Air Knife Drying Applications

Remove the water. Protect the surface. Match the air system to the production line.

An air-knife drying system should be selected from the moisture condition required at the next process, not from knife length, blower horsepower or a high outlet-velocity claim alone. Establish the incoming water load, product geometry, line speed, exposure time and permitted surface condition. Then compare compressed-air and blower-fed arrangements at the installed operating point, including air preparation, ducting, drainage, control and total energy.

For accessible free water, the useful action is often physical displacement: moving droplets and films off the product into a controlled collection path. Where the requirement also includes evaporation, removal from recesses, absorbed moisture or prevention of later condensation, those duties need separate treatment. A surface that looks dry beside the knife may still be unsuitable for labelling, coating, packaging or assembly downstream.

Required page asset: Drying application comparison matrix, with editable application, utility and production-result records.

Page boundary: Surface-moisture removal and the comparison of its air-supply architectures. This article does not replace the selected equipment instructions or a product-specific process qualification.

1. Start with the downstream requirement, not with an air knife

The strongest enquiry begins with a production problem: labels lift after washing, a code becomes unreadable, liquid carries into packaging, droplets remain in a recess, or a rinsed component enters assembly wetter than permitted. Each describes a different drying duty. Asking only for a one-metre air knife leaves the supplier to guess what successful operation means.

Identify the exact surface that matters and the next operation it must support. The outer wall of a sealed container, its closure seam, the underside of a tray and a prepared coating surface can require different coverage and cleanliness. Drying the easiest face does not establish that the critical face is ready. A drawing with the relevant zones marked is often more useful than another nominal airflow estimate.

Define the acceptance location and time. An observation immediately after the jet can differ from the condition after a turn, a conveyor transfer or a wait before labelling. Water hidden beneath a rim may emerge later. A cold container can acquire new condensate after the initial rinse water has been removed. The air system must therefore be assessed against the required condition where the process uses it.

Retain the production envelope: product sizes, materials, speeds, spacing, lane count, wettest credible incoming condition and permissible downtime. A setup that works on one product at reduced speed is not automatically a production solution. The output of this first step is a drying requirement that competing suppliers can interpret consistently. [S1; N1]

2. Define what “dry” means for this application

“Completely dry” is not a usable specification unless the assessment method and boundary are stated. A process may require no visible droplets on a code zone, a bounded residual-water mass, no liquid released during a defined tilt, acceptable adhesive performance or another documented outcome. These are related but not interchangeable measures.

Separate total residual moisture from local wetness. Ten grams distributed across a large sheet and ten grams concentrated inside a critical pocket can have very different consequences. Similarly, a mean result across several containers can hide a small number of unacceptable seams. Record both the useful overall indicator and the local condition that can cause a failure.

The criterion should be achievable and relevant. A requirement to evaporate every last trace of water can introduce unnecessary heat and residence time when the real need is a clean, droplet-free external code area. Conversely, a visual-only criterion can be inadequate where a residual film affects adhesion or corrosion. Use the process requirement to decide the method, rather than selecting a convenient measurement and assuming it proves everything.

Record a borderline or unavailable result honestly. “Not assessed at the required location” is different from “dryness achieved.” Avoid converting a missing measurement into a pass because the equipment runs without an alarm. [N1]

3. Separate displacement, evaporation and removal of absorbed moisture

An air knife can strip accessible water from a surface through the action of the moving air. In that mode, the water leaves predominantly as liquid droplets, a film or collected drainage. The process is not required to supply the latent heat needed to turn all that water into vapour. Paxton describes its blower-fed knives as shearing water from washed products; that mechanism is the limited manufacturer foundation used here. [R2]

Evaporation is a different duty. It transfers water into the air as vapour and requires both a heat supply and a route for carrying the vapour away. Heat can come from the product, surrounding air, a heater or another process stream. If evaporation cools a sensitive product or increases enclosure humidity, those consequences belong in the design.

Absorbed moisture or water retained in pores can be more difficult again. Removing surface droplets does not show that moisture within a porous material has reached a required final level. Product thickness, internal transport, temperature and allowable residence time may govern a separate drying process. The name air knife should not conceal that difference.

A useful sequence can be drainage first, mechanical or pneumatic displacement second, and a smaller evaporation stage only for the remainder. That arrangement must be developed for the actual material and handling constraints. A wiper can damage a delicate finish; tilting can spill a product; heat can affect seals or labels. No stage becomes appropriate solely because it reduces compressed-air demand.

Keep water balances separate from energy balances. A drain collection measurement supports liquid removal. A humidity change supports a vapour-transfer assessment only with the relevant airflow and state information. Neither alone proves that the required product zones are dry. [S1; N1]

4. Quantify the incoming water load at the correct boundary

The incoming condition is the water presented to the drying section, after any upstream drain zone. Do not use the rinse-pump delivery as the drying load: most of the rinse may already have returned to a tank. Conversely, a nominally small rinse flow can leave troublesome pockets if the product traps water.

