DA-07-061

Development of a Ground-Source Heat Pump System with Ground Heat Exchanger Utilizing the Cast-in-Place Concrete Pile Foundations of Buildings Kentaro Sekine

Ryozo Ooka, DrEng

Mutsumj Yokoi, DrEng

Member ASHRAE

Yoshiro Shiba

ABSTRACT Ground-source (geothermal) heat pump (GSHP) systems can achieve a higher coefficient of performance than conventional air-source heat pump (ASHP) systems. However. GSHP systems are not widespread in .Japan because oftheir expensive boring costs. The authors have developed a GSHP .system that uses the cast-in-place concrete pile foundations of a building as heat exchangers to reduce the initial boring cost. In this system, .some U-tubes are arranged around the surface of a cast-in-place concrete pile foundation. The heat exchange capability ofthis system, subterranean temperature changes, and heat pump performance were investigated in a full-scale experiment. As a result, the average values for heat rejection were 186-20! W/m (per pile, 25 W/ m per pair of tubes) while cooling. The average COP of this system was 4.89 while cooling, rendering this .system about J. 7 times more effective in energy-saving terms than the more typical ASHP .systems. The initial cost of construction per unit for heat extraction and rejection is US$0.79/W (approximately f79/W)for this system, whereas it is USS3/W (¥300/fV) for existing .standard borehole .systems. Therefore, this system is expected to be commercially viable.

SuckHo Hwang

the heat island phenomena, as this system does not emit exhaust heat into the atmosphere during air conditioning. However, GSHP systems are not popular in Japan except for experimental versions. This is primarily due to the high cost of boring to run piping underground. For example, such boring costs average about US$30/m (approximately ¥3.000/m) in the USA, whereas the same work is about US$100/m (¥10,000/m) in Japan. Thus, even if the heat pump performance in GSHP systems is more effective than that ofthe more common ASHP systems, the GSHP systems are unable to recoup the initial piping costs within their life cycles. Recently, a GSHP system that uses the foundation piles of buildings as a heat exchanger (so-called energy pile system) was introduced into some buildings in order to reduce the initial boring cost (Hamada et al. 1997; Arup 2002; Presetschnik and Huber 2005).

INTRODUCTION

However, an effective and low-cost design method for energy pile systems has not yet been developed. In addition, most energy pile systems have used precast, prestressed concrete pile or steel pipe pile. Recently, in urban areas in Japan, the cast-in-place concrete foundation piles of buildings have been used for reasons of traffic circumstances when carrying the piles or for cost reduction.

Ground-source heat pump (GSHP) systems can achieve a higher coefficient of performance than conventional airsource heat pump (ASHP) systems because the ground, which functions as the heat source or sink, is at a higher temperature In winter and lower in summer than the air temperature (Kavanaugh 1992; Kavanaugh and Rafferty 1997). In addition, there will likely be some mitigation agaitist the effects of

The authors have developed a GSHP system (energy pile system) that uses the cast-in-place concrete pile foundation of buildings. In this research, a full-scale experiment was conducted. The heat exchange capability of this system, subterranean temperature changes, and performance of the heat pump were investigated. Furthermore, the construction costs of this system were also examined.

Kcnlaro Sekine is a research engineer at Ihc Building Engineering Research Institute Technology Center. Taisei Corporation, Kanagawa, Japan, and Mutsumi Yokoi is a senior engineer at Design Division. Taisei Corporation. Tokyo. Ryozo Ooka is an associate professor and SuckHo Hwang is a graduate student at the University of Tokyo, Japan. Yoshiro Shiba is a deputy manager ofthe Development Department, Zeneral Healpump Industry Co., Ltd., Nagoya, Japan.

558

©2007 ASHRAE.

SYSTEM OUTLINE The usual diameters of cast-in-place concrete pile foundations are from t .5 to 4.0 m. In this system, some U-tubes are arranged around the exterior cast-in-place concrete pile foundations, as shown in Figure I. as the ideal location. U-tubes are fixed, not directly to the hoops ofthe main reinforcement of the pile., but to the spacers for preventing eccentricity ofthe pile. At this location, the U-tubes are positioned outside the diameter ofthe pile detentiined by the structural design and the core ofthe piling (and concrete will be placed around the U-tubes), and will cause no partial loss of area of the piling and no loss in structural strength. The arrangement can also ensure a larger space between U-tubes than in the conventional method of installing U-tubes inside piles., and, thus, the heat interference between U-tubcs is smaller.

