An Air Conditioner Sizing Benchmark for High Performance Homes

26 Comments Read/write comments

High-performance home

One of the most frustrating parts of my job as a Home Energy Rating (HERS) provider is dealing with the size of air conditioners installed in ENERGY STAR homes. My frustration has spilled over here in the Energy Vanguard blog several times, and the topic in all its manifestations (HVAC sizing, Manual J load calculations, the HVAC industry...) has been one of my favorites to write about. I've got an idea of how to make things better now, though. Keep reading.

Today it happened again. I was checking HERS rating files, and, as I usually do, I applied my AC sizing rule of thumb to see if the HVAC contractors treated the homes as a true high performance homes. Turns out they didn't. For the three homes in question, the air conditioning capacity was one ton for each 529 square feet, 544 sf, and 781 sf. Those numbers are not good. The first two, actually, are terrible.

In case you're new to this subject, air conditioners should be sized to meet the cooling load of the house, and bigger is NOT better. If the AC is too big, it doesn't run long enough to dehumidify well, and the constant on-and-off cycles will shorten the life of the equipment. A Manual J load calculation is how you find out what size air conditioner the house needs.

So, here's my idea, and I'm starting a campaign for it. HVAC contractors like rules of thumb, so let's take that idea and adapt it. If you're a builder, home buyer, HVAC contractor, or real estate agent, here's a way that you can do a quick analysis to see if the air conditioner is oversized: Find the conditioned floor area. Find the air conditioner's capacity in tons. Divide the former by the latter. What you get is the number of square feet of conditioned floor area per ton of AC capacity.

I told you above that 529, 544, and 781 aren't good, so what is a good number? As a rough guide, you can use the following:

High Performance Home = 1000 sf/ton or more

We've done a lot of load calculations here at Energy Vanguard, and I don't think we've had a single new home come in lower than 1000 sf/ton. I built a house here in Georgia that came in at about 2000 square feet per ton, and we recently worked on another one that was higher than that. This number, 1000 sf/ton, should be your baseline. Nothing lower than that should be acceptable.

Now before you get all in a huff and accuse me of being a hypocrite for promoting a rule of thumb, let me explain that I'm not saying the rule of thumb above is a subsitute for Manual J. Yeah, we could come up with a number for each location (e.g., 1100 sf/ton in Dallas, 24,000 sf/ton in Seattle, 0.00000000000000000000000001 sf/ton in Hell...), but you still have to do a room-by-room Manual J load calculation to select equipment and design ducts properly.

Where this rule of thumb could really shine is for these three things:

1. Helping home builders know if their HVAC contractor is in the ballpark of proper HVAC sizing

If a home builder is taking care of the building envelope, getting the ducts sealed, and doing all the things you have to do to build a high performance home, they can use this as a guide in negotiating with their HVAC contractor. When the HVAC contractor tells the builder what they want to install, the builder can quickly see how it compares to our benchmark. (Of course, if the builder uses a third party for HVAC design, then they don't need to use this benchmark because they'll have a better number.)

2. Helping home buyers know if they might be buying a home with an oversized air conditioner

Home buyers can get their calculators out and do this simple calculation. If they're buying what's supposed to be an energy efficient home and the number is far below 1000 sf/ton, then either the house isn't really so efficient or the system is oversized. In either case, they can steer clear of it and look for a real high performance home.

3. Simplifying programs like ENERGY STAR new homes

The ENERGY STAR new homes program has been great for moving the whole new homes market toward greater energy efficiency. If you've taken a look at the new Version 3 guidelines, however, you probably felt a bit woozy afterward. They've made the program too complex, especially for HVAC. A rule of thumb like this could reduce everyones' headaches.

Size Matters

You know what's really good about this new rule of thumb? It satisfies our natural urge to believe that bigger is better! No one wants to say to their neighbor:

"Hey, Bob, I bet mine's smaller than yours."

