Download experimental comparison of four borehole heat exchangers

EXPERIMENTAL COMPARISON OF FOUR BOREHOLE HEAT
EXCHANGERS
J. ACUÑA(a), B. PALM (b)
The Royal Institute of Technology, Brinellvägen 68,
Stockholm, 100 44, Sweden, fax number 0046-8-20416
(a)
[email protected]
(b)
[email protected]
KEYWORDS: Borehole Heat Exchanger, Borehole Thermal Resistance, Pressure Drop
ABSTRACT
The most common way to exchange heat with the bedrock in ground source heat pump applications is
circulating a secondary fluid through a closed U-pipe loop in a vertical borehole. This fluid transports the
heat from the rock to the ground source heat pump evaporator. The quality of the heat exchange with the
ground and the necessary pumping power to generate the fluid circulation are dependent on the type of fluid
and its flow conditions along the pipe. Four different borehole heat exchangers are tested using ethyl alcohol
with 20% volume concentration. The fluid temperatures are logged at the borehole inlet, bottom, and outlet.
The collectors are compared based on their borehole thermal resistance and pressure drop at different flow
rates. The results indicate that the pipe dimensions play an important roll, spacers might not contribute to
better heat transfer, and inner micro fins in the pipes improve the performance of the collectors.
1. INTRODUCTION
The ground is an attractive energy source for heating and cooling applications. It has near constant
temperature along the year regardless of the ambient temperature variations, which allows having a reliable
and long lasting resource which permits delivering a constant amount of heat with stable temperatures over
the years. Profit from the ground energy is normally obtained by means of a Borehole Heat Exchanger
(BHE) installed in vertical boreholes. The most popular BHE consists of a closed U-pipe, through which a
secondary fluid travels down and then up while collecting the heat from the bedrock. The collected energy is
subsequently delivered to a so called ground source heat pump (GSHP). The heat transfer between the
secondary fluid and the surrounding ground depends on the arrangement of the BHE flow channels, the
convective heat transfer in the ducts, and the thermal properties of the BHEs as well as the filling material
(groundwater in this case). The thermal resistances associated with these different parts are normally added
together and called borehole thermal resistance Rb, normally determined between the secondary fluid and the
borehole wall.
An increase of one degree Celsius of the out coming fluid from the borehole can give a COP increase in the
heat pump of approximately 3%. Therefore, it is of great importance to use the most appropriate BHE in
order to efficiently extract heat from the ground. This paper presents a comparison of four different Borehole
Heat Exchangers which are tested at the same heat pump installation.
Considering the borehole as a cylindrical heat exchanger with two inner tubes inserted into a shell (the
borehole wall), an energy balance could be made by equalizing the absorbed heat by the fluid with the
Newton’s law of cooling, as presented in equation 1. Since the fluid is heated as it travels through the BHE
channels (in the heating mode, which is the case for the actual study), the temperature arises between
different points when the fluid travels down and upwards.
•
q ' = ρ V CpΔT =
1
ΔTml
Rb
equation 1
ΔTml =
(Tbhw − Tin ) − (Tbhw − Tout )
(T − Tin )
ln bhw
(Tbhw − Tout )
equation 2
8th IIR Gustav Lorentzen Conference on Natural Working Fluids, Copenhagen, 2008
The left hand side of equation 1 represents the heat absorbed by the fluid between two temperature
measurement points. This heat will vary in accordance with the flow conditions inside the collector pipes.
The right hand side of equation 1 represents the overall heat transferred from the borehole wall to the
secondary fluid, where the logarithmic mean temperature difference is calculated as expressed in equation 2.
Since the absorbed heat is calculated for the downward and upward channels, the values assigned to Tin and
Tout will be the inlet and bottom temperature for the downward collector pipe, and the bottom and outlet
temperature for the upward collector pipe, respectively. The heat absorbed by the fluid obtained with
equation 1 is divided by the total length of the studied section (the active borehole length) in order to
estimate the heat extraction per meter [W/m].
As expressed in (Hellström 1991), there is a convective heat transfer at the borehole wall, and heat
conduction in the pipe walls and filling material (groundwater in this case). There are also contact resistances
between the different materials at the interfaces. It is of course desirable that BHEs have as low Rb as
possible. The convection in the secondary fluid side during laminar flow may give rise to higher thermal
resistances between the secondary fluid and the BHE pipes than in turbulent flow. Therefore, it is usually
desired to keep the flow within the turbulent region. However, the pumping power required to induce
turbulence must be regulated so that the best heat transfer conditions are achieved in the borehole at the
lowest energy cost, i.e. using the smallest amount of energy to pump the flow. The pumping power is
proportional to the pressure drop in the BHE channels, as expressed in equation 3.
