Letters: January/February 2006
Average New York Energy Use?
The article by Mr. Henry Gifford in the Sept/Oct ’05 issue of Home Energy, “Third Street,” (p. 24), shows the “Average NYC Apartment” heating energy use as 24 Btu/ft2/HDD. With 5,000 HDD, that means heating energy of 120,000 Btu/ft2/yr. He also claims the “Average NYC Apartment” uses 103,262 Btu/ft2/yr for hot water. Then he proceeds to make comparisons with the buildings described in his article.
The following data come from “What Type of Heating Uses Least Energy?” by Segeler, Trunk, and De Pinto, in the August 1973 Heating/Piping/Air Conditioning magazine. These actual measured data are from existing New York City apartment buildings that were built and being operated before anyone ever heard of an energy code, before the 1973 Arab oil embargo, before energy conservation was well known, and when the price of oil was pennies per gallon. These buildings are likely uninsulated with singleglazed and drafty windows.
For 18 New York City Housing Authority buildings with 16,934 dwelling units using oil, the average source energy for heating was 82,000 Btu/ft2/yr, and for 12 buildings with 3,616 dwelling units using district steam it was 76,000 Btu/ft2/yr.The energy was assumed to be from #6 heating oil with 145,000 Btu per gallon and steam was assumed 67.1% efficient. Thus, the steam-heated buildings used 51,000 Btu/ft2/yr at the site, or 10.2 Btu/ft/HDD. For 16 additional privately owned New York City apartment buildings with 27,500,000 ft2 total, the source energy consumption for heating ranged from 35,000 to 88,000 Btu/ft2/yr, or 7.0 to 17.6 Btu/ft2/HDD.Thus, for 46 old New York City apartment buildings with tens of thousands of dwelling units, not even one of those buildings used anywhere near the amount of heating energy/ft2 that Mr. Gifford claims is for an “Average NYC Apartment” today.
For hot water, the 18 New York City Housing Authority buildings using oil averaged 43,000 Btu/ft2/yr with a range from 36,000 to 52,000 Btu/ft2/yr.The steam buildings averaged 52,000 Btu/ft2/yr source energy, with a range from 42,000 to 60,000 Btu/ft2/yr. The privately owned buildings used between 22,000 and 49,000 Btu/ft2/yr.Thus, these old New York City apartment buildings averaged less than half the hot water energy consumption/ft2, and as little as 21%, compared with what Mr. Gifford assumed was an “Average NYC Apartment” today.
For 27 more pre-1973 New York City electrically heated buildings, which were likely insulated, the average site energy use for heating was 27,000 Btu/ft2/yr, or 5.4 Btu/ft2/HDD. This is less than the 6.58 Btu/ft2/HDD for the new 2001 apartment building at 535 East 5th Street.
Author Henry Gifford responds:
Thank you for shedding light on the question of how much energy New York City apartment houses use, which, I am sad to say, increases the number of studies I have heard of from two to three. When I wrote about Third Street I knew of the study described in the article, and one other where Peter Judd compiled fuel use by apartment, not per ft2, so direct comparison of results is impossible. Judd's work can perhaps shed light on the question of why one study would find fuel use per ft2 to be double what another study finds.
Judd’s “Overheated City” is posted on my Web site at HenryGifford.com, and describes “the 600% spread” between the most efficient and the least efficient buildings within each of five building age categories, while the differences between categories are minor. Judd says, “It is not a matter of the age of the building and its level of insulation or even equipment. Old buildings can be operated at least as efficiently as modern buildings. No one type of structure or equipment is necessarily any more efficient in actual use than any other. The critical quality is management, meaning support for staff and close monitoring of performance.” Judd goes on to report that buildings newly renovated by the government use far more energy than average. (For some reason, he is no longer a government employee.)