For a moving flat surface with an explicitly assumed uniform film, a preliminary inventory is:

Water mass rate = liquid density × wetted width × conveyor speed × film thickness.

Use kg/m³, metres, metres per second and metres to obtain kg/s. Convert micrometres to metres before multiplication. The expression describes the assumed film entering the boundary. It does not predict how much of it a particular knife can remove. Ribs, rivulets, droplets, perforations and changing orientation can invalidate the uniform-film model.

For discrete products, use water mass per item multiplied by the actual item rate. Establish whether that mass includes the underside, fixtures and carriers. A wet carrier can re-wet a product even when the product itself has been treated. Where water loading varies widely, retain the distribution and worst credible product family instead of using only a convenient mean.

Document how the loading was obtained: measured, calculated from a declared model, estimated or provided by the process owner. A transparent preliminary estimate helps define a supplier enquiry; it should not be relabelled a measured plant duty. Record product temperature and rinse chemistry where they affect wetting or residue.

The inventory can expose a poor starting point. If a large fraction of the water can drain during existing transfer time, the knife should be sized for the remaining task rather than compensating for a missing drain path. If a process change increases carryover, the drying requirement must change with it. [N1, N2]

5. Understand why the same jet dries one surface and not another

A smooth exposed surface, a rough casting and a narrow groove do not present the same removal problem. Droplets and films interact with surface texture, wetting, geometry and the direction of gravity. Detergent, oil or a change in coating can alter that interaction even when the water quantity appears unchanged.

The engineering implication is practical: record the actual product condition used during the trial. A polished sample that drains readily may not represent the production finish. A clean-water demonstration may not represent a process liquid containing surfactant or a persistent oily residue. Those differences should be visible before a favourable demonstration becomes a purchase basis.

Momentum directed towards the surface can disturb water, but useful removal also needs a path away from the surface. Excessively aggressive impingement can disperse fine droplets without delivering them to collection. A different angle or an upstream drainage feature may solve the task more effectively than increasing the jet pressure.

Avoid inventing a universal velocity-to-water-removal equation. The interaction depends on more than an outlet velocity, and a catalogue point may be measured at a different distance or in unrestricted air. Use the candidate supplier's applicable guidance, then judge the actual product result. The arithmetic in this article screens inventories, geometry, operating points and consumption; it is not a universal surface-drying model.

When troubleshooting, ask what changed in the surface and liquid before assuming the blower has lost capacity. A different washer chemistry, a worn coating or altered product temperature can change drying while the air equipment remains mechanically unchanged. [N1]

6. Map product geometry, movement and stability

Divide the product into useful drying zones: broad faces, leading and trailing edges, necks, closures, pockets, undersides and conveyor-contact areas. Mark which are directly exposed, which are shadowed and which can release retained water after the air treatment. This zone map determines whether one knife, several knives or a combination of knives and targeted nozzles deserves evaluation.

A long knife is not automatically a full-product solution. A support rail can obstruct its jet. A container shoulder can shield the neck. A tray can carry water underneath a flange. An underside can remain wet where it rests on a belt. Additional pressure cannot create direct access through a solid obstruction.

Include product movement through the full section. A turn, rotation or transfer may expose a hidden face, but it can also release water onto an already dried area. The order of the air stations should follow the intended water path. There is no universal top-first or side-first sequence that applies to every product and conveyor arrangement.

Product stability is another limit. Light empty containers, flexible films and small components can move, flutter or tip under an uneven air force. Supporting or guiding the product can improve repeatability, but the support itself may create a new shadow or water trap. Review that interaction with the conveyor designer rather than treating guidance as separate from drying.

Record the permitted product-position tolerance. A successful setup that relies on an exact hand-held position is difficult to reproduce on a moving line. Adjustable mounts should have a repeatable datum, a clear recipe position and clearance for the largest permitted product. [S1; N1]

7. Retain line speed, pitch and useful exposure time

For an established useful treatment zone of length L along the travel direction, and constant product speed v, exposure time is L/v. The useful length is the region in which the process receives adequate action, not simply the physical width of the knife body or the region in which a person can feel airflow.

For discrete products, centre-to-centre pitch and conveyor speed determine the item rate when the conveyor moves continuously without slip. Using v in m/s and pitch p in metres, items per minute are 60v/p per occupied lane. Missing products, accumulation, variable spacing and multiple lanes require their actual occupancy record.

Distinguish treatment time for one surface point from the complete occupied window of one product. A surface point traverses the useful zone. The knife may need to remain effective from the leading edge entering until the trailing edge leaves. Using only the surface-point exposure as the valve-open time can leave much of a long product untreated.

A faster line can reduce treatment time, increase incoming water per minute and reduce gaps between treatment windows simultaneously. These changes are not interchangeable. The supplier should see the maximum production case as well as normal operation, and the control design should account for speed changes and restart behaviour. [S1; N1, N2]

8. Select coverage, angle and stand-off as one arrangement

Define knife length along its actual active slot. When a straight knife is skewed in plan, its cross-conveyor projection can be shorter than its physical length. For a simplified planar arrangement with angle alpha measured from the cross-conveyor direction, the projected span is L cos(alpha). This geometry does not establish the effective drying footprint; edge effects and product height still matter.

