Ultrasonic Atomization: How Piezo Vibration Makes a Droplet You Can't Feel

Most nozzles make droplets by forcing liquid through a hole and letting a fast gas or a sharp edge tear it apart. Ultrasonic atomization skips the gas and the high-pressure liquid line entirely. A piezoelectric element vibrates a small horn or a thin tapered surface at a very high frequency, typically 20–120 kHz, and the liquid film riding on that surface gets shredded into droplets as it leaves the tip. The breakup is driven by the vibration, not by shear.

That single difference changes three things buyers care about: the droplet size you can hold, how much compressed air you burn, and how even the deposit is across the part. This guide is written for the engineer evaluating an ultrasonic atomizing nozzle against a pneumatic air atomizer for the same duty: coating a web, conditioning a gas stream, depositing a release agent. It covers how the physics actually works, what the honest droplet and flow ranges are, where the technology wins, where it loses, and what you have to specify so the first installation is not a science project.

What the Vibration Actually Does

The classic way to atomise, a pressure nozzle or an air atomizer, uses kinetic energy to pull a liquid sheet apart. Ultrasonic atomization uses a vibrating surface instead. A liquid film only a few hundred microns thick is fed onto the face of a horn vibrating at its resonant frequency. The vibration excites capillary waves on the film surface, and at the right amplitude the wave crests detach as droplets. The process is called capillary-wave atomization, and it was mapped in the 1960s by Robert Lang, whose relation still describes the result: the droplet diameter runs close to a third of the capillary wavelength set by the vibration frequency.

Two engineering facts fall out of that physics, and both matter more than any catalogue drawing:

• Higher frequency → smaller droplets. The capillary wavelength shrinks as frequency rises, and the droplet size follows it. A 60 kHz tip makes a finer mist than a 20 kHz tip on the same liquid.
• Droplet size is nearly independent of liquid flow rate. Unlike a pneumatic nozzle, where turning up the air changes the mean droplet a lot, an ultrasonic nozzle holds its Sauter mean diameter (SMD) fairly steady across its usable feed range. The size is set by frequency and liquid properties; the feed rate only sets throughput.

That second point is the reason ultrasonic nozzles show up where a pneumatic nozzle would throw too much scatter. If your process is a coating that must land at a controlled thickness, a droplet size that does not wander when the pump surges is worth a lot.

The Frequency-Size Relationship

For water-like liquids the capillary wavelength follows λ ≈ (8πσ/ρf²)^(1/3), where σ is surface tension, ρ is density and f is the vibration frequency. Multiply by roughly 0.34 to get the expected mean droplet diameter. That is a published physics relation, not a vendor claim, so it is worth showing for the frequencies you will actually be offered:

Frequency	Capillary wavelength (water)	Expected SMD (water)	Typical duty
20 kHz	~165 µm	~55 µm	Wider depositing patterns, higher feed
40 kHz	~105 µm	~35 µm	General coating and humidification
60 kHz	~80 µm	~27 µm	Thin-film coating, fine humidification
120 kHz	~50 µm	~17 µm	Precision films, low feed rates
~2.4 MHz	~7 µm	~2 µm	Medical nebulizer class (different device)

Real nozzles do not hit the relation exactly, tip geometry, the liquid feed method and the drive amplitude all shift the result, but the trend is real, and it is the first thing to check when someone quotes a droplet number: does the frequency support it? A 20 kHz tip claiming 10 µm SMD on water is a red flag.

Real Droplet and Flow Ranges (Not Catalogue Promises)

A production ultrasonic atomizing nozzle, the kind built around a vibrating capillary or a wide-band horn, lands in a droplet band of about 10–60 µm SMD, with liquid feed in the 1–60 L/h range per nozzle depending on the tip width and the drive power. Narrow-film tips run fine at 10–20 µm for low flow; wider depositing nozzles spread to 30–60 µm for higher throughput. If a supplier quotes sub-5 µm at industrial flow rates, they are quoting the nebulizer end of the technology, not the industrial coating end.

