Content-type multipart/related;

 boundary="Boundary_(ID_zTXmGzgRN/vat3N59tzoEw)"; type="multipart/alternative"

X-Priority 3

X-MSMail-priority Normal


Cathedralized Attics




HVAC, Roofing & Insulation


Performance & Specification




Tenth Joint Engineering Society Conference (JESC)


Lafayette, Louisiana


January 26, 2006






Myron Katz, Building Science Consultant, Wisznia Associates AIA, New Orleans


Norman M. Witriol, Professor of Physics, Louisiana Tech University, Ruston


Jinson J. Erinjeri, Graduate Research Assistant, Louisiana Tech University, Ruston















Myron Katz, PhD.

302 Walnut Street

New Orleans, La 70118


Energy Consultant and Building Scientist





·     Ventilated attics which contain HVAC system components are scientifically unjustifiable in our climate. 


·     Placing the thermal and air-flow boundary of conditioned space within inches of the roof system creates a cathedralized attic. 


·     This proximity potentially creates an interdependent roofing and insulation system that can degrade or enhance performance and durability.  


·     This situation also drastically affects HVAC loads and performance. 


·     This talk will clarify and/or debunk commonly held beliefs regarding, ASHRAE s recent 152 standard for Distribution System Effectiveness, ACCA s Manual J standard for HVAC sizing, IECC s roofing and attic codes and their interplay in the setting of a cathedralized attic.






This talk is an outgrowth of a research project within the Trenchless Technology Center, Louisiana Tech University, Ruston, Louisiana.  Among the research objectives was the investigation and characterization of residential duct leakage in Louisiana.






We gratefully acknowledge the Louisiana Department of Natural Resources (under a DOE grant) for its sponsorship of this research, DNR Interagency agreement No. 2030-04-03.  During the course of this research project, our research team at Louisiana Tech University received valuable support and guidance from the DNR Energy Section staff, in particular Paula Ridgeway, Harvey Landry, Buddy Justice, Howard Hershberg, and Tangular Williams.


I. Ventilated attics which contain HVAC system components are scientifically unjustifiable in our climate.  Placing the thermal and air-flow boundary of conditioned space within inches of the roof system creates a cathedralized attic. 




A home normally has an attic which is open to the outside.  The insulation is installed between the attic and the home, usually at the bottom of the attic.  The problem is that there are so many holes (associated with electrical wiring, AC ducts, ceiling lamp fixtures, bathroom fans, ceiling registers, attic access portals, etc.) connecting the attic and the home that the home essentially has a large open window to outside via the attic.  We all know that if you keep your windows open when you run the heating/cooling system you are wasting a lot of energy.  This is essentially what is happening in the attic.  We measure this connection in a home energy rating. 







Because it is very difficult to seal the holes connecting the attic and the home, an effective and affordable solution is to change the home to include the attic (like a finished attic), and insulate the underside of the roof, thereby creating a cathedral ceiling in the attic.   Because the attic is now part of the conditioned space, leaks to the attic are internal leaks, namely not to the outside.




Cathedralized Attic






The elimination of air leaks and heat flows to/from outside by simultaneously air sealing and insulating the attic just below the roof is estimated to save over 40% of a Louisiana homeowner’s heating and air conditioning bill.

How does a Cathedralized attic affect building cooling loads?



Supply duct leak (%)

Return duct leak (%)

Total duct leak (%)

Unvented roof

cooling load reduction (%)

Simulation results









Measured results









6” duct disconnect








Please notice the deep connection between the study of cathedralized attics and avoiding duct leakage in homes with HVAC equipment in the attic.  The original proponents of unvented attics (the first name they used) believed that avoiding duct leakage would ensure that this solution would save more energy than the same home with a vented attic.







Our first major report


On Testing HVAC Duct Leakage in Existing Residential Buildings

in North Louisiana,


was published in 2003.  Within that report, we established that average duct leakage at 25 Pa was about 400 cfm and inferred from statistical evidence that average duct leakage in our 55 home sample was just under 30% of rated flow.  [4]




The current project’s goals require that we quantify the dollar savings the average homeowner would receive by completely sealing those duct systems.  This talk presents the first precise fruits of that research.


Energy and Cooling Load Savings Derived by Sealing Duct Systems




Depend upon


1.   Complete energy audits of the homes, i.e., collection of all data needed as input to the energy auditing software.


2.   Review of various ACCA, ASHRAE, and ASTM standards regarding


a.   Duct leakage testing protocols


b.   AC load sizing


c.    House-leakiness testing


d.   Distribution System Efficiency


3.   Deeply understanding what the energy models measured, estimated and calculated what they don’t.




There was so much controversy about duct testing that we created our own method!





Generalized Subtraction Correction Algorithm, published in ASTM's Journal of Testing and Evaluation, November 2004, unified and enhanced duct leakage, house leakiness and zone pressure dependence testing into a fast and robust diagnostic approach for measuring and analyzing the advective flows most important to energy auditing.


2004 ASHRAE standard 152




It wasn’t until 2004, that ASHRAE came out with standard 152.  This not only proposed their specifications for the duct leakage measuring algorithm but it also described a more complete data collection to estimate residential HVAC Distribution System Efficiency.  Defined as




distribution system efficiency the ratio between the energy consumption by the equipment if the distribution system had no losses (gains for cooling) to the outdoors or effect on the equipment or building loads and the energy consumed by the same equipment connected to the distribution system under test.






