What would a “no air-conditioning” design in Phoenix AZ entail?
I am designing a small (600 sq ft) guest house in a Phoenix AZ suburb and in my quest for it to be as energy efficient as possible, I was struck by the question: is it even possible to make it such that it doesn’t require any air conditioning at all? (No “cheater” alternatives like swamp cooling, either)
I had an energy audit performed on my existing house that came with many potential upgrades modeled. I was struck by just how little impact almost anything I did would make on our cooling load. Our (many) windows are shaded at all times so solar gain means nothing but upgrading them to have a much higher r-value showed essentially no benefit in the model. Upgrading our R-2 (or so) slump block walls to R-10 would save us just a few hundred kWh a year. Sealing air leaks would save even less. And on and on. My July energy bill shows roughly 2,000 kWh for A/C alone for my 2,500 sq ft ranch home. No amount of modeling indicated anything but a small drop in energy and certainly nothing even remotely close to even pretending to go without A/C.
Given that, I have no idea at all what a no A/C design would even look like, here, if it’s even possible at all.
It couldn’t use the traditional standby of tons of thermal mass (e.g., 2 ft thick adobe walls) since Phoenix is in a heat bubble and summers never dip very much below 95-100, even at night.
Huge amounts of insulation (R-100+) on the walls seems unlikely since R-10 does little and every doubling beyond that provides exponentially (maybe?) less benefit, each time.
Roof insulation definitely does help — I got July down to 2,000 kWh largely because of upgrading the attic to R-50 (from maybe R-15 or less). But, again, that only went so far.
Thoughts? Is it a pipe dream?
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
In comparing a heating load in a cold climate to a cooling load in a hot climate, there are two major differences. In a cold climate you have solar heat gain and waste heat (things you do for other purposes, such as cooking, just being in the house (50-100W per person)). There's no equivalent for either of these for cooling. Active cooling is pretty much your only option, unless you place your house substantially underground.
Kurt,
Indoors, as Trevor correctly notes, you've got heat gains from appliances, lighting, people, and dogs. Without air conditioning, more insulation doesn't do much.
Before 1940, millions of people in hot climates lived in homes without air conditioning. The homes were hot, of course, especially in climates like yours, where "summers never dip very much below 95-100, even at night." People sat in front of electric fans. Some installed swamp coolers. (Swamp coolers aren't "cheating," by the way. More information here: "Saving Energy With an Evaporative Cooler.")
All of that said -- most people in hot climates prefer to live in a house with air conditioning, especially if the alternative is living in a house where the interior temperature is 95 to 100 degrees.
You might list all the sources of heat gain. Also look at AC efficiency.
You can get the AC load in a 600 sq ft building down to negligible cost (although at some point, PV solar is more cost effective).
Not recommended, but if you wanted to avoid AC, then yes - fans, evaporative cooling, thermal mass, good air sealing and movement of large amounts of air through the building at night (when even August should be in the low 80s).
Have you determined that this guest house w/out active cooling would even be code compliant?
A no-AC strategy has to fall back on cave technology. R-value is almost pointless- what you need is thermal mass, extremely window area, and extremely limited power use indoors, and (seasonally) nighttime ventilation. It will still be pretty hot in summer, but it can be designed to not be dangerously hot for mammalian occupants.
Take a look at some traditional adobe structures or hogans of that region for design clues. Exterior shutters on windows to reject solar gain, whitewashed high mass walls & roofs to limit heating of the high mass walls. Traditional homes in hotter-drier parts of India use a similar high mass/low gains strategy, along with parapets and a flat roof used for socializing & sleeping outdoors when it's too hot indoors.
Kurt,
Your endeavor is one of futility. Unless you want to live in a cave deep inside a mountain. There is NO WAY to achieve your aspirations of having a non-AC home built to survive a Phoenix summer.
I've lived in Phx and the summer months have a 3-4 month monsoon season which sees high humidity/dew points. You want to know what misery feels like? Try a 110F day with a 60F degree dew point. Good luck residing in a home with those temps and humidity levels without AC.
Spend the $2,000 a get a ductless mini split installed. Problem solved.
