through which a number of gages in an area are laid or, in the 

 case of a single rosette, on a line through the hatch. 



VL CONCLUSIONS 



The strain gage data was linear. When reduced, the 

 principal stress in most areas was very close to the theoretical. 

 In some areas, particularly between inserts, the strains were 

 generally 10-15% higher than expected. The stress versus 

 gage number for the various areas are found in Figures 4-13, 

 plotted for a pressure of 3300 psi. 



The load was apparently not ideally transmitted across the 

 hatch, see Figure 8. This was probably due to the addition of 

 an -ring and slight misalignment of the hatch. 



After all the gage readings had been recorded it was found 

 that two of the electrical hull penetrators leaked. They were 

 replaced and the vessel repressurized to 3300 psi. The 

 penetrators did not leak during this test. Subsequently, all 

 penetrators have been modified to enable better seating. 



