Filter Press for Mining: Selection, Sizing, and Cycle Math

Last updated: 2026-10-30

The right size for a filter press for mining duty falls out of the cake, not the catalogue: a duty of 40 t/h of dry solids at 18% cake moisture and a 90-minute cycle needs roughly 2,800 m² of effective filtration area — delivered as two large presses of about 1,400 m² each — and every other decision, from membrane plates to feed pump, follows from that number.

Key Takeaways

• Size from cake mass per cycle: dry solids feed, cake moisture, and cycle time fix plate area.
• Membrane plates repay their premium wherever cake moisture gates haulage, stacking, or transport.
• Cloth selection by particle size and pH drives cycle economics more than plate format does.
• Keep feed solids above roughly 25% by weight; thin feeds stretch fill time and destroy capacity.
• Budget cloth washing, drip trays, and cake handling early — they are real cost, not accessories.

This page is the equipment-level selection guide inside our solid liquid separation cluster: duty definition, plate construction choice, cloth selection, sizing arithmetic with a full worked example, troubleshooting, and cost. For why the moisture target matters, see filter cake moisture; for squeezing more tonnes from an installed press, see filter press cycle optimization.

Solid-Liquid Separation: selection rules

Where Does a Filter Press for Mining Fit in a Hydromet Flowsheet?

A filter press is a batch-operated pressure filter that separates solids from slurry by pumping feed into chambers between cloth-covered plates until the chambers are full of cake, then opening to discharge the cake one plate gap at a time. It sits at the end of the solid-liquid train — after the thickener, after leaching or precipitation — wherever the flowsheet demands a handleable, transportable solid rather than a pumpable slurry.

Four duties dominate in mining and hydrometallurgy, and they do not reward the same hardware:

Dry-stack tailings. Tailings dewatering to a stackable cake is the largest single application. The cake must hold a stable slope on the stack, which usually means 15–20% moisture or better on fine tailings (indicative, wet basis, to be confirmed by test work). The press replaces or supplements the tailings dam and is justified as much by water recovery and permitting as by the cake itself; the wider flowsheet logic is covered in our dry-stack tailings application page.

Concentrate dewatering. Base-metal concentrates ship with contractual moisture limits, typically 8–12% (wet basis, indicative). Pressure filtration is usually the only technology that reaches these levels on flotation concentrates, and the membrane squeeze is what closes the last few points.

Solvent-extraction crud and interstage solids. Small, occasional, unpleasant. Crud presses run infrequently at modest plate counts; what matters is fast changeover, easy cleaning, and corrosion-resistant materials rather than throughput.

Precipitate and neutralization solids. Hydroxide precipitates, gypsum from neutralization, and jarosite-type residues are fine, often sticky, sometimes thixotropic — the duties where cloth selection and cake discharge design make or break the installation.

Two working definitions anchor everything that follows. Cake moisture is a mass fraction of water that the discharged cake carries, quoted on a wet basis (water mass divided by total wet cake mass) unless a vendor states otherwise — always confirm the basis before comparing guarantees. Cloth blinding is a progressive blockage of the weave that fines, precipitates, and slimy particles cause, and it announces itself as fill time creeping upward cycle after cycle long before anyone measures it directly.

How Do You Size a Filter Press for Mining Throughput and Cycle Time?

Sizing is four multiplications and one division: the press must discharge enough cake per cycle, at the chosen cycle time, to match upstream solids production.

The worked example. Duty: 40 t/h of dry solids (dry basis) from a gold tailings thickener underflow. Target cake: 18% moisture (wet basis). Assumed cycle: 90 minutes, all-in.

