Storage algorithms for reserving and freeing contiguous blocks of storage space were investigated. A simulation program was constructed and run for each method; several different distributions of the parameters that characterize the request for storage space were used. The performance of a noncontiguous storage algorithm was simulated for comparison with the other methods. The effect of disk flaws on the performance of several algorithms was also simulated. The simulation runs showed that the algorithm currently in use at LLL performs best for theoretical distributions of file size and life but that it is not best-suited for the distributions of size and life drawn from the samplings of the LLL machines. The results did not indicate a better-suited choice for the LLL data because samples from different LLL systems resulted in different rankings among the various algorithms. 9 tables.

A simple model for calculating the temperature-dependent deposition of silica scale from geothermal brine is outlined. A finite-difference technique for solving the moving boundary problem is described. The implicit formulation is solved directly, using matrix inverse methods appropriate for banded matrices. The multimaterial problem is defined in a way that maintains the compact matrix bandwidth of a single material. Sample problems are presented in dimensionless form, including a comparison between the finite-difference results and an analytical solution for a single-material slab problem. These initial studies show that the temperature at the brine-scale interface is not significantly perturbed by the growth of a thin scale layer, although pipe thickness and different experimental scale-deposition curves significantly affect the total amount of scale forming over a given period of time.

This report compares two models for converting beta backscatter count readings into thickness measurements. The necessary formulas to be used in an unweighted and weighted regression analysis are listed. The question of whether one should perform a regression analysis using the five available standard thicknesses or whether one should, in addition to these standard thicknesses, use zero as a standard thickness is decided. A weighted regression analysis is compared with an unweighted one for each model. The ''best'' model is selected, and the conclusions of the analysis are presented.

Effort continues to determine the parameters for maximizing the operating efficiency of the off-gas systems and the venturi scrubber. Results for the month show a decrease in particulate emissions from the system. Engineering designs are being finalized for enclosures to permit the introduction of radioactive waste into the incinerator and the handling of the resultant ash. Concurrent with these improvements, the cooling system capacity will be increased to handle more heat in the off-gas as a result of continuous operation. Additional cooling capacity will also be required to protect the burning chamber enclosure. In July, a HEPA filter was installed on the final exhaust to demonstrate that premature loading of the filter would not be a problem when incinerating radioactive waste.