Subsequent to its release, the Residential Services Network, (RESNET) adopted ASHRAE 152 as THE method for testing ducts and the logic to drive the software for estimating home energy use and cooling and heating loads.






Home energy auditing software followed suit with the introduction of RemRate 12 which calculates all of these result in a manner consistent with the new standard.  We used this software in our work.




Over the last few months our research team has revisited the same homes in our original 55-home sample to collect the extra data needed for the ASHRAE 152 standard and RemRate 12.  Two of these homes had sufficient data for complete analysis at the time of this talk.  In our sample, they are homes numbered 40 and 43.




Both homes are in Ruston Louisiana.  They both have all HVAC and ducts in attic, R-3 insulation on the ducts, ranch-style homes, one story, slab-on-grade, 80 AFUE furnaces in attic, medium color roofs, no radiant barriers in attic, windows with average shading, and gas water heaters with 0.56 EF.  They differ as follows.





Home 40

Home 43

Area of condition space   

Sq ft



Attic Floor Insulation       




Wall Insulation




Window area

Sq Ft



Window type


Single, wood

Double/LowE wood


Effective Leakage Area     

Sq Inches



Supply Leakage       

CFM @ 25 Pa



Return Leakage    

CFM @ 25 Pa



Flow though blower  




Area of supply ducts        

Sq ft



Area of return ducts        

Sq ft



Heating capacity        




Cooling capacity    




AC efficiency      




Water Heater Location  








Percent Duct Leakage













Notice that the Home 43 is much tighter, has much less duct leakage, less duct leakage as fraction of blower flow, more insulation, much better windows and less window area.  However, Home 40 has a higher SEER air conditioner.






Modeling Energy Use for a Cathedralized Attic




        Increase the volume of the home by about 10000 cu ft.


        Change the insulation for the attic from the floor to R-19 vaulted with a surface area = 1.5 times the floor area.


        Reduce the house-leakiness by 50%.


        Set Duct Leakage to zero.


        Relocate entire HVAC and Duct system into conditioned space.






We set the reduction in house-leakiness to 50% because of the following study.

Infiltration with a Cathedralized Attic







Notice by reading the 3rd row of data, that the infiltration rate of a home with a cathedralized attic was less than ½ of that with a vented attic.




If all of the duct system is in the attic, a cathedralized attic can be expected to save about 25-30% of the annual heating and cooling bill.  For most such homes this improvement will pay for itself in less than five years.




Probably much faster … as the next graphs explain.

Save 25-30% of HVAC Energy






The graphed data utilize the input energy auditing data for homes 40 (light blue) and 43 (dark magenta) which are located in Ruston, Louisiana.  However, the climate and energy costs data were set to New Orleans as of January, 2006.  $0.11 / kwh and $1.10 / CCF of natural gas.




What to do




1.   Seal all vents to outside and make all “attic ventilation” inoperable including: attic fans, ridge vents, gable vents, soffit vents; these openings should be repaired on both sides of the roof decking, i.e., the roof perforations and protrusions should be removed and the area re-roofed ― except as noted in the last step.  Carefully seal the area between the rafters at the low end of the roof.  Do not remove, close-up or disable the kitchen exhaust stack, the exhaust vents for water heaters, furnaces or plumbing stacks.

2.   Using light-weight fiberglass or nylon cloth, drape and then staple the material to the studs of the attic gable walls and rafters of the attic ceiling to enclose the cavities between them.

3.   Blow high-density, high-borate-content cellulose insulation into the cavities between rafters, studs of gable walls and the areas between the joists of the attic floor that do not cover conditioned space.

4.   If you have a gas water heater or gas furnace in your attic or home or an attached garage, enclose the equipment in air-tight cabinets connected on the top and bottom to the outside via two 12” ducts and install digital carbon monoxide detectors in those areas (to comply with M1703, the combustion air chapter, of the 2003 International Residential Code).





Drape and then staple onto rafters.






Cathedralized Attic



What it costs




1.   The actual cost of cathedralizing your attic will depend greatly on whether the roof/attic components were recently replaced or have yet to be “fixed.”  If timed with other repairs, the roof work, proper insulation installation and mechanical system integration can lead to major cost savings.

2.   Installation of the cellulose insulation should be done professionally.  Expect the installed cost to be less than $1 per sq foot of roof, wall or floor area.  In estimating the total cost, consider that an attic with 1500 sq ft of attic floor area has less than 2200 sq ft of roof area unless the roof has a very high pitch.

3.   The total cost for an average attic is conservatively estimated at around $3000.






1.    If you have all of your duct system in your attic, this improvement can be expected to save about 25-30% of your annual heating and cooling bill.  For most such homes this improvement will pay for itself in less than five years.

2.    Your next air conditioner could be 40% smaller.

3.    Allows “FAN ON” cooling which avoids uneven cooling while it provides enhanced comfort, better indoor air quality and more durability.

4.    No Sweat!

5.    Safer and more durable home because it keeps the home from “sucking”.

6.    More comfort when using your heating or cooling system; and more comfort when you do not have electricity or choose not to use it. 

7.    Increased resistance to hurricanes and a safer shelter.

8.    Roof leaks will be less likely and when they occur, they will do less damage.

9.    Better protection against mold, wood rotting fungus, termites and fire.

10.           More durable against future construction activities in your attic.

11.           Extra storage and living area.

12.           Allows the use of modern ceiling (light) cans without additional infiltration.

13.           Make your home “part of the solution” rather than “part of the problem”.

14.           More advanced energy options that can save much more energy are easier to add after this improvement.











Your next air conditioner will be smaller.



