My grandfather, William L. Holladay, was a refrigeration and cooling engineer. Decades ago, he wrote a pioneering, speculative article on ground-source heat pumps, “The Heat Pump: What it does, and what it may do someday.” The article appeared in the October 1948 issue of Engineering and Science Monthly. (For a basic explanation of how a heat pump works, and the difference between an air-source heat pump and a ground-source heat pump, see Heat Pumps.)
Even back in 1948, my grandfather realized that the Achilles’ heel of ground-source heat pumps was their high cost. He wrote, “To use earth heat, a hole must be dug, and the cost, while not always predictable, will surely be high: maybe from 25 to 50 per cent of the entire project.”
In the six decades since my grandfather’s article was published, engineers have not given up on ground-source heat pumps (GSHPs). Among those who remain enthusiastic about these systems are a group of true believers who are convinced that (a) this is a wonderful technology, and (b) there must be a way to bring costs down.
Ground-source heat pumps are still expensive. One fan of the technology, Brian Clark Howard, has provided the following guideline to system costs: “In our book Geothermal HVAC, my co-author, Jay Egg, crunches the numbers for a typical homeowner, based on his 20+ years in the business. For a home geothermal system, he estimates the total installed cost at $42,000.”
This estimate is similar to those made by two installers of ground-source heat pumps from Maine, Jeff Gagnon and Jim Godbout. In an episode of the Green Architects’ Lounge series on the GBA website, Gagnon estimated the cost of a residential GSHP system (including the cost of a drilled well) at $30,000 to $42,000, while Godbout gave an estimate of $40,000 to $50,000.
Blame the backhoe operators and well drillers?
Back in 1995, Steve Kavanaugh, a professor of mechanical engineering at the University of Alabama and a nationally known expert on GSHPs, wrote a paper called “Cost Containment for Ground-Source Heat Pumps”. In that paper, Kavanaugh noted, “High installation costs have been identified as a major barrier to wider application of ground-source heat pumps (GSHPs), often referred to as geothermal heat pumps. The primary reason cited for higher cost is the ground loop.”
After looking into ground loop costs, however, Kavanaugh disagreed with industry experts who blamed backhoe operators and well drillers for the high cost of GSHPs. Kavanaugh wrote, “This study indicates the labor, overhead, and profit in horizontal loop installation [are] actually lower than the prevailing cost of equivalent activities in the construction industry.”
Since 1995, the cost of GSHP systems has gone up sharply. When Kavanaugh recently analyzed the reasons for the skyrocketing costs, he again found that the main culprit for high system costs is not the cost of the ground loop; it’s the high cost of the equipment installed in customers’ basements.
Equipment costs are rising faster than ground loop costs
In a 2011 study, Kavanaugh analyzed the cost data for of forty GSHP systems. The results of his research were published in the October 2012 issues of ASHRAE Journal, in an article titled “Long-Term Commercial GSHP Performance.”
Kavanaugh and his fellow researchers concluded that, since his original 1995 survey of GSHP systems, the cost of the HVAC components for GSHP systems has increased by 177%, while ground loop costs have only increased by 52%. Kavanaugh wrote, “The percentage of ground loop costs to total GSHP system cost have declined from 38.5% in 1995 … to 25.5% in 2011. …Thus, attempts to optimize GSHP cost by focusing primarily on the ground loop seemed illogical.”
Kavanaugh was recently interviewed by Energy Design Update, a monthly newsletter. In that February 2013 interview, Kavanaugh said, “My conclusion — which is the opposite of the industry, as the industry is calling for the cost of the loop to come down — is that, considering inflation, the cost of the loop has come down in areas with mature, competitive markets, and can’t go much further. Equipment costs, meanwhile are going through the ceiling.”
Inflated efficiency claims
Kavanaugh has long been concerned about the exaggerated energy efficiency claims made by GSHP manufacturers. When I first interviewed Kavanaugh on the topic back in 2007, he decried competing efficiency claims by GSHP manufacturers and air-source heat pump manufacturers as “a lying contest.”
