High temperature stirling - was: late night testing on home bru hydro (a.e.h)
Morris Dovey
> "Ulysses" <eatm...@spamola.com/> wrote in message
> news:VvmSj.198$aj...@fe093.usenetserver.com...
>> 724 degrees F is pretty serious.
>
> Damn right! Just as a single point of comparison, the steam temperature from
> the naval nuclear reactors I once operated was hundreds of degrees less.
I was mulling over the comparison with naval reactors when it (finally!)
dawned on me that the tube at the trough's focus could actually /be/ the
hot head of a Stirling engine (fluidyne or other type).
Assuming a cold side temperature of 70F (perhaps as low as 55F if water
cooled), this should provide a considerable improvement in hot/cold side
temperature differential.
The new questions are: What might I use as a working fluid for a
fluidyne with that high a hot-side temperature, and do I need to plan on
using something other than air as the working gas?
Hmm. Time to go fire up the software model with new temperature
parameters...
--
Morris Dovey
DeSoto Solar
DeSoto, Iowa USA
http://www.iedu.com/DeSoto/Stirling/Heat.html
daestrom
"Morris Dovey" <mrd...@iedu.com> wrote in message
news:481a757a$0$87075$815e...@news.qwest.net...
> Vaughn Simon wrote:
>> "Ulysses" <eatm...@spamola.com/> wrote in message
>> news:VvmSj.198$aj...@fe093.usenetserver.com...
>
>>> 724 degrees F is pretty serious.
>>
>> Damn right! Just as a single point of comparison, the steam
>> temperature from the naval nuclear reactors I once operated was hundreds
>> of degrees less.
>
> I was mulling over the comparison with naval reactors when it (finally!)
> dawned on me that the tube at the trough's focus could actually /be/ the
> hot head of a Stirling engine (fluidyne or other type).
>
> Assuming a cold side temperature of 70F (perhaps as low as 55F if water
> cooled), this should provide a considerable improvement in hot/cold side
> temperature differential.
>
> The new questions are: What might I use as a working fluid for a fluidyne
> with that high a hot-side temperature, and do I need to plan on using
> something other than air as the working gas?
>
Well, when you consider that water or water/anti-freeze would have to be
seriously pressurized to avoid steam if you wanted to use it as just the
'primary' coolant to carry heat to the actual engine, that would probably be
ruled out. Some oil/organic compounds could be used, but many tend to break
down with high temperatures leaving a gooey, tarry mess.
But you mentioned putting the working fluid right through the collector.
That might be tricking with a long-trough. You can't go too narrow with the
tube due to increased friction, so a long trough collector pipe would
increase the volume of the 'hot side' quite a bit. Just means you need a
'cold side' that is similar in size (well, actually smaller by the ratio of
temperatures). And the 'displacement' of the liquid in the fluidyne would
have to move an 'appreciable amount' of the gas from hot to cold, so with a
large hot and cold, you need a large 'displacement' (either really long
stroke, very wide bore, or combination).
You don't need any sort of phase change for the working fluid (unlike
Rankine cycle), so I should think any gas that behaves close to the ideal
gas law at these temperatures / pressures would suffice. The only reason to
not use air would be if the high-temperature oxygen causes some corrosion
problem. Material selection for the piping and such would seem to be key.
daestrom
Neon John
<daestrom@NO_SPAM_HEREtwcny.rr.com> wrote:
>You don't need any sort of phase change for the working fluid (unlike
>Rankine cycle), so I should think any gas that behaves close to the ideal
>gas law at these temperatures / pressures would suffice. The only reason to
>not use air would be if the high-temperature oxygen causes some corrosion
>problem. Material selection for the piping and such would seem to be key.
I'd start out using a high BP liquid instead of gas. Something such as the
heat transfer oils that are ubiquitous in the petrochem industry. Good for
several hundred degrees at essentially atmospheric pressure, higher with
modest (less than 100psi) pressure.
Though I haven't done the calcs, I suspect that the upper limit will be set by
re-radiation off the back side (ironically, the side pointing toward the sun)
of the absorber, assuming no fancy infrared optics or insulation is deployed.
John
--
John De Armond
See my website for my current email address
http://www.neon-john.com
http://www.johndearmond.com <-- best little blog on the net!
Tellico Plains, Occupied TN
Why the US is losing its competitivve edge:"It used to be that the USA was pretty good at
producing stuff teenaged boys could lose a finger or two playing with."-James Niccol
Morris Dovey
> Well, when you consider that water or water/anti-freeze would have to be
> seriously pressurized to avoid steam if you wanted to use it as just the
> 'primary' coolant to carry heat to the actual engine, that would
> probably be ruled out. Some oil/organic compounds could be used, but
> many tend to break down with high temperatures leaving a gooey, tarry mess.
The original approach was to use the trough only to heat air that flowed
(by natural convection) into the heat exchanger inside the hot head of
the fluidyne. Since the body of the fluidyne was constructed of PVC
(which doesn't seem to tolerate temperatures much above 165F) we had a
bit of a problem with overheating...
> But you mentioned putting the working fluid right through the collector.
> That might be tricking with a long-trough. You can't go too narrow with
> the tube due to increased friction, so a long trough collector pipe
> would increase the volume of the 'hot side' quite a bit. Just means you
> need a 'cold side' that is similar in size (well, actually smaller by
> the ratio of temperatures). And the 'displacement' of the liquid in the
> fluidyne would have to move an 'appreciable amount' of the gas from hot
> to cold, so with a large hot and cold, you need a large 'displacement'
> (either really long stroke, very wide bore, or combination).
On this engine, I think I can keep the inside of the hot head dry - but
there will be at least one gas/fluid interface to deal with.
At the moment, I'm thinking of a single 8' hot head tube at the focus.
The air in this tube will expand into the cold head. I'm thinking that
the cold head can be a couple of 8' lengths of baseboard radiator
fintube located under/behind the trough. At least part of the cold head
will need to be fluid filled (by the fluid displacer) to provide an
expansion/contraction "arena" in which the displacer can move.
Both hot cold heads will need to be connected to a fluid reservoir, to
which the open "tuning" tube also connects. The open end of the tuning
tube is where I plan to extract the power.
> You don't need any sort of phase change for the working fluid (unlike
> Rankine cycle), so I should think any gas that behaves close to the
> ideal gas law at these temperatures / pressures would suffice. The only
> reason to not use air would be if the high-temperature oxygen causes
> some corrosion problem. Material selection for the piping and such
> would seem to be key.
Understood. I think I'd like a fluid that remains liquid (without
boiling) in the 40F - 800F range. I was actually more worried about a
diesel-type ignition with air as the working gas, but I'm not sure
that's a real problem.
Material selection looks like iron pipe for the hot head, copper
fintubes for the cold head, steel for the reservoir, and iron pipe again
for the tuning tube / power connection.
Bruce Richmond
Might want to use multiple base board tubes in parallel to give the
same cross section as the rest of the tubing. Not sure how hot things
will be at the cold head but you may need to take that into
consideration if you are using soder joints.
> Both hot cold heads will need to be connected to a fluid reservoir, to
> which the open "tuning" tube also connects. The open end of the tuning
> tube is where I plan to extract the power.
>
> > You don't need any sort of phase change for the working fluid (unlike
> > Rankine cycle), so I should think any gas that behaves close to the
> > ideal gas law at these temperatures / pressures would suffice. The only
> > reason to not use air would be if the high-temperature oxygen causes
> > some corrosion problem. Material selection for the piping and such
> > would seem to be key.
>
> Understood. I think I'd like a fluid that remains liquid (without
> boiling) in the 40F - 800F range. I was actually more worried about a
> diesel-type ignition with air as the working gas, but I'm not sure
> that's a real problem.
For a working fluid you might want to consider brake fluid. It is
designed to deal with high temps. Since you seem to be getting away
from the "found anywhere" concept I will again suggest helium. It's
not that hard to get and would improve efficiency. And since you will
be dealing with materials that could withstand a fair amount of
pressure you may want to consider some way of closing the system so
you can run higher pressure.
> Material selection looks like iron pipe for the hot head, copper
> fintubes for the cold head, steel for the reservoir, and iron pipe again
> for the tuning tube / power connection.
Copper might be better for the hot head. It will have no problem with
the temperture and conducts heat better. Adding fins inside and
outside of the tube would give it more surface area for heat
transfer.
