Assessment of Hybrid Geothermal Heat Pump Systems ...infohouse.p2ric.org/ref/43/42893.pdf3 either desuperheaters on the heat pumps or, where large amounts of hot water are needed,

F E D E R A L E N E R G Y M A N A G E M E N T P R O G R A M

Internet: http://www.eren.doe.gov/femp

TechnologyInstallationReview

A case study onenergy-efficienttechnologies

Prepared by theNew TechnologyDemonstrationProgram

No portion of this publi-cation may be alteredin any form without priorwritten consent from theU.S. Department of Energyand the authoring nationallaboratory.

both residential and commercial/institutionalbuildings. However, GHPs often havehigher first costs than conventional systemsmaking short-term economics unattractive.This disadvantage can be magnified in commercialbuildings, many of which have much larger coolingneeds than heating needs, especially for buildingslocated in climates typical of the southern UnitedStates. For GHP systems using closed-loop verticalground heat exchangers, this load imbalance can resultin a ground temperature increase over time causingsystem performance deterioration. Increasing the sizeof the ground heat exchanger or increasing the distancebetween adjacent heat exchanger boreholes can post-pone the temperature increase problem but will alsoresult in higher system cost. An alternative, lower costapproach for such applications can be use of a hybridGHP design. In hybrid GHPs, the ground heatexchanger size is reduced and an auxiliary heat rejecter(e.g., a cooling tower or some other option) is used tohandle the excess heat rejection loads during buildingcooling operation. The extent to which the ground heatexchanger size can be reduced in a hybrid GHP system

will vary with location and climate, but itmust be at least large enough to handle

the building heating requirements.Hybrid GHPs can also be used

for sites where the geologicalconditions or the available

ground surface will notallow a ground heatexchanger large enoughfor the building coolingloads to be installed.

A number of recentreports and researchpapers have beenpublished that dealwith both design ofhybrid GHPs andoperating experience

with a few installations.

ASHRAE (1995) andKavanaugh and Rafferty

(1997) both discuss advan-tages of hybrid GHPs and

present design procedures. Theformer sizes the auxiliary heat

rejecter based on the differencebetween monthly average heating and cool-

ing needs of the building and offers general guidelines

Assessment of Hybrid Geothermal Heat Pump SystemsGeothermal heat pumps offer attractive choice for space conditioningand water heating

IntroductionThe purpose of this Technology Installation Review isto provide an overview of hybrid geothermal heatpump systems. It presents the results of recent researchon these systems, looks at system types, energy sav-ings, maintenance considerations, and measured tech-nology performance from several examples.

Using the ground as a thermal energy source and/or aheat sink for heat pumps has long been recognized tohave a number of advantages over the similar use ofambient air. Ground temperatures at about 3-ft depthor lower are much less variable than ambient air tem-peratures. Further, soil or rock at these depths is usu-ally warmer than ambient air during the coldest wintermonths and cooler than ambient air during the summermonths. This fact leads directly to cooler condensa-tion temperatures (during cooling operation) andwarmer evaporating temperatures (during heating) fora heat pump with consequent improved energy effi-ciency. It also results in increased heating and coolingcapacity at extreme temperatures, thereby reducing oreliminating the need for auxiliary heat.

Heat pump systems that make use of theground in this way are called ground-source or geothermal heat pumps(GHPs). GHPs are also known bya variety of other names: geo-exchange heat pumps, ground-coupled heat pumps,earth-coupled heat pumps,ground-source systems,ground-water source heatpumps, well water heatpumps, solar energy heatpumps, and a few othervariations. Some namesare used to describe moreaccurately the specificapplication; however,most are the result of mar-keting efforts and the needto associate (or disassociate)the heat pump systems fromother systems.

Why Hybrid GHPs?The advantages of GHPs over conven-tional alternatives make them a very attractivechoice for space conditioning and water heating for

DOE/EE-0258

Matrix ofground heatexchangersin verticalbores shownwith optionalcooling towerfor hybridsystems

2

closed loop is “conditioned” by exchang-ing heat with one or more geothermal heatsources or sinks, such as the ground,groundwater, surface water, wastewaterstreams, or potable water supplies (whereallowed). Figure 1 illustrates the varioustypes of geothermal source/sinks that maybe used in GHP systems. Hybrid GHP sys-tem designs using outdoor air as a supple-ment heat sink can be configured with anyof these GHP source/sinks. Figure 2 is aschematic illustration of a hybrid GHPusing a closed loop vertical ground heatexchanger and a cooling tower as an auxil-iary heat rejecter.

for integration of the heat rejecter into thesystem piping. The latter bases heatrejecter sizing on peak loads at design con-ditions and the difference betweenrequired ground heat exchanger boreholelengths for heating and cooling.

Kavanaugh (1998) revises and extends thedesign procedures presented in the abovetwo publications. In addition, a controlmethod is proposed for balancing the cool-ing and heating loads on the ground heatexchanger to limit long-term temperatureincrease. The revised procedure is appliedto an office building in three climates,and first cost and operating cost issuesare discussed.

Yavuzturk and Spitler (2000) present acomparative study investigating severalcontrol strategies for hybrid GHP systems.The strategies investigated include setpoint control (operating the auxiliaryrejecter whenever the heat pump enteringor exiting fluid temperature exceeds a settemperature), differential temperature con-trol (operating the auxiliary rejecter when-ever the difference between heat pumpfluid temperature and ambient air tempera-ture exceeds a set value), and operation of theauxiliary rejecter to remove heat from theground heat exchanger field during night-time hours. The purpose of the last strategyis to use the tower to attempt to balance theannual heat rejection and heat absorptionloads on the ground heat exchanger. A20-year life-cycle cost analysis is con-ducted to compare each control strategy.

Thornton (2000) performed an analysis ofa hybrid GHP for a building at the U.S.Navy’s Oceana Naval Air Station. Overallsystem performance was compared forsimilar control strategies as studied byYavuzturk for both 1-year and 10-year per-formance periods.

