ay 252 and Jranse stations TE and Liss sere survicicathy active to sield sx data. Those froa Teak Stations 35 ami 56, with the excepticn of sclrur trsa Station 52, bad cecayed to near Sacaground levels. Thersai seutroa Jaca vere obtained froa gold and cobalt saaples which were flwa to saaiia for evaluation. Telemetered records showed that 31] neutron detectors Fusctiooed aechanically fi.e., the sample wheel was turning) catil impact. There was an elec- trical salfuaction on Station 20 during the entire tine of ‘fall; also, Statioa GO data were Tost because its transmitter went off the air at % seconds after launch. Station €52 operated properly until 175 seconds after burst at which tine there was an electrical malfuncticn. The background was still sufficiently high to sask foil activity duriag the tine Station 252 was functioning. other detectors functioned until impact. All Background count aaskea sulfur data on the Teak stations aod all data on the three Orange stations. Teleaecered data were obtained from fission foils at Teak Stations 38, 59, aol 209. For reasons given in Section 3.2.4, the uncertainty in values given in telemtered data is act least a factor of 3. 3.2.4 Conclusions and Recoomcodations Background caused oore difficulty than was anticipared in the evaiusliuu of telemetered data. There is ewidence that an appreciable portion of this interference activity cam from aaterials outside the detector itself. When instruments at Godiva LI were calibrated, access to che exposed detectors was not allowed until approximately 2O minutes after burst because of radiation hazard in the vicinity of che machine. cones ispacted i2ss than 20 minutes after burst, late the calibration curves backeard in time. Since all instrument nose it was necessary to extrapo- As a result of different induced background activities at calibration aod in the actual event, the slopes of the observed decay curves differed in the two cases. The flux as read from the data would, therefore, depend upon the tire at sbich it was read out. This caused the uncertainty cited in Section 3.2.3 The mucbers given io Table 3.1 represent the average of telemecered data read out at 200 seconds and 1000 seconds after burst. Because of the high probability of recovery or instrmece pacaages firua the ocean, laboratory counting of recovered activation samples is probabl; the sicplest acd certainly the most accurate technique for ceasuzsemeat of neutron ay 102

Select target paragraph3