•
E pump =
ΔP ⋅ V
ΔPf = f
η pump
ρ ⋅ w2 L
2
⋅
D
equation 4
equation 3
It is strictly necessary to account for this power that will, in the long term, imply an energy cost. The
pressure drop ∆P in the pipes occurs mainly due to friction and it is greater at higher velocities. It can be
estimated for a fully developed flow with equation 4. The friction factor f estimation depends on whether the
flow is laminar or turbulent. This is solved by using equation 5 for laminar flows and the correlation
suggested by (Gnielinski 1976) for turbulent flow, expressed in equation 6.
f =
64
Re
(for Re≤2300)
equation 5
f =
1
(0,79 ⋅ ln Re− 1,64)
2
(for Re≥ 2300)
equation 6
Re =
w⋅ D
ν
equation 7
Where the Reynolds number is calculated as expressed in equation 7. The calculated pressure drop does not
include the losses in bends and turns. It is important to mention that the variations of the secondary fluid
thermophysical properties with the temperature changes inside the BHEs must be considered in order to have
an accurate calculation. These properties are well known from (Melinder 2007).
2. THE TEST RIG
The experimental work is carried out in a heat pump installation where a building with 19 apartments of 40
m2 each is supplied with domestic hot water and comfort heating by two GSHPs of 16 and 32 kW,
respectively. The 16 kW heat pump also supplies comfort heating in case the latter does not have sufficient
capacity to cover the building demand. The energy source is 6 boreholes out of which 4 have been selected
for the present study. The secondary fluid used in this installation is an aqueous solution of Ethanol 20%
volume concentration, with a freezing point of -8 °C (Melinder 2007). The boreholes are water filled, have a
diameter of 140 mm and a depth of 260 meters. Each borehole is equipped with a different BHE. The
borehole main characteristics are tabulated in Table 1.
The boreholes have also been characterized through detailed groundwater measurements in one of the
boreholes, showing that there is not any relevant ground water flow in the installation (Acuña et al, 2008).
Moreover, detailed borehole deviation measurements have been carried out every ten meters along their
8th IIR Gustav Lorentzen Conference on Natural Working Fluids, Copenhagen, 2008
length in order to determine how the borehole diverges from the expected drilling direction. This allows
identifying to what extent the BHE pipes are laying on one of the borehole wall sides, among others. The
total horizontal deviation values are also tabulated in Table 1. Last, but not least, a thermal response test has
been done at 2 of the boreholes, giving indication of the mean undisturbed ground temperature, and the
average ground thermal conductivity. These resulted to be about 8,7 oC and 3,75 W/m K, respectively.
Table 1. General borehole characteristics
Borehole
number
Active
borehole
length [m]
Total horizontal
deviation at 260 from
the start point [m]
2
4
251,6
254,5
84,9
64,1
PE 40x3,7mm
PE 40x2,4mm
5
242,7
75,7
PE 40x2,4mm
6
245,7
96,9
PE 40x2,4mm
BHE details
3 Pipe collector
U-pipe collector
U-pipe with spacers: 1,3 cm separation
between the pipes. Separated 2 m from each
other at the top 50 meters, and 3 m along the
rest of the collector
U-pipe with micro fins in helix form. The
helix direction is inverted periodically along
the BHE length
Figure 1 illustrates the system including 5 boreholes. The inlet and outlet lines of the boreholes are
connected to a common manifold. The volumetric flow in each collector is controlled by means of STA-D
regulation valves located in the return lines and with a frequency control circulation pump. The volumetric
flows are measured in each collector with an inductive instrument of the type Brunata HGS9-R6184/1B2M24, also located in the return lines.
Figure 1. Sketch of the experimental rig
It is also observed in Figure 1 that each borehole heat exchanger is instrumented with thermocouples for
temperature measurements at the bottom and outlet points. Bottom thermocouples are located about half a
meter higher than the collector bottom return line. The inlet temperature is common for the collectors and it
is measured at a point located before the inlet manifold. All the temperature measurement points are located
inside the collector pipes, i.e. the measurement point is in direct contact with the secondary fluid. The
thermocouples used are of the type T, “quick disconnect” TMQSS-M050U-150 from Omega Eng. The total
pressure drop in the collectors are also measured during the tests by connecting an instrument of the type
TESTO 526-2 between to the pressure taps located at the collector inlet and outlet lines.