Bruce
Jeff
> Vaughn Simon wrote:
>> "Ulysses" <eatm...@spamola.com/> wrote in message
>> news:VvmSj.198$aj...@fe093.usenetserver.com...
>
>>> 724 degrees F is pretty serious.
>>
>> Damn right! Just as a single point of comparison, the steam
>> temperature from the naval nuclear reactors I once operated was
>> hundreds of degrees less.
>
> I was mulling over the comparison with naval reactors when it (finally!)
> dawned on me that the tube at the trough's focus could actually /be/ the
> hot head of a Stirling engine (fluidyne or other type).
Then, you've considered acoustic stirlings?
>
While I was googling for some background info for you, I ran across this:
http://www.youtube.com/watch?v=TWbMTk1rQ0g
Jeff
Morris Dovey
> Might want to use multiple base board tubes in parallel to give the
> same cross section as the rest of the tubing. Not sure how hot
> things will be at the cold head but you may need to take that into
> consideration if you are using soder joints.
We did a design walk-through this afternoon, and agreed to start out
with a single 8' run of fin tube for the cold head. There won't be any
solder joints anywhere - joins will either be threaded, welded, or brazed.
> For a working fluid you might want to consider brake fluid. It is
> designed to deal with high temps. Since you seem to be getting away
> from the "found anywhere" concept I will again suggest helium. It's
> not that hard to get and would improve efficiency. And since you
> will be dealing with materials that could withstand a fair amount of
> pressure you may want to consider some way of closing the system so
> you can run higher pressure.
Hmm - brake fluid would never have entered my mind. We're going to start
out with water, and I made a note (to myself) so as to not forget this
option if we have problems with water. Thanks - it's always good a Good
Thing to have a "Plan 'B'" :-)
>> Material selection looks like iron pipe for the hot head, copper
>> fintubes for the cold head, steel for the reservoir, and iron pipe
>> again for the tuning tube / power connection.
>
> Copper might be better for the hot head. It will have no problem
> with the temperture and conducts heat better. Adding fins inside and
> outside of the tube would give it more surface area for heat
> transfer.
We talked about this and decided to start out with the iron pipe we
already have - but will switch to copper if we need to. I suspect that
the /second/ unit we build will have a copper hot head even if the iron
works satisfactorily.
We also talked about adding a "clip on" semicircular reflective shield
to the sun side of the hot head to reduce radiation losses. That, also,
will probably come later.
The hot head will be 3/4"ID x 8' iron pipe. The regenerator tube will be
3/4"ID x 18" iron pipe. The cold head will be 3/4"ID x 8' copper fin
tube, and both hot and cold heads will connect to an (initially
water-filled) expansion chamber. Some of the calculations went like this:
The volume of the hot head is 42.412 cu in (0.184 gal) and when the air
inside is heated from 70F to 700F, it will expand (assuming constant
presure) to 424.12 cu in (1.836 gal).
The difference between the expanded and unexpanded air volumes (1.836
gal - 0.184 gal = 1.652 gal) is the volume we'd like to see the engine
pump with each cycle. Our PVC engine has run fairly consistently in the
1 to 1.5 Hz range. If this engine runs at a (convenient for calculation)
1 Hz, then it should pump 1.652 gal/sec, 99.12 gal/min, and 5947.2 gal/hour.
We already know that won't happen, because the cold head temperature
won't return to 70F for each cycle. Still, even if the cold head
temperature averages somewhere around 300F, it should still pump at a
reasonable speed. If the temperature seems unreasonably high we'll adopt
your suggestion to add a parallel 8' run of fin tube to the cold head -
but I'm hoping that isn't necessary.
Morris Dovey
>> I was mulling over the comparison with naval reactors when it
>> (finally!) dawned on me that the tube at the trough's focus could
>> actually /be/ the hot head of a Stirling engine (fluidyne or other type).
>
> Then, you've considered acoustic stirlings?
>>
> While I was googling for some background info for you, I ran across this:
>
> http://www.youtube.com/watch?v=TWbMTk1rQ0g
That's neat! I've been nudged several times to look at the acoustic
Stirlings, but I haven't seen anything yet that encouraged me to attempt
scaling one up into the multi-horsepower range - and all of the
non-fluidyne Stirlings require shop capabilities that I don't have and
can't afford.
Morris Dovey
> I'd start out using a high BP liquid instead of gas. Something such
> as the heat transfer oils that are ubiquitous in the petrochem
> industry. Good for several hundred degrees at essentially
> atmospheric pressure, higher with modest (less than 100psi) pressure.
The fluidyne needs both a liquid and a gas. We're planning to start off
with water and air because of cost and availability. With 700F air from
the hot head, we'll probably make a bit of steam.
> Though I haven't done the calcs, I suspect that the upper limit will
> be set by re-radiation off the back side (ironically, the side
> pointing toward the sun) of the absorber, assuming no fancy infrared
> optics or insulation is deployed.
We gave this a bit of thought and decided to try a bare focal tube for
our initial design. If the engine runs at all (and we think it will)
then we'll probably try a "half-pipe" reflector on the back side to see
if that'll make any difference in performance.
I'm guessing that our most significant limitation will be the amount of
heat we can shed at the cold head without spending more than strikes us
as reasonable.
Bruce Richmond
You need to be working with absolute temps. Add 460 to your F temps.
So (700+460)/(70+460)=1160/530=2.19 as opposed to the 10 you got using
F degrees.
Morris Dovey
> You need to be working with absolute temps. Add 460 to your F temps.
>
> So (700+460)/(70+460)=1160/530=2.19 as opposed to the 10 you got using
> F degrees.
Yes! Thank you! Methinks my brain gears are missing a few teeth - you've
made me really glad I posted before we started welding.
Actually, the lesser ratio makes construction noticeably easier (and
less expensive <g>)
Thanks again.
Bruce Richmond
> Bruce Richmond wrote:
> > Might want to use multiple base board tubes in parallel to give the
> > same cross section as the rest of the tubing. Not sure how hot
> > things will be at the cold head but you may need to take that into
> > consideration if you are using soder joints.
>
> We did a design walk-through this afternoon, and agreed to start out
> with a single 8' run of fin tube for the cold head. There won't be any
> solder joints anywhere - joins will either be threaded, welded, or brazed.
Whatever the method try to make the transitions as smooth as
possible. Any turbulance or other resistance to flow will reduce the
net output as they must be overcome when the fluid moves back and
forth.
> > For a working fluid you might want to consider brake fluid. It is
> > designed to deal with high temps. Since you seem to be getting away
> > from the "found anywhere" concept I will again suggest helium. It's
> > not that hard to get and would improve efficiency. And since you
> > will be dealing with materials that could withstand a fair amount of
> > pressure you may want to consider some way of closing the system so
> > you can run higher pressure.
>
> Hmm - brake fluid would never have entered my mind. We're going to start
> out with water, and I made a note (to myself) so as to not forget this
> option if we have problems with water. Thanks - it's always good a Good
> Thing to have a "Plan 'B'" :-)
>
> >> Material selection looks like iron pipe for the hot head, copper
> >> fintubes for the cold head, steel for the reservoir, and iron pipe
> >> again for the tuning tube / power connection.
By "reservoir" I assume you mean the bottom tube and parts of the
vertical tubes that are constantly filled with liquid. How about the
regenerator or top tube?
> > Copper might be better for the hot head. It will have no problem
> > with the temperture and conducts heat better. Adding fins inside and
> > outside of the tube would give it more surface area for heat
> > transfer.
>
> We talked about this and decided to start out with the iron pipe we
> already have - but will switch to copper if we need to. I suspect that
> the /second/ unit we build will have a copper hot head even if the iron
> works satisfactorily.
>
> We also talked about adding a "clip on" semicircular reflective shield
> to the sun side of the hot head to reduce radiation losses. That, also,
> will probably come later.
>
> The hot head will be 3/4"ID x 8' iron pipe. The regenerator tube will be
> 3/4"ID x 18" iron pipe. The cold head will be 3/4"ID x 8' copper fin
> tube, and both hot and cold heads will connect to an (initially
> water-filled) expansion chamber. Some of the calculations went like this:
The flow resistance of 3/4" pipe will be much greater than the 3" PVC
you were using. I know larger pipe gets expensive quick but you might
try a second system with 1" pipe for comparison. My previous
suggestion to use multiple baseboard pipes on the cold head was to
match the cross sectional area of the larger pipe I had assumed you
would be using for the hot head.
