around many, many times.
I should note that I am rather far removed from being a highly experienced steam guy, I have spent a lot of time devouring information, yes, but there is still a lot to learn.
One thing I do know for sure is that the 8 to 9 lbs per hp hr that I stated in my earlier post, regarding my calculations as to the counterflow engine, it would be lucky to be a real world 12 to 13 lbs per hp hr. The uniflow engines are the same way, there is a substantial drop from the theoretical efficiency to the dyno efficiency to the ROAD efficiency.
I will link yet again to this:
[www.steamcar.net]
A few quotes from the above link about the SES project.
"The SES system was designed to the practical limit of both parameters, settling eventually on 1000 psi and 1000 ºF to the expander inlet at all times."
"Interestingly, temperature had little effect on performance in the range above some 750 ºF."
"Certainly, SES could have saved itself lots of headaches had it known this from the beginning, because lowering the temperature while receiving the same performance would have eliminated much of that expensive research into exotic lubricants and wear properties. Lower temperature would have also made the use of the finicky throttle valve easier, a valve that was hoped would not be necessary but in practice the fully automatic control was too difficult to achieve without it. It also turned out that, contrary to the data of specific steam consumption obtained on a dynamometer, overall efficiency did not suffer all that much in the car, even improved for some operating regimes, at a lower than 1000 psi inlet pressure."
I think that what is happening is that with the very hot steam and the smaller engines the greater heat transfer from the greater temperature differential disallows the theoretical efficiency that 1,000 F steam "should" give.
Also, Skinner didn't employ high temperatures or high pressures, as Ken stated, they used VERY low condenser pressures.
Ken,
One solution for "hard core acceleration" I have come up with is a good oll' bypass valve for the condenser, which lets the engine "bark" with the best of them! Just strap in a larger water tank. Also, where in the heck can someone in the U.S.A. maintain 200 mph for more then 2 miles on the open road? How about 150 mph on the open road, 2 what 3 miles at most. 100 mph, well on the interstate for a while or in the middle of the desert.
The above sounds sarcastic, but seriously, really seriously, what are your, or anyone else's who is reading this, estimate as to the greatest distance that one could travel in America on a regular basis at 100 mph, 150 mph and 200 mph? The answers to those questions dictate to a massive extent the ratio or maximum firing rate to the reserve capacity of the boiler.
My work with the above mentioned spreadsheet has shown me that outputs well above 1,000 hp are possible with enough reserve and a moderate firing rate, that would be with a relatively large engine, boiler pressure of 2,000 psi and tire spinning torque at 500 psi and 70% cutoff, that would give 1/4 mile times and speeds that would have to be seen to be believed and would be better then any car licensed to drive on the road. That is no joke or exaggeration.
At sustained high outputs condensing becomes a big issue, the best solution I have found with my feed water heater and condenser spreadsheets is to abandon the "fan" because at those speeds it is called a "propeller" and use an Exair or similar air accelerator with tight ducting for the condenser(s) and an engine exhaust powered air pump to power the air accelerators.
As for the boiler, yet again the spreadsheets, I have written dozens of them over the years, with various tube layouts and configurations. I firmly believe that it is safe to say that an overall heat transfer coefficient of 75,000 btu sq ft per hour is more then possible with a forced re-circulation type boiler. That includes the superheater, evaporation and economizer sections of the boiler. If anyone reading this gets around to writing their own boiler analysis spreadsheet, I have found that the minimum step rate for each pass of tubes is 100. That is 100 calculations for the heat transfer per bank of tubes. Any less and the output from the spreadsheet is deranged. I calculate the temp rise of the tubes and influence of the flow rates on the heat transfer of the superheater and the economizer, when I was investigating finned tubing I even had it calculate the fin temp rise. After a lot of work I don't see that fins, from an overall standpoint, are of a great advantage. At very high outputs the fin rise makes then uneconomical and bulky. The heat transfer to boiling water is extremely complex so I use Wolverine tubes db3cal spreadsheet to cross reference my boiler designs. I will attach their program to this post since they supply it for free to anyone. I also use free source X Steam for Excel for the steam properties, I will also attach it. I used those two programs and old engineering books as the source material for the spreadsheets. . . that and a LOT of time.
I wish I could supply my more complex spreadsheets to people, but they are similar to someones cluttered desk, that person knows where everything is, but nobody else does. . .
