Tom, you make some xlnt comments on the SV-2, but these cover SE in general. Indeed, the whole spin on isothermal heating & cooling is...bogus. However, hearing such from you, is like a light in the darkness. But, isn't this raining on the parade, and what you've complained about me doing ???
No doubt, the iosothermal processes combined with regen severely hampers any SE design. From a theoretical view, the 4 main processes (isochoric, isobaric, isothermal, adiabatic) limited by a 4 process cycle with 2 prs of alternating processes, only yield 6 potential cycles: Stirling, Ericsson, Otto, Brayton, Carnot, and invalid (so far) cycle with sq PV plot of 2 isochoric & 2 isobaric processes. However, loosening this 'alternating 2 process' limit (within 4 process limit) allows for Atkinson & Diesel (the academics consider Diesel as isobaric heating vs isochoric Otto heating).
Conventional wisdom is to consider these cycles as simply isothermal or adiabatic, whereby Stirling & Ericsson are isothermal, and Otto & Brayton are adiabatic. Carnot is left for textbook chatter due to low output, and Diesel & Atkinson are considered adiabatic versions of Otto. OK, in a few lines, I've nailed nearly all hot gas cycles, but this conventional view exposes that the 2 isothermal cycles - Srirling & Ericsson - are also regen cycles. I'm still OK with this conclusion where 'all iosthermal cycles are regen cycles', BUT...all regen cycles are not isothermal, since Brayton can be regen, whether low end piston scheme like Proeschel pitched or high end (tho limited) CCGT.
I never had much interest in cutting metal for a fancy paperweight, so I spent years quantifying these cycles by process via ratio analysis (aka RA) and graphic analysis (aka GA). In the SE community, guys like Senft, Urieli (et al) honed in a thing called 'load' which is the ratio of input heat to regen heat, but they never found a simple way to express this (across a range of cycles). That post I did on how much is regen heat attempts to address this, but appears to have flown under the radar. The point is that regen heat exceeds input in common SE designs when the compression ratio hovers around the thermal ratio. The only way for Stirling or Ericsson to decrease load is to have a compression ratio far greater than thermal ratio, whereby input heat greatly exceeds regen heat. So, a low load design would have a PV plot with a long 'banner' like plot (think LTD) vs short 'flag' like PV plot (think textbook SE).
Note, that this has noting to do with any particular mech, open vs closed cycle, pressurization, etc, but further analysis will expose the effect/s of compression ratio on cycle and engine. As I like to say, simple cutting metal until the old lady's on your back or the grant moola dries up, doesn't cut it...as many guys have found out. I agree with exploring beyond isothermal cycles, and tend to focus on adiabatic cycles, but these require a relatively higher compression ratio than anything Stirling. The good news is we're surround by adiabatic cycles (Otto, Diesel, Brayton, a few Atkinson), but the bad news remains...how does one get the heat in externally, and how does one make a simple mech that addresses the relatively high compression ratio req'd ???
The only engine I ever saw Arch pitch was something he named the Nascent Engine akin Tailer's thermal lag. Arch pitched this 20 yrs ago, before the internet exploded. Meanwhile, few new that Tailer's patent stemmed from the 1987 ORNL paper by Chen & West (yep, that's Colin West) but I had a back channel to ORNL, so chuckled over Tailer's patent. Interestingly, Tailer pitched his thermal lag as a pseudo Otto, but I consider it a pseudo Carnot which is much more limited (like LTD, Gloy, Warbrooke, etc).
matt brown wrote: ↑Sat Dec 17, 2022 4:11 pm
Tom, you make some xlnt comments on the SV-2, but these cover SE in general. Indeed, the whole spin on isothermal heating & cooling is...bogus. However, hearing such from you, is like a light in the darkness. But, isn't this raining on the parade, and what you've complained about me doing ???
No doubt, the iosothermal processes combined with regen severely hampers any SE design. From a theoretical view, the 4 main processes (isochoric, isobaric, isothermal, adiabatic) limited by a 4 process cycle with 2 prs of alternating processes, only yield 6 potential cycles: Stirling, Ericsson, Otto, Brayton, Carnot, and invalid (so far) cycle with sq PV plot of 2 isochoric & 2 isobaric processes. However, loosening this 'alternating 2 process' limit (within 4 process limit) allows for Atkinson & Diesel (the academics consider Diesel as isobaric heating vs isochoric Otto heating).
