Thermal Properties Problem
Will Cleaver
I've been talking with Eerik form Solar Fire and the solar
concentrators - http://www.youtube.com/watch?feature=player_embedded&v=CXJgAmft2jI
He presents a simple problem and am wondering if anyone here can help
with the solution?
"I can calculate how much energy is reflected by the mirrors, and I
can determine a boiler radius and so discard any energy that falls
outside the boiler due to focal aberration throughout the day (some
light we allow to miss to keep a reasonable boiler size.).
What I can't do is the optimization of the boiler size; i.e. at what
point does increasing the boiler size create more losses than gains
(through convection with the air and radiating heat back into the
environment).
There's relatively simple equations to implement this, but I can only
easily understand the math part. How to judge the emissivity of steel
(does what kind of steel make a difference, how it's cut, will some
sort of paint help etc.), heat transfer coefficient through the steel
into the boiler (depending on steam temperature), etc.
With these equations in hand and reasonable values representing a
typical boiler (or many types of boilers, or general equations to
determine boiler properties etc.) then I can run an optimization to
output the ideal boiler size considering the intended application
temperature etc.
For now, I just try to keep the concentration factor at around ~200,
which seems a good value.
second problem:
What should we assume about losses in the steam pipe per meter square?
We used about 4-5cm of glass wool (resistant to 1200 C); good, too
much, not enough ?
Cheerio
Eerik"
Dorkmo
http://www1.eere.energy.gov/industry/bestpractices/software.html
On Nov 26, 3:11 pm, Will Cleaver <willclea...@gmail.com> wrote:
> Hi Guys,
>
> I've been talking with Eerik form Solar Fire and the solar
> concentrators -http://www.youtube.com/watch?feature=player_embedded&v=CXJgAmft2jI
jamie clarke
This means that the heat transfer through the steel can be neglected as the process will be nearly adiabatic. Heat transfer from the boiler would only begin to occur when the sun was gone. As an evacuated borosilicate shell for the boiler would probably be expensive you may need this link:
http://www.engineeringtoolbox.com/overall-heat-transfer-coefficients-d_284.html
Mild steel has a greater heat transfer coefficient then stainless steel or cast iron.
This is to do with the crystal structure and carbon content.
Black surfaces have an emissivity closer to 1 and highly polished reflective materials have and emissivity closer to 0. Highly polished silver has an emissivity of 0.02.
Emissivity calculations assume the material is thicker then the wavelength of the optical photon.
Other geometrical considerations are very important. A sphere will have the least surface to the greatest volume, this will make heat transfer away from the material. A tetrahedron has the greatest amount of surface and the least amount of volume of any regular geometry.
Thanks steam guys! I must have learnt something
jamie clarke
Steam to mild steel to air= 11.3W/m^2 K (boiler radiation to environment)
Water to mild steel to water= 340 W/m^2 K (feed water entering boiler)
Steam to mild steel to water= 1050W/m^2 K (Cooling power of a heat exchanger)
Will Cleaver
Thanks for the info Jamie.
From Eerik;
"We looking for more basic level stuff, such as what happens when light hits a surface. The easy equation that tells you how much energy you'll absorb based on steel thickness, concentration factor
Mike Stone
> Black surfaces have an emissivity closer to 1 and highly polished reflective materials have and emissivity closer to 0. Highly polished silver has an emissivity of 0.02.
You can also look at 'black body approximation'. Make a coiled tube with one end closed and one end open. Put the closed end in your boiler. Shine your light into the open end.
Light going into the tube will reflect off the inside surface of the tube many times, with a certain amount of absorption every time. The path that allows light to reflect all the way in and all the way back out is vanishingly small, so the hole approximates a perfectly black, perfectly absorptive surface.
Better yet, it allows you to insulate your boiler well and pipe the light in through the tube. The only uninsulated surface radiating heat to the atmosphere will be the tube itself.
Copper pipe painted black inside should give you good absorption and heat transfer.
