BlogApplication notes
Why has my ultrasonic cleaning become less effective?
- Type
- Application notes
- Published
- 20 November 2026
- Prepared by
- Technical team
- Product group
- Cleaning & Surface Prep
Running the cycle longer is the usual response when ultrasonic cleaning weakens, but the conditions that soften bubble collapse stay, and long runs can erode soft metal surfaces. What actually weakens cleaning can be a bath too hot or cold, too little degassing, the wrong cleaner concentration, a load blocking the sound, or a changed level or worn transducers, and several can arrive together.
Some days the usual ultrasonic cleaner runs the same parts for the same time, yet chips stay in blind holes and an oily sheen shows on the surface. A longer cycle seems to help a little, then the rejects return the next week, and the tank sounds different from before. It looks like a breakdown, but a service call may come back with nothing wrong.
When cleaning weakens, under-cleaned parts move to the next step and rewashing costs time and energy. Trying to make up for it with longer cycles can bring cavitation erosion on soft metal surfaces (Glasscock, 1995). It is also worth remembering that ultrasound does not replace the cleaner. The cleaner emulsifies oil and wets and dissolves the soil; ultrasound speeds that action close to the surface (Fuchs, 2014). Cleaning power therefore depends both on how many bubbles burst and how hard, and on how much work the liquid can still do.
An ultrasonic cleaner uses transducers, the parts that turn electricity into vibration, on the tank bottom or walls to send sound, usually at tens of kHz, into the liquid. As the sound repeatedly squeezes and stretches the liquid, bubbles grow from tiny nuclei during the stretch (Leighton, 1994; Mason & Lorimer, 2002). They grow over several cycles and then collapse sharply during the squeeze, producing local high pressure, shock waves and microjets aimed at the surface. This is cavitation (Mason & Lorimer, 2002; Verhaagen & Fernández Rivas, 2016). That force lifts particles and films off the surface and drives cleaning liquid into narrow gaps and blind holes (Fuchs, 1995, 2014; Yusof et al., 2016). How hard the bubbles collapse depends strongly on temperature, dissolved gas, surface tension and the sound field, the way sound is spread through the tank (Mason & Lorimer, 2002; Niemczewski, 2007; Nomura & Nakagawa, 2001).
In such cases the technical team sets up the five hypotheses below and rules them out one by one. The start is always the same: measure temperature and level, then run a foil test, in which aluminium foil is dipped briefly and the holes and wrinkles show where cavitation acts (IEC, 1987), and compare the result with a record made under the same conditions when cleaning was good.
Hypothesis 1: bath too hot or cold
Bath temperature moves within a day and with the seasons. The times to watch are tanks run all day with the heater on, summer afternoons, and at the other end winter mornings or the hours just after a fresh fill. A typical scene is a tank where parts that cleaned well in the first morning batch come out less clean as the afternoon goes on, and steam rises when the lid is lifted. The opposite scene, in the same tank, is oily soil left only on the first batch of a winter Monday.
Temperature affects cleaning in two opposite ways. As it rises, viscosity and surface tension fall and the cleaner's chemistry speeds up, so oil and soil loosen more easily (Fuchs, 1995, 2014). At the same time vapour pressure, the liquid's tendency to turn to gas, rises, and more vapour fills each bubble. Vapour and gas inside a bubble act as a cushion during collapse, so the more vapour, the softer the collapse (Leighton, 1994; Mason & Lorimer, 2002). Near the boiling point the liquid starts to boil at the transducer face and at bubble sites, and cavitation weakens sharply (Fuchs, 1995, 2014).
So which temperature is best? A study of tap-water cavitation under practical cleaning conditions found the highest intensity below 20 °C, with a second peak possible between 20 and 40 °C during heating, its position depending on liquid height and air content (Niemczewski, 2007). There is, in other words, no single best temperature; it differs by liquid and by tank. And since the energy the transducers put into the liquid ends up as heat, the bath slowly warms during continuous running even with the heater off.
Measuring the actual bath temperature, not the heater set point, is where to begin. A thermometer at the centre and in a corner, read at the start and end of each batch, shows whether the bath climbs through the afternoon or sits below the recommended range in the morning. A pattern that worsens in the afternoon suggests a rise during running; oily soil left in winter or just after a fresh fill suggests a bath below the cleaner's range. Foil tests at several temperatures show directly where cavitation is strongest in that tank (IEC, 1987; Niemczewski, 2007).
Set the standard temperature within the cleaner maker's recommended range, at the point where foil tests showed the strongest cavitation. If the bath goes over the range by afternoon, hold the upper limit with lid-open time, rest intervals or a cooling coil. If it sits below the range, load parts only once it reaches the target, and write the preheating time into the work standard.
