Detailed fragmentation data for different blast delay timing tests at the Red Dog mine in Alaska

are analyzed. This work is a continuation of previous work in which the crusher output was found

to be better described by the three-parameter Swebrec function than by the Rosin-Rammler

function. For a crusher model, the x50 variable sufficed to describe changing crusher input while

b and xmax could be considered fixed. For the crusher output from aggregated stockpile rock, as

measured by digital image analysis using Split-Online, the constancy of b was not as apparent. In

this paper detailed stockpile data, single Split data series, and sieved belt cuts of crusher product

are analyzed using the Swebrec and truncated Rosin-Rammler functions. The variability of b is

confirmed. By using specific fraction passing sizes a shortcut is presented that removes the need

to do curve fitting to determine the three Swebrec function parameters. Prediction equations are

presented for these parameters for the tests involving the effect of inter-hole delay time on crusher

output. The slope value at x50 was found to be independent of timing, unrelated to the estimated

milling properties, and well approximated by 0,5/(x75-x25).