Vacuum Filtration vs Pressure Filtration: Which Wins?

Last updated: 2026-11-03

Vacuum filtration vs pressure filtration comes down to one variable: the applied pressure differential against the cake’s resistance to flow. Practical vacuum systems generate at most about 80 kPa (0.08 MPa) of differential; pressure filtration delivers 0.4-2.5 MPa, 5 to 30 times the driving force. The winning technology is the one whose cake resistance stays under that ceiling, as the solid-liquid separation overview explains.

Key Takeaways

• Vacuum tops out near 80 kPa (0.08 MPa) of differential; pressure filtration delivers 0.4-2.5 MPa, 5-30 times more.
• Cake resistance, not pump capacity, sets the moisture ceiling: compressible cakes stop responding beyond roughly 300 kPa.
• Vacuum wins on continuous operation and cost per square meter; pressure wins on final moisture and fine-particle capture.
• Leaf tests at 100, 300, and 600 kPa resolve the choice for a specific slurry in a single lab campaign.
• Expect 14-20% moisture from vacuum on coarse concentrates versus 8-14% (indicative) from membrane pressure filtration.

What Separates Vacuum Filtration From Pressure Filtration?

Three definitions anchor the comparison:

• Vacuum filtration is a solid-liquid separation that drives liquor through a filter medium by holding the downstream side below atmospheric pressure, so the atmosphere itself supplies the differential. Because the wet side sits at ambient pressure, the practical ceiling is roughly 80 kPa (0.08 MPa, 12 psi) minus losses in pipes, cloth, and the barometric leg.
• Pressure filtration is a separation that applies pump, compressed-air, or hydraulic pressure above the medium, typically 0.4-2.5 MPa (58-360 psi), with membrane presses adding a mechanical squeeze of up to about 1.5-2.0 MPa on the formed cake.
• A vacuum belt filter is a continuous machine that forms, washes, and dewateres the cake on a traveling permeable belt over stationary drain boxes, discharging continuously.
• A filter press is a batch machine that traps slurry between recessed cloth-covered plates, then squeezes and blows the cake dry before plates open.

The physics that separates them is brutal in its simplicity. Filtration rate scales with roughly the square of the pressure differential only for incompressible cakes; for the compressible cakes that dominate hydrometallurgy, rate gains flatten once the cake collapses onto the cloth. That is why the extra megapascals of pressure filtration buy drier cake and finer-solids capture more than they buy raw speed — and why vacuum remains viable at all. Where specific cake resistance is low (coarse, crystalline, free-draining solids), 80 kPa is plenty and the cheaper technology wins. Where resistance is high (fine precipitates, hydroxide flocs, clay-laden tailings), vacuum never builds a dischargeable cake and the choice is made for you.

One further consequence deserves emphasis before any equipment is discussed. The driving force is fixed by the technology, but the resistance is fixed by the slurry — so the entire comparison reduces to matching one against the other. Engineers who start from machines (a brand of belt, a plate size) rather than from measured cake behavior routinely buy either far more area than the duty needs or a press that the slurry blinds in a month. Confirm which case you occupy with test work on a pilot-scale vacuum filtration unit before any sizing exercise; the cost of a week of leaf testing is trivial against a mis-sized installation.

Vacuum Filtration vs Pressure Filtration: Driving Force and Cake Resistance Limits

Vacuum filtration vs pressure filtration is ultimately a question of where the slurry sits relative to each technology’s cake-resistance ceiling. Specific cake resistance, α (m/kg), measures how much flow resistance a unit mass of deposited cake offers; it is the single number that predicts everything else. Coarse flotation concentrates typically show α near 10^10-10^11 m/kg, well within vacuum reach. Fine precipitates and neutralization sludges can exceed 10^12-10^13 m/kg (indicative), where vacuum dewatering stalls entirely — liquor barely migrates, the cake cracks, and air rushes through the cracks doing no work.

Treat every range in the table as typical or indicative and to be confirmed by test work on your own slurry. Two structural consequences follow. First, the trade between the technologies is area versus pressure: vacuum compensates for weak driving force with cheap, continuous square meters, while pressure filtration buys a stronger differential at higher cost per square meter and pays for it in batch handling. Second, the deeper driving-force comparison of pressure filters and filter presses shows the same pattern in equipment terms — the choice is rarely about which machine is better in the abstract, only which ceiling the slurry violates.

A practical reading of the ceiling row: if leaf testing shows specific cake resistance below about 10^11 m/kg with weak compressibility, vacuum deserves a full economic evaluation and often wins outright. If resistance sits between 10^11 and 10^12 m/kg (indicative), the decision hinges on the moisture specification — vacuum may form a cake but cannot dry it far enough. Above 10^12 m/kg, stop evaluating vacuum and spend the budget on membrane press test work instead. These thresholds are indicative order-of-magnitude guides; the compressibility index fitted from your own leaf data overrides them every time, because two slurries with identical resistance can behave oppositely under increasing pressure.