> The volume of the hot head is 42.412 cu in (0.184 gal) and when the air
> inside is heated from 70F to 700F, it will expand (assuming constant
> presure) to 424.12 cu in (1.836 gal).
>
> The difference between the expanded and unexpanded air volumes (1.836
> gal - 0.184 gal = 1.652 gal) is the volume we'd like to see the engine
> pump with each cycle. Our PVC engine has run fairly consistently in the
> 1 to 1.5 Hz range. If this engine runs at a (convenient for calculation)
> 1 Hz, then it should pump 1.652 gal/sec, 99.12 gal/min, and 5947.2 gal/hour.
>
> We already know that won't happen, because the cold head temperature
> won't return to 70F for each cycle. Still, even if the cold head
> temperature averages somewhere around 300F, it should still pump at a
> reasonable speed. If the temperature seems unreasonably high we'll adopt
> your suggestion to add a parallel 8' run of fin tube to the cold head -
> but I'm hoping that isn't necessary.
You seem to be under the impression that you can use the full stroke
of the moving fluid to pump with. I am quite certain you will find
there is much less power to be had. Going back to the analogy of a
pendulum, there is energy to be had as it swings down to vertical.
But if you take that energy there is none left to carry it up the
remainder of the swing. The clock works adds just a bit of energy to
each stroke to make up for the energy lost to friction. Use that
energy to power something else and it will stop swinging.
The energy added on each stroke is the movement seen in the cold
head. Without it the motion in the hot head will quickly die. You
can take some of that energy and have a reduced amplitude in the hot
head, but some of the added energy must go to replacing the energy
lost in the movement of the fluid.
As a quick calculation, your reflector has an area of 2 meters ^ 2.
That means it gathers about 2 KW of power. At most the efficiency of
the system can be (1160-530)/1160=54%. I would be very surprised if
you could obtain half of the theoretical efficiency. That would put
the max output at about 500 watts. That's really not too bad when you
look at what a 300 watt low temp Stirling engine looks like.
http://www.nmri.go.jp/eng/khirata/stirling/kiriki/yama1/300pfrm.html
or a 1 KW
http://www.nmri.go.jp/eng/khirata/stirling/kiriki/yama2/index.html
Bruce
Morris Dovey
> By "reservoir" I assume you mean the bottom tube and parts of the
> vertical tubes that are constantly filled with liquid. How about the
> regenerator or top tube?
The regenerator will be 3/4" iron pipe filled with 1/8" copper tubes.
It'll be wrapped with fiberglass and the whole enclosed in a length of
scrap aluminum tubing.
> The flow resistance of 3/4" pipe will be much greater than the 3" PVC
> you were using. I know larger pipe gets expensive quick but you might
> try a second system with 1" pipe for comparison. My previous
> suggestion to use multiple baseboard pipes on the cold head was to
> match the cross sectional area of the larger pipe I had assumed you
> would be using for the hot head.
Understood (the PVC engine is 4" PVC pipe, BTW). It's an affordability
issue - we'd use larger diameter and copper if we could (we're able to
use fintube only because I intercepted two pieces on their way to the
local landfill).
I wish I know enough to accurately predetermine the behavior with tubing
material and ID as independent variables.
> You seem to be under the impression that you can use the full stroke
> of the moving fluid to pump with. I am quite certain you will find
> there is much less power to be had. Going back to the analogy of a
> pendulum, there is energy to be had as it swings down to vertical.
> But if you take that energy there is none left to carry it up the
> remainder of the swing. The clock works adds just a bit of energy to
> each stroke to make up for the energy lost to friction. Use that
> energy to power something else and it will stop swinging.
Hell, I'm not even under the impression that I can make it work at all
on the first (second, third,..) attempt. :-)
We did find that the PVC version seemed to do better when it had a load
to work against. We're still not sure if that's a basic characteristic
of the device or if adding load doesn't in some way correct a deficient
design...
> The energy added on each stroke is the movement seen in the cold
> head. Without it the motion in the hot head will quickly die. You
> can take some of that energy and have a reduced amplitude in the hot
> head, but some of the added energy must go to replacing the energy
> lost in the movement of the fluid.
I think I'll need some quiet time to digest this - I had been thinking
that the energy added at each stroke was the heat absorbed at the hot head.
> As a quick calculation, your reflector has an area of 2 meters ^ 2.
> That means it gathers about 2 KW of power. At most the efficiency of
> the system can be (1160-530)/1160=54%. I would be very surprised if
> you could obtain half of the theoretical efficiency. That would put
> the max output at about 500 watts. That's really not too bad when you
> look at what a 300 watt low temp Stirling engine looks like.
I think we'd consider 27% efficiency to be incredible success, worthy of
a major Iowa celebration (as in: "Barkeep, brew and corn dogs for
everyone!")
A 500W output with less than US$250 worth of materials (retail) from an
engine whose only moving parts were air and water seems to me an
acceptable outcome. Even a 300W output wouldn't be too shabby.
If we see a 300W output, of course, that'd just mean that we'd have to
see how close we could come to the theoretical - that's just human nature...
> http://www.nmri.go.jp/eng/khirata/stirling/kiriki/yama1/300pfrm.html
>
> or a 1 KW
>
> http://www.nmri.go.jp/eng/khirata/stirling/kiriki/yama2/index.html
These guys obviously have much deeper pockets than I - but I think we're
not all trying to solve the same problems...
The questions of the day (this day only - tomorrow's will be different)
are whether it can be used to pump brine out of a rice field, and
whether it can pump water to/through a water purification system - and
how many can be stowed in a C-130 cargo bay and whether knocked-down
kits can survive a parachute drop and be deployed in under an hour by an
average person with no training.
Bruce Richmond
wrote:
> "Morris Dovey" <mrdo...@iedu.com> wrote in message
>
> news:481a757a$0$87075$815e...@news.qwest.net...
>
>
>
>
>
> > Vaughn Simon wrote:
Just noticed this. On the examples built so far the hot and cold
pipes were the same diameter. Making the cold pipe smaller will
increase the movement of the fluid in that pipe for a given
displacement volume. That would translate into a higher velocity with
more momentum. This would definitely be a means of tuning the
system. With previous designs the hot and cold pipes were not all
that different in terms of absolute temperture. With this new set-up
the hot tube will be about twice the temperture of the cold tube.
Might need to take that into account when sizing the tubing.
> And the 'displacement' of the liquid in the fluidyne would
> have to move an 'appreciable amount' of the gas from hot to cold, so with a
> large hot and cold, you need a large 'displacement' (either really long
> stroke, very wide bore, or combination).
>
> You don't need any sort of phase change for the working fluid (unlike
> Rankine cycle), so I should think any gas that behaves close to the ideal
> gas law at these temperatures / pressures would suffice. The only reason to
> not use air would be if the high-temperature oxygen causes some corrosion
> problem. Material selection for the piping and such would seem to be key.
>
> daestrom
The lower molecular weight of helium allows it to change temperture
quicker which increases the efficiency of the engine. Hydrogen would
work even better but is more difficult and dangerous to work with.
>
>
> > Hmm. Time to go fire up the software model with new temperature
> > parameters...
>
> > --
> > Morris Dovey
> > DeSoto Solar
> > DeSoto, Iowa USA
> >http://www.iedu.com/DeSoto/Stirling/Heat.html- Hide quoted text -
>
> - Show quoted text -- Hide quoted text -
>
> - Show quoted text -
Bruce Richmond
Heat energy flows through a heat engine and in the process some of it
is converted to mechanical energy. The large movement seen in the hot
tube and the open tube is mostly due to harmonic pressure fluctuations
in the air trapped in the loop. It is roughly equivelent to the
displacer piston in a mechanical Sterling engine. Its motion does not
produce the power, it just pushes the air to where it needs to be for
the heating and cooling cycles. The power, both for moving the
displacer and for net output, is produced by the power piston, which
is much smaller than the displacer. The cold tube is roughly
equivelent to the power piston. If you put a valve in it so that you
can block it you would see that the large motions in the other tubes
will quickly stop when you block flow in the cold tube.
> > As a quick calculation, your reflector has an area of 2 meters ^ 2.