Stand-off should be measured between named datums: for example, the slot lip and a specified product surface. An unlabelled distance can become meaningless when two suppliers use different knife geometries. Include the tallest product, lateral wander, machine vibration and clearance needed to avoid contact.

Angle must also have a clear definition. Plan-view skew, inclination towards the surface and downward direction along a container are different adjustments. Record them separately. A drawing or dimensional setup record is preferable to a note saying “angle approximately right.” Do not transplant a numerical angle from an unrelated installation.

Aim to move water towards a deliberate exit without sending it back into an incoming wet surface or an adjacent finished area. Opposed jets can interfere, and a top jet can put water onto a previously dried side. The correct arrangement depends on the product and travel sequence. Where a supplier recommends staggering or a particular orientation, preserve that recommendation with the model and application conditions.

Vortron's manufacturer guidance identifies the joint importance of gap, knife length, quantity, distance and matching blower conditions.

The accepted setup should be reproducible after changeover. Mark mounting positions and record which adjustments are permitted for each product family. A later wet-product complaint may be a positioning problem rather than a utility-capacity problem. [S1; N1]

9. Treat slot uniformity and the supply plenum as performance features

A useful knife needs the intended distribution along its active length. A locally obstructed slot, damaged lip, unsupported adjustment or unequal feed can create a wet stripe even when the common inlet pressure appears satisfactory. Increasing common pressure may improve the stripe only by over-treating the rest of the product.

Specify the exact knife configuration: active length, slot arrangement, feed connections, permitted orientation, materials and any manufacturer-defined adjustment. Do not treat every knife with the same length as having the same flow characteristic. Paxton's published product information, for example, states flow against pressure per unit knife length for a defined product configuration. Such a record illustrates why configuration matters; it is not a universal air-knife sizing table. [R5]

Changing a slot gap changes the relationship between supply pressure and airflow. It can also move a blower to a different operating point. A larger opening therefore does not guarantee a stronger effective jet at the product. The combined blower, duct and knife system must be considered.

Inspect and service the lip only under the applicable safe isolation and maintenance instructions. Uncontrolled scraping, drilling or shim changes can damage a precision feature or change its consumption. Maintain the intended cleanliness and avoid introducing particles into an air stream serving a sensitive surface.

For long or multi-feed arrangements, ask the supplier how uniformity is maintained and what evidence supports it at the installed condition. Record any limits on feed arrangement, restrictions and active length.

10. Size compressed-air supply from motive consumption

A compressed-air knife uses supplied compressed air to create the working stream. Some designs also entrain substantial surrounding air. The air drawn from the room is not additional compressor delivery. Keep compressed-air input, entrained flow and total mixed stream separate; EXAIR describes surrounding-air entrainment for its Full-Flow Air Knife. [R3]

For identical knives, a constant-flow consumption screen is number of knives multiplied by the applicable reference flow per knife multiplied by actual air-on time. Use the manufacturer's flow at the selected inlet condition, with the units and reference state retained. A length-normalised figure may be multiplied by active length only when the identified product information supports that use.

A compressed-air requirement must include the installed flow path. Consider the common filter, regulator, valve, manifold and individual feeds under simultaneous operation. A pressure reading at a remote regulator does not establish knife-plenum pressure while several banks are on. Correct restrictions before raising the plant setting to compensate for them.

For pulsed service, retain establishment, steady action and shutoff tail. A long pipe downstream of the valve stores gas that may discharge after the command ends and require replenishment during the next event. If the complete upstream measurement already includes that volume, do not add it again.

The desired operating pressure is the lowest supported condition that achieves the required drying throughout the operating envelope, not an arbitrary minimum or the device's maximum rating. Quality, temperature, product stability and response remain part of that decision. A reduction in knife consumption is not automatically the same percentage reduction in compressor-station power.

11. Select a blower at the installed duty point

A blower-fed knife obtains its air from a dedicated low-pressure supply system. The duty point must represent the knife bank, ducting, filters, silencers and inlet conditions together. A blower's maximum free-delivery volume and its maximum pressure usually describe different conditions; combining them creates a fictitious capability.

Ask for the relevant pressure-flow curve, speed, air state and power information. Identify whether pressure is a static rise or total-pressure rise and where it is measured. A catalogue flow can refer to inlet actual volume, an outlet condition or another stated basis. Preserve the definition before comparing it with a knife requirement.

At a common junction, parallel knives share the pressure available at that junction, while their flows add. Along one branch, the pressure requirement includes the knife inlet requirement plus the applicable branch losses. Adding knife pressures from parallel branches is incorrect. Ignoring the shared trunk loss is equally incorrect.

A preliminary pressure budget can identify a shortfall at a nominated flow. It does not solve the final operating point when the flow itself changes. If a blower provides less pressure than the network requires at that flow, the system must be solved using the actual curves; it cannot be declared adequate merely because the motor has spare nameplate power.

