Figure 2-19 shows a detail of the geometric model of the 40.0 inch diameter
reinforcement configuration which is composed of Parts 3, 4, and 5 from Model
No, 2. Again notice the close geometric modeling.
The loadings and boundary conditions applied to Model No. 2 are shown in
Figure 2-15. The two 11,000 lb/in. ring loads shown acting at the junction
of Parts 8-9 and 10-11 represent the simulated bolt or clamp loads. The
internal pressure was set at 1000.0 psig. The boundary condition at the
edge of the shell represents the symmetry condition found along any diametral
line in a uniform sphere under uniform internal pressure. Some of the maxi-
mum stresses and the location and direction in which they act are shown in
Figures 2-19 and 2-20.
(c) Model No. 3. This model, as shown in figure 2-16, is
composed of four parts as follows:
Type of Shell
Variable (to model the taper
transition and part of the
The loadings and boundary conditions applied to Model No. 3 are shown in
Figure 2-16. The basic loading was 1000 psig internal pressure. The addi-
tional 446 psig pressure shown acting upon Part 4 represents the bearing
pressure of the inserted window upon the reinforcement flange. This figure
is attained by assuming the total bearing load caused by the 1000 psig acting
on the area of the window is borne by a 4.0 inch ring surface of the
reinforcement plate. The only restraint applied to the inner edge of the
10.0 inch diameter reinforcement ring was one fully constraining rotation.
It was assumed that the inserted window configuration would be fairly thick
and act as a rotation constraint. The shell edge was considered in Model No.
2. Some of the maximum stresses are shown in Figure 2-21. In addition the
radial displacement of the inner edge of the reinforcement ring and the
moment that will be applied to the flange of the inserted window is shown.
This 4.8 mil displacement and the 15,000 moment load should be used
in the design of the window.
(4) Categorization of Stress. At this point in the "design by
analysis" procedure, the stresses as calculated above should be broken down
into the stress categories as earlier defined. All stresses developed in
this analysis were principal stresses because the models were composed of
shells of revolution. The computer program used breaks the stresses down
into total stresses acting on the surfaces and the membrane stress in each
direction all along the shell. The submission of the computer printout, in
this case, with some particular points of interest further defined, should be
sufficient documentation. The computer printout shall not be included in
this chapter. However, as an example at the junction of Parts 4 and 5 in
Model No. 2 (see Figure 2-15) the computer has printed out,