It would have to be underground, and even then, you wouldn't get away from A/C entirely, I'll bet, short of being waaaay underground.
My own feeling is that between the temperatures and the water shortage, people just shouldn't BE living in that area in great numbers. But a whole lot of folks seem to disagree with me.
Thanks for all the replies. That does confirm my suspicion that it's not possible.
One of the reasons for the thought exercise, though, was to also tease out strategies that I might not have heard of or thought of and to expose areas where I didn't know I didn't know something.
For example, I hadn't factored in the INTERNAL heat gains. I tend to focus on LEDs everywhere and almost exclusively Energy Star compliant appliances but I can see that it would all add up.
But... that also exposes another question. If internally generated heat is one of the main reasons why it's not possible to go without active cooling (to include swamp coolers in the mix), then why does the outside temp still dominate the need for cooling inside?
That is to say, the internal heat load in my house is going to be relatively static throughout the year since it's the same people and the same things. Yet I use 2,000 kWh on my A/C units in July and ZERO kWh in November. That 2,000 kWh difference has to be 100% because of the outside temp, since there's no other different heating source.
That makes sense to the point of being obvious and barely worth saying... but if it really is that simple, then why doesn't extra insulation make that big of a dent? Why do the energy models show incredibly diminishing returns to insulation to the point that no amount entered in the model would substantially diminish the effects of the outside heat gain?
Given all that, let's rephrase the question to ask if it's possible to build a house in Phoenix that has active cooling ONLY for internally generated heat and is otherwise immune to the outside temperature?
>"...why does the outside temp still dominate the need for cooling inside?"
Outdoor temperature does NOT dominate the need for cooling inside.
The dominant factor in cooling loads is solar gain, primarily through roofs and windows (not walls) in most houses. In high-R houses or houses with a lot of thermal mass it's mostly windows. Roof decks can hit 150F even when it's only 75F out. Windows take in a lot of heat (even north facing windows with no direct sun) in bright, reflective landscapes.
High thermal mass roof & walls, solar reflective exterior finishes, small windows, exterior shades are what it takes to quell cooling loads.
Kurt,
Q. "Is it possible to build a house in Phoenix that has active cooling ONLY for internally generated heat and is otherwise immune to the outside temperature?"
A. No. If the outdoor temperature is (for example) 100 degrees F, and the indoor temperature is 76 degrees F, there will always be heat flow through the wall and roof assemblies, from the exterior to the interior. Insulation can slow the rate of heat flow, but not stop it; that's physics.
Since these concepts seem to be new to you, I suggest you read these two articles:
"Calculating Cooling Loads"
"My House is Too Hot"
Kurt, I calculate around $30/mo to cool a 600 sqft building with moderate R values. What do you get?
Jon, I was only able to model my current home because of the energy audit I had done on it through the local power company. I don't have access to modeling software going forward, so have no numbers for the guest house.
What were your assumptions in the model you ran, that gave you the $30/mo?
An indirect way to lower heat gain is dense leafy trees shading the sun. But don't expect to not need AC. You might be able to improvise some solutions, a wide diameter 100ft deep concrete "rod" extending into the earth with its excellent conductivity though how long till the earth around it is heat saturated? A poor mans geothermal cooler. I have no idea how much coolness you will get out of it.
The basic idea is doubling insulation will cut heat flow in half. At low R values this is simple, R2 to R4 is no big deal, R10 to R20 is no big deal but doubling gets to diminishing returns and unwieldy numbers, R20-40 is not crazy but R40 to R 80 is not likely to be installation friendly or cost effective.
The best you can do is very low air infiltration, eliminate thermal bridging, low SHGC windows and well insulated. As for interior gains people are hard to eliminate, lights can be LED but things like dryers never use, and your fridge is probably equivalent to an extra 1 to 1.5 humans. Eliminate phantom loads, you may be surprised how much a cable box uses.