> > That means it gathers about 2 KW of power. At most the efficiency of
> > the system can be (1160-530)/1160=54%. I would be very surprised if
> > you could obtain half of the theoretical efficiency. That would put
> > the max output at about 500 watts. That's really not too bad when you
> > look at what a 300 watt low temp Stirling engine looks like.
>
> I think we'd consider 27% efficiency to be incredible success, worthy of
> a major Iowa celebration (as in: "Barkeep, brew and corn dogs for
> everyone!")
Notice I said I would be surprised if you got that. It may be
possible though with enough tweaking.
> A 500W output with less than US$250 worth of materials (retail) from an
> engine whose only moving parts were air and water seems to me an
> acceptable outcome. Even a 300W output wouldn't be too shabby.
>
> If we see a 300W output, of course, that'd just mean that we'd have to
> see how close we could come to the theoretical - that's just human nature...
of course :)
> >http://www.nmri.go.jp/eng/khirata/stirling/kiriki/yama1/300pfrm.html
>
> > or a 1 KW
>
> >http://www.nmri.go.jp/eng/khirata/stirling/kiriki/yama2/index.html
>
> These guys obviously have much deeper pockets than I - but I think we're
> not all trying to solve the same problems...
>
> The questions of the day (this day only - tomorrow's will be different)
> are whether it can be used to pump brine out of a rice field, and
> whether it can pump water to/through a water purification system - and
> how many can be stowed in a C-130 cargo bay and whether knocked-down
> kits can survive a parachute drop and be deployed in under an hour by an
> average person with no training.
How much elevation change to pump the brine out?
Bruce
Morris Dovey
> How much elevation change to pump the brine out?
AFAICT from looking at the news videos, only about 2-3 feet.
Morris Dovey
>> But you mentioned putting the working fluid right through the
>> collector. That might be tricking with a long-trough. You can't go
>> too narrow with the tube due to increased friction, so a long
>> trough collector pipe would increase the volume of the 'hot side'
>> quite a bit. Just means you need a 'cold side' that is similar in
>> size (well, actually smaller by the ratio of temperatures).
>
> Just noticed this. On the examples built so far the hot and cold
> pipes were the same diameter. Making the cold pipe smaller will
> increase the movement of the fluid in that pipe for a given
> displacement volume. That would translate into a higher velocity
> with more momentum. This would definitely be a means of tuning the
> system. With previous designs the hot and cold pipes were not all
> that different in terms of absolute temperture. With this new set-up
> the hot tube will be about twice the temperture of the cold tube.
> Might need to take that into account when sizing the tubing.
So far, the experimental models have all used a single size tube
throughout - save for the regenerator tube. I think it'll be a good idea
to tinker with the diameters of the various elements. The single-size
approach doesn't seem to be a bad starting point.
>> And the 'displacement' of the liquid in the fluidyne would have to
>> move an 'appreciable amount' of the gas from hot to cold, so with a
>> large hot and cold, you need a large 'displacement' (either really
>> long stroke, very wide bore, or combination).
Yesterday I wrote a program to calculate "displacer" dimensions based on
hot head dimensions and best guesses at hot and cold side temperatures.
This time I remembered to add an f2k(t) function to calculate the
expansion ratio in K instead of degrees Fahrenheit. :-)
> The lower molecular weight of helium allows it to change temperture
> quicker which increases the efficiency of the engine. Hydrogen would
> work even better but is more difficult and dangerous to work with.
You're right - but plain old air is going to have to do the job for this
engine. The H/He options will await an application in a more controlled
environment with trained personnel.
Bruce Richmond
One simple inexpensive thing you could try to improve the efficiency
would be to apply stove black to the hot pipe. I realize it will be
"black pipe", but such pipe is often not painted at all or has glossy
paint. Flat black paint may work as well, but paint can also act as
an insulator. In theory I believe the stove black would also help the
cold pipe radiate heat, but I don't recall ever seing it used on
baseboard heating elements.
Bruce
Bruce Richmond
> Bruce Richmond wrote:
> > On May 20, 11:55 am, Morris Dovey <mrdo...@iedu.com> wrote:
> >> The questions of the day (this day only - tomorrow's will be different)
> >> are whether it can be used to pump brine out of a rice field, and
> > How much elevation change to pump the brine out?
>
> AFAICT from looking at the news videos, only about 2-3 feet.
>
To get a better feel of what we are looking at I did some quick
calculations.
A gallon of water weighs about 8 lbs.
Lifting it 2 feet would take 16 ft. lbs. of work.
300 Watts = .4 hp = 220 ft. lbs. per second.
220/16=13.75 gal./sec
A 3/4" ID pipe has a cross sectional area of .44 sq in.
There are 231 cubic inches per gallon.
13.75 X 231 = 3176 cubic inches.
That would fill 3176/.44=7218.75 inches or 601.6 feet of pipe.
So the water would have to be flowing through the pipe at 600 ft/sec
or a bit over 400 mph to pump 13.75 gal/sec.
If the water is going up and down 8 ft during the cycle it would have
to make 75 cycles per second.
If the cycle is working at 1.5 hz then the water would have to be
going up and down 400 feet during the cycle.
A 4" pipe has a cross sectional area of 12.57 sq in or 28.6 times that
of the 3/4" pipe. That would reduce the velocity to about 21 ft/sec
for 13.75 gal/sec.
If we have the water in a 4" pipe going through a 7 ft cycle at 1.5 hz
it would lift 6.875 gal/sec the 2 feet and use 150 Watts, neglecting
losses.
But then like I said before, we can't expect to use the full stroke of
the fluid to pump without killing the cycle.
Let's say we can only use about 5% of the full travel for pumping.
That would mean about 1/3 gal per sec or 20 gal/min or 1200 gal/hr.
Might not sound as impressive as the 6000 gal/hr you estimated
earlier, but it would be a lot less work than lifting a 5 gal bucket 4
times a minute.
That is using 4" pipe. Seems the 3/4" would be too restrictive.
Making the whole thing out of 4" metal pipe would be expensive. How
about cutting the reflector into four 2 ft lengths for 4 pumps. Use a
2 ft section of thin wall 4" copper pipe with some sort of insulator
between it and pvc pipe for most of the pump. A single 8' section of
baseboard pipe could be cut into four 2 ft sections to be used in
parallel for the cold head.
Another possibility would be to mount 4" pvc a bit off from the focal
point so it would give say a 2" wide band of 200 F deg heat rather
than a 3/8" wide band of 700 F deg heat.
Just some things to think about.
Bruce
Morris Dovey
<snipped a lot of good thought>
Let's shift our thinking just a little bit. I'm hoping that that "little
bit" makes for much improved operation.
The plan is to move only air (no fluid, or as little fluid as possible)
in the 3/4" pipe because there'd be just too much resistance to flow.
An absolute maximum of the fluid flow will take place in 4" pipe, for
/exactly/ the reasons you put forth.
I know that the 3/4" ID is restrictive, even for air, but the
temperature within the hot head will have some kind of inverse
relationship to the hot head ID (IOW, the larger the volume, the lower
the temperature, and vice versa). I also understand that the greater the
temperature differential between the hot head and the cold head, the
greater the possible engine efficiency. The only practical way I have to
maximize that differential is by maximizing the hot head temperature.
So what I want to do is to find that balance point where I have the
highest hot head temperature /and/ the least airflow losses. I'm certain
that both can't be maximized simultaneously, so I'm going to be working
to find the best trade-off I can make using off-the-shelf supplies from
my local hardware/building store.
AFAICT so far, the engine's frequency is directly related to the height
to which the working fluid is displaced. That's counter-intuitive to me,
because I expected some kind of mass relationship, but I can cope. <g>
If that's true, then (up to a point) I can make the engine run faster by
shortening the vertical displacement of the working fluid. The speed
limit for the engine now under construction (I think) will be imposed by
the restricted airflow in the hothead/regenerator/cold head path - and,
as per usual, I'll need to play with the height of the fluid in the
tuning tube to attempt to approach some reasonable degree of resonance
between the Carnot cycle and the working fluid's natural (simple
harmonic?) frequency.
Your help is always much appreciated (I've been in over my head from the
very beginning).
Morris Dovey
> That is using 4" pipe. Seems the 3/4" would be too restrictive.