*($% the db3cal program is too large to attach here, so here is the link to their website to download it [www.wlv.com]
Caleb Ramsby
Edited 1 time(s). Last edit at 01/14/2015 02:05AM by Caleb Ramsby.
|
Re: Design of Reciprocating Piston Expanders January 14, 2015 11:12AM |
Registered: 15 years ago Posts: 180 |
this is an interesting development:
[www.launchpnt.com]
it's a cam-less engine. I get how it works, it's pretty clever. the two springs actuate on a linear cam, so as soon as the voice coil moves the valve, it will snap into the opposite position, without any electricity. So holding the valve in either open or closed position consumes no electricity.
I think it could be adapted to a steam engine. probably longer valve stems to keep the coils cool
Dan
[www.launchpnt.com]
it's a cam-less engine. I get how it works, it's pretty clever. the two springs actuate on a linear cam, so as soon as the voice coil moves the valve, it will snap into the opposite position, without any electricity. So holding the valve in either open or closed position consumes no electricity.
I think it could be adapted to a steam engine. probably longer valve stems to keep the coils cool
Dan
|
Re: Design of Reciprocating Piston Expanders January 14, 2015 12:38PM |
Registered: 19 years ago Posts: 238 |
Hi Dan
In your original post you asked 'Why oh why would anyone go back to trying single piston single stroke steam engines?'
The answer is that a single cylinder is less expensive for prototyping, just read the page you have posted. They put all that technology on a single cylinder Rotax 500cc engine!
The electronic latching valve has possibilities but consider the following.
1) The company indicates the valve at present has a 5000 rpm limit (4 stroke), that equates to only 2500 rpm for a steam engine (2 stroke) which is not very high for a modern compact high speed design as you propose.
2) When the vehicle is beginning to move the mechanism will have to open the poppet valve against throttle pressure, there will be no compression to assist in opening the valve so the coil will have to be much stronger than an IC engine.
3) If the compression pressure increased above the throttle pressure there is a possibility of the poppet valve opening and the latch holding it there. Possible damage from contact with the piston.
4) A modern SI or CI engine requires one or more revolutions for all the position sensors to register and allow the ECU to calculate the timing. If a steam engine is stationary how will the ECU calculate which valves to open to get the correct direction of rotation for the present crankshaft position. Possible but I believe it will require a number of extra sensors, or a starter motor and clutch which for me defeats the point.
Stumpf who was one of the fathers of the unaflow, designed a number of single acting unaflow automobile engines. Steamobile used them in their road trucks. Not saying that I agree with the design but just making you aware of something other than ship engines.
Brian
P.S. Thank you Ken and Caleb for the fantastic information and insight which makes this forum the best education for modern steam automobile enthusiasts.
In your original post you asked 'Why oh why would anyone go back to trying single piston single stroke steam engines?'
The answer is that a single cylinder is less expensive for prototyping, just read the page you have posted. They put all that technology on a single cylinder Rotax 500cc engine!
The electronic latching valve has possibilities but consider the following.
1) The company indicates the valve at present has a 5000 rpm limit (4 stroke), that equates to only 2500 rpm for a steam engine (2 stroke) which is not very high for a modern compact high speed design as you propose.
2) When the vehicle is beginning to move the mechanism will have to open the poppet valve against throttle pressure, there will be no compression to assist in opening the valve so the coil will have to be much stronger than an IC engine.
3) If the compression pressure increased above the throttle pressure there is a possibility of the poppet valve opening and the latch holding it there. Possible damage from contact with the piston.
4) A modern SI or CI engine requires one or more revolutions for all the position sensors to register and allow the ECU to calculate the timing. If a steam engine is stationary how will the ECU calculate which valves to open to get the correct direction of rotation for the present crankshaft position. Possible but I believe it will require a number of extra sensors, or a starter motor and clutch which for me defeats the point.
Stumpf who was one of the fathers of the unaflow, designed a number of single acting unaflow automobile engines. Steamobile used them in their road trucks. Not saying that I agree with the design but just making you aware of something other than ship engines.