·      More comfort when using your heating or cooling system; and more comfort when you do not have electricity or choose not to use it.


·      Allows “FAN ON” cooling which avoids uneven cooling, while it provides enhanced comfort, better indoor air quality and more durability.


·      NO SWEAT!




The previous graphs on cooling load savings and energy savings all assumed a 78 F set point.  However




there is more energy to save




When a home in a hot and humid climate cannot reach 78 F and 50% RH, usually associated with distribution losses, over-sizing, and infiltration, the homeowner will tend to respond by setting the thermostat to a lower temperature.  The AC unit can reach a 72 F set-point during a large percentage of the cooling season even if it cannot get there in near peak-load conditions. 




The consequential increase in energy use is likely to be far greater




than ASHRAE 152 predicts (because the standard insists on the use of 78 F only) even though ASHRAE 152 purports to output all SEASONAL COOLING ENERGY INCREASES from distribution problems. 


NO SWEAT!   (continued)


there is more cooling load to save


Aggravating this problem is condensed moisture 


When the home is cooled below dew point, as it would be in this example,


the latent load on the home from diffusion and infiltration is vastly greater


than ACCA Manual J predicts. 


If fiberglass or cellulose is the site of such condensation, the R-values will decline greatly adding to the sensible load. 


None of this is even hinted at in the ASHRAE standard and I think these effects can easily dwarf what the standard does output.




This is the output screen of Elite Software’s RHVAC which conforms to ACCA Manual J version 8, the latest.  Notice that ZERO moisture is associated with wall, ceilings or floors; namely vapor diffusion is ignored.



Text Box: ACCA Manual J doesn’t utilize Diffusion when sizing an AC unit.  





·      Dominant Supply Leaks in Vented Attics cause Houses that Suck!




It's quite bad for a home in the south to "suck", i.e., the not uncommon situation that the normal flow of air from the AC unit's duct leakage to outside is mostly from the supply ducts; the result is that the home is depressurized with respect to outside.  It then "sucks" air in from outside whenever the AC unit is running.  This problem is fixed by either putting ducts into conditioned space or sealing the ducts.  When the problem is not fixed, it causes the home to pull in hot and humid air and greatly adds to the energy costs and puts undue burden upon the building envelope where moisture intrusion can lead to catastrophic building failure or severe indoor air quality problems.  Among the possible results are mold, termites, wood-rotting fungus and, of course, high energy bills.   Not a very uncommon result is mold-related illness.  Cathedralized attics tend to avoid this problem.





Attic Fans … always a bad idea.




Cathedralized attics lose the attic fan.  Attic fans have never been a good idea.  This was confirmed by the Louisiana Home Builders in a study they released in 1983, it was later reconfirmed by a series of studies by scientists and engineers from national professional associations.  Removing attic fans is always a good idea even if no other steps are taken. 


[11, 12]


Attic Fans!    Bad!



·    HVAC Operation Across Closed Doors Causes Infiltration.




Although not as big an energy waster as duct leakage, a home can have large energy losses caused by infiltration literally induced by a very tight HVAC system.  (Ironically, sealing the ducts in an attic can exacerbate this problem.)  When a home is conditioned by a standard central cooling system, it is usually the case that the supply register(s) are not in the same room as the return register(s).  This poses a problem when doors are closed between a room containing a supply register and rooms containing the nearest return register.  What happens is that the room with the supply registers becomes slightly pressurized with respect to outside thereby providing the driving force to push air out of the home at the cracks around the windows.  Similarly in the room with the return register, the pressure becomes slightly lower than outside, this brings in outside air.






The resulting energy loss is very large and, even worse this grossly compromises the cooling system ability to properly dehumidify the home.  This problem is not worse in a home with a cathedralized attic; on the contrary, a cathedralized attic can make a good solution work better. 




Jump Ducts work better in a Cathedralized Attic.




One solution to this pressurized room problem is to provide a “jump duct” that connects the air near the ceiling of the rooms on each side of a doorway; the duct can run in that attic.  This solution removes the pressure problem, but if the jump-duct leaks or must run through a hot or humid attic, which is often the case when the attic is ventilated, energy losses can be significant.  Jump-ducts better in a cathedralized attic.





However, the pressurized room problem can also be solved with a “transfer grill” which does not need an attic for installation; this solution equalizes pressures and provides for the return flow by using a common wall’s stud cavity.




Transfer Grilles can get the job done as well.


Adequate Dehumidification



Adequate dehumidification of a home cooled by the most commonly used cooling systems can be more difficult when the home’s cooling system runs for shorter periods of time.  Although this is a possible consequence of a cathedralized attic, the avoided duct leakage and infiltration provided by a cathedralized attic will probably more than compensate for this “short-cycling” problem.  Even this is not problem if the home has more than one air conditioner or if that unit has two compressor speeds.  However, in the worst case, when extra dehumidification is needed, very cost-effective stand-alone dehumidification equipment costs less than $300 to buy, $200 to install and $300 a year to run.  Even in that scenario, the extra cost of running the stand-alone dehumidifier is more than offset by the additional cooling energy savings thereby provided because the home will be more comfortable at higher indoor temperatures since the air is dryer.




“… while the additional sensible load resulting from duct leakage may be lower in homes with vented attics, the additional latent load is likely to be higher.”