>
> Making the whole thing out of 4" metal pipe would be expensive. How
> about cutting the reflector into four 2 ft lengths for 4 pumps. Use a
> 2 ft section of thin wall 4" copper pipe with some sort of insulator
> between it and pvc pipe for most of the pump. A single 8' section of
> baseboard pipe could be cut into four 2 ft sections to be used in
> parallel for the cold head.
>
> Another possibility would be to mount 4" pvc a bit off from the focal
> point so it would give say a 2" wide band of 200 F deg heat rather
> than a 3/8" wide band of 700 F deg heat.
>
> Just some things to think about.
Hmm - thinking...
How about if I use 1.5" copper for the hot head and rotate my mirror
material so that the collector is 8' wide and 4' long? That should boost
the temperature right up there and simultaneously shorten hot and cold
heads by 50%.
It shouldn't produce temperatures much hotter than 1400F...
Bruce Richmond
> Bruce Richmond wrote:
>
> <snipped a lot of good thought>
>
> Let's shift our thinking just a little bit. I'm hoping that that "little
> bit" makes for much improved operation.
>
> The plan is to move only air (no fluid, or as little fluid as possible)
> in the 3/4" pipe because there'd be just too much resistance to flow.
>
> An absolute maximum of the fluid flow will take place in 4" pipe, for
> /exactly/ the reasons you put forth.
>
> I know that the 3/4" ID is restrictive, even for air, but the
> temperature within the hot head will have some kind of inverse
> relationship to the hot head ID (IOW, the larger the volume, the lower
> the temperature, and vice versa).
While the relationship does exist you still have to work within
certain limits. A gasoline engine gives more power for a given size
if you increase the compression ratio, but you can't just raise it as
high as you would like to. On pump gasoline it is difficult to avoid
detonation at ratios higher than 12:1. You also get more power by
increasing the frequency of the cycle, or in the gasoline engine the
rpm. But again if you try to go too far with this the engine will
break. I suspect that to use all the power of your 8' reflector you
will need to use large diameter pipe. To use smaller pipe would mean
increasing either the stroke length or frequency of the cycle, either
one of which can present problems. The idea of using small pipe for
the air and large pipe for the fluid seems reasonable.
> I also understand that the greater the
> temperature differential between the hot head and the cold head, the
> greater the possible engine efficiency. The only practical way I have to
> maximize that differential is by maximizing the hot head temperature.
I'm starting to think getting this to work using 724F on the hot side
is going to cost more than it's worth. Raising the hot side temp
increases efficiency but the smaller pipe decreases it. I also have
doubts about the actual temp of the air in the hot pipe. If the 724F
only applies to a 3/8" stripe down one side of the pipe, that heat
will spread around the pipe and average out the temp. I suspect the
average temp in the hot head will end up little if any higher than it
would if the light was focused on a stripe the full width of the pipe.
By adding the half pipe reflector you had mentioned before you could
even make the stripe wider than the pipe and let the half pipe
redirect it at the other side of the pipe. That might get the
localized temp back down to where PVC could take it. Might still want
to use copper for the hot head due to its better heat transfer
properties, but with the lower temp it could be directly connected to
PVC.
BTW, have you ever taken temp readings of the air and water at the hot
and cold heads?
> So what I want to do is to find that balance point where I have the
> highest hot head temperature /and/ the least airflow losses. I'm certain
> that both can't be maximized simultaneously, so I'm going to be working
> to find the best trade-off I can make using off-the-shelf supplies from
> my local hardware/building store.
>
> AFAICT so far, the engine's frequency is directly related to the height
> to which the working fluid is displaced. That's counter-intuitive to me,
> because I expected some kind of mass relationship, but I can cope. <g>
Think of the pendulum where the longer it is the bigger the swing and
the lower the frequency. How much weight is at the end doesn't make
much difference, just how far it is from the pivot point.
> If that's true, then (up to a point) I can make the engine run faster by
> shortening the vertical displacement of the working fluid. The speed
> limit for the engine now under construction (I think) will be imposed by
> the restricted airflow in the hothead/regenerator/cold head path - and,
> as per usual, I'll need to play with the height of the fluid in the
> tuning tube to attempt to approach some reasonable degree of resonance
> between the Carnot cycle and the working fluid's natural (simple
> harmonic?) frequency.
Putting a load on the engine should shorten the displacement. The
energy applied to the load would otherwise have gone into displacing
the fluid further. Using smaller pipe for the air chamber should
decrease the displacement and increase the frequency. Think of the
air chamber as a spring. As the displacement reduces its volume the
pressure goes up. Reduce the air chamber volume and the pressure will
rise quicker, reversing the motion of the fluid. Think of it as a
shorter, stiffer spring. You may find that going to 3/4" pipe for
most of the air chamber makes for too stiff a spring. If that is the
case I'm quite sure you will be able to use PVC for the air pipe to
the cold side of the regenerator. If I'm correct there should be a
considerable temp drop across the regenerator.
I got thinking about the heat flow through the water. It contributes
nothing to the power since the power is derived from the expansion and
contraction of the gas. So anything that would slow the transfer of
heat through the water pipe would improve efficiency. That is
basically what the regenerator does in the air pipe. Adding a
regenerator to the water pipe would have the same effect as having a
longer pipe without the increase in actual mass moving back and
forth. Wrapping the hot end of the water pipe in fiberglass would
also help.
While the water pipe regenerator will improve efficiency it may not be
worth the added expense. At this point it looks like the reflector
can put out more heat than the engine can handle. The money might be
better spent going to the next size bigger on the PVC pipe.
> Your help is always much appreciated (I've been in over my head from the
> very beginning).
I'm afraid it's the blind leading the blind here. I know some
thermodynamics and physics in general, but I never heard of a fluidyne
engine before reading your web page. Considering how much info can be
found on them it seems you are at the leading edge of the
technology :)
Bruce
Bruce Richmond
> Bruce Richmond wrote:
> > That is using 4" pipe. Seems the 3/4" would be too restrictive.
>
> > Making the whole thing out of 4" metal pipe would be expensive. How
> > about cutting the reflector into four 2 ft lengths for 4 pumps. Use a
> > 2 ft section of thin wall 4" copper pipe with some sort of insulator
> > between it and pvc pipe for most of the pump. A single 8' section of
> > baseboard pipe could be cut into four 2 ft sections to be used in
> > parallel for the cold head.
>
> > Another possibility would be to mount 4" pvc a bit off from the focal
> > point so it would give say a 2" wide band of 200 F deg heat rather
> > than a 3/8" wide band of 700 F deg heat.
>
> > Just some things to think about.
>
> Hmm - thinking...
>
> How about if I use 1.5" copper for the hot head and rotate my mirror
> material so that the collector is 8' wide and 4' long? That should boost
> the temperature right up there and simultaneously shorten hot and cold
> heads by 50%.
Should be better than 3/4" pipe I think. Shorter head should also go
better with the shorter displacement/higher frequency idea.
> It shouldn't produce temperatures much hotter than 1400F...
May want to try just half of your standard reflectors first.
Morris Dovey
> On May 23, 3:47 am, Morris Dovey <mrdo...@iedu.com> wrote:
>> I know that the 3/4" ID is restrictive, even for air, but the
>> temperature within the hot head will have some kind of inverse
>> relationship to the hot head ID (IOW, the larger the volume, the
>> lower the temperature, and vice versa).
>
> While the relationship does exist you still have to work within
> certain limits. A gasoline engine gives more power for a given size
> if you increase the compression ratio, but you can't just raise it as
> high as you would like to. On pump gasoline it is difficult to avoid
> detonation at ratios higher than 12:1. You also get more power by
> increasing the frequency of the cycle, or in the gasoline engine the
> rpm. But again if you try to go too far with this the engine will
> break. I suspect that to use all the power of your 8' reflector you
> will need to use large diameter pipe. To use smaller pipe would mean
> increasing either the stroke length or frequency of the cycle, either
> one of which can present problems. The idea of using small pipe for
> the air and large pipe for the fluid seems reasonable.
Yes to the limits - I just don't know where they are. I expect I need to
work my way through a whole range of tubing size combinations to find
where the trade-offs are. I think I'd rather go through all of that than
try to guess at the optimal point and lock onto the first combination
that worked.
>> I also understand that the greater the temperature differential
>> between the hot head and the cold head, the greater the possible
>> engine efficiency. The only practical way I have to maximize that
>> differential is by maximizing the hot head temperature.