Filter loading, intake temperature, duct changes and altered slot settings can move the operating point. Assess the service condition as well as a new clean installation. Retain supplier limits for speed, motor input, temperature and stable operation. Do not create an improvised high-resistance operating condition by blocking outlets to obtain a more favourable reading.

Spraying Systems explicitly distinguishes blower packages from compressed-air knives in its product range. That distinction supports keeping their supply designs separate, rather than treating a blower as a drop-in source for every compressed-air knife. [R1; N1]

12. Compare the two supply architectures against the same drying task

A broad continuously occupied line may favour evaluation of a blower-fed arrangement. A compact intermittent duty may favour evaluation of a controlled compressed-air arrangement. These are starting points for comparison, not a rule that either architecture always has lower operating cost.

Paxton's comparison identifies use frequency and conveyor conditions as selection factors and distinguishes blower operation from frequently switched compressed-air service. The exact starting limits and operating costs remain product- and installation-specific; no universal starts-per-hour rule is imposed here. [R2]

A hybrid can also deserve assessment: gravity drainage for bulk liquid, a blower for broad surfaces and a small controlled jet for a difficult recess. The combined system must be counted as a whole. Do not compare the compressed-air part of one candidate against the complete electrical consumption of another.

Evaluate only candidates that meet mandatory product, operational and safety conditions. A low energy number cannot compensate for a wet code zone, damaged packaging or uncontrolled mist. Once the task is supported, compare capital, utilities, maintenance and operating flexibility on consistent assumptions. [S1; N1]

13. Define air quality where the stream reaches the product

Air quality depends on the entire path. In a compressed-air arrangement, the final hose, local treatment, materials and any intentional lubrication can change the delivered condition. In an entraining knife, surrounding air mixes with motive air. In a blower arrangement, the intake location and subsequent air path influence what reaches the surface.

Classify the contact boundary. External container drying is not the same as air directed into an open food package or onto a critical coating surface. The required evidence should come from the process and equipment requirements. Do not invent one universal cleanliness class or assume that all external drying is noncritical.

Separate particles, liquid water, water vapour, oil aerosol, oil vapour and biological concerns where relevant. A filter label does not address every contaminant. An oil-free compressor claim does not establish the complete point-of-use condition, and a stainless-steel knife does not make the mixed air sterile.

Intake placement is particularly important when a blower is located close to the wet process. It may draw mist, cleaning chemicals or hot humid exhaust back into the working stream. Moving the intake can improve conditions, but the resulting duct and filter losses belong in the blower selection.

Record cleaning and maintenance materials as part of the quality boundary. A replacement hose, a lubricant used during assembly or an inadequately cleaned enclosure can affect the result. The correct response is a controlled specification and appropriate evidence, not simply adding a finer filter without considering pressure loss or the actual contamination mechanism. [S1; N1]

14. Keep temperature, humidity and condensation distinct

A warm air stream can support evaporation, but increased temperature alone does not establish a drying result. The water has to receive the necessary heat, and the resulting vapour has to leave the product vicinity. A warm saturated recirculating enclosure may perform very differently from a warm stream supplied with suitable ventilation.

For a simple evaporation balance on a dry-air basis, water carried away per unit time is dry-air mass flow multiplied by the increase in humidity ratio. Both humidity ratios must use the same definition, and the air state and measurement boundary must be known. Using relative humidity as though it were a water mass fraction gives an incorrect result.

Pressure dew point in a compressed-air line and atmospheric dew point around a product are different conditions. A low line pressure-dew-point number is useful for the supply assessment but does not, by itself, prove that a cold product will remain free of condensation after the jet mixes with room air.

Condensation can occur when the surface is below the dew point of the air actually contacting it. The relevant condition may change after the product leaves the enclosure. Removing rinse droplets and preventing new atmospheric condensation are therefore separate requirements. A chilled container that re-wets downstream may need changes in timing, environment or product conditioning rather than a more powerful knife.

Also protect upper temperature limits. Labels, films, seals, coatings and the product itself can be temperature-sensitive. Blower discharge heat should not be assumed beneficial. Specify the permitted delivered-air and surface-temperature envelope and retain the selected package's actual information.

15. Give the removed liquid a controlled destination

Water displaced from a product remains inside the process boundary until it is captured, drained or exhausted appropriately. A clean-looking surface can be surrounded by mist that later deposits on the same product, nearby machines or the floor. The design should therefore show the liquid path as clearly as the air path.

Consider drain-pan slope, splash containment, liquid compatibility, accessible cleaning and the destination of collected fluid. Process wastewater may not be suitable for uncontrolled disposal. Avoid creating a stagnant area beneath a nominally effective drying station or returning contaminated water to a process that requires separation.

Enclosure ventilation has to accommodate supplied air, entrained room air, displaced droplets and any evaporated moisture. A very small opening can raise enclosure pressure or redirect flow; a very open arrangement can fail to contain mist or noise. The extraction designer should establish the appropriate balance for the actual substances and geometry.