Global warming is just adding to your problems. I had relatives who lived in a trailer in Tucson with no AC and did okay with just fans, but that was over a half century ago. My brother in Austin used to talk about 30+ days per summer of 100+F daily high temps. Now he's talking about the number of 110+F daily high temps. I grew up in New England with the ground covered by snow from Dec to Feb 60 years ago. No longer. IMHO places like Phoenix and Florida are in danger of being almost inhabitable in 50 years, since climate change is not being addressed.
Phoenix average July temp went from 90 in 1950, to 97 in 2018.
Phoenix average daily high July temp went from 98 in 1950, to 105 in 2018.
Phoenix average daily low July temp went from 76 in 1950, to 88 in 2018. That daily low temp is what you need for natural air cooling the house overnight, so it has become intolerable for human occupancy without some cooling system. (Online data source: WeatherUnderground)
Yeah, the ongoing and worsening climate change is going to affect everybody, but likely places like Phoenix more than most. I don't know if the current increased average temperatures is related to that just yet, though -- it's more likely due to the heat island created by adding some 4 million people(!!) since 1950.
" If internally generated heat is one of the main reasons why it's not possible to go without active cooling (to include swamp coolers in the mix), then why does the outside temp still dominate the need for cooling inside?"
It's really the combination of internal gains an external temperature. The external temperature is definitely the bigger driver. If that outside temperature rarely drops below your comfort temperature, say 75F, then you have NO means of cooling the house - which means any heat generated in the house will begin to build up, and it has nowhere to go (until you get well over the ambient outside temperature).
I think this might be a good analogy. Imagine wearing a winter coat on a warm, sunny day. You are going to be sweltering in that coat; much, much more uncomfortable than if you didn't have the coat on. While it's counter-intuitive, ALL that discomfort is from the internal heat gain of your body. The easy way to see this is to imagine replacing your body in that coat with a mannequin, and realise that mannequin is actually cooler than it would be without the coat.
>"It's really the combination of internal gains an external temperature. The external temperature is definitely the bigger driver."
Outdoor air temps may be a bigger driver than internal gains in most houses, but it's far from the biggest driver of cooling loads.
In an insulated low mass house with windows the outdoor air temperature is secondary to window gains. The surface temperature of the roof (think "large unglazed absorptive solar collector") easily reaches temperatures 50-80F hotter than the air temperatures, and the surface temps of sun exposed walls can hit peak temps well above ambient too. With lower mass construction the delay in those higher temperatures affecting interior ceiling/wall temperatures (and thus the cooling load) is fairly short, less than 2 hours.
In house with HIGH mass roof & walls the sensible load on the interior from air temperature and even direct solar gains on the roof gets averaged out over the course of a day, lowering the peak & average loads. At dawn the surface temperature of the roof in arid or semi-arid locations like Las Vegas NV or Phoenix AZwill be a few tens of degrees COOLER that the outdoor air temperature due to nighttime radiation into the cold night sky, offsetting some of the solar heat gains through roofs/walls.
But with windows the gains are intense and immediate, with no offsetting overnight radiation to help out. Exterior shutters/shades are more effective than heat rejecting soft low-E coating windows, but the latter still help.
Using high solar reflective index (SRI) roofing reduces the surface temperatures to only a few 10s of degrees above the ambient air temperature, but doesn't reduce the nighttime radiational cooling, reducing the net load from roof gains whether high mass or low (but high mass will still reduce the peaks more than low mass construction.)
To a lesser degree, high SRI wall finishes will help too, primarily on unshaded east/south/west facing walls. The direct solar gains of walls are much lower than roofs, partly due to the incident sun angle, but also due to much higher convection on vertical surfaces compared to angled or flat surfaces.
There is a reason why buildings in the warmer-sunnier parts of the Mediterranean look like this:
http://www.visitgreece.gr/deployedFiles/StaticFiles/Photos/Generic%20Contents/Nisia/santorini_stair_view_560.jpg
https://images.oyster.com/photos/street-and-surrounding-area--v9861073-720.jpg
Note the limited amount of window area, the color of the roofs & walls, and high mass masonry construction rather than 2x4s with thin sheathing & cladding with a half inch of gypsum on the interior.