>
> I'm starting to think getting this to work using 724F on the hot side
> is going to cost more than it's worth. Raising the hot side temp
> increases efficiency but the smaller pipe decreases it. I also have
> doubts about the actual temp of the air in the hot pipe. If the 724F
> only applies to a 3/8" stripe down one side of the pipe, that heat
> will spread around the pipe and average out the temp. I suspect the
> average temp in the hot head will end up little if any higher than it
> would if the light was focused on a stripe the full width of the
> pipe.
>
> By adding the half pipe reflector you had mentioned before you could
> even make the stripe wider than the pipe and let the half pipe
> redirect it at the other side of the pipe. That might get the
> localized temp back down to where PVC could take it. Might still
> want to use copper for the hot head due to its better heat transfer
> properties, but with the lower temp it could be directly connected to
> PVC.
It doesn't make much sense to even consider PVC with a concentrator. The
PVC loses integrity above 165F - and at the temperatures produced by the
concentrator, we can't even use solder to hold things together.
> BTW, have you ever taken temp readings of the air and water at the
> hot and cold heads?
No - the best we can do is take temperature readings of the tubes from
the outside.
>> AFAICT so far, the engine's frequency is directly related to the
>> height to which the working fluid is displaced. That's
>> counter-intuitive to me, because I expected some kind of mass
>> relationship, but I can cope. <g>
>
> Think of the pendulum where the longer it is the bigger the swing and
> the lower the frequency. How much weight is at the end doesn't make
> much difference, just how far it is from the pivot point.
Ok. Then if the mass of the water isn't a consideration, and if the
water functions only as a displacer, I should be working to reduce the
amount of water toward the minimum required to work as a displacer.
>> If that's true, then (up to a point) I can make the engine run
>> faster by shortening the vertical displacement of the working
>> fluid. The speed limit for the engine now under construction (I
>> think) will be imposed by the restricted airflow in the
>> hothead/regenerator/cold head path - and, as per usual, I'll need
>> to play with the height of the fluid in the tuning tube to attempt
>> to approach some reasonable degree of resonance between the Carnot
>> cycle and the working fluid's natural (simple harmonic?)
>> frequency.
>
> Putting a load on the engine should shorten the displacement. The
> energy applied to the load would otherwise have gone into displacing
> the fluid further. Using smaller pipe for the air chamber should
> decrease the displacement and increase the frequency.
Right. The displacement is reduced because the volume being expanded is
less. With the reduced volume, I'm hoping for more rapid heating and
/faster/ expansion - which should, indeed, increase the frequency.
> Think of the air chamber as a spring. As the displacement reduces
> its volume the pressure goes up. Reduce the air chamber volume and
> the pressure will rise quicker, reversing the motion of the fluid.
Now at least /I/ am on shakier ground. My reading is that this expansion
in a Stirling's hot head is isobaric - meaning that it takes place at
constant pressure. The classic Carnot pressure/volume plot shows a
relatively slow pressure drop during the expansion and a large, sudden
drop at the end of the expansion. Similarly I see the contraction phase
described as isobaric - during the time that the volume is contracting,
there's a relatively slow pressure increase with a large, sudden jump in
pressure at the end of the contraction.
I'm having difficulty making that match up with the picture you're
presenting - what am I missing?
> Think of it as a shorter, stiffer spring. You may find that going to
> 3/4" pipe for most of the air chamber makes for too stiff a spring.
> If that is the case I'm quite sure you will be able to use PVC for
> the air pipe to the cold side of the regenerator. If I'm correct
> there should be a considerable temp drop across the regenerator.
A large temperature drop across the regenerator is desirable, no? It
would mean that heat is being well-conserved between cycles.
> I got thinking about the heat flow through the water. It contributes
> nothing to the power since the power is derived from the expansion
> and contraction of the gas. So anything that would slow the transfer
> of heat through the water pipe would improve efficiency.
Yuppers. I've even given some thought to putting a small free-floating
piston at the air-water interface to reduce that transfer. It's on my
list of "things to try".
> That is basically what the regenerator does in the air pipe. Adding
> a regenerator to the water pipe would have the same effect as having
> a longer pipe without the increase in actual mass moving back and
> forth. Wrapping the hot end of the water pipe in fiberglass would
> also help.
I don't think that's the case. At the regenerator I want to capture heat
during the expansion phase to be released during the contraction phase.
At the hot head-displacer interface I want to inhibit heat transfer
across that interface at all times.
My regenerator is built as a bundle of small tubes within the outer tube
to have a maximum of surface area for grabbing heat from the air during
expansion and giving it back during contraction.
I'm a bit antsy about the regenerator inhibiting airflow, but recognize
the benefits. This is another Stirling trade-off point to be resolved.
However, I'm /really/ sure that I don't want to interpose a flow
restriction of any kind in the fluid flow.
> While the water pipe regenerator will improve efficiency it may not
> be worth the added expense. At this point it looks like the
> reflector can put out more heat than the engine can handle. The
> money might be better spent going to the next size bigger on the PVC
> pipe.
As I mentioned before PVC isn't an issue in this all-metal engine. More
heat than the engine can handle means (I think) overheating to the point
where it blows all its fluid out the tuning tube... :-(
>> Your help is always much appreciated (I've been in over my head
>> from the very beginning).
>
> I'm afraid it's the blind leading the blind here. I know some
> thermodynamics and physics in general, but I never heard of a
> fluidyne engine before reading your web page. Considering how much
> info can be found on them it seems you are at the leading edge of the
> technology :)
Fluidynes have been around for a while, and aren't new with me. You can
find a fair number of videos of 'em running on YouTube. What's
surprising is that fluidynes, and Stirling cycle engines in general,
have been so little commercialized other than as toys.
Bruce Richmond
> Bruce Richmond wrote:
> > On May 23, 3:47 am, Morris Dovey <mrdo...@iedu.com> wrote:
> >> I know that the 3/4" ID is restrictive, even for air, but the
> >> temperature within the hot head will have some kind of inverse
> >> relationship to the hot head ID (IOW, the larger the volume, the
> >> lower the temperature, and vice versa).
>
> > While the relationship does exist you still have to work within
> > certain limits. A gasoline engine gives more power for a given size
> > if you increase the compression ratio, but you can't just raise it as
> > high as you would like to. On pump gasoline it is difficult to avoid
> > detonation at ratios higher than 12:1. You also get more power by
> > increasing the frequency of the cycle, or in the gasoline engine the
> > rpm. But again if you try to go too far with this the engine will
> > break. I suspect that to use all the power of your 8' reflector you
> > will need to use large diameter pipe. To use smaller pipe would mean
> > increasing either the stroke length or frequency of the cycle, either
> > one of which can present problems. The idea of using small pipe for
> > the air and large pipe for the fluid seems reasonable.
>
> Yes to the limits - I just don't know where they are. I expect I need to
> work my way through a whole range of tubing size combinations to find
> where the trade-offs are. I think I'd rather go through all of that than
> try to guess at the optimal point and lock onto the first combination
> that worked.
>
I certainly didn't mean to dictate a particular size, just trying to
go in the right general direction from the start.
>
>
>
> >> I also understand that the greater the temperature differential
> >> between the hot head and the cold head, the greater the possible
> >> engine efficiency. The only practical way I have to maximize that
> >> differential is by maximizing the hot head temperature.
>
> > I'm starting to think getting this to work using 724F on the hot side
> > is going to cost more than it's worth. Raising the hot side temp
> > increases efficiency but the smaller pipe decreases it. I also have
> > doubts about the actual temp of the air in the hot pipe. If the 724F
> > only applies to a 3/8" stripe down one side of the pipe, that heat
> > will spread around the pipe and average out the temp. I suspect the
> > average temp in the hot head will end up little if any higher than it
> > would if the light was focused on a stripe the full width of the
> > pipe.
>
> > By adding the half pipe reflector you had mentioned before you could
> > even make the stripe wider than the pipe and let the half pipe
> > redirect it at the other side of the pipe. That might get the
> > localized temp back down to where PVC could take it. Might still
> > want to use copper for the hot head due to its better heat transfer
> > properties, but with the lower temp it could be directly connected to
> > PVC.
>
> It doesn't make much sense to even consider PVC with a concentrator. The
> PVC loses integrity above 165F - and at the temperatures produced by the
> concentrator, we can't even use solder to hold things together.
