A flotation cell is an agitated, aerated tank through which slurry flows, with the valuable mineral carried into a froth that overflows a lip while the remaining slurry passes to the next cell. Each particle needs sufficient time in the cells to contact a bubble, so cell sizing is fundamentally a matter of flow and residence time. If 11 m³ of slurry arrives each minute and the required mean residence time is 20 minutes, the cells must hold about 11 × 20 = 220 m³ of slurry, and the installed volume must be somewhat larger because part of each cell is occupied by air and by the rotor and its housing.
Sizing a conventional flotation bank is a short calculation, but each input requires care. This article sets out the steps in the order a designer works through them, with one worked example throughout, and then covers conditioners and column cells.
Slurry flow
Flotation cells hold slurry rather than dry solids, so the dry solids rate is converted to a volumetric flow of solids plus water:
where
- slurry volume flow (m³/h)
- dry solids feed rate (t/h)
- solids density (t/m³, numerically the specific gravity)
- solids content by mass (fraction, for example 0.35)
Worked example: a copper rougher treats 300 t/h of dry solids at 35% solids by mass, with a solids specific gravity of 2.8.
Qs = 300 × (1 ÷ 2.8 + 0.65 ÷ 0.35) = 300 × (0.357 + 1.857) = 664 m³/h, or 11.07 m³/min.
Pulp density has a large effect. The same 300 t/h at 30% solids would be 807 m³/h, about 21% more volume for the same tonnage, so the design should use the lowest density at which the plant will realistically operate, or check both ends of the range. Recirculating streams, such as cleaner tailings returned to the rougher feed, must be added to the flow.
Plant residence time
The laboratory kinetics test gives the flotation time required to reach the target recovery. In the companion kinetics test article, the rougher reached the target copper recovery of 88.5% at 8 minutes. This time is multiplied by a scale-up factor that accounts for short-circuiting in continuous cells and the other differences between laboratory and plant:
where
- design residence time of the plant bank (min)
- laboratory flotation time that gives the target recovery (min)
- scale-up factor from laboratory to plant (dimensionless)
With a scale-up factor of 2.5, the plant rougher requires 2.5 × 8 = 20 minutes. Published factors for conventional cells mostly fall between about 2 and 4, and the value for the duty should be obtained from the laboratory and the equipment supplier.
Total cell volume
Part of every cell is occupied by air and part by the rotor, stator and other internals, and an allowance for this space ensures that the slurry still receives the required time:
where
- total nominal cell volume needed in the bank (m³)
- slurry flow (m³/min when the time is in minutes)
- allowance for air in the pulp and for internals (fraction of cell volume)
- ,
- number of cells, and the nominal volume of one cell (m³)
With an allowance ε = 0.15, meaning 15% of the cell volume is unavailable to slurry (the supplier's figure for the selected machine should be used):
Vtotal = 11.07 × 20 ÷ (1 − 0.15) = 260 m³.
Number and size of cells
The total volume is divided by the nominal volume of the selected cell and rounded up. With 50 m³ cells, 260 ÷ 50 = 5.2, so six cells are required. The installed volume of 300 m³ gives an actual residence time of 300 × 0.85 ÷ 11.07 = 23 minutes, a modest margin for higher throughput or lower pulp density.
The number of cells is a design decision in its own right. Fewer, larger cells have lower capital and operating costs, but each well-mixed cell allows some pulp through quickly, whereas more cells in series behave more like the laboratory batch cell. As the kinetics article shows, a single well-mixed cell needs almost four times the batch time to reach the same recovery, and six cells about 1.2 times. Suppliers give guidance on the minimum number of cells per bank for each duty, and a bank must also have enough cells to tolerate one being offline for maintenance if the plant cannot be stopped.
| Cell size | Cells needed for 260 m³ | Installed volume (m³) | Actual residence time (min) |
|---|---|---|---|
| 30 m³ | 9 | 270 | 20.7 |
| 50 m³ | 6 | 300 | 23.0 |
| 100 m³ | 3 | 300 | 23.0 |
Three 100 m³ cells give the same installed volume as six 50 m³ cells but would perform worse because of short-circuiting, so the six-cell option is the better starting point in this case.
Conditioner sizing
A conditioner is an agitated tank that allows the reagents time to act before flotation. It is sized in the same way, without the air allowance. For 5 minutes of conditioning in the worked example, the live volume is 11.07 × 5 = 55 m³, plus freeboard above the slurry level.
