h a l f b a k e r yThe Out-of-Focus Group.
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
Please log in.
Before you can vote, you need to register.
Please log in or create an account.
|
Oil filled electric radiators have an electric heating element which heats the oil, and the oil conveys the heat to the fins, which convey the heat to the air.
I propose a device which would look very similarly, but instead of having oil inside, would have water and low pressure steam.
The heating
element would heat the water, which would boil, increasing the pressure.
The increased pressure would cause the temperature of steam that's inside the fins, to increase, so that it's above the temperature of the metal of the fins. Heat would be conducted out of the steam, into the fins, into the air.
Due to the removal of heat, the steam in contact with the fins would condense and drip downward to the heating element.
The heat provided by this water and steam filled radiator would be just as even as that provided by the common oil filled version, but the device would be much lighter, since most of the H2O would be in the form of steam (which of course has a low density).
Lighter weight makes it easier to move around, and reduces shipping costs.
The only minus is that unlike an oil filled radiator, it wouldn't be silent -- there'd be the constant noise of water boiling.
Hex
http://en.wikipedia...ranium_hexafluoride
Free to good home ... [8th of 7, Oct 19 2009]
Phase diagrams
http://en.wikipedia...m#2D_phase_diagrams
Essential reading [8th of 7, Oct 20 2009]
[link]
|
| |
than and, well, the tendency to explode. |
|
| |
It may be lighter, but not because some of the water is steam, m'kay? |
|
| |
Why would it tend to explode? |
|
| |
Implode, maybe, but an explosion could only occur if the temperature of the steam inside the radiator were over 212F. |
|
| |
The only reason to buy an electric radiator (instead of some alternative type of electric heater) is that it's got such a large surface area, which allows it to warm the room, while being merely warm to the touch. |
|
| |
The surface of the radiator should reach 120F, at most. If hotter than this, burn injuries due to touching the heater become possible. |
|
| |
To prevent such overheating, there would be an internal temperature switch, which would also serve to shut off the element, if the steam reaches too high a pressure. This internal temp switch would be placed at the highest point in the system, where (under normal operation) it will be surrounded by steam; if the radiator has a tip-over shutoff switch, we can be sure that this will be the case at all times. |
|
| |
For even more safety, we could add a pressure switch, which could also shut off the heating element if the pressure inside ever rose enough to be close to atmospheric temperature, but unless there were a leak, allowing air in, that switch should never activate. (Why? Because if there's no air inside, only H2O, and if the temp is never above 120F, then the pressure will always be below 1 atm.) |
|
| |
This of course would be in addition to the external thermostat, which would turn the heater on or off in response to the temperature of the room the heater is in. |
|
| |
phoenix, surely the weight of a quart or so of water, and a gallon or two of steam, will be less than that of the gallon or two of oil in an oil-filled radiator. |
|
| |
If the air is pumped out of the radiator, leaving only water vapour, then the water will equilibrate the pressure by evaporating. The vapour pressure of water at 20C is 23.38 millibar; atmospheric pressure is about 1000 millibar, depending on the prevailing weather. Thus the casing will have to be strong enough to withstand 1 atmosphere external pressure without imploding. |
|
| |
So, let us assume that the interior of the radiator is filled with Nitrogen at 1 Atm to reduce corrosion and prevent implosion. At 50C, there will be no significant increase in internal pressure; but there will be relatively little evaporation of water as the water is nowhere near its boiling point (at 1 Atm pressure). |
|
| |
Heat distribution away from the heating element will therefore depend on convection of Nitrogen with a little water vapour. This is relatively inefficient; it would be more effective to have a simple free-air convector heater. |
|
| |
(This would work rather better if you selected a working fluid with a boiling point around 50C. Candidates would be Acetone or Carbon Disulphide) |
|
| |
Why *can't* the casing be strong enough to withstand 1 atmosphere external pressure without imploding? |
|
| |
As for filling the interior with nitrogen... that would go *completely* against the point of the idea, which is to use pure water/steam, so that a heat pipe effect occurs between the heating element and the surface of the fins. |
|
| |
As for preventing corrosion -- I don't see why we'd need to do anything more than apply a thin layer of paint to the interior of the radiator, to do that. |
|
| |
I *might* consider switching to a different refrigerant, instead of water, one whose vapor pressure at room temperature (20C) is moderately higher, to reduce the pressure differential, but I would still use only the one fluid. |
|
| |