The honest comparison against pneumatic air atomizing is a window, not a line:

Mechanism	Droplet (SMD)	Liquid feed (per nozzle)	Compressed air	Uniformity driver
Ultrasonic (piezo)	10–60 µm	1–60 L/h	None (electric, few–tens W)	Vibration frequency
Ultrasonic-assisted air atomizing	10–30 µm	1–20 L/h	~4 Nm³/h	Vibration + air shear
Pneumatic internal mix	10–35 µm	0.5–120 L/h	3–12 Nm³/h	Gas/liquid shear
Pneumatic external mix	25–80 µm	2–200 L/h	7–14 Nm³/h	Gas/liquid shear
High-pressure misting	10–120 µm	0.05–1.5 L/min	None	Pressure + orifice

The takeaway is not “ultrasonic is smaller”. Pneumatic internal mix can go finer. The takeaway is ultrasonic gets to the fine end without an air compressor, and it holds the droplet size steady as you change flow, which pneumatic nozzles do poorly.

Where Ultrasonic Beats Pneumatic Air Atomizing

The win condition is deposit uniformity with low overspray, not peak flow:

• Thin-film coatings on medical devices, sensors, and electronic substrates, where a 20 µm fog landing evenly beats a 40 µm cone with a heavy centre.
• Fuel cell and battery electrodes, where coating weight variation directly kills cell performance.
• Pharmaceutical and nutraceutical deposition, where dose uniformity is a compliance issue, not a nice-to-have.
• Release agents and lubricants on sticky or temperature-sensitive webs, where you want a mist, not a stream, and you do not want to pay for compressed air across a 10-nozzle line.
• Spray drying feed, where a fine, relatively mono-disperse droplet makes the dryer shorter and the powder tighter in size.
• Small-area humidification or gas conditioning, where a fog must be generated near the point of use without a compressor or a high-pressure pump.

A pneumatic nozzle can do most of these. It just costs more in air and lands a wider droplet distribution, so the coating edge is softer. Where the process is sensitive to the fine tail of the distribution, and coating weight and dose uniformity both are, the ultrasonic nozzle’s tighter distribution is the differentiator, not just the mean size.

Where You Should NOT Use Ultrasonic

Be honest about the limits:

• High liquid throughput. If you need hundreds of L/h through one point, a pneumatic or a full-cone hydraulic nozzle is the right tool. Ultrasonic tips feed a thin film; they don’t move bulk liquid.
• Viscous or particle-heavy fluids. The feed path is a capillary or narrow slot. Slurries, high-solid suspensions, and thick coatings clog the tip. Pneumatic external mix handles viscosity far better.
• Cheap, dirty, or recycled water. Ultrasonic tips want filtered, low-solid feed. A spiral full-cone nozzle will run on murky water all day; an ultrasonic tip will choke.
• Large-area flood coverage. A single ultrasonic nozzle covers a narrow band. Wide-area cooling or dust suppression is better served by misting or hollow-cone hydraulic nozzles.
• High ambient dust or splashing. The vibrating tip is exposed; debris landing on the horn face perturbs the film and atomization becomes uneven.

How to Select an Ultrasonic Nozzle

Size it the same way you size any atomizer, just with frequency as the extra dial:

1. Pick the droplet you need first. Below ~20 µm for fine films; 30–60 µm for wider depositing patterns. That sets your frequency band.
2. Match feed rate to the tip width. Low flow (1–10 L/h) → narrow capillary tip; higher flow (10–60 L/h) → wider depositing nozzle. Confirm the liquid viscosity is within the tip’s rated range.
3. Filter the feed. A 25–50 µm line filter ahead of the nozzle pays for itself in stopped clog calls.
4. Confirm the drive power and drive unit. Count on a few watts for a small tip up to a few tens of watts for a wide depositing nozzle. The drive electronics must hold the resonant frequency as the tip warms and loads change. A fixed-frequency drive drifts off resonance and the nozzle silently gets coarser.
5. Check standoff and pattern overlap. Ultrasonic nozzles make a narrow cone or a fan; place them so patterns overlap correctly across the web or part. The low spray velocity means the deposit edge is soft, a feature for uniformity, but it also means you cannot “aim” the spray the way you aim a pneumatic one.
6. Plan a containment zone. Because the spray is so fine and slow, drafts move it. On a coating line, extraction or a containment shroud is part of the installation, not an afterthought.