Brian
P.S. Thank you Ken and Caleb for the fantastic information and insight which makes this forum the best education for modern steam automobile enthusiasts.
|
Re: Design of Reciprocating Piston Expanders January 14, 2015 01:26PM |
Registered: 15 years ago Posts: 641 |
Brian McMorran Wrote:
-------------------------------------------------------
> Hi Dan
>
> In your original post you asked 'Why oh why would
> anyone go back to trying single piston single
> stroke steam engines?'
> The answer is that a single cylinder is less
> expensive for prototyping, just read the page you
> have posted. They put all that technology on a
> single cylinder Rotax 500cc engine!
>
> The electronic latching valve has possibilities
> but consider the following.
>
> 1) The company indicates the valve at present has
> a 5000 rpm limit (4 stroke), that equates to only
> 2500 rpm for a steam engine (2 stroke) which is
> not very high for a modern compact high speed
> design as you propose.
Hi all:
Many interesting points to ponder here!
As a reminder I provide a link to an interesting youtube clip: www.youtube.com/watch?v=ccjTMQwKWNs
--the Dan Gelbart engine addresses several issues recently discussed, including electric control over intake valve duration and compression relief during the exhaust cycle.
BTW as our illustrious President has pointed out, this has already been tried (many years ago) in spite of the patent that Gelbart has managed to get.
FWIW--I just wanted to make sure that all active participants in this thread were aware.
Bill
Edited 1 time(s). Last edit at 01/14/2015 01:43PM by Bill Hinote.
-------------------------------------------------------
> Hi Dan
>
> In your original post you asked 'Why oh why would
> anyone go back to trying single piston single
> stroke steam engines?'
> The answer is that a single cylinder is less
> expensive for prototyping, just read the page you
> have posted. They put all that technology on a
> single cylinder Rotax 500cc engine!
>
> The electronic latching valve has possibilities
> but consider the following.
>
> 1) The company indicates the valve at present has
> a 5000 rpm limit (4 stroke), that equates to only
> 2500 rpm for a steam engine (2 stroke) which is
> not very high for a modern compact high speed
> design as you propose.
Hi all:
Many interesting points to ponder here!
As a reminder I provide a link to an interesting youtube clip: www.youtube.com/watch?v=ccjTMQwKWNs
--the Dan Gelbart engine addresses several issues recently discussed, including electric control over intake valve duration and compression relief during the exhaust cycle.
BTW as our illustrious President has pointed out, this has already been tried (many years ago) in spite of the patent that Gelbart has managed to get.
FWIW--I just wanted to make sure that all active participants in this thread were aware.
Bill
Edited 1 time(s). Last edit at 01/14/2015 01:43PM by Bill Hinote.
|
Re: Design of Reciprocating Piston Expanders January 14, 2015 03:30PM |
Admin Registered: 19 years ago Posts: 2,120 |
As a guy who works for General Motors Powertrain World Headquarters, I can tell you very often companies developing new technologies (versus evolutionary updates) use single cylinder engines for the development work. They have been very common in the auto industry since they started doing fuel standardization testing back in the 1920s. In fact, at least two companies specifically supply engines just for the purpose,
FEV
Ricardo
There are any number of reasons for this. Fewer cylinders limit the number of variables you have to keep track of when doing basic testing. The single cylinder research engines are highly modularized to a degree that would be counterproductive in production engines, this allows for fast and simple fabrication of very specific configuration engines. First cost is going to be lower. Reduced fabrication gets your testing done more rapidly.
When I wrote up the Janicki proposal I specified a single cylinder test engine to speed up and improve the quality of development:
Janicki proposal
Single cylinder research engines are simply how the professionals do things.
FEV
Ricardo
There are any number of reasons for this. Fewer cylinders limit the number of variables you have to keep track of when doing basic testing. The single cylinder research engines are highly modularized to a degree that would be counterproductive in production engines, this allows for fast and simple fabrication of very specific configuration engines. First cost is going to be lower. Reduced fabrication gets your testing done more rapidly.
When I wrote up the Janicki proposal I specified a single cylinder test engine to speed up and improve the quality of development:
Janicki proposal
Single cylinder research engines are simply how the professionals do things.
|
Re: Design of Reciprocating Piston Expanders January 14, 2015 03:56PM |
Registered: 19 years ago Posts: 1,268 |
Hi,
Certainly a lot of great information coming out here.
