musingsheader image
2 Helpful?

Window-Mounted Air Conditioners Save Energy

Compared to homes with central air conditioning, homes with window units have lower cooling costs

Posted on May 25 2012 by Martin Holladay

Window-mounted air conditioners (also called room air conditioners) aren’t particularly efficient; the best available models have an EEREnergy-efficiency rating or energy-efficiency ratio. As most commonly used, EER is the operating efficiency of a room air conditioner, measured in Btus of cooling output divided by the power consumption in watt-hours; the higher the EER, the greater the efficiency. of about 10 or 11. Central air conditioners (also called whole-house air conditioners or split-system air conditioners) are significantly more efficient; it’s possible to buy one with an EER of 14 or even 15.

So if you care about energy efficiency, you should use a central air conditioner, not a window air conditioner — right? Well, not necessarily.

Air conditioners use a lot of electricity

According to the American Council for an Energy-Efficient Economy, the electricity used for home air conditioning represents almost 5% of all the electricity produced in the U.S. So if you are going to air condition your house, it would appear to make sense to buy the most efficient unit you can find.

There are two metrics used to rate the efficiency of central air conditioners: the Energy Efficiency Ratio (EER) and the Seasonal Energy Efficiency RatioSeasonal Energy Efficiency Ratio (SEER) is the total cooling output (in BTU) of an air conditioner or heat pump during its normal annual usage period divided by its total energy input (in Watt-hours) during the same period. The units of SEER are Btu/W·h. SEER measures how efficiently a residential central cooling system operates over an entire cooling season. The relationship between SEER and EER depends on location, because equipment performance varies with climate factors like air temperature and humidity. (SEER(SEER) The efficiency of central air conditioners is rated by the Seasonal Energy Efficiency Ratio. The higher the SEER rating of a unit, the more energy efficient it is. The SEER rating is Btu of cooling output during a typical hot season divided by the total electric energy in watt-hours to run the unit. For residential air conditioners, the federal minimum is 13 SEER. For an Energy Star unit, 14 SEER. Manufacturers sell 18-20 SEER units, but they are expensive. ). The unit for both metrics is BtuBritish thermal unit, the amount of heat required to raise one pound of water (about a pint) one degree Fahrenheit in temperature—about the heat content of one wooden kitchen match. One Btu is equivalent to 0.293 watt-hours or 1,055 joules. /W•h.

EER is the cooling capacity of the appliance (in Btu/h) at an outdoor temperature of 95°F divided by the current draw of the appliance in watts.

SEERSeasonal Energy Efficiency Ratio (SEER) is the total cooling output (in BTU) of an air conditioner or heat pump during its normal annual usage period divided by its total energy input (in Watt-hours) during the same period. The units of SEER are Btu/W·h. SEER measures how efficiently a residential central cooling system operates over an entire cooling season. The relationship between SEER and EER depends on location, because equipment performance varies with climate factors like air temperature and humidity. is the total cooling output (in Btu) over the cooling season divided by the total electrical energy input (in watt-hours) over the cooling season. For typical residential air conditioners, EER equals about 0.875 SEER; however, this ratio varies somewhat for different air conditioner models.

EER and SEER ratings

Unlike central air conditioners — which are rated by both EER and SEER — room air conditioners (window units) are rated only by EER.

Room air conditioners that use at least 10% less energy than the federal standard are eligible for an Energy StarLabeling system sponsored by the Environmental Protection Agency and the US Department of Energy for labeling the most energy-efficient products on the market; applies to a wide range of products, from computers and office equipment to refrigerators and air conditioners. label. When shopping for a room air conditioner, look for an Energy Star unit with an EER of 10.7 or more.

The idea behind the SEER rating is to allow consumers to compare the seasonal efficiency (rather than the peak efficiency) of split-system air conditioners. If you are in the market for a new central air conditioner, look for an Energy Star unit with a SEER of at least 14.5 and an EER of at least 12.

Some critics have noted that air conditioner manufacturers design equipment to achieve a high score on the laboratory SEER test, even when some of the engineering solutions used to achieve high test scores result in poor performance or efficiency in the field. For more information on perceived flaws in the SEER standard and SEER testing protocols, see SEER Ratings Challenged.

John Proctor, the president of Proctor Engineering Group in San Rafael, California, has performed useful research into possible flaws in the SEER rating method. Proctor (with co-author Gabriel Cohn) has publshed a paper on the topic: Two-Stage High Efficiency Air Conditioners: Laboratory Ratings vs. Residential Installation Performance. Proctor and Cohn wrote, “The increased installation of high Seasonal Energy Efficiency Ratio (SEER) air conditioners along with utility program rebates for these units prompted a study of the measured performance of these systems. This project assessed the performance of these systems in the climate zones found in the mid-Atlantic region of the U.S. ... The data were analyzed to assess the relationship between laboratory testing and real world performance. This study found causes for concern including: actual seasonal energy efficiency ratios between 59% and 84% of the rated SEERs, constant fan operation substantially degrading seasonal efficiencies and reducing dehumidification, latent loads that exceed Manual J estimates, and sensible loads substantially lower than Manual J estimates.”

According to some experts, problems with the SEER rating system may make EER a more useful metric for comparing the efficiency of central air conditioners.

Window units versus central air conditioning

Even though central air conditioners are more efficient than window-mounted air conditioners, homes equipped with window-mounted air conditioners use less energy for cooling than homes with central air conditioning.

For some readers, this fact will be unsurprising. After all, homes with central air conditioning usually maintain lower temperatures in more rooms than homes with window-mounted air conditioners.

Other readers — including those who assume that equipment efficiency is the most important factor in energy use — may be surprised to learn that homes with window-mounted air conditioners are energy misers.

