Two types of propellant were used, one a slower burning, single- 

 base type (M6) and the other a faster burning, double-base type (M26). 

 The anchor gun system had been designed to be used with the M26; however, 

 computer simulation indicated that satisfactory performance could be 

 achieved with the M6, and its low cost and ready availability made the 

 M6 a very attractive choice. The exact burn rate of the new propellant 

 had not been determined in closed bomb tests. To be safe, reduced 

 charges, which still provided ample performance, were used initially 

 while peak gun barrel pressures were measured with copper crush gages 

 mounted in each cartridge assembly. 



Installation P1 (permanent site, anchor no. 1) was fired with 15 

 pounds (6.8 kg) of M6, producing a peak gun barrel pressure of 34 ksi 

 (234 MPa) , somewhat higher than expected. The designed maximum allowable 

 operating pressure for the gun is 35 ksi: (241 MPa) therefore, the use of 

 reduced charges was justified until M6 performance could be accurately 

 defined. Gun barrel pressures derived from M26 firings also were 

 higher than expected. The reason for these differences was attributed 

 to: 



1 . Elevated burning rates (higher than typical for similar propellant) 

 that may have existed in the untested batches of propellant utilized. 



2. A lower- than- calculated chamber volume caused by the large negative 

 tolerances on the inside diameter of the hollow piston that were accepted 

 in order to reduce piston cost. 



3. The response of the crush gages to peaks, including the effect of 

 standing wave peaks caused by the nonhomogeneous distribution of the 

 charge within the gun barrel, as opposed to the computer prediction's 

 accounting only for the average pressure at any instant throughout the 

 entire chamber volume. 



Once the system was calibrated with a few firings, slight modifications 

 could be made to vary performance with reasonable confidence. This 

 confidence resulted from measured performance repeatibility. Charge 

 weights were then varied to yield measured peaks of about 30 ksi (207 

 MPa). This pressure provided maximum gun life as well as ample energy 

 for installing the flukes so as to achieve holding capacities well in 

 excess of requirements. 



Fluke penetration was measured in each case when the anchor load 

 line was taut by subtracting the measured water depth and the measured 

 length of line above water from the entire known line length. This 

 measured line penetration, plus the distance from the connection on the 

 fluke to the fluke tip, yielded fluke penetration to the tip. 



The required load to retrieve each piston was recorded. Over the 

 course of the operation, six of the pistons were not recovered. The 

 first expended piston was left in place during installation P2 primarily 

 to determine whether the piston when still in contact with the fluke 

 would affect its performance. This would be particularly important if 

 the seafloor contained large pockets of coral sand /aggregate rather than 



19 