Dorkmo
http://www.redrok.com/concept.htm#absorptivity
http://en.wikipedia.org/wiki/Selective_surface
http://en.wikipedia.org/wiki/Concentrating_solar_power
http://en.wikipedia.org/wiki/Absorbance
http://en.wikipedia.org/wiki/Stefan-Boltzmann_law
http://galileo.phys.virginia.edu/classes/252/black_body_radiation.html
heres something about insulating pipes
http://www.wbdg.org/design/midg_design.php
Dorkmo
it was, it would just take longer to heat up and cool down? i think
the only consideration for thickness would be how much pressure from
the steam is on the container is? does any of the sound right? maybe i
should stop talking lol.
On Nov 27, 1:30 pm, Will Cleaver <willclea...@gmail.com> wrote:
> Thanks for the link Dorkmo. That is more whole-systems type stuff, for
> large plants.
>
> Thanks for the info Jamie.
>
> From Eerik;
> "We looking for more basic level stuff, such as what happens when light
> hits a surface. The easy equation that tells you how much energy you'll
> absorb based on steel thickness, concentration factor
> and temperature inside the boiler. it would only gain ~5 %"
>
> On 27 November 2011 15:53, jamie clarke <jamieclarke...@gmail.com> wrote:
>
>
>
>
>
>
>
> > Heat transfer coefficients for Mild steel:
>
> > Steam to mild steel to air= 11.3W/m^2 K (boiler radiation
> > to environment)
> > Water to mild steel to water= 340 W/m^2 K (feed water
> > entering boiler)
> > Steam to mild steel to water= 1050W/m^2 K (Cooling power of a
> > heat exchanger)
>
> > On Sun, Nov 27, 2011 at 3:34 PM, jamie clarke <jamieclarke...@gmail.com>wrote:
>
> >> The emissivity of a perfect vacuum is zero. If you encase the boiler in
> >> borosilicate glass and evacuate the air then any heat transfer from the
> >> boiler to the atmosphere will be radiant conduction and convection will be
> >> negligible.
>
> >> This means that the heat transfer through the steel can be neglected as
> >> the process will be nearly adiabatic. Heat transfer from the boiler would
> >> only begin to occur when the sun was gone. As an evacuated borosilicate
> >> shell for the boiler would probably be expensive you may need this link:
>
> >>http://www.engineeringtoolbox.com/overall-heat-transfer-coefficients-...
jamie clarke
Heat travels through a vacuum as an infared photon so in essence heat can be just a form of light but our eyes do not detect that wavelength. Red visible light is 540 nanometer waves and infared light(heat) is 1 micron waves.
So imagine you are a electromagnetic wave traveling through a crystal(steel)
If you are infared and the surface is less than a micron(paint) the crest and trough of your wave will pass through the 1 micron layer of paint without being absorbed.
If the layer of paint was thicker then 1 micron it would essentially be infinitely thick because the infared photon would not pass through the paint but be absorbed by the paint and then the paint would have to transfer its energy to the steel by conduction.
Conduction is different to radiation and conduction is transmitted by PHONONS and not photons. Phonons are sound waves that travel through a crystal.
Dunno if this makes any sense to anyone but photons either pass through a surface or there energy is completely absorbed by the atoms in that surface and re-emitted(Radiant heat transfer). Phonons are sound waves and they propagate through the atomic lattice of a surface. (conduction)
So in the case of emissivity heat behaves as a light wave and in the case of coeffecient of heat transfer it behaves as a sound wave.
Sound needs a material to propagate through but light can propagate through a vacuum.
Its some pretty interesting physics :D
Eerik Wissenz
Hey all,
Thanks for the replies on this problem. Many of the links may help.
To answer quickly a few points. The thickness of the steel is important since it will determine how fast the heat flows into the water. The thicker the steel, the slower the heat will get to the other side, and so the hotter the hot (solar absorbing side) will be. The hotter the absorption surface, the more energy it will emit back into the environment. And yes, of course if the steel is too thin then the boiler can explode.
The problem is not simply a black body problem, because the goal is to to take the energy away through the steam. Unfortunately, the boiler can't be encased in a vacuum since it needs a surface to be exposed to the incoming light. So, even if we assumed steel acted as a black body, we don't know it's temperature because it is being cooled continuously by the evaporation of steam. I.e. if the focal point is 700 C and the steam temperature is at 200 C, then the absorbing surface will be somewhere between the two depending on the properties and thickness of the material.
