Case Study: Electrostatic Misting for Poultry Plant Surface Sanitation

A surface that is wet is not a surface that has been sanitized. On a poultry further-processing line that gap shows up hours later. An ATP swab above 150 RLU follows, then a held batch and a re-clean of equipment the schedule had already released. A wand at the end of a hose cannot tell the two states apart. The operator sees the face that takes the spray, and never sees the face behind the frame.

This case covers a North American poultry further-processing plant. Its surface sanitation step ran on portable pressure sprayers and a peracetic acid type sanitizer before the move to electrostatic misting. The route covered contact surfaces, trays and baskets, and equipment frames across three zones, about 12,000 m² per shift. Solution use sat near 720 litres per shift, and about one in five swab points returned above 150 RLU. The aisle between stations stayed wet enough that the plant logged slip incidents there twice in one quarter.

Case Snapshot

The table below sets the reference configuration used for the rest of this account.

Item	Reference configuration
Industry	Poultry further processing, raw and cooked product contact surfaces
Duty	Surface sanitation of contact surfaces, trays and baskets, and equipment frames between production blocks
Treated area	About 12,000 m² per shift across three zones
Original setup	Portable pressure sprayer, 4 to 8 bar at the wand, about 25 percent deposition on mixed geometry, 720 L of working solution per shift
System chosen	Fixed electrostatic misting tunnel, 30 misting nozzles at 0.40 mm orifice and 10 bar, 6 second bursts on a 60 second cycle
Sanitizer and criterion	Peracetic acid type sanitizer at 100 to 200 ppm active, 1 minute contact, ATP swab criterion 150 RLU

Every figure below is the generic shape of this class of surface sanitation retrofit, not a plant record.

The treated area is the figure that drives the rest. Sanitizer volume, head count, pump size, burst timing and swab route all scale from square metres of surface per shift. A plant that starts from head count and works back to area usually ends up with the wrong figure for both.

The Challenge

The sanitation block between production blocks had three cost centres: sanitizer volume, route time and the risk of a failed verification. All three traced back to one number, the fraction of sprayed sanitizer that landed on a surface which needed sanitizing.

The route was a person with a wand, so coverage depended on where the wand was pointed and how long it stayed there. Faces square to the wand took a flooding dose and dripped. Faces behind a rail, under a belt or inside a basket took vapour and drift.

The verification result followed the geometry. About one in five points on the 30 point shift route came back above 150 RLU. The failures concentrated at hinge points, rail undersides, basket corners and the inside faces of frames. Those are the places a wand cannot see.

The volume followed the same geometry. The wand delivered 720 L of working solution per shift to place a sanitizing dose on 12,000 m². That is 60 mL of solution for every square metre treated. Most of the excess ended on the floor and in the drain rather than on a contact face.

Why a Hand Sprayer Covers Only the Face It Can See

Spray deposition is a line of sight process until something steers the drops. Published measurements of manual application put deposition at 30 to 50 percent on flat, square-on targets. On back faces, shadowed pockets and the far side of a frame, it falls under 20 percent.

Shadowing is geometric, and the reach of a spray into a pocket runs at roughly the pocket’s own width in depth. A gap 40 mm wide and 60 mm deep takes almost no deposit from a nozzle sitting in front of it. Hinge barrels, rail undersides, basket corners and frame backs all present a shielded surface to a wand.

A drop leaving a wand tip at 20 to 40 m/s slows within a few hundred millimetres. It then falls or drifts, because neither gravity nor air drag turns a corner.

Raising the dose does not fix a shadow. More volume thickens the film on faces that already passed, adds runoff and leaves the shielded face dry. The plant had tried a longer dwell and a second pass at each station before the retrofit.

How a Charged Droplet Turns a Corner

Charge changes the steering, not the chemistry. A charged drop entering the field around a grounded target follows the field lines, and those lines bend around a part. A drop that would have flown past an edge is pulled onto the back face instead. Published work on charged-drop application reports deposition gains of 2 to 4 times the uncharged case on mixed geometry.

Induction charging is the arrangement used here, and the electrostatic spraying guide covers the same physics for coating duty. An electrode at 20 to 60 kV sits near the nozzle, and the liquid takes up charge as it breaks into drops. Published charge-to-mass ratios for induction charged sprays sit in the 0.5 to 3 mC/kg band. That is enough to move a trajectory over a 300 to 500 mm flight.

The target must be grounded through a defined resistance. A plastic basket or a rubber belted conveyor isolates the part, the field collapses around it and the wrap goes with the field. This plant bonded the tunnel frame, the rails and the drip tray to building earth, then read the resistance at each station at shift start.