Conventional wisdom is to consider these cycles as simply isothermal or adiabatic, whereby Stirling & Ericsson are isothermal, and Otto & Brayton are adiabatic. Carnot is left for textbook chatter due to low output, and Diesel & Atkinson are considered adiabatic versions of Otto. OK, in a few lines, I've nailed nearly all hot gas cycles, but this conventional view exposes that the 2 isothermal cycles - Srirling & Ericsson - are also regen cycles. I'm still OK with this conclusion where 'all iosthermal cycles are regen cycles', BUT...all regen cycles are not isothermal, since Brayton can be regen, whether low end piston scheme like Proeschel pitched or high end (tho limited) CCGT.
I never had much interest in cutting metal for a fancy paperweight, so I spent years quantifying these cycles by process via ratio analysis (aka RA) and graphic analysis (aka GA). In the SE community, guys like Senft, Urieli (et al) honed in a thing called 'load' which is the ratio of input heat to regen heat, but they never found a simple way to express this (across a range of cycles). That post I did on how much is regen heat attempts to address this, but appears to have flown under the radar. The point is that regen heat exceeds input in common SE designs when the compression ratio hovers around the thermal ratio. The only way for Stirling or Ericsson to decrease load is to have a compression ratio far greater than thermal ratio, whereby input heat greatly exceeds regen heat. So, a low load design would have a PV plot with a long 'banner' like plot (think LTD) vs short 'flag' like PV plot (think textbook SE).
Note, that this has noting to do with any particular mech, open vs closed cycle, pressurization, etc, but further analysis will expose the effect/s of compression ratio on cycle and engine. As I like to say, simple cutting metal until the old lady's on your back or the grant moola dries up, doesn't cut it...as many guys have found out. I agree with exploring beyond isothermal cycles, and tend to focus on adiabatic cycles, but these require a relatively higher compression ratio than anything Stirling. The good news is we're surround by adiabatic cycles (Otto, Diesel, Brayton, a few Atkinson), but the bad news remains...how does one get the heat in externally, and how does one make a simple mech that addresses the relatively high compression ratio req'd ???
The only engine I ever saw Arch pitch was something he named the Nascent Engine akin Tailer's thermal lag. Arch pitched this 20 yrs ago, before the internet exploded. Meanwhile, few new that Tailer's patent stemmed from the 1987 ORNL paper by Chen & West (yep, that's Colin West) but I had a back channel to ORNL, so chuckled over Tailer's patent. Interestingly, Tailer pitched his thermal lag as a pseudo Otto, but I consider it a pseudo Carnot which is much more limited (like LTD, Gloy, Warbrooke, etc).
"But, isn't this raining on the parade, and what you've complained about me doing ???"
I don't believe I've complained of you "raining on the parade".
"I agree with exploring beyond isothermal cycles, and tend to focus on adiabatic cycles, but these require a relatively higher compression ratio than anything Stirling"
What does adiabatic have to do with compression ratio one way or the other? Nothing whatsoever.
My complaint about your posts is mostly that you seem to string a bunch of big words together with no apparent comprehension of the actual meaning. Your posts make little if any sense but you then pretend others are ignorant of the terminology. Your posts simply do not make sense (it seems to me anyway) and I find trying to untangle whatever point you might be trying to make, more often than not a big waste of time.
Nothing personal, but do you know the meaning of the word "adiabatic". You don"r think it can apply to a Stirling because they don't have a high enough compression ratio?
What?
Cloud formation is adiabatic, or a result of an adiabatic process. Compression ratio has little if anything to do with it.
"What does adiabatic have to do with compression ratio one way or the other? Nothing whatsoever."
An adiabatic process has very distinct inter related PVT values that vary by the gamma of the gas. I almost hate to inject another...big word...but gamma is the thermo term for the ratio of constant pressure heat capacity to constant volume heat capacity or in thermo buzz gamma=Cp/Cv. The monatomic gases are considered 1.4 and the diatomic are considered 1.6, tho both are close approximations. So, diatomic air has gamma=1.4 while monatomic helium has gamma=1.6, and no, I don't make this stuff up (see endless thermo youtube videos if doubting thomas). Pre depression, it was req'd ME homework to make an 'adiabatic index' (table) where the dT and dP were determined vs dV. Constructed correctly, this table has dV listed in a left side column akin compression ratio (1,2,3,4,5,6...) and then 2 adjacent columns for corresponding dP & dT, and usually Profs would only require a mono and dia table. Thus, ME students learned the PVT relationships for adiabatic processes and how to calculate them...merely apply correct gamma exponent to dV in question and viola.
