Liquid Nitrogen as building cooling?

Discussion in 'Highly Technical' started by imported_Tick, Jun 14, 2006.

  1. imported_Tick

    imported_Tick Diamond Member

    Joined:
    Feb 17, 2005
    Messages:
    4,685
    Likes Received:
    0
    Modern air conditioners are horribly inefficient. A possible solution that occured to me was to do the cooling in large plants, that can be very efficient, and ship the cold to people. The obvious way to do this is a compressed gas. LN2 seemed a good choice, because it only results in nitrogen, so it could easily be vented to atmosphere. Further, LN2 can be bought cheaper by the gallon than milk. I figured the distribution system would be that you would have a volume of liquid nitrogen delivered every 3 months or so, and you would store it in a cryogenic container. Then, you would just use it to chill incomming air. I fail to see how it could not be more efficient than the current system, but what concerned me was the question of how much liquid nitrogen would be needed. Would it be like a cubic yard per month, or 200 cubic yards a month? I don't have the thermodynamics skills to calculate that. So, I was wondering if this was a feasible idea.
     
  2. jagec

    jagec Lifer

    Joined:
    Apr 30, 2004
    Messages:
    24,442
    Likes Received:
    0
    Modern air conditioners really aren't all that inefficient. Maybe window units, but a central, large-scale unit to do all the cooling wouldn't be all that much better compared to commercial grade HVAC systems.

    You'd need a LOT of LN2...assuming that the majority of heat is removed by boiling the LN2, and not by warming the gas produced, it's about 161 kJ/L. According to this website I just found, a small 300 sq ft office with a couple people in it will require about 1.5-2 kW of cooling during the daylight hours, and .5 at night (link). Using the lower figure, that's 43,200 kJ for an 8-hour workday, plus 28800 at night...or 447L of LN2 (or .58 cubic yards per day, to use your units). Keep in mind this is for ONE smallish room...an entire building will have much, much greater cooling needs.
     
  3. Gibsons

    Gibsons Lifer

    Joined:
    Aug 14, 2001
    Messages:
    12,486
    Likes Received:
    4
  4. Mark R

    Mark R Diamond Member

    Joined:
    Oct 9, 1999
    Messages:
    8,496
    Likes Received:
    0
    It comes down to efficiency and thermodynamics.

    It becomes more and more difficult to cool something, the greater the difference between your cold material and your heatsink. The maximum theoretical effectiveness (not efficiency), also called COP, of a refrigerator is given by COP = Hot temp / Temp difference [all temperatures in Kelvin].

    So, in the case of a typical AC system, the heatsink temperature is about 50 C (323 K), and the cold side is about 5 C. This gives a maximum theoretically possible COP of about 7. In practice, AC systems aren't perfectly efficient, and a typical modern system will have a COP of about 3.5 (an efficiency of about 50%).

    What about liquid nitrogen? Lets assume a similar heatsink temperature, but this time the cold side is about 90 K. Maximum possible COP in this case is 1.3. Indeed, because insulation becomes very difficult at these temperatures, cryo coolers are much less efficient than AC - 15% efficiency is considered excellent. So typical COP for a cryocooler would be about 0.2.

    So, for a typical AC system, each unit of electricity you use will shift 3.5 units of heat - in a Liquid nitrogen system, each unit of electricity will only shift 0.2 units of heat.
     
  5. herm0016

    herm0016 Diamond Member

    Joined:
    Feb 26, 2005
    Messages:
    5,680
    Likes Received:
    3
    air conditoning? what air conditioning...??? lol, the only building here that has air is the theatre in which i work. its uses a standard condeser/evporator setup with a fairly inert gas for the transfer agent.
     
  6. Born2bwire

    Born2bwire Diamond Member

    Joined:
    Oct 28, 2005
    Messages:
    9,843
    Likes Received:
    0
    The city of Chicago also has large ice houses to chill water and send that out as a means of cooling. They did a lot of construction on campus in Urbana to lay the pipes in for the cold water. Most of the climate control for the Uni's buildings use steam or chilled water. The steam is produced at the town's power plant and heats the buildings, runs the dishwashers, etc. The cooling system is new and is up and running now. As far as I can tell, it does a good job. The idea of using conduction methods of heating and cooling versus traditional convection are not new. Sometimes they use the water in radiators to heat or cool forced air, or they will run pipes in the flooring to pump the water through.

    LN2 would probably a difficult system to work with. It would be difficult to pipe through the building like water or natural gas because the gas would quickly boil and then you have to pump both a liquid and a gas. You would also have to find a way to distribute the LN2 throughout an area without it boiling off before reaching all the way through the room. You also have problems of ice forming around the storage systems and delivery systems. There is a large LN2 tank outside my building and even in 100 degree weather there are large blocks of ice that form around the taps and outlets. Water is better in my mind because you can cool it to a low temperature and it will still be a liquid and it has a large heat capacity so it is able to absorb a lot of heat from the environment.
     