Vacuum Filtration vs Pressure Filtration on Cake Moisture and Throughput

Three duty types cover most hydrometallurgical dewatering. The figures below are typical or indicative ranges compiled from operating practice and must be confirmed by leaf or press tests; cake moisture basis is wet mass.

Energy basis: pump and compressor power per tonne of dry solids deposited, typical plant figures, to be confirmed by measurement. Moisture targets for transport and downstream drying should be set against measured filter cake moisture data rather than vendor tables.

The verdicts are deliberately concrete:

• For coarse flotation concentrates with P80 above 100 µm, vacuum filtration is preferred because the cake resistance is low enough that 80 kPa already delivers 14-20% moisture at the lowest cost per square meter of any continuous option.
• For fine tailings and neutralized slurries below 40 µm P80, pressure filtration is preferred because vacuum never builds a dischargeable cake and residual moisture exceeds transport and storage limits.
• For hydroxide or gypsum sludge with highly compressible cakes, membrane pressure filtration is preferred because a mechanical squeeze at 1.5 MPa consolidates what no amount of vacuum area can.
• For duties needing multi-stage counter-current washing on an open cake, vacuum filtration is preferred because displacement washing is visible, adjustable, and continuous.

Where the slurry sits between these categories, size the equipment with the method in filter press selection for mining and verify with pilot work before committing.

Two caveats keep these verdicts honest. First, moisture basis matters: figures here are wet-mass percentages, and converting between wet and dry basis shifts every number by several points — mixing the two bases is the most common error in vendor comparisons. Second, throughput scales down sharply with cake thickness targets; a duty specified at 10 mm cake will show roughly half the specific throughput of the same slurry at 25 mm (indicative), so always compare figures at equal cake thickness before concluding that one technology is faster.

A 7-Step Selection Sequence: How Do You Choose Between Them?

Filter selection in mining should follow a fixed sequence, not vendor persuasion. In practice, vacuum filtration vs pressure filtration is decided in the laboratory weeks before any purchase order, and the sequence below is how experienced engineers get there without wasted pilot work. Run these seven steps in order:

1. Characterize the slurry. Measure P80, solids content, density, and rheology. A yield-stress slurry above roughly 30-40 Pa (indicative) will feed a press poorly and form uneven vacuum cakes; flag it early.
2. Run leaf tests at 100, 300, and 600 kPa. A three-point pressure series on a leaf test apparatus (or 20-80 kPa equivalent for vacuum simulation) shows whether cake formation rate responds to pressure — the fastest screening signal available.
3. Compute specific cake resistance and compressibility. Fit α versus pressure from the leaf data. A compressibility index below about 0.3 (typical) favors vacuum; above 0.5, only pressure with a membrane squeeze will dry the cake.
4. Check the moisture specification and downstream drying cost. If the spec is below roughly 15% w/w (typical transport threshold), vacuum usually cannot meet it directly and the drying penalty must be priced into the comparison.
5. Evaluate the wash requirement. If wash ratio exceeds roughly 2 m3/t solids, compare continuous displacement washing on vacuum equipment against repulping and re-filtration between press cycles.
6. Evaluate energy and labor. Add vacuum pump power, compressor demand, cloth changes, and operator hours per tonne; batch presses carry per-cycle labor that continuous vacuum machines avoid.
7. Decide on total cost per tonne of dry solids. Combine capex amortization, energy, labor, cloth consumption, and any drying or disposal penalties — then commit to pilot confirmation.

Worked arithmetic from a recent indicative campaign: leaf tests gave 0.22 t/m2/h of dry solids on the vacuum case and 0.75 t/m2/h on the membrane press case. For a 24 t/h dry-solids duty, vacuum needs 24/0.22 ≈ 110 m2 of active area, while pressure needs 24/0.75 ≈ 32 m2 — an area ratio of about 3.4:1. That ratio, multiplied by the installed cost per square meter of each technology, is the core of the capital comparison. Both duties can be reproduced in-house with a pilot-scale plate-and-frame filter press and matched vacuum equipment before the numbers are frozen. Full step-by-step guidance for the vacuum side is given in the vacuum belt filter selection guide.