Once you get this PVT relationship, you'll see why any adiabatic cycle loves a high compression ratio (ICE or ECE). Unless you dream up some weird cycle, any adiabatic cycle will follow known ICE where heat input is either constant volume or constant pressure, and constant pressure heat input gets a strike against it right off, due to dissimilar compression vs expansion volumes (thermal ratio function which further complicate most mech and cycle schemes).
A diesel engine requires a high compression ratio for a fairly obvious reason: Diesel engines rely on the heat of compression for ignition. Adiabatic compression helps to accomplishes that. Isothermal, obviously would not.
There is however no intrinsic relation binding adiabatic exclusively with high compression as you have implied.
What do you think the term adiabatic means actually?
"Once you get this PVT relationship, you'll see why any adiabatic cycle loves a high compression ratio (ICE or ECE)"
Really? PV and T are state variables. "Adiabatic" is a process.
An adiabatic process does not require a high compression ratio. It doesn't "love" a high compression ratio.
Stirling engines don't have fuel requiring a high compression ratio for ignition. That does not mean compression and expansion in a Stirling cannot be adiabatic. You seem to be conflating "high compression" with "adiabatic" as if one had to go hand in hand with the other, just because you saw the words used together in a textbook.
"the relatively high compression ratio req'd ??"
An adiabatic process does not require a high compression ratio
I suppose my compressor conversion proposal kind of veers off the beaten path so there is some question regarding how it, whatever it ends up looking like, should be classified within engine taxonomy.
Personally I'd avoid thermodynamics terminology when possible. Not because I don't understand it but because it is mostly outdated, from an actual physics point of view.
Thermodynamics grew out of the whole Carnot & company milieu that viewed heat as a "thing" or substance rather than just another form of energy.
Take adiabatic.
The definition of the word necessitates making a distinction between heat and "work".
What however is the actual distinction between heat and work in light of the kinetic theory of heat?
When a gas molecule bumps into a piston it can transfers "work" if it is an adiabatic process but not heat.
I suppose, in order to be able to communicate at all we need words with agreed upon meanings and classification of ideas and whatnot but personally I have no interest in debating such issues. Much of it is imaginary nonsense, making hard distinctions where there really isn't any.
Wow, that post was a challenge due to wrap issue creeping in...
Anyways, Tom, that's how inter related adiabatic PVT values are for gases. The main problem for an adiabatic ECE cycle will remain getting enough heat in and how, since the only way will be constant volume heating or constant pressure heating. An open cycle might nix typical 'cooling' restraints, and irregular 3 or 5 legged cycle might reduce mechanical restraints, but irregular cycles tend to suffer from both lower output and lower efficiency.
matt brown wrote: ↑Sun Dec 18, 2022 5:49 pm
Wow, that post was a challenge due to wrap issue creeping in...
Anyways, Tom, that's how inter related adiabatic PVT values are for gases. The main problem for an adiabatic ECE cycle will remain getting enough heat in and how, since the only way will be constant volume heating or constant pressure heating. An open cycle might nix typical 'cooling' restraints, and irregular 3 or 5 legged cycle might reduce mechanical restraints, but irregular cycles tend to suffer from both lower output and lower efficiency.
I'm trying to deduce exactly what it is you're trying to say, or prove.
Just so maybe we can have some common nomenclature for discussing ... whatever, I'll just ask again:
What do you think the term adiabatic means actually?
There is an old post in here somewhere.
To quote Geoff V: "...unlike most academic theory, which considers the heat transfer to be isothermal, in practice, due to the speed of the cycle, the heat transfer is almost certainly adiabatic, an experiment carried out by John Archibald demonstrates this to my satisfaction ..."
I've already posted video of what I assume to be said experiment.
I guess you can disagree. As pointed out, the whole concept of "isothermal" expansion or contraction is mostly mythology in the first place. The actual reality is that a typical external combustion engine (Stirling) is already operating with mostly adiabatic expansion and compression.
Somehow though, you present this (an ECE operating adiabatically) as some kind of impossible to solve problem.
matt brown wrote: ↑Sun Dec 18, 2022 5:49 pm
Wow, that post was a challenge due to wrap issue creeping in...