Five years later, Kavanaugh’s opinions on the issue haven’t changed. In the February 2013 EDU interview, Kavanaugh said, “I believe that the GSHP industry has created a deceptive rating system to counter the deceptive air-source heat pump rating system. I feel it really hurts the industry, as it promises unrealistic energy efficiency numbers when, in fact, they are less efficient units that what the ratings suggest.”
The problem dates back to 2000. On January 1, 2000, a new standard (ISO standard 13256-1) replaced two earlier standards (ARI 325 and ARI 330) used to rate the efficiency of GSHPs. Whereas the ARI standards included an allowance for the electricity used by pumps to draw well water or circulate fluid through ground loops, the new ISO standard eliminated all pumping energy from COP calculations. The effect was that COP ratings jumped up: for the same piece of equipment, the COP rating under the new ISO standard was higher than the COP rating under the old ARI standards.
For consumers, the advantage of the old ARI method was that it at least made a stab at including pump and fan energy — unlike the current ISO method, which simply abandons any attempt to account for these essential energy inputs.
Dan Ellis is the president of ClimateMaster, a GSHP manufacturer. Back in 2007, Ellis defended the new ISO standard. “The rating should be used to compare one box to another,” Ellis told me. “In the residential market, no one is going straight from the rating and saying, ‘That is what you will get in your real house.’”
In spite of Ellis’s assertion, however, GSHP manufacturers often blur the line between unit COP and system COP. For example, this is how another GSHP manufacturer, WaterFurnace, boasts about a heat pump with a COP of 5: “A WaterFurnace geothermal system … can deliver an astounding five units of energy for every one unit of electrical energy used.” Since this calculation ignores pump and fan energy, the statement is false.
In the February 2013 EDU interview, Kavanaugh said, “To calculate performance on these multi- and variable-capacity models, the standard [ISO Standard 13256-1] calls for water temperature in the loop to be 68°F, which is ridiculous, because loops operate at much higher temperatures in cooling. Essentially, what you have there is a something similar to rating the efficiency of a car or truck … when it’s rolling down the hill. If the evaporator coil is 80.6°F and the water coil (condenser) is 68°F, you can get a ridiculously high efficiency reading. On top of that, these calculations assume that the fan has no static pressure. … When you take that piece out of the rating, you get a very deceptive, high efficiency rating.”
[Author’s postscript: Tony Cooper, a distributor of GSHP equipment in Frisco, Texas, posted a blog in response to this article. Cooper asserts that any discussion of inflated efficiency claims “is absolutely irrelevant” because “homeowners don’t know or care what SEER, EER, or COP mean.”]
Most rating methods are flawed
Of course, all efficiency rating systems are subject to criticism: for example, the AFUE of gas furnaces ignores electricity use, while SEER ratings overstate the performance of air conditioners during humid weather. Similarly, the heating season performance factor (HSPF) does not accurately predict the installed performance of air-source heat-pumps.
Kavanaugh notes, “In defense of water-source heat pump manufacturers’ desire to use this deceptive rating system, it is important to understand the system used by competing air-source heat pumps. In cooling, the SEER, if it’s a single-speed machine, will be determined when the inside air temperature is at 80°F and the outside air is 82°F. With these conditions, you honestly might as well open your windows. … Equipment manufacturers are playing games, getting really high COP and SEER, which all are really inflated. … Owners have to be really careful when someone quotes them COP or EER. Ratings have huge gaps in them, and don’t include pump operation, either. Some designers and installers do sloppy work and end up putting huge pumps on to compensate for poor piping plans. The big pumps take a huge chunk out of potential efficiency.”
