Today’s thread is about sampling and dilution (key risk #3 of my pinned tweet). I have real time mining pictures below.
Punchline is tolerances are tight and dilution has been ~10%.
But data coming from just 1 or 2 benches ain’t worth schist.
1. geology is not so kind to have straight 1m lines, so sample grades at contact point will be diluted.
2. RC and DH have good corr., but BH systematically underreport high grade (we have data to support this — prob. because of fines)
Time to take some channel samples to take a real estimate of grade.
We’ll report these when our lab is operational.
We have a belt sampler at the crusher to have final grade reconciliation.
That looks like sub 0.5m tolerances to me.
@NeilRingdahl has been a critic of ours (in the good sense of the word). Wonder what he thinks.