In Phoenix during summer the nighttime temps are hovering at 100F at midnight and drop into the low 90's by 5am and the heat-up begins when the sun comes up. Temps push to 115F during the day.
The "heat island" created by concrete, asphalt and rooftops traps this heat and prevents it from cooling off at night. That's why central Phoenix doesn't cool down at night and this drives up the average temps. The outskirts of Phoenix, areas like Anthem or Buckeye, see better nighttime cool downs as they still have areas of open desert and haven't been developed like central Phoenix.
Then there is the monsoon season which makes a swamp cooler COMPLETELY USELESS as the air is saturated with moisture and swamp coolers need dry air to work properly. So the 3 hottest months of the year also have the highest dew points.
Like it was mentioned, just get a $2,000 minisplit and pay the $30 a month to run it. Problem solved. If you want to build a house without AC, finding a better and milder climate would make your endeavor easier. Try a Zone 4B or 5B climate and something in a higher elevation. The vast diurnal temps from daytime to nighttime are better suited for no AC and you can use thermal mass to help temper the swings.
I'm in a Zone 4B desert climate at 5,000 feet. It can go from 95F (day) to 65F(night) in the summer. I still run minisplit AC and set it at 70F and forget it (I like being comfortable). Maybe $25 per month to run it during summer. Minisplit is off during fall and spring. Winter, if it's cold, the unit may come on a few times for heat. Otherwise it's off during winter. House is R-23 ICF with 6" concrete core and a R-40 SIP roof. In climates like mine, ICF/thermal mass, really shine.
>"In Phoenix during summer the nighttime temps are hovering at 100F at midnight and drop into the low 90's by 5am and the heat-up begins when the sun comes up. "
For a visual on Phoenix mid-summer norms see:
https://weatherspark.com/m/2460/7/Average-Weather-in-July-in-Phoenix-Arizona-United-States#Sections-Temperature
>" House is R-23 ICF with 6" concrete core and a R-40 SIP roof. In climates like mine, ICF/thermal mass, really shine."
True, but that's still not enough to skip the cooling for most people.
With high mass, high SRI construction it will be possible to keep the indoor temps under human body temp using night time ventilation strategies, but without mechanical cooling it won't be possible to keep it under 90F. Humans can acclimatize to 90-95F environments, but it requires less & lighter clothing, staying well hydrated, and taking care to avoid strenuous activities during peak weeks that could induce heat stroke.
> makes a swamp cooler COMPLETELY USELESS as the air is saturated with moisture
Peter, can you put some figures on this?
I see dew points around 65F (far from saturation) where evaporative cooling is useful, especially at the elevated interior temperatures that a no-AC plan would require. Perhaps 3 pints/hour for 600 sq ft.
Jon,
If you haven't lived in Phoenix in a home that only had evaporation (swamp) cooling during the summer months. The data figures are meaningless. The Phoenix climate is NOT what it was back in the 1950's before development and the heat island. The hundreds of thousands of outdoor pools, golf courses with greens that are watered constantly, the concrete, asphalt, roof tops, cars, the extreme heat, etc.
It's very miserable trying to run a swamp cooler in monsoon season without modern AC. It's 110F outside with 60+ degree dew points. The temps at night are in the mid to upper 90's. Been there, done that.
Swamp coolers do OK in the late fall, winter and early spring but can't provide proper cooling during the summer monsoons in Phoenix. They worked back in the 1950's - 1980's but the buildup and heat island of the 2000's in the Phoenix area has created such a climate change that swamp coolers are no longer recommended or used. The Phoenix climate has seen it's avg highs and lows climb by 10 degrees since the 1950s.
Older homes just get a modern AC installed. Nobody wants to mess with swamp coolers anymore. They don't provide proper cooling in the Phoenix summer anymore. Try living out a Phoenix summer without AC and just a swamp cooler.
The Phoenix news did a story in 2010 on how evaporation coolers are going away in Phoenix and no longer used. As of 2010 LESS THAN 3% of Phoenix residents rely solely on a swamp cooler. Today, that stat is about 1%.
http://archive.azcentral.com/business/articles/20101010biz-evaporative-coolers-disappearing-from-Phoenix-area-homes-1001.html
Sure, it makes a big difference if you are talking about the typical Phoenix resident or the highly unusual person who is determined to operate without conventional AC. I have done it (using all of the techniques listed above) in a somewhat less extreme climate.