For those interested in the data behind this question, I recommend a blog by fellow energy nerd Tim Grejtak, “The Fascinating World of Air Conditioner Efficiency.” According to Grejtak’s analysis, “Window ACs use less energy on a per cooling degree day and per square footage basis by a factor of 1.75.”

So why do homes with window-mounted air conditioners use less energy for cooling? There are several reasons:

  • People living in homes with central air conditioning tend to keep every room in the house cool — even unoccupied rooms — while people living in homes with window-mounted air conditioners are more likely to just cool one or two rooms.
  • Central air conditioners often have ducts located in unconditioned attics. Research has shown that air conditioning systems with ducts in unconditioned attics have duct losses amounting to about 20% of the air conditioner’s cooling output.
  • Since window air conditioners are often noisy, homeowners remember to turn them off when they leave a room. By contrast, some homeowners with central air conditioning often leave their air conditioning system on all day, even when the home is unoccupied. As explained in an article by Brian Palmer, “The quiet and unobtrusive functioning of central air conditioners ... can lead to accidental overuse.”

The bottom line

There are several valuable lessons arising from this analysis:

  • Efficient equipment is no bargain if it encourages waste.
  • Two or three window-mounted air conditioners cost less to install than central air conditioning.
  • Bad ductwork is the Achilles’ heel of central air conditioning systems. If you have central air conditioning, be sure to seal all of the seams in your ductwork and insulate ducts well.
  • If you are designing a new home, be sure to locate all ductwork inside of the home's conditioned envelope.
  • If you aren’t home, turn off your air conditioner.
  • When you’re sleeping, you may only need a 6,000 Btu/h air conditioner to cool your bedroom. It’s wasteful to use a 2-ton central air conditioner for such a small load.
  • If you use one or two window-mounted air conditioners instead of central air conditioning, pat yourself on the back. You probably have a lower carbon footprint than your neighbors who have high-SEER central air conditioners.
  • The most efficient way to air condition your house is probably with ductless minisplit units. The only problem with this approach is the high cost of the equipment. However, if you have a new superinsulated house, and if ductless minisplit units can provide both heating and cooling, then these units may be a good, low-cost solution.
  • If you live in a dry climate, consider using an evaporative cooler to cool your house. The cost to operate an evaporative cooler is only a fraction of the cost to operate an air conditioner.

Remember: you don’t need a slash

I’ll end with a gentle reminder: the correct abbreviation for “air conditioner” is AC (or A.C.). It is not A/C. There is no slash.

It isn’t a fraction. No one is dividing A by C. It’s an abbreviation for two words. So, please — don’t include a slash.

Last week’s blog: “Choosing an Energy-Efficient Refrigerator.”

Tags: , , , , , ,

Image Credits:

  1. D Griebeling
  2. Traci Lawson

May 25, 2012 8:34 AM ET

Seems to me
by 5C8rvfuWev

you've also made a good argument for zoning with programmable thermostats.

May 25, 2012 8:53 AM ET

Edited May 25, 2012 8:54 AM ET.

Response to Joe W
by Martin Holladay

Installing a zoned forced-air distribution system can save energy if done right. It can also waste energy. Not all HVAC contractors are capable of designing and installing forced-air zoning equipment.

The problem with a zoned forced-air system is knowing what to do with the extra air flow when a motorized damper shuts off a zone. As Jeri Donadee explained in a useful JLC article, "When only one zone in a multi-zone system calls for heat [or cooling], there needs to be some way to dissipate the extra cfm output of the furnace. Some brands of zone control ignore this problem, and let the high airflow howl through the small duct. Other brands will allow for the other zone(s) to open slightly and allow for the air to “leak” into areas that do not actually require conditioning. A third option is to install a bypass damper that allows the excess air to be recirculated back to the return. How the “extra” air is handled is a matter of contractor preference. The surplus air issue is much less of a problem with a two-stage gas furnace or a two-stage heat pump, especially one with a variable-speed fan — one more reason for installing two-stage equipment."

May 25, 2012 10:03 AM ET

Origin of the slash
by Dick Russell

Martin, where did the slash in A/C come from anyway? If someone puts in AC, is he getting alternating current? I suppose in context AC will be interpreted correctly. I'll try to remember.

May 25, 2012 10:48 AM ET

Edited May 25, 2012 10:53 AM ET.

Why do some abbreviations include slashes?
by Martin Holladay

The errant slashes that show up in some abbreviations -- including a few abbreviations that have been around for decades, like c/o and w/o -- are of mysterious origin and impossible to justify.

In technical writing, where a slash is most often used to indicate division or a fraction, the introduction of an unjustified slash (in my opinion) serves only to introduce confusion.

May 25, 2012 11:15 AM ET

Edited May 25, 2012 11:16 AM ET.

What do you think about sizing?
by Eric Sandeen

I asked on an older article, but - it's older. So I'll re-ask here.

On the cooling load calculation article, the "Blasnik / Bailes" rule of thumb for a code-built house is 1 ton (12,000 BTU) per 1000 square feet.

The Energy Star guide is 12,000BTU for 450-550 square feet - twice the capacity.

What do you think? It's a selfish question because I'm about to replace a through-wall unit, and I'm trying to pick the right BTU (though I must choose between about 10k and about 8k, so not a huge range anyway).

May 25, 2012 11:34 AM ET

Response to Eric Sandeen
by Martin Holladay

One ton of cooling per 450 to 550 square feet is an old rule of thumb that makes sense for older houses.

For a new superinsulated house, a better rule of thumb is one ton of cooling per 1,000 square feet.

However, it's important to note that these rules of thumb are no more than a rough guide. If the room includes a lot of glazing -- especially unshaded east-facing or west-facing glazing -- you'll need more cooling capacity than you would for a room with smaller windows.

