Cold Comfort for Utilities

To err is human, but to blame it on someone else is even more human.
- Murphy's Laws

While no one has great data on Denver air conditioning installations, word has it that builders install refrigerated air in about 15 percent of new homes-less in entry-level housing, more in move-up homes. Yet four years after move-in, more than 50 percent of homeowners are equipped with AC.

Why the big differential'? Well, homebuyers have learned that production builders tend to price AC higher than the after-market installers, so builders don't profit from selling and installing AC. But they also don't get the headaches, either.

The Fort Collins Housing Performance Study (long overdue; will be in print this September) illustrates that residential air conditioners indeed have their headaches (see "A 'No-Cooling' Package," the December 2000 issue). The quick solution seems easy: install a unit that's larger than it needs to be. As one installer put it, "I've never been sued for putting in too large a unit." But the quick solution is causing a plethora of problems. Consumers pay more now for the installations. Eventually, they are likely to pay a lot more to operate them. It's time for a change here.

Oversizing mania

More than 65 houses in the Fort Collins study were equipped with AC. Equipment-sizing calculations were subsequently performed by the utility on those homes. Whereas ASHRAE engineering guidelines indicate that AC equipment should be sized to meet 115 percent of the cooling load at the design cooling temperature (91 degrees F in Fort Collins), the actual units were sized at an average of 208 percent of the cooling load. That's nearly twice what should have been required.

Here's the short list of what happens when you oversize equipment. First, you typically have to oversize the furnace, too, in order to handle the higher airflow level required for the AC; often, the after-market contractor comes up short here. Second, the equipment tends to short-cycle, which provides less comfort, less dehumidification during the dog days of July, and can lead to higher cooling bills. Third, the equipment may not last as long. Fourth, the client pays an extra $300 to $500 for the installation. Finally, the utility is put in a bind.

Why is oversizing so commonplace'? Three reasons: 1) almost no one is performing equipment-sizing calculations; 2) almost no one is carefully designing ductwork to deliver cooling to the second floor in taller homes; and 3) most ductwork is phenomenally leaky. In fact, we have the leakiest ductwork in the nation. Given this situation, adding an extra ton or two is the round peg that is forced into the square hole.

The leaky ductwork item is a killer. If leaky ducts can't deliver enough air during the heating season, you can often live with the result because warm air rises. But during the summer, the result is a freeze-fry Goldilocks house: 78 degrees F upstairs, 73 degrees F on the main floor and about 60 degrees in the basement (cool air sinks).

The utility bind

Oversized residential AC equipment is also a utility killer in Colorado. In many homes, AC doesn't operate very often just a few weeks of the summer. But it is always turned on during the hottest stretches. More than in most parts of the country, that means utilities have to expand their power plant capacity just to meet a sharp spike in customer demand. That's like having to keep a Ferrari in the garage for a once-a-week trip to the grocery store; it's an awfully expensive way to get the job done.

Instead of building the new plants themselves, utilities can let independent power producers build those new plants. Then the local utilities buy that new power on the open market. But that power is always most expensive during the hottest summer days.

Either way, many Colorado utilities probably lose money when their residential customers turn on their air conditioners. In one utility's jurisdiction, last summer it cost $160 to supply power to the average residential AC customer, while the current rate structure only brought in $60 of revenue from that average AC customer.

Utility options

In the long term, utilities nationwide will have to change the way they bill their residential customers. Those customers who use the most power at periods of peak demand must eventually pay a higher rate. (Customers with oversized AC will pay even more.) To do this, utilities will need to install new meters that measure when and at what rate electricity is consumed, not just how much is consumed. Such an effort will cost hundreds of millions of dollars in Colorado alone. It is already starting in California markets.

In the mid term, utilities need to send a signal to builders to change several key construction practices in the residential sector. Such changes, described below, could prevent utilities from having to build so many power plants. To get those changes, utilities will probably have to pay builders some incentive money, at least for a year or so. In the short term, Xcel Energy is already offering two incentives. First, they are providing rebates to customers who buy higher efficiency AC equipment (12 SEER vs. the conventional 10 SEER). This is a very incomplete approach. Second, Xcel and a few others are planning to pay residential customers $25 for the right to slightly control when their AC units can turn on (15 minutes on, then 15 minutes off). This is a band-aid with its own share of equity problems.

Four keys

What should Colorado builders do to improve AC performance and how should Colorado utilities help?

First, they should reward builders who reduce future cooling requirements within each new building shell. Since windows let in half a home's summertime heat gains, focus on windows by paying builders a small sum to upgrade with low-e models. Limit the incentive to windows that let in only about half as much summertime solar gains as the average window. That means picking windows with a solar heat gain coefficient (SHGC) of 35 percent or less, compared to 65 percent or so for the normal double-glazed window.

Second, toss out the age-old "rules of thumb" for sizing equipment and ductwork. Require a design layout and cooling load calculation for every new home-even for those homes sold initially without AC. And as Hank Rutkowski has repeatedly stated (see "The guy who wrote the sizing book," October 2000 issue), those calculations should be performed using accurate design temperatures and "aggressive assumptions." It takes less than an hour to perform sizing calculations for both the equipment and the ductwork itself ... for homeowners who will live with the consequences for decades. For a production builder, changing any particular model to reflect a different orientation to the sun takes just minutes.

Third, give incentives to builders who have their installers seal their ductwork with mastic, caulk or aeroseal (not "temporary tape"). When all the supply and return-air grilles are sealed off, 70 percent to 80 percent of the HVAC equipment's normal airflow can leak out of the average Colorado home's ductwork today. The new national standard is no more than 10 percent leakage, meaning our average ductwork is about seven times leakier than it should be. Early results from the field indicate that the best way to achieve this is to totally eliminate use of building cavities as return-air chases.

Fourth, whenever residential AC is installed, provide incentives for builders to "commission" their systems. A commissioned AC system is one that has been tested for airflow and refrigerant charge. Right now, no one is checking. Studies from California show that AC units with insufficient airflow and the wrong refrigerant charge can experience a major loss in performance.

Since Xcel is currently devoting several million dollars to reduce residential air conditioning loads, the above suggestions aren't pure fantasy. Xcel could put a time limit on the incentives they offer builders to "do it right the first time." But even if utilities don't directly support the above efforts, builders should consider heading in that direction anyway, and sooner rather than later.

Steve Andrews consults with builders for E-Star Colorado and writes on energy issues (sbandrews@att.net). E-Star (www.e-star.com), is a nonprofit home energy rating system that works with both new and existing homes statewide.

2008 Built Green Colorado

Home Builders Association of Metro Denver, 9033 E. Easter Place, Suite 200, Centennial, CO 80112
(303) 778-1400 fax: (303) 733-9440  info@builtgreen.org

Last Updated: 10/05/2007