Castini>lacc concrete pile Diiimuti'r 1.3 to 4.0 m

al Hea Pump

U-bend 21 u..;,- ...;.. ^..-uLr Ji.i Foundnlion piles Main reinforcfmeiit

The U-tubes are normally made of high-grade or crosslinked polyethylene. These U-tubes usually have a 21 to 28 mm inside diameter. The U-tube used in this system is made of polyethylene, which is very steady chemically and is also strong in temperature change. Thus, there is no worry about which polyethylene has deteriorated even if concrete curing (heating, swelling/shrinking) occurs. In the case of the application of 1.5 m cast-in-place concrete pile to the usual office building in Japan, one pile is usually set per 30 to 40 m" floor area each. In this study, the heat extraction/rejection rate for one pile was about 180 W/m. Therefore, if pile length is 30 m (as is very common in Japan), the heat extraction/rejection rate per pile became 5.4 kW. This corresponds to the air-conditioning load for two to three floor areas (i.e., about one-third ofthe total air-conditioning load of an eight-story ofUcc building, which is a very popular building design in Japan). The heat source for air conditioning in office buildings is usually divided into several parts. In this case, geothermal heat from this pile system can be used as one part ofthe heat source.

Core of the piling

Figure 1 Outline ofthe heat exchange system using cast-inplace concrete piles.

Experimental Equipment Outiine

test to measure the ground thermal conductivity in this site, which was 1.40 W/mK, an ordinary value in Japan. There were two cast-in-place concrete piles (both 1.5 m in diameter, 20 m in length) around which eight U-tubes (outside diameter, 34.0 mm; inside diameter, 28.8 mm) were installed in parallel in this experiment. It is possible to control the number of U-tubes in operation by opening and closing their valves. Before the experiment, an optimum number of pairs of U-tubes was not clarified. Thus, as many pair of U-tubcs as possible were arranged, then an optimum number of pairs would be dctennincd by changing the number of pairs operated in this experiment. Opening or closing the valves installed in the tubes can alter the number of U-tubes used. In an actual building, the number of U-tubes to be used will be determined based on pile diameter and heat extraction/rejection ratio.

An experimental institution was built on site at the University of Tokyo in Chiba. Chiba is east of Tokyo, and the average annual air temperature is about 15.4X, with the average air temperature in August being about 26.4°C and in January about 5.4°C. Accordingly, both heating and cooling functions are necessary. A plan ofthe experimental institution and the system configuration is shown in Figure 2. The results of this experiment were obtained at a local site. Thus, people should apply these results to other sites carefully. A borehole tog at this site is also shown in Figure 2 in order to show the locality of this experiment. We conducted a thermal response

The system used with this experimental equipment consists of a water-to-water heat pump with a reciprocating compressor (4.6 kW cooling, 5.7 kW heating). Cold and hot water circulates through a fan-coil unit and a radiation panel in two examination rooms, respectively, as shown in Figure 2. The flow of cold and hot water is 27 L/min (0.00045 mVs). The flow ofthe heat source (sink) water is 33 L/min (0.00055 re? Is). This system has two rooms. One has a fan-coil unit installed. The other has a radiation air conditioner. Thermostat and electrical valves control the amount of water supplied to the fan-coil unit.

The other air-conditioning load is supplemented with the usual air-source heat pump. These methods are quite reasonable and applicable. In this paper, basic perfonnance and the construction cost of this system are examined.

FULL-SCALE EXPERiMENT

ASHRAE Transactions

559

4S37 5

4517 5

.TOO

depth story

Fun uoiliinit

1

Preparalion room

i\l>onircntrooraA

lHin panel

(j j

& Oantiol panel

roumJalion pile A

—v..

U-bends ISW

I 1250 .,,.. I

4 5 6 7

U A

1500

oolor

loam

1.7

3.2

^^k

1.3 03

-"-'^'-

clay loam tuffclav siltv dav

darii brown

1

pals r«d brown

CIHV fine

9 10

groundwabr evel

11

•w

-1-3 14 15

Foundation pile B

12.67

fine sand

pale dark brovm

~I7~' 13 19 20

"'*

3250

soil

,•}

E;^>i;nment mom A

(tx)thriTBl heal <: Ubdcrgivuind [ neasuremmit po

log

t—.-H 1250

soil map

6O0O

lixpcf iniait ruom A Fun coil unit Cioling 3.000 kcal/h Healing 3.800 kcal/h Experiment riHim B KudiHUon panel 39linc 1.940 I,/h Ctwiing .1.000 kcal/h lieating 1,800 kcal/h