1. Wet cake mass. At 18% moisture, solids are 82% of the cake, so wet cake = 40 ÷ 0.82 = 48.8 t/h.
2. Cake per cycle. A 90-minute cycle means each discharge must deliver 48.8 × 1.5 = 73.2 t of wet cake.
3. Cake volume per cycle. Mineral cakes bulk at roughly 1.4–1.8 t/m³; take 1.5 t/m³ (indicative, to be confirmed on your solids). Volume = 73.2 ÷ 1.5 = 48.8 m³.
4. Chambers required. A 2,000 × 2,000 mm plate offers about 3.9 m² of face area per side and, at a 35 mm chamber depth, about 0.137 m³ of chamber volume per plate. Chambers = 48.8 ÷ 0.137 ≈ 356 plates.
5. Total effective area. 356 plates × 7.8 m² per plate ≈ 2,780 m².
6. Machine layout. One press this large is a single point of failure and cannot be re-clothed while running, so the standard answer is two presses of roughly 1,400 m² each (about 178 plates) in parallel, each handling half the duty.

The specific throughput line deserves attention because it is where optimistic vendor quotations hide. Fine, slow-filtering tailings sit at the low end of 8–25 kg/(m²·h); coarse, fast-filtering concentrates and washed residues reach 30–60. If a quotation implies a number above the range your particle-size distribution justifies, ask which filtration rate — measured on your slurry, not on their test mineral — supports it.

Where the 90 minutes go. A realistic cycle budget for fine tailings at 50–55% feed solids (indicative): fill 30 min, pressure-hold 25 min, membrane squeeze 10 min, core blow and drain 7 min, open-discharge-close 18 min. Nearly half the cycle is not filtration at all — it is mechanical dead time, which is why the judgment below about feed solids matters more than any single equipment specification.

One judgment that only an engineer who has commissioned these units would emphasize: feed solids below about 25% by weight wreck cycle economics in a way that datasheets do not show. Fill time stretches until the press spends most of its life pumping water through a cake that barely exists, and since fill time is dead time paid for at full installed-plate cost, a 25-minute fill on a 90-minute cycle quietly destroys the throughput the plate area was bought to deliver. Thickening upstream to 45–55% solids costs a fraction of the plate area needed to filter that water instead; our thickener sizing guide covers that step.

Plate and Frame or Membrane Recessed Plate: Which Construction Suits Which Duty?

Two constructions dominate. In the plate and frame filter press, alternating plates and frames form the chambers; the cake grows unrestrained until the chambers are full. In the membrane (diaphragm) recessed-plate press, the cake first fills the chamber at feed pressure, then flexible diaphragms — inflated with water or air at 8–16 bar (indicative) — compress the cake, expelling liquor that feed pressure alone cannot remove.

A membrane filter press is a recessed-plate design that adds those flexible squeeze diaphragms behind the cloth, and its value is measured in one currency: filter cake moisture points. The squeeze typically strips 4–8 percentage points of moisture relative to the same cake filtered without squeeze (indicative, fine mineral slurries, to be confirmed by test work), and it also yields a more uniform cake across the plate face, which matters for discharge reliability.

Reduced to two decision sentences: For a dry-stack tailings or fine precipitate duty where cake moisture gates haulage, stacking stability, or downstream transport, a membrane press is preferred because the diaphragm squeeze buys those 4–8 moisture points that chamber pressure alone leaves in the cake. For a coarse, free-filtering concentrate above roughly 100 µm where vacuum or simple pressure filtration already approaches the contractual moisture limit, a plate and frame construction is preferred because the squeeze adds little moisture benefit while adding diaphragms, squeeze-water systems, hoses, and a larger maintenance surface.

For readers weighing the press against the alternatives, the head-to-head economics against clarifying-thickening duty are worked through in thickener vs filter press, and the boundary between press technology families — chamber presses versus horizontal pressure filters — is treated in pressure filter vs filter press. Where the slurry is too fine or too abrasive for cloth-based filtration at acceptable cost, a decanter is the usual fallback; its fit for mineral slurries is assessed in decanter centrifuge for mineral slurry.

How Should Filter Cloth and Feed Pumping Be Chosen?