The 724 degrees is when the 4' wide panel concentrates the light into
a 3/8" wide beam. If it is only concentrated into a 4" wide beam it
will be at a much lower temp. Concentrate it into a 6" wide beam and
it will be an even lower temp. At some point the temp will be low
enough to work with PVC. It may not be as efficient as could be done
with metal, but on a per dollar basis it could be a winner.
> > BTW, have you ever taken temp readings of the air and water at the
> > hot and cold heads?
>
> No - the best we can do is take temperature readings of the tubes from
> the outside.
>
Shouldn't be that hard to mount thermometers at either head. Might
not be able to get real time variations ,but it should be good enough.
> >> AFAICT so far, the engine's frequency is directly related to the
> >> height to which the working fluid is displaced. That's
> >> counter-intuitive to me, because I expected some kind of mass
> >> relationship, but I can cope. <g>
>
> > Think of the pendulum where the longer it is the bigger the swing and
> > the lower the frequency. How much weight is at the end doesn't make
> > much difference, just how far it is from the pivot point.
>
> Ok. Then if the mass of the water isn't a consideration, and if the
> water functions only as a displacer, I should be working to reduce the
> amount of water toward the minimum required to work as a displacer.
Water is cheap. I wouldn't worry too much about reducing how much is
used.
The way I am seeing it, starting with the water at the high point in
the open tube, the water flows back from the open tube to the hot tube
raising the level there. At the "T" where the hot tube goes up there
is less resistance to the water going straight across than making the
turn to go up the cold tube. With the water rising in the hot tube
and little change in the cold tube the trapped air is being
compressed. With air being forced from the hot side to the cold side
it cools reducing its volume and the rise in pressure. The
regenerator helps here, quickly cooling a large volume of air while
heating the metal in the regenerator. With enough cooling it will
aproach a constant pressure. Eventually the air in the cold tube is
near the temp of the tube itself ruducing heat flow and contraction.
But the momentum of the water flowing in the lower pipe continues to
compress the trapped air. As the pressure rises it slows the flow in
the hot tube and increases the flow up the cold tube. With the water
at its highest point in the hot tube it has given up its momentum to
compress the air, and the compressed air, along with gravity, will now
start the water flowing back down the hot tube. The momentum of the
water floing up the cold tube will give an extra little shove to get
the water flowing back down the cold tube. When the momentum of the
water floing up the cold tube runs out it too will start flowing
down. The fact that the air is flowing into the hot tube and being
heated means the pressure drop from the fluid level dropping in the
hot tube will be less than it would be without the heat being added.
Add enough heat and the pressure will aproach constant despite the
falling water in the hot tube. As the water rises in the open tube it
will add to the resistance of the water flowing into that tube,
eventually stopping the flow and starting the cycle over again.
> > Think of it as a shorter, stiffer spring. You may find that going to
> > 3/4" pipe for most of the air chamber makes for too stiff a spring.
> > If that is the case I'm quite sure you will be able to use PVC for
> > the air pipe to the cold side of the regenerator. If I'm correct
> > there should be a considerable temp drop across the regenerator.
>
> A large temperature drop across the regenerator is desirable, no? It
> would mean that heat is being well-conserved between cycles.
Yes, the drop is desireable. There is however an optimal drop. It is
not just a "the more the better" thing.
> > I got thinking about the heat flow through the water. It contributes
> > nothing to the power since the power is derived from the expansion
> > and contraction of the gas. So anything that would slow the transfer
> > of heat through the water pipe would improve efficiency.
>
> Yuppers. I've even given some thought to putting a small free-floating
> piston at the air-water interface to reduce that transfer. It's on my
> list of "things to try".
I was thinking about adding a piston to reduce splashing as the
frequency increased. Using it to also block heat transfer would be a
good idea.
> > That is basically what the regenerator does in the air pipe. Adding
> > a regenerator to the water pipe would have the same effect as having
> > a longer pipe without the increase in actual mass moving back and
> > forth. Wrapping the hot end of the water pipe in fiberglass would
> > also help.
>
> I don't think that's the case. At the regenerator I want to capture heat
> during the expansion phase to be released during the contraction phase.
> At the hot head-displacer interface I want to inhibit heat transfer
> across that interface at all times.
You want the heat to flow through the air pipe where some of it can be
converted to mechanical energy. Heat that flows through the water is
just wasted. The fiberglass around the hot end of the water pipe will
prevent heat loss to the air around the pipe. The hotter the water in
that end of the pipe the more it resists additional heat flowing into
it. At the cold end we want the water as cool as possible to help
cool the air. Leaving that end of the pipe unwrapped lets the pipe
shed some heat before it gets to the air/water interface.
> My regenerator is built as a bundle of small tubes within the outer tube
> to have a maximum of surface area for grabbing heat from the air during
> expansion and giving it back during contraction.
>
> I'm a bit antsy about the regenerator inhibiting airflow, but recognize
> the benefits. This is another Stirling trade-off point to be resolved.
Just make the regenerator(s) a bit larger in diameter to reduce the
flow restriction through them.
> However, I'm /really/ sure that I don't want to interpose a flow
> restriction of any kind in the fluid flow.
>
> > While the water pipe regenerator will improve efficiency it may not
> > be worth the added expense. At this point it looks like the
> > reflector can put out more heat than the engine can handle. The
> > money might be better spent going to the next size bigger on the PVC
> > pipe.
>
> As I mentioned before PVC isn't an issue in this all-metal engine. More
> heat than the engine can handle means (I think) overheating to the point
> where it blows all its fluid out the tuning tube... :-(
Or the water starts splashing uncontrolably at high frequency, or the
stroke becomes so long that it eats up too much energy in flow
resistance.
> >> Your help is always much appreciated (I've been in over my head
> >> from the very beginning).
>
> > I'm afraid it's the blind leading the blind here. I know some
> > thermodynamics and physics in general, but I never heard of a
> > fluidyne engine before reading your web page. Considering how much
> > info can be found on them it seems you are at the leading edge of the
> > technology :)
>
> Fluidynes have been around for a while, and aren't new with me. You can
> find a fair number of videos of 'em running on YouTube. What's
> surprising is that fluidynes, and Stirling cycle engines in general,
> have been so little commercialized other than as toys.
Yes, it does seem strange.
Bruce
Morris Dovey
> On May 24, 10:56 pm, Morris Dovey <mrdo...@iedu.com> wrote:
>> Ok. Then if the mass of the water isn't a consideration, and if the
>> water functions only as a displacer, I should be working to reduce
>> the amount of water toward the minimum required to work as a
>> displacer.
>
> Water is cheap. I wouldn't worry too much about reducing how much is
> used.
I think water /isn't/ cheap here. The more of it there is, the greater
overhead it presents. I'll talk a bit about this below, but in the
meantime think about a mechanical Stirling engine built with Un
(unobtanium, which has no mass) pistons, connecting rods, and crankshaft...
<snip Carnot cycle description>
That was my original perspective. It was changed by (for lack or a
better term) communing with the 4" PVC engine...
The engine actually starts out at rest, with everything at ambient
temperature and at least close to atmospheric pressure.
As heat is added to the hot head, the air volume in the hot head expands
- and the hot and cold head "pistons" are depressed equally, and the
output (tuning) piston is raised - a fairly predictable behavior, all in
all.
The big clue is the behavior of the engine when the hot head has been
heated to (again for lack of a better term) its maximum temperature - a
temperature that it seems to not want to exceed.
At that point (/exactly/ at that point) the balance is lost. The hot
head is stuck at that high temperature, but the cold head is bleeding
heat as fast as it can - and the air within it begins contracting.
At that point I see a very small motion, a slight twitch, in the cold
piston, but hear a rushing in the regenerator tube as replacement hot
air is drawn from the hot head.
The first few oscillations tend to be quite small, but build to full
scale in fewer than a dozen cycles - I think more as a result of the
regenerator plumbing coming up to temperature than anything else.
Everything I've seen seems to indicate that the water is a "necessary
evil" - necessary because a displacer is necessary, but evil because it
contributes no energy while presenting an energy "overhead".
The water's mass, if and only if the pendulum's period exactly matches
the period of the Carnot cycle, resembles a flywheel - but if there's
any mismatch at all, the water's inertia will work /against/ the
production of mechanical energy by the Carnot cycle - without delivering
the benefits a flywheel provides to a rotary engine...