But they will say:

"Dude, you got only 529 square feet per ton. I've got 1500! BwaaHaaaHaaaa"

Because psychologically, size matters and bigger is better. So let's give 'em a way claim the numerical high ground with this new benchmark. (This use of the reciprocal function is a form of mathematical aikido.)

Now we just need a catchy name, a jingle, and some good PR for our new benchmark. Got any ideas?

Comments

boone guyton

What about dual stage air to air heat pump systems? I know they work better for heating does it also work the same way for AC? At first stage it is basically working at half load.

Allison Bailes

Boone G.: Yes, there are some types of air conditioners that overcome the problems of oversizing. Dual stage, variable speed, and variable refrigerant flow are all technologies that help. Not that many houses have them, though, because they're more expensive. It's a rare builder who'll put one of those in a spec home because of the cost.

Dan Sunderland

Allison, again you really are on the right track. I am absolutely amazed (and not really) how resistant the HVAC industry to this type of thinking! When we are more interested in "things that work" than our standards of practices, we will ALL be better off! 
Dan @EC&D;, Inc.

Allison Bailes

Dan: There are some good HVAC contractors out there who get it. Sadly, I see the work of many who don't. I think this new benchmark will help empower builders and home buyers so that we can start to move the industry.

Christopher Cadwell

This is absolutely the right track. 
 
I have brought up this idea of simplifying how homes are rated, by just simply using a benchmark of nominal AC sizing. I have even mentioned it against experts, who even on a cursory thought said it was a good idea. 
 
I have told mechanical engineers that it is their job to design the shell around the AC system. But really it takes a day long meeting with the contractor, subs, and architects, with the engineer doing Man J, D, S while designs are brought up and discussed.  
 
The idea of this exact sizing benchmark was not my own, and was continually mentioned to me by my original trainer. 
 
What I realized is that by concentrating on nominal AC sizing as a benchmark for even rating homes, it simplified the entire process, and quality assurance was now in the hands of the builder, who has a comfortable home or not. That is to say that it would be easy to rate a home just by verifying the nominal installed AC sizing. 
 
1000 sq ft a ton is a benchmark for our extremely hot desert climates as well. I would like to modify that to 1000 sq ft per "nominal ton."  
 
To add to that I have done some science this summer on AC sizing here in summer design temperatures of plus 106 degrees. I have been able to install up to 700 sq ft per ton on just a typical house, typical windows, typical insulation, and poor ducts in the attic. I was able to hold 79 degrees in the house with it. If I can do that on a house with no real energy features, that should tell you something about houses with better windows, etc. 
 
One thing to note about SEER is that there is an underlying assumption of sizing attached to it. Think about that. Also that some studies have been done and found that SEER in no way was a predictor of energy results, and this was because micro-climate had more affect on the systems than the "boxes" could overcome. The studies also did not factor in sizing nuances to my knowledge. 
 
My prediction is that this type of bench-marking needs to happen to simplify the industry. My proposition is that the sizing benchmarks should be micro-climate specific. 1000 sq ft a ton is the starting point, because if I can do that in my climate where we commonly see 114 days, "where else is worse?" - DEATH VALLEY 
 
Once we get to that level then we can worry about staging and modulation, that will then have their place, as an improvement on the above.

Allison Bailes

Chris C.: Thanks. Yes, we need to make it simpler. What I like best about this, though, is that it empowers builders and home buyers.

Christopher Cadwell

"What I like best about this, though, is that it empowers builders and home buyers."  
 
Yes - that is probably more important than all the technical jive. 
 
Since home buyers buy on emotion and justify on logic, the home builders just mirror that because they are in tune with their customers. 
 
So there needs to be the emotional link and if it is "oneupsmanship" by this manner then so be it. 
 
"Dude, you got only 529 square feet per ton. I've got 1500! BwaaHaaaHaaaa" - that idea may be a key to propelling the industry into the needed change. 
 