May 25, 2012 1:52 PM ET

Edited May 25, 2012 2:00 PM ET.

no ducts, no problem
by Dana Dorsett

Inverter drive ductless mini-splits, while considerably more expensive than window-shakers, will out perform them (and ducted whole-house systems) with margin. The part-load performance is well above the typical EER, and oversizing them for the peak loads slightly should result in better SEER type performance too, since it'll spend far more time in it's highest-efficiency low-compressor speed operating mode, but still won't short-cycle. See figures 14& 15, p. 19 (p.27 in .pdf pagination) in this document:

At lower compressor and blower speeds at outdoor temps between 80-90F the laboratory measured COP of the unit is literally twice the manufacturer's reported number (which it taken at a mid-level compressor and blower speed.) Oversizing slightly guarantees that it'll outperform the SEER number, and by more than just a little bit. With the high turndown ratios cycling losses even at 2x oversizing would be miniscule compared to the performance gained. The primary downside to 2x oversizing is the high air-flow volume of the bigger units resulting in noise & chill effects, but even 1.25x oversizing for the actual peak load bumps up the curve by quite a bit.

Referring to figure 12 in the same document (back 2 pages), the min-output performance of a the nominally 1-ton unit between 80-90F is about 6KBTU/hr, which is about that of the smallest window air conditioners. But the COP performance at that minimum (in figure 14) is about 8, which translates to an as-operated EER of about 27,or about 2x the efficiency of the very best window units.

This superb part-load efficiency makes them more amenable to "set and forget" strategies of use, since arriving home to an 85-90F house in the late afternoon and turning it on forces it to run at maximum speed/lowest efficiency to play catch up, which will in many cases result in higher overall power use, and in EVERY case represents a higher PEAK power load. Setting it to 78F for the day most of the heat is pumped out at a COP 3-5x more efficiently during the cooler part of the day than at max speed at or near the peak outdoor air temp. (Not a simple model, but works in general.)

May 25, 2012 1:57 PM ET

Edited May 25, 2012 1:58 PM ET.

Cooling and Heating Losses
by greenhouse437

Has anyone studied the losses--both AC and heating (heating if the AC is kept in the winter)--through the window AC side panels which are sometimes cloth-thin corrugated pieces that slide out to cover the window gap? Also leakage through the AC unit itself. There are outer and inner covers for this I know but I doubt this problem was factored into the efficiency ratings.

May 25, 2012 2:12 PM ET

Response to David Goldman
by Martin Holladay

The efficiency (EER) ratings of room air conditioners do not consider air leakage. The amount of air leakage will vary widely, depending on the care used during installation. Installation in a custom through-the-wall box, carefully gasketed, is obviously preferable to window mounting.

You're right that the corrugated side panels on most window-mounted air conditioners leave a lot to be desired.

I am certainly aware of all of the deficiencies of room air conditioners, which is why I wrote that "the most efficient way to air condition your house is probably with ductless minisplit units." Dana Dorsett apparently agrees with that conclusion.

May 25, 2012 2:30 PM ET

mini-splits are not necessarily THE most efficient way...
by Dana Dorsett

In some (but not all) instances the most efficient way to air-condition would be with ground source heat pumps (GSHP) but the very high first costs of GSHP often make the power savings of that modestly higher efficiency more expensive than making up the difference with grid-tied rooftop photovoltaic panels. Short of that, ductless mini/multi-splits are a super-efficient (and much happier) medium for most.

May 25, 2012 11:12 PM ET

Edited May 25, 2012 11:14 PM ET.

Purpose of windows
by Curt Kinder

Windows, complicated and expensive holes in exterior walls, are installed to afford light, view and ventilation to the room's occupants. They are not intended to be support and frame noisy, ugly, wet, air leaking metal boxes to cool rooms.

I grudgingly support their use in nothern zones for the 10-20 days per year occupants need some cooling and dehu to get some sleep.

I am a dyed-in-the-wool Floridian steeped in full meal deal ducted and zoned central sir solutions.

That said, I'm in the midst of my first minsplit install and am truly amazed by the efficiency, quiet, versatility and ease of installation of minisplits.

My own house has geo, and experiences the extraordinary efficiency a well-executed GSHP system affords - 3400 SF heated and cooled for $400 per year.

May 26, 2012 5:33 AM ET

Response to Curt Kinder
by Martin Holladay

While my article compares the energy used for cooling in homes with window-mounted air conditioners to the energy used for cooling in homes with central air conditioning, I agree with you that window-mounted air conditioners are noisy, ugly, and wet.

In fact, my article noted that one reason that people with window-mounted air conditioners save money is precisely because they are so noisy that they remember to turn them off.

For those who can afford the cost of installation, ductless minisplits are the way to go.

May 26, 2012 9:57 AM ET

I agree with the remarks about unnecessary operation
by Curt Kinder

Central AC does induce a sort of "moral hazard" in the form of inadvertent, unnecessary consumption.

I grew up in NE Mass, where AC was needed on blessedly few days, but I would have paid a pretty penny to have one on those few nights I couldn't sleep, and that discomfort is one of the things that drove me to learn about refrigeration and AC.

I went to college in Penna, where is is quite hot in fall. Window units were against regulations, so my first challenge upon arrival was to configure a window AC so that it looked like a fan from the outside. I did, it worked, and my room was popular on hot days. Beer keg coolers were illegal as well, but that's another story...

Society has sort of a comfort / luxury creep - how many cars are sold without AC, cruise control, and a fairly decent sound system today. I don't have the numbers handy, but I'd bet very few new homes are built w/o central AC, even far up north.

