20 THE SHORTER-TERM BIOLOGICAL HAZARDS OF A FALLOUT FIELD masses of insoluble particulates can he removed from sloped surfaces by water films. The tests were of an exploratory nature and only simulated roofing surfaces were used; the basic objective was simply to test the capability of a water film in moving the contaminant. A 4’ x 4’ panel was mounted on a tilting easel and set at an angle of 14° 3” rise in 12’7). 1 x 1” ribs fixed to the panel in the direction of slope divided the panel into 4 sections of equal area, carefully removed in 1-foot increments and weighed. The latter measurements were used as material balance checks against the measurements obtained from the wetted pancls and to determina uniformity of dust deposition, *The results of the tests appear on the fol- lowing table. Two methods of applying wash water were The first was by means of a header, or tion was by a garden spray nozzle fixed in position above the panel and adjusted to spray two adjacent test sections. Water delivery rates were a function of the characteristics of the delivery systems. They ranged from 0.3 to 1.0 gal/min/lineal foot. The total wash water used per panel in each test run was collected in a funnel placed under the lower edge and emptying into a jar. The contaminant was simulated by calcium carbonate dust with a particle size range of 44 «to 150 yn. This material was dusted onto the panel from a 4-foot long shaker held several feet above the surface. During dusting, the shaker was moved back and forth over the panel in the direction of slope to effect uniform deposition. The tests were conducted in the following The wash water was turned on prior to dusting. A measured charge of dust was shaken onto the 4’ x 4’ panel as uniformly as possible, both in surface distribution and time. The rate of deposition was about 0.3 to 0.5 gm/min/ft? and the total deposition was very close to 1 gm/ft? in each case. When the shaker charge was exhausted, the water was turned off. The collected wash water was filtered and the solid content weighed. The residual solids on the wetted test sections were Typeof washdown Avg.re- itofroof efficiency Rete gpm/ width . Smooth aluminum... _| Header)... -.-. Smooth aluminum treated with Aerosol O. T...._-...- Header... 3. Smooth aluminum 1.0 moval % ever, the spray used in the third test success- fully wetted the entire surface and the removal efficiency was again correspondingly ir proved. The corroded and painted surfaces were contribution system. The assumption was made that the spray would perform at least as well as the header on these surfaces and therefore 152.6 was not tested. From the standpoint of practical application, it is difficult to imagine that 180.4 &@ spray would be necessary to achieve the sol O. Th... 2... Spray... 4, Corroded aluminum__| Header-. 3 -6 197.8 199. 6 with flat white alkyd... .--- a. Header-_. 6. Simulated gravel Header... surface... ...--.-+ -6 1.9 197.2 48.6 surface.....-.---. Spray-._ 9 32. 6 5. Aluminum painted 2 Average of two values, The limitations of this series of tests are obvious. However, the simple objective of demonstrating the ability of water to transport sizeable masses of particulates was realized. The results are sufficiently encouraging to justify further investigation. Certain behavior characteristics exhibited by the washdown system during the tests were noted. On reasonably amooth surfaces, the contaminant was effectively removed wherever the water film was maintained. In test number 1, the film divided into individual rivulets about half way down the slope. The paths of the rivulets were relatively fixed and as a con- sequence, portions of the test sections were unwetted and uncleaned, as indicated by the removal efficiency. How- even a smooth metal roof would be devoid of surface irregularities and it would appear that, required uniformity of water distribution. treated with Aero- 7. Simulated gravel The pretreatment with the wetting agent in the second test was an effort to promote more uniform wetting and was partially successful efficiencies were realized with the header dis- Water Nr deliver a distributed water film to 2 of the 4 test sections. The second method of applica- manner. Type of surfaces lively poor efficiencies were obtained. ducive to a uniform water film; hence, good SUMMARY OF WASHDOWN TESTS the 4 sections. tested. The dust on the unwetted test panels was similarly removed and weighed, Test surfaces were mounted within distribution pipe, mounted across the upper end of the sloped panel and perforated so eas to THE APPLICATION OF AUTOMATIC WASHDOWN TO PITCHED ROOFS As a result, rela- As one might expect, removal from the coarse irregularity of a gravel surface is }ess effective. Since gravel surfaced roofs are normally flat or only gently sloped, performance may be 21 expected to be poorer than indicated in these tosts, A second question concerning mass transport is associated with a roof washdown system. This involves the capability of a water flow to move the contamination collected in a roof gutter. This aspect was tested qualitatively. Water was passed through a slightly inclined 4-inch diameter cylinder. CaCO, dust was discharged into the water stream in the cylinder from a vibrating feeder at an arbitrary rate of about 80 gm/min. This was done at water flow rates of 4 and 9 gal/min. As determined visually, all of the dust was transported along the cylinder and out from the end for as long as the feed was continued. Again the limitations of the test are obvious and need not be enumerated. On the other hand, it has been demonstrated that reason- ably heavy amounts of insoluble particulates can be flushed through a gutter. As in the surface washdown tests, the results of this experiment may be taken as justification for further experimentation.