I have wanted to create a "super efficient" certification or even brand. This kind of stamp could go miles toward a display of emotional value, that may be compelling to buyers emotions. 
 
Just some ideas.

John Nicholas

Allison, 
 
Great thought! Don't think smaller AC for you motto or jingle. Think More for Money! or More in your pocket! 
 
Also, what about a rule of thumb for those of us in heating climates?

PJ

Mr. Bailes, 
 
As I am sure you know, contractors insert values(Like ACH, System Loses) into their Manual J calculations that give them the answers they are comfortable with. Tell me what answer you want I'll give it to you. If you want 1000 sq/ft per ton, No problem, I land my calculation right on it. 
 
For old houses, no one really has a clue what many of these values are until tests have been run. 
 
What real value is the calculation if I don't know the inputs and many of them like ACH can have wide tolerances (.35-1.5)? 
 
Also, the Manual J is a pretty crude calculation and tends to focus our minds on one number that occurs at a worst case rather than on the fact that loads change every minute of day from no load to a max condition. Should we not shift our focus from what happens 1% to 2% of the time to how the loads change. Even if your Manual J loads are perfect, your system is too big 97.5% of the time. 
 
Thanks, 
Pj 
 
 
 
 
 

Allison Bailes

John N.: Oversized heating systems don't have the same drawbacks as oversized AC's do. It's still better to get them sized correctly, but there's no general rule of thumb there that works. The problem is that heating design temperatures are all over the map. Cooling design temperatures are in a much narrower range, which is how I was able to pick a number (1000 sf/ton) that can act as a baseline. 
 
PJ: Great points. For the benchmark I'm proposing, I'm focusing on new homes, though, so the variation should be less. I'm also not saying the HVAC contractor should use this to do their sizing. It's mainly to give builders and home buyers a way to evaluate the systems HVAC contractors propose or install.

Trish Holder

Nice job breaking this down for builders and homeowners -- most of whom have the mechanical aptitude of a sandwich.

Allison Bailes

Trish H.: Thanks. The good thing is that they don't have to have mechanical aptitude. They just need to have a basis for knowing how to judge if a system is oversized or not.

Andrew J. Courts Jr.

Good piece. Nailed ENERGY STAR's overkill on V3. Likely the death of program.

Allison Bailes

Andrew: Thanks! I hope V3 isn't the death of ENERGY STAR new homes. Time will tell.

Christopher Cadwell

"A very powerful AC system, suped up for homes; able maintain comfort at half the watts of the competition." = getting something more - not taking something away. 
 
Kind of like a Tesla car, super fast like a muscle car, only it is electric. Who wouldn't want that? 
 
I predict that the Prius will soon have a little sports car counterpart, just for the new generation, who wants a suped up hybrid. 
 
Just some ideas. 
 
I would like to further these ideas into a think tank, of opinion leaders and see if we can do some market testing with phrases as above.

Kent Robinson

Allison, I think you have a good idea. Pj also has a good point about Man J being oversized most of the time. Wouldn't it make more sense to manage humidity with a whole house dehumidifier than trying to use the AC to dehumidfy, even if you could get your HVAC guy to do it right.

nichole l reber

2000 square feet per ton? Crikey! Especially in GA, where people think they'll die if they experience a single drop of sweat.  
Your idea to simplify programs like Energy Star's are also really good, especially in this time of overlapping green building programs. Too much information just confuses everyone, especially home buyers.  
Cheers!

David Butler

Late to the party today, but I do have a few comments on the comments... 
 
@Allison, 2-stage AC doesn't really overcome the oversizing problem. It may mitigate the moisture impact a little bit, but *only* if the high stage is sized to the load, which is rarely the case in my experience. I don't know how many times I've heard of contractors selling 2-stage by touting the high stage as "reserve power" for those extra hot days or big parties. I guess it's ok to do this, but don't then expect the system to achieve its rated SEER. Dealers never reveal that little detail. And BTW, almost the same benefit can be accomplished with an $8 relay to reduce fan speed on dh call, assuming the extra latent capacity is needed in the first place. 
 