   Issues Related to Venting of Attics and Cathedral Ceilings (ASHRAE 1999)











Adequate Ventilation and Fresh Air




A home with a cathedralized attic is about twice as tight as a home without such a system -- decreasing natural infiltration which for almost all homes is the primary source of fresh air from outside.  For the small number of homes in LA which are unusually tight, this problem can easily be remedied.   (See previous slide!)

Seal the Ducts!

Although a cathedralized attic will drastically lower the energy wasted by having a duct system in a conventional, ventilated attic, almost without exception, ducts need repair work as well.  Average duct leakage in Louisiana has been tested to be 30% of normal flow. This is the primary reason why a cathedralized attic saves so much energy because with this improvement, duct leakage is no longer to outside of the conditioned envelope of the home.  [4]



However, if the ducts do not get special sealing repair work, much of the heating and cooling will be directed away from normal living areas.  A simple and very cost-effective job of literally painting on a thick, paint-like coating, called “mastic”, should be applied to all joints of the duct work and between the duct system and the blower and the duct system and the registers.  The result will be more directly applied cooling and heating to where the people live!  Energy will be saved because comfort will be attained more quickly and via shorter run times; the unit will operate for shorter periods of time. .


·    Increased resistance to hurricanes and a safer shelter.




Flexible Vinyl Soffit Vents Were Prone to Blow Out




Rain Penetrated 10 Feet or More into Attic Spaces










Although this “solves” the problem of hurricane resistance in a vented attic, clearly it is better still to have an attic with no ventilation at all!


·    Roof leaks will be less likely and when they occur, they will do less damage.


 1.   More durable system because moisture vapor flows are better accommodated through the sheathing when moisture penetrates the underlayment.


2.   Fewer roofing components that can be torn off in a major wind event mean less roof damage.


3.   Fewer roofing protrusions mean a simpler roof and one less prone to leak.


4.   Less susceptible to intruding high pressure that can help lift off roofing components.


5.   Insulation acts a secondary underlayment to protect the home from minor roof failure because of the vast moisture storage capacity of cellulose.


6.   The borates in cellulose retard intrusion of termites and wood-rotting fungus at minor roof failures.




·    Better protection against mold, wood rotting fungus, termites and fire.




These are all consequences of the borates in the cellulose.



Why Cellulose?



Cellulose is the insulation of choice because as compared to any other insulation it provides the best combination of cost, ease to install correctly, high R-values, more infiltration barrier, increased moisture holding capacity of the wall and thereby increases its durability to moisture load, contains borate which retards fire, mold, wood-rotting fungus, termites, and roaches, does not impede moisture flows and does not release toxic gases when burned.  Moreover, cellulose is a much less energy-intensive product than fiberglass and is usually made from recycled newspaper.  Cellulose is often installed wetter than wood often becomes after a flood or rain leak, mold doesn’t grow there because of the borates, so it will help protect your wall and roof structure after the future storm.






·    More durable against future construction activities in your attic.




1.               The next workman who enters your attic is not very likely to perforate or remove the insulation because the rafter cavities are not in the way of that job whether alarm, electrician, plumber, AC repair man, etc.


2.               Few workers are willing to deal with what appears to be a roofing component because a water leak is likely to show up soon enough after he leaves to alert the homeowner.


3.               Can’t be inadvertently stepped on, so less likely to be accidentally disturbed.


4.               AC condensate will not fall into it.


·    Extra storage and living area.




1.     Semi-conditioned space can now safely store temperature or humidity sensitive items.


2.     Extra living area that often does not need conditioning at all!  Good for a work out area in summer or playroom during the rest of the year.


3.     Conversion to real living area is many steps closer.


4.     No kneewalls needed!






·    Allows modern recessed lighting without additional infiltration.


Recessed cans are often part of renovations of older homes.  These usually provide leaks and thus infiltration between conditioned rooms and ventilated attics above.  Making them or choosing them to be air-tight is a challenge but once accomplished, the fixture will not allow a compact fluorescent lamp to operate as efficiently.  This is because as the lamp heats up its energy consumption stays the same, but the light output decreases.  If the air flow is unobstructed into a cathedralized attic both problems are solved.






·    Make your home “part of the solution” rather than “part of the problem”.


 What community or global problems does this solve?


1.               Helps to avoid heat islands.

2.               Lowers energy use and thus slows the causes of global warming.

3.               Utilizes recycled materials.

4.               Makes your home more resistant to storms.

5.               Provides better and more inhabitable shelter when there is no external artificial power.






·    Make your home better suited for advanced energy solutions.




What works better in a Cathedralized Attic?

1.              Ground source Heat pumps

2.              Sealed Combustion Heaters

3.              Chilled water distribution systems.

4.              Jump Ducts

5.              Dehumidifiers can double as heaters.

6.              Recessed Cans with Compact Fluorescent Lamps.

7.              Tighter homes can utilized mechanical ventilation and have cleaner air.




·    Avoid Carbon Monoxide and Improve Heater Efficiency




When air conditioning equipment is in an attic, the heating system is usually a part thereof.  In Louisiana, most often, such equipment is a gas-fired furnace that is designed to run in a ventilated attic.  The efficiency of such a furnace is usually 80%, that is, 80% of the heat energy of the natural gas gets into the air stream leaving the furnace on its way to the home via the duct system.  Such equipment must have an adequate supply of oxygen just outside the blower cabinet otherwise carbon monoxide may become a life-threatening byproduct of combustion.