Cloth selection is decided by two properties of the slurry: particle size and chemistry. The rule of thumb: the weave must be fine enough that the initial cloudy filtrate clears within one to two minutes as a cake bridge forms over the pores, yet open enough that filtration flux stays economic. Choose too fine and the press blinds; choose too open and fines report to the filtrate, loading the clarifier and returning solids to the wrong side of the water balance.

Chemistry sets the fabric. Polypropylene is the default for pH 2–13 below roughly 90 °C (indicative); polyester offers higher strength and acid resistance but hydrolyzes in hot alkaline liquors; polyamide suits alkaline duties with abrasive solids. Oxidizing liquors — ferric, chlorinated — shorten fabric life regardless of polymer, one more reason cloth life figures must come from test work on your slurry.

Blinding prevention is a cycle-discipline problem more than a fabric problem. Four measures keep permeability: maintain enough feed solids for an immediate cake bridge; core-blow every cycle so residual cake does not dry into the weave; schedule cloth washes (acid for carbonates, caustic for slimy precipitates) before fill time degrades by 20%, not after; and trend fill time per cycle as the most telling health metric on the unit.

Feed pumping is the other half of the cycle. The standard arrangement is a centrifugal pump for rapid initial fill at low resistance, handing over to a hydraulically driven piston-diaphragm pump as cake resistance builds, terminating at 8–12 bar (indicative); squeeze water for membrane plates needs its own clean supply at 14–16 bar. Two commissioning realities: size the low-pressure fill pump generously, because it sets the first and least productive third of the cycle; and never let the pump curve and the press guarantee be quoted by different vendors without someone checking the intersection.

Seven Troubleshooting Scenarios: Symptom, Cause, and Fix

1. Cake not forming, or thin and soupy. Symptom: chambers discharge slurry, not cake. Cause: feed solids too low for a bridge to form, cloth too open for the particle size, or feed pump losing prime. Fix: raise feed solids via the thickener, drop to a tighter weave, verify pump suction and fill-valve sequencing before blaming the cloth.
2. Cake moisture above target. Symptom: cake passes squeeze but arrives wet. Cause: insufficient squeeze pressure or time, a failed diaphragm, uneven cake across the plate, or fill pressure that never reached terminal value. Fix: log squeeze-water pressure and cake thickness per plate; restore terminal feed pressure; repair or isolate the failed plate — one dead diaphragm wet-downs the whole discharge.
3. Fill time growing cycle over cycle. Symptom: the classic blinding signature. Cause: fines loading the weave, carbonates or sulfates crystallizing in the cloth, or washes skipped during a production push. Fix: chemical cloth wash immediately, review wash chemistry against the liquor analysis, and tighten the wash interval; if recovery is incomplete, the weave was wrong for the duty from the start.
4. Cloth wear at plate edges and corners. Symptom: filtrate runs cloudy early in cycle; cloth life far below quoted. Cause: plate misalignment, worn sealing surfaces, over-tight clamping, or wrinkled cloth concentrating abrasion. Fix: inspect sealing faces at every cloth change, verify plate-shift alignment, and insist on cloth cut and sewn to the plate pattern rather than field-trimmed.
5. Leakage between plates during fill. Symptom: dribbles at the plate-pack edges. Cause: insufficient hydraulic closing force, worn gaskets, cake crust on sealing faces from the previous cycle, or feed pressure exceeding the plate-pack design. Fix: check closing force against the vendor curve, clean sealing faces every discharge (plate shakers help), and replace gaskets as a set rather than individually.
6. Slow fill at normal pressure. Symptom: pressure builds correctly but flow is low from the first minute. Cause: throttled or undersized fill line, feed pump far left on its curve, air locked in the feed manifold, or — most commonly — cloth already partially blinded before the cycle began. Fix: verify line sizing and valve positions, bleed the manifold, and treat persistent slowness as a cloth problem, not a pump problem.
7. Plates hanging up during discharge. Symptom: plate shifter stalls or plates transport unevenly, occasionally pulling cloth. Cause: cake sticking to one cloth face, unequal cake thickness twisting the pack, worn shifter drive components, or cloth overhang catching. Fix: address cake release first — cloth surface finish, squeeze profile, discharge vibration — then mechanically; a new shifter motor on a sticking-cake problem wastes a shift and changes nothing.