A filter designed for dry dust is not automatically an appropriate liquid-mist separator. Likewise, a general-purpose vacuum collector should not be assumed suitable for process liquids, solvents or reactive materials. Match the removal and collection mechanism to the material and cleaning requirements.

Inspect for secondary wetting sources: dripping upper surfaces, wet guide rails, a saturated conveyor, a downstream spray or contact with wet packaging. The knife can be working correctly while the overall product path remains defective. A drying trial should follow the product through the next relevant transfer rather than ending at the most favourable viewing position. [S1; N1]

16. Build a drying sequence that survives real production

For a compressed-air bank, define the permission to run, product detection, useful-air establishment and required end of treatment. Retain the distinction between an electrical command and air that has reached an effective condition at the product. A short valve response specification does not include every delay in a long manifold.

For a blower-fed arrangement, use the selected manufacturer's supported running and control modes. Repeatedly starting a blower for every product can be inappropriate. Speed control, staged operation, an approved standby mode or continuous running may be suitable, but their limits and energy must be evaluated. Do not close outlets or add a bypass without the supplier's operating guidance.

The sequence must handle empty gaps, conveyor stops, restart, jam recovery, missing products and recipe changes. During a stop, continued air may overheat, move or mark one product, while immediate loss of drying can leave incoming material unsuitable. The required response depends on the product and machine; neither continuous blowing nor immediate shutdown is universally correct.

Define the disposition of products that passed during an unready condition. A blower-running signal may not prove the required knife pressure. A valid pressure signal may not prove all surfaces were exposed. Production needs a practical means to identify interrupted treatment and decide whether affected items require further assessment.

Keep ordinary product control separate from protective functions. Guarding, isolation, emergency response and safe maintenance must come from the responsible machine design. A product sensor or a generic pressure switch should not be described as a validated safety system. [S1; N1]

17. Make changeover and maintenance repeatable

A drying arrangement can lose performance through small changes: a shifted bracket, a replaced flexible duct with a tighter bend, a new filter type, a worn slot lip or a different product guide. Record the configuration needed to reproduce the accepted result rather than leaving it in the memory of one operator.

Each product recipe should identify active stations, permitted speed, position datums, required utility conditions and any special handling. Avoid undocumented adjustment of a common regulator to correct one product when another product shares the same outlet. Changes to knife pressure or blower speed should remain within the selected equipment and process limits.

For maintenance access, consider isolation of every relevant energy source and retained pressure. A disconnected electrical command does not necessarily remove stored air. A stationary conveyor does not establish that surrounding mechanisms or hot parts are safe to approach. Follow the actual equipment procedure; this article is not an instruction to service a live knife or blower.

Condition-based observations can include filter differential at a defined flow, knife pressure, drive condition, unusual sound, discharge temperature, slot cleanliness, bracket position and product wetness. Use the manufacturer's maintenance requirements and the site's operating evidence. No universal replacement interval is imposed.

After a material change, repeat the affected production observation. Replacing one knife does not automatically require redoing every unrelated study, but it does require enough evidence to show that the relevant product zones and operating cases remain supported. Keep the as-found and as-left configuration identifiable. [N1]

18. Establish drying performance through useful product observations

A suitable trial links incoming condition, equipment setup, utilities and downstream result. Begin with a defined product and moisture requirement, then record the circumstances in which that requirement is assessed. Changing line speed, product temperature and knife arrangement at once can make a favourable result difficult to interpret.

Use a method appropriate to the surface and product. Weighing can support a residual-mass estimate when the product itself is stable and the timing is controlled. Collection or blotting methods may be more appropriate for particular accessible zones, but their recovery limitations must be understood. Visual observation can support a visual requirement without pretending to quantify invisible films.

Preserve individual results where local failures matter. An average can remain acceptable while one closure or seam is wet. Where a maximum per-item criterion exists, report how many assessed items met it as well as the mean. The sample plan should represent the actual operating cases; this article does not prescribe a universal sample size or acceptance probability.

Compare first products after startup, steady running, largest and smallest products, maximum line speed and the relevant service condition. Include a downstream observation after any movement that can release trapped liquid or expose the product to condensation. A record of every second at one easy duty is not the same as coverage of all relevant duties.

A supplier demonstration, a production trial and formal product release have different responsibilities. Use the observations to establish the drying capability, then leave release decisions with the designated process owner. A written article or a worked example does not approve a specific installation. [N1]

19. Diagnose the pattern of wetness before changing the source

The wetness pattern often narrows the investigation. A persistent stripe points towards coverage or uniformity; a wet underside points towards access or conveyor contact; a problem confined to restart points towards readiness or timing. These are hypotheses to investigate, not automatic diagnoses.

Do not infer inadequate compressor capacity from one wet product. Similarly, do not infer a knife defect from one pressure gauge. Align the observations in time and preserve the installed state. Repeatedly increasing pressure without identifying the limiting mechanism can add air use, noise and mist while leaving the critical surface unchanged.