...and no matter what, the water presents a power-robbing friction load.
<about the regenerator>
> Yes, the drop is desireable. There is however an optimal drop. It
> is not just a "the more the better" thing.
I had to think about that before I could nod in agreement, but yes - if
the regenerator could magically grab /all/ of the heat, the engine
couldn't run. Trying to engineer an optimal regenerator will probably
require a lot of experimentation, a lot of serious data acquisition, and
enough curve-fitting to keep a computer busy for a while. It probably
won't get done in my lifetime. :-(
> You want the heat to flow through the air pipe where some of it can
> be converted to mechanical energy. Heat that flows through the water
> is just wasted. The fiberglass around the hot end of the water pipe
> will prevent heat loss to the air around the pipe. The hotter the
> water in that end of the pipe the more it resists additional heat
> flowing into it. At the cold end we want the water as cool as
> possible to help cool the air. Leaving that end of the pipe
> unwrapped lets the pipe shed some heat before it gets to the
> air/water interface.
I think we want to keep the water as cool as possible by minimizing the
air-to-water heat transfer on both sides.
On the PVC engine, which has metal heads (and which locates the heat
exchanger /inside/ the hot head), both the hot head and the regenerator
tube are completely wrapped in fiberglass insulation.
>> My regenerator is built as a bundle of small tubes within the outer
>> tube to have a maximum of surface area for grabbing heat from the
>> air during expansion and giving it back during contraction.
>>
>> I'm a bit antsy about the regenerator inhibiting airflow, but
>> recognize the benefits. This is another Stirling trade-off point to
>> be resolved.
>
> Just make the regenerator(s) a bit larger in diameter to reduce the
> flow restriction through them.
Another trade-off. Every time there's a diameter change some power is
lost. I suspect there's a diameter transition geometry that minimizes
the power loss, but I doubt that I can go there with the tools (or
budget) at hand. I would guess that a general regenerator solution might
make a terrific thesis project for some ambitious student <nudge, nudge>.
Bruce Richmond
> Bruce Richmond wrote:
> > On May 24, 10:56 pm, Morris Dovey <mrdo...@iedu.com> wrote:
> >> Ok. Then if the mass of the water isn't a consideration, and if the
> >> water functions only as a displacer, I should be working to reduce
> >> the amount of water toward the minimum required to work as a
> >> displacer.
>
> > Water is cheap. I wouldn't worry too much about reducing how much is
> > used.
>
> I think water /isn't/ cheap here. The more of it there is, the greater
> overhead it presents. I'll talk a bit about this below, but in the
> meantime think about a mechanical Stirling engine built with Un
> (unobtanium, which has no mass) pistons, connecting rods, and crankshaft...
The mass of the pistons, connecting rods, and crankshaft make little
difference in how much power an engine produces at a steady speed.
The heavier parts add a bit of friction but other than that they do
not take any more energy to keep moving at a constant speed. If the
engine needs to accelerate, as in a race car, then the heavier
pistons, connecting rods, and crankshaft act much like a bigger
flywheel.
Thank you for this description of the start up. I've got to build one
of these things :) What have you been using as heat sources up to
now?
> Everything I've seen seems to indicate that the water is a "necessary
> evil" - necessary because a displacer is necessary, but evil because it
> contributes no energy while presenting an energy "overhead".
>
> The water's mass, if and only if the pendulum's period exactly matches
> the period of the Carnot cycle, resembles a flywheel - but if there's
> any mismatch at all, the water's inertia will work /against/ the
> production of mechanical energy by the Carnot cycle - without delivering
> the benefits a flywheel provides to a rotary engine...
>
> ...and no matter what, the water presents a power-robbing friction load.
Agreed on the friction load.
Best set up I can think of at the moment would be to have large radius
bends in the hot and open pipes so they almost form a U, with the cold
pipe rising from a T fitting in the middle. The water could be kept
down to where it is just above the bottom of the U at the lowest point
in the cycle. If you use a piston over the water it will have to stay
in the straight pipe above the curve. The cold pipe water has less
variation in height so the water would be further up in that tube,
about the mid point of the cycle in the hot tube. Cold tube diameter
and air chamber volume can be used to adjust the frequency. For
experimental purposes you might want to put in a section of pipe like
a slide trumbone. After the needed volume is established it can be
made with fewer curves and no slide.
> <about the regenerator>
>
> > Yes, the drop is desireable. There is however an optimal drop. It
> > is not just a "the more the better" thing.
>
> I had to think about that before I could nod in agreement, but yes - if
> the regenerator could magically grab /all/ of the heat, the engine
> couldn't run. Trying to engineer an optimal regenerator will probably
> require a lot of experimentation, a lot of serious data acquisition, and
> enough curve-fitting to keep a computer busy for a while. It probably
> won't get done in my lifetime. :-(
I think you should be able to get close with a few tries.
> > You want the heat to flow through the air pipe where some of it can
> > be converted to mechanical energy. Heat that flows through the water
> > is just wasted. The fiberglass around the hot end of the water pipe
> > will prevent heat loss to the air around the pipe. The hotter the
> > water in that end of the pipe the more it resists additional heat
> > flowing into it. At the cold end we want the water as cool as
> > possible to help cool the air. Leaving that end of the pipe
> > unwrapped lets the pipe shed some heat before it gets to the
> > air/water interface.
>
> I think we want to keep the water as cool as possible by minimizing the
> air-to-water heat transfer on both sides.
Not sure how you would do that other than the floating piston.
> On the PVC engine, which has metal heads (and which locates the heat
> exchanger /inside/ the hot head), both the hot head and the regenerator
> tube are completely wrapped in fiberglass insulation.
>
> >> My regenerator is built as a bundle of small tubes within the outer
> >> tube to have a maximum of surface area for grabbing heat from the
> >> air during expansion and giving it back during contraction.
>
> >> I'm a bit antsy about the regenerator inhibiting airflow, but
> >> recognize the benefits. This is another Stirling trade-off point to
> >> be resolved.
>
> > Just make the regenerator(s) a bit larger in diameter to reduce the
> > flow restriction through them.
>
> Another trade-off. Every time there's a diameter change some power is
> lost. I suspect there's a diameter transition geometry that minimizes
> the power loss, but I doubt that I can go there with the tools (or
> budget) at hand. I would guess that a general regenerator solution might
> make a terrific thesis project for some ambitious student <nudge, nudge>.
Hope that's not directed at me. I graduated in the class of '71.
Cone shaped nozzles will reduce turbulance when making a diameter
change. Might be able to find off the shelf reducers that smooth the
transitions a bit.
Bruce
Morris Dovey
> On May 25, 10:54 am, Morris Dovey <mrdo...@iedu.com> wrote:
> Thank you for this description of the start up. I've got to build
> one of these things :) What have you been using as heat sources up
> to now?
Remember that the only version built and tested so far have been a
couple of low-temperature engines. Just about anything hot: Heat gun,
hair dryer, propane torch, oxy-acetylene torch, water heater elements,
toaster-oven heat elements, a bearing-heater element, and (I almost
forgot <g>) the sun.
>> Trying to engineer an optimal regenerator will probably require a
>> lot of experimentation, a lot of serious data acquisition, and
>> enough curve-fitting to keep a computer busy for a while.
>
> I think you should be able to get close with a few tries.
I think I could optimize a single regenerator with 6 or 8 tries, but to
arrive at a general solution that allowed a designer to calculate a
solution for a new engine with a high degree of confidence won't be
quite so easy.
>> I think we want to keep the water as cool as possible by minimizing
>> the air-to-water heat transfer on both sides.
>
> Not sure how you would do that other than the floating piston.
I think that there might be a balance point between minimal interface
surface area and length of the stroke.
>> I would guess that a general regenerator solution might make a
>> terrific thesis project for some ambitious student <nudge, nudge>.
>
> Hope that's not directed at me. I graduated in the class of '71.
Nope, I was thinking about even younger folks - current students who
have access good labs, mentors, and (perhaps) collaborators close at hand.
> Cone shaped nozzles will reduce turbulance when making a diameter
> change. Might be able to find off the shelf reducers that smooth the
> transitions a bit.
I'd wager that the optimal solution will more resemble an asymetrical
ogee shape - but I wouldn't be willing to wager /much/. :-)