A Worked Example: The Air Bill You Do Not Buy

Say you are coating a 100 mm-wide web with a release agent and you need a 25 µm deposit with tight uniformity. A pneumatic internal-mix nozzle at 15 µm SMD would work but burns ~5 Nm³/h of air per nozzle and throws a wider distribution. An ultrasonic tip set to ~40 kHz gives a steady ~25 µm SMD across a 5–20 L/h feed with no air line. Across an 8-nozzle line that is roughly 40 Nm³/h of compressed air you did not buy, and a coating edge that stays in spec.

Put a number on it. At a conservative $0.03 per Nm³ and 8,000 operating hours a year, 40 Nm³/h is about $9,600 a year in delivered air. The eight ultrasonic drives draw maybe 120 W in total, 0.12 kW, which at $0.12/kWh is about $115 a year in electricity. That is an order-of-magnitude difference of roughly eighty to one, and it is why the technology earns its place on continuous coating lines despite the higher first cost of the hardware. The trade is real: you pay more for the nozzle and drive, and the payback is measured in months when the line runs continuously.

Operating Cost Beyond the Air

The running-cost comparison is not only electricity versus compressed air. Three more items decide the annual picture:

• Maintenance. An ultrasonic tip has no orifice to erode in the classic sense, but the horn face does erode slowly in abrasive duty and the piezo element is a consumable with a life in the thousands of hours. Spare tips and drives belong in the budget.
• Liquid preparation. Filtration is stricter than for a hydraulic nozzle, a 25–50 µm filter and, for some duties, softened or deionised water to stop scale on the vibrating face.
• Cycle time and yield. This is where ultrasonic usually wins the total-cost argument: a tighter droplet distribution means less over-coating, less rework and less material loss. On a coating where even two percent better material retention pays for the nozzle several times over, the air savings are secondary.

Failure Modes and Troubleshooting

Ultrasonic nozzles fail differently from pneumatic ones, and most field complaints are drive or feed problems rather than atomization problems:

Symptom	Likely cause	Fix
Spray coarsens or spits intermittently	Drive drifted off resonance, or feed above the film capacity	Confirm drive locks on resonance; reduce feed; check amplitude
Droplet size changes as the shift warms up	Tip temperature shifts resonance; drive not tracking	Use a frequency-tracking drive; confirm power supply margin
No spray at all, liquid drips off the tip	Tip face not vibrating: drive fault, cable, or broken piezo	Check drive output and cable; replace tip/piezo assembly
Uneven pattern, heavier on one side	Tip face fouled or damaged; film not covering the face evenly	Clean the face per the manual; inspect for erosion or chips
Clogging at the feed slot	Solids or scale in the liquid	Upgrade filtration; check chemistry against feed-path material
Liquid pools and will not atomise	Feed above capacity, or viscosity above the tip rating	Lower feed; heat the liquid; move to a wider tip

The pattern is consistent: ultrasonic atomization is a resonant system, and every failure mode except simple clogging is a symptom of losing resonance or starving the film. When in doubt, check the drive first. It is the cheapest diagnosis and the most common cause.

What an OEM Buyer Should Put on the Spec

If the ultrasonic nozzle is going into a product or a line you sell, the spec sheet needs more than a droplet number:

1. Frequency and expected SMD at a defined feed rate and liquid temperature, never a bare “10 µm”.
2. Feed rate range and the viscosity window the tip was validated against.
3. Drive requirements: voltage, power, and whether the drive tracks resonance automatically.
4. Tip material and the liquid chemistries it is rated for.
5. Filtration requirement: what size filter protects the feed path.
6. Standoff and pattern geometry at the intended mounting.
7. Life target: expected tip and piezo life under continuous duty, and the replacement-part strategy.
8. Test liquid and conditions behind every quoted figure.