When production cannot be interrupted, record the evidence available and the unresolved questions. A measured pressure trend can help distinguish common supply loss from a local restriction without claiming to establish every surface mechanism. Plan the smallest practical next investigation that can resolve the specific uncertainty. [S1; N1]

20. Compare complete energy and operating cost

Compressed-air consumption and blower electrical input are unlike quantities. Compare them through a declared energy boundary while preserving each original measurement. For compressed air, the important economic question is how station electricity changes when the drying demand changes. For a blower, include actual electrical input during production, standby and other operating states.

An allocated compressed-air energy intensity can support an explicitly labelled estimate. With specific power s in kW per reference m³/min and air volume V in reference m³, allocated energy is sV/60 in kWh. The division by 60 is essential. This calculation does not prove the marginal station saving: unload behaviour, sequencing and other users can change the result.

The total comparison should include extraction, heaters, auxiliary nozzles and changed cooling or ventilation where they are inside the decision boundary. Include maintenance and consumables separately. A blower-only motor figure should not be compared with a compressed-air system including every auxiliary, or vice versa.

Define the production denominator. Energy per accepted product can reveal a different outcome from energy per attempted pass. Re-drying, rejects and reduced throughput can erase an apparent utility saving. A faster line can use more total energy while using less per accepted item; both statements should remain visible.

A simple payback estimate is eligible installed cost divided by positive annual net saving, using compatible periods and stated assumptions. It is not meaningful when the net saving is zero or negative, and it is not a substitute for a project-specific investment analysis. Keep tariffs and quoted costs dated when used for an actual customer; the worked numbers below are fictional teaching inputs.

The final choice should preserve product quality, operating rate and maintainability. A technology label alone cannot establish the preferred investment. [S1; N1, N2]

21. Worked example 1 — Reduce the water load before adding thermal duty

Fictional teaching case. A flat conveyor carries a wetted width of 0.80 m at 0.30 m/s. Assume one assessed face has a uniform incoming water film 75 micrometres thick and liquid density of 1,000 kg/m³. The model excludes edge pools, underside water and carrier moisture.

The assessed area rate is 0.80 × 0.30 = 0.24 m²/s, or 864 m²/h. The equivalent film loading is 1,000 × 75 × 10^-6 = 0.075 kg/m². Incoming water is therefore 864 × 0.075 = 64.80 kg/h.

Suppose an approved upstream drain arrangement removes 70% of that incoming water. The load reaching the air section becomes 64.80 × 0.30 = 19.44 kg/h. This does not mean the knife itself has become 70% more efficient; the upstream duty has changed.

Assume the air section leaves an equivalent residual loading of 0.00050 kg/m² on the assessed face. Water leaving on that face is 864 × 0.00050 = 0.432 kg/h. The air section must remove 19.44 − 0.432 = 19.008 kg/h, corresponding to approximately 97.78% of its incoming water by this mass balance.

The last row assumes, solely for this illustration, a latent heat of 2,400 kJ/kg and that all 19.008 kg/h is evaporated. It excludes sensible heating, losses and equipment efficiency. If the air knife instead displaces most of that water into a drain, the full latent-duty calculation does not describe the knife's removal mechanism and must not be used as its electrical-power requirement.

A separate evaporation screen illustrates another boundary. At an assumed dry-air flow of 1,200 kg/h and an inlet-to-outlet humidity-ratio increase of 0.002 kg water/kg dry air, the stream carries away 2.40 kg/h as added vapour. That is far below 19.008 kg/h. Under those assumptions, a claim that the entire removal occurred by evaporation is inconsistent; collected liquid, mist, other air streams or a mistaken input would need investigation.

The design conclusion is to establish the actual incoming load and removal mechanism. The calculation does not select a knife, certify the residual-film requirement or establish that 70% drainage is available on a real line. [N2]

22. Worked example 2 — Knife length and sensor timing can both miss the product

Fictional teaching case. A product stream needs coverage across 0.90 m. A candidate knife has a 1.00 m active slot and is skewed 30 degrees in plan from the cross-conveyor direction. Treat the slot projection as a geometric screen only.

Cross-conveyor projection is 1.00 × cos(30 degrees) = 0.8660 m. The nominal shortfall is about 34.0 mm before accounting for edge performance or product wander. The geometric slot length required to project 0.90 m at the same angle is 0.90/cos(30 degrees) = 1.0392 m. That is not a final recommended knife length: the effective drying contour and installation clearance remain necessary.

Now consider product travel at 0.50 m/s through an established useful treatment zone 0.060 m long. Each surface point has 0.060/0.50 = 0.120 s of exposure. If the process requires an illustrative 0.180 s under that air condition, the zone would need to be 0.50 × 0.180 = 0.090 m long, or another justified change would be required.

For a product 0.120 m long, the complete occupied interval is (0.120 + 0.060)/0.50 = 0.360 s. With a 0.300 m pitch, products arrive every 0.600 s, equivalent to 100 items/min per occupied lane. Ideal product-occupied duty is therefore 0.360/0.600 = 60%, not the 20% obtained by dividing surface-point exposure by the arrival interval.

