Reference ID: MET-9202 | Process Engineering Reference Sheets Calculation Guide
Introduction & Context
Multi-stage size reduction is the backbone of solids processing in food, pharmaceutical, and mineral plants.
By splitting the total reduction ratio across several mechanical stages—each operating inside an empirically-proven “safe” window—engineers avoid excessive heat, screen blinding, and off-spec particles.
The worksheet below formalises the algorithm you would embed in a plant-design spreadsheet or Python sizing tool: it converts a user-supplied feed diameter \(D_f\), product diameter \(D_p\), and Bond Work Index \(W_i\) into the minimum number of identical stages and the specific energy demand while forcing every intermediate size to stay inside Bond’s law boundaries.
Methodology & Formulas
Global reduction demand
\[
R_{\text{total}} = \frac{D_f}{D_p} \quad \text{with } D_f,\, D_p \text{ in µm}
\]
Number of identical stages
Fix an economically-optimum stage ratio \(R_{\text{stage}}\).
Enforce \(R_{\text{stage}}\in[2,10]\) (see validity table).
Obtain the exact (non-integer) stage count
\[
n_{\text{exact}} = \frac{\ln R_{\text{total}}}{\ln R_{\text{stage}}}
\]
and round upward to the next integer:
\[
n_{\text{stages}} = \lceil n_{\text{exact}} \rceil
\]
Recompute the real ratio per stage so the product stays exactly 1:
\[
R_{\text{actual}} = R_{\text{total}}^{1/n_{\text{stages}}}
\]
Intermediate diameters
For \(i = 1\ldots n_{\text{stages}}\)
\[
D_i = \frac{D_{i-1}}{R_{\text{actual}}}, \quad \text{with } D_0 = D_f
\]
Every \(D_i\) must sit inside Bond’s validity window (table).
Specific energy (Bond, 1952)
For each stage \(i\)
\[
E_i = 10\, W_i \left( \frac{1}{\sqrt{D_i}} - \frac{1}{\sqrt{D_{i-1}}} \right)
\quad \text{kWh ton}^{-1}
\]
Total specific energy
\[
E_{\text{total}} = \sum_{i=1}^{n_{\text{stages}}} E_i
\]
Empirical regime limits mirrored from code
Parameter
Lower bound
Upper bound
Units / remark
Stage reduction ratio
2
10
dimensionless
Particle size
50
10000
µm (Bond’s law validity)
The algorithm raises a fatal error if any calculated intermediate size or ratio transgresses the tabulated limits, ensuring the design stays inside well-documented industrial experience.
Splitting the duty becomes economical once the required reduction ratio exceeds roughly 8–10:1. A single device forced to cover the entire ratio usually suffers from:
Low energy efficiency because fine particles are over-ground while coarse ones are still under-worked.
Excessive wear and heat generation, raising maintenance and cooling costs.
Limited ability to optimize media size, speed, or classifier settings for each part of the size spectrum.
A two- or three-stage line lets each mill target a narrower range, cutting specific energy consumption 10–30% and improving product size accuracy.
Base the selection on feed F80, target P80, abrasion index, clay content, moisture, and downstream liberation requirements:
Primary stage: use robust crushers that tolerate tramp metal (jaw or gyratory) to reduce run-of-mine ore to ~100–150 mm.
Secondary stage: choose cone or HPGR to bring the stream to 15–25 mm; HPGR suits harder, more abrasive feeds and improves downstream grinding efficiency by micro-cracking.
Tertiary or regrind stage: use ball, Vertimill, or VRM to reach sub-millimeter or micron sizes, balancing energy, steel wear, and classification sharpness.
Simulate each configuration with software such as Bruno or JKSimMet, then compare CAPEX, OPEX, and power draw to rank the flowsheets.
Implement an interlocked cascade control strategy:
Bin level to feeder VFD: maintains choke-feed conditions on the primary crusher.
Power or hydraulic pressure on cone/HPGR to close the CSS or roller speed, preventing overload.
Mill load to water ratio and hydrocyclone pressure: keeps slurry density and circulating load inside design limits.
