"Scrubber stack losses are now calculated by weighing and analyzing the liquor from the scrubber each time it is drained.... This method of calculation can be in error by a considerable amount because of changes in operating conditions. The figures obtained are inherently deceptive, because a scrubber operating at a very low efficiency for a month would collect little uranium, thus showing a low loss. It were then repaired and operated carefully for the next month. the calculations would show a high loss, provided the process conditions did not change."

• "Investigation of Methods of Measuring and
  Reporting Uranium Losses to the Atmosphere."
  --Memo to A.F. Pennak from E.W. Randle. National Lead Co. of Ohio, 3/12/71

"The 1955 discharge of 15.4 kg U...for Plant 1 is the total measured from the start of sampling in September through December. No production records are available which would provide a basis for extrapolating the four-month measured loss to an estimate for the entire year. Therefore, the discharge measured in September-December is assumed to be the discharge for the entire year."

• MW Boback: History of FMPC Radionuclide Discharges, Nov. 1985

Between March 15, 1978 and June 14, 1978 "a significant dust loss occurred in the Plant 9 dust collector serving the NPR furnace and the crucible burnout area," caused by "extensive damage to the blow ring assembly and two bags pulled loose from their upper mountings....The standard operating procedure for the Plant 9 dust collectors specifically requires daily inspection of the collector for broken bags...A report is to be prepared for each inspection. These inspections have not been made for at least one year. No one in Plant 5 supervision knew they had been discontinued. ...If daily inspection of the collector had been carried out as called for in the standard operating procedure, the loss would have been quite small, perhaps as low as 3 kilograms."

• W.J. Adams, "Incident Report Covering the Dust Loss in the Plant 9 G-901-1034 dust collector," memorandum to S.F. Audia, July 14, 1978

"There is no procedure for inspection of stack probes; some have not been checked since initial installation - up to 30 years ago."

**

"A leak test of the sampler of G4-15 indicated that the system continued to have a flow of about 1/2 the indicated rate when the intake nozzle was blocked. FMPC personnel stated that they had been having many problems with system leakage."

**

"On the South Buffalo stack the filter was wet. The filter was not secured by a filter-holder gasket, allowing a significant leakage path around the collecting media."

• Oak Ridge Associated Universities (ORAU), "Environmental Program Review," Final Report, October 1985.