Why an Oversized Air Conditioner Is a Bad Idea
Apparently, I haven’t written a single article this summer about oversized air conditioners. Shame on me! I’ve written several in previous summers, mostly about how HVAC contractors bypass the HVAC design process by using rules of thumb and how they mess up the Manual J load calculations when they do go the preferred route. I’ve never written an article focusing on the reasons to size an air conditioner properly, though, so here it is.
Reason #1 – Comfort
When an air conditioner runs, it does two jobs. It lowers the temperature of the air, and it removes moisture from the air. To do the second one, the AC has to run for a while.
As the air passes over the evaporator coil, it encounters a very cold surface. Ideally, when the air passes over that cold surface, the air temperature drops about 20° F. In places where we have higher relative humidity in the summer, the other important process that happens when the air hits the coil is that the temperature of the coil is below the dew point of the air. As a result, water vapor condenses on the coil.
Here’s the thing, though. Water vapor condenses on coils in oversized air conditioners, too, but you have to get enough condensation on the coil for the water to start dripping down into the pan below the coil. Even then, you’re not there yet. You still have to get enough water in the pan for it drain to the outside.
Until the water that condenses actually makes it to the outside, you haven’t really dehumidified the air. Why? Because that water on the coil can evaporate and get back into the air in the home. If you make the mistake of leaving your thermostat in the Fan-On position instead of Auto (Don’t do that!), the water on the coil gets back into the air even quicker.
Oversized air conditioners don’t run for a long time because they satisfy the cooling load quickly and then shut off. Properly sized air conditioners run longer, so if you want your air conditioner to dehumidify your home as well as cool it, don’t let the HVAC contractor oversize it.
Reason #2 – Wear and tear
If you live in the desert, what I just said above doesn’t matter to you. There’s no water vapor to condense on the coil, and if there were, your dew point is probably close to absolute zero. (Well, OK, that might be a bit of an exaggeration.) The second reason definitely should matter to you, however. The thing that wears equipment out is starting up and shutting down. The more it happens, the shorter the life of the equipment.
When an air conditioner is oversized, it starts up and shuts down a lot more because it runs for only a short time to meet the thermostat setpoint. Then a few minutes later, it comes on again and runs for a short time. Over the course of a day, an oversized air conditioner can have a lot more start-ups and shut-downs than a properly sized air conditioner. That means you’ll probably be repairing it more often and replacing it sooner.
Will you save money?
People in the industry used to think you’d save money on your air conditioning bills with a properly sized air conditioner, but that thinking has changed. John Proctor wrote an article for Home Energy Magazine that shows only a small savings for the homeowner (but the utilities benefit by lower peak loads). You should save money on the upfront cost, though, because you’re putting in a smaller AC. (As Proctor points out in his article, oversizing isn’t always the worst problem, though. You have to be a subscriber to Home Energy Magazine to read the article, but it’s well worth the read if you can get it.)
How do you determine the proper size?
For new homes, use the full HVAC design process, which starts with Manual J, the heating and cooling load calculation protocol from the Air Conditioning Contractors of America, the trade association for AC contractors. For existing homes, the best way to do it is to see how long your current air conditioner runs when it’s at the design conditions. Of course, you want to do this while your AC is still in good shape, not when it’s on its last legs. If it runs for only 5 to 10 minutes before shutting off, it’s definitely oversized. If your runtimes are over half an hour at design conditions, it’s probably sized close to the actual cooling load.
If you’re building a new home, remodeling an existing home, or just getting a new air conditioner in an existing home, ask your builder, remodeler, or HVAC contractor how they’re planning to size the air conditioner. If they tell you they’re basing it on the size of your house, don’t let them do it. There’s a lot more to it than square footage of conditioned floor area, and of course, I haven’t touched at all on the capabilities of different types of equipment (single speed, fixed capacity, multi-stage, mini-splits…). More on that in future articles.
Oversized AC, Screwed-up Manual J, ENERGY STAR HVAC Tirade!
Got Manual J? Don’t Assume It’s Correct.
We Are the 99% — Design Temperatures & Oversized HVAC Systems
This Post Has 18 Comments
Heat Pumps in Northern climates are often oversized to reduce running backup heat and will save money.
Most, two stage units only come in One ton increments so many times you need to move up in tonnage a half a ton.
