Title: TESTING AND EVALUATION OF THE MODIFIED DESIGN OF THE 25-DISK ROTARY MICROFILTER

This report details redesign of a commercially available rotary microfilter to meet the operational and maintenance requirements for radioactive service. Personnel developed the design and coordinated procurement of two filters followed by testing of one unit. System testing examined the ability to rinse soluble material from the system, filtration performance using several insoluble solids loadings, effectiveness in washing sludge, amount of wear to parts and maintenance of the system including the insertion and removal of the filter stack, and the ability to flush solids from the system. The test program examined flushing the filter for soluble material by filling the system with a Rhodamine WT dye solution. Results showed that draining the system and rinsing with 50 gallons of water resulted in grater than 100X reduction of the dye concentration. Personnel determined filter performance using various amounts of insoluble sludge solids ranging from 0.06 to 15 weight percent (wt%) insoluble solids in a 3 molar (M) sodium simulated supernate. Through approximately 120 hours of start-and-stop (i.e., day shift) operation and various insoluble solids loadings, the filter produced filtration rates between 3 and 7 gallons per minute (gpm) (0.12-0.29 gpm/ft{sup 2}) for a 25-disk filter. Personnel washed approximately 80 gallons of simulated sludge using 207 gallons of inhibited water. Washing occurred at constant volume with wash water fed to a well mixed tank at the same rate as filtrate removal. Performance measurement involved collecting and analyzing samples throughout the washing for density and sodium content. Results showed an effective washing, mimicking a predicted dilution calculation for a well mixed tank and reducing the sodium concentration from 3.2 M to less than 0.3 M. Filtration rates during the washing process ranged between 3 and 4.3 gpm for one filter unit. The filter system then concentrated the washed 15 wt% insoluble solids slurry to approximately 20 wt% insoluble solids with no operational problems with the exception of the entrainment of air due to leaking packing in the feed pump. Prior to the air entrainment, the filtration rate was approximately 4.2 gpm for one filter assembly with the process fluid temperature adjusted to 35 C. Personnel measured the turbidity of filtrate samples from all phases of testing. All samples measured were less than 3 NTU, with the majority of samples less than 1 NTU. Thus, all measurements fell below the process acceptance criterion of less than 5 NTU. After slurry operations, personnel rinsed the filter with the equivalent of 250 gallons of water by re-circulating 50 gallons of water. The residual sludge solids remaining on the filter stack weighed approximately 685 grams. This amount of solids corresponds to an equivalent activity of 15.1 curies (Ci) beta and 0.38 Ci gamma radiation dose for Sludge Batch 4. Workers completely disassembled the filter system and examined it for signs of wear and component operation. An evaluation by a John Crane Inc. representative concluded that the wear observed on the mechanical seal resulted primarily from the numerous stops and starts, the abrasive nature of the process fluid and the possibility that the seal faces did not receive enough lubrication from the process fluid. No measurable slurry bypassed the mechanical seal. While it is extremely difficult to predict the life of the seal, the vendor representative indicates a minimum of one year in present service is reasonable. Changing the seal face material from silicon-carbide to a graphite-impregnated silicon-carbide is expected to double the life of the seal. Replacement with an air seal might be expected to increase lifetime to five years. The bottom bushing showed wear due to a misalignment during the manufacture of the filter tank. Minor adjustments to the alignment with shims and replacement of the graphite bushing with a superior material will greatly reduce this wear pattern.