Suppose a sensor detects the leading edge 0.250 m before the leading boundary of the useful zone. Transit time is 0.500 s. Useful air takes 0.080 s to establish after the on-command. The ideal on-command is consequently 0.420 s after detection. Issuing it at 0.500 s leaves the leading edge travelling 40 mm before useful air is established.

The ideal electrical command width is 0.400 s, or 66.67% of the 0.600 s arrival interval, while the effective occupied window is 60%. Airflow during establishment and decay is not necessarily constant, so neither duty fraction alone establishes measured consumption.

These values assume constant speed, fixed geometry and repeatable delays. A real controller must handle speed changes, missing items and merged windows for the same bank. The important result is that width coverage, surface exposure, product occupancy and valve timing are four separate questions. A single “one-metre knife at 100 items/min” specification does not resolve them. [N2]

23. Worked example 3 — A blower can meet the clean case and miss the service case

Fictional teaching case. Two identical parallel knives require a combined 720 actual m³/h at a specified knife-plenum pressure 9.0 kPa above the local discharge environment. Each takes 360 actual m³/h at that same stated condition. All target-point curve flows are assumed reconciled to this same knife-flow basis. Both knives are supplied from a common trunk.

For this simplified pressure screen, assume compatible static-pressure boundaries, with connecting-section velocity effects already represented in the stated losses. The common trunk loses 1.8 kPa at the combined flow, each governing branch loses 0.7 kPa at its own flow, and the inlet filter contributes 0.5 kPa on the clean case.

Required blower pressure rise at the nominated flow is 9.0 + 1.8 + 0.7 + 0.5 = 12.0 kPa. The two parallel knife requirements are not added as 18.0 kPa. Their flows add; each governing branch is assessed at the pressure it needs.

Assume Candidate A supplies a rise of 13.0 kPa at the nominated flow and declared speed. Its clean-condition arithmetic margin is 1.0 kPa. If the inlet filter restriction rises to 2.0 kPa while the other stipulated values are evaluated at the same target flow, required rise becomes 13.5 kPa. Candidate A has a 0.5 kPa target-point shortfall.

The service-case calculation does not prove the system will continue to pass 720 m³/h at a lower pressure. Its actual flow and pressure require the coupled blower and system curves. The source may move to a lower-flow duty point, with consequent changes in slot velocity and drying capability.

For a separate geometric screen, assume each knife has a 0.75 m active slot and a 0.0010 m effective open gap, with the entire stated per-knife flow evaluated at the slot-exit condition. Open area is 0.00075 m². Flow of 360 m³/h is 0.100 m³/s, giving mean slot velocity 133.33 m/s by Q/A. If the 360 m³/h were instead quoted at the blower inlet, a density-state conversion would be required before this division.

That mean slot value is not the velocity over every part of the product and does not establish drying. Nor is a simple pressure-to-velocity conversion adequate when density changes, losses and real slot geometry matter. Candidate selection still requires applicable curves, temperature and motor limits, uniformity and the product result. The example's purpose is to prevent free-delivery flow, parallel pressures and differing volumetric-flow bases from being mixed. [N2]

24. Worked example 4 — Duty and station response can reverse the energy preference

Fictional teaching case. Three compressed-air knives each consume 0.55 reference m³/min at the specified condition. Their combined active demand is 1.65 reference m³/min. The production window is 16 h/day, and the knives are effectively active for 35% of it. Assume this flow-time boundary already includes the air being compared and excludes unrelated plant users.

Daily reference volume is 1.65 × 60 × 16 × 0.35 = 554.40 m³/day. At an assumed allocated specific power of 7.2 kW per reference m³/min, assigned energy is 554.40 × 7.2/60 = 66.528 kWh/day. This is an allocation, not a measured marginal station saving.

A candidate blower arrangement is assumed to use 5.8 kW for 10 h/day and 1.1 kW in its supported standby mode for 6 h/day. Its daily input is 64.60 kWh. Extraction consumes 0.8 kW for 16 h in both alternatives, adding 12.80 kWh/day to each. The totals are therefore 79.328 kWh for the allocated compressed-air case and 77.40 kWh for the blower case: a difference of only 1.928 kWh/day.

Now assess a different continuous-duty scenario, with the compressed-air knives active for 90% of the same 16 h window and the blower running at 5.8 kW throughout. Assigned compressed-air energy becomes 171.072 kWh/day, and blower input becomes 92.80 kWh/day. With identical extraction on both sides, the arithmetic difference is 78.272 kWh/day. The larger difference arises from changed duty assumptions, not a universal blower saving percentage.

The last row introduces a separate fictional observation: assume the compressor station actually reduces input by only 42.0 kWh/day when the intermittent knife demand is removed. Adding the blower's 64.6 kWh then increases net electricity by 22.6 kWh/day, before any other change. It would be incorrect to retain the allocated 66.528 kWh as a verified saving after this observation.