A few studies of actual hybrid GHP instal-lations have been reported as well.Phetteplace and Sullivan (1998) describe asystem installed in an administration build-ing at the U.S. Army’s Fort Polk facility.The paper presents performance data overa 22-month period. Singh and Foster (1998)explore first-cost savings resulting fromhybrid GHP designs in two buildings—anoffice building and an elementary school.Both of these cases are examples of instal-lations where site geological characteris-tics or surface area limitations precluded

use of a ground heat exchanger largeenough to handle the total cooling needs.

Technology DescriptionGHP system typesThere are several types of geothermal heatpump systems that can be used for buildingspace conditioning and water heating. Thecommon denominator is that water sourceheat pumps exchange heat between indoorair (for space heating or cooling) or water(for heating or chilling water) and a liquid(either water or a water-coolant mixture)flowing in a closed loop. The liquid in the


Figure 1. Various geothermal source/sinks that can be applied to geothermal heatpump systems in commercial or residential applications.

Commercial Applications:

Residential Applications:

Optional coolingtower forhybrid systems

Matrix of ground heatexchangers in vertical bores

3

either desuperheaters on the heat pumps or,where large amounts of hot water areneeded, dedicated water-to-water units.

GHP systems save money because theyuse less energy and improve energy con-sumption patterns. The 4003 GHPs infamily housing at Fort Polk reduced thesummer electric peak demand of that cityof 12,000 people by 43% and increasedthe annual electric load factor from 0.52 to0.62. Federal sites can purchase electricityat lower costs when their load character-istics improve so dramatically. (For moredetails on the Fort Polk story seewww.eren.doe.gov/femp/financing/tecspec.html. Click on “Geothermal HeatPumps.” Also see Hughes and Shonder[1998].)

Maintenance Considerations—Cooling TowersFor hybrid GHPs using cooling towers asthe auxiliary heat rejecter, tower mainte-nance is an issue that must be considered.Cooling towers add additional mainte-nance time and cost to the HVAC system.ASHRAE (1996) offers recommendationsfor a tower maintenance program includ-ing daily and weekly visual inspections forcleanliness, control component function-ing, and sump water levels. These recom-mendations are intended to encouragetower maintenance and operating practicesto reduce Legionella risks.

Geothermal heat source/heat sinkoptionsFor the government to receive the bestvalue from GHP technology, installationcontractors and Federal site personnel needto determine which geothermal heatsource/sink type or combination of types ismost economical for each site. The orderof preference is not universal, but it gener-ally is as follows.

Groundwater already being pumped. Isgroundwater currently being pumped tothe surface? Some Federal sites pumpgroundwater to the surface, treat it, andre-inject it as a part of groundwaterremediation projects in areas near build-ings. Tapping into such an already existingheat source/sink may be economical. Aplate and frame heat exchanger can be usedto transfer heat between the groundwater


Figure 2. Hybrid GHP system schematic—tower isolated from building and ground loop.

The closed loop in a GHP system mayserve one or many heat pumps, dependingon the application. For example, militaryfamily housing might be served with sys-tems having one heat pump per living unit,each with its own vertical ground heatexchanger (GHX). Larger facilities mighthave many heat pumps on a common loopwith a central variable-speed pumping sta-tion and one large vertical GHX. The sche-matic in Figure 2 is an example of a largecentral system type. Individual watersource heat pump units provide heatingand cooling to each zone within the build-ing. These units are connected to a com-mon interior building pipe loop, which isused as a heat source or sink as needed.Dedicated water-to-water heat pumps mayalso be connected to the loop to meet build-ing water heating needs or to preheat or coolventilation air as shown in this example.The building loop is connected to theground source/sink system (a verticalground heat exchanger in this case) via acentral pumping station. The pumping sta-tion may be configured with a few large orseveral small single-speed pumps in paral-lel or with variable-speed pumps. Multipleparallel pumps or variable speed pumpsoffer the possibility of reducing pumpingpower during periods when building loadis lower than the design values. In thishybrid system example, the cooling tower(or other heat rejection unit) is connected inseries with the ground heat exchanger andis isolated from the building and groundpiping loops with a plate heat exchanger. Awet or evaporative tower is shown in theexample, but dry towers may also be used.

Systems using dry towers would consumemore energy because the tower would haveto reject heat to the ambient air dry bulbtemperature rather than the lower wet bulbtemperature, as is the case with evaporativetowers.

Energy Saving Mechanismsand BenefitsGHPs save energy and money because theequipment operates more efficiently thanin conventional systems. The compressorsin the individual heat pump units of a GHPsystem operate much more efficiently thanthose in air source units because the geo-thermal source/sink temperature is farmore stable than that of outdoor air and hasmuch less severe high and low extremes.In addition, air need be moved only on oneside of the GHP, and less power is neededto move the liquid on the other side thanwould be needed to move air. Unlike airsource units, GHP systems do not needdefrost cycles or backup electric resistanceheat at low outdoor air temperatures.

Common loop GHP systems recover heatas part of their design. In cooler weather,the heat pumps serving the building perim-eter extract heat from the common loop toprovide space heating, while units servingcore areas are cooling space and rejectingheat to the common loop. When the com-mon loop is in balance, no net conditioningis required from the geothermal source;under some operating conditions, the offsetbetween heating and cooling units reducesthe thermal load on the source. Recoveredheat also can be used to heat water, using

4

and a common loop serving individual heatpump units in nearby buildings. Afterremediation is completed, the pumping onthe groundwater side of the heat exchangercan be re-optimized for the HVAC applica-tion and continued using the same supplyand reinjection wells.

Stationary surface water. Are there largevolumes of stationary surface water on thesite, owned by the government and withgovernment use restrictions, near buildingswith significant heating and cooling loads?Heat exchange with surface waterimpoundments such as reservoirs, runoffretention basins, reflecting pools, ponds,and lakes, may be economical. A commonloop serving individual heat pump units innearby buildings can be submergeddirectly into the body of water. If the looppipe material could be subject to damagefrom local wildlife, or if the water is usedfor recreational or other purposes thatmight interfere with this approach, an on-shore pump house with a plate and frameheat exchanger and protected intake fromand discharge to the body of water could beconsidered. The issue of sensitivity oflocal fish to water temperature changesmay need to be considered as well.