 If the home is too well sealed to outside, natural infiltration is too small and no other steps are taken to provide sufficient oxygen to the home or the furnace, the danger of carbon monoxide poisoning will be that much more likely.  Although very few homes have been built in greater New Orleans tight enough to cause this problem, the threat of death must take precedence.  That is why carbon monoxide detectors are a required part of this improvement.   And that is also why we have specified the code-compliant installation of two 12” flexible ducts from outside connected to an air-tight cabinet enclosing the furnace. [16]  The detector will tell you that extra steps are needed to safely inhabit a tight home with a cathedralized attic and an 80% gas furnace.  In this case, the gas furnace should be rechecked for proper installation.  However, improved ventilation of outside air to the living space can also improve indoor air quality.  The method recommended is to directly connect a small duct from outside to the return plenum of the blower.  Although the initial recommended cathedralized attic specifications are the least capital intensive, they do not generate the lowest operating costs or the greatest safety. 

The next time the home’s air conditioning equipment or heating system needs replacement, the furnace should be replaced with a sealed-combustion furnace or heat pump.  A sealed-combustion furnace is 90% efficient and should only cost a few hundred dollars more but by design, it pipes-in its own oxygen from outside as part of the installation.  This furnace does not need an extra cabinet or ducts to outside.  Although this system costs a little more than an 80% furnace it may take more than a few years to recover the cost since New Orleans has such a short heating season. 


Alternatively, an (air source) heat pump will actually deliver heat to a home for less money than even a 100% gas furnace.  It does operate at slightly lower cooling efficiencies but this penalty pails in comparison to the cooling and energy savings of a cathedralized attic.   This product is much more common in Louisiana than 90% furnaces and will, therefore, be easier to find and purchase at a competitive price.




The most efficient heating system is also the most efficient cooling system; a ground-source heat pump.  Such a system has no combustion at all and therefore is a terrific appliance in a cathedralized attic.  Although such a system costs about 30 to 50% more to install it can be expected to save more than that in annual energy use so it is a cost-effective alternative to standard cooling equipment.




Hydronic or Chilled-Water Distribution is Easiest in a Cathedralized Attic.




Chilled-water cooling and heating distribution is also called “hydronic”.  Although this is a mature technology and is used in 1/3 of all commercial space, it is rarely employed in homes.



The reason is that commercial hydronic equipment is not manufactured to be easy installed into the relatively thin walls or ceiling cavities found in homes.



However, with a cathedralized attic, the relatively bulky equipment can easily be installed on the floor of such an attic and will operate as well as in a commercial installation and should cause a commercial technician no more of a challenge than he expects to find in an office building. 

With this technology it is possible to provide each room with its only water-to-air heat-exchanger which can directly heat or cool only that room; it can have its own thermostat as well.  Such a system also provides for dehumidification for 100% of the cooling season, not just when the thermostat calls for cooling.  Such a system is vastly superior to ducts and is much more appropriate in a cathedralized attic than a ventilated one.  Installing a chilled-water HVAC distribution system only requires that the tradesman have basic plumbing, carpentry and electrical skills.  The equipment is more expensive than ducts but the energy saving will generate savings sufficient to pay for this improvement in less than five years.


(The three pictures in this section were taken at the home of Myron Katz.  The first two pictures show two installed fan-coils the first picture was taken from within the dining room looking into the living room; the second picture is a view of one of these same fan-coils taken from the attic looking down.  The third picture shows a commercial fan-coil sitting on the floor of the cathedralized attic; it is not installed.)




Alternatively, a more effective insulation system employing a radiant barrier can be placed between the rafters.


The system with the radiant barrier has twice the installed cost but may not save twice as much in annual energy use; however, the avoided installed cost of future air conditioning equipment and the improved roof sheathing durability make this choice cost-competitive.  It also increases the home’s comfort and durability as well. 


Using TechshieldTM, radiant barriers or barrel tile shingles can be installed much more cheaply while reroofing than during an in-the-attic insulation installation; however, some asphalt shingle manufacturers (but not Elk) will not warranty such a roof / radiant barrier system.




A perforated radiant barrier is placed shiny-side-down right below the roof decking.  An air space is provided below it and above the insulation.

Part II. Cathedralized Attics

Roof Durability   vs.

                 Energy Efficiency




“What is the impact of tile color on energy use in homes with vented roofs?




        The cooling load due to heat gain from the (vented) attic was 76% lower for the white tile roof compared to a black shingle roof.


        But for typical houses with shingle roofs, heat gain from the attic is about 10% of the total cooling load.


        yields a 7.6% reduction in total cooling load    [3]



“Can unvented roofs be used with asphalt shingles?




“If asphalt shingles can be used on vented roofs in the hot-dry climate, where shingle temperatures can approach 200 F, then asphalt shingles can be used on unvented roofs in the hot-humid climate where the shingle temperature will be less than 170 F. 




“The Asphalt Roofing Manufacturers Association officially does not approve the use of shingles over unvented roofs, however, there has been recent movement to address the lack of technical information to support that view.  As of yet, unpublished information tends to suggest that the long-term durability of better grade shingles (>25 yr warranty) will be slightly affected or unaffected when used in unvented roof applications. Where a manufacturer can substantiate that a 5 F to 10 F maximum increase in shingle temperature will materially affect the shingle durability, then, in that case, it would not be prudent to use asphalt shingles over unvented roof assemblies.    [3]



What is the impact of tile roofs on attic and roof temperatures?