Wnit-hmir mdcr

Oeothennal heat pump Cooling:4.6kW(CoIdwaior:12/7''C,CooKngwatcr:30O5'^) Hcating;5 7 kW (Hoi \witer:40/45'<:. Heal souce water: 14/ Q'C) l'ileclric consumption, i .6 kW (Cooling). 1.8 kW lUtatmg) lit) V Refrigerant. R407C CoUJ and Hot water pump How 33 L/miii.head 12m. electric motor outpulfl.25kW Heat xouce water pump now:33 L/miii,head:25m. electriu molor output tl 75 kW

Figure 2

( »Hn-plaix cnncrai: p i c dijimJar:! i m. flengthiOin) L-bcnds 8 pet ptk

Plan ofthe experimental institution and system configuration.

Experiment Outline The heat pump in this system operated from 9:00 to 18:00. Monday to Friday, as in typical office buildings. It was not operated on Saturday or Sunday. In summer (from June to September), heat was discharged (sunk) into the ground. Conversely, in winter (from December to March), heat was extracted (sourced) from the ground. A list ofthe measurement items is shown in Table I. COOLING AND HEATING RESULTS IN 2003 Underground Temperature Figure 3 shows the variations in the underground temperature at measuring points A and B and air temperature in 2003. The subterranean temperature was about 20°C at I m below ground level (G.L. -1 m),about 19°CatG.L.-10m,andabout 560

'Ivcilntcnc luhc iiuick : ;g.Binm.lHiglhi 20m

17°C at G.L. -19 m in both measuring points A and B. as shown in Figure 6, at the start ofthe air-conditioning operation (7/16). The subterranean temperature at each point gradually rose thereafter. The subterranean temperature at G.L. -1 m had reached about 25°C by the time the air-conditioning operation ended. The subterranean temperature at G.L.-I m was significantly influenced by the ambient air temperature. However, subterranean temperatures only changed a few degrees at G.L. -10 m and G.L. -19 m throughout the year. At the start ofthe heating operation (12/25), the subterranean temperature at G.L. -10 m was about I9°C, while that at G.L. -19 m was about 17°C. The subterranean temperatures at each point fell gradually after the start of this operation. The subterranean temperatures at G.L. -10m and G.L. -19 m stabilized at about 15°C during the heating operation in February and remained nearly constant until the operation ended (3/28). ASHRAE Transactions

Table 1.

The Measurement Items

Measured Item

Measuring Equipment (Permissible Range)

Measurement Point

Subterranean temperature

T-type thermocouples (±1°C)

Depth: 1 m, 10 m. 19 m

U-bend surface temperature

T-type thermocouples ( i l ' C )

Depth: 1 m, 10 m. 19 m

Heat source/sink water temperature

Platinum measurement resistor (±0.5°C)

In the pipe

Cold and hot water temperature

Platinum measurement resistor (±0.5°C)

In the pipe

Water flow

Flow meter (±2%)

In the pipe

Electrical power used

Electric power meter

Power panel

Outside temperature, relative humidity, wind velocity, wind direction, quantity of solar radiation, rainfall

Measure Point B Cooling

Heating

a «i

Foundalioo KleA 1250 6/28

Figure 3

8/17

10/6

11/25

1/14

3/4

4/23

6/28

8/17

10/6

11/25

1/14

3/4

4/23

Underground and air temperature variations (lejt, point A; right, point B).

Heat Source/Sink Water Temperature The air temperature and heat source/sink water temperatures for cooling and heating are shown in Figure 4. The heat sink water temperature at the start of the cooling operation was about 20°C, rose gradually after that., and reached about 29^0 just before the end of the cooling period. The average temperature of the heat sink water and air during the cooling period were about 24.5°C and 29.2°C, respectively. The heat sink water temperature was about 4.7*^0 lower than the air temperature on average for the cooling period, and the maximum difference between the water and air temperatures was 12,3°C for the cooling period. On the other hand, the heat source water temperature at the beginning of heating was about I7°C and fell gradually after the start of the operation, and remained at about i3°C from early January until the end of March. The average air temperature during the heating period was about 9.9°C, while the minimum was about 1.1 °C. The heat source water was about 3.1 ''C higher than the air temperature on average for the heating period. The maximum difference between the water and air temperatures was 11.9°C for that period. Thus, using the groundwater as a heat source or sink was more ASHRAE Transactions

effective than using ambient air. Accordingly, GSHP is expected to be more effeetive than ASHP both in terms of cooling and heating.