Charge is not a substitute for line of sight inside a shielded pocket. The field does not reach deep into a gap whose walls carry the shielding charge. Three ring stations at different azimuths put every face square to a nozzle at some point on the route. Charge handles the edges, the returns and the residual shadow between stations.

Atomization Options for a Surface Sanitation Route

Three families of hardware compete for a duty like this one. They differ far more in where the drops go than in the chemistry they carry.

Method	Droplet band	Deposition on mixed geometry	Shadow and back face coverage	Chemical use
Pressure spray, hand held or fixed, no charge	80 to 250 µm at a fine tip, 200 to 600 µm at a flooding tip	30 to 50 percent square on, under 20 percent in shadow, published range	Poor, line of sight only	Highest, since most of the mass misses the target or runs off
Air-assisted atomizing	30 to 100 µm	30 to 50 percent, published range	Fair, because the air stream can carry drops past an edge	Moderate, and the compressed air adds a separate energy cost
Electrostatic misting with induction charging	30 to 80 µm	60 to 85 percent, published range	Good on exposed edges and back faces, none inside a shielded pocket	Lowest, because the smallest mass per square metre reaches the target

The droplet band matters less than the steering in electrostatic misting. A fixed array of misting nozzles buys repeatability over a wand. It does not change where the drops go until charge is added. Without an electrode the same array still leaves the shadow behind the rail.

Fogging nozzles that make 20 to 60 µm drops with no electrode sit between the two lower rows of that table. They suit humidification and dust suppression, where coverage of a volume matters more than dose on a defined surface. They are a weak answer for a dose.

A mist spray nozzle selected only for its droplet size misses the shielded face for the same reason. Droplet size sets flight time and carry distance, but not direction. A high pressure fog nozzle at 50 to 70 bar makes a very fine fog at high velocity. That velocity breaks up the film, raises runoff and loses the fine end of the cloud to the room air.

The Solution

The route became a tunnel with three ring stations over about 4.5 m of conveyor. Each station carries a ring of heads aimed across the opening, fed from one manifold and switched by one solenoid. At 4.5 m/min a part group spends about 60 seconds inside and passes one station every 20 seconds.

Head count came from coverage geometry rather than a catalogue page. The perimeter of a 1,200 mm by 900 mm opening is 2 x (1,200 + 900) = 4,200 mm. A 65 degree full cone at a 400 mm standoff lays a pattern width on the target plane. That width is 2 x 400 x tan(32.5) = 510 mm. Spacing at 440 mm gives about 14 percent overlap, inside the 10 to 15 percent margin used for even coverage. That gives 10 heads per ring and 30 across the tunnel.

Metering followed the sanitizer specification. An inline dosing pump holds the working solution at 0.2 percent by volume of a peracetic acid type concentrate. That places the active ingredient inside the 100 to 200 ppm band used in published practice for food contact surface sanitizing. One minute of contact inside the tunnel matches the published contact requirement for this chemistry, and the residue drains rather than being rinsed.

Grounding and records completed the installation. Each station has a bonded drip tray, a pressure and flow switch on the manifold and a counter on the solenoid. The counter is the record. A shift with fewer bursts than the route requires shows up in the log rather than in a swab result. The spray applications overview covers how the same deposition logic applies to other wet surfaces in a plant.

Droplet Size, Sink Rate and Air Movement

Droplet size sets three things at once: how long a drop stays airborne, how far it carries, and how much air movement it can survive.

Settling velocity for a small sphere in still air follows Stokes, v = (ρp − ρa) g d² / (18 μ). For a water drop at 20 °C, ρp is 998 kg/m³, ρa is about 1.2 kg/m³ and μ is 1.81 x 10⁻⁵ Pa·s. A 50 µm drop returns 0.075 m/s, and the same expression gives 0.027 m/s at 30 µm and 0.19 m/s at 80 µm.

Published measured terminal velocities for water drops land near 0.07 m/s at 50 µm and 0.25 m/s at 100 µm. The simple expression overstates the 100 µm case at about 0.30 m/s. Drag leaves the Stokes regime above roughly 60 µm. The published figure is the safer design value at the coarse end of a mist band.

Room air moves faster than these drops fall. Supply diffusers, doorways, conveyor movement and foot traffic commonly produce 0.1 to 0.3 m/s in a food processing room. That is above the settling velocity of everything finer than about 90 µm. A 50 µm drop in a 0.2 m/s draught does not descend, it drifts.

