### STP162: Tests on the Binomial Sampling Theory for Heterogeneous Coals

Visman, J.

*Senior Scientific Officer, Department of Mines and Technical Surveys, Calgary, Alberta, Canada*

Pages: 12 Published: Jan 1955

**Abstract**

Broken coal leaving the mine can be considered to consist of two components, “coal” and “rock,” found from a density-separation at roughly 1.60 specific gravity. The object of the following study is to investigate whether this two-component mixture—which is admittedly artificial—can be used for predicting the considerable variances in ash found in most raw coals, by applying the binomial formula for ash developed earlier (2). The ash variance for single pieces, it is noted, may vary from, say, 10 to about 3000, due to the sometimes strongly bimodal distribution, whereas the trend variance is only of the order of 10. The ash variance is also considerably larger than the variances of other constituents, for instance, moisture or sulfur. The sampling procedure for ash is therefore in a class apart, and it stands to reason that even a first approximation of this “binomial variance” would be of great practical value in finding the minimum weight and number of increments for a preassigned accuracy.

**Paper ID:** STP46269S

**Committee/Subcommittee:** D05.23

**DOI:** 10.1520/STP46269S

ASTM International is a member of CrossRef.