I am just finishing the stuff that Lohring first posted. They sure tested a lot of things that didn't work. From the above discussions it seems that our worst design problem is in both cylinder lubrication at high temperatures and piston ring sealing at such high mean effective pressures.
A properly designed piston ring "should" seal with just a small amount of pressure against the cylinder wall. The pressure at the back of the ring pushing it to the cylinder wall and the pressure between the ring and the cylinder wall pushing the ring inward to the piston "should" almost balance out with just enough extra pressure against the cylinder wall to make the seal.
If the ring seals at the top then all of the back pressure is pushing the ring into the cylinder wall. If it seals at the bottom then very little back pressure is pushing the ring into the cylinder wall. As a ring wears I imagine that it would seal more at the top of the ring and increase the pressure pushing it into the cylinder wall. Many production rings have different surface profiles from square or tapered to barrel shaped.
Ken, of course I am not giving up on my compound engine, however I am seeing more possible troubles with rings and lubrication than I had planed on. This is a problem, regardless of the type of engine, that I believe will only be solved with laboratory and test bench. With my big cylinder diameters for the low pressure stage of the compound I am reluctant to give up hydrodynamic lubrication.
As I mentioned before with boundary lubrication having a coefficient of friction of about 0.2 and hydrodynamic lubrication having a coefficient of friction of about 0.004 there is a difference of about 50 times. With big piston diameters this is important. One piston vs. fifty of them for the same friction.
Best,
Bill G.
PS Bill H. what does FWIW mean?
Certainly a lot of great information coming out here.
I am just finishing the stuff that Lohring first posted. They sure tested a lot of things that didn't work. From the above discussions it seems that our worst design problem is in both cylinder lubrication at high temperatures and piston ring sealing at such high mean effective pressures.
A properly designed piston ring "should" seal with just a small amount of pressure against the cylinder wall. The pressure at the back of the ring pushing it to the cylinder wall and the pressure between the ring and the cylinder wall pushing the ring inward to the piston "should" almost balance out with just enough extra pressure against the cylinder wall to make the seal.
If the ring seals at the top then all of the back pressure is pushing the ring into the cylinder wall. If it seals at the bottom then very little back pressure is pushing the ring into the cylinder wall. As a ring wears I imagine that it would seal more at the top of the ring and increase the pressure pushing it into the cylinder wall. Many production rings have different surface profiles from square or tapered to barrel shaped.
Ken, of course I am not giving up on my compound engine, however I am seeing more possible troubles with rings and lubrication than I had planed on. This is a problem, regardless of the type of engine, that I believe will only be solved with laboratory and test bench. With my big cylinder diameters for the low pressure stage of the compound I am reluctant to give up hydrodynamic lubrication.
As I mentioned before with boundary lubrication having a coefficient of friction of about 0.2 and hydrodynamic lubrication having a coefficient of friction of about 0.004 there is a difference of about 50 times. With big piston diameters this is important. One piston vs. fifty of them for the same friction.
Best,
Bill G.
PS Bill H. what does FWIW mean?
|
Re: Design of Reciprocating Piston Expanders January 14, 2015 04:34PM |
Registered: 19 years ago Posts: 238 |
Hello Bill
Good video from Dan Gelbart and makes me want to investigate the subject more.
I am not dismissing electronics in any way because it opens up many avenues when experimenting with an engine on the dynamometer. However for a production engine that is a different matter, at least for my particular interests.
A good friend designed a mechanical hydraulic valve operation for his relatively slow speed engine but still had the problem when there was leakage, how could the valves be reset simply without adding to the complexity or cranking the engine?
I wondered about electro-hydraulic operation for engine testing, again limited to slow speed and problematic for static starts.
Brian
Good video from Dan Gelbart and makes me want to investigate the subject more.
I am not dismissing electronics in any way because it opens up many avenues when experimenting with an engine on the dynamometer. However for a production engine that is a different matter, at least for my particular interests.
A good friend designed a mechanical hydraulic valve operation for his relatively slow speed engine but still had the problem when there was leakage, how could the valves be reset simply without adding to the complexity or cranking the engine?
I wondered about electro-hydraulic operation for engine testing, again limited to slow speed and problematic for static starts.