A spec that quotes droplet size without a frequency, a feed rate and a liquid temperature is not a spec. The physics is public, the capillary-wave relation above, so the supplier’s number can be checked against the frequency before you commit.

Bottom Line

Ultrasonic atomization is the right call when uniformity and low overspray beat raw flow, and when you would rather not run a compressor. It holds a tight droplet size as flow changes, needs only electric drive, and lands in the 10–60 µm band that pneumatic air atomizers also reach, but it gets there differently, and that difference is what earns its place on a precision coating line.

Frequently Asked Questions

What droplet size do ultrasonic atomizing nozzles produce? Production units sit in the 10–60 µm band (Sauter mean diameter). Fine-film designs reach ~10–20 µm at low flow; wider depositing patterns run 30–60 µm. Frequency sets the size: roughly 55 µm at 20 kHz, 35 µm at 40 kHz, 17 µm at 120 kHz for water-like liquids.

Do ultrasonic nozzles need compressed air? No. Breakup is by the piezo-driven vibrating surface, not by gas shear. You need a clean liquid supply and a low-power electrical drive, a few watts to a few tens of watts per nozzle. No air line, no compressor.

When should I pick ultrasonic over pneumatic air atomizing? Pick ultrasonic when coating uniformity and low overspray matter more than throughput: medical device coating, fuel cell electrodes, release agents on sticky webs, pharmaceutical deposition. Pick pneumatic when you need high liquid flow or are spraying a viscous fluid a thin vibrating film cannot feed.

Why is my ultrasonic nozzle coarser than the datasheet? Check the drive first. If it is not tracking resonance as the tip warms, the amplitude drops and the spray coarsens. Then check feed rate. Above the film capacity the tip cannot atomise the surplus. Then check the tip face for fouling.

Can ultrasonic atomize viscous liquids? Only within a window. Above roughly 20–50 cP, depending on the tip, the film breaks down and atomization becomes uneven. Heating the liquid to drop viscosity is the usual fix; for genuinely viscous fluids a pneumatic external-mix nozzle is the better tool.

How much power does an ultrasonic nozzle draw? A few watts for a small tip up to a few tens of watts for a wide depositing nozzle, plus drive losses. Against a compressor feeding pneumatic nozzles on the same duty, the difference is typically one to two orders of magnitude.

What liquid preparation does an ultrasonic tip need? Filtered feed (25–50 µm), low solids, and for hard water a scale-control plan. Scale on the vibrating face kills atomization. That is stricter than a hydraulic nozzle needs, and it belongs in the installation budget.

Is an ultrasonic-assisted air nozzle the same thing? No. Some suppliers combine a vibrating tip with a small air assist to widen the pattern or push the spray onto the part. The hybrid still needs air (a few Nm³/h), so it gives up the no-air advantage to gain reach. It sits between a pure ultrasonic and a pneumatic atomizer in cost and capability.

A Sizing Checklist

• Droplet target fixed by the process outcome (coating weight, evaporation distance), not by “as fine as possible”.
• Frequency band chosen to support the droplet target. Check the capillary-wave relation.
• Feed rate matched to tip width; viscosity confirmed within rating.
• 25–50 µm feed filtration in place.
• Drive unit with resonance tracking; power supply sized with margin.
• Standoff and pattern overlap worked out for the web or part geometry.
• Containment or extraction planned for the slow, fine spray.
• Spare tip and drive budgeted; life target agreed with the supplier.

The BoreJet air atomizing nozzle range includes both pneumatic and ultrasonic-assisted models, with air consumption and droplet figures per unit, so the air-versus-electric comparison can be run on real numbers. For the droplet physics behind the choice, the droplet size guide explains SMD and the air-to-liquid lever in detail, and if your duty is continuous coating, the air cost guide shows why the air bill usually decides the payback. Send your duty, web width, line speed, target film weight, to our application team and we will point you at the technology and model that match.