Heat Extraction/Rejection from/into the Ground The averages for the heat extraetion and rejection either from or into the ground of the foundation piles A and B at the outset ofthe cooling and heating periods are shown in Figure 5. The maximum values for heat rejection were 158 W/m (pile A) and 164 W/m (pile B), while the average values for heat rejection were 100 W/m (pile A) and 120 W/m (pile B) while cooling. The heat rejection per paired U-tube was about 12.515 W/m during the cooling period. Pile B seems to have higher heat rejection than Pile A. Possible reasons for this finding could be that the iocal soil thermal properties around pile A and B are different, mass flows in U-tubes around pile A and B are different due to the difference of pipe friction, etc. The maximum values for heat extraetion were 119 W/m (pile A) and 124 W/m (pile B), while the average values were 44 W/m (pile A) and 52 W/m (pile B) while heating. Heat extraetion per paired U-tube was about 6-7 W/m during the heating period. 561

9 X

X X

X

X

** X

X

'^ H<

X-'''emperaiure( 13; 00)

yx X

'^

^ : Heal Source Water TempefHiure

.tllf

* » '

f=

y.

2

X :Temp«ature(n:OO) ^ : Heat Sink Water Tempeialure

-

6/2S

7/18

Figure 4

8/7

W16

8/27

10/6

12/5

1/14

2/3

2/23

3/14

4/3

4/23

Heat sink/source water temperatures—left, cooling (sink): right, heating (source).

Average of

I

12/25

Air-TemMrBtir

V \

/

^

*^ ><

\

/

X

X X A

A

A

'A

A

X A

A A *

A *

'Ay

A

X ^\^

X

A

Average <^ Air-Temperature

1 ^ A

1 r* A

•

8. -

A

A'*i

/

*

**

•5

A.

A

A-

A

•" A

L

A

A

• :PileA 7/8

7/18

7/28

X

X •^

Pile Ii 8/7

8/17

8/27

9/6

9/16

9/26

10/6

12/5

12/25 I/I4

if A* V *

2/3

2/23

* 11

3/14

4/3 4/23

Figure 5 Heat extraction/rejection from/into the ground— left, cooling (sink): right, heating (source).

Coefficient of Performance The coefficients of performance (COP) for this system, the heat sink/source water temperature and air-conditioning load while cooling and heating., are shown in Figure 6. The data were calculated as the total in an hour from values measured at 10-minute IntetA'als. While cooling, the tnaximum COP was 6.4. while the average was 3.7. When the airconditioning load was high or the heat sink water temperature was low in August, the COP recorded high values. While heating, the maximum COP was 5.0, while the average was 3.2. When the air-conditioning load was high or the heat source water temperature was high in January, the COP recorded high values. The results showed that COP was affected more by the air-conditioning load than by the source/sink temperature. • Although the expected average heat extraction/rejection values were 160 W/m per pile (i.e., 20 W/m per U-tube), the actual value was much less than expected. It is thought that this was due to the low air-conditioning load on this system; therefore, an additional load was installed, and the cooling experiment was repeated in 2004. These resuhs will be described in next section. 562

Figure 6 shows both trends during each day and a seasonal trend. As to a seasonal trend, the results of each month are shown in different symbols in this figure. The plotting with the same symbols shows a trend during each day. The experiment rootn, where a dummy load of 2.5 kW was placed, was controlled at a constant air-conditioning load. However, the actual load is not maintained at a constant level because it is dependent on outdoor temperature, solar radiation, and other factors. COOLiNG RESULTS IN 2004 The cooling results in 2004 are shown in Figures 7-9. Figure 7 shows heat rejection dropping during each day. Figure 8 shows sink temperature increasing during each day, and Figure 9 shows the COP drop during each day. The maximum values for heat rejection were 259 (pile A) and 278 W/m (pile B), while the average values were 204 {pile A) and 220 W/m (pile B) from 6/16 to 8/21. The average values for heat rejection were I86(pile A) and201 W/m (pile B) while cooling. These attained the authors' expectations (160 W/m per pile). Here, the COP for ASHP was calculated from the air temperature measured at the experimental site and the performance curve of ASHRAE Transactions

X July ^ Augast * September

15

20

25

30

35

0,0

Heal Sink Water Temperaturef C]

20

40

Load of Air-Conditioning(kW]

0

A O

(m

o X 10

12

14

16

IS

80

60

20

0.0

Heat Source Water Teniperaturet°C]

2.0

4.0

:Deceniber January February iMarch 6.0

8.0

Load of Air-Conditioning[kW]

Figure 6 Coefficient of performance (top, cooling; bottom, heating).

a typical ASHP. The COPs for this system (GSHP) and the ASHP are shown in Figure 10. The average COPs for this system and ASHPs were 4.89 and 2.90, respectively, while cooling. Thus, this system is about 1.7 times more efficient than the more common ASHP systems.