Anyways, Tom, that's how inter related adiabatic PVT values are for gases. The main problem for an adiabatic ECE cycle will remain getting enough heat in and how, since the only way will be constant volume heating or constant pressure heating. An open cycle might nix typical 'cooling' restraints, and irregular 3 or 5 legged cycle might reduce mechanical restraints, but irregular cycles tend to suffer from both lower output and lower efficiency.
I think what you may be overlooking is the "work" element.
Don't know what you mean exactly by "cooling restraint", but with adiabatic expansion cooling takes place as a result of work output rather than heat removal, if that was not obvious already. I've been saying the same thing since 2012. With adiabatic expansion the volume increases, so the heat is more spread out and there is a temperature drop, but what your table doesn't account for, apparently, is WORK output. Heat transfered out, not as heat but as work, which results in an additional drop in temperature.
With enough cooling as a result of adiabatic expansion with work output there is no need for additional cooling by conduction of heat to a sink. No 'cooling restraint'.
Then as a result of the adiabatic cooling, the gas contracts and compression work is performed by atmospheric pressure, putting heat back into the system. The "heat of compression" from work done on the gas by outside atmospheric pressure.
Due to the low temperature that was produced by adiabatic expansion, there is a very great ∆T and completely contrary to academic silliness, there is rapid heat addition (from the heat source at the hot end) during adiabatic compression. (Actually more like contraction as a result of the cooling during expansion + work output but after that, actual compression by atmospheric pressure + work input nonetheless).
This is about the opposite of academia which asserts that isothermal heat rejection takes place during compression. That, IMO is clearly impossible given that the temperature of the gas has dropped low enough for it to contract.
If the gas is contracting as a result of it becoming cold then it can only be taking in heat.
I don't think standard thermodynamic modeling even has any proper terminology to describe such a process where compression is accompanied by simultaneous heat input. "Isothermal heat rejection" is certainly not appropriate as heat is being taken in, both as heat and as work. "Adiabatic compression" is at least half right sort of, because no heat is going out, but strictly speaking "adiabatic" compression does not allow for or generally even consider the possibility of heat input during compression.
True, the gas is getting hotter due to work input and compression, but it is still much colder than the heat source, so would also be rapidly taking in heat.
All these simultaneous heat inputs combine at TDC.
By definition, an adiabatic process has no heat 'flow'. Geoff V post used poor wording on adiabatic heat transfer, but OK later on. Nevertheless, I got his meaning, where SE compression & expansion are too fast to be isothermal and thus approach adiabatic.
Consider a modified SE engine/cycle with 4 ideal processes where isothermal expansion is replaced by adiabatic expansion. With no heat input during adiabatic expansion, less Wpos is generated (per cycle). However, what isn't apparent without calcs is that this modified cycle is less efficient than similar ideal Stirling. Yep, less power AND less eff per cycle, and provided a similar adiabatic compression could be shoe horned into scheme in lieu of isothermal compression, this 'Otto' version would have even less power and eff (per cycle).
I'm starting to see what you're scheming, but here's how I see it...regardless of expansion process, (1) Wpos ends when engine pressure equals ambient pressure, and ambient pressure will limit isobaric compression via dT end of expansion vs ambient (2) further expansion is possible depending upon mech, but would be Wneg, tho might gain some advantage depending upon scheme (like a flame licker)
matt brown wrote: ↑Mon Dec 19, 2022 1:51 am
By definition, an adiabatic process has no heat 'flow'. Geoff V post used poor wording on adiabatic heat transfer, but OK later on. Nevertheless, I got his meaning, where SE compression & expansion are too fast to be isothermal and thus approach adiabatic.
Consider a modified SE engine/cycle with 4 ideal processes where isothermal expansion is replaced by adiabatic expansion. With no heat input during adiabatic expansion, less Wpos is generated * (per cycle). However, what isn't apparent without calcs is that this modified cycle is less efficient than similar ideal Stirling. Yep, less power AND less eff per cycle, and provided a similar adiabatic compression could be shoe horned into scheme in lieu of isothermal compression, this 'Otto' version would have even less power and eff (per cycle).
I'm starting to see what you're scheming, but here's how I see it...regardless of expansion process, (1) Wpos ends when engine pressure equals ambient pressure, and ambient pressure will limit isobaric compression via dT end of expansion vs ambient (2) further expansion is possible depending upon mech, but would be Wneg, tho might gain some advantage depending upon scheme (like a flame licker)
* I wouldn't say "no heat input".
The way I conceptualize it is something like baseball. Atmospheric pressure throws the pitch. Heat input takes the swing. The idea is to combine and concentrate the two forces to get the greatest reaction. Hit the piston out of the park.
