Parabolic Trough Cooker Testing and Results

Energy Absorption

Although the constructed parabolic trough system intercepted one square meter of solar
radiation, there are complications associated with assuming the widely accepted sunlight
irradiance of 1000 W/m2 being perfectly converted to thermal energy at the black
collecting pipe. The tests for energy absorption were performed over a series of days with
differing sunlight intensity levels, all being less than the 1000 W/m2 as measured by a
pyranometer on the trough. New Mylar can be assumed to perfectly reflect the incoming
light to the focus of the trough; however, the parabolic shape is not absolutely perfect and
scatters a significant portion of the incoming light, in addition to the fact that the Mylar has
been slightly degraded due to mechanical and ultraviolet wear throughout its life. When
the sunlight finally reaches the pipe, the black paint coating can be assumed to convert the
light to heat, but also reflects a significant portion to the surroundings. Because of these
inefficiencies, it was necessary to test the overall solar energy to thermal energy
conversion properties of the system.

Description of Test and Results

The black collecting pipe was filled with water and capped at both ends. In order to make a
variety of measurements of the system, thermocouples were placed at four locations on the
steel pipe and three locations within the water, leaving the last of eight channels on the
computer for the pyranometer described before. The steel pipe thermocouples were
placed mid-pipe and the water thermocouples were placed at the quarter length, mid-span,
and three-quarter length inside the pipe. The tests were run on mostly sunny days, where
the likeliness of a cloud disturbing the measurements was insignificant. In order to
determine the temperatures of the steel and water for the energy equations, the four
measurements for the steel and three measurements for the water were averaged using an
arithmetic mean. Care was taken to isolate the steel pipe thermocouples from the wind,
which was noticed to greatly affect temperature readings. The water thermocouples were
also checked to verify that they were not touching the steel pipe before each test, as that
would lead to overestimation of the water temperature.

The following results average five of nine tests that had enough sunlight to heat the water
in the system to boiling and no equipment or experimental failures were observed. The
excluded tests had variety of setbacks including clouds blocking the sun, thermocouples
becoming unattached, and data logger problems that prevented proper recording of data.

Measurements and values used

The time versus temperature curves for heating water from 30 °C to 100 °C along with the
surrounding steel heat absorber pipe are plotted in Figure 17.
Since the pasteurizing device will be operating in the full range of temperatures tested, a
first-order slope approximation was fitted to both the water and steel curves in order to
determine the temperature change over time.

Summary of test results

These calculations yield a total realized heating power for our system of 316.2 watts, a 34%
efficient conversion of sunlight to thermal energy. With this conservative estimate of
power, the estimated water pasteurization ability of the device was 4.19 liters/hour. Since
these tests were run during winter days with low ambient temperatures, this value likely
represents the lower estimate of the power of the device.

Figure 1: Temperature curve for trough absorber

Valve

Throughout the design process of the parabolic trough system, different options for a
thermostatic valve to control the temperature and flow of water from the terminal end of
the device were investigated. The initial idea of constructing a novel design using the
properties of bimetallic materials as a thermometer was largely abandoned after it was
realized that widely available automobile thermostats would provide a similar function. A
manufacturer of thermostatic valves was also contacted for a quote, $112, which turned
out to be prohibitively expensive for the application of this project.

A quick water tightness check of the thermostat purchased at the local automotive parts
store showed that the device leaked around a metal on metal contact point within the
device. Since the thermostatic valve needs to be absolutely watertight to avoid allowing
unpasteurized water into the pasteurized water reservoir, a modification was needed to
prevent this leak. The application of the silicone used to seal the box cookers formed a
quasi gasket that prevented any water from leaking around the internal parts of the
thermostat.

The thermostat purchased is rated at 91°C, which was later verified to be the temperature
the device begins to open at. In the system, the valve remains shut during periods when
the water has not reached a high enough temperature to be considered safe. However,
once the water reaches 91°C, well above the pasteurization temperature, the valve begins
to allow the flow of water into the clean reservoir. If the system falls below this
temperature, the valve closes again, and the water will not pass until the system heats back
up. With the thermostatic valve and the geometry of the trough, this device can be
completely passive, and purify water throughout the day without human intervention.

Figure 2: Schematic of Modified Thermostat. The blue blocks represent the silicone "gasket" addition to the
thermostat.

Water Flow and Temperature

In order to verify the functionality of the parabolic trough as a pasteurizer, it was necessary
to determine the temperature of water released from the system and the steady state flow
rate that could be realized by the system. It is important that the temperature within the
absorber is maintained at a high enough temperature to destroy pathogens contained in
the water flowing through the absorber pipe.

Using the calculations discussed in the section above, the conservative estimation of the
steady state water flow rate is about 4.2 liters per hour. Figure 19 displays temperature
data recorded and plotted as temperature versus time. Initially the absorber was not
purged with water resulting in a high temperature recorded for the air in the absorber.
Since the thermal conductivity of air is significantly lower than water, it took
approximately 30 minutes for the valve to open even though the valve is rated to open at
91 °C; the air temperature near the valve was near 110 °C.

Once the valve opened, the back pressure of the system needed to be adjusted in order to
reach steady state. If the pressure on the outlet side of the valve is not adjusted, the
temperature of the water near the valve rises to approximately 102 °C, the valve opens and
the water is rapidly expelled from the system, and then the valve closes. The system
operating in this fashion does not reach steady state. In order to reach a continuous flow,
the pressure on the outlet side of the valve must be adjusted to reach steady state and a
continuous flow rate. When this was done, a sustainable flow was achieved around 52
minutes into the experiment and was maintained until the end of the experiment at 75
minutes. Constant flow would have been maintained as long as solar irradiance was
maintained at a similar level. The flow rate was calculated to be 465mL over a five-minute
time span, which is equivalent to 5.58 L per hour. In addition, water was calculated to be
at sufficient pasteurization temperature for a period of approximately 81 seconds before
exiting the system.

Figure 3: Plot of temperature versus time for the water pasteurization trough
system. The red data points indicate internal temperature near valve and blue data

points correspond to temperature readings 25 cm from the valve.

Power Cycle

We started our exploration of creating a thermal to mechanical energy system by analyzing
the model Rankine cycle. The only design parameter that we could determine was the
energy input due to the parabolic trough. We needed to find a way to convert the thermal
energy to mechanical energy. We began with a few centrifugal pumps that we thought
might spin if we applied a pressurized vapor through them, thus acting as a turbine. The
pumps that we found were too large for any practical design for our system scale. For
example, we analyzed a submersible well pump, but found that it required too great of a
mass flow for use to consider it as a viable option. It was also determined that it is difficult
to find a small scale turbine to used on a small system, and that a turbine may not be the
best conversion device for a system of our scale. This led us to consider a standard
reciprocating engine or a steam engine as the thermal to mechanical conversion device for
a small system. We are currently in need of a reciprocating engine to test this type of
design. Perhaps a two stroke weed-whacker engine would suffice, which we will determine
in the next few months.