What Does a Filter Press for Mining Cost to Buy and Operate?

Capital cost scales with effective area, membrane content, and automation. Indicative installed equipment costs (large-format presses, excluding building, upstream thickener, and cake conveying): plate-and-frame roughly 250–550 USD per m² of effective filtration area, membrane roughly 400–900 USD/m², with automated plate-shifting and cloth-washing systems at the upper end. These are planning ranges only — scope varies widely between vendors, and every figure must be confirmed by quotation.

Operating cost has four parts, and the honest way to compare quotations is per tonne of dry solids:

The calculation basis above is per tonne of dry solids processed at the worked-example duty, with ranges reflecting feed fineness and liquor chemistry; all values are indicative and to be confirmed by test work and vendor quotation. The cheapest press on the table usually carries the thinnest cloth-washing and spares scope, and a 30% saving on cloth unit price evaporates if the fabric blinds twice as fast.

What Do Commissioned Installations Show? (Indicative Case Data)

Three anonymized data points from commissioning and operating records, each indicative and tied to its stated basis:

• Gold tailings dry-stack, two-press layout. 42 t/h dry solids (dry basis), achieved cake moisture 17.8% (wet basis) against an 18% guarantee, all-in cycle 88 minutes including a 4-minute core blow. The squeeze-water system, not the feed pump, proved the commissioning bottleneck.
• Copper concentrate shipment duty. Membrane press at 12 bar squeeze: concentrate moisture 9.5% (wet basis) on a flotation concentrate with P80 near 55 µm; cloth life above 5,000 cycles with monthly acid washes — the cleanest duty a cloth can see.
• Nickel hydroxide precipitate. Same press family, different behavior: cake moisture 31% (wet basis), cloth life near 2,000 cycles, weekly caustic washes. Identical hardware, opposite cloth strategies — the slurry, not the press, writes the operating manual.

The cross-case pattern worth carrying into procurement: cycle time targets were met on all three sites, but only after fill time was trended and cloth washing scheduled by trend rather than by calendar.

A 25-Item Checklist for Reviewing a Vendor Technical Offer

Before any purchase order, walk the technical offer against this checklist: a vendor who answers every line without prompting has commissioned equipment before.

FAQ: Filter Press Questions Project Managers Ask Most

How is plate area calculated from a solids throughput?

Work backward from the cake. Divide dry solids per cycle by the cake solids fraction to get wet cake mass, divide by cake bulk density to get cake volume, then divide by chamber volume per plate to get the plate count. At 40 t/h of dry solids, 18% moisture, and a 90-minute cycle, that chain gives roughly 356 chambers of 0.137 m³ — about 2,780 m² of effective area, normally two presses in parallel. Confirm cake density and moisture by test work first.

What cycle time should be assumed before filtration test work is available?

Use 90–120 minutes all-in for fine tailings and precipitates, and 60–90 minutes for coarser, free-filtering solids, then treat both as placeholders. The honest breakdown is fill plus filtration plus squeeze plus blow plus discharge, and the non-filtration share is typically 40–50% of the cycle. Any vendor quoting a total cycle without stating each element is quoting a number that has never been commissioned.

Why does wet-basis versus dry-basis moisture matter so much?

Because the same cake can be described two ways and both sound fine. A cake at 18% moisture on a wet basis is 22% water on a dry basis, so a vendor quoting the larger dry-basis number can appear two points better than a competitor quoting wet basis. Every moisture guarantee should state the basis, the test method, and the sampling point; guarantees without a stated basis are not comparable and should be returned for clarification before commercial evaluation.

Can a filter press run continuously?

