Model:Morrell Mi SMC SAG

Morrell SMC SAG & ball mill Model

This is a SAG or AG mill plus ball mill model that estimates the specific energy consumption (E_SAG) using the equations of Morrell (2004).

Testwork Required

• The SMC™ test, stored in the "DWT" testwork table (Drop Weight Test).
• Bond ball mill work index. The method requires all the elements of the ball mill work index test to be filled in:
  • Ball mill Wi test closing screen size, umclosing, µm
  • Ball mill Wi test product size, P₈₀, µm
  • Ball mill Wi test feed size, F₈₀, µm
  • Ball mill Wi grams per revolution at test conclusion, gpr

Formulae

The method is superficially similar to Bond (among other models), but with a variable exponent on size defined by: $f(x) = - 0.295 - \frac{x}{10^6}$

Where x is the particle 80% passing size in µm.

$E_{total} = 4 \times Mia \left( 750^{f(750)} - F_{80}^{f(F_{80})} \right) \times CFsag + 4 \times Mib \left( P_{80}^{f(P_{80})} - 750^{f(750)} \right)$^{Morrell, 2004}

$E_{SAG} = 4 \times Mia \left( P_{80}^{f(P_{80})} - F_{80}^{f(F_{80})} \right) \times CFsag$^{Morrell, 2004}

$E_{ball} = E_{total} - E_{SAG}$^{Morrell, 2004}

CF_sag calibration factor

The CF_sag factor is used to reflect the effect of pebble crushing and pre-crushing on the overall circuit performance. This value is automatically selected if you leave this field blank in the model configuration settings.

• Base value, SAB circuit (no pebble crushing, ~6 inch SAG feed), CF_sag = 1.00
• Pebble crushing, SABC circuit, CF_sag = 0.95

Transfer size

Morrell specifically discourages use of transfer size in calculations such as those in SAGMILLING.COM for a variety of reasons (see discussion in Morrell, 2011). Unfortunately, the alternative method proposed by Morrell is a complex equation that requires pre-selected mill sizes and a series of unpublished calibration factors. This alternative method is not practical for the calculations in SAGMILLING.COM, so a transfer size approach is used instead.

The modelled transfer sizes are probably "Bond-compatible" and not similar to what would be measured directly in a plant survey. Plant survey results would need to be subjected to a phantom cyclone calculation before they are comparable to the model predictions.