This paper was presented at the Twenty-first Annual National Conference of the Society of Aeronautical Weight Engineers at Seattle, Washington, May 14-17, 1962. During the past few years, extensive testing has been performed with model tanks to determine the weight of ice and frost that might form on a liquid oxygen tank under varying climatic conditions. In extreme instances, weights as high as 1500 pounds in fair weather and 10,000 pounds in inclement weather were extrapolated from model test data.
In a recent test, a camera was placed in an ICBM nose cone to photograph the booster during separation. These pictures clearly showed that 40 percent of the liquid oxygen tank was still covered with ice and frost at separation. It had previously been assumed that any ice and frost formed on the LO2 tank during filling would be shaken off by vibration, or melted away during boost through the atmosphere.
Additional tests showed that model tank extrapolations were considerably higher than actual ice and frost formations observed on missiles. The test data were analyzed by standard statistical techniques, revealing the most important variables affecting ice and frost formation. These were found to be, in order of importance: hold period, ambient temperature, wind velocity, and relative humidity. The results obtained from this statistical analysis, when correlated with other test data, yield explanation for the ice and frost formation phenomena on a missile liquid oxygen tank.