“In the same climate, tile roofs of any available color will cause the roof and attic to operate at lower temperatures than the same roof with asphalt shingle of any available color. Simulations and early measurements predicted a maximum increase in roof sheathing temperature of 17 F for unvented tile roofs over vented tile roofs. Measurements through an entire summer season in Las Vegas showed a maximum increase of 20 F. The sheathing temperature never exceeded 155 F, and typically varied between 70 F at night and 130 F during the day.” [3]



TenWolde developed and later verified (TenWolde 1988 and 1997, respectively) a predictive roof temperature model. In the 1997 report, he described a predictive roof temperature model especially for sloped wood-based roof systems. This model shows that the surface temperature of plywood roof sheathing is dominated by solar gain and the heat exchange between the surface and ambient air, not by attic ventilation.  




Diurnal (daily cyclic) temperature variation and hourly sheathing temperature histories are also influenced by the radiant energy absorptivity of the roofing surface, roof pitch, and, to a lesser extent, insulation and attic ventilation. The TenWolde model predicts that wet plywood sheathing dries quickly under warm summer conditions, even if ventilation is minimal. For example, if plywood is installed at 60% moisture content, the moisture content is roughly 15% after 1 week and falls to 8% in roughly 2 weeks. The model also indicates that the absorptivity of solar (radiant) energy by the roofing material has the greatest effect on increasing or reducing the average temperature of the plywood roof sheathing.  If the absorptivity of the roofing material is 0.92, the model predicts the maximum hourly temperature for the roof sheathing plywood as 60°C and the maximum predicted exterior roof membrane temperature as 66°C. If the absorptivity is changed to 0.2, supposedly representing a metal roof system, both the maximum predicted sheathing temperature and maximum predicted membrane temperature drop to 35°C. Roof pitch has only a moderate influence on reducing the exterior surface temperature and the average temperature of the plywood. The model also predicts that the presence of insulation installed directly on the underside of the sheathing has virtually no influence on sheathing temperature on the top surface, but raises the average sheathing temperature relative to that of the top surface.




When the ventilation rate in uninsulated systems is increased from 8 to about 21 air changes per hour, almost no decrease of the top surface sheathing temperature or the average sheathing temperature is predicted.”    [17]









"                                                          SECTION R806


R806.1 Ventilation required. Enclosed attics and enclosed rafter spaces formed where ceilings are applied directly to the underside of roof rafters shall have cross ventilation for each separate space by ventilating openings protected against the entrance of rain or snow. Ventilating openings shall be provided with corrosion-resistant wire mesh, with 1/8 inch (3.2 mm) minimum to 1/4 inch (6.4 mm) maximum openings.




“But we NEVER did it that way!”

Most building professionals will complain that cathedralized attic use is incorrect and detrimental to the durability or comfort of the home.  These people are relying on “I know how it has always been built” vs. applying modern building science solutions.  This specification is a code-compliant choice used for more than a decade in various regions of the country.  In fact, you will find cathedralized attics at the Louisiana House exhibit currently under construction on LSU’s campus.

 http//  [18]




“I hear that the IRC and the IECC have already been altered to explicitly allow unvented and cathedralized attics.  Am awaiting documentation and will share.” 

Thu, 19 Jan 2006


Claudette H. Reichel, Ed.D.,

Professor (Extension Housing Specialist)  [19]

Addendum to Florida Building Code,

effective date October 1, 2005



"R4409. Conditioned attic assemblies Unvented conditioned attic assemblies (spaces between the ceiling joists of the top story and the roof rafters) are permitted under the following conditions


1. No interior vapor retarders are installed on the ceiling side (attic floor) of the unvented attic assembly.


2. An air-impermeable insulation is applied in direct contact to the underside/ interior of the structural roof deck. “Air-impermeable” shall be defined by ASTM E 283.


3. Shingles shall be installed as shown


a. For asphalt roofing shingles A 1-perm


  (57.4 mg/s · m2·Pa) or less vapor retarder


  (determined using Procedure B of ASTM E 96)


  is placed to the exterior of the structural roof deck;


  i.e. just above the roof structural sheathing.


b. For wood shingles and shakes a minimum continuous ¼ inch (6 mm) vented air space separates the shingles/shakes and the roofing felt placed over the structural sheathing."








Permeance of some roofing components.







Code Objections


“I consider that the Florida Building Code, R4409., requirement for "Unvented conditioned attic assemblies" to have an air impermeable insulation system placed directly below the roof sheathing to be an inappropriate requirement.  This construction assembly will entrap moisture between the roofing system and the insulation, allowing it to condense on the underside of the roofing after dark, leading to the premature deterioration of the sheathing.  I recommend providing a vented air space between the insulation and the sheathing to allow moisture laden air to escape.”

Sheldon J. Leavitt, AIA, P.E.;   

Architect and Professional Engineer

Member ASTM Committee E 6 Performance of Building Construction (1989 to Present)



‘“Air-impermeable” shall be defined by ASTM E 283.’



‘This “Standard Test Method for Determining Rate of Air Leakage though Exterior Windows, Curtain Walls, and Doors Under Specified Pressure Differences Across the Specimen” does not define “air-impermeable” as stated in subparagraph 2 of “R4409. Conditioned attic assemblies.”  I consider the references to be inappropriate, as used.’  [22]

Sheldon J. Leavitt, Jan 22, 2006



The Opinion of a Very Experienced New Orleans Roofer.




I have a gut feeling that TYVEK, in tandem with something else (?) may turn out to be the answer of all answers for the best of both worlds when it comes to keeping water and moisture from attacking objects stored beneath it; even when water ponds on the surface of the TYVEK. My one experience with this product was an uptown (New Orleans home) where we stored valuable books in the open yard under a make-shift TYVEK tarp -like shelter. In time, the tarp began to hold water and actually pond water, up to 6 ", and after several months of our weather, the books were not damaged in any way by dampness or moisture of any (?). Further, I have never seen Tyvek deteriorate due to UV exposure or shred from time left to open weather (like most other house wraps).