Kurt, I'm glad you are asking this question. More people need to ask out of the box questions because the answers to them (generally) lead to progress. Conventional wisdom is a poor substitute for actually thinking about a subject. Sadly, you live in an area where it probably isn't possible to do without air conditioning. That's because it is my understanding that Phoenix's nighttime temperature in summer never gets below 70 degrees F.
In dry hot arid environments that have sufficient cooling at night it is very possible to go without air conditioning. There are three things that are required if those conditions are met:
1. A whole house fan that is turned on at night after the temperature has gone down to below 70 F or so. This brings in cool air to replace the air that has gradually warmed during the day.
2. Sufficient insulation and airtightness in the house that your home acts much like a thermos bottle after the "coolth" recharging at night.
3. Ceiling fans in most rooms so that as the temperature moves gradually to the less comfortable high 70s F during the day you can still be comfortable.
Lest you think it is a pipe dream to go without AC I do without it in an area where most everybody uses AC. "BUT" I live in an area that cools to below 70F in summer at night and it is a low humidity region. Those are absolute requirements before you even attempt to go without AC in summer in a hot summer climate. However, there is a surprising high number of places in the western USA where those conditions exist. Unhappily, I don't think Phoenix is one of them.
> This brings in cool air to replace the air that has gradually warmed during the day.
I think it's clearer to say "to cool down the thermal mass". Air doesn't hold enough heat to worry about.
> Ceiling fans in most rooms so that as the temperature moves
Yes. Partly because the fan increases evaporation from skin and partly because it helps move heat into the passive thermal mass (that is gradually running out of capacity). The latter continues to be important at night (eg increasing heat removal rate from a concrete floor) - even in unoccupied rooms.
Whole house fans are much more important than ceiling fans to cool a home at night. My home has little thermal mass but is well insulated and airtight. It works great. Thermal mass can be important but it isn't necessary for all homes. Ceiling fans are just for daytime comfort, at least in my house. I have a crawl space with no thermal mass to hold heat.
>"1. A whole house fan that is turned on at night after the temperature has gone down to below 70 F or so. "
Below 70F at night? Huh?
The average daily lows are only below 70F there from about the beginning of October through half-past May. From late May through the end of September the overnight lows will usually be well above 70F.
The hard part of THAT solution is to build the guest house somewhere else, far away from the intended Phoenix 'burbs! :-)
In Phoenix there are many days in mid-summer that don't drop below 90F at night. For most of July into early August the average daily lows are in the mid-80s F. That's the average. During major heat events it might not drop below the mid 90s. See:
https://weatherspark.com/m/2460/7/Average-Weather-in-July-in-Phoenix-Arizona-United-States#Sections-Temperature
Dana, please reread my post.
OK, you were making a more generic solution for locations much cooler than Phoenix- fair enough!
Nightime ventilation works great in say, the Columbia Basin or most of Wyoming, northern UT, etc, and those would be GREAT places to build the guest house, eh? ;-)
But on a more sober note, with climate change it may become necessary to abandon some of these warmer locations before the end of century. Average overnight lows 85F are one thing, overnight lows over 100F on a 5 day heat wave are lethal. Phoenix has already seen days where the overnight lows were in the mid-90s, and those mid summer overnight lows seem to be creeping up faster than the overall rate of global warming.
Eric,
Dana had some very good points - I read your post and I'm sure he read it well. The big problem in the Phoenix area is that the night whole house fans just don't work well. Few houses have basements - most are built on slabs, multistory. The "cold water temp" for the fresh water coming into the house may be in the low 70's and that's a water line buried in the ground.
I've been in Phoenix during the early part of May (with data loggers)- before the "hot season" and the mid afternoon outside temp was already above 100F in the shade. Lots dirt and pavement which heats up and then takes quite sometime to cool down. If the outside night time temp doesn't drop below 85F, the whole house fan doesn't gain much to pull the heat out of the interior building mass. Especially when you might only reach 85F just for a couple of hours early in the morning.