There simply aren't enough hard core environmentalists in my area to risk basing a business model on deprivation. While most folks would barely notice loss of AC in walk in closets, pantries, laundry rooms and the like, it is an uphill battle convincing them of that.

My first minisplit project is a 4 zone system going into, of all things, a downmarket rental consisting of two used doublewide mobile homes. The fact the sold the job is that the crawl space ductwork is shot, so the minis compared reasonably well with a conventional system plus a reduct.

What I need to have happen is for the minis to support more heads at lower tonnages - A 3 ton unit supports 4 heads of minimum 9 kbtuh capacity - what I really need are two ton units able to support up to 4 heads at 6 kbtuh each. Most bedrooms don't need 6 kbtuh, never mind 9 kbtuh.

May 26, 2012 1:49 PM ET

riding on curt kinder
by 5C8rvfuWev

you said:

"what I really need are two ton units able to support up to 4 heads at 6 kbtuh each. Most bedrooms don't need 6 kbtuh, never mind 9 kbtuh."

this is exactly where I've gotten stuck, too. I've talked to 2 installers so far and, when I raise the question, they both try to sell me on a ducted system which they say is "more flexible."

Not at all sure of where to take that thought --

May 26, 2012 10:51 PM ET

Those 10-12 days a year...
by Eric Sandeen

Yeah, up north, our old house only needs AC for 10-12 days a year, so we do live with the ugly, wet, noisy boxes in windows (well one lives in a hole previously punched through the back wall).

I'm still almost considering mini-splits though because I have a sad hunch those 10-12 days per year may be steadily increasing as time goes by, so maybe it'd be worth the expense.

But once I have that fantastic mini-split system I'll probably be way too tempted to run it all the time, as you say...

May 27, 2012 2:45 AM ET

Edited May 27, 2012 3:02 AM ET.

watch out
by David Martin

Holy hidden terabecquerels of Fukushima Daiichi, Energy Nerd!!!

I'm afraid manufacturers and installers of oversized, centralized, systematized HVAC equipment and their associated leaky, uninsulated ductwork in unconditioned spaces might not be too amused with your latest musings. Are you not concerned that you might overturn their entire business model?

May 27, 2012 5:15 AM ET

Edited May 27, 2012 5:15 AM ET.

Response to David Martin
by Martin Holladay

It's harder than you think to overturn a business model that supplies comfort or luxury, even when the business model includes very high operating costs and environmental hazards.

As Curt Kinder wrote, "There simply aren't enough hard-core environmentalists in my area to risk basing a business model on deprivation."

May 27, 2012 1:20 PM ET

Edited May 28, 2012 12:06 PM ET.

Ear muffs work better than jackets?
by Mike MacFarland

The implication of this article is analogous to stating "studies show that ear muffs keep ears warmer than jackets, so choose ear muffs over jackets on a cold day." Never mind that ear muffs hinder your ability to hear, are typically unsightly, and are usually only purchased by those who practice pocket protection. And because those who wear 'muffs are likely the same ones who haven't experienced the luxury of a comfortable jacket, to imply that jackets and 'muffs can be compared, at all, is like saying that hamburger is better than a Filet.
Room air conditioners and their more expensive overseas cousins, the ductless mini-splits, are a mere tool that may be able to meet performance and comfort targets, in limited situations. Home Performance contractors should be very cautious with statements, such as those in previous comments, that seem to support ductless mini splits as the efficiency silver bullet- especially when we cannot easily measure the delivered performance of these units. Better still is to evaluate homes which are silently kept affordably comfortable without temperature stratification over a very narrow range, like 3 degrees, and without unfavorable hotel-room-noisemakers placed on every wall.
How does a room air conditioner or ductless mini split assist in the circulation of ventilation air throughout every living space?
Our customers all want efficient, healthy, and comfortable homes. Only giving them one out of three, with a ductless mini or a window mounted AC unit (note the absence of a slash), is like throwing 'muffs on- whilst neglecting the jacket. I say let's compare only jackets to jackets, and concentrate on measuring and improving jacket performance. And the best place to start is by convincing jacket manufacturers to begin making smaller than 1-1/2 ton coats!

May 27, 2012 11:29 PM ET

No window units please
by James Morgan

Down our way there are a few hardy souls who soldier on through our long hot humid summers without any AC at all but for the rest of us central ducted systems have really been the only option for decades. Unlike Mike M we are welcoming split ductless systems especially for smaller better-insulated homes and I have hopes that costs will drop enough in the next few years to make them an attractive option for low-budget projects. Meanwhile their much higher unit efficiency especially in the heating cycle is bound to put pressure on the manufacturers of conventional systems to upgrade their technology.

Mike, we're accustomed to many of the more efficient appliances hailing from Europe but with names like Sanyo, Mitsubishi, Fujitsu and Daikin dominating the field I think you'll find that this technology originates in a completely different corner of the world.

May 28, 2012 4:43 AM ET

Response to Mike MacFarland
by Martin Holladay

Q. "How does a room air conditioner or ductless minisplit assist in the circulation of ventilation air throughout every living space?"

A. A ductless minisplit is used for space heating and cooling; it is not intended to be a mechanical ventilation system. Split-system air conditioners are not intended to provide mechanical ventilation either. If you want to include a mechanical ventilation system in your house -- and you should -- then it needs be be designed and installed.

Q. "Our customers all want efficient, healthy, and comfortable homes. Only giving them one out of three, with a ductless mini or a window mounted AC unit (note the absence of a slash), is like throwing 'muffs on- whilst neglecting the jacket."

A. Of course a house should be efficient, healthy, and comfortable. All over the world, homeowners are living in efficient, healthy, and comfortable homes that are cooled with ductless minisplits. It's certainly possible to design and build an inefficient, unhealthy, and uncomfortable house that includes a ductless minisplit unit, but I am not advocating bad design or construction practices.