For the high-duty allocation scenario, a fictional tariff of ₹9/kWh and 300 equivalent days gives 78.272 × 9 × 300 = ₹211,334.40/year gross arithmetic benefit. With an assumed additional annual maintenance cost of ₹18,000, net benefit is ₹193,334.40/year. At an assumed installed investment of ₹420,000, simple payback is approximately 2.17 years. This remains a scenario estimate, not a quotation or realised project result.

Product output still governs eligibility. Lower utility use achieved by slowing the conveyor, leaving wet seams or increasing re-drying is not an equivalent duty. Report the accepted output and moisture result alongside the utility comparison. [N2]

25. Drying application comparison matrix

Use the matrix to identify candidate routes before selecting equipment. The entries describe engineering questions, not a ranking of products or a claim that every route will be suitable. Preserve the incoming condition and required downstream result for all candidates.

A practical assessment should distinguish eligible, unresolved and unsuitable candidates. An unresolved product-temperature limit is not evidence of suitability, and a higher cost is not evidence of technical failure. State the missing information precisely enough for a supplier or process owner to answer it.

Compare the future production envelope as well as today's easiest product. A slightly more expensive arrangement may offer valuable changeover range, but that benefit should be described rather than assigned an invented monetary value. Keep the proposal tied to what the supplier can actually support. [S1; N1]

26. Editable application and comparison records

The following records are manually editable working tables. Complete them with actual project information; blank cells are not default settings. Preserve the original evidence when assumptions or configurations change.

Record A — Product, moisture and production requirement

Record B — Candidate drying configuration and utilities

Record C — Production observation and decision

Attach relevant manufacturer curves and site observations by their actual document identities. Do not paste private customer records into public website demonstrations. The public page can show blank forms and fictional examples while the enquiry workflow keeps project records access-controlled. [N1]

27. Frequently asked questions

Does an air knife dry mainly by heat or by blowing water away?

For accessible free liquid, an air knife can remove water through physical displacement. Some evaporation may occur, but it should not be assumed to account for the whole removal. Where the specification includes evaporation or internal moisture reduction, establish the heat, air-state and residence-time requirements separately. A collected-liquid route and a vapour-removal route are different parts of the process.

Is a blower always more economical than compressed air?

No. The comparison depends on the accepted drying task, occupied running time, blower standby behaviour and actual compressor-station response. Broad sustained duties often justify evaluation of a blower, while short intermittent duties deserve a careful control and energy comparison. Include extraction and other auxiliaries consistently, and distinguish allocated compressed-air energy from measured avoidable electricity.

Can air-knife length be selected from conveyor width alone?

Conveyor width is only one input. The effective span also depends on active length, skew, product height, position variation and edge performance. The useful treatment zone along travel determines exposure, while recesses and undersides may need a different arrangement. A knife can be longer than the conveyor width and still leave a critical area untreated.

Why does the product become wet again after leaving the knife?

Investigate retained water released during movement, wet guide rails or conveyor contact, overhead dripping, neighbouring jets and condensation. These are distinct mechanisms. A cold surface exposed to humid surroundings can acquire new condensate even after rinse water has been removed. Observe the product at the point where dryness matters, not only beside the air knife.

Is more pressure the right response to poor drying?

Not necessarily. First compare coverage, incoming water, product speed, slot condition, local supply and the liquid exit path. Increasing pressure can increase consumption and mist without reaching a hidden feature or correcting a timing problem. Any changed condition must remain within the equipment limits and preserve product stability and surface requirements.

Does an amplification ratio reduce the required compressor size by that ratio?

No. An entraining knife's total stream can include surrounding air that did not pass through the compressor. Compressor demand is based on motive-air consumption plus other actual compressed-air users. The total mixed stream matters to product interaction, enclosure airflow and air quality, but an amplification number alone does not establish useful drying capability or station electricity.

Can a pressure dew point confirm that the finished product is dry?

No. Pressure dew point describes a stated compressed-air condition. Product dryness depends on its incoming liquid, surface, exposure and surroundings. Air can also mix with ambient air before reaching the product. Preventing later condensation requires the relevant surface temperature and the dew point of the contacting atmosphere, not just the compressor-room treatment record.

Should the blower switch off in every empty product gap?

Use the selected blower's supported control modes and starting limits. Rapid per-item switching may be unsuitable, and useful airflow may not be established when the next item arrives. An approved standby or speed-control arrangement can be evaluated, but it must preserve operating limits and drying readiness. For compressed-air banks, pulsing still needs correct coverage and actual-air timing.

How should dryness be measured on a complex product?

Begin with the critical zones and the downstream consequence. Combine an appropriate overall measure with local observations where necessary. Record the method, observation time and handling before assessment. A low mean residual mass does not establish that every seam or recess is acceptable, and a visual check does not automatically quantify an invisible film.

What information should be sent for a drying-system proposal?

Provide product drawings or suitable images, critical wet areas, liquid description, incoming condition, speed and spacing, required final result, available utilities and installation constraints. Include the current configuration and observations of the problem. A useful proposal should identify the supported operating point, complete equipment scope, control and drainage responsibilities, and the product conditions still requiring assessment. [S1; N1]