These data, and the higher energy use by newly renovated buildings with presumably more efficient equipment, tell me that the larger a building gets, the less insulation and other shell measures matter, and the more important temperature control becomes. To theorize the extreme, a oneroom house will always have the same temperature in all the rooms. A two-room house will probably be almost as evenly heated as a one-room house. But, it’s unlikely that a 100-family building will have the same temperature throughout, especially as wind and air leakage and solar gain change minute by minute. All too often the result is open windows, or people running air conditioners in the winter.
Overheating in large buildings is caused by underheating. That is, the coldest room in the building, which might have a heater blocked by a built-in bookshelf, or have a broken heater, will result in a phone call that has a high chance of resulting in the heat getting turned up buildingwide. I believe that the 600% spread can be explained only by open windows, and not by differences in combustion efficiency or insulation levels.
Tellingly, the heating energy champions of the study you mentioned are electrically heated buildings, which typically have thermostats in each apartment or room. It is also interesting to note that when two virtually identical apartment complexes, which are next door to each other, look identical to me, and share many management functions, were compared, one complex used 20% more energy than the other. (See “Which Type of Heating Uses Less Energy?” Table 4, on my Web site.)
Perhaps the choice of buildings studied explains the large difference in what the studies found. The 1973 study looked only at large buildings, with multiple full-time employees, while the other studies included numerous smaller buildings, which might not have even one full-time employee. The higher cost of overheating a large building, combined with the attention full-time employees can give, could explain higher fuel use in smaller buildings.
New York City was a very different place in the 1970s. The streets were barely paved, the subways were a mess, and the city government was flirting with bankruptcy. Rent control, combined with higher inflation in the ’70s, helped make buildings so hopelessly unprofitable that small- and medium-sized buildings were being abandoned by the thousands. My uncle, who was a firefighter, told me about changing shifts by standing in the street in front of the firehouse, waiting for the trucks to drive up and the crews to get off, boarding the trucks, and driving to one of the fires that were almost always burning. This was not a time when a tenant complaining about discomfort got a lot of attention, so there is some reason to believe buildings were not as hot in those days.
Sad to say, equipment arrangements probably have become worse since the 1970s. A typical apartment house then had an oil-fired boiler. Years later, many of these were replaced with atmospheric gas boilers. Atmospheric gas boilers have obscene standby losses compared to boilers with motorized burners, especially when connected to a 60-ft-tall brick chimney. Also, it became popular to add air venting to steam systems in an attempt to get “more steam” from the same boiler with the same firing rate, or to “move the steam faster,” which can result in buildingwide overheating.
All this strikes me as reinforcing themes in my article: Stopping the overheating is most important, reducing standby losses is next on the list, and boiler efficiency is least important.
Can Lighting Caution
While I appreciate the research presented by the article “Further Wrestling with Recessed-Can Lights,” (Sept/Oct ’05, p. 30), I think a word or two of caution is in order.An IC lighting can is not a sealed lighting can. IC cans have been tested, evaluated, and listed with whatever holes the manufacturer designed into them. Even with insulation in contact with the exterior of the cans, there will generally be some air movement through the holes into the insulation, and this is verified with air leakage tests. This air movement was taken into account when the lamp was designed and to some extent may provide cooling for some parts of the lamp assembly.
The thermal cutoff switch in these cans is a last line of protection for the internal parts of the lamp. Operation at temperatures approaching the thermal cutoff temperature is not a good idea. Sealing the holes and operating the lamp for many hours just below the cutoff temperature would not be a very safe operating condition. While not causing immediate failure of the internal components, the life of the insulation and other plastic parts of the lamp may be shortened. Some of these parts may be provided cooling via the air leakage through the can holes.
Furthermore, from a liability point of view, filling or taping these holes on IC cans will violate the listing on the lamp and would therefore be a violation of the National Electrical Code (NEC), Section 110.3(B). Where the NEC has been legislated into law (most parts of the United States), the taping of the holes would therefore be illegal. Taping these holes should not be done, and particularly not by professionals who are involved in installing or servicing these devices for the public.