On average, ductless minisplit units are far more efficient than split-system air conditioners with typical duct systems as installed in the U.S.

May 28, 2012 9:28 AM ET

Don't get me wrong, I'm a deep south central HVAC duct guy
by Curt Kinder

There is no doubt that a ducted system's ability to put 10 CFM into the walk in closet or pantry, 20 CFM into a small bathroom, and otherwise be adjusted to meet all the CFM requirements shown in a carefully executed room-by-room Manual J load calc. Done right it works. Some of my systems have four zones working off a single air handler.

The trouble is, doing ducts right is time consuming and therefore expensive. That coupled with the fact that ductwork placed in attics adds another 30% or so to the total load means there is a heckuva lot of room for improvement.

Of course ducts should be placed in conditioned space, but 80+% homes down here have them up in the attic. There are only two ways to remedy that: faom the attic or rebuild ducts in a series of soffits and chases. Both are very expensive but do work. We are able to reduce system tonnage by 1/3 or so via sprayfoam.

So if minis could be effectively, efficiently deployed in lieu of lots of attic ducts and sprayfoam there should be savings both upfront and down the road, since mini SEER values are as high or higher than high end ducted systems

The rest of the world seems to get by without ductwork...why can't we?

May 28, 2012 12:04 PM ET

Edited May 28, 2012 12:11 PM ET.

Reply to Martin
by Mike MacFarland

Thanks for your reply. I enjoy your blog immensely, and only occasionally disagree.  The mini split is such a time.  For example, Chapter 20 of ASHRAE Fundamentals (2009) starts with "Room air distribution systems are intended to provide thermal comfort and ventilation for space occupants and processes."  I am a strong believer in mixed air systems, using engineered diffusers and right sized ducts, to deliver measured quantities of proper velocity air to individual rooms, as calculated in Manual J/D.  We install stand-alone HRV's as well as exhaust-only systems to exceed ventilation standards. With properly sized heating and cooling systems in our hot, dry climate, we deeply bury radial designed ductwork in unconditioned attic space and measure distribution system delivery efficiencies above 90%.  
A product of proper ceiling-based air distribution systems is that they produce  almost perfect mixing of air  in cooling mode, with air change effectiveness approaching one. By nearly eliminating temperature stratification within spaces by delivering higher velocity air within the unoccupied zones, we deliver more occupant comfort, and better distribute ventilation air, than other methods.  I reassert my point that ductless minis and window mounted units should be classified as a method and tool which can, in some situations, provide acceptable performance. And without many unsightly boxes mounted within  the numerous rooms within a home, they fall well short of being a single, space conditioning system capable of providing code-acceptable heating installation in our state (CA) (68F at a distance 3' above the floor in every room).
Thanks again for the great information you provide on a weekly basis. It's a Sat AM highlight to know there is a new Musings to read with the morning paper.

(James- good point, I edited my previous comment to read 'overseas' as a better adjective.)

May 28, 2012 1:57 PM ET

Edited May 28, 2012 1:58 PM ET.

Reply to Mike MacFarland
by Martin Holladay

It sounds like you do an excellent job of duct design. That's good.

It's possible to use space-conditioning ductwork for providing ventilation, but it's not ideal. If the home uses ductwork for the space conditioning system, using the same ductwork to deliver ventilation air often makes sense. However, that's not how it's done in Europe, and that's not how it's done in hydonically heated homes in New England, where dedicated ventilation ductwork is more common.

It's very rare to see a well designed, well installed, tight duct system installed within a home's conditioned envelope. If you can manage that trick, bravo. However, your system will still incur an energy penalty compared to a ductless minisplit system, because you'll use electrical energy to blow all that air through all that ductwork, and you'll have to overcome the static pressure of the duct system to do so.

May 28, 2012 8:04 PM ET

Edited May 28, 2012 8:05 PM ET.

Reply to Martin
by Mike MacFarland

Yes, and doing good design is what I hope to inspire others to do in an industry with wonderfully skilled tradesmen who just need bosses who will pay them to do their best work. Through very careful design, we find that our total external static pressures average only about 60% of the manufacturers maximum (at 0.30" w.c. versus 0.50" manuf.), despite pushing over 500 CFM per ton across the coil, having high efficiency filtration, and despite ensuring perfectly tight ("LO") ducts in every installation ( And at an average of 6 CFM/Watt, we don't feel like we are paying any higher price than ductless mini's (exactly how do you measure that to compare, anyhow?). Our standards are high, and that's the message we need to get across to the industry- we need very high HVAC standards. Anyone who cares about quality can achieve these "A grade delivery efficiencies" in un-conditioned attic spaces, with conventional, simple equipment, with very reasonable installation times. And obviously your audience consists of these very smart folks who care very much about what they do, and I am just one of many who learn something new everyday in this wonderful field and from great info provided here and elsewhere.

And to further clarify on the ventilation, I'm not advocating an integrated space conditioning and ventilation system. The point is that proper room air diffusion will better mix the fresh ventilation air throughout the home (a bi-product of good room air diffusion), at least during the heating and cooling seasons, which, for us, is about 3/4 of the year. I fully agree that it's not ideal to use this ductwork for ventilation, and a stand-alone approach will enable much more efficient ventilation.

Thanks again Martin for your time and good work.

May 29, 2012 6:30 AM ET

Time for a haiku
by Peter Hastings 4C

As Curt Kinder wrote, "There simply aren't enough hard-core environmentalists in my area to risk basing a business model on deprivation."

If there are to be real benefits to the environment then proposed improvements must be not merely effective but adopted widely.