Maybe, as you suggest, acetone, but certainly not carbon disulfide. The latter has too low an autoignition temperature (90C), and since I never heard of it till you mentioned it, and I searched on wikipedia, it's probably more expensive. |
|
| |
// Why *can't* the casing be strong enough // |
|
| |
It can. But that means a strong structure, so either thick walls, or very strong, stiff material, like Titanium - which is rather expensive, and this is presumably a mass-market product. |
|
| |
// not carbon disulfide // |
|
| |
Come now, where's your sense of adventure ? CS2 isn't expensive, and what's wrong with a cheap domestic appliance containing a couple of kilos of spontaneously-flammable, toxic, corrosive liquid being heated above its boiling point ? It's not like it's Hex, or anything. So, there may be the odd product liability issue to address, but if they don't sell, you can always flog them off to the military as cheap and cheerful hyperbaric weapons. |
|
| |
Steel or copper will not corrode in a pure water atmosphere. Very little will. If you pump it to a near perfect vacuum, then add the right amount of water, the unit will equilabrate at a sub 1 atmosphere pressure with partial liquid water and partial water vapor. |
|
| |
Likewise, 1 atmosphere of pressure is not all that much, your garden hose provides at least as much, and doesn't tend to crush things. (Seriously, 1 atmosphere=14.7 PSI. Household water pressure is a minimum of 15 PSI, typically noticeably higher) Thin copper household pipes are genrally rated for greater than 100 PSI, and plenty strong enough (even in compression). |
|
| |
// Steel ... will not corrode in a pure water atmosphere. // |
|
| |
So why are corrosion inhibitors added to central heating systems ? |
|
| |
Answer: any system of dissimilar metals will generate a potential difference. Even if the water is pure to start with, a few ions will be liberated to get the reactions going. Eventually one metal will oxidise and nascent hydrogen will be released. |
|
| |
Household water piping is operating in tension, not compression. And it is a tubular structure .... [goldbb] is proposing a device with large flat regions to act as dissipators (convectors). Look at how an automotive radiator is constructed - do you think it could withstand having all the air vac'd out of it ? |
|
| |
We are of course agreed on the equilibration - see the numbers in previous anno. |
|
| |
lets talk some reality about water: |
|
| |
It freezes. Your device will need to be rupture proof if the entire contents freeze, furthermore the functionality must not be impaired. |
|
| |
It causes corrosion: Even without oxygen present water will allow corrosion, especially hot water, especially boiling water. |
|
| |
It will conduct electricity. |
|
| |
If you contain it in a strong container, that container will need to rupture failsafe in the event of an overpressure. |
|
| |
Thermostats suspended in water are never treated as failsafe. An additional safety mechanism is always present. |
|
| |
First Corrosion: Household piping either for heat or water
is not a pure water system. There is dissolved oxygen,
there is leakage, there are other problems. Water does
not corrode, oxygen does. And while some oxygen is
unavoidable, a minimal amount is not going to act
sufficiently to clog the system. Absolute worst case, a
small sacrificial anode would be sufficient to preserve the
unit for decades. And other than that anode, why would
you have a dissimilar metal? Make it all out of stainless. |
|
| |
Second, operation under pressure: Round tubing is equally
resistant to uniform pressure from the inside and the out.
Most heat exchangers are made of round tubing with fins
attached (look closer at a car radiator). Also, I repeat, 15
psi is nothing. High quality metal snips, exerting a their
cutting force over an area less than 1/100" of a square
inch, with an average human grip strength of 25 pounds are
exerting 2500 psi. These snips will have difficulty cutting
through .040" Stainless. 15 psi is nothing. |
|
| |
Third properties of water: Pure water does not conduct,
but I'll ignore that for the moment, since you will have
some metal ions floating around. Next question, why do
you care? The thermostat does not need to be immersed.
The temperature response lag on the outer surface will be
minimal with metal as thin as this requires. This also
means that you can use 2 or 3 thermostats without any
difficulty. Simply build the device with a margin between
maximum temperature and fail temperature and this is not
a problem. Likewise, your electric heaters are not
immersed in water. They are embedded in a large metal
block that makes up the bottom portion of your heat
exchanger.
Next, freezing: Surprisingly enough household hot-water
pipes are not built with a pressure relief against freezing.
This unit is meant for indoor use, and is actively heated,
there should not be a major issue here. If there is the
unit is destroyed, which is why it is labeled do not store
below "x" where x is 20 degrees above the freezing point at
the given pressure. If you are truly worried about this, a
small amount of ethylene glycol in the mix will reduce
your heat transfer rate, but will also lower the freezing
temperature.
