(Figure 3-12) manufactured by Phillips Trawl Products Ltd. of 

 England and a molded polystyrene sphere (Figure 3-13) manu- 

 factured by Plasticos De Galicia, S. A. of Vigo, Spain. Both of 

 these spheres are manufactured primarily for use on trawl fish 

 nets. The former, designated the FE-18, is rated for 1000 meters; 

 the latter, called the Marola 200, is rated for 1300 meters. Each 

 cost about $2.25 for a 6 lb. buoyancy sphere, thus promising to 

 reduce the total floatation cost to less than $3500. The order-of- 

 magnitude price saving over glass, and the ready availability of 

 large quantities were very attractive. About 100 floats of each type 

 were immediately procured for evaluation tests. It was quickly 

 found that the FE-18 float did not have sufficient depth capability 

 and the Marola 200 exhibited excessive cold flow resulting in short 

 life at depth. Refer to the next chapter for a summary of test 

 results. 



Each manufacturer promjptly improved his float to meet our 

 specifications. Phillips increased the wall thickness to provide 

 the "XX" float which can withstand the pressure at 1250 meters 

 and Plasticos De Galicia molded a "Special" Marola 200 using a 

 "high grade" non-impact- resistant polystyrene for much reduced 

 cold flow. It should be noted that the Marola float is encased in a 

 waffled outer polyethylene jacket to cushion the inner polystyrene 

 sphere against impact loads. The space between the inner sphere 

 and outer jacket is free -flooding. 



The principal parameters affecting float life are pressure dif- 

 ference from maximum rated pressure and water temperature. A 

 large number of test points have been correlated with available 

 creep failure laws. At this time it does not appear that one should 

 hope for an average float life much in excess of one year at 1000 

 meters. For this reason excess buoyancy was used, and other materials 

 and float designs must be evaluated for the future long-lived tele- 

 scope. Thermal-setting plastics appear worthy of further inves- 

 tigation in this regard. 



3.2.0 Float Attachment Device 



Development of a suitable float attachment device was a vexing 

 problem. This device had to meet the following requirements: 



209 