90%E x 0.1%A << 10%E x 10%A

Not moral living
but comfort with lower costs
must be the sales pitch

May 29, 2012 7:09 AM ET

Another environmental haiku
by Martin Holladay

What do we need now?
Rainfall wanes while heat waxes;
Please: carbon taxes!

May 29, 2012 1:56 PM ET

Great, timely article
by Carl Seville

Interesting coincidence that you put up this post as I am getting proposals for mini splits to replace the window AC units and single ancient floor furnace in my house. I am hanging on in my 700 SF (historic) shack that I am no allowed to remove to build my new house. I did spray foam the floor and roofline a while back so the house is relatively tight, although I can't do much to the walls as they have wood siding with no sheathing. I don't think the city would approve removal and replacement of the windows and siding, so I'm hanging tight for now.

The floor furnace does an adequate, although inefficient job of heating the house. The rare cold spell tends to make the house pretty uncomfortable. I have three ENERGY STAR window units that give me terrific control over the temperature, I only use them when I need them (78 degrees as I write this with the windows open in Georgia). It's extremely rare for me have any AC on other than my bedroom at night. I get a perverse pleasure of listening to my neighbor's heat pump run (winter and summer) whenever my windows are open.

The higher first cost of the mini splits over a single system is hard to justify, but I've wanted to experience them first hand so I can speak from experience, and I've decided that having the room by room control, and the higher efficiency is worth the expense.

One interesting point about mini splits. Apparently the single head units are noticeably more efficient than the multi-head type. Mitsubishi rates their single units at up to 28 SEER while the multi head models only go as high as 18 SEER. As pointed out, SEER rating isn't everything, but it still looks like single head units may be a better route for some application. Now if we could only get people to design homes small and simply enough that ductless minis will actually work, and find people who are willing to live with them.

As a side note, I think I'm going to keep my gas fired gravity floor furnace just in case the power goes out in the winter. I can always light the pilot and heat the house without electricity.

May 29, 2012 8:57 PM ET

Jevon's paradox?
by Rachel White

Does anyone else see the Jevon's paradox at work here? I'd encourage anyone who isn't familiar with this notion to read Martin's post from a couple of years ago on it (just search for Jevon's paradox and it will pop right up). Simpy out, the Jevon's paradox states that improvements in efficiency fuel greater consumption. Of course efficiency alone can't account for excessive consumption, but it sure may play a role.

May 29, 2012 11:03 PM ET

Evaporative Cooling for the Southwest is a no brainer
by Kevin Dickson, MSME

I just couldn't let this go on without mentioning the best option for hot dry climates - swamp coolers.

"on a Btu-of-cooling-per-kWh-of-electricity basis, the best evaporative cooling systems are on the order of five times more efficient than SEER 13 central air conditioning (CAC) systems and demand is less by a factor of four or more. Further, additional water use at the site (home) amounts to only about 3 percent of the water use of an average residential customer." From an NREL report.

Haters will note there is quite a bit of maintenance. I agree, so I designed mine to be indoors with an outdoor air source. If you have to pay the $300/yr in maintenance to a contractor, this can destroy your savings.

May 30, 2012 5:43 AM ET

Response to Rachel White
by Martin Holladay

I think you're right -- the Jevons paradox is at work here.

Here's the link: The Jevons Paradox.

May 30, 2012 5:50 AM ET

Edited May 30, 2012 6:21 AM ET.

Response to Kevin Dickson
by Martin Holladay

Thanks for the important reminder about evaporative coolers. You're right. I will edit my article to include a sentence or two about evaporative coolers.

An article in the GBA Encyclopedia includes quite a bit of information on evaporative coolers, including the following:

"Evaporative coolers, also known as swamp coolers, use a fan to blow outside air over pads that have been soaked with water, causing the water to evaporate and the air to drop in temperature. As long as they are installed in a location with a dry climate (in the U.S., that generally means west of the Rocky Mountains), they will use far less electricity to cool a house than central air conditioning. ...

"If outdoor air is dry, an evaporative cooler can lower the temperature of incoming air between 15 F° and 40 F°.

"Evaporative coolers add moisture to the air, a plus in arid regions. In regions where relative humidity is high, evaporative coolers can't cool the air effectively because the air is too damp for further evaporation.

"Houses in climates where the outdoor wet-bulb temperature exceeds 70° F for no more than 1% of the cooling season can be comfortably cooled with an evaporative cooler. That area includes most of the Western U.S. Exceptions include regions that regularly experience several weeks of hot, humid weather. ...

"Indirect coolers use the evaporation process to lower the temperature of a heat exchanger. Outdoor air passing through the heat exchanger falls in temperature but doesn't pick up any additional moisture.

"The best known indirect evaporative cooler is the Coolerado.

"Two-stage evaporative coolers first cool the air in an indirect evaporative cooler. After that, the air passes through a conventional direct evaporative cooler, further lowering the temperature of the air while raising its moisture content.

"The best known two-stage evaporative cooler is the OASys."

May 31, 2012 4:44 AM ET

Edited May 31, 2012 5:14 AM ET.

Evaporative Cooling
by Kevin Dickson, MSME

Just a few caveats:

Coolerado and OAsys are higher tech and more efficient than conventional coolers. They are also at least five times more expensive.

Evaporative coolers for window mounting can cost as little as $300-$400, AND can cool the entire house:

As great a solution as they are, they aren't gaining any market share in the new home market. This despite my constant blogging and even a zoning law in Denver that allows a swamp cooler in the side setback but not a compressor.

They just can't seem to shake their "low income" status.

May 31, 2012 10:12 AM ET

Air Leakage Around Window Units
by Kohta Ueno

Has anyone studied the losses--both AC and heating (heating if the AC is kept in the winter)--through the window AC side panels which are sometimes cloth-thin corrugated pieces that slide out to cover the window gap? Also leakage through the AC unit itself.

the only information I have seen around is Steven Winter's study on NYC air conditioner sleeves (not window units)--but it is analogous.