Carry distance comes from the same relaxation time, τ = ρp d² / (18 μ). A drop released at 45 m/s from a 0.40 mm orifice at 10 bar decays with a τ of 7.7 ms at 50 µm. It travels about 45 x 0.0077 = 0.35 m before its initial velocity is spent. At 30 µm the carry is only about 0.12 m, and at 80 µm it is about 0.9 m.

The fine end of the band is therefore the constraint on standoff. A 30 µm drop covers 120 mm on its own momentum and then drifts. It reaches a 400 mm target only with a gentle co-flow behind it. That is why the stations use a low velocity shroud rather than shaping air. It runs a 350 to 400 mm standoff and a 40 to 50 kV electrode.

Flow, Pressure and Chemistry Per Shift

Flow through a nozzle of known orifice follows the orifice equation. For water in bar and millimetre units the working form is Q (L/min) ≈ 0.65 · d (mm)² · √P (bar), with a discharge coefficient near 0.95.

At a 0.40 mm orifice and 10 bar that gives Q ≈ 0.65 x 0.40² x √10 = 0.65 x 0.16 x 3.162 = 0.329 L/min per nozzle. Thirty nozzles give 30 x 0.329 = 9.9 L/min, or 2.6 gpm. Trimming the manifold to 7 bar gives 0.65 x 0.16 x 2.646 = 0.275 L/min per head, so the pressure window sets flow from 0.28 to 0.33 L/min per nozzle, and the pump duty is 9.9 x 10 / 612 = 0.16 kW.

Step	Value	Arithmetic
Treated area per shift	12,000 m²	route scope across three zones
Landing target	15 mL/m²	a 15 µm equivalent film, since 1 mL/m² equals 1 µm
Volume that must land	180 L per shift	12,000 x 0.015
Deposition, pressure spray	25 percent	published range, mixed geometry
Working solution, pressure spray	720 L per shift	180 / 0.25
Deposition, charged drops	70 percent	published range, in the 2 to 4 times gain band
Working solution, electrostatic misting	257 L per shift	180 / 0.70
Reduction	2.8 times	720 / 257
Nozzle on-time	26 min per shift	257 / 9.9
Bursts of 6 seconds	260 per shift	26 x 60 / 6
Concentrate at 0.2 percent v/v	0.51 L per shift	257 x 0.002
Annual solution at 300 shifts	77 m³	257 x 300 / 1,000

The reverse check is the part worth keeping. Subtract the 180 L that had to land from the 257 L sprayed and the difference is 77 L. That is 30 percent of what left the nozzles, which matches the 70 percent deposition assumption. The same subtraction on the manual route leaves 540 L unaccounted from 720 L, or 75 percent.

Orifice wear is the drift a metered dose cannot see on its own. Flow scales with the square of the diameter. A 5 percent growth in orifice diameter passes (1.05)² − 1 = 10.3 percent more flow at the same pressure. It arrives without a setting change. A solenoid counter cannot show that drift, so the plant reads manifold pressure and nozzle flow at the annual service.

Concentration is the other half of the dose. A 0.2 percent by volume dilution of a 5 to 15 percent peracetic acid type concentrate puts the active ingredient between 100 and 300 ppm. Published practice for this chemistry sits at 100 to 200 ppm active with about one minute of contact. The assay range across concentrates is why the dilution is set from the label assay rather than the drum volume.

Coverage Sampling and What ATP Tells You

ATP swabs read the biological soil left on a surface in relative light units. A reading of 150 RLU is a common industry criterion for a food contact surface. It is a screen rather than an absolute count, so a surface can pass the screen while still sitting outside its coverage target.

The shift route used 30 points, chosen rather than randomised. They were hinges and latches, rail undersides, basket corners, frame backs, belt undersides, drain covers and the contact faces of the tunnel itself. Those are the points a wand could not reach and a ring does reach.

Sample size decides what a swab result can support. Thirty points per shift runs a trend but does not prove a change. To separate a 20 percent fail rate from a 5 percent fail rate at 80 percent power, a two proportion comparison is needed. It requires 2 x (1.96 + 0.84)² x 0.125 x 0.875 / 0.15² = 76 to 80 swabs per arm. That is two shifts on the old route and three on the new one.

A single rate needs a different number. The expression n = z² p (1 − p) / e² uses p at 0.15, e at 0.05 and z at 1.96. It returns 3.84 x 0.1275 / 0.0025. That is about 196. So roughly 200 swabs place one rate inside a 5 point margin at 95 percent confidence. Swab a defined 100 cm² template after the surface drains and before production restarts, and read inside the meter’s hold window.