Equipment vendors always want to sell more equipment
The aims of manufacturers and vendors of ground-source heat pumps may differ from the aims of a builder or homeowner. “Ground-source heat pump manufacturers often sell pump kits with two pumps when only one pump is needed,” Kavanaugh told me when I interviewed him recently. “For example, consider the use of 3/4-inch U-tubes [for vertical ground loops] for a 4-ton heat pump with which manufacturers typically provide a kit with two pumps. What you need is 1-inch U-tubes and upsized headers from the size (1-1/4 inch) that the GHP vendors often recommend to one that provides less head loss (1-1/2 inch). Then you only need one pump. This will result in an 8 to 10% higher system efficiency and the savings on the pump will more than offset the cost of the larger U-tubes and headers. But the vendors prefer to sell two pumps, because manufacturers always like to sell more equipment. Installers tend to buy into the vendors’ recommendations about the size of the ground loop. These loops are typically smaller and shorter than they should be. It’s important that the ground loop be correctly sized. But when costs become too high for the client, some manufacturers and vendors would rather see a larger fraction of the money go to equipment and a smaller fraction to the ground loop — because they aren’t making any money on the ground loop.”
Clearly, it doesn’t make much sense for a contractor designing a ground loop to rely on advice from an equipment vendor. Instead, Kavanaugh advises system designers to use LoopLink software from GeoConnections, Inc.
Why is there such a dearth of field data?
Many residential energy experts — including Marc Rosenbaum, Andy Shapiro, and Henry Gifford — have bemoaned the lack of performance data on installed residential GSHP systems. When I spoke to Kavanaugh recently, he acknowledged the problem. “What is needed is field data (and a lot of it) to verify installation quality and what actually works well,” he said. “The Department of Energy seems more interested in pie-in-the-sky type of equipment with a low probability of being affordable rather than measured field data. So that is a major source of funding that is not available. We tried to get a research project funded through ASHRAE to gather field data on commercial building ground-source heat pump performance, and it was rejected three times. But I guess they felt it wasn’t research-y enough. Fortunately we were able to obtain funding through the Electric Power Research Institute, the Tennessee Valley Authority, and the Southern Company. Ironically, the publication arm of ASHRAE was much more interested than the research folks, since results of the project have been published by the ASHRAE Journal. ”
While Kavanaugh’s study of the performance of commercial GSHP installations is certainly useful, no comparable study of residential GSHP systems has yet been made.
Kavanaugh is still loyal to the technology
Kavanaugh is aware that the high cost of GSHP systems puts the technology out of reach for most homeowners. The February 2013 EDU interview quotes him as saying, “The economics in the commercial building sector are much better than those in the residential. I really have some concerns, and am disappointed for the residential sector because heat pump technology is primarily restricted to high-end applications, because of economics. … It is often not a good bargain for the average customer.”
Nevertheless, Kavanaugh remains a fan of the technology. “I can assure you, I would never have an air-based heat pump in my home,” Kavanaugh told me.
Kavanaugh still hopes that the cost of GSHP systems will eventually drop. “It is a product that has a 20% smaller compressor than an air-source heat pump,” Kavanaugh told me. “Compared to a split system, it should have half the sheet metal, half the aluminum, and half the copper in it. It should have less than half the controls in it. If the product were widely used, if it were common, it would cost less than a split-system air conditioner/heat pump. But the current situation is that that device costs more than a split system unit, and it should not. … If you could install the system inside the building, with the pump and ductwork, for the equivalent of an air-source system, the economics would be better. The only premium would be for the ground loop.”
These systems don’t make sense for homes
If you’ve read this far, you probably realize that the answer to the question posed at the top of the page — are affordable ground-source heat pumps on the horizon? — is no. When it comes to residential heating and cooling loads, the trend is toward smaller loads (due to improved building envelopes with reduced air leakage, thicker insulation, and better windows), not larger loads. A home with a good thermal envelope doesn’t need much heating or cooling, and it certainly doesn’t need a $42,000 HVAC system.
For that matter, it probably doesn’t even need a $28,000 HVAC system. Now that a few ductless minisplit units can heat or cool your home with an average COP of 2.0 or more, residential ground-source heat pumps, like solar thermal systems, just don’t make any sense.
Martin Holladay’s previous blog: “Ventilation Rates and Human Health.”