CONCLUSIONS 1.

The authors have developed a GSHP system using the castin-place concrete pile foundations of a building as heat exchangers in order to reduce the initial boring cost.

2.

In this system, eight U-tubes are arranged around the outer surface of cast-in-place concrete pile foundations.

3.

The heat exchange capability of this system, the subterranean temperature change, and performance of the heat pump were investigated in a tijll-scale experiment.

4.

The average values for heat rejection were 186-201 W/m (per pile, 25 W/m per pair of tubes) while cooling.

5.

The average COP for this system was 4.89 while eooling, so this system is about 1.7 times more efficient than the more common ASHP system.

6.

The initial cost of construction per heat extraction and rejection unit is US$0.79/W (approximately ¥79/W) for this system, whereas it is US$3/W (approximately ¥3()0/W) for the standard borehole system.

7.

This system is expected to be commercially viable.

EXAMINATION OF CONSTRUCTION COST A comparison ofthe construction costs between the usual borehole system and our proposed system is shown in Table 2. Here, a single U-tubc is assumed to be used in the usual borehole system. The heat extraction and rejection per unit length of a single U-tube is assumed to be 40 W/m. The boring cost is ordinarily US$IO()/m (approximately ¥ 10,000/m) in Japan. The heat extraction and rejection capabilities ofthe proposed system are based on the cooling experiment performed in 2004. The construction cost for the proposed system is based on an example introduced in an actual building. The cost of construction per heat extraction and rejection unit of the proposed system is 75% cheaper than that of a borehole system. Accordingly, the proposed system is expected to pay for itself within ten years. ASHRAE Transactions

563

-~ a-

o S

Figure 7 Heat rejection into the ground in 2004.

Figure 8 Air temperature and heat sink water temperature in 2004.

564

ASHRAE Transactions

'

Figure 9

'

I

*

I

•

I

•

I

'

t

_J_L_L.±.J_L

'

I

I

I

I

r

t

I

' '

i I I 1 1 I I I, I

COP change for cooling in 2004.

3SHP

•

• •

^i no 'IIV-I tin. »H. ASHP

Figure 10 Units COR

Table 2.

Comparison of Cost

Form of Heat Exchange

Borehole Type (Single L-Tube)

Proposed System (Cast-in-Place Concrete Pile Type) (8 Pairs of L-Tubes)

Heat extraction and rejection per unit of heat exchange, W/m

40

200

Boring costs, US$/m Piping eosts, US$/m

71

Additional labor costs for foundation and piling work, 87

US$/m Total cost, USS/m

120

158

Total cost per extraction and rejection heat unit, USS/W

3

0.79

ASHRAE Transactions

565

8.

9.

Regarding operational problems related to this approach, the amount of work to connect U-tubes to reinforcing bars cannot be disregarded. Simplification of the construction method will be examined in future research. The experimental results of this research were obtained at a particular local site. A lot of other research will be required in the future to develop this system to apply generally to real office buildings.

REFERENCES Arup Geotechnies. 2002. DTI Partners in Innovation 2002. Ground storage of building heat energy. Overview report. Hamada. Y., K. Ochifuji. K. Nagano, and M. Nakamura. 1997. Study on the heating and cooling by long-term

566

heat storage with underground vertical U-tubes. Pro-

ceedings ofMEGASTOCK '97 1:37^2. Kavanaugh, S.P. 1992. Field test of a vertical groundcoupled heat pump in Alabama. ASHRAE Transactions 98(2):607~15. Kavanaugh, S.P., and K. Rafferty. 1997. Ground-Source Heat Pumps. Design of Geothermal Systems for Commercial and Institutional Buildings. Atlanta: American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc. Presetschnik, A., and H. Huber. 2005. Analysis of a ground coupled heat pump heating and eooling system for a multi-story office building. Proceedings ofthe Sth International Energy Agency. Heat Pump Conference 2005,

pp. 4-8.

ASHRAE Transactions