Energy, Consumables and Capital Cost

All figures below are typical or indicative ranges, per tonne of dry solids unless stated, and require confirmation for any specific project. The cost comparison below groups the recurring and capital items side by side:

Basis notes: energy figures assume plant-scale measurement of pump and compressor draw per tonne of dry solids deposited; cloth life assumes neutral-to-mildly-acidic slurries at ambient temperature; capital figures exclude civils, buildings, and slurry handling upstream, and should be re-based on vendor quotations for any real project. Two lines deserve comment. Vacuum energy is governed by air flow, not liquor flow — the pump moves mostly air through the cake and cloth, which is why cracked or thin cakes inflate it so sharply. Press energy is governed by the squeeze: a membrane cycle without blow-down costs measurably less than one with extended air blowing, so cycle design is the main energy lever.

The structural trade is larger cheap area against smaller expensive area. When the 3.4:1 area ratio from the previous section is multiplied through these unit costs, the installed-capital gap narrows sharply on easy duties and disappears or inverts on hard ones — which is precisely why the leaf-test data, not the unit costs, must come first. Energy, cloth, and labor rarely overturn an area-based capital comparison; they shift it by tens of percent, while a wrong technology choice shifts it by multiples.

What Goes Wrong With Each, and What Does It Cost?

Seven recurring failures cover most of the pain plant engineers report. They mix both technologies deliberately, because plants rarely operate one in isolation and the diagnostic habits transfer. Symptom, cause, and action follow for each, ordered roughly by how often field surveys encounter them.

1. Cake cracking and air breakthrough (vacuum). Moisture rises and vacuum pump load spikes. Cause: a shrinking fine cake cracks, and air short-circuits through the cracks instead of displacing liquor. Action: reduce air-rate setpoint once formation completes, add a wash liquor blanket over the crack, or switch to a shallower, faster-turning cycle.
2. Cloth blinding (both). Formation time lengthens cycle by cycle. Cause: fine solids and precipitates lodging in the weave. Action: schedule chemical wash-in-place, review cloth weave against particle size, and consult the filter cloth blinding guide before changing equipment.
3. Filtrate solids carryover (both). Downstream clarity drops. Cause: cloth damage, seam failure, or precoat loss. Action: inspect and patch cloths on a scheduled basis; verify filtrate turbidity per section rather than per shift.
4. Plate and diaphragm wear (pressure). Squeeze pressure decays or filtrate appears in the drain. Cause: diaphragm fatigue, abrasive slurry, off-design squeeze pressure. Action: enforce squeeze-pressure limits, replace diaphragms on cycle count rather than failure, and check feed distribution.
5. Uneven cake washing (vacuum). Wash ratio climbs without improving solubles removal. Cause: maldistributed wash water and a non-uniform cake. Action: repipe wash headers for even distribution, level the belt, and confirm cake thickness profile along the width.
6. Vacuum loss from seals and barometric leg (vacuum). Gauge vacuum is fine at the pump but weak at the pan. Cause: seal wear, leaking drain-box gaskets, or an undersized barometric leg losing prime. Action: pressure-test the system cold, reseal drain boxes, and verify leg height against design.
7. Batch cycle time creep (pressure). Cycle grows from 45 to 90 minutes over months. Cause: progressive cloth blinding, feed pump wear, or slurry density drift. Action: trend cycle phases separately — fill, form, squeeze, blow, discharge — and fix the phase that moved, not the whole cycle.

Each of these costs real money: a cracked-cake vacuum duty can double energy per tonne, and a creeping press cycle silently halves plant throughput before anyone files a report. Note also how several causes — blinding, feed density drift, distribution problems — appear under both technologies; the slurry, not the machine, is usually the root cause, and curing it upstream (classification, reagent control, feed density) is cheaper than any mechanical fix.

Field Data: Two Dewatering Duties Compared

Two anonymized duties from recent project work illustrate how the numbers land (all figures indicative, basis: pilot campaigns on representative slurries, to be confirmed at scale).

Duty A, a copper flotation concentrate with P80 around 120 µm and 55% solids feed: vacuum belt testing produced a 22 mm cake at 16% moisture and 0.45 t/m2/h of dry solids, with wash efficiency near 92% soluble removal at 1.0 m3/t wash ratio. The matching press test reached 11% moisture but could not match the continuous throughput without a threefold installed-area comparison turning against it. Vacuum won on total cost per tonne.

Duty B, a neutralized iron-hydroxide sludge with P80 below 30 µm: vacuum testing formed a 6 mm cake that cracked before discharge, holding 34% moisture (wet basis) with pump air consumption near 1.8 Nm3/m2/min — economically hopeless. Membrane pressing at 1.5 MPa squeeze gave 21% moisture at 0.18 t/m2/h. Pressure filtration won by default; no vacuum configuration came within 10 percentage points of the moisture target.