Moving surface water. Are large volumesof reliable moving surface water (e.g.,large rivers with reliable flow and modestcurrent), owned by the government andwith government use restrictions, on thesite near buildings with substantial space?Use of this type of water for heat exchangewith a submerged pipe loop may be eco-nomical. If use of a submerged loop is notfeasible, an on-shore pump house with aplate and frame heat exchanger and pro-tected intake from and discharge to themoving body of water could be consid-ered. Issues such as historical high and lowwater conditions, debris flow, sensitivity ofaquatic life to temperature changes, andcommercial and recreational traffic wouldrequire serious attention.

Wastewater streams. Does the site havelarge-volume, reliable flowing wastewaterstreams near buildings with significantsquare footage? If so, those streams mayoffer economical heat exchange. A commonloop serving individual heat pump units innearby buildings could be conditioned by aplate and frame heat exchanger in contactwith the wastewater. Heat exchanger

maintenance must be considered, as wellas the stability of the missions of the facili-ties that are the source of the wastewater.

Groundwater. Are large quantities ofgroundwater available at a reasonabledepth at the site, as well as an acceptableand economical means of disposal, nearbuildings with significant heating andcooling loads? Heat exchange withgroundwater may be economical evenwhen the project must bear the cost ofdeveloping the supply and discharge wells.The groundwater can be used with a plateand frame heat exchanger to condition acommon loop serving individual heatpump units in nearby buildings. Poorwater quality might require the use ofexpensive heat exchanger materials, andadditional maintenance and aquiferre-injection in some formations might beexpensive. Local environmental regula-tions on the use of groundwater need to beconsidered. In some cases, a double-walled heat exchanger may be required tominimize the risk of contamination of theground-water with the fluid used in thecommon building loop.

Standing column well. Standing columnwell GHP systems are similar to standardgroundwater GHPs, but since water is re-circulated between the well and the build-ing, only one well may be required (largerprojects may have several wells in paral-lel). Standing column wells are feasible inareas with near surface bedrock. Deepbores are drilled, creating a long standingcolumn of water from the static water leveldown to the bottom of the bore. Water isrecirculated from one end of the column tothe other. During peak heat rejection orextraction periods, the system can bleedpart of the water to the surface (lake, pond,stream, etc.) rather than reinjecting it all,causing water inflow to the column fromthe surrounding rock formation. Thiscools the column during heat rejection(building cooling), heats it during heatextraction (building heating), and reducesthe required bore length.

Ground heat exchangers. Does the sitehave sufficient land area to accommodateground heat exchangers near buildingswith significant square footage? If so, heatexchange with the ground, using vertical orhorizontal loops, may be economical. Acentral ground heat exchanger can condition

a common loop serving individual heatpump units in nearby buildings. Alterna-tively, each heat pump, or small group ofheat pumps, can have its own ground loop.Horizontal loops require considerablymore land area but may be less expensiveto install, depending on the types of soiland rock formations encountered in drill-ing. Ground heat exchangers are an optionalmost anywhere. They are placed last inthis list not because they are less economi-cal than other geothermal options, butmerely to ensure that other options are con-sidered where they exist.

Specific hybrid systemconsiderationsSites should be examined to determinewhether a hybrid GHP system might beadvantageous. For applications where therequired geothermal heat sink capacity forcooling is much greater than the requiredheat source capacity for heating, hybridsystems can lead to significantly reducedinstallation costs. This is particularly truefor commercial or institutional buildings inwarm climates. A rule of thumb suggestedby Kavanaugh (1997) for defining a warmclimate location is one where the averageground temperature at 30-ft depth exceeds60°F. Hybrid designs may also have eco-nomic advantages in such buildings inmore moderate climates (30-ft ground tem-perature between 50-60°F) particularly forbuildings with low ventilation air require-ments or where heat recovery units or otherconditioning means are used to reduceventilation air heat gains/losses. Hybriddesigns can also allow use of GHP systems(and access to most of their advantages)where site subsurface geological conditionsor surface area limitations prevent installa-tion of geothermal source/sink systemswith sufficient capacity to fully meet thebuilding heat rejection needs for cooling.Auxiliary heat rejecter location and main-tenance requirements must be considered.

Technology Performance

Building 1562 – Fort Polk, LA

In 1993, Building 1562, a 24,000-ft2

administration building at Fort Polk, LA,was renovated including conversion of theexisting HVAC system to a hybrid GHP sys-tem. The new system consisted of 14 water


5

source heat pumps (total cooling capacityof about 120 tons) connected to a commonbuilding loop that was conditioned by avertical ground heat exchanger grid withseventy, 200-ft-deep boreholes on 10-ftcenters and a 78-ton wet cooling tower.(Kavanaugh and Rafferty [1997] recom-mend that bore-to-bore spacing in verticalGHX grids be at least 20-25 ft for applica-tions that are cooling load dominated tominimize long-term ground temperaturerise effects.) A hybrid system was selectedby the designer primarily because of limitedspace available for borehole installationand because high internal personnel andcomputer loads together with the warm cli-mate location led to design cooling loadsmuch greater than heating loads. Phetteplaceand Sullivan (1998) presented operatingdata collected over a 22-month period.Cooling tower operation was initiatedwhenever the common loop fluid tempera-ture exceeded 97°F and was terminatedwhen the loop temperature fell below 95°F.