Brian
|
Re: Design of Reciprocating Piston Expanders January 14, 2015 06:40PM |
Registered: 19 years ago Posts: 657 |
Hey Guys,
As for valves and their movement. All of this searching for the "magical" valve and valve gear is in my opinion nonsense. If using a poppet valve just use a camshaft, if using a piston, D or Corliss valve use the Walschearts, Joy or Stephenson type of valve gear. My personal preference is the piston type with the Walschearts valve gear.
With slide valves the key is to have a large steam lap, this gives a larger valve opening for any given cutoff, the key with the piston valve is to make it BIG and make the ports STRAIGHT AND SHORT.
Use the heck out of Professor Halls program that I linked to in my earlier post and see what I am talking about.
Also read this discussion about the differences between the Caprotti actuated poppet valves and the Walschearts operated piston valve.
[5at.co.uk]
Caleb Ramsby
As for valves and their movement. All of this searching for the "magical" valve and valve gear is in my opinion nonsense. If using a poppet valve just use a camshaft, if using a piston, D or Corliss valve use the Walschearts, Joy or Stephenson type of valve gear. My personal preference is the piston type with the Walschearts valve gear.
With slide valves the key is to have a large steam lap, this gives a larger valve opening for any given cutoff, the key with the piston valve is to make it BIG and make the ports STRAIGHT AND SHORT.
Use the heck out of Professor Halls program that I linked to in my earlier post and see what I am talking about.
Also read this discussion about the differences between the Caprotti actuated poppet valves and the Walschearts operated piston valve.
[5at.co.uk]
Caleb Ramsby
|
Re: Design of Reciprocating Piston Expanders January 14, 2015 06:54PM |
Registered: 11 years ago Posts: 202 |
Caleb Ramsby Wrote:
-------------------------------------------------------
> RESOURCE LINK
>
> [oac.cdlib.org]
> /
>
> The above link is to the Bancroft library and
> gives one access to free online-available note and
> diary's of Abner Doble from his earliest
> experiments up to his work on the failed Ultimax.
>
>
> Having, over the years, read through all of those
> papers 3 or more times, I can say that it is the
> most important information regarding light steam
> power available on the internet. More so for what
> failed and why, then what worked.
>
> Don't just read it, STUDY IT!
>
> Caleb Ramsby
Thank you for posting this link. There's enough information there for several engineering courses. It will take me a long time for even a first pass through.
Lohring Miller
-------------------------------------------------------
> RESOURCE LINK
>
> [oac.cdlib.org]
> /
>
> The above link is to the Bancroft library and
> gives one access to free online-available note and
> diary's of Abner Doble from his earliest
> experiments up to his work on the failed Ultimax.
>
>
> Having, over the years, read through all of those
> papers 3 or more times, I can say that it is the
> most important information regarding light steam
> power available on the internet. More so for what
> failed and why, then what worked.
>
> Don't just read it, STUDY IT!
>
> Caleb Ramsby
Thank you for posting this link. There's enough information there for several engineering courses. It will take me a long time for even a first pass through.
Lohring Miller
|
Re: Design of Reciprocating Piston Expanders January 14, 2015 07:12PM |
Registered: 15 years ago Posts: 641 |
lohring Wrote:
-------------------------------------------------------
> Caleb Ramsby Wrote:
> --------------------------------------------------
> -----
> > RESOURCE LINK
> >
> >
> [oac.cdlib.org]
>
> > /
> >
> > The above link is to the Bancroft library and
> > gives one access to free online-available note
> and
> > diary's of Abner Doble from his earliest
> > experiments up to his work on the failed
> Ultimax.
> >
> >
> > Having, over the years, read through all of
> those
> > papers 3 or more times, I can say that it is
> the
> > most important information regarding light
> steam
> > power available on the internet. More so for
> what
> > failed and why, then what worked.
> >
> > Don't just read it, STUDY IT!
Nah--
don't just study it-----------------build it and and find out how it performs in the real world!!
All the rest is just serious BS IMO.
Time to put up,---------------or shut up here!!
B.
-------------------------------------------------------
> Caleb Ramsby Wrote:
> --------------------------------------------------
> -----
> > RESOURCE LINK
> >
> >
> [oac.cdlib.org]
>
> > /
> >
> > The above link is to the Bancroft library and
> > gives one access to free online-available note
> and
> > diary's of Abner Doble from his earliest
> > experiments up to his work on the failed
> Ultimax.
