Hybrid Car – More Fun with Less Gas

Air infiltration heat losses

register ::  Login Password  :: Lost Password?
Posted by Ralph Doncaster on December 22, 2007, 1:34 am
I read a post from October asking about this.  My recent emails to
people involved in the Canadian Energuide program addresses this

Here's another report I found on the significance of air infiltration
(albeit for high-rise buildings, but still relevant).

If I use the low number of 12W/m^2 mentioned in the report @-18C, that
works out to 284 BTU/F heat loss due to air infiltration.  My previous
calculation of 486BTU/F falls nicely in the middle of the 12-25W/m^2
numbers in the report.

Over the past week I have been averaging ~155kWh/day and only able to
keep the house at 15-17C.  With the outside temperature around -5C and
factoring ~300BTU/F (weighted avg R20 envelope * 6000sqft) for
resistive heat loss I'm left with at least another 300BTU/F loss to
air infiltration.   Adding 600W of heat from occupants (2 adults, 4
children), 500W for solar gains (almost 20m^2 of south-facing glazing)
and 100W of heat from use of the propane stove I get another 100BTU/F
of heat loss (which I would attribute to air infiltration).

These numbers point to a natural air infiltration for my house of
~0.3ACH, or about 25% of the  50PA blower door test of 1.28ACH(I've
done some air sealing since then, and estimate I would test @1.2ACH
now).  This leads me to suspect that my initial model of 0.4ACH (based
on Tim Weber's paper) was too high.  With Hot2000 modelling 0.167ACH
of natural air infiltration for 1.5@50, it ends up under-estimating
the air infiltration heat losses by more than half.

Posted by nicksanspam on December 22, 2007, 12:53 pm

12 W/m^2 would be 12x3.41/10.76 = 3.8 Btu PER HOUR PER SQUARE FOOT, no?
Where's the original temperature difference in degrees C?

... 1 W is 3.41 Btu/h, so 1W/m^2C is 3.41/10.76ft^2/m^2/1.8F/C
= 0.176 Btu/h-F-ft^2, no? And 1 Btu/h-F-ft^2 is 5.68 W/m^2C.
OTOH, US R5.68 is metric R1.

With no h?

Sounds like you have colder weather and less sun. Where I live near Phila,
1 ft^2 of R2 south glazing with 80% solar transmsission would collect about
0.8x1000 = 800 Btu on an average January day and lose 24h(65F-30F)1ft^2/R2
= 420, for a net gain of 380 Btu, ie 113 Wh, if it's a window, or about
800-6h(65F-30F)1ft^2/R2 = 695 Btu net, if it's a low-mass sunspace wall vs
a window. So 20 m^2 would collect about 113x20x10.76 = 24.3 kWh/day, ie
about 1 kW over 24 hours, if it's a window, or 2 kW, if it's a low-mass
sunspace wall.


Posted by Jeff on December 22, 2007, 1:50 pm
 nicksanspam@ece.villanova.edu wrote:

   You missed the -18C. It's in the article also.

   BTW, I've downloaded but not installed the simulation software
mentioned (Hot2000). Any experience with this Nick? It's new to me.


Posted by nicksanspam on December 22, 2007, 2:44 pm
Then why the F, and where's the h, in "284 Btu/F"?

No. I write simple BASIC simulations that use hourly TMY2 or Energy Plus
weather data, with lots of flexibility and no hidden assumptions nor
mysterious bugs (after I fix them :-)


Posted by Jeff on December 22, 2007, 5:38 pm
 nicksanspam@ece.villanova.edu wrote:

   I had assumed that he  did the conversion for the 6000 ft^2 and at
his temperature, but then that opens up other questions!

  I am trying to wrap my mind around air exchange heat loss. In theory I
can see the sensible heat per cubic foot. But I don't understand what
affects the air exchanges a house makes.  Does a temperature or pressure
gradiant drive air exchanges or is this a relatively fixed number? It's
all more than I have a clue about. It's a little hard for me to think of
my house exchanging half it's air or so through my painted over windows
and the cracks in more door. That and air exchange means air is coming
and going. Is it coming from some place and going elsewhere? It doesn't
seem reasonable that it is leaking in and out at the same place.

   Lets say we have a 10,000 ft^3 house. If it is exchanging half that
per hour that is 83 CFM. What total size air hole would that be going


This Thread
Bookmark this thread:
  • Subject
  • Author
  • Date
please rate this thread