The observed data show that, over theperiod of monitoring, the amount of heatrejected to the ground (in cooling) was

about 43 times as great as that extractedduring heating—an indication that thetower was not used effectively to reducethe thermal load on the ground in thisinstallation. Phetteplace and Sullivannoted that some increase in ground tem-perature occurred between the two summersover which the data were taken. This isattributed to the large imbalance betweencooling and heating loads on the groundloop and the close spacing of the bore-holes. They recommended that the towersetpoint controls be revised to either ini-tiate tower operation at a lower fluidsetpoint temperature or to operate thetower whenever possible to minimize theground heat rejection load. Even thoughthis would result in increased tower energyconsumption, it is estimated that systemenergy consumption would be reducedbecause the heat pumps would operate athigher efficiency (due to lower enteringfluid temperatures). The data showed thatthe tower fan and pump only accountedfor 4% of the system energy over the22-month monitoring period while the heatpumps used 77% of the total. Thus,

improved efficiency of the heat pumpscould more than compensate for even largeincreases in tower energy.

It was also noted that the building loop cir-culation pumps accounted for 19% of thetotal system energy use over the monitoredperiod. Loop circulation is provided bytwo 15-hp pumps, one of which operatescontinuously. The system could be modifiedto use solenoid valves at each heat pump(to shut off fluid flow when the heat pumpis off) and variable speed loop pumps toreduce loop flow during low load periods.Phetteplace and Sullivan estimated thatsuch a pump control strategy could havereduced the loop pumping power by 45%over the study period resulting in a 8.5%reduction in system energy use.

As of this writing, the system at Ft. Polk’sbuilding 1562 has not been modified toincorporate any of the recommendationsmade by Phetteplace and Sullivan. How-ever, they are preparing to retrofit a groupof their buildings with GHP systems underFEMP’s new GHP Super ESPC (seesidebar below) and this may include


GHP Super ESPC

The Federal Energy Management Program (FEMP) has implemented a “Super ESPC” to streamline the process of procuring GHP-centeredenergy saving retrofit projects for Federal facilities. In an ESPC (energy savings performance contract), an energy services contractor (ESCO)bears the costs of implementing energy-saving measures in exchange for fixed (annual or monthly) payments from the resulting cost savings.Under the Super ESPC, a GHP system must be the major focus but other energy saving measures (e.g., lighting improvements) can beincluded if they make the project more economical. FEMP has selected and preapproved a pool of ESCOs with which Federal agencies cancontract. Under the Super ESPC, delivery orders can be awarded quickly, and facility managers have the assurance that all of the selectedESCOs are qualified to deliver top-quality GHP-centered energy efficiency projects.

The following contractors have been selected for this Super ESPC:

• Constellation Energy Source, Baltimore, MD; www.cesource.com

• Duke Solutions, Charlotte, NC; www.dukesolutions.com

• Energy Performance Services, Inc., King of Prussia, PA; www.energy-assets.com

• Enron Energy Services Operations, Inc., Alpharetta, GA; www.enron.com

• Trane Company Asset Management Services, St. Paul, MN; www.trane.com

Among the advantages of GHP-centered projects under the Super ESPC are:

• Alignment with ESPC statutory authority and full compliance with all Federal procurement regulations is assured.

• New GHP-based HVAC and water heating systems can be acquired at no capital cost to the Federal facility.

• Super ESPC contracts were awarded to large financially stable ESCO teams that can offer financing at low rates. In addition, these ESCOshad to demonstrate their GHP capabilities through past performance and a specific proposal for a large initial project.

• ESPC project cost savings are guaranteed to exceed payments to the ESCO for services and debt.

6

upgrading the controls of the existinghybrid GHP system (Phetteplace 2001).

Paragon Center – Allentown, PA

The Paragon Center in Allentown, PA, isan 80,000-ft2, four-story office building. Itwas intended to have a full GHP system forbuilding HVAC with a ground heatexchanger consisting of fifty-five, 500-ftdeep boreholes to meet a 200-ton designcapacity (Singh and Foster 1998). Althoughan early test bore indicated that the sitecould accommodate the deep bores, subse-quent drilling in the intended ground looplocation was unable to go lower than about110-125 ft due to high water flow in a lime-stone strata. Drilling at that location wouldrequire casing the boreholes. For the original500-ft borehole design casing would haveincreased installation costs beyond theallowable budget. It was determined that ahybrid GHP design with eighty-eight,125-ft cased boreholes and a 120-ton cool-ing tower could be built within the budgetallowing the building to enjoy most of theadvantages of GHP technology. Singh andFoster note that the building HVAC systemalso employs heat recovery units to reduceheat losses and gains from ventilation airand uses variable speed pump control tolimit pump energy at off-peak conditions.The building operating energy cost isreported to be less than $1.00/ft2 annually,including demand charges.

Elementary School – West Atlantic

City, NJ

Singh and Foster (1998) also discussed aplanned expansion of an elementaryschool in West Atlantic City, NJ. Theexpanded structure will have 85,000 ft2 ofconditioned area and will be occupied9 months of the year. Total cooling loadincluding ventilation air is 275 tons. Heatrecovery equipment included in the HVACdesign covers 25 tons of the load, leaving250 tons to be covered by the main HVACsystem. It was intended to use a full GHPsystem with ninety, 400-ft bores. How-ever, the relatively small site (210 ft by580 ft) was not large enough to fit all of theboreholes needed. A hybrid GHP will beinstalled with a ground heat exchangersystem of sixty-six, 400-ft bores covering133 tons of the design load and a 117-toncooling tower to make up the difference.

Ground heat exchanger installed cost isabout $9.70 per bore ft ($255,600 total).No operating data is reported for the sys-tem, but Singh and Foster estimated thatthe hybrid GHP would generate annualenergy cost savings of about $4500 andmaintenance savings of $5000 comparedto a conventional boiler/tower water loopheat pump system. The estimated installa-tion cost of the hybrid GHP is $1,139,100compared to $1,118,300 for the conven-tional alternative, so the estimated simplepayback is 2.2 years.

Technology DemonstrationDetailed comparative analyses have beenperformed on full and hybrid GHP systemsby Yavuzturk and Spitler (2000) on a smalloffice building in two locations and byThornton (2000) on a Navy trainingcenter building.