Hypothesis 2: too little degassing
A fresh bath can look clean and still clean worse. Look at this hypothesis just after a change of liquid, after a large one-off top-up for evaporation, and on tanks with pump and filter loops that draw in air. A common example is a shop that changes the bath on Monday morning and puts the first basket in straight away. Coarse streams of bubbles keep rising over the transducers and the sound is rough, yet the parts come out less clean than the week before.
The trouble is dissolved air. Water and water-based cleaners hold dissolved air, and each time a bubble expands in the sound field, slightly more gas diffuses in than diffuses out, so the bubble gradually fills with gas. This is called rectified diffusion (Leighton, 1994). A gas-filled bubble cannot collapse hard; it only oscillates, or grows and floats up to the surface (Leighton, 1994; Mason & Lorimer, 2002). That is why a fresh fill shows plenty of visible bubbles yet cleans worse (Fuchs, 1995, 2014).
Degassing means running the sound at start-up so that it grows these gas bubbles and drives them out through the surface (Fuchs, 1995, 2014). A recent study that varied the dissolved air and measured cavitation erosion found that once air passed saturation, gas-filled bubbles and microbubbles increased and erosion fell (Xia et al., 2025). Cleaning relies on the same collapse force, so an air-rich bath tends to clean less well too.
Weak cleaning only right after a fresh fill or a large top-up makes too little degassing the likely cause. At start-up the bubbles may be coarse and the sound rough, then as degassing proceeds the bubbles thin out and the sound evens out; watch and listen for that change. On tanks with circulation, check for air drawn in at the suction or the return, or return liquid falling onto the surface and folding air in. A foil test at the same spot before and after degassing makes the difference visible (Glasscock, 1995; IEC, 1987).
Put degassing into the work standard rather than leaving it to habit. After a fresh fill or a large top-up, run the tank empty in degas mode for the time the equipment maker advises, and record it. Replace evaporation little and often rather than in one large addition, with a short degas after each. Tidy up the circulation path by bringing the end of the return line below the surface and stopping air from entering at the suction.
Hypothesis 3: wrong concentration
Cleaner concentration drifts without showing it. It tends to leave its range in tanks used a long time without a change, in processes that put heavily oiled parts straight in, and where evaporation is made up with water alone or, the other way, concentrate is added by eye. In a tank that has taken oily parts in the same bath day after day, a thin oil film comes back onto parts as they come out, and an oil slick forms on the surface. In a tank where concentrate has been poured in generously, foam covers the surface instead, and stains remain after rinsing.
The surfactants in a cleaner lower the surface tension of water so it wets narrow gaps, and they wrap loosened oil in small droplets so it cannot redeposit (Fuchs, 2014). With each wash these components are spent holding soil and leave on the parts, so concentration falls. As it falls, the bath can no longer hold what the ultrasound loosens; soil settles back on the parts, and oil builds up on and in the liquid.
Surfactants also change how the bubbles behave. Adsorbed on bubble surfaces, they have been shown to slow bubble coalescence (Sunartio et al., 2007), and an anionic surfactant was observed to raise cavitation activity at low concentration but lower it at higher concentration (Lee et al., 2005). More is not better, then: chemistry and bubble behaviour line up when concentration sits in the maker's range. Too high, and foam rises, rinsing gets harder and the risk of residue can grow.
The parts coming out and the surface of the bath give the first answer. Oil film returning to parts, an oil layer on the surface or a steady fall in what gets cleaned in a fixed time suggests low concentration or soil build-up. Foam over the surface or more stains after rinsing points to over-concentration. Measure concentration on a fixed schedule with the refractometer or titration kit the maker specifies, and record it. A bath heavy with oil does not recover even at the right concentration, so oil content becomes the basis for deciding when to change it.
If concentration is low, add concentrate; if an oil layer keeps forming, skim or overflow it off, or change the bath. If it is over the range, dilute with water, degas, and measure again. Pre-washing or draining heavily oiled parts cuts the soil carried into the tank, which keeps the same bath in range for longer.
Hypothesis 4: load blocking the sound
When results differ by position in the basket, look at the load. It shows up when more parts go in per batch to keep up with output, when baskets have fine mesh or solid sides, when parts sit directly on the tank bottom, and with parts full of blind holes. After the number of parts per basket goes up, parts on top and at the edges may come out clean while only those in the middle and lower layers stay dirty. Resetting temperature and concentration does not change those spots.
Sound does not spread evenly through a tank. Waves reflected from the walls and the surface overlap into standing waves, giving alternating spots of high and near-zero pressure (Fuchs, 2014; Mason & Lorimer, 2002). Parts and baskets added to the tank absorb and reflect part of the sound, so cavitation weakens behind them like a shadow (Fuchs, 2014). The bubbles absorb energy too: a study modelling the sound field in a cleaning vessel found that more bubbles dissipate more energy, lowering the pressure that reaches the object (Nomura & Nakagawa, 2001).