Party Walls Newsletter
4/1/2011 The Hidden Cost of Room Air Conditioners

To quantify the effect room air conditioners have on infiltration and, consequently,
heating energy use and cost, SWA developed a method for measuring the
effective leakage area of a particular room air conditioner using a modified blower
door test. Tests were conducted on sixteen room ACs across eleven multifamily
buildings in New York City. The findings revealed the average room air conditioner
from this sample had a leakage area of 6 square inches, which when scaled up to
the NYC multifamily building sector, equates to an estimated 167,000 square foot
hole – almost the size of a typical Manhattan city block! On an annual basis, this
immense opening translates into an operating cost penalty between $130-$180
million and the discharge of 375,000-525,000 tons of CO2 into the atmosphere.

While the sample size was small and further research is needed to ensure a
statistically significant study, this paper identifies the technical obstacles that must
be overcome to reduce the heating energy penalty of room air conditioners.
“There Are Holes in Our Walls” presents the findings of our work and details the basic
design considerations of room ACs, the human factors in their application, the
negative impacts on building envelope efficiency, and the possible solutions to
those impacts, both available and proposed.

Not sure how much good covers do... probably depends entirely on how well they are installed and how good their seals are--that's like asking, "how good is weather stripping on doors?" I've seen it installed great, I've seen it where you see daylight around it.

Of course, recommending that folks pull the AC units out of the windows in the winter is key.

May 31, 2012 10:48 AM ET

Swamp Coolers are dinosaurs
by kim shanahan

There are far better low cost cooling solutions for the high Southwest deserts than swamp coolers. Historically they have been roof mounted with a single dump and no ductwork. That means a big whole in your roof, which for climate zones 4 & 5 means huge energy losses during the longer heating season. The typical solution was to slide in a "cookie sheet" to ineffectively mitigate the heat loss. When ducted system are installed the ducts needed are significantly larger than those for AC units. Also they use a ton of water; not good in drought-stricken areas. They also are ineffective if the windows are not left open a crack, which most home-owners are relectant to do when they leave their homes empty durng the work day. Coming home at 5:30 to a closed up sweltering home and then cracking the windows and turning on the swamp cooler means making dinner in a hot house while the cool moist air starts moving around.

The better solution in the Southwest is to build a tight house and install HRVs or ERV that run 24/7 (not a fraction) on low speeds. This is the same principal espoused by your grandfather who opened up the windows at night and shut them during the day. Even when we see daytime temperatures near 100 we typically get nightime temperatures in the 60s or lower. HRVs and ERVs
as developed in Canada were primarily designed to save heat but they work equally well to conserve cool and allow for pretty even temperature modulations. I know they work because we put them in 25 affordable homes in Santa Fe back in 2007- 2008. No AC, tight homes, healthy air exchanges and cool temperatures with closed windows. Elegant solution

May 31, 2012 11:04 AM ET

Response to Kim Shanahan
by Martin Holladay

You have accurately described the problems with badly designed and badly installed evaporative coolers. However, just because badly designed and badly installed systems exist is no reason to condemn the technology.

You are a fan of ERVs. However, trust me -- it's certainly possible to have a badly designed and badly installed ERV. So what?

Anyone who wants to install an evaportive cooler needs to design the system well, install it well, and use it appropriately. As long as you live in a dry climate and have enough water to run the system, you'll be comfortable and save a lot of electricity compared to an air conditioned house.

Neither ERVs nor HRVs can be used for cooling. If you operate an ERV or HRV on a day when the outdoor temperature is higher than the indoor temperature, you'll be introducing heat into the house and raising the indoor temperature.

To flush the air out of your house at night when the outdoor air temperature is low, what you need is a whole-house fan, not an ERV or HRV.

Jun 1, 2012 9:05 AM ET

Mini Splits Don't give you fresh air.
by Andrew Coates

I think one thing that people forget is that most mini splits are installed without any system for refreshing the air. This makes them more efficient but eventually the air gets stale. You see them used in dark offices all over the world where they run 24 hours a day in the middle of the building and everyone gets sick. It has also become everyones solution for solving bad design in hot climates, especially in the third world.

Jun 1, 2012 9:33 AM ET

Response to Andrew Coates
by Martin Holladay

Andrew Coates,
You worry that ductless minisplit units don't provide fresh air. However, neither do window-mounted air conditioners or split-system air conditioners. All of these different types of equipment are space cooling systems, not ventilation systems.

if you want to install a ventilation system in your home, you are free to do so. In some cases, the ventilation system uses existing space-heating and space-cooling ductwork to distribute the fresh air. In other cases, the ventilation system has dedicated ventilation ductwork. But in all cases, the ventilation system needs to be designed and installed -- it doesn't come as part of the cooling equipment.

Jun 1, 2012 10:37 AM ET

HRVs can cool
by kim shanahan

"Neither ERVs nor HRVs can be used for cooling." True, they are not designed to do so, but they can modulate temperatures relative to what is being brought in as fresh air. And in that respect they work. When we had exterior temperatures near 100 the interior tempertures were in the high 70s - low 80's, similar to what one can get from evap coolers. The cool night air in the mid 50s to low 60s brought interior air temps at night to the high 60s to low 70s. So the interior temperature swings were much smaller than exterior, And we had fresh air. We were quite surprised the HRV worked as well as it did to help keep the home cool. In fact, we wondered why someone couldn't design an evap feature on the return air side of the Lifebreath furnace with its built-in HRV to truly create a cooling function.

Jun 1, 2012 10:45 AM ET

Edited Jun 1, 2012 10:46 AM ET.

