Page 416 - Practical Design Ships and Floating Structures

39 1

2.3 Temperature Dktriburion within the Flame
Figure 2 shows the measured temperature distribution within the combustion flame at e2sec and
t=24sec for the case Z=2Omm. Although plate heating face temperature increases significantly, the
measured temperature distribution within the combustion flame is almost the same through the
measurement. That is, the thermal-flow field becomes stable in extremely short time.
Figure 3 shows the averaged temperature distributions for the cases where Z=12,20 and 28mm. The
followings can be seen; 1) thin temperature boundary layers of about 1-2mm thickness are formed
close the steel plate, 2) the temperature falls down rapidly upward from the plate surface and 3) the
temperature around the nozzle rises when Z is small.
Figure 4 shows the averaged temperature distributions on the plane 1.5mm upward from the plate
surface for the case where Z=12mm. This figure shows that lower temperature is observed on the
center region of the plane, this is because the core of the gas flame, in which combustion does not
occurs, touches the plate.
2.4 Discussion

The experimental results mentioned above indicate that thermal-flow field within the combustion
flame remains almost unchanged. The local surface heat transfer between heating gas and steel plate is
dominated by the turbulent energy of the flow field. These lead us to a hypothesis that the distribution
of gas temperature near the plate surface and local heat transfer coefficient are time independent and
they depend only on the distance from the torch.
The equation that expresses the relation between gas temperature right on the plate TG, plate heating
surface temperature Ts, heat flux from gas to plate q and local heat transfer coefficient a is given as
q = -aT, +UT,. (1)
According to the proposed hypothesis, Eq. (1) leads to a linear relationship between q and Ts because
we assume that a and TG remain unchanged in time.

3 INVERSE HEAT CONDUCTION ANALYSIS OF SPOT HEATING EXPERIMENT

The discussion mentioned in the previous section is based on the result obtained from the experiment
using a small torch whose heating power is much lower than that of the line heating torches used in
shipyards. The proposed hypothesis should be verified in the experiment using a high power line
heating torch. In this section, we shall show that a linear relationship between q and Ts, which is
anticipated from the hypothesis, exists in the results of spot heating experiment using a high power line
heating torch used in shipyards.

3.1 hkperimental Apparatus
A circular mild steel plate of diameter 300mm and 6mm thickness is arranged horizontally and a torch
with a #2000 nozzle is positioned above the plate. The center of the plate is heated by a flame of
oxyacetylene. Table 2 shows the pressure and the flux of acetylene and oxygen. The heating power of
this gas flame is higher by far than that of the methane torch used in the L.X.F. experiment. The
distance between the nozzle and the plate is arranged to be 18.5mm. The time histories of the plate
back face temperature are measured by using thermoelectric couples welded on the plate back surface.
The outputs of thermocouples are recorded on a personal computer by every 0.5sec. The points, at
which temperature is measured, are arranged in the radial direction from the center to the point 104mm
apart from the center. Hereafter, r denotes the distance from the center. The intervals of the points are
4mm up to the point at which AOmm, and 8mm for the points at which r>4(hnm. In the experiment,
heating is ceased within about 10 seconds because the heating surface at the center melts about 0.5 mm
deep at that time.