Yavuzturk and Spitler consider a smalloffice building (14,205-ft2 conditionedspace) in Houston, TX, and Tulsa, OK, cli-mates. Annual loads for this building inHouston were calculated to be 7.5 millionBtus heating and 181.6 million Btus cool-ing. For the Tulsa location, the calculatedannual loads were 50.1 million Btus heat-ing and 133.8 million Btus cooling. Designcooling loads for the building were esti-mated to be about 15 tons in both locations.The hybrid GHP system layout is based onthe schematic in Figure 2 where the cool-ing tower is isolated from the ground heatexchanger and building loops with a plateheat exchanger. The plate heat exchangeris plumbed in series with the ground heatexchanger. The ground loop is modeledusing a short time step approach developedby Yavuzturk and Spitler (1999). Totalsystem simulation for both full and hybridGHP cases is done using the TRNSYScode (Klein et al. 1996) with the groundloop model incorporated into a TRNSYSmodule. Eight different hybrid GHPsystem control strategies within three maincategories were investigated to determinetheir impacts on system energy consump-tion and overall system operation. Thespecific strategies within each maincategory that yielded the lowest system(building heat pumps plus tower pumpand fan) energy use and life-cycle cost aresummarized in this review. System

maintenance costs were not quantified forthe life-cycle cost estimates in these studiesand no annual electricity escalation ratewas considered.

Hybrid Case 1. The cooling tower is acti-vated when the fluid temperature leavingthe heat pumps exceeds a specified setpoint. In this study, the set point tempera-ture was 96.5°F.

Hybrid Case 2. This approach uses thedifference between the heat pump fluidexit temperature and the ambient wet-bulbtemperature. Cooling tower operation isinitiated whenever the temperature differ-ence exceeds 3.6°F and is stopped whenthe difference falls below 2.7°F .

Hybrid Case 3. This strategy is based onusing the cooling tower to reject heat fromthe ground to avoid long-term temperaturerise. This effect is achieved by operatingthe tower for six hours daily (from mid-night to 6 a.m.) year-round. To avoid highloop temperatures, a secondary set pointcontrol is included to operate the towerwhenever the fluid temperature enteringthe heat pumps exceeds 96.5°F.

Summary results for the subject building inthe Houston and Tulsa locations are givenin Tables 1 and 2, respectively, and also inFigures 3 and 4. Case 1 results in the leastusage and energy consumption for thecooling tower of the three hybrid controlapproaches. Case 2 yields the lowest val-ues for annual system energy use and life-cycle cost for both locations while case 3has the lowest system first cost. The case 2strategy operates the tower a very largenumber of hours and generally underadvantageous outdoor ambient (lowertemperature) conditions. This approachtended to cool the ground over time result-ing in lower ground loop fluid temperature.This resulted in lower heat pump energyuse due to more efficient heat pump opera-tion in the cooling season. Tower usageand energy consumption is greatest for thisapproach, but overall system energy usecompared to the full GHP design is 18%lower for the Houston location and 6%lower for the Tulsa location. Clearly, thehybrid approach yields greater energy sav-ing benefits in the warmer (higher coolingload) location. Energy use for the groundloop and building loop circulation pump is


7

also a significant factor in total systemenergy consumption as noted above indiscussion of the Ft. Polk system. In analy-sis of the Yavuzturk and Spitler casestudy for this review, however, it is assumedto be the same for the full GHP and thehybrid GHP cases. The building heatpump flow requirements will be the samefor all systems and the total pressure lossfor the GHX and tower loops is assumednot to vary significantly among the fourcases studied.

Figure 4 shows the comparison in systemfirst cost between full and hybrid GHPdesigns for both locations. First cost of theground heat exchanger plus auxiliary heatrejecter for the hybrid cases is lower thanfor the full GHP ground loop in both loca-tions. This difference is most dramatic inHouston where the hybrid system reducesinstallation costs by more than 50%.


Table 1. Summary of Hybrid GHP case study for 14,025 ft2 office building in Houston, Texas (from Yavuzturk 2000)Heating degree-days = 1434; Cooling degree-days = 2889* Annual heating load = 7.5 million Btus Annual cooling load = 181.6 million Btus

Base case—no Hybrid Hybrid Hybridcooling tower Case 11 Case 22 Case 33

Number of boreholes in ground heat exchanger 36 @ 250 ft 12 @ 250 ft 12 @ 250 ft 12 @ 250 ft

Cost of ground heat exchanger4 $54,000 $18,000 $18,000 $18,000

Maximum fluid temperature entering heat pumps in 20-year period (°F) 96.6 96.3 80.5 96.0

Minimum fluid temperature entering heat pumps in 20-year period (°F) 71.3 67.3 40.5 54.1

Design capacity of cooling tower (tons) 22.5 11.5 8.5

Cost of cooling tower and plateheat exchanger including controlsand auxiliary equipment5 - $8,662 $4,427 $3,272

Total cost of ground heat exchanger and cooling tower equipment $54,000 $26,662 $22,427 $21,272

Present value of 20 years of electricity costs6 $19,611 $19,413 $16,011 $20,573

Present value of total costs $73,611 $46,075 $38,438 $41,845

Annual energy use (kWh) Heat pumps 24,425 23,877 17,792 24,453 Cooling tower fan - 260 1,847 1,006 Cooling tower pump - 42 302 164 Total system 24,425 24,179 19,941 25,623

* From Air Force Manual AFM 88-29, “Engineering Weather Data,” July 1978.1 Set point control to limit fluid temperature exiting heat pumps to 96.5°F or less.2 Differential temperature control activates tower when difference between heat pump exiting fluid temperature and air wet-bulb temperature exceeds 3.6°F and turnstower off when this difference falls below 2.7°F.3 Tower operated between midnight and 6 a.m. year-round to cool ground heat exchanger field (attempt to balance heating and cooling loads on ground); secondary setpoint control to limit fluid temperature entering heat pumps to 96.5°F or less.4 Estimated at $6.00/ft of borehole, including horizontal runs and connections (Kavanaugh and Rafferty 1997).5 Estimated at $385/ton of tower design capacity based on data from Means (1999).6 Assumes $0.07 per kWh cost of electricity; no price escalation rate assumed. A 6% discount rate is used for present value computation.