Column cell sizing by carrying capacity
Column cells are tall and narrow, with wash water added to the froth to clean it, and they are used mainly in cleaning stages. Their size is often limited not by residence time but by the mass of concentrate that the froth can carry over the lip per square metre of column. This limit is the carrying capacity:
where
- carrying capacity: the most concentrate the froth can carry over the lip (g/min per cm² of column area)
- empirical constant: 0.068 from Espinosa-Gomez and co-workers, mostly small columns; about half that has been proposed for large industrial columns
- 80% passing size of the concentrate (µm)
- density of the concentrate solids (g/cm³)
- ,
- concentrate mass rate (g/min) and required column area (cm²)
Worked example: a cleaner column must produce 12 t/h of concentrate with d80 = 40 µm and a solids density of 4.2 g/cm³.
With a = 0.068, Ca = 0.068 × 40 × 4.2 = 11.4 g/min·cm². The concentrate rate is 12,000,000 ÷ 60 = 200,000 g/min, so the area required is 200,000 ÷ 11.4 = 17,500 cm², or 1.75 m², corresponding to a column about 1.5 m in diameter. With the more conservative value a = 0.035 proposed for large industrial columns, Ca = 5.9 g/min·cm², the area is 3.4 m² and the diameter about 2.1 m.
The two results differ by a factor of two, which is why column sizing should be confirmed by pilot work or the supplier's data for the ore. Columns are also checked for residence time by the same flow and time method as conventional cells, and the larger of the two areas governs.
Final design checks
The calculation should be repeated at the highest flow the plant will see. Launders and concentrate pumps must be checked for the concentrate mass and froth volume, particularly in the early cells where most of the mass floats. Suppliers also check the concentrate load per metre of lip and per square metre of froth surface, adding crowders and launders where needed. Finally, banks are stepped so that slurry flows by gravity from cell to cell, so the fall available in the plant layout must be confirmed.
Glossary
- Bank
- A row of flotation cells connected in series, so slurry flows from one to the next.
- Residence time
- The average time slurry spends in a cell or bank: volume divided by flow.
- Air allowance
- The fraction of the cell volume taken up by air bubbles and internals, which slurry cannot use.
- Conditioner
- An agitated tank that gives reagents time to act before flotation.
- Column cell
- A tall flotation cell with a deep froth washed by water, mostly used for cleaning.
- Carrying capacity
- The most concentrate a froth can carry over the lip per unit of column area per minute.
- Lip loading
- Concentrate mass overflowing per metre of froth lip; a check on froth removal.
- Short-circuiting
- Slurry leaving a cell sooner than average and so having less chance to float.
Frequently asked questions
How is flotation cell volume calculated?
Multiply the slurry flow by the plant residence time, divide by one minus the air allowance, then divide by the volume of one cell and round up. The plant residence time is the laboratory time multiplied by a scale-up factor.
What air allowance should be used?
It depends on the machine and the duty. The equipment supplier's figure for the selected cell should be used; the worked example uses 15% to illustrate the method.
How many cells should a flotation bank have?
Enough to keep short-circuiting low and to allow for maintenance. More cells in series behave more like the laboratory batch test, and suppliers give minimum cell numbers for each duty.
How is a column cell sized?
Usually by carrying capacity, which sets the area needed to carry the concentrate over the lip, with a check on residence time. The result should be confirmed by pilot work or supplier data, because published carrying capacity constants vary by a factor of two.
Discuss your plant with a metallurgist
Tiegel is a Perth metallurgy company. We help with testwork, plant optimisation, metal accounting and investigations, and we build the Tiegel Suite software.
Talk to a metallurgistReferences and further reading
- Wills, B.A. and Finch, J.A. (2016). Wills' Mineral Processing Technology, 8th edition. Butterworth-Heinemann. Chapter 12, Froth flotation.
- Finch, J.A. and Dobby, G.S. (1990). Column Flotation. Pergamon Press.
- Espinosa-Gomez, R., Finch, J.A. and Johnson, N.W. (1988). Carrying capacity limitations in flotation columns. Column Flotation '88, Society for Mining, Metallurgy and Exploration. Link
Worked examples and charts marked illustrative use invented figures, not data from any operation.