← Back to Home

Ultrasonic cleaner specs: three numbers that mislead

desk gearmaker toolsultrasonic cleaner

An ultrasonic cleaner is one of those products where the spec sheet looks simple and the purchase decision still goes wrong. A listing gives you three numbers: 40 kHz, 60 W, 2 L. All three are real. None of them answers the question that matters, which is whether the machine will lift flux off the board you are holding.

I did not buy six cleaners and dunk parts in them. What I actually did was copy two manufacturer technical tables line by line, add the decision rules from Branson's official Ultrasonic Cleaning Primer, convert everything to one comparable unit, and then check six live listings against it. Every figure here traces back to the official documents listed at the end or to a product page read on 2026-11-13. Prices move by region and batch, and the price container was empty from my location at capture time, so no prices are quoted. The product links below are affiliate links; ordering through them may earn this site a commission, which changes none of the findings.

// FLOW_01Article roadmap5 NODES01STEP_01TL;DR: cut the shortlist with W/L first02STEP_02The "60 W" on the listing is three lines in the official table03STEP_03Power density: 39.6 W/L down to 10.7 W/L inside one series04STEP_04Only a 2-3x power window is useful; past that you get surface cavitation05STEP_05Frequency is a force setting, not a tierEND_OF_FLOW
Article outline · auto-generated from section headings

TL;DR: cut the shortlist with W/L first

TL;DR: the word "power" on a listing has at least three meanings, and the stated capacity is only 77-93% of the geometric tank volume. What actually separates two machines at the same price is power density, meaning effective ultrasonic watts divided by rated litres. Small parts and boards want a small tank with high density; greasy parts need heat; large parts need a large tank, but a large tank is not a harder cleaning bath.

The "60 W" on the listing is three lines in the official table

Elma publishes a per-model product sheet for the Elmasonic EASY range, and it puts three power figures side by side for the same machine. On the EASY 30 H: 280 W total power consumption, 80 W effective ultrasonic power, 320 W maximum ultrasonic peak performance, with 200 W of heat listed separately.

Copy three heated models out of that sheet and two identities appear:

ModelTotalEffective ultrasonicPeakHeatRated volume
EASY 30 H280 W80 W320 W200 W2.75 L
EASY 40 H340 W140 W560 W200 W4.25 L
EASY 60 H550 W150 W600 W400 W5.75 L

First identity: total equals effective plus heat, on all three rows (280=80+200, 340=140+200, 550=150+400). Second: peak is exactly four times effective, again on all three. The footnote in the same PDF explains why peak is not a capability number: it is a multiplier applied to the waveform, "Due to the form of signal the maximum peak ultrasonic power calculates with factor 4 or 8".

So a lone three-digit wattage on a listing can be total draw, effective ultrasonic power, or peak. Those three differ by a factor of four, and only the middle one describes the cleaning field. The CREWORKS 2 L listing is comparatively honest here: it separates 60 W from a 100 W heater, so you can add up 160 W yourself. Most listings leave a single number and no way to tell which.

CREWORKS 2L ultrasonic cleaner official product photo
CREWORKS 2L ultrasonic cleaner official product photo · 图片来源:官方页面 / 品牌官网

Power density: 39.6 W/L down to 10.7 W/L inside one series

Volume and power do not grow together. Divide the official effective wattage by the official rated volume across the nine Elmasonic E Plus models and the range is wider than expected:

ModelRatedEffective ultrasonicW/L
EP100.2 gal (0.76 L)30 W39.6
EP20H0.5 gal (1.89 L)35 W18.5
EP30H0.7 gal (2.65 L)80 W30.2
EP40H1.1 gal (4.16 L)120 W28.8
EP60H1.5 gal (5.68 L)150 W26.4
EP100H2.5 gal (9.46 L)150 W15.9
EP120H3.4 gal (12.87 L)200 W15.5
EP180H4.8 gal (18.17 L)200 W11.0
EP300H7.4 gal (28.01 L)300 W10.7

Top to bottom is 3.7x. The row worth staring at is EP20H: it costs more than the EP10, adds heat, and holds 2.5x the liquid, yet effective ultrasonic power only moves from 30 W to 35 W. Its density is less than half the EP10's. Buying one size up buys more water, not stronger cavitation.