The transferable lesson is that both duties behaved exactly as their specific cake resistance predicted, and the leaf tests that cost days predicted outcomes that would otherwise have cost capital. A third, smaller duty confirms the pattern: a gypsum sludge from a neutralization circuit, tested at the same campaign, showed 0.08 t/m2/h on vacuum against 0.25 t/m2/h on the press at 23% moisture (indicative) — a 3.1:1 area ratio that, priced against the capital figures in the previous section, put the two options within 15% of each other on installed cost before energy was counted. Three numbers to carry away: 3.4:1 area ratio (duty A-type arithmetic), 34% versus 21% moisture on the same fine sludge, and 92% wash solubles removal from a single vacuum displacement wash.

Frequently Asked Questions

What is the fundamental difference between vacuum and pressure filtration?

Vacuum filtration pulls liquor through the medium by creating sub-atmospheric pressure below it, so the differential cannot exceed about 80 kPa (0.08 MPa). Pressure filtration pushes liquor through from above at 0.4-2.5 MPa. The larger differential lets pressure equipment handle finer, more compressible cakes and produce lower cake moisture.

Why can’t vacuum filtration handle fine tailings effectively?

Fine particles create high specific cake resistance, often above 10^12 m/kg for tailings and sludges (indicative). At only 80 kPa of differential, liquor flow through such a cake becomes negligible, the cake forms too slowly or cracks on dewatering, and moisture stays above transport limits. Pressure filtration’s 5-30 times stronger driving force overcomes that resistance.

How much drier is a pressure-filtered cake than a vacuum-filtered one?

On comparable slurries, membrane pressure filtration typically delivers 4-8 percentage points lower cake moisture than vacuum (indicative): 10-14% versus 14-20% on coarse concentrates, and 16-24% versus 25-32% on fine slurries. The exact gap depends on compressibility and squeeze pressure and must be confirmed by test work on your own slurry.

Is vacuum filtration always cheaper to operate?

No. On coarse, free-draining duties, vacuum wins clearly: lower energy, minimal labor, continuous operation. On fine or compressible slurries, vacuum’s weak driving force forces enormous filter areas, high air consumption through cracked cakes, and reprocessing, which can push its cost per tonne above a well-run membrane press. Run the leaf tests first.

What leaf test pressures should I use to compare the two technologies?

Run leaf tests at 100, 300, and 600 kPa to bracket pressure filtration, and 20-80 kPa to simulate vacuum conditions (typical practice). From the three pressure points, fit cake resistance and compressibility: weak response to pressure means a compressible cake that only a membrane squeeze will dry, while strong response suggests either technology can work.

How do I convert a leaf test result into full-scale filter area?

Divide the required dry-solids throughput by the leaf-test specific throughput, then apply a scale-up safety factor — typically 0.7-0.8 on leaf results (indicative) — to allow for cloth aging, feed variation, and downtime. For a 24 t/h duty testing at 0.22 t/m2/h, that gives roughly 110/0.75 ≈ 150 m2 with the factor applied, before round-numbering to a standard machine size.

When is the extra capital of a membrane filter press justified?

When any of three conditions holds: the moisture specification is below roughly 15% and drying is expensive; the slurry is fine or compressible enough that vacuum cannot form a dischargeable cake; or wash solubles recovery justifies squeeze-assisted washing. Otherwise, the cheaper continuous vacuum area usually wins on total cost per tonne of dry solids.

Conclusion

The vacuum filtration vs pressure filtration decision is settled by two numbers from a few days of leaf testing: specific cake resistance and the moisture specification. Below roughly 10^11 m/kg of cake resistance with a moisture target above 14%, continuous vacuum filtration is the cheaper machine in nearly every arrangement; above that resistance, or below that moisture, membrane pressure filtration is not a luxury but the only option that works. Do not size either technology from vendor tables — run the 100/300/600 kPa series on your own slurry, compute the area ratio between the two duties, and price both flowsheets on total cost per tonne. If you want that program scoped and priced, request a quote for a pilot filtration campaign, or browse the full resources library for the underlying methods.

References

• Rushton, A., Ward, A.S. and Holdich, R.G., Solid-Liquid Filtration and Separation Technology, 2nd ed., Wiley-VCH.
• Wakeman, R.J. and Tarleton, E.S., Solid/Liquid Separation: Principles of Industrial Equipment Design and Scale-Up, Elsevier.
• Svarovsky, L., Solid-Liquid Separation, 4th ed., Butterworth-Heinemann.
• Purchas, D.B. and Sutherland, K., Handbook of Filter Media, 2nd ed., Elsevier.
• SME Mineral Processing and Extractive Metallurgy Handbook, Society for Mining, Metallurgy and Exploration, Englewood, CO.
• Wills, B.A. and Finch, J.A., Wills’ Mineral Processing Technology, 8th ed., Butterworth-Heinemann.
• Tarleton, E.S. and Wakeman, R.J., Solid/Liquid Separation: Equipment Selection and Process Design, Butterworth-Heinemann.

Changelog

• 2026-11-03: First publication.