Thornton (2000) performed a comparativeanalysis of hybrid and full GHP systemsfor Building 137 at the U.S. Navy OceanaNaval Air Station. Building 137 is a train-ing facility with about 21,000 ft2 of officeand classroom space. In addition, there area number of large, open-bay hangarsattached to the building. The building ispresently heated and cooled using a combi-nation of steam from a central base steamgenerating station (for the hangar areas)and water source heat pumps connected toa common building water loop (for theclassroom/office area). When the buildingis being cooled, condenser heat from theheat pumps is rejected through a 100-toncooling tower. The HVAC system for theoffice/classroom portion of the buildingwill be renovated to a GHP-based systemunder an energy savings performance con-tract (ESPC). This will be one of the firstsuch arrangements to be put in place under

the GHP “Super ESPC” recently approvedby FEMP (see sidebar, page 5).

In Thornton’s analysis, the TRNSYS codewas used to model the building and allHVAC system components. The annualheating load for the office/classroom por-tion of the building was calculated to be199.6 million Btus while the annual cool-ing load was 439.1 million Btus. There are21 individual water source heat pumpsused to provide heating and cooling with atotal nominal cooling capacity of about75 tons. Total length of the ground heatexchanger for the full GHP baseline casewas 11,400 ft. The ground heat exchangerwas simulated using the DST algorithm(Pahud and Hellstrom 1996) which wascast as a TRNSYS module. Thornton con-sidered several different hybrid systemconfigurations and control strategies basedon using the existing cooling tower as the

8

Table 2. Summary of Hybrid GHP case study for 14,025 ft2 office building in Tulsa, Oklahoma (from Yavuzturk 2000)Heating degree-days = 3680; Cooling degree-days = 1949* Annual heating load = 50.1 million Btus Annual cooling load = 133.8 million Btus


Number of boreholes in ground heat exchanger 16 @ 240 ft 9 @ 240 ft 9 @ 240 ft 9 @ 240 ft


Maximum fluid temperature entering heat pumps in 20-year period (°F) 96.4 96.9 79.0 97.9

Minimum fluid temperature entering heat pumps in 20-year period (°F) 50.2 39.8 24.2 39.2

Design capacity of cooling tower (tons) - 17.0 11.0 5.5

Cost of cooling tower and plate heat exchanger including controlsand auxiliary equipment5 - $6,545 $4,235 $2,118

Total cost of ground heat exchanger and cooling tower equipment $23,040 $19,505 $17,195 $15,078


Present value of total costs $39,039 $35,493 $32,171 $32,672

Annual energy use (kWh) Heat pumps 19,927 19,813 16,463 20,769 Cooling tower fan - 86 1,882 984 Cooling tower pump - 14 308 161 Total system 19,927 19,913 18,653 21,914

* From Air Force Manual AFM 88-29, “Engineering Weather Data,” July 1978.1 Set point control to limit fluid temperature exiting heat pumps to 96.5°F or less.2 Differential temperature control activates tower when difference between heat pump exiting fluid temperature and air wet-bulb temperature exceeds 3.6°F and turnstower off when this difference falls below 2.7°F.3 Tower operated between midnight and 6 a.m. year-round to cool ground heat exchanger field (attempt to balance heating and cooling loads on ground); secondary setpoint control to limit fluid temperature entering heat pumps to 96.5°F or less.4 Estimated at $6.00/ft of borehole, with horizontal runs and connections (Kavanaugh and Rafferty 1997).5 Estimated at $385/ton of tower design capacity based on data from Means (1999).6 Assumes $0.07 per kWh cost of electricity; no price escalation rate assumed. A 6% discount rate is used for present value computation.

Tulsa, Ok Houston, TX

30,000

25,000

20,000

15,000

10,000

5,000

0Full GHP Hybrid GHP

Case 1Hybrid GHP

Case 2Hybrid GHP

Case 3

Annu

al s

yste

m E

nerg

y Us

e (k

Wh)

19,927

24,425

19,913

24,179

18,65319,941

21,964

25,623

Figure 3. Full vs. hybrid GHP system (heat pumps, tower pump, and tower fan) energyuse for 14,025-ft2 office building—Houston, Texas, and Tulsa, Oklahoma, locations.


auxiliary heat rejecter. The three that pro-duced the lowest hybrid GHP system(building heat pumps plus tower fan andcirculating pump) energy usage are sum-marized below.

Oceana Hybrid Case 1. For this system,the tower is piped in series with the groundheat exchanger and is not isolated from theGHX loop as shown in Figure 5. Thetower is activated when the fluid tempera-ture in the system fluid loop exceeds aspecified set point. In this case, the setpoint temperature was 80°F. Ground heatexchanger size for this case was 7,500 ft vs.11,400 ft for the base case.

Oceana Hybrid Case 2. For this system,the tower is isolated from the ground heatexchanger with a plate heat exchanger asshown in Figure 6. The auxiliary heatrejection circuit (tower) operates when thefluid temperature in the tower fluid loop

9

exceeds 70°F. Main loop flow was splitbetween the plate and ground heatexchangers such that the total heat rejec-tion to the ground in summer equaled theheat extraction in winter. The requiredsplit in loop flow to achieve ground ther-mal load balance for this case was 90% tothe plate heat exchanger. Ground heatexchanger size for this case was 4,650 ftvs. 11,400 ft for the base case.


Figure 4. Full vs. hybrid GHP system first cost for 14,025-ft2 office building—Houston,Texas, and Tulsa, Oklahoma, locations.