A foil study of cavitation in large cleaning tanks likewise found ultrasonic activity unevenly distributed, and concluded that moving parts during cleaning gives more uniform results (Glasscock, 1995). Fine-mesh baskets, stacked parts and parts resting directly on the tank bottom all block the path of the sound or damp the radiating surface (Fuchs, 2014). Faces where parts touch and the insides of blind holes see little exchange of liquid and clean especially slowly.
Running the foil test in three states shows the effect of the load at once. Test the empty tank, the tank with an empty basket and the tank with a real load under the same conditions, and compare the pattern of holes and wrinkles (Glasscock, 1995; IEC, 1987). Defects that began when output was raised make this hypothesis stronger. Even the same part, placed the wrong way, can trap air in a blind hole and keep the liquid out, so check orientation as well.
Lay parts in a single layer, not touching one another, and keep the basket off the tank bottom. Replace fine-mesh or solid-sided baskets with open mesh or wire racks. Raising and lowering the basket slowly during cleaning, or turning it partway through, reduces the shadows (Glasscock, 1995). Tilt parts with blind holes so the air can escape.
Hypothesis 5: level or transducers
If one zone of the tank stays weak after every other condition is set, or an old tank's results have slowly worsened, this is the hypothesis left. It is worth checking on tanks filled by eye with no level mark, on heated tanks that lose a lot to evaporation, and on units that have run on the same transducers for years. Picture a tank in use for several years, topped up by eye, where at some point only the parts in one corner start to come out less clean. Changing temperature, degassing, concentration and load leaves that corner as it was.
The standing-wave pattern in a tank depends on liquid height; change the level and both the strong-cavitation spots and the response to temperature shift, even in the same tank (Niemczewski, 2007; Nomura & Nakagawa, 2001). As evaporation lowers the level, parts may stick out of the liquid or the sound field may drift from the design. Over long use, transducers can come unbonded or the radiating plate can thin from cavitation erosion, cutting output, and because the change is gradual it is hard to notice (Fuchs, 1995, 2014).
Before blaming the transducers, know the limits of the foil test. It is a simple way to see cavitation distribution, but it is qualitative and depends on foil thickness, temperature and handling, so results compare only when repeated under identical conditions (IEC, 1987; Verhaagen & Fernández Rivas, 2016). If the same zone keeps coming out weak under the same conditions, suspect the transducers beneath it. Where a more quantitative check is needed, measuring cavitation noise with a hydrophone, a microphone that goes into the liquid, is set out in an international technical specification (IEC, 2024).
Mark the maker's design level on the tank and set the liquid to it before each day's start. Run foil tests regularly with the same foil, temperature, level and time, and keep the photos, so that slow change can be seen by comparing one photo with the next. If a zone stays weak, have the equipment maker check the transducers and the generator, the unit that sends them their electrical signal.
Testing the hypotheses
Much can be checked before taking the unit apart. First write down when the trouble started and where it shows. Trouble only in the afternoon brings hypothesis 1 to mind first; trouble right after a change or a large top-up, hypothesis 2; oil film returning more and more, hypothesis 3; one place in the basket, hypothesis 4; one zone of the tank, hypothesis 5. Timing and position alone cut the list a great deal.
Next come the easy measurements. Compare the actual level with the mark and the temperature at the centre and in a corner with the recommended range, and check whether a degas run followed the last change or top-up. Then measure concentration and look for an oil layer, foam and cloudiness. If a value is out at this stage, correct it first and check the result again on the same parts (Fuchs, 2014; Niemczewski, 2007).
If every value is right and cleaning is still weak, use foil tests as a control. Test first in the empty tank with those conditions set; if it perforates evenly, put the equipment hypothesis aside, then test again with a real load and compare. A large difference points to hypothesis 4, so change the load and basket (Glasscock, 1995). If the same zone stays weak even in the empty tank with the level set, it points to the transducer side of hypothesis 5, and it is time to have them checked (IEC, 1987, 2024).
What makes diagnosis easy is a baseline record. Keep the temperature, level, concentration, degas time, load count and an empty-tank foil photo from a period of good cleaning on one sheet; when trouble starts, set today's values beside it and see at once what changed. The foil test is qualitative, so it compares only with the same foil, position and time as the baseline photo (IEC, 1987; Verhaagen & Fernández Rivas, 2016). Judge the cleaning result too: pick a few representative parts, assess them by the methods already used on the floor, such as a visual check, a wipe test or a water-break test, and record the results alongside.
The hypotheses often overlap. On a summer afternoon a rising temperature, a lower level from evaporation and the higher concentration that follows all arrive together. In that case restore the level first, degas after topping up, then bring the temperature down and measure concentration again. Change one condition at a time and record the result, so it is clear which action worked; it is done only when the same good result holds over several runs after the fix.