8th IIR Gustav Lorentzen Conference on Natural Working Fluids, Copenhagen, 2008
3. THE EXPERIMENTS AND FUNDAMENT FOR CALCULATIONS
All the Borehole Heat Exchangers have been operating according to the house demand since the system was
put in operation. The experiments for this article have been carried out during April and May 2008, and the
measurement condition has been that both heat pumps are on so that the same energy demand from the
boreholes is required. The volumetric flow in the boreholes is adjusted previous to the heat pump start up.
Temperatures have been measured at different flow conditions in the BHEs. It was shown in (Acuña et al.
2008) that the temperature conditions along the borehole vary with time during the heat extraction period and
take some time to stabilize. Therefore, the temperature and the flow measurements are taken every 60
seconds and an average is taken for a 30 minutes long period when the conditions have stabilized after the
heat pump start up.
Each minute, the fluid density, kinematic viscosity and heat capacity are calculated at the measured
temperature using the results from (Melinder 2007). Hence, the Reynolds number is calculated with equation
7, obtaining the fluid velocity from the measured flow and the pipe inner dimensions. The friction factor is
estimated with equation 6 or equation 5, and the pressure drop with equation 4. Finally, the heat absorbed per
meter by the secondary fluid is calculated for the down and up going channel of each collector using
equation 1. The latter is immediately used in order to calculate the borehole thermal resistance Rb for each of
the collector channels as well, assuming a temperature value for the borehole wall.
Choosing the appropriate borehole wall temperature for the calculations was relatively difficult. The rock
type at the borehole installation is granite and the undisturbed ground temperature of the site has been
carefully measured every one meter along one of the boreholes before the system was put into operation. The
measurement was done using an optical fiber cable coupled to a Sentinel-DTS instrument from
Hydroresearch AB. The cable is located in BH4, between the collector and the borehole wall. The
undisturbed ground temperature profile is shown in Figure 3, where it can be observed that the rock
temperature has a minimum value at a depth of about 110 m, and that the gradient then becomes positive
with an increase of about 1°C for each 100 m. The groundwater temperature profile during heat pump
operation has approximately the same shape as the undisturbed ground temperature but with lower
temperature values due to the heat extraction from the secondary fluid. This is known thanks to the same
optic fiber cable, which was used to measure temperatures during heat pump operation. This was helpful for
making logical estimations for the borehole wall temperature when calculating the logarithmic mean
temperature difference. Figure 2 shows measured values for the groundwater for a whole heat pump cycle.
It is observed in Figure 2 how the groundwater temperature changes during and after the heat pump
operation cycle. It becomes lower after the heat pump starts due to the heat extracted from the secondary
fluid. Moreover, after the heat pump goes off, it is observed how the groundwater temperatures come back to
the initial level. Accounting for this, it is possible to estimate the borehole wall temperatures.
The borehole wall temperature was assumed to be constant and equal to 7,2 °C for the calculations in this
report. This is of course possible only by neglecting the negative and positive gradient that the ground and
groundwater temperatures present before and after a depth of about 100 meters, respectively. Neglecting the
geothermal gradient is a common assumption according to (Claesson 1988).
The undisturbed ground temperature is determined as shown in Figure 3, where it can be observed that the
average values are in accordance with the average determined by the thermal response test. The Reynolds
numbers obtained during the calculations always have a value above 3300 for all the BHEs, indicating
turbulent flow (Re>2300). The exception is BH2, where the flow on the way up is divided into the two
upward channels and becomes laminar for the flow 1,5 m3/h.
8th IIR Gustav Lorentzen Conference on Natural Working Fluids, Copenhagen, 2008
Figure 2. Groundwater temperature during a heat pump cycle
Figure 3. Undisturbed ground temperature
4. RESULTS AND DISCUSSION
Figure 4 shows the heat extraction values from each borehole at the 3 tested volumetric flows. The heat
extraction is calculated within the active borehole length and not considering the whole borehole depth. The
active borehole length for each BHE is shown in Table 1. This factor makes a small difference when
comparing the collectors.
It can be observed in Figure 4 that the measured heat extraction normally increases with increasing flow
rate. The exceptions are BH2 and BH6 at 2,5 m3/h, where the extracted heat is slightly lower than the one
corresponding to 1,8 m3/h. This could be due to the fact that all the measurements were not carried out
during the same day, which could alter the borehole temperature conditions. Besides this, the inlet
temperature values are not always the same during all the measurements and this creates a lower or higher
temperature difference between the secondary fluid and the ground, which could have influence on the
results.
It can also be inferred from Figure 5 that the borehole equipped with spacers (BH5) does not give any
benefits since the heat extraction values are more or less the same as compared with the normal U-pipe
collector (BH4). These two BHEs were always measured at the same time and at the same flow, which gives
the same inlet temperature when they are measured at each of the volumetric flow rates. An important factor
to mention from Figure 4 is also that, at the two lowest volumetric flow rates, the best heat extraction
performance is obtained in BH6 whilst the worst in BH2.