Crest's Powersonic benchtop line is an independent sample, and it agrees:

ModelVolumeUltrasonic powerW/L
P2302.8 L80 W28.6
P3604.2 L100 W23.8
P5005.4 L120 W22.2
P110012.3 L180 W14.6
P12009.2 L200 W21.7
P180019.6 L240 W12.2
P260026.1 L300 W11.5

A 2.49x spread. Look at P1100 versus P1200: P1200 holds 74.8% of the volume but carries 111% of the power, so density climbs from 14.6 to 21.7. Same 45 kHz, same manufacturer, same shelf tier, a 49% density gap that no listing will show you, because no listing divides watts by litres.

On the retail side: the CREWORKS 6 L is 180 W into 1.6 gal (6.06 L), or 29.7 W/L; the CREWORKS 2 L is 60 W into 2 L, or 30.0 W/L. Same brand, near-identical density, which suggests these lines are built to a fixed density. That is also why "the bigger tank cleans harder" fails: it spreads the same field through more liquid.

Only a 2-3x power window is useful; past that you get surface cavitation

Branson's official primer contains a quantified statement that almost nobody quotes. Cavitation starts only when local pressure drops below the liquid's vapour pressure, so there is a threshold. Push power higher and at some point the liquid can no longer absorb it: violent cavitation forms right at the transducer face, the liquid layer turns elastic, and energy stops reaching the bath. The document calls this surface cavitation.

Then comes the sentence that matters: "The sonic range over which cavitation can exist between threshold and maximum is usually separated by no more than a ratio of 2 or 3 to 1." A tank's useful power window is typically 2x to 3x wide. Below the threshold there is no cavitation; above the ceiling you are heating electricity for nothing.

The purchasing consequence: if a machine already runs at the top of its window, going from 100 W to 300 W will not triple the cleaning. Real gains come from frequency, temperature, chemistry and loading. Conversely, a tank that sits at 10 W/L cannot be rescued by buying the next size up.

Frequency is a force setting, not a tier

Listings present frequency as a hierarchy: 40 kHz above 28 kHz, 48 kHz above 40 kHz. The manufacturers describe it as a trade. Branson again: "As the frequency increases, the number of these cavities also increases but the energy released by each cavity decreases making higher frequencies ideal for small particle removal without substrate damage." More bubbles, each one weaker.

A manufacturer technical page puts numbers on the bands: roughly 25-40 kHz gives larger bubbles that collapse with greater force but distribute less evenly, suited to heavy soils and robust metal; 80-170 kHz gives smaller, gentler bubbles spread more evenly, for delicate parts and electronics.

So frequency chooses force plus uniformity, not quality. Among the six units here, Elma runs 37 kHz, Crest offers the same model at 45 kHz or 132 kHz, and the CREWORKS and VEVOR units run 40 kHz, with one dual-frequency machine in the mix. A single Crest chassis offered at two frequencies is itself the proof that frequency is a configuration, not a grade.

Rated capacity is 77-93% of the geometric tank

"2 L" does not mean you can pour in 2 L. Official tables usually give both the rated volume and the internal dimensions, and multiplying the dimensions settles it. Across Crest's seven models, P500's 5.4 L is only 76.5% of its geometric tank, P1200 reaches 90.3%, and the rest sit at 83-87%. Across Elma's nine, EP10 is 77.8% and EP20H is 93.0%.

That is a 76.5%-93.0% band. The reason is mundane: rounded corners, a fill line, and clearance for parts and basket mean the rated figure is usable volume, not geometric volume. The problem is that many listings give only the capacity, so you cannot tell which one you are being quoted.

Two examples, one good and one thin. The VEVOR 1.3 L publishes 155 x 140 x 65 mm, which works out to 1.41 L geometric against a 1.3 L rating, or 92.2%, and the two agree. The CREWORKS 6 L states 1.6 gallon and no internal dimensions at all, so I can confirm the capacity basis but cannot recompute it. That is not a defect; it is a gap you have to close yourself.

The practical rule: measure the largest part first, then look for internal dimensions, not capacity. A 100 x 100 mm board will not enter a tank that is 118 mm wide, no matter what the litre figure says.

Loading: 70% of the cross-section, 50 mesh, one inch off the bottom

Branson's primer gives three numbers you can act on immediately, and they beat any review:

First, the sum of the parts' cross-sectional areas should not exceed 70% of the tank's cross-sectional area, or the parts block the sound