The instrument is calibrated against 150 mg of EDTA (usually - the actual weight is flexible, as long as it is weighed accurately). Samples of grain are ground in a U-D Cyclone grinder, fitted with a 1.0 mm screen, and are analyzed as received. Sample weight for analysis is 250 mg. Again, the weight is flexible, provided that it is accurately recorded, but the standard error per test increases with sample weights of less than 150 mg, while sample weights in excess of 300 mg are difficult to handle. The sample is weighed into a piece of tin foil, and is then wrapped tightly in the foil, to exclude air. Weighed, wrapped samples are then loaded onto the carousel, and analyzed in sequence. Up to 34 samples can be loaded onto the carousel. Results are expressed on a constant moisture basis, either 13.5% or moisture-free. Moisture is determined on all samples by the AACC Approved Method No. 44-15A. We are developing calibrations for all frequently-analyzed materials on the NIRSystems Models 5000 and 6500 for moisture.
This is the method. The procedure for employing the method here includes:
Maintenance is carried out every day according to the instrument software. Further maintenance is carried out periodically according to the Manufacturers' manual.
The coefficient of variability (our measure of analytical precision) is normally about 0.6% of the mean, as compared to about 1.2 % for the Kjeldahl method.