Tulsa, Ok Houston, TX

60,000

50,000

40,000

30,000

20,000

10,000


Case 1Hybrid GHP

Case 2Hybrid GHP

Case 3

Syst

em F

irst C

ost (

$)

23,040

54,000

19,505

26,662

17,195

22,427

15,078

21,272

Oceana Hybrid Case 3. For this system,the tower is isolated from the ground heatexchanger with a plate heat exchanger asshown in Figure 6. The auxiliary heatrejection circuit (tower) operates when thefluid temperature in the tower fluid loopexceeds 70°F. Main loop flow was splitbetween the plate and ground heatexchangers so that the maximum fluidtemperature entering the building heat

pumps was kept below 95°F. The requiredsplit in loop flow for this case was 42% tothe plate heat exchanger. Ground heatexchanger size for this case was the sameas for case 2.

Summary results for Thornton’s ten-yearanalysis for the full GHP base case and thethree hybrid GHP cases are presented inTable 3. As in Yavuzturk and Spitler’sanalysis, the setpoint control strategy ofcase 1 results in the least tower usage andenergy consumption. Case 3 yields thelowest value for both first cost and life-cycle cost. For this case study, the first costis taken to be the GHX cost plus the cost ofthe plate heat exchanger and tower loopcirculation pump. Tower cost is not includedbecause a tower already exists for thisbuilding. All other system costs (for theinternal building systems and controls,GHX pump and controls, etc.) are assumedto be the same for all systems examined.

For the Oceana site, annual energy use forthe building heat pumps is almost the samefor all cases (full and hybrid GHPs) in con-trast to the results seen for the previouscase study assessment for Houston andTulsa locations. At Oceana, the cooling-to-heating (C/H) load ratio for the buildingstudied was about 2.2—about 20% lowerthan for the office building studied in Tulsa(C/H ratio about 2.7) and much lower thanfor the building studied in Houston (C/Hratio about 24.2). Thornton’s Oceanastudy results show that in the hybrid GHPcases, the building heat pumps used moreenergy in the heating season than for thebase case full GHP due to the lower mini-mum fluid temperatures. Cooling seasonheat pump energy use ranged from slightlylower to about the same for the hybridGHP systems. With the added energy useof the tower fan and pumps, overall systemenergy use for the hybrid GHP cases stud-ied at Oceana ranges from 1% to 13%higher than for the base case.

Figure 7 shows the comparison of first costbetween the full and hybrid GHPs for theOceana study. For case 1 (with tower notisolated from GHX and building), theGHX size and system first cost are bothreduced by about 34%. In cases 2 and 3,with an isolated tower circuit, it is possible

Figure 5. Hybrid GHP system schematic—tower unisolated from ground andbuilding loop.

10

Figure 6. Hybrid GHP system schematic for Oceana analysis cases 2 and 3—towerisolated from building and GHX and piped in parallel with GHX.

Table 3. Summary of Oceana Hybrid GHP case study - existing tower (from Thornton 2000)Heating degree-days = 3639; Cooling degree-days = 1485* Annual heating load = 199.6 million Btus Annual cooling load = 439.1 million Btus


Number of boreholes in ground heat exchanger - ten-year design 76 @ 150 ft 50 @ 150 ft 31 @ 150 ft 31 @ 150 ft


Maximum fluid temperature entering heat pumps in ten year period (°F) 94.9 84.6 92.5 95.1

Minimum fluid temperature entering heat pumps in ten year period (°F) 63.3 53.7 35.5 44.4

Existing cooling tower capacity (tons) - 100.0 100.0 100.0

Cost of plate heat exchanger and tower circuit pump5 - - $3,800 $3,800

Total cost of ground heat exchanger, plate heat exchanger,and tower pump $92,454 $60,825 $41,512 $41,512


First-year electricity costs6 $3,950 $4,007 $4,470 $4,335

Present value of first costs plus electricity costs $121,523 $90,317 $74,410 $73,418

Annual energy use (kWh) Heat pumps 70,027 69,083 72,234 71,632 Cooling tower fan - 573 3,115 1,379 Cooling tower pump - 1,391 3,902 3,840 Total system 70,027 71,047 79,251 76,861

* From Air Force Manual AFM 88-29, “Engineering Weather Data,” July 1978.1 Unisolated tower; tower operated whenever loop temperature exceeds 80°F.2 Isolated tower with plate heat exchanger; tower operated to achieve balance between heating and cooling loads on ground heat exchanger.3 Isolated tower with plate heat exchanger; tower operated with setpoint control to keep maximum fluid temperature entering heat pumps at or below 95°F.4 Actual cost of test bores drilled at Oceana site for ground thermal conductivity testing was $8.11 per ft of borehole, including horizontal runs and connections (personalcommunication with B. Koshka, Trane Co. December, 2000).5 Plate heat exchanger estimated at $25.50 per ton based on data presented by Kavanaugh and Rafferty 1997; tower circuit pump cost estimated at $1250.6 Electricity cost $0.0564 per kWh (rate charged to Oceana buildings by U.S. Navy Public Works Center - Norfolk); no price escalation rate assumed. A 6% discountrate is used for present value computation.

to operate with lower temperatures in theGHX (using an antifreeze solution). TheGHX size is reduced about 59% and sys-tem first cost by about 55% in these cases.The hybrid systems had life-cycle costsfrom 26% lower (for case 1) to 40% lower(for case 3) than that for the base case.

Summary. Based on the results from thetwo case studies analyzed for this review,the following observations are made.

• Hybrid GHP systems can significantlyreduce system first costs even when atower needs to be purchased. Costs canbe reduced by more than 50% for veryhighly cooling dominated applicationssuch as the small office building inHouston (cooling-to-heating load ratioof 24:1).


11


Figure 7. Full vs. hybrid GHP system first cost comparison for Bldg 137 Oceana NAS,Virginia (existing cooling tower).

100,000

80,000

60,000

40,000

20,000


Case 1Hybrid GHP

Case 2Hybrid GHP

Case 3

Syst

em F

irst C

ost (

$)

92,454

60,825

41,512

• For applications where a suitable toweralready exists (as at the Oceana studysite), a hybrid system can result in sys-tem cost reductions of more than 50%even when the building is not overlycooling load dominated.