No — it is inherently a batch machine, and continuity comes from layout, not the press itself. Single-press installations accept cyclical flow to the cake conveyor; two- or three-press installations stagger their cycles so discharge and cloth maintenance on one unit overlap with filtration on another. For duties where truly continuous operation is essential and the solids are too fine for cloth filtration economics, a decanter centrifuge is the conventional alternative.

Does higher squeeze pressure always mean lower cake moisture?

Only up to a point. Moisture falls steeply as squeeze pressure rises to roughly 8–12 bar and then flattens, because the remaining water is bound in fine pores and around particles rather than free. Beyond that knee, added pressure buys longer squeeze times and membrane stress more than moisture points. The useful lever at constant pressure is squeeze duration and cake uniformity, which is why fill practice influences moisture as much as the squeeze-water pump does.

How often must filter cloth be replaced, and what drives the cost?

Between roughly 2,000 and 8,000 cycles in mining duties (indicative), with the spread set by particle sharpness, liquor chemistry, and washing discipline — not by press brand. Abrasive concentrates with regular acid washing reach the high end; slimy precipitates with skipped washes fall well below. Cost is best budgeted per tonne of dry solids processed, which converts cloth from a maintenance surprise into a predictable operating line item comparable across vendors.

What single operating metric best predicts trouble?

Fill time per cycle, trended daily. Every failure mode that matters — cloth blinding, feed solids drifting low, pump wear, valve leakage — shows up as fill time lengthening weeks before throughput or moisture reports register the loss. A press whose fill time has grown 20% over a month is telling the operating team exactly which day to schedule a cloth wash, and that one trend chart is the cheapest reliability upgrade on any installation.

Specification Support

If you are specifying a press against a defined duty, the shortest path to firm numbers is a short technical exchange: send the dry solids rate, feed solids, particle-size distribution, liquor pH, and the moisture target, and our process group will return a preliminary plate-area calculation, a membrane-versus-plate-and-frame recommendation, and an indicative cycle budget — via the request-a-quote page, or browse the resources library first if the flowsheet is still moving. Every performance figure on this page is indicative until confirmed by test work on your slurry.

Filter Press Area Calculator from Leaf Test Data, step by step

Conclusion

A filter press for mining duty is sized the moment three numbers are fixed — dry solids feed, cake moisture, and cycle time — and the 40 t/h worked example converts them into roughly 2,800 m² of effective area, two parallel presses, and a cloth and feed-pump strategy chosen by particle size rather than by catalogue. Work the sizing chain before the quotation, insist on stated bases for every guarantee, and instrument fill time from the first week. Bring us your slurry analysis, and we will run the numbers with you.

References

1. Society for Mining, Metallurgy & Exploration, SME Mineral Processing Handbook (solid-liquid separation and tailings dewatering sections). https://www.smenet.org/
2. Rushton, A., Ward, A.S. & Holdich, R.G., Solid-Liquid Filtration and Separation Technology, 2nd edition, Wiley-VCH, 2000.
3. Green, D.W. & Southard, M.Z. (eds.), Perry’s Chemical Engineers’ Handbook, 9th edition, McGraw-Hill, 2019 (filtration and separation chapters).
4. Wakeman, R.J. & Tarleton, E.S., Solid/Liquid Separation: Principles of Industrial Filtration, Elsevier, 2005.
5. U.S. Geological Survey, Mineral Commodity Summaries, annual series (copper, nickel, gold commodity chapters). https://www.usgs.gov/centers/national-minerals-information-center
6. ASME Boiler and Pressure Vessel Code, Section VIII, Division 1, Rules for Construction of Pressure Vessels, ASME International (squeeze-water and feed-system pressure parts).
7. ASTM A240/A240M, Standard Specification for Chromium and Chromium-Nickel Stainless Steel Plate, Sheet, and Strip, ASTM International (plate and wetted-part materials).

Changelog

• 2026-10-30: First published.