@Chris, the SEER rating procedure does make an assumption about sizing. I don't recall the details, but it does not assume perfect sizing. 
 
@Kent, the latent capacity which comes along with AC isn't free, but it is by far the least expensive way to remove moisture. 
 
@PJ, as Allison said, the existing home issue is a bit OT, but you brought up a good point... HVAC dealers who specialize in retrofit should all own an use a blower door and duct blaster. Those that don't are just guessin' 
 
That being said, as I've oft noted in other forums, a homeowner can get an accurate estimate of the home's true load by cycle timing, for those so inclined. Service contractors should be all over this by installing data loggers in their client's homes. That way, when it's time to replace the system, there's no question about what size is needed. If an envelope retrofit is planned, it's useful (although not always possible) to wait a year before changing the HVAC, so that cycle times can be monitored post-retrofit. (Sorry, didn't mean to get sidetracked.) 
 
As for why worry about sizing so much when even a perfectly sized system is oversized 97% of the time... Well, full variable systems (VRF) get around that problem, BUT they are very expensive. As it turns out, a single-stage system sized properly has a much lower life-cycle cost and does a good job maintaining moisture levels (and comfort) in most of the country, if house is built correctly and system is properly designed. Even a VRF system won't overcome poor envelope and duct design. 
 
BTW, a system that's 25% or 30% oversized will be somewhat less efficient, and will remove less moisture, but in many cases, will work just fine. The problem we're all twisted up about is the egregious oversizing we're seeing in high performance homes, where it's not unusual to see seeing systems that are 2x or even 2.5x the true load. Here's the reason... it takes the evaporator coil 8 to 15 minutes to start dripping water. If the system never stays on that long, it will not remove ANY moisture. In humid climates (most of the US), now you got problems. Not to mention the additional energy lost during startup, especially if ducts are in the attic. 
 
What I like about the AC sizing benchmark is that it would finally force HVAC dealers and builders to collaborate on plan review, rather than "here are the plans, tell me how much". And at last, it would create value for proper system design.

bob

excellent article !!

Jim

Your moving in the right direction. As we tighten the building envelope idealy using spray foam insulation to prevent air infiltration the old rules of thumb will cause serious short cycling problems. However, ANY rule of thumb can get you in trouble if you don't consider you are cooling CUBIC FEET NOT SQUARE FEET. An 8' cieling home will require far less tonnage than a 10-12' cieling home if both homes are 1,8000sq/ft. Pretty basic stuff.

David Butler

@Jim, I think you missed the point. System would still have to be sized based on Manual J (do I hear an echo in here?) 
 
Enforcing a minimum SF-to-tonnage ratio would lead to more efficient homes as well as causing HVAC contractors to sharpen their pencils when doing MJ, exactly the opposite of what's happening now.

bob

10' and 12' ceilings don't have to add nuch to the cooling/heating load if done right. Use some of the space to house the ductwork inside conditioned space above hallways. There is only the additional 2'-4' of wall space to consider, it doesn't add much to the cooling load as the increased cubic ft would imply.

Chandler Pernell

Really like your flip on "bigger is better". We have been combatting this logic for 35 years and never considered flipping the numbers in order to perceive a properly sized air conditioner "bigger". EXCELLENT.

Timm Hamm

You mentioned "high performance homes". I agree about your sizing "rule of thumb". But what about non-high performance homes"? Homes that I go into to perform an energy audit that show air leakage 3 times the acceptable rate and duct leakage 4 times the acceptable rate?

Jerry

I understand the whole high sf/ton what hasn't been touched on here is how to circulate the air in the home. If a 3000 sf home has 3 tons of ac 3*400cfm/ton=1200cfm. This won't circulate the air sufficiently without some zoning or dampers someplace. How do you get the proper circulation?