Finally, rupture fail safe: Pressure relief valves are a
simple item, found on every hot water heater in
existence. Set it up such that water at the triggering
pressure expands and thus cools sufficiently that the
emitted steam is not a risk to humans, and you don't have
a problem. The addition of baffles or similar give it time
to cool before emitting to atmosphere. |
|
| |
If most of these were a major problem, steam radiators
would not exist. And they've been around for quite a
while. |
|
| |
every steam radiator that I have ever seen was a circulating system with boiler, condensers, and quite a lot of redundancy for safety and reliability. An electric heater in a sealed cavity that also contains water is simply not a more efficient or safer way to heat a room than a fan circulating or element radiating electric heater. |
|
| |
Um, this is a small circulating system with boiler, condenser, and you can build in as much safety and redundancy as you like. |
|
| |
That being said, I'm not commenting on the efficiency, thermo is not my favorite subject. This unit would be lighter(1/4-1/2 full of water rather than 100% full of oil). For that lighter weight, it would provide the same convective surface as an oil filled radiator, making it more efficient than a simple radiant element, and not requiring the electricty for the circulating fan of a forced air unit. I don't have the numbers available, but the idea !is! technically feasible, which is my primary argument. |
|
| |
// making it more efficient than a simple radiant element // |
|
| |
No, it won't. The thermodynamic efficiency will be the same. |
|
| |
If you have a 1000W heat source, then once the device has equilibrated with its environment,it will dissipate 1000W of heat, irrespective of the physical shape of the device. Some parts of the device may be at different temperatures, and indeed it is quite possible for these parts to exceed their melting point - but, over all, the efficiency will be 1. |
|
| |
Oil filled radiators exist to distribute energy at a uniform, relatively low, temperature over a large surface area. Convector heaters have an element that may be at several hundred degrees C, in a protective enclosure, as do fan heaters. |
|
| |
The primary concern in all these designs is to maintain the exterior of the device at a temperature which will not cause harm to the user if they make contact with it. |
|
| |
If you consider the phase diagram for water <link> the relevant data are the critical temperatures and pressures relating to the liquid-gas phase transition. |
|
| |
I was using efficiency in the functional, rather than thermodynamic sense. A strong natural convection current will distribute heat more effectively through a room than a radiating unit. This will provide more uniform heat, generally a desirable result. |
|
| |
Why the argument over niggling details? |
|
| |
Efficiency:
The unit will not be any more thermodynamically efficient than a similar oil-filled radiator.
Heat in = heat out. Period.
If you want more heat output without higher surface temperatures, increase the surface area and/or add a fan to provide more airflow than natural convection alone. |
|
| |
Buckling:
Wide, flat fins would most likely buckle under 14.7 psi. Imagine a pile of tubes soldered together side-to-side, then slit along the solder line. Now, image two panels stamped into that shape or something similar. They'll support 1 atm just fine. |
|
| |
All the concerns about safety:
Build in appropriate safeguards:
1: If the unit gets too hot, shut it off.
2: If the heating element gets too hot, shut it off.
3: If the unit is empty, shut it off.
4: If the unit overpressures, shut it off.
5: If any of the sensors fail a self-check, shut it off.
etc...
With appropriate monitoring and cross-checking of sensors, this can be a multiple-redundant system. |
|
| |
Corrosion:
Sure, water can corrode metal. Make it out of something that corrodes slowly and provide a sacrificial anode. Done. |
|
| |
Maintenance of vacuum:
Not mentioned yet, but vacuum is tough to maintain. One tiny leak, and the unit is broken. Contamination of the working gas with air will destroy the efficiency of a heatpipe. Even this is not a huge problem. Just make the heatpipe radiator unit as a single weldment. Fill and suction-drain the unit at the factory, then solder-seal the fill port. Use a contacting-element heat source (a-la electric stove burners heating your water kettle). |
|
| |
All the objections to this idea have been over design detail issues which could be easily overcome. |
|
| |
If it makes it less likely to implode (buckle), I'd be more than willing to switch from a wide flat fin design, to round cylindrical tubes. We might want the tubes to have fins attached, for more surface area, but that's not a problem. |
|
| |
Actually, using cylindrical tubes instead of wide flat fins presents another advantage - they could be individually produced (evacuated of air, have water added, and soldered closed), then attached to the base heating unit using thermal adhesive. |
|
| |
Since each tube is a very simple component, reliability would by higher, and production costs would be lower. |
|
| |