Nevertheless, it is not recommended to compare the collectors by looking at the extracted heat due to the fact
that not all the measurements were taken at the same time, which could cause different inlet and groundwater
temperatures at different measurement occasions.
Figure 5 shows the calculated borehole thermal resistance values for the downward and upward channels for
each of the borehole heat exchangers at different flows. As previously mentioned, these values are obtained
assuming a borehole wall temperature equal to 7,2 °C. The first fact reflected in Figure 5 is that the borehole
thermal resistance is always higher for the upward channel in the BHEs. This might partially have to do with
8th IIR Gustav Lorentzen Conference on Natural Working Fluids, Copenhagen, 2008
possible thermal shunt circuiting, i.e. heat transferred from the upward to the downward channel. This
situation could change in unsymmetrical BHEs where the down and upward channels are not equal.
Heat Extracted [KW]
Heat extracted at different flows
16,0
14,0
12,0
10,0
8,0
6,0
4,0
2,0
0,0
14,7
14,3
8,6
10,08
10,12
11,9
12,2
9,7
13,7
12,9
11,2
9,5
1,5 m3/h
1,8 m3/h
2,5 m3/h
BH2 (1-2)
BH4
BH5
BH6
Borehole
Figure 4. Heat absorbed by the secondary fluid at different flows
It is worth making a closer comparison of BH4 and BH5. The first 50 meters of BH5 have spacers every two
meters, while the other 210 meters are equipped with spacers every three meters. The work done by (
Claesson 1988) presented some values obtained from a computer model where calculations have been made
for three possible positions of the collector channels in the borehole. The results show that the best results are
obtained when the two channels are separated by a radial distance such that they are almost in contact with
the borehole wall. This is not the case in BH5 where the separation between pipes in BH5 is 13 mm in the
best case. This could be a reason why the thermal performance of BH4 and BH5 is very similar, even
presenting more positive values for the normal U-pipe collector.
Regarding the BHE with inner micro fins (BH6), it is observed that the thermal resistance values are the
lowest of all the collectors, implying that this BHE has the best thermal performance with the exception of at
2,5 m3/h, where BH5 is best. One of the possible small factors why this BH6 shows better performance
might be the exact position of the collector pipes with respect to the borehole wall. It was shown in Table 1
that BH6 is the borehole which deviated most from the vertical, which makes it likely for the tube in this
hole to have better contact with the borehole wall.
Thermal Resistance [K/(W/m)]
Borehole Thermal Resistances (BH wall temp = 7,2 C)
0,30
down 1,5 m3/h
0,25
up 1,5 m3/h
0,20
down 1,8 m3/h
0,15
up 1,8 m3//h
0,10
down 2,5 m3/h
0,05
up 2,5 m3/h
0,00
BH2 (1-2)
BH4
BH5
BH6
Borehole
Figure 5. Borehole Thermal Resistances for down and upward BHE channels at different flow rates
Nevertheless, BH2 is also considerably deviated, but this collector still presents a higher borehole thermal
resistance. This is partially explained by the higher wall thickness of this collector (3,7 mm) that makes the
pipe wall thermal resistance become 0,0752 K/(W/m) instead of 0,0484 K/(W/m), which is the case for the
other collectors. The former is 35% higher than the one corresponding to the latter. The question about which
flow direction to choose in the three pipe BHE, as (Eskilson 1987) states, arises when the upward and
8th IIR Gustav Lorentzen Conference on Natural Working Fluids, Copenhagen, 2008
downward BHE channels are unsymmetrical, such as the case of the three pipe collector. This will be further
investigated later in this research project in order to compare the performance of this BHE when running it in
the opposite direction.
The calculated and experimental values for the pressure drops in the BHEs are plotted in Figure 6, Figure 7,
Figure 8 and Figure 9. With the exception of BH6, it is generally observed that the calculated pressure drop
is slightly lower than the experimental values. This is attributed to the fact that the accessories such as
elbows, bends, bottom part of the collector, are not considered in the calculation.
144,53
129,73
1,5 m3/h
54,21
75,59
75,00
48,35
Pressure drop [KPa]
Experimental
160
140
120
100
80
60
40
20
0
1,8 m3/h
2,5 m3/h
Calculated
Pressure drop [KPa]
160
140
120
100
80
60
40
20
0
BH4
160
140
120
100
80
60
40
20
0
149,60
129,54
87,33
78,22
61,43
56,41
Experimental
BH5
BH6
128,36
81,10
56,54
77,98
56,68
1,5 m3/h
1,8 m3/h
2,5 m3/h
Experimental
Calculated
Figure 8. Pressure drop in BH5
160
140
120
100
80
60
40
20
0
1,8 m3/h
Calculated
Figure 7. Pressure drop in BH4
149,63
1,5 m3/h
2,5 m3/h
Figure 6. Pressure drop in BH2
Pressure drop [KPa]
Pressure drop [KPa]
BH2
131,51
70,03
49,10
Experimental
136,20
77,31
58,02
1,5 m3/h
1,8 m3/h
2,5 m3/h
Calculated
Figure 9. Pressure drop in BH6
The calculated and the experimental pressure drop values for BH6 are very similar to each other,
demonstrating that neglecting the micro fins roughness is a good approximation. Moreover, it is observed in
Figure 9 that the calculated value is higher than the experimental one, which unexpectedly shows that the
real pressure drop in the collector with micro fins is in fact lower. It is in fact shown that the BH6 has the
lowest pressure drop of all the BHEs, including BH4 and BH5 which are common U-pipe collectors with the
same dimensions. This is surprising, and it is a reason to study further the micro fins arrangement.
The difference between the calculated values for BH4, BH5 and BH6 are attributed to the fact that the flow
rate and the temperature levels were not always exactly the same.
BH2 presents pressure drop values relatively equal to the other collectors. This would not be expected due to
the fact the flow on the way up is divided in two channels, which would reduce the pressure drop on the way
up. However, the pipe inner diameter is 32,8 mm and not 35,2 mm as for the other collectors, and this will
cause the opposite effect over the pressure drop as expressed mathematically in equation 2.
The pressure drop analysis indicates that BH6 is the best option, since the required pumping power for the
secondary fluid would be slightly lower for this collector.
5. NOMENCLATURE AND ABBREVIATIONS
BH2…
BH4…
BH5…
BH6…
BHE…
Borehole 2 – Three pipe collector – mode: 1 pipe down and 2 up
Borehole 4 – U pipe collector
Borehole 5 – U pipe collector with spacers
Borehole 6 – U pipe collector with inner micro fins
Borehole Heat Exchanger
8th IIR Gustav Lorentzen Conference on Natural Working Fluids, Copenhagen, 2008
GSHP…
Ground Source Heat Pump
Epump…
Rb…
q’…
Tbhw…
d…
ρ…
Cp…
Pumping power [W]
Borehole Thermal Resistance [K/(W/m)]
Heat absorbed by the secondary fluid per meter active borehole [W/m].
Borehole wall temperature [°C]
Distance between temperature measurement points [m]
Fluid density
Fluid specific heat capacity [J/KgK]
•
V…
∆T…
∆P…
∆Pf…
f…
ηpump…
w…
L…
D…
Re…
ν…
∆Tml…
Tin…
Tout…
Tug…
Secondary fluid volumetric flow [m3/s]
Temperature difference between two points [K]
Pressure drop [Pa]
Pressure drop due to friction [Pa]
Friction factor
Pump efficiency
Velocity [m/s]
Total tube length [m]
Inner pipe diameter [m]
Reynolds number
Kinematic viscosity of the fluid [m2/s]
logarithmic mean temperature difference [K]
Inlet temperature at the down or up going channel [K]
Outlet temperature at the down or up going channel [K]
Undisturbed ground temperature [°C]
REFERENCES
Acuña J. Björn Palm, Peter Hill (2008) Characterization of boreholes - results from a U-pipe borehole heat
exchanger installation. IEA Heat Pump Conference. s4.19 . Zurich, Switzerland
Claesson, J. P. E. (1988). "Conductive heat extraction to a deep borehole: Thermal Analyses and
Dimensioning Rules." Energy 13(6): 509-527.
Eskilson, P. (1987). Thermal Analyses of Heat Extraction Boreholes. Department of Mathematical Physics.
Lund, Sweden., Lund Institute of Technology.
Gnielinski, V. (1976). "New equations for heat and mass transfer in turbulent pipe and channel flows." Int.
Chem. Eng. 16: 359-368.
Hellström, G. (1991). Ground Heat Storage. Thermal Analyses of Duct Storage Systems. Theory. Dep. of
Mathematical Physics. Lund, Sweden, University of Lund, Sweden.
Melinder, Å. (2007). Thermophysical Properties of Aqueous Solution Used as Secondary Working Fluids.
Energy Technology. Stockholm, The Royal Institute of Technoloy. Doctor.
8th IIR Gustav Lorentzen Conference on Natural Working Fluids, Copenhagen, 2008