Information Technology Reference
In-Depth Information
each copper wall for continuous monitoring of their temperature. In this way, it was
possible to maintain the temperature of the sidewalls uniform within ±0.1 C of the
desired temperature. In order to measure the temperature distribution inside the test
section eight 1.0 mm thermocouples were inserted in the back wall through eight
holes. They were arranged in two horizontal planes perpendicular to the front and
back walls. In this way, two ranks of four evenly spaced thermocouples in the
y direction measured the temperatures at two water depths in the z direction. This
arrangement minimized heat conduction along the thermocouples and therefore
reduced the measurement error. All thermocouples are type k (NiCr-NiAl) sheathed
with stainless steel and were calibrated with an accuracy of ±0.1 C. The ther-
mocouples outputs were recorded by a data acquisition system at preselected time
intervals between two consecutive measurements. These reservoirs, which are also
called constant temperature baths, were connected to the coolant flow system by
rubber tubes, which were insulated by foam pipes. Alcohol was used as the coolant
fluid in the constant temperature baths.
The experimental case were performed with the cold wall maintained at 0 C
and the hot wall at T
H
= 12 C. This is an important interval to explore the
maximum density phenomenon. The water was carefully siphoned into the test
section to avoid introduction of air. To obtain the initial configuration of the solid
phase of the fluid, it was necessary to freeze the substance with the test section
turned 90 to the left before the beginning of the experiment (Fig.
12.4
). In this
way, the two constant temperature baths were set at low temperatures. After
conclusion of the freezing process, the solid was heated up to a temperature close
to its fusion point. In this way, the control point temperature was gradually
increased and the temperature in the solid was carefully monitored. This heating
regime was continued until all the system reached 0 C. At this moment, the rest of
the test section was filled with water at 0 C. The desired water temperature (0 C)
was obtained by mixing cold water with crushed ice and allowing sufficient time
for the equilibrium. The water was carefully siphoned into the test section to avoid
introduction of air. The test section was turned 90 back to its original position, as
shown by Fig.
12.4
. Next, the valve attached to the hot wall was closed and the
temperature of the bath connected to the valve was increased. When the temper-
ature bath attained the desired temperature, the valve was reopened and the left
wall was heated to T
H
within 1 and 4 min depending on the T
H
value. Once the
initial conditions were attained, the melting run was started. Thermocouple data
were collected at 1-min intervals. The temperature of the sidewalls was maintained
constant throughout the data run with the help of the electrical resistances. Since
the solid phase occupies a greater volume than the liquid phase, a feed line
connected to the top wall was used to fill the test section with water at 0 C.
To visualize the flow patterns, the water was seeded with a small amount of
pliolite particles. Pliolite is a solid white resin with a specific gravity of 1.05 g/cm
3
that is insoluble in water. These tracer particles of small diameter (\53 lm) are
neutrally buoyant. A beam from a 5.0 mW helium-neon laser was used as light
source. The laser beam passed through a cylindrical glass rod to produce a sheet of
laser light before passing through the test section wall. Photographs of the flow
Search WWH ::
Custom Search