Weigh the entire extraction boundary
Use one identified kernel lot and one defined run interval. Weigh prepared kernel, every water addition, collected milk, residual meal and separately collected losses. Keep cleaning water outside the product balance. Sample the milk and meal at the same stated boundary.
Compare composition as well as output
Report collected milk mass per kernel mass separately from milk fat and total solids. Added water changes the first ratio without creating coconut fat. A fat-recovery estimate needs measured fat in the kernel feed and the collected product on a consistent mass basis; no fixed recovery is assumed.
Inspect what remains in the meal
Record meal condition, any re-pressing pass and the sampling method. A second pass changes both additions and product composition, so retain separate first-pass and combined results. Trial acceptance should include the milk specification and the destination of the meal.