The laws of Heat transfer have been around since Fred Flintstone. ACCA simply dumbed them down, removed the fact that loads vary over time, and trademarked the term Manual J for financial reasons. I would argue any “procedure” that focuses on a condition that will occur less than 1% of the time is counter productive to comfort and silly on the face of it.
On old houses, without extensive testing, contractors have no clue, and are simply guessing at the input values are for infiltration, duct leakage, and all the other values, so with ten contractors you will get eleven loads and they will all go to their grave telling you they are right. The truth is contractors learn quickly what inputs give them the loads they like, then they are right.
What great science when every answer is right!
See ya next week,
Pj: I had
Pj: I had a feeling I’d see a comment from you on this one. As I mentioned at the end of the article, I didn’t want to get into all the different equipment types here, but yes, heat pumps in cold climates need to be oversized to meet the heating load without running too much strip heat. Actually, though, ACCA does acknowledge that loads change, and if you read Manual J, you’ll find that it talks about part-load, design-load, and extreme-load conditions and why we should design for design loads.
Allison — A few comments —
Allison — A few comments — Air Conditioners in “the desert” — that is most of California, Nevada, Arizona, Colorado, West Texas, etc. do have moisture in the indoor air and do produce condensate and do dehumidification — even though it is a waste. Smaller ACs will make it easier to get the higher CFM per ton you need to reduce the dehumidification in favor of sensible cooling. In addition in dry climates the fan should remain on after the compressor shuts off for an amount of time proportional to the amount of water on the coil.
The most efficient AC is an oversized two speed unit with the high speed disabled.
I have had homebuyers tell me
I have had homebuyers tell me, more or less, “If I want my house to cool to 68 degrees, I want it to cool to 68 degrees. I don’t care about the outside temperature and all of that design BS.!’
How do HVAC contractors deal with customers like that? Right now, they are dealing with it by fudging their sizing.
“People in the industry
“People in the industry used to think you’d save money on your air conditioning bills with a properly sized air conditioner”
-This is true in my experience with every job that I have done it, even to the exclusion of other energy improvements. I believe that saying downsizing does not save energy is False information. Take it from someone who actively does it. I have several case studies, that it does save energy in a harsh desert climate.
to Chris C. When experiments
to Chris C. When experiments have been done with the same house with both the original and final ACs being the same efficiency, same duct system, etc. The results have been mixed. Nothing overwhelming. On the other hand when Chris Cadwell puts in a new AC, he is sure the new smaller one is properly charged and probably has higher CFM per ton than the old one – a great advantage in the dry climate. Am I right Chris? I also believe you often keep the bigger indoor coil — which is another advantage.
How does the 20°F
How does the 20°F temperature drop work exactly? I assume this is supposed to be set up during the install. Say two identical AC’s roll off the assembly line, one ships to Miami and the other to Phoenix – totally different design temps. How does each unit get dialed in to achieve the 20° F temp drop and proper dehumidification needed in the install climate? I’d assume there would be a lot of issues with doing this accurately. Someone please explain or point me to an article on this.
Great comment, PJ. When it
Great comment, PJ. When it comes down to it, there are ultimately only 5 or 6 sizes to choose from. As a wise man once said: “Why take a long cut to a guess when you can take a short cut to a guess?” And I firmly believe this is the mentality of most contractors. I am also a firm believer in good design, but I think that expecting contractors and programs to use an unweildy and non-transparent means of getting there will continue to get us the results we are seeing now.
Allison, as always a straightforward, entertaining, and well crafted piece.
@ Thomas B. The 20 degree F
@ Thomas B. The 20 degree F temperature drop is an incorrect rule of thumb. The temperature drop changes all the time depending on the indoor and outdoor conditions. More importantly, the airflow is what needs to be done well AND MEASURED. The target temperature split is a little better than the 20 degree rule of thumb, but measuring the airflow is the right way to do it.
See the following California Energy Commission web info on measuring methods as well as the less effective, but useful temperature split targets. http://www.energy.ca.gov/2008publications/CEC-400-2008-004/CEC-400-2008-004-CMF.PDF Look at Table RA3.2-3 and Section 3.3. for technical information.