Particle-size analyzer feedback: trims the tertiary-mill tonnage set-point to hit the target P80.
Program hard interlocks so a high-level or high-power alarm in any stage throttles upstream feeders first, protecting mills and conveyors.
Use Bond-based models for crushers and the Morrell–Barratt equation or SGI test data for ball/HIG-mills to calculate specific energies (kWh t-1). Express each stage with:
Wi (work index) or Mi parameters.
Reduction ratio Ri = F80,i / P80,i.
Energy equation Es = 10 × Wi (1/√P80 – 1/√F80).
Sum Es for all stages; then iterate the split points so the derivative of total energy with respect to each ratio approaches zero—this gives the minimum overall kWh t-1 while respecting downstream liberation limits and equipment max-power ratings.
Worked Example: Multi-Stage Grinding of Peppercorns
This example designs a multi-stage grinding system to reduce whole peppercorns from a coarse feed to a fine powder. The goal is to achieve a specified product size while optimizing energy efficiency and adhering to mechanical limits.
Bond Work Index for the spice, \( W_i = 15.0 \, \text{kWh/ton} \)
Ambient temperature, \( T = 25.0 \, ^\circ \text{C} \)
Calculate the total reduction ratio:
\[
R_{\text{total}} = \frac{D_f}{D_p} = \frac{10000.0}{100.0} = 100.0
\]
Determine the number of stages. Using a target per-stage ratio of \( R_{\text{stage}} = 5.0 \) (within the empirical range of 2 to 10), the exact number is:
\[
n_{\text{exact}} = \frac{\log(R_{\text{total}})}{\log(R_{\text{stage}})} = \frac{\log(100.0)}{\log(5.0)} = 2.861
\]
Rounded up to the nearest whole number, \( n_{\text{stages}} = 3 \).
Compute the actual reduction ratio per stage for three identical stages:
\[
R_{\text{actual}} = R_{\text{total}}^{1/n_{\text{stages}}} = 100.0^{1/3} = 4.642
\]
Calculate intermediate particle sizes after each stage:
After stage 1: \( D_1 = \frac{D_f}{R_{\text{actual}}} = \frac{10000.0}{4.642} = 2154.435 \, \mu m \)
After stage 2: \( D_2 = \frac{D_1}{R_{\text{actual}}} = \frac{2154.435}{4.642} = 464.159 \, \mu m \)
After stage 3: \( D_3 = \frac{D_2}{R_{\text{actual}}} = \frac{464.159}{4.642} = 100.000 \, \mu m \), which matches the target.
Calculate total energy consumption using Bond's Law. The energy per stage is \( E = 10 \times W_i \times \left( \frac{1}{\sqrt{D_p}} - \frac{1}{\sqrt{D_f}} \right) \) for each stage's input and output sizes. Summing over all three stages:
\[
E_{\text{total}} = 13.5 \, \text{kWh/ton}
\]
This value is derived from the staged calculations, with intermediate energy contributions ensuring no over-grinding.
Convert temperature for any kinetic analysis (if required for heat-sensitive materials):
\[
T_{\text{Kelvin}} = T + 273.15 = 25.0 + 273.15 = 298.15 \, \text{K}
\]
Validate against empirical ranges:
Particle sizes: \( D_f = 10000.0 \, \mu m \) and \( D_p = 100.0 \, \mu m \) are within the Bond's Law range of 50 to 10,000 µm.
Per-stage reduction ratio: \( R_{\text{actual}} = 4.642 \) is within the recommended range of 2 to 10.
Final Answer: The design requires 3 stages with a reduction ratio of 4.642 per stage, intermediate particle sizes of 2154.435 µm and 464.159 µm, and a total energy consumption of 13.5 kWh/ton. The temperature for kinetic calculations is 298.15 K if needed.
"Un projet n'est jamais trop grand s'il est bien conçu."— André Citroën
"La difficulté attire l'homme de caractère, car c'est en l'étreignant qu'il se réalise."— Charles de Gaulle