I've also been in Phoenix in late September when the same hot temperatures were also occuring... and again I'd brought along some data logging equipment. It (temp) just doesn't drop as fast outside anymore..
Dennis,
Everything you are saying I know. I'm sorry you didn't get that. I said twice in my original post that Phoenix isn't a suitable candidate for that reason. What's the confusion here? Is it the fact that I commended Kurt for the question? No question is dumb and I wanted Kurt to know that.
Follow some of the examples of homes built in the middle east... and mansions from the US south. High ceilings. Let the heat rise and hang up high. Only use the ceiling fans to bring the heat down as the house cools off.
Insulate under the roof... and let the attic help moderate the temperature gain, don't just insulate the attic.
Plan on good shading for the south facing walls... awning or equivalent that can still allow the light in.. but shade the walls.
Don't place the washer and dryer inside of the building envelope you wish to keep cold. Put them in a garage or the equivalent of a separated mud room. You could easily gain one or two degrees F into the living space while you are drying clothes -- EVEN WITH a heat pump dryer. (I know, because I've been doing dryer testing and have measure the impacts -in dryer testing in Phoenix).
Use a heat pump hot water heater, it can pull the heat from the living space while providing the hot potable water you want.
I'd still design with a minisplit... when its been 100F outside for a week at a time (day and night)... eventually the heat will soak through the house unless you have a lot of mass inside the house.
You could use brick inside the house - put all the insulation on the outside of the brick and gain mass that way... certainly not common design in Phoenix. In any case insulation on the outside of the house framing does provide some thermal breaks from the heat gain into the house through the walls. But the walls are not likely to be your big problem -- unless you used 2x4 framing with little insulation.
If you can get access to a infrared camera - take a look at the heat gain patterns on the inside of a few houses that you like... then think about those heat flows and how you can solve those gains.
>when its been 100F outside for a week at a time (day and night)...
That doesn't happen (yet) even in aptly named Death Valley, but that type of weather is deadly. There have been three day runs in Death valley that never dropped below 100F, (twice this past July) but not a full week.
But that's not to say a heat wave of that magnitude in populated areas won't happen before the end of this century even if greenhouse gas emissions stopped today.
I was wondering about this in the terms of semantics. What about installing a Heat pump water heater? From what I have heard they basically do the same thing as AC but change that heat into hot water. Would this be able to lower the temp enough in a small well-insulated house so "technically" we don't have a traditional A/C but we could cool the house?
Thoughts?
I am sure everyone here knows that when they say temps in Phoenix are 115F. Those temps are recorded in the shade. In the sun, that temp would be around 130F. On a house, unless it is 100% shaded, it's dealing with the intense desert sun which adds heat to the structure.
Vehicles parked outside in the summer can see interior temps of 160F. ASU recorded a car dash reaching 192F when parked in the Phoenix summer sun. Suffice it to say the summer sun plays a role in the homes overall heat gain.
Zero solar gain would really help a home but that would mean a home without windows. We are back to Dana's idea. Just live in a cave and you might survive a Phoenix summer without AC.
Azad,
Q. "Would a HPWH be able to lower the temp enough in a small well-insulated house so technically we don't have a traditional AC but we could cool the house?"
A. Let's do the math.
A BTU is the amount of heat required to raise the temperature of one pound of water by one degree Fahrenheit. A family uses about 40 gallons of hot water a day; that's 320 pounds of water. In Phoenix, you might need to raise the temperature of the water from 70 degrees to 140 degrees; that's a delta-T of 70.
So a HPWH provides 70 * 320 = 22,400 BTU/day, or 933 BTU/h. One ton of cooling is 12,000 BTU/h, so the HPWH provides 7.7% of one ton of cooling (also known as 0.08 ton). That's not much.
So unless you use a ground loop or well, you would have to dump $10/mo or so of hot water down the drain (for a 600 sq ft well insulated building in Phoenix). Use air source AC and add PV solar if you are concerned about operating cost.