intact coral, because the piston could possibly restrict or prevent 

 fluke keying. There appeared to be no discernible degradation in perfor- 

 mance. In fact, it is possible that leaving the piston in place enhanced 

 performance if the coral was intact. Pistons were not recovered during 

 the last five installations at the temporary site (T4 through T8) . 

 These pistons were well-used, and the piston was not affecting holding 

 capacity; therefore, to save time they were left in place. Piston loads 

 varied from a low of about 30 kips (135 kN) to over 100 kips (445 kN) 

 for pullout. The piston has at most one -fifth the resisting area of the 

 anchor flukes, and it is loaded only at one end. Compared to the central- 

 ly loaded flukes, therefore, there is about a factor of 10 between what 

 the piston and fluke should hold. Piston retrieval provided a good 

 basis for judging the ultimate capacity of each anchor fluke. Based 

 upon the piston loads, ultimate holding capacity would vary between 

 300,000 (1.35 MN) and 1,000,000 (4.45 MN) pounds. The weak link in the 

 system is therefore the anchor cable, which was rated at 230,000 pounds 

 (1.02 MN) minimum breaking strength when new. 



The piston load was not solely relied upon to judge anchor perfor- 

 mance. Each anchor was proof -loaded to 100 to 120 kips (445 to 535 kN) 

 in direct uplift. One fluke was loaded to 150 kips (670 kN) and another 

 was loaded briefly to 165 kips (735 kN) ; this latter load was caused by 

 wave surge on the barge. 



Three unsuccessful firing attempts occurred because of malfunction 

 of the S/A device, and one installation was delayed by the discovery and 

 correction of a broken circuit in an electrical cable immediately before 

 deployment of the anchor system. The S/A malfunctions were readily 

 diagnosed and remedied by the onsite CEL project engineers. Although 

 the same problems should not recur, such problems are typical of proto- 

 types such as the CEL 100K Propellant Anchor System in its present 

 configuration and should be expected, at a diminishing frequency, in 

 possible future similar operations. 



FINDINGS 

 Holding Capacity 



Although it is not definitely known at this point, the long-term 

 capacity of the anchors can be inferred from the short-term proof-loading 

 and piston pullout loads. 



Based upon the minimum piston pullout load of 30,000 to 35,000 

 pounds (135 to 155 kN) , the respective areas of the piston and the 

 coral/sand fluke, and the respective mechanical advantages associated 

 with pulling on the end of the piston and on the center of the fluke, it 

 is estimated that the short-term capacity is at least 300,000 pounds 

 (1.35 MN) ; hence, the strength of the main anchor load line (230,000 

 pounds, or 1.02 MN) is a lower limit to short-term capacity. Each fluke 

 was proof-loaded to over 100,000 pounds (445 kN) . This resulted in 



20 



