[BLDG-SIM] Increased R value Credit for Concrete and Themal Massing

stvgates at pacbell.net stvgates at pacbell.net
Wed Dec 3 13:01:43 PST 2003


My area of expertise is simulation of mechanical systems, and I do not claim to be an expert in heavy mass walls.  However I'll stick my neck out and add my thoughts to this discussion, and invite others to respond.  

I agree with Curt Petersen.  But, in addition to being frequency dependent, my understanding is that heavy-mass wall performance is also dependent on the average daily temperature differential across the wall, the placement of mass vs. insulation, and interior heat gains.  

1)  Mass vs. Resistance (vs. daily outdoor temperature swing)

Assume a lightweight interior space, having no internal loads, is to be maintained at 70F.  If the average daily outdoor temperature is also 70F, and the daily outdoor temperature is swinging +/- 15F about the average, then a heavy construction will maintain the interior temperature very close to 70F.  In this situation, the heavy mass wall will perform better than a resistance wall.

If instead the average daily temperature is 30F, with the same daily swing, then the interior temperature will be close to 30F.  This is also true of a resistance wall.  However, the daily (not instantaneous) amount of heat required to maintain the interior temperature at 70F is now a function of the resistance, not the impedance, and the low-resistance heavy-mass wall will require considerably more heat to maintain 70F.

In other words, for a heavy mass wall to get credit for impedance, the outdoor temperature swing must overlap the desired interior temperature.  The closer the average outdoor temperature to the indoor, the greater the effect of the impedance.  This has significant ramifications for different climates.  For moderate climates or moderate seasons where the outdoor temperature swing overlaps the desired interior temperature for the majority of hours, heavy mass walls can work well.  For more extreme climates or seasons, it's hard to see how a heavy mass wall can be of significant benefit.  

I live in Sacramento, Calif where the average lo/hi summer temperatures are 60F and 95F.  Here, a mass wall can work well.  However, in winter, the average temperatures are around 40F/60F, and I wouldn't be willing to trade a well-insulated house for one built out of uninsulated concrete!  

2) Placement of mass vs. insulation (vs. solar gains)

The above is a little too simplistic because it ignores the placement of mass vs. insulation, and solar gains.  Assume the sun is shining strongly on a wall.  If the insulation is to the outside of the mass, then the outer wall surface temperature (sol-air temperature) will be hotter than if the mass is to the outside.  This is because the mass readily conducts heat inward, whereas insulation doesn't.  Since re-radiation is proportional to the fourth power of absolute temperature, placing the insulation to the outside will cause the wall to instantaneously reject a greater portion of the solar gain.  Placing the mass to the outside allows the wall to capture more of the solar gain; part re-readiates at night, but part conducts into the space.  So which is better?  It depends on whether you are more concerned about winter or summer performance, and how sunny those seasons are.

I recently participated in a study of a refrigerated warehouse that demonstrated this effect.  Holding all factors constant except the placement of mass vs. insulation, mass to the outside of the wall increased annual cooling loads because of the increased capture of solar gains.  (And by the way, this effect was captured in DOE-2).

3)  Effect of interior loads

High interiors loads and/or solar gains thru windows, as well as when they occur bias all of the above.  


Conclusion?  For residential buildings, I believe that a high-resistance shell with a moderate amount of interior mass is the most cost-effective approach for most climates.  While overly simple, a key concept is the "time constant" of a house, which is the product of resistance and capacitance.  The interior temperature decays to the exterior temperature as a function of the time constant (  exp(1/RC) ).  The greater the time constant, the slower the decay rate.  In general it is cheaper to achieve a given time constant by adding insulation instead of thermal mass.

I just built a two-story house and am currently testing this theory.  The interior mass consists of a concrete slab with about 60% tile on the first floor, and 5/8" sheetrock interior walls.  The exterior walls are 2x6 studs with damp-spray rock-wool insulation (R22 in a 5-1/2" cavity, with no voids).  The windows are fiberglass, low-e2, with overhangs on the east, south, and west exposures; most windows facing north and south (~14% of floor area).  The attic has blown rock-wool.  The roof is metal over 2" fiberglass insulation, with a radiant barrier on the underside.  I opted for a metal roof instead of tile because we live in earthquake country, and I wanted the roof to be as light as possible, while still durable.  The kitchen appliances are all electric, as the house is well-sealed and I didn't want to worry about the NOx and CO produced by a gas stove.  All bathroom exhaust fans are on timers to ensure moisture removal after bathing.

We moved in late last summer.  During September, the outdoor temperature range was typically 60F to 95F.  The interior temperature stayed in the low-70's without any air conditioning, and swung about 2-3F.  We ran the whole house fan a couple of nights just to see how cool we could get it, but didn't need to.  I don't know about winter performance yet, but so far we have been heating the entire house (3500 sq.ft.) using two 20,000 Btuh, 80% efficient, thermostatically-controlled gas fireplaces.  Outdoor temperatures have been as low as 35F, but we have not yet needed the conventional forced-air gas furnaces to maintain 70F inside, and most of the time the fireplaces are off.  

This house hardly fits the classic definition of a "passive solar" house, but it performs like one because of its high time constant, achieved primarily through insulation rather than mass.  While the tile floors were great during the summer, I am waiting until March to assess the comfort of these floors with colder ground temperatures (the slab is not insulated).  If anyone wants an update later on, send me your address.

Comments?

----- Original Message ----- 
  From: Curtis Pedersen 
  To: BLDG-SIM at gard.com 
  Cc: BLDG-SIM at gard.com 
  Sent: Tuesday, December 02, 2003 11:21 AM
  Subject: [BLDG-SIM] Increased R value Credit for Concrete and Themal Massing


  Dave:

  This concept falls into the "cheater" category. What happens is that the proponents of heavy construction elements show that the steady periodic behavior of a heavy structure, at some well-selected frequency, is better than the behavior at steady state. The electrical analogy would be to replace resistance with capacitive impedance. Of course it depends on the frequency, and it is not resistance, so the steady state (zero frequency) result will be different. 

  A reasonable building simulation tool properly accounts for both the capacitance and the resistance of a building element. The "effective R" is bogus. 

  Curtis Pedersen
  Professor Emeritus of Mechanical Engineering
  University of Illinois



  On Tuesday, December 2, 2003, at 11:26 AM, David Stewart wrote:


    Dear Folks
     
    Has anyone explored the increased R value credits for thermal massing for example Dow T-Mass etc. and can one simply increase the R value to the 'effective R value while changing the mass in DOE.  I would like some more background on this credit with 3rd party validation
     
    Thanks
     
    Dave Stewart
     
    David C. Stewart & Associates Inc.
    16 Shawinigan Road
    Dartmouth, NS B2W 3A3
     
    Website: http://dcsa.ca
     
    Tel: 902 462 8111
    Fax 902 435 6646
    <David C Stewart MS P. Eng..vcf>

===========================
You received this e-mail because you are subscribed 
to the BLDG-SIM at GARD.COM mailing list.  To unsubscribe 
from this mailing list send a blank message to 
BLDG-SIM-UNSUBSCRIBE at GARD.COM
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.onebuilding.org/pipermail/bldg-sim-onebuilding.org/attachments/20031203/d08d4fab/attachment-0002.htm>


More information about the Bldg-sim mailing list