Response to Kim Shanahan
by Martin Holladay

I don't really disagree with you, but running an HRV during the day, when outdoor temperatures are higher than indoor temperatures, still raises the indoor temperature. So under those conditions, the HRV provides heating, not cooling. That may be inevitable if you want fresh air, but it is still a fact.

At night, of course, when the temperature drops to the 60s, it's a different story. But that's when you want a powerful fan, not an HRV.

Jun 1, 2012 3:49 PM ET

PTHPs, PTACs, and some Window Units have a Ventilation Damper
by Kevin Dickson, MSME


All the best window and through-the-wall units have a small damper to provide optional ventilation.

See page 5 of this document -

Kim, I agree that swamp coolers shouldn't be used for new, low energy homes. But that still leaves a huge market.

Jun 1, 2012 4:09 PM ET

Response to Kevin Dickson
by Martin Holladay

Thanks for the information on through-the-wall air conditioners that have a ventilation option.

Jun 4, 2012 6:48 AM ET

Air Infiltration Through Window AC Units
by Len Moskowitz

Window AC units are notorious air infiltration leak paths. Consider what even one window unit would do to a Passive House's ACH numbers.

Jun 4, 2012 7:52 AM ET

Edited Jun 4, 2012 7:53 AM ET.

Response to Len Moskowitz
by Martin Holladay

If a Passivhaus needs cooling, most designers would specify a ductless minisplit unit.

For ordinary Americans looking to save energy, however, a room air conditioner can make sense. For the most airtight installation, a room air conditioner can be installed in a through-the-wall sleeve that includes gasketing at the perimeter.

Jun 5, 2012 11:32 AM ET

Portable units
by Chris Churchill

What do you think of the portable AC/heat pump units that use a duct through the window for the outside air?

Jun 5, 2012 11:48 AM ET

Edited Jun 5, 2012 3:09 PM ET.

Response to Chris Churchill
by Martin Holladay

Portable air conditioners that have a flexible duct between the unit and a window are not using the duct to pull in outdoor air; they are using the duct to exhaust hot air.

Air conditioners remove heat from a room. To work properly, they have to send that heat outdoors. If a portable air conditioner has no duct to exhaust heat, it won't cool the room where it is located.

I'm sure that these units use a lot less electricity than central air conditioning. However, the Achilles' heel is the duct connection through the window. If you can rig up a piece of gasketed plywood that limits air leakage at the area where the duct goes through the window, I imagine that such an air conditioner wouldn't be any worse than the typical window-mounted unit.

[Later edit: in fact, these units are significantly less efficient than window-mounted units; see the comments that follow.]

Jun 5, 2012 2:35 PM ET

Martin, I don't want to get
by Chris Churchill

I don't want to get too far off subject here, but I've really been wondering about these units. It seems that you wouldn't want to be exhausting the room air that you've already paid to cool or heat. If the unit has brought a room down to 70F during a hot day, then continuing to use the unit would cause it to remove the 70F air from the room to condense the refrigeration gases. The user has already paid for that 70F air! The same is true in heat mode. I can't settle the fact that there's not both a intake and exhaust duct on these things. Thoughts?

Jun 5, 2012 2:47 PM ET

Response to Chris Churchill
by Martin Holladay

I think you are right. Because these units aren't using outdoor air to cool the condenser coils -- they're using indoor air that then gets exhausted -- I imagine that they are less efficient than window-mounted units. In a way, these portable air conditioners are contributing to air leakage.

Thanks for pointing this out. Below is an illustration of a Frigidaire portable air conditioner.

Frigidaire portable air conditioner.jpg

Jun 5, 2012 3:08 PM ET

More on portable air conditioner efficiency
by Martin Holladay

From a Consumer Reports web page on portable air conditioners:

"Our past tests of three large portable air conditioners, each with a claimed cooling capacity of 10,000 Btu/hr, showed that they delivered less relief from the heat than their manufacturers touted. What’s more, these $400 to $500 appliances were pricey, especially compared with the window air conditioners we recently tested. (Use our free calculator to determine what size air conditioner(s) you need.)

"The portable units we tested had a single exhaust hose that routed air from inside the room over the air conditioners’ condenser coils and finally to the outside. A typical portable air conditioner comes with a low-profile vent adapter bracket that you place in a double-hung window to allow the hot, humid air to exhaust (generally no tools are required for this installation).

"While portable air conditioners might be convenient, those we tested delivered only about half of their cooling capacity—that means they operated with an energy-efficiency ratio (EER) of about 5 or 6. Compare that with the minimum EER of just under 10 for the window air conditioners we’ve recently test (we tested window models with an EER as high as 12).

"Why are portable units are so inefficient? The air they exhaust to the outside comes from inside the room you’re trying to cool (as well as from adjacent spaces). The energy used to cool and dehumidify the air is essentially wasted since it gets sent out the vent hose.

"What’s more, this setup means that warm, humid outside air will leak back into the room, and that air must then be cooled and dehumidified. The result: inefficient operation."

Jun 5, 2012 4:09 PM ET

Excellent info! Thanks
by Chris Churchill

Excellent info! Thanks Martin. These units would be a great option if the manufacturers made the easy fix of adding an intake duct.

Jun 8, 2012 3:16 AM ET

Edited Jun 8, 2012 3:20 AM ET.

Dual Hose Units Have Been Around a While
by Kevin Dickson, MSME

Dual hose portable AC units don't cost significantly more than a single hose unit.
Problem solved.

And be careful using a cheap window unit in a wall sleeve. The outdoor intake ducts are usually on the side, and must not be plugged.

Register for a free account and join the conversation

Get a free account and join the conversation!
Become a GBA PRO!