Author and Recessed Can Man Larry Armanda responds:
Thank you for your comments on “Further Wrestling with Recessed-Can Lights,” which appeared in the Sept/Oct issue of Home Energy. The article addresses standard IC-rated recessed can lights that have been installed in inaccessible attic spaces and covered with some type of thermal insulation. This also does not meet current energy codes. Air sealed fixtures should have been installed. As you mentioned, there is some air movement through these holes into the insulation. This air movement is my concern, since it doesn’t stop at the insulation. My years of field-testing buildings tells me that this air movement through the insulation is substantial, as was indicated in Table 2, showing the results of a pressure pan test of recessed-can lights that I conducted on my own home.
The energy penalty due to this air leakage is substantial, and with the everescalating cost of fuel—not to mention our dependency on foreign fuel supplies and global warming—this air leakage can no longer be ignored. Another concern is the structural damage that occurs when warm moist air moves into the ceiling cavity, condensing on roof rafters and sheathing, and rotting the roof structure. This air leakage causes ice dams, which I wrote about in the Jan/Feb ’01 issue of Home Energy (“A Recessed Can of Worms,” p. 42). The ice dam that was caused by this recessed-can air leakage was roughly the size of a small sport-utility vehicle and probably consumed as much energy. When this ice broke loose from the roof, it crushed two A/C units on the ground below.
NEC Section 110.3 (B) reads: “Installation and Use: Listed or labeled equipment shall be installed and used in accordance with any instructions included in the listing or labeling”. The International Energy Conservation Code has requirements for non-IC rated fixtures to be “installed inside a sealed box constructed of a minimum of 1/2-inch gypsum wallboard or constructed from a preformed polymeric vapor barrier, or other material manufactured for this purpose, while maintaining required clearances of not less than 1/2 inch from combustible material and not less than 3 inches from insulating material.” This method of air sealing these types of recessed-can lights has been used by weatherization professionals for many years, and to the best of my knowledge with no problems. Manufacturers have been producing retrofit kits for a variety of manufacturers’ fixtures to address this issue of air leakage. Examples include the Lithonia Air-Tite Tape kit, similar to the kit I used in my research, the Halo Air Tite Super Trim kit, and the Do It Best’s Builders Best complete retrofit kit, which can be universally mounted to Halo and Juno fixtures.
As energy professionals, we are constantly battling code and energy issues. When these cans are installed in inaccessible attics, the energy professionals are left with the task of trying to eliminate the loss of energy and possible structural damage that results.
I installed six 6-inch recessed lights in my kitchen recently.The attic is above the kitchen, which made the installation easy, but now I am trying to find out the best way to plug up the six 6-inch holes that I have in the ceiling of the kitchen.The Halo fixtures I bought at Home Depot are IC rated, so they can be installed in insulation. I have blown-in insulation in the attic, but I don't think that is going to be good enough to keep the heat out of the attic. I live in Grand Rapids, Michigan, where it’s cold in the winter, and we get a good amount of snow.
What’s the best way to fill the holes? Is it best to create a box out of foam board or drywall to put around each light fixture in the attic? If I do install an airtight box around the fixture, does there need to be any clearance around the light? I thank you a ton for any help you might be able to give me, so that I can finish this project before the weather turns cold.
Larry Armanda answered this question in plenty of time before the cold set in. Here is his response:
Thank you for your question on air sealing recessed-can light fixtures. You are correct to be concerned about the amount of air that exfiltrates through this type of fixture. Not only do these fixtures move air, they also move moisture. When this moisture is exposed to a cold surface, which is most likely most of the fall, winter, and spring in Grand Rapids, air sealing becomes important and costeffective. I would not use foam board material to build the box above the fixture. A better choice, which is less costly, is building a box from drywall board. As far as clearances go, keep 3 inches of space around and over the top of the fixture. This clearance meets the International Energy Conservation Code. A word of caution, though. Do not insulate over the top of the air sealed box, since this could elevate the temperature inside the light fixture, causing nuisance tripping of the thermal overload. Using the reflector type of lamps will help with nuisance tripping.