The flight of the ball is "adiabatic". It isn't that adding heat is bad or might not be helpful to some degree but generally speaking it's just impossible once the ball is in flight. Running after it swinging the bat is just a waste of energy.
Some of the most efficient old steam engines had valves to cut off steam and limit steam input into the cylinder to a short blast at TDC. That initial blast has to be allowed to fully expand and cool, not just for efficiency but so the piston can return.
Even a turbine really does the same thing. A jet of gas hits one blade which then flies off and then the jet is hitting the next blade or section. (With the possible exception of a Tesla turbine).
Continuing to supply heat after the piston has reached velocity is just stuffing up the return path. It doesn't contribute appreciably to power output. Unless the engine is designed to move very slowly and uses valves or something, but in a Stirling type engine with no valves the displacer is used as a kind of heat valve to limit heat input. (Or by otherwise restricting heat input in some way - thermal lag for example)
Take the "rice engine".
Observably, it seems to run better, with more power when heat input is limited, after being taken off the burner.
Why?
I think, because less heat input results in more effective adiabatic cooling making the return stroke, not only possible and easier but with enough cooling you get positive work input from atmospheric pressure. (If the engine isn't stuffed up by excess heat input during expansion)
Did you watch this video?
Take particular note of his last statement: "the contraction of the gas is also a power stroke".
The recognition that expansion in a Stirling engine tends toward being adiabatic and working with that rather than trying to somehow force or engineer a so-called "ideal" isothermal expansion means, to me, anyway, active cooling, with a water jacket or something is not necessary. In fact, it seems detrimental, robbing power from the expanding gas. More heat wasted and less heat available for work output.
So... does that make the cold side of a Stirling engine redundant or actually harmful? I tend to believe so. It is a form of unnecessary cooling. The cooling should be primarily, if not entirely the result of work output.
The 2nd law says that's impossible, but observation suggests otherwise. A Stirling engine without a flywheel works. It runs too fast for conductive cooling apparently, so cooling via adiabatic expansion with work out appears to be more than adequate.
So why not eliminate the cold side of the displacer chamber and see what happens?
But how?
No point in even trying if it is actually " impossible" so I don't think too many people ever gave it any thought.
My idea is to replace the cold side with a sealed "air spring". But I have no concrete proof yet that this could actually work.
To work, in theory, heat input would need to be limited to short bursts at or near TDC followed by expansion and cooling due to work output (not heat rejection to a sink).
With limited heat input and cooling by work output, atmospheric pressure effects the return "power stroke".
At TDC another short blast of heat and the cycle repeats.
Polish_20221219_130123765.jpg (461.64 KiB) Viewed 2329 times
Theoretically, if heat input is actually *ALL* transformed into work output, then the only thing really preventing such a Tesla type ambient heat powered engine is the lack of any real *perfect* insulation.
As an originally unintended "proof of concept". A few years ago I was doing various experiments trying to retain heat to see how long a Stirling engine would run on a cup of hot water, ice and so forth but while doing that, I remembered something I had speculated about years earlier.
That was, could adiabatic expansion with work output be making the cold side of the engine colder than ambient?
Nobody on the forum thought so at the time. Heat has to be let out through the cold side or the engine would overheat.
Ten years later I got the idea to try it anyway, while filming a different experiment, since I already had the insulation handy.
I fully expected the engine would slow down and stop due to overheating as soon as I covered the "sink"
After recording that, I went back and watched the video with a stop watch to see how much it slowed down. and found that It didn't even slow down, it ran faster!
I did several more experiments with the same results.
My conclusion: my speculation from about 10 years earlier was right. Adiabatic cooling and the conversion of heat into work actually kept the cold side cold. The insulation prevented heat from destroying this refrigerating effect, so insulating the cold side increased the temperature difference, allowing the engine to run better.
That or the engine is acting as a heat pump causing heat to build up on the hot side. (Or both).
Whatever the case the results were consistent. The engines all ran better when I attempted to eliminate the cold "sink".
The Science/Physics forum kicked me out for displaying a supposed "perpetual motion" device, which was against their policy, and they concluded that the insulation increased cooling of the engine by increasing the surface area or something but assumed some such explanation and deeming the phenomenon not worth further investigation and kicked me out.
I think it's worth enough further investigation to set up a dedicated workshop. Maybe I'm just wasting time and money but I don't think so.