> >
> >
> > Having, over the years, read through all of
> those
> > papers 3 or more times, I can say that it is
> the
> > most important information regarding light
> steam
> > power available on the internet. More so for
> what
> > failed and why, then what worked.
> >
> > Don't just read it, STUDY IT!
Nah--
don't just study it-----------------build it and and find out how it performs in the real world!!
All the rest is just serious BS IMO.
Time to put up,---------------or shut up here!!
B.
|
Re: Design of Reciprocating Piston Expanders January 14, 2015 07:45PM |
Registered: 19 years ago Posts: 657 |
Hey B.
As far as I know the only one who has ever posted here that has produced their own car of their own design that was capable of real world performance was Ted Pritchard. I don't think that Carter ever posted here, nor Barrett. Not sure though since people don't always use their names as uhh, their name. All of the above are dead.
Everyone else here that has made things, have either cobbled together other peoples stuff or made small slow things. The best of the best are the Vanderbilt Cup Replicas with the Bryan engines in them, the creator of them never posted here.
I couldn't agree with you more about the BS of theory vs practice! That is precisely why I have not posted anything here in over two years and am extremely reluctant to post anything more.
However, if one ignores the how and why of the prior arts success's and failures then one is simply an idiot and a fool.
Caleb Ramsby
As far as I know the only one who has ever posted here that has produced their own car of their own design that was capable of real world performance was Ted Pritchard. I don't think that Carter ever posted here, nor Barrett. Not sure though since people don't always use their names as uhh, their name. All of the above are dead.
Everyone else here that has made things, have either cobbled together other peoples stuff or made small slow things. The best of the best are the Vanderbilt Cup Replicas with the Bryan engines in them, the creator of them never posted here.
I couldn't agree with you more about the BS of theory vs practice! That is precisely why I have not posted anything here in over two years and am extremely reluctant to post anything more.
However, if one ignores the how and why of the prior arts success's and failures then one is simply an idiot and a fool.
Caleb Ramsby
|
Re: Design of Reciprocating Piston Expanders January 14, 2015 09:05PM |
Registered: 15 years ago Posts: 641 |
Caleb Ramsby Wrote:
> I couldn't agree with you more about the BS of
> theory vs practice! That is precisely why I have
> not posted anything here in over two years and am
> extremely reluctant to post anything more.
>
> However, if one ignores the how and why of the
> prior arts success's and failures then one is
> simply an idiot and a fool.
Hi Caleb:
There are so many dreamers here!
I hope that we don't extinguish the dreamers--they provide us with the hope to succeed, beyond our dreams.
There are also a lot of realists who might take the current "state of the art" and move it forward a little bit from where it currently stands. I think I might be counted among those--but will allow others to assess as such.
To all: Please do NOT quash the dreams of those who dream beyond our current realities!!
Happy dreams to all,
Bill H.
> I couldn't agree with you more about the BS of
> theory vs practice! That is precisely why I have
> not posted anything here in over two years and am
> extremely reluctant to post anything more.
>
> However, if one ignores the how and why of the
> prior arts success's and failures then one is
> simply an idiot and a fool.
Hi Caleb:
There are so many dreamers here!
I hope that we don't extinguish the dreamers--they provide us with the hope to succeed, beyond our dreams.
There are also a lot of realists who might take the current "state of the art" and move it forward a little bit from where it currently stands. I think I might be counted among those--but will allow others to assess as such.
To all: Please do NOT quash the dreams of those who dream beyond our current realities!!
Happy dreams to all,
Bill H.
|
Re: Design of Reciprocating Piston Expanders January 14, 2015 09:48PM |
Registered: 19 years ago Posts: 657 |
Sorry, only registered users may post in this forum.
All files from this thread
| File Name | File Size | Posted by | Date | ||
|---|---|---|---|---|---|
| Surface area and clearance % to bore and stroke.xls | 333 KB | open | download | Caleb Ramsby | 01/13/2015 | Read message |
| XSteam_Excel_v2.6_US.xls | 397 KB | open | download | Caleb Ramsby | 01/14/2015 | Read message |
| X_Steam_Tables.bas | 153.6 KB | open | download | Caleb Ramsby | 01/14/2015 | Read message |
| Uniflow Engine Section Details.png | 218.8 KB | open | download | lohring | 01/19/2015 | Read message |