  7. f95toli

    f95toli Golden Member

    Joined:
    Nov 21, 2002
    Messages:
    1,547
    Likes Received:
    0
    No, it is possible to distribute LN2 using pipes. At the university where I used to work there are a couple of big LN2 tanks outside the building and then there all pipes going to all the LN2 cooled equipment (mainly pumps, LN2 is used in cryopumps and coldtraps in e.g. diffusion pumps). It is expensive but possible.
    However, it is still be useless for cooling a whole building (although occasionally I do use it the reduce air humidity in the lab where a I work, a bucket of nitrogen is a good way to remove moisture from the air)

    Btw, the blocks of ice you mention are probably forming at the pressure release valves. If it indeed forming at the taps there is something wrong with the insulation.


     
  8. silverpig

    silverpig Lifer

    Joined:
    Jul 29, 2001
    Messages:
    27,710
    Likes Received:
    0
    Also, while LN2 is cheap, a good cryogenic dewar is expensive.
     
  9. dkozloski

    dkozloski Diamond Member

    Joined:
    Oct 9, 1999
    Messages:
    3,005
    Likes Received:
    0
    We pump ground water. It comes out of the ground at about 38F and is treated with a polymer to prevent scaling of the heat transfer coils in the air handlers. Heat exchangers have the tubes scrubbed once a year. Simple, no cryogenics, no compressors.
     
  10. Duddy

    Duddy Diamond Member

    Joined:
    Jul 22, 2002
    Messages:
    4,640
    Likes Received:
    0
    Expensive.
     
  11. Jeff7

    Jeff7 Lifer

    Joined:
    Jan 4, 2001
    Messages:
    41,584
    Likes Received:
    2
    There's the problem right there - distribution. If it were to be piped, the pipes would need to be kept supercold or else the nitrogen would just boil away, and then you'd better have pipes that can withstand incredible pressure. So the pipes would be either encased in cooling equipment, or else they'd have to have very thick walls that can withstand extreme cold. And in that case, there's inefficiency introduced. The cold is "leaking" out through the pipe then. Look at when a rocket is launched which uses liquid oxygen. The New Horizons probe for example, was launched on an Atlas V rocket. The outside of the rocket shed a lot of ice from the vibration of the engines igniting - the outside of the booster was chilled by the liquid oxygen within, onto which water condensed and then froze. The fact that the thermal difference was felt on the outside of the rocket represents an inefficiency, just as it would in a liquid nitrogen pipe.

    Another option that I just came up with - a magnetically levitated pipe suspended within a larger evacuated pipe. But again, both the inner and outer pipes would have to be built to withstand the pressure differential.


    And one other issue - how do you think they make liquid nitrogen? It's likely done with coolers of the same family as that in your air conditioner. Plus with an air conditioner, you're talking about a thermal differential of maybe 30-40 degrees - outside temp may be 100-110F, while inside it'd be close to 70F. Liquid nitrogen - the thermal differential is more like 390F. More expensive and complicated equipment is needed to produce that kind of temperature.

    In short, liquid nitrogen cooling - not safe, not practical, and very expensive.
     
  12. Howard

    Howard Lifer

    Joined:
    Oct 14, 1999
    Messages:
    47,964
    Likes Received:
    1
    Doesn't it displace oxygen?
     
  13. f95toli

    f95toli Golden Member

    Joined:
    Nov 21, 2002
    Messages:
    1,547
    Likes Received:
    0
    You should have read my post. It is definitly possible to distribute nitrogen using pipes. Yes, it is expensive because they need to be insulated but is some cases it is worth it, it is still more efficient to keep the LN2 in big storage tanks and then disitribute it using pipes than to use many small dewars.
    And I don't understand you comment about the pressure. Liquid nitrogen does of course boil off but as long as you have e.g. a spring loaded pressure release valve that is not a problem, the pressure is not high. Moreover, you can use standard stainles steel pipes, you do not need a very thick wall.
    In fact, I often use ordinary plastic tubes when I transfer liquid nitrogen from transport dewars to cryostats. The plastic becomes very stiff an brittle but as long as you do not try to bend it while it is still cold it is safe.




     
  14. OSX

    OSX Senior member

    Joined:
    Feb 9, 2006
    Messages:
    660
    Likes Received:
    0
    The large amount of piping needed to transport the liquid LN2 would likely be the Achilles' heel of the system, because you'd likely pick up a large amount of heat from the surrounding environment.