 ..  I would feel comfortable using it on my decking anytime.” [23]


Dick Schmidt, President, Snow White Roofers,

New Orleans, La, January 24, 2006


Myron’s note: TYVEK has a permeance of 50 perms.

Part III. Cathedralized Attics Improve HVAC loads and Performance.




What is the impact of tile color on roof temperature in unvented roofs?   [3]


“To our knowledge, published data does not exist to directly address the question of the impact of tile color on roof temperature, and more important, annual space conditioning energy consumption. However, gathering available information, the following analysis is made.


“Measured data reported from the Florida Solar Energy Center (FSEC) shows that white concrete tile from Monier has a solar reflectance of 0.75 while the same concrete tile in dark red has a solar reflectance of 0.20.   An FSEC study, over the summer of 1997, used a series of six foot wide attics over a cooled common space to compare attic temperatures between different roof constructions, including the aforementioned white and red tile. Mid-attic air temperature was 8 F warmer on average for the red tile roof compared to white tile roof (88 vs. 80), and 25 F warmer at maximum (122 vs. 96). Compared to a black shingle roof, the measured temperature difference across R-19 ceiling insulation was 76% lower for the white tile roof and 23% lower for the red tile roof.  Thus, the cooling load due to heat gain from the attic was 53% lower for the white tile roof compared to the red tile roof.


“In the FSEC study, the red tile roof performed about the same as a black shingle roof with a radiant barrier.  For typical houses with shingle roofs, heat gain from the attic is about 10% of the total cooling load. Heat gain from the attic will contribute less than 10% of the total cooling load for shingle roofs with radiant barrier and tile roofs. Taking the 76% heat gain reduction for white tile multiplied by 10% yields a 7.6% reduction in total cooling load. Taking the 23% heat gain reduction for red tile multiplied by 10% yields a 2.3% reduction in total cooling load. Thus, the white tile roof will reduce the total cooling load about 5% more than the red tile roof. However, the heating season must also be considered.


“During the heating season, a red tile roof will perform better than a white tile roof because it allows more heat gain to the attic, increasing attic temperature and reducing heat loss from the house. This benefit/disbenefit relationship is illustrated in detailed simulations performed using the FSEC 3.0 computer program.  Annual cooling and heating load simulations for Las Vegas showed that, compared to a roof with black shingles, a white tile roof decreased the cooling load by 9% and increased the heating load by 3%, for an annual space conditioning load reduction of 2%. After applying the local cost of electricity for cooling, and the cost of natural gas for heating, the annual space conditioning cost reduction for white tile compared to black shingles was 4%. Thus, the annual space conditioning cost reduction for the white tile roof was less than half the cooling cost reduction, due to the white tile disbenefit in the heating season. While a red tile roof was not simulated for the published study, it would have had less cooling load reduction, and less heating load increase than the white tile roof. In summary, one half of the 5% difference in cooling load reduction between white tile and red tile from the FSEC study would yield a 2.5% difference in annual load reduction due to tile color. (white tile vs red tile (.53*.10*.5)=.027).


“In Las Vegas and Tucson, the BSC Building America Program is constructing homes with red or brown tile roofs and sealed cathedralized attics. The sealed cathedralized attic puts the entire space conditioning air distribution system inside the air and thermal boundary of the building. Because this construction eliminates duct heat gain/loss and duct leakage to outdoors, saving about 20% of the total space conditioning load, the 3% or less difference due to tile color is small.

“The vented roof design was insulated to R-30 while the unvented roof design was insulated to R-22. Hence, the results are not exactly comparable and the unvented roof should have more cooling load reduction than shown.


“What is the impact of unvented roof design on building cooling loads?

“The answer to this question lies primarily with the amount of duct leakage to outdoors in the vented roof design. Based on simulations, if 10% of the air handler flow was leaking on the return side, and 5% was leaking on the supply side, the unvented design would save about 8% on cooling load. If duct leakage was 15% on the return and 10% on the supply, the unvented design would save about 14% on cooling load. Based on measurements taken during a side-by-side test3 in Las Vegas, the unvented design reduced cooling load as follows

Supply duct leak


Return duct leak


Total duct leak (%)

Unvented roof

cooling load reduction (%)

Simulation results











Measured results











6" duct disconnected





Interpolating between total duct leakage and cooling load reduction, the measured and predicted values agree reasonably well.”  [3]

Problems with ASHRAE 152




1.     When the attic pressure is more than 10 Pa different from outside, while the home is depressurized to 25 Pa, the measurement of duct leakage is not authorized.  The authors recommend that you open a door to outside.  But what if there isn't a door to open?  Do you want the rater to cut a hole in the roof?



2.     Calculating REAL duct leakage from Measured depends upon a difficult to impossibly accurate measurement of supply duct pressure with respect to attic pressure.  Moreover, the standard asks for the wrong measurement it requests the difference between inside the supply plenum and outside, but what is really needed is the difference between inside the supply plenum and the attic.




