DISCUSSION 



The experimental results and the analytical results had excellent 

 agreement. Even when the viewport failed at the low-pressure face corner, 

 there was no failure at theO-ring groove. The design recommendations 

 (Figures 18 and 19a) of this report predicted failure for this viewport and, 

 as shown in Figure A-3, failure occurred. 



It might also be interesting to note the flexural behavior of this 

 viewport as shown in Table A-1 . For the free boundary condition, no tensile 

 stresses appeared at the center of the low-pressure face as they did for the 

 fixed boundary case. 



Although minimum research time negated any investigation of 

 mesh-size influence on theO-ring groove, the results were in agreement with 

 the experimental results. Also, the analytical simulation of the groove was 

 a worst case solution. This means there was no rounding, of the corners in 

 the simulated model while in the experimental case the corners had a 1/8-inch 

 radius. 



FINDING 



A conventional 0-ring design with the groove located in the conical 

 face of the viewport (Figure A-2) satisfactorily fulfills the requirements for 

 viewport sealing in three ways: operationally, economically, and structurally. 



40 