Thanks Alison, <
On high volume, production RNC, we never know what direction a house will face. There are no North arrows on the plans, therefore we assume worst case to CYA. All houses of that design, get the same AC, same duct system, and same heater. Builders won’t even let us downsize AC’s if we know the orientation as two homeowners would start a fight over why they had the same house but different size units. In southern climates we have to oversize heaters to get the CFM for the ACs. So if are Manual J’s are perfect, the AC system is oversized at least 75% of the time, remembering that even the “properly sized units” were oversized 99% of the time, our duct systems are wrong 75% of the time, and our Furnace is oversized 100% of the time.
On replacements, Thomas says it best. Guessing at unknowns to make loads come out to what you want is no different that just guessing the size. To bad the focus is on a load that may happen for only seconds a year, rather than the constantly changing nature of reality. By taking the time variable out of the model in order to dumb down the math to a 3rd grade level, we have lost focus on both comfort and efficiency. Our most comfortable and efficient systems are those that modulate capacity to constantly changing conditions, not simply turning off and on at full tilt.
But the old load calc stick can be fun to watch, and listening to all the guys tell you their loads are right is priceless.
Region matters a lot (both
Region matters a lot (both peak temperature and dehumidification need). Oversized single-speed very bad in SE, but oversized 2-speed with variable speed blower might be wonderful there – huge evaporator for mild conditions (high CFM/ton, too), but ability to increase latent work with high compressor/low fan. The big Q. is how to get consumers to value professional design, selection, and installation, to differentiate the “product” as including important services.
We’re going through A/C
We’re going through A/C replacement at our house. Downsizing from a 3 ton to a 2 ton in a 1600sqft Oklahoma home. 800sqft per ton, but expect power bills and humidity control to improve.
@John P: we can’t always win
@John P: we can’t always win the battle to keep hvac out of the attic. Don’t you agree that cycle losses in this case are significantly greater, especially in hot climates? When attic is scorching hot, insulation can only do so much to keep the air handler and duct components from getting warm between cycles. At the start of each cycle, the system dumps hot air into the house, and all the heat absorbed by the metal components eventually transfers into the return and supply air. The larger the system, the more times this happens over the course of a summer.
@Allison: as you noted, with heat pumps there’s an offsetting benefit to sizing a bit larger. But it’s important to qualify that. Energy Star allows up to 25% oversize based on cooling load in certain climate zones, and I sometimes go even further. But I don’t think anyone would argue that a system that’s double the design load or worse is a non-issue. Unfortunately, this is what often happens in high performance homes when the contractor doesn’t have any idea what the actual design load is.
@Bill: I always discuss
@Bill: I always discuss temperature settings with clients during the design phase. I recall one lady who said she sets her stat at 68 in summer. Digging further, I learned her 100-year old home had single-pane glass and little insulation. I asked her to buy a thermometer to monitor the areas where she spends most of her time (being careful to avoid lamps, sun and suppy registers). Turns out the e-m stat was off 2 degrees and there was an additional 4 degree swing between stat location and the living areas. After moving into her new home, the client happily reports being very comfortable at 75F. I don’t think many people would be comfortable at 68F.
@ David Butler Lets start
@ David Butler Lets start with what we seem to agree on. Attics are a bad place for ducts, particularly if they are poorly insulated and long (large surface area).
The cycling thought cuts both ways. The smaller unit’s compressor-on cycle is longer than for a larger unit. During that longer on cycle the temperature difference between the supply air and the attic is at its greatest, thus the highest heat loss per minute. The off-cycle in both cases is controlled by the thermostat deadband and the heat gain of the structure, therefore it is darn near the same length of time. During the off cycle, the heat gain is limited by the volume of air in the duct. There are less total cycles with the smaller unit, there is more time spent at the higher heat gain. Making the smaller unit have a higher conduction heat gain than the larger unit — (all else being equal eg. duct surface area).
great blog, thank you and
great blog, thank you and keep it up!
Just a comment, but even in
Just a comment, but even in desert-like climates in the southwestern US, condensation can be an issue on compressors and refrigerant lines, so make certain yours are all insulated properly. Even very low humidity in the summer can result in mold growth, especially if your A/C unit is hit by sprinklers, which many are.
Great article. And make sure
Great article. And make sure you keep that evaporator coil clean.
Comments are closed.