• For heavily cooling dominated sites,hybrid GHPs can result in heat pumpand system energy savings comparedto full GHPs when the supplementaryheat rejecter is operated enough hoursto reduce the average heat pump enter-ing fluid temperature during the cool-ing season.

• The authors of both case studies pointout that none of the hybrid systemdesigns they examined have been opti-mized. A design optimization methodis needed to balance GHX size, supple-mental heat rejecter size and type, con-trol strategy, and electric rate structureto achieve lowest life-cycle or first costdesigns for a given location.

ReferencesASHRAE. 1995. Commercial/institu-tional ground-source heat pumps engi-neering manual. Atlanta: AmericanSociety of Heating, Refrigerating, andAir-Conditioning Engineers, Inc.

ASHRAE. 1996. 1996 ASHRAE hand-book—Systems and equipment, chapter36. Atlanta: American Society of Heating,Refrigerating, and Air-Conditioning Engi-neers, Inc.

Hughes, P.J., and J. Shonder. 1998. Theevaluation of a 4000-home geothermalheat pump retrofit at Fort Polk, Louisiana:final report. Report No. ORNL/CON-460Oak Ridge: Oak Ridge National Laboratory.

Kavanaugh, S.P., and K. Rafferty. 1997.Ground-source heat pumps: Design ofgeothermal systems for commercial andinstitutional buildings. Atlanta: AmericanSociety of Heating, Refrigerating, and Air-Conditioning Engineers, Inc.

Kavanaugh, S.P. 1998. A design methodfor hybrid ground-source heat pumps.ASHRAE Transactions 104 (2): 691-698.

Klein, S.A., et al. 1996. TRNSYS Manual,a transient simulation program. Madison:Solar Engineering Laboratory. Universityof Wisconsin-Madison.

Means. 1999. Means mechanical costdata. Kingston: R.S. Means Co.

Pahud, D., and G. Hellstrom. 1996. Thenew duct ground heat model for TRNSYS.Eurotherm Seminar N* 49, Eindhoven,The Netherlands, pp 127-136.

Phetteplace, G., and W. Sullivan. 1998.Performance of a hybrid ground-coupledheat pump system. ASHRAE Transactions104 (2): 763-770.

Phetteplace, G. 2001. Personalcommunication.

Singh, J.B., and G. Foster. 1998. Advan-tages of using the hybrid geothermaloption. The Second Stockton Interna-tional Geothermal Conference, The Rich-ard Stockton College of New Jersey.(http://styx.geo-phys.stockton.edu/pro-ceedings/hybri/singh/singh.PDF).

Thornton, J. 2000. A detailed TRNSYSanalysis of building 137; Oceana Naval AirStation, Virginia Beach, Virginia. Projectreport to Oak Ridge National Laboratory.

Yavuzturk, C., and J. Spitler. 2000. Com-parative study of operating and controlstrategies for hybrid ground-source heatpump systems using a short time step simu-lation model. ASHRAE Transactions106 (2).

DisclaimerThis report was sponsored by the United States Department of Energy, Office of Federal Energy ManagementPrograms. Neither the United States Government nor any agency or contractor thereof, nor any of their employees,makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness,or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringeprivately owned rights. Reference herein to any specific commercial product, process, or service by trade name, mark,manufacturer, or otherwise, does not necessarily constitute or imply its endorsement, recommendation, or favoring bythe United States Government or any agency or contractor thereof. The views and opinions of authors expressedherein do not necessarily state or reflect those of the United States Government or any agency or contractor thereof.

For More Information

FEMP Help Desk(800) 363-3732International callers please use(703) 287-8391Web site: www.eren.doe.gov/femp

General Contacts

Ted CollinsNew Technology Demonstration Program ManagerFederal Energy Management ProgramU.S. Department of Energy1000 Independence Ave., SW, EE-92Washington, D.C. 20585Phone: (202) 586-8017Fax: (202) [email protected]

Steven A. ParkerPacific Northwest National LaboratoryP.O. Box 999, MSIN: K5-08Richland, WA 99352Phone: (509) 375-6366Fax: (509) [email protected]

Technical Contact

Van BaxterOak Ridge National LaboratoryP.O. Box 2008, MS-6070 Bldg 3147Oak Ridge, TN 37831-6070Phone: (865) 574-2104Fax: (865) [email protected]

Produced for the U.S. Departmentof Energy (DOE) by the Oak RidgeNational Laboratory

DOE/EE-0258

December 2001


Printed with a renewable-source ink on paper containing at least 50% wastepaper, including 20% post consumer waste

About FEMP’s New Technology Demonstration ProgramThe Energy Policy Act of 1992, and subsequentExecutive Orders, mandate that energy consumptionin Federal buildings be reduced by 35% from 1985levels by the year 2010. To achieve this goal, the U.S.Department of Energy’s Federal Energy ManagementProgram (FEMP) is sponsoring a series of programs toreduce energy consumption at Federal installationsnationwide. One of these programs, the New Tech-nology Demonstration Program (NTDP), is tasked toaccelerate the introduction of energy-efficient andrenewable technologies into the Federal sector and toimprove the rate of technology transfer.

As part of this effort FEMP is sponsoring a series ofpublications that are designed to disseminate informa-tion on new and emerging technologies. NewTechnology Demonstration Program publicationscomprise three separate series:

Federal Technology Alerts—longer summaryreports that provide details on energy-efficient, water-conserving, and renewable-energy technologies thathave been selected for further study for possibleimplementation in the Federal sector. Additionalinformation on Federal Technology Alerts (FTAs) isprovided in the next column.

Technology Installation Reviews—concise reportsdescribing a new technology and providing case studyresults, typically from another demonstration programor pilot project.

Technology Focuses—brief information on new,energy-efficient, environmentally friendly technolo-gies of potential interest to the Federal sector.

Documents

Assessment of Hybrid Geothermal Heat Pump Systems ...infohouse.p2ric.org/ref/43/42893.pdf3 either desuperheaters on the heat pumps or, where large amounts of hot water are needed,