Limits and cautions
Specific values for temperature, concentration and degassing time follow the cleaner and equipment makers' ranges. Unlike typical tanks running water-based cleaners, cleaning with flammable solvents, high-frequency cleaning at hundreds of kHz and above, and parts sensitive to cavitation erosion, such as soft metals or coated surfaces, need separate review (Glasscock, 1995). Where poor cleaning bears on product safety or specification compliance, do not decide on floor judgement alone; confirm with the equipment and cleaner suppliers.
Never put hands into a running tank; pain has been reported from contact with 20 to 40 kHz cleaners (Health Canada, 1991). Close the lid to cut airborne noise and cleaner mist, and wear gloves and eye protection as the cleaner's safety data sheet, the SDS, directs. Heating alkaline or acid cleaners increases vapour and mist, so confirm ventilation. Leave electrical checks and repair of transducers and generators to the equipment maker or a qualified technician.
Sources
- Fuchs, F. J. (1995). Ultrasonic cleaning: Fundamental theory and application. NASA Technical Reports Server, conference paper. https://ntrs.nasa.gov/citations/19950025362
- Fuchs, F. J. (2014). Ultrasonic cleaning and washing of surfaces. In J. A. Gallego-Juárez & K. F. Graff (Eds.), Power ultrasonics: Applications of high-intensity ultrasound (chapter 19). Woodhead Publishing, Elsevier. https://shop.elsevier.com/books/power-ultrasonics/gallego-juarez/978-1-78242-028-6
- Glasscock, B. H. (1995). Cavitation effects in ultrasonic cleaning baths. In The 1995 research reports: NASA/ASEE Summer Faculty Fellowship Program. NASA. https://ntrs.nasa.gov/citations/19960020723
- Health Canada. (1991). Guidelines for the safe use of ultrasound: Part II, Industrial and commercial applications (Safety Code 24). https://www.canada.ca/en/health-canada/services/environmental-workplace-health/reports-publications/radiation/guidelines-safe-use-ultrasound-part-industrial-commercial-applications-safety-code-24.html
- International Electrotechnical Commission [IEC]. (1987). IEC TR 60886:1987, Investigations on test procedures for ultrasonic cleaners. https://webstore.iec.ch/publication/3816
- International Electrotechnical Commission [IEC]. (2024). IEC TS 63001:2024, Measurement of cavitation noise in ultrasonic baths and ultrasonic reactors. https://webstore.iec.ch/publication/91816
- Lee, J., Kentish, S. E., Matula, T. J., & Ashokkumar, M. (2005). Effect of surfactants on inertial cavitation activity in a pulsed acoustic field. The Journal of Physical Chemistry B, 109, 16860–16865. https://doi.org/10.1021/jp0533271
- Leighton, T. G. (1994). The acoustic bubble. Academic Press. https://shop.elsevier.com/books/the-acoustic-bubble/leighton/978-0-12-441920-9
- Mason, T. J., & Lorimer, J. P. (2002). Applied sonochemistry: The uses of power ultrasound in chemistry and processing. Wiley-VCH. https://doi.org/10.1002/352760054X
- Niemczewski, B. (2007). Observations of water cavitation intensity under practical ultrasonic cleaning conditions. Ultrasonics Sonochemistry, 14, 13–18. https://doi.org/10.1016/j.ultsonch.2005.11.009
- Nomura, S., & Nakagawa, M. (2001). Analysis of an ultrasonic field attenuated by oscillating cavitation bubbles. Acoustical Science and Technology, 22(4), 283–292. https://doi.org/10.1250/ast.22.283
- Sunartio, D., Ashokkumar, M., & Grieser, F. (2007). Study of the coalescence of acoustic bubbles as a function of frequency, power, and water-soluble additives. Journal of the American Chemical Society. https://doi.org/10.1021/ja068980w
- Verhaagen, B., & Fernández Rivas, D. (2016). Measuring cavitation and its cleaning effect. Ultrasonics Sonochemistry, 29, 619–628. https://research.utwente.nl/en/publications/measuring-cavitation-and-its-cleaning-effect/
- Xia, D., Wu, J., & Su, K. (2025). Classification of regimes determining ultrasonic cavitation erosion in aqueous solutions containing dissolved air. Ultrasonics Sonochemistry, 116. https://doi.org/10.1016/j.ultsonch.2025.107324
- Yusof, N. S. M., Babgi, B., Alghamdi, Y., Aksu, M., Madhavan, J., & Ashokkumar, M. (2016). Physical and chemical effects of acoustic cavitation in selected ultrasonic cleaning applications. Ultrasonics Sonochemistry, 29, 568–576. https://doi.org/10.1016/j.ultsonch.2015.06.013