3.     The ASHRAE 152 standard outputs two measures of the distribution system's ability to cool/heat a home.  (See page 2.)




delivery effectiveness the ratio of the thermal energy transferred to or from the conditioned space to the thermal energy transferred at the equipment distribution system heat exchanger. Energy delivered to or from the conditioned space includes distribution system losses to the conditioned space.




distribution system efficiency the ratio between the energy consumption by the equipment if the distribution system had no losses (gains for cooling) to the outdoors or effect on the equipment or building loads and the energy consumed by the same equipment connected to the distribution system under test.




For quite a while I was dumbfounded about how to distinguish them…  However, I now think I'm on the right track.  While the first measures what percentage of the cooling already produced by the evaporator coil gets to the home.  The second, distribution system efficiency, measures the ratio of energy used by the cooling system when there are no losses to when there are losses.  Significantly, distribution system efficiency is degraded by increases in cooling load or decreases in equipment efficiency -- neither of these is considered in delivery effectiveness.




In case this sounds too technical, here is an example  Suppose duct leakage is dominated by supply duct leakage.  This will make the home "suck" as Joe Lstiburek likes to say.  I think that a home that sucks has more cooling load than the same home that doesn't.  Delivery Effectiveness is degraded by the duct losses only, but Distribution System Efficiency is also degraded by increases in cooling load.




4.     Ok, so now you want to calculate/estimate the increase in cooling load on the home from a dominating supply leak.  What data will you most certainly need?  You need to know the leakiness of the home and you need to know the pressure-coupling ratio between the home and the attic.  (This is the ratio between pressures in the attic and home when the home is depressurized with respect to outside.)  Guess what, ASHRAE 152 doesn't even require as input the Leakiness of the Home.  And, although it requires measurement of the pressure-coupling ratio of the home and attic, that datum is apparently not used in any calculation.  It is only used to forbid the use of the standard, below pressure-coupling ratio’s of 60%.




5.     The standard does not properly consider the real R-value of duct insulation in a hot and humid climate.  The best energy raters in our climate know that metal ducts with loosely attached, fiberglass insulation will tend to become soaked.  So what R-value do you want to assign such systems?  Integrated into this problem is duct leakage... especially if the leak is near such insulation.  In such a case, even more moisture is likely to accumulate.




6.   See NO SWEAT! Above.




[1] Moisture Control in Buildings, Heinz Trechsel, ed. ASTM, Philadelphia, PA, 1994. ISBN 0-8031-2051-6.


[2] Builder’s Guide Hot-Humid Climates, Joseph Lstiburek, Energy Efficient Building Association, 2000.


[3] Measurement of Attic Temperatures and Cooling Energy Use In Vented and Sealed Attics In Las Vegas, Nevada, Armin F. Rudd, Joseph W. Lstiburek and Neil A. Moyer, EEBA Excellence, The Journal of the Energy Efficient Building Association, Minneapolis, MN, Spring 1997.  http//


[4]  On Testing HVAC Duct Leakage in Existing Residential Buildings in North Louisiana, Norman M. Witriol, Jinson J. Erinjeri, Myron Katz, Robert McKim and Raja Nassar, Trenchless Technology Center, Louisiana Tech University, Ruston, Louisiana, 2003.


[5] Generalized Subtraction Correction Algorithm, Myron Katz, Norman M. Witriol, and Jinson J. Erinjeri, ASTM, Journal of Testing and Evaluation, November 2004, Vol. 32, No. 6.


[6]  ANSI/ASHRAE Standard 152-2004, Method of Test for Determining the Design and Seasonal Efficiencies of Residential Thermal Distribution Systems, American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc. ISSN 1041-2336.



[7] Personal Communication, Steve Baden, Executive Director, Residential Energy Services Network, Dec 1, 2005, to take effect, July 1st 2006.


[8] Personal Communication, Dave Roberts, Senior Software Developer, Architectural Energy Corp, Dec 2, 2006.


[9] Celbar Insulation Systems, http// 


[10] Report Summary output, RHVAC, Elite Software,


[11] Report of Louisiana Home Builder’s Assn on Energy Flows, 1983.


[12] Venting Of Attics Cathedral Ceilings, William B. Rose, Anton TenWolde, ASHRAE Journal, October 2002.  http//


[13] Issues Related to Venting of Attics and Cathedral Ceilings, Anton TenWolde, William B. Rose, CH-99-11-4, http//


[14] Building for Extreme Climates, Joseph Lstiburek, Talks given on December 15, 2005 hosted by the LSU Ag Center.


[15] Certified Lighting Efficiency Professional, Course Materials, Association of Energy Engineers, 2003.


[16] 2003 International Residential Code of One- and Two-Family Dwellings, International Code Council, 2003.  ISBN 1-892395-58-4.




][17] Roof Temperature Histories in Matched Attics in Mississippi and Wisconsin, Jerrold E. Winandy, H. Michael Barnes, and Cherilyn A. Hatfield, United States Department of Agriculture Forest Service Forest Products Laboratory Research Paper FPL-RP-589. http//


[18] Visits to Louisiana House by this author, http//


[19] Personal Communication, Claudette Reichel, Professor Extension Housing Specialist,, .


[20] Personal Communication, Dennis Stroer, Past-President National Energy Raters Association,, January, 2006.


[21] Personal Communication, Sheldon Leavitt, Architect and Professional Engineer, Member – ASTM Committee E-6 Performance of Building Construction. Jan 20, 2006.


[22] Personal Communication, Sheldon Leavitt, Architect and Professional Engineer, Member – ASTM Committee E-6 Performance of Building Construction. Jan 22, 2006.


[23] Personal Communication, Dick Schmidt, President, Snow White Roofers, New Orleans. Jan 24, 2006.