The goal of all this is to be able to determine precisely the amount of energy that is absorbed and emitted over the entire absorptive surface. By modeling that the computer can then discard any region that emits equal or more energy than it absorbs over the course of the day.
The problem is well understood for combustion boilers, but in these the heat transfer is gas and radiation to surface, and there are other parameters like the need to have enough space for the oxygen-fuel to mix and burn properly. With solar concentration the initial heat transfer is only radiation to surface, and there is no need of volume for combustion. However, the smaller and hotter one makes the focal point, the more precise and expensive the concentrator has to be; but too wide a focal point and the heat transfer is very inefficient and investing in more precision is worth (this also depends on the intended temperature of the application). So a correct modeling of the heat transfers is required to determine the most cost effective solution.
Best regards
Eerik
Eerik Wissenz
Free Access to Solar Energy
www.solarfire.org
SiSustainable - Web Development
www.sisustainable.com
Eerik Wissenz
jamie clarke
Alot of the wavelengths will pass through the glass to heat up the boiler.
Eerik Wissenz
Encasing the boiler in glass is not used in high concentrations, only lower concentrations. At high concentration about as much or more light is reflected off the glass than is saved through vacuum insolation. High concentration is already by nature fairly efficient since the large difference in temperature tends to dominate other factors. That's why this isn't such an important question and I'm guessing would only be about ~5% efficient.
Eerik
jamie clarke
I have been pondering the question quite a lot but it we had an array of sf32's in the future what would be the highest temperature you could achieve?
To rephrase the question, whats the highest temperature you could theoretically achieve and what would have to be different to achieve that?
Eerik Wissenz
The maximum theoretical temperature for solar concentration is 5300 C, as beyond this temperature we could start heating the sun surface rather than the other way around.
For Solar Fire, the focal temperature varies during the day. 800-1000 C at noon and maybe 600~ in morning/afternoon; it of course depends where you are on the planet and when you start measuring focal temperature, but say for a 7-8 hour production in the tropics 700 C is probably around the good number. But also, focal point temperature varies over the absorption surface (i.e. some hotspots may be significantly hotter than the edges), but to consider the whole absorptive surface I need to know how the heat behaves in the metal, considering it's being cooled by steam evaporation. The hottest solar fire has been measured at is 900 degrees, but that was on a machine significantly smaller than the new large one which have also a more precise technique, so going over 1000 C should not be a problem but I won't claim so until I've measured it. However, higher and higher temperatures aren't really desired since it fatigues the metal. 700 C is already fairly hot.
Best regards
Eerik
jamie clarke
I think i provided the values and hopefully with both pieces of information you can proceed.
Thank you for the information Eerik, im even more excited now.
Daniel Connell
I'd only go with steel if you're looking to contain some serious pressure. Which I guess you might.
In terms of geometry, glass, vacuum etc, I've been thinking about something along these lines:
http://www.solarflower.org/boiler_01.jpg
Which is a metal cylinder insulated on every surface except the bottom, which is a steep, black, inverted cone capped with high temp glass. The concentrated light enters through the glass and hits the metal. Most of the energy is absorbed through and into the water, that which is reflected bounces again, about four or five times depending on the angle. The cone should be able to take higher pressure, so you could have only this in copper, welded to a thicker steel cylinder.
I wouldn't bother with vacuums, to be honest, they need to be pretty much complete to offer any insulation, and pinhole leaks are hard to avoid. Also black body radiation losses at these kinds of temperatures can run surprisingly high. Just glass and air does a pretty good job, you don't need it perfect.
As far as calculation goes I generally take these kinds of problems to either CR4 ( http://cr4.globalspec.com ) or the Physics Forum ( http://www.physicsforums.com ), but expect both to tell you all the ways you should be doing it different before they finally get around to figuring you a useful answer.
Bloody engineers.
But I suspect this is a bit too real-worldy to get an accurate answer for. Probably better make one...
Daniel.
jamie clarke
He suggested having fins within the boiler and also suggested copper.
Ive had an idea, how about having the pressure container that daniel has drawn, made of sheet metal. Then within this steel container have the preheated water. Then a copper coil immersed could generate the steam.
The main variables are he discussed are temperature difference, the area of the boiler, the coefficient of heat transfer, the specific heat capacity of the water and the flow rate among probably a ton of other things.