The plant acceptance band followed from the same arithmetic. It was under 5 percent of the 30 point route above 150 RLU on a rolling four shift basis. No single point failed three shifts running.

The Results

The table compares twelve months before the retrofit with twelve months after, on the same route scope and the same 300 shift year.

Metric	Before	After	Unit
Working solution per shift	720	257	L
Working solution per square metre	60	21	mL/m²
Sanitizer concentrate per shift	1.44	0.51	L
Deposition on mixed geometry	25	70	percent
Solution sent to the drain per shift	720	257	L
Swab points above 150 RLU	about 20	under 5	percent of route
Operator hours on the route	18	11	hours per shift
Aisle slip incidents	two in one quarter	none logged	incidents
Annual solution at 300 shifts	216	77	m³

Two figures carry a caveat. Deposition is a mass balance rather than a sensor reading, so it inherits the uncertainty of the landing target. A plant setting that target at 25 mL/m² would scale every volume by two thirds.

The honest cost is maintenance. Electrodes, nozzle tips and bonded drip trays add a routine the manual route never had. That is about 3 hours a month for nozzle inspection, electrode cleaning and grounding checks, plus an annual flow check across all 30 heads. No currency figure is given here, because payback depends on local prices for labour, water and concentrate. The 2.8 times ratio is the transferable part, since it holds at any price.

Reading the Numbers Against Published Ranges

None of the figures above are unique to this plant. They sit inside bands that are published, and any plant can check its own numbers against the same bands.

• Droplet band of 30 to 80 µm: the published range for a charged misting head of this class.
• Deposition gain of 2 to 4 times: the published range for charged drop application against the uncharged case on mixed geometry.
• Pressure spray deposition of 30 to 50 percent square on and under 20 percent in shadow: published manual application measurements.
• Electrostatic misting deposition of 60 to 85 percent: the published band for charged application onto grounded, exposed geometry.
• Settling velocities of 0.07 m/s at 50 µm and 0.25 m/s at 100 µm: published measured terminal velocities for water drops.
• Working solution at 100 to 200 ppm active with about one minute of contact: published food contact surface sanitizing practice.
• ATP criterion of 150 RLU: a common industry screen, not a universal legal limit.

Two assumptions carry the whole arithmetic, and both are visible in the table. The treated area is 12,000 m² per shift and the landing target is 15 mL/m². Halve the area and every volume halves with it while every ratio holds, and changing the landing target scales the volumes in direct proportion.

Why This Matters

The saving in this case came from deposition efficiency rather than from nozzle hardware. The plant did not buy a smaller pump or a cheaper sanitizer. It changed where the drops went, and the volume, the drain load, the floor condition and the swab result all followed from that one change.

Three thresholds carry into any surface sanitation route. Keep treated area in square metres per shift as the leading figure, because every other number scales from it. Keep air movement in the misting zone under about 0.1 m/s, below the settling velocity of the drop band in use. Keep the target grounded through a defined resistance, because a charged drop with no field to follow behaves exactly like an uncharged one.

An electrostatic misting route also changes what can be documented. A solenoid counter, a manifold pressure switch and a dosing pump rate produce a per shift record. A wand produces nothing beyond an account of where it was pointed. An audit can trend a log, and it cannot trend an account.

The geometry rule transfers without modification. Where a pocket is roughly as deep as it is wide, no field and no pattern reaches the bottom from a single angle. The answer is a second nozzle angle or a second station rather than more volume. The misting nozzle selection guide walks through the pattern, standoff and droplet chain used to produce the 30 head layout above.

What to Send for a Sizing

A sizing for this duty runs on the same inputs every time. Those inputs are treated area in square metres per shift, the surface types and their materials, and the sanitizer chemistry with its concentration target. The remaining inputs are the required contact time, the conveyor or batch geometry, and the opening dimensions. Last are the flow and pressure available and the grounding path at each station.

Droplet band and charge setting follow from the deposition target and from the air movement in the room. Head count follows from the opening perimeter and the pattern width at the standoff the room allows. A pattern and deposition check on a single station, run against the plant’s own swab route, confirms the deposit target before 30 heads are ordered.

Send the treated area, the surface list, the sanitizer specification and the opening dimensions to the BoreJet application team through the contact page. The sizing returns a head count, a standoff, a pressure window and a burst timing. Each value is checked against the deposit target rather than chosen from a catalogue page. The charged and uncharged heads used for electrostatic misting here are listed in the misting nozzles range.