Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Air-exchange number

  1. Jan 25, 2006 #1
    I'm wondering about a term in ventilation called the air-exchange number. It is used to calculate a buildings heat loss through infiltration. It's measured in 1/h (h=hour), and is calculated as L/V where L is m3/h and V is the room volum in m3. I'm wondering about this L, is this the amount of air brought into the room by the vent system? I just can't see what else it could be...

    I'm supposed to evaluate the accuracy of a specific HVAC program. I thought I would first calculate (by hand) a rooms different u-values. Then I will heat up the room to a certain temp, let it cool for a certain time period, and then measure that room and adjacent rooms' temps. This will give me a loss of Watts "before and after". Now I'm not quite sure what to do with these values, not even sure if I'm on the right track, so if anyone has any ideas as to what I can do I'd really appreciate it :-)
    Last edited: Jan 25, 2006
  2. jcsd
  3. Jan 25, 2006 #2


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    Hi TSN. My area of expertise isn't in ventalation or A/C systems but I've heard of exchange rates. This is simply the number of times all of the air in a given room or building is replaced per unit time. If for example, you have a building with 100 cubic meters of volume, and you have 500 cubic meters of air going into that building per hour, there are 5 air exchanges per hour, so I would guess that the L in your equation is just the air flow per hour. In the example here, that was 500 cubic meters per hour.

    You should be able to see the applicability of this fairly quickly. If we assume the air comes in at some temperature and the building is maintained at another constant temperature, all of that air must be heated or cooled to the temperature of the building before it leaves. The assumption is that the air entering the building will be heated or cooled to the building temperature due to the large surface area inside the building. If it is being blown through the building so fast it doesn't heat or cool to the building temperature, the building temperature will have to change in responce.

    Regarding #2, are you saying that you have an HVAC program that you're trying to verify is accurate? Can you be more specific about what the program inputs are and what is being predicted? The heating or cooling of the air entering a building due to air changes should be straight forward and not need any testing to verify IMO. What would need verifying is the estimations the program might make of heat transfer between the inside of the building and the environment. In that case, some testing along the lines you described might be very practical.
  4. Jan 25, 2006 #3
    Thanks. Try to follow this thought:
    I have four adjacent rooms, pluss one above and below. I note the temp in all of them. Now I heat up my test-room to a certain temp. Once this temp is reached, I note the temp in the adjacent rooms. Then I wait for the whole system to cool down to its initial state. Now I have a time value, but I'm not sure where/how to utilize it for some good. Suggestions?
  5. Jan 25, 2006 #4


    User Avatar

    Staff: Mentor

    Infiltration and ventilation are two different things. Ventilation is air that is brought into the building via the HVAC system while infiltration is air that finds its way into the building through cracks, seams, porous materials, etc. Infiltration rates are often surprisingly high.

    If you are literally going to test a room, it can be done in two ways: First by heating the room a specific number of degrees and via Newton's law of cooling, calculate the energy expended while it cools and thus the rate of heat loss (the intial rate is probably what you want). Second, apply a specific amount of heat to hold a specific temperature in a steady-state condition.

    For the first method, the biggest difficulty is in determining the thermal inertia of the room: the amount of heat lost is calculated form the specific heat capacity of the room. For the steady-state condition, that is irrelevant, since you directly measure heat-flow via the heat you are adding.
  6. Jan 25, 2006 #5
    I think this is where I'm a bit confused. So I let it cool down, let's say this takes 3 hours, then you mention Newton's law of cooling, which I think of as

    {{dT} \over {dt}} = - U \cdot A \cdot \left( {T - T_S } \right)

    Is that right at all? If so I just need to brush up on my diff knowledge in order to solve it...if you don't have any tips? How do I implement those three hours in this equation?
  7. Jan 25, 2006 #6


    User Avatar

    Staff: Mentor

    Ehh, actually, thinking about it a little more, I'm not sure I would do that. There's a simpler way: find the initial rate of change in temperature and multiply by the specific heat of the room. But again, calculating the specific heat of the room is a little iffy.
  8. Jan 26, 2006 #7
    So if my room is V m3. I heat up the room a certain temp, then let it cool, and I measure the temp drop rate to be i.e. 10K in 2 hours. Then I use the following equation:

    \dot Q = C_P \cdot V \cdot \rho \cdot \dot T

    This gives me Watt value, but I'm not excactly sure what it tells me. Perhaps the value for the temp drop should be negative, so that it tells me that it would require so many Watts to cool the room 10K in 2 hours If you could verify this it would be great.

    Also, if there are several items in the room that have different heat capacity, can I then just add these together or how would that work?
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook