Lab 6: Heat Pipe

A Lab Report

Submitted By

Neel Nadpara

ME 3264 Applied Measurements Laboratory

Section 004L

Department of Mechanical Engineering

University of Connecticut

Lab Preformed On:

March 9, 2015

1|Page
Summary:
There are numerous engineering scenarios that require the transfer of heat from one location to
another; this heat transfer can be achieved via the use of heat pipes. Heat pipes are an effective
way to transfer heat from one solid interface location to another by means of thermal
conductivity and phase transition. A heat pipe is a passive heat transfer device that can transfer
heat with a much smaller temperature drop as compared to high thermal conductivity materials
such as copper. In general, heat pipes are consisted of a shell (wall), a wick and a small amount
of working fluid. At the hot interface (evaporator section) the working fluid inside the heat pipe
absorbs thermal energy from the heat source causing the fluid to vaporize. This vapor then
travels along the heat pipe to the extreme end of the device to the cold interface (condenser
section) where the fluid condensates back into a liquid thereby releasing the latent heat. The
working fluid returns back to the evaporator by means of a wick which forms a capillary force.
Since heat pipes rely on heat transfer by boiling and condensation for which the heat transfer
coefficients are very high, they are very effective thermal conductors. In this heat pipe lab our
goal was to understand the heat transfer characteristics of a heat pipe, observe the thermal
response time of a heat pipe compared to a solid copper rod, measure the response time and
temperature profile across a heat pipe, and calculate the effective thermal conductivity of a heat
pipe compared to a solid material with a high thermal conductivity. In this lab seven
measurements of temperature were taken at seven different points along the heat pipe over a time
period. Three k-type thermocouples were positioned underneath the heater element along the
evaporator section and four k-type thermocouples were fitted underneath the condenser section.
Monitoring the readings of temperature from LabVIEW, the thermal power input to the heat pipe
was supplied by a heater element that was controlled by the output voltage of an autotransformer.
Using these reading, the effective thermal conductivity of the heat pipe was calculated using the
equation:
𝐿 𝐿
( 𝑒 +𝐿𝑎 + 𝑐 )𝑞ℎ𝑝
𝑘𝑒𝑓𝑓 = (𝑇 2 2
. (Equation 1)
𝑒,𝑎𝑣𝑔 −𝑇𝑐,𝑎𝑣𝑔 )𝐴𝑐

What we concluded from this lab is that heat pipes are very effective heat transfer devices and
are much more efficient in transferring thermal heat than their solid counterparts based on the
working fluid within and the environment and temperature differences in which they operate.
Moreover, based on the results in the lab the effective thermal conductivity of the heat pipe is
dependent on the orientation of the heat pipe, with a more consistent and reputable reading
coming from the heat pipe in the horizontal position while the vertical heat pipe had a higher
thermal conductivity as heat input increased due to gravitational effects that are present in the
vertical orientation.

2|Page
Experimental Procedure:

The first thing that we did in order to start this lab was to make qualitative observations between
a non-instrumented heat pipe (copper-water and glass-acetone heat pipes were tested) and a solid
copper rod. The goal was notice the difference in rate of heat transfer between a copper rod and
an un-instrumented heat pipe. This was done by dunking the copper rod and the heat pipe rod
simultaneously into hot water (heated to a boil via a hot plate), waiting until the rods heated up,
and then dunking the rods in cold water (ice water), until they became cold to the touch.
However, for the glass-acetone heat pipe the heat pipe was allowed to cool prior to being dunked
in the cold water to prevent the glass from cracking due to thermal shock. Additionally, since a
heat pipe that was cut on one side to show the internal wicking structure was given it was also
tested to be compared as a hollow copper tube to the previous examples. The procedure was
designed to show how efficiently the different rod/heat pipes were in effective heat transfer
versus time.

In the next step, the core of the lab was started. A 0.20 m horizontally oriented copper-water heat
pipe is obtained. A flexible heater with a width of 45 mm is wrapped around one end of the heat
pipe and is secured while the other 155 mm of the heat pipe is left exposed to ambient air. The
heated section acts as the evaporator while the unheated section acts as the condenser which is
cooled by natural convection. Three k-type thermocouples were attached to the heated
evaporator section (TC1-TC3) under the heater while 4 k-type thermocouples were secured to
the condenser section of the heat pipe (TC4-TC7). A schematic of the set-up can be seen below
in Figure 1. The heat pipe thermocouple assembly was connected to a data acquisition system
which is linked to LabVIEW. This facilitated the measurement for ambient temperature of the
heat pipe in the horizontal position. Once the readings were all relatively equal, data acquisition
was started and the heater element was turned on. With the heater element turned on, power was
added to the heat pipe system by adjusting voltage and current. The initial reading was
application was of 6.0±.1 W and was applied to the evaporator section of the heat pipe. Power
was calculated using the equation seen below:

Power = Voltage × Current (𝐏 = 𝐕𝐈) (Equation 2)

Power in this case was also the heat input and was referred to qhp in our calculations. After
waiting for 15 minutes or when the temperature readings in each section leveled off to steady
state, readings of temperature were recorded from the thermocouples. This process was repeated
until the temperature in the heat pipe reached over 150°C. Once the temperature reached 150°C,
the heat pipe was turned vertically and the power stepped down back to 6.0±.1 W systematically
as temperature fluctuations were recorded at 15 minute time intervals to allow the system to
reach steady state. The purpose of the measurements was to quantify the performance of the heat
pipe and compare it to other heat transfer devices. This quantification is known as the effective
thermal conductivity or keff. Thermal conductivity is calculated by determining the thermal

3|Page
resistance of the heat pipe. Using the equation for thermal resistance (as seen below) the keff
(effective thermal conductivity) can be solved for (also seen below).

𝐋𝐞 𝐋
Te,avg −Tc,avg +𝐋𝐚 + 𝐜
𝟐 𝟐
rhp = = (Equation 3)
qhp 𝐤 𝐞𝐟𝐟 ×𝐀𝐜

𝐋 𝐋
( 𝐞 +𝐋𝐚 + 𝐜 )𝐪𝐡𝐩
𝐤 𝐞𝐟𝐟 = (𝐓𝟐 𝟐
(Equation 4)
𝐞,𝐚𝐯𝐠 −𝐓𝐜,𝐚𝐯𝐠 )𝐀 𝐜

Figure 1: Schematic of the horizontal heat pipe setup (the heat pipe is just flipped for the
vertical setup). Also the left section acts as the evaporator while the right acts as the
condenser.

Results and Discussion:

Qualitative Observations:

For the qualitative portion of the experiment, after dunking the copper rod and the un-
instrumented heat pipe into the boiling water we observed the speed in which the heat pipe
increased in temperature as opposed to the copper rod. Based on what we felt, it took close to 10
seconds for the heat pipe to reach a point where the its temperature was almost untouchable
while the copper rod spent close to a minute to experience the effects of the hot water. Similarly
when dunking both rods in the cold water, the heat pipe’s change in temperature was almost
immediate while the copper rod took quite a bit of time to cool down. With the glass heat pipe
the trend of quick heating and cooling continued despite to poor thermal conductivity of glass.
When the hollow copper was subjected to the same treatment, the hollow pipe heated up and
cooled down faster than the solid copper but slower than the heat pipe. These results make sense

4|Page
since a heat pipe should be able to transfer heat faster than a solid material since they use boiling
and condensing fluids which have much higher thermal conductivities compared to that of solid
materials. Also, the glass heat pipe should perform better than a solid copper rod for the same
reason, however it preforms poorly compared to the copper heat pipe as the copper in the shell of
the heat pipe has a higher thermal conductivity allowing heat to be transferred to the working
fluid much faster. Additionally, the hollow copper pipe should perform better than the solid
copper rod since it has less mass to heat up, however it will perform worse than the heat pipes
since it still relies on the thermal conductivity of the solid material.

Quantitative Results:

Heat Pipe Evaporator Time vs. Temperature

200
180
160
Temperature (°C)

140
120
100 TC1
80 TC2
60 Horizontal Setup Vertical Setup TC3
40
20
0
0 1000 2000 3000 4000 5000 6000
Seconds (Sec)

Figure 2: Graph showing the time vs. temperature of the evaporator section of the heat
pipe.

5|Page
 Heat Pipe Condenser Time vs. Temperature
180
160
140
Temperature (°C)

120
100 TC4
80 TC5

60 TC6
Horizontal Setup Vertical Setup
40 TC7

20
0
0 1000 2000 3000 4000 5000 6000
Seconds (Sec)

Figure 3: Graph showing the time vs. temperature of the condenser section of the heat pipe.

Average Heat Exchanger Sections Time vs. Temperature

200
180
160
140
Temperature (°C)

120
100
Average Evaporator
80
Average Condenser
60
Horizontal Setup Vertical Setup
40
20
0
0 1000 2000 3000 4000 5000 6000
Seconds (Sec)

Figure 4: Graph showing the time vs. temperature of the evaporator and condenser
sections of the heat pipe so that they may be compared.

6|Page
 Voltage, Current, and Calculated Power
Input Measured Calculated
Current (A) Voltage (V) Power (W)
0.52 11.4 5.928
Horizontal Orientation 0.74 16.2 11.988
0.9 19.9 17.91
0.74 16.3 12.062
Vertical Orientation
0.52 11.4 5.928

Horizontal Heat Pipe

K Effective (W/m*k)
Average Evaporator

Average Condenser

Resistance of Heat

Area of Heat Pipe

Diameter of Heat
(TC1, TC2, TC3)

(TC4, TC5, TC6,

Difference (°C)

Evaporator (m)
Heat Input (W)

Condenser (m)

Pipe (ohms)
TC7) (°C)

Length of

Length of

Pipe (m)

(m2)
(°C)

5.928 79.09019 76.58607 2.50412 0.155 0.045 0.422 0.005 2.00E-05 11836.5
11.988 126.8317 122.2727 4.55909 0.155 0.045 0.380 0.005 2.00E-05 13147.3
17.91 160.1455 153.4644 6.6811 0.155 0.045 0.373 0.005 2.00E-05 13403.5
Average K Effective 12795.8

Vertical Heat Pipe

K Effective (W/m*k)
Average Evaporator

Average Condenser

Resistance of Heat

Area of Heat Pipe

Diameter of Heat
(TC1, TC2, TC3)

(TC4, TC5, TC6,

Difference (°C)

Evaporator (m)
Heat Input (W)

Condenser (m)

Pipe (ohms)
TC7) (°C)

Length of

Length of

Pipe (m)

(m2)
(°C)

12.062 173.3435 167.3421 6.00139 0.155 0.045 0.498 0.005 2.00E-05 10049.3
5.928 130.7617 127.4974 3.26435 0.155 0.045 0.551 0.005 2.00E-05 9079.92
Average K Effective 9564.63

7|Page
 Input Power to Heat Pipe vs. Temperature Drop
8

6
Teperature Drop (°C)

4
Horizontal Setup
3 Vertical Setup

0
0 5 10 15 20
Input Power (W)

Figure 5: Graph showing the input power vs. temperature drop of the of the heat pipe for
the horizontal and vertical setups.

Input Power vs. Thermal Conductivity

16000
Effective Thermal Conductivity (W/m*K)

14000
12000
10000
8000
Horizontal Heat Pipe
6000
Vertical Heat Pipe
4000
2000
0
0 5 10 15 20
Heat Input (W)

Figure 6: Graph showing the input power vs. thermal conductivity of the of the heat pipe
for the horizontal and vertical setups.

8|Page
Discussion:

The graph in Figure 4 above shows the time vs. temperature graph of the condenser and
evaporator as power into the heat pipe was increased. From this graph we can observe the
closeness in shape of the temperature distributions in the pipe. We also notice the gradual
increased difference in temperature as a greater amount of power, and therefore thermal energy
was definitely added to the system. Similar observations can be seen in the vertical heat pipe
temperature vs. time side of the graph as from the horizontal heat pipe graph. In this case power
was systematically decreased in the system and therefore temperature decreased in a similar
fashion. Based on the graph we also saw about a 10°C difference in temperature between the
horizontal and vertical heat pipes in favor of the vertical for similar power input values. Based on
the input power vs. temperature drop curve in Figure 5 we can see a strong linear relationship
between both variables. Based on this graph we also notice a greater change in temperature
between the horizontal and vertical positions of the heat pipe based on power inputs with a
greater change occurring in the vertical position than the horizontal position. Using measured
values, thermal conductivity was calculated by using the average temperatures in the condenser
and evaporator at each time interval along with known power inputs and dimensions of the heat
pipe. There are only two variables that change when you calculate the effective thermal
conductivity. They are the input power and the temperature drop (the temperature difference
between the condenser and evaporator sections of the heat pipe). For the horizontal setup, the
input power for each of the heating cycles were 6, 12, and 18W while the thermal conductivities
were 2.36 W/K, 2.62 W/K, and 2.68 W/K. Typically thermal conductivity is assumed to be
constant, however in this lab the values change. This is surprising considering that the heat
addition and removal in this experiment are relatively low. In metals conductivity is primarily a
result of free elections. In this case the heat pipe’s shell is likely a copper alloy as in alloys the
change in electrical conductivity is usually smaller and thus thermal conductivity increases with
temperature which is what occurred in our case. Moreover, as a greater amount of heat is lost as
the heat input increases the effective thermal conductivity would increase. Also, based on the
input power vs. effective thermal conductivity in Figure 6 we can see that the relative position of
the heat pipe had to do a lot with values of thermal conductivity. For the most part thermal
conductivity in both heat pipes had a consistent slope. In both cases as power increased the
effective thermal conductivity also increased. However, when comparing the horizontal to the
vertical it becomes very obvious that the vertical heat pipe preforms poorly. To illustrate, when
the input power is 12±.1 W in the vertical setup the thermal conductivity is 2.01 K/W while in
the horizontal setup the thermal conductivity was 1.82 K/W with a power input of 6 ±.1 W. This
is likely due to the fact that gravity is now at play. Gravity would work against the vaporized
working fluid trying to travel from the evaporator to the condenser this in turn would lead to
slower heat transfer meaning that more heat can be transferred to the surroundings. Moreover, a
vertical orientation will enhance the capillary force in providing condensed working fluid back to
the evaporator section. This in turn means that there is more fluid at the evaporator than there
would be in a horizontal setup. This would be ideal for a high temperature heat transfer

9|Page
application. However, in this situation a large amount of fluid in the evaporator would mean that
it would take longer to vaporize the working fluid leading to a slower heat transfer rate which in
turn could lead to more heat lost to the surroundings. The largest source of calculation error has
to do with power input due to it having the greatest uncertainty of the tree primary contributors
for error.

Data and Error Analysis:

The primary focus on this lab was to determine the thermal conductivity of the heat pipe.
Uncertainty was introduced in this lab with power measurements, measurements of length and
diameter, and temperature measurements. For power measurements uncertainty was ± 0.1 W, for
length measurements it was ±0.01 mm, and for temperature measurements uncertainty was
±0.001oC. Some sources of error in the lab came from our power generator. Measurements for
current tended to jump around slightly during the input of power into the heat pipe. Another
source of error could have come from how accurate we were in our measurements for the lengths
of the evaporator and condenser section, which could have affected our calculations for thermal
conductivity. Using our calculations and the RMS formula that is derived below we saw that our
relative uncertainty due to power input tended to decrease as power increased according to the
table below.
1) P = IV

2) δP = δ(IV)
∂P ∂P
3) δP = δI + ∂V δV
∂I (Equation 5-9)
4) δP = VδI + IδV

5) 𝐑𝐌𝐒 = √(𝐕𝛅𝐈)𝟐 + (𝐈𝛅𝐕)𝟐

Uncertainty of Input Power

Input Measured Calculated Uncertainty
Current (A) Voltage (V) Power (W)

Horizontal Orientation 0.52 11.4 5.928 ±0.117

0.74 16.2 11.988 ±0.167
0.90 19.9 17.91 ±0.207
Vertical Orientation 0.74 16.3 12.062 ±0.168
0.52 11.4 5.928 ±0.117

10 | P a g e
There were numerous possible sources of error for this lab. For instance, the heat pipe was
exposed to atmospheric conditions. However, these conditions were not uniform as a door was
open allowing there to be convection currents and wind in the room therefore there could be a
substantial convection heat transfer at certain times while it may be less at other times leading to
error. Moreover, the temperature was monitored by k-type thermocouples. Thermocouples
require proper calibration in order to work properly and they were setup prior to our experiment
meaning there could be error due to this. A more controlled environment would provide more
reliable data with a clearer steady state and the data would have less noise. Also, every
thermocouple type has a specified temperature range for k-types it is typically -200°C to 1250°C.
While the experiment took place in this range, it is important to note that the thermocouple is
most accurate at the median of this range but the experiment took place outside of the median
thus there could be some error due to this. Also, when adjusting the power input there were
changes in both voltage and current. Thus, it was difficult to maintain consistent power increases
and decreases which in turn lead to error. Additionally, it is possible that steady state was never
reached in some situations which is particularly possible considering that a gust of air could
prevent steady state. Furthermore, the method used to determine steady state was to read the
graph on LabVIEW that compares the temperature variation for minute to ensure that the
variation is less than ±.5 ˚C. Thus, if time to reach steady state was increased from 15 minutes to
something like 25 to 30 minutes more reliable data with less error would have been produced.
Improvement in this lab could easily be achieved by waiting for more time in between
temperature measurements so that we could actually know if the steady state temperature at the
particular power input was correct. Another thing we could have done with more time was to
continue to increase the power into the system at smaller increments so that more data points
could be obtained which would show trends better (you can only assume a linear relationship
when you have 2 data points but if you had 3-5 you could see parabolic or exponential relations).

Conclusion:

Based on our values for thermal conductivity, we saw that the effective thermal conductivity
remained relatively constant as temperature increased while the heat pipe was in the horizontal
position at an average of 12795 W/m*K. When the heat pipe was turned to the vertical position
thermal conductivity drastically changed between different values of input power from 9079 to
10049 W/m*K. Because we knew that the heat transfer capability of a heat pipe is not unlimited
we have to talk about the reasons why the heat pipe has limitations. For a wicked heat pipe in
which the return of liquid to the condenser section relies on capillary action of the wick, the most
important limitation is the capillary limit. The capillary limit is encountered when the wick is not
capable of producing enough pressure to pump the liquid from the condenser back to the
evaporator. When this heat transfer limitation is reached, the evaporator temperature will
increase sharply because there is not enough liquid in the wick structure to evaporate and absorb
this heat input. This observation would have been seen if the power was further increased into

11 | P a g e
the system pushing the temperature of the heat pipe past 150°C. When we flipped the heat pipe
to the vertical position, and we saw the high change in thermal conductivity, we can conclude
that the heat pipe was not being effective because its process of evaporation and condensation
occurring in the condenser and evaporator were being affected by the change in position and
gravity. The overall high rate of thermal conductivity of the heat pipe is 11180.2 W/m*K when
the horizontal and vertical setups are averaged. This in turn may be compared to the relative
thermal conductivity of an average copper rod, where the effective thermal conductivity is 401
W/m*K. Based on these numbers one can see how useful a heat pipe can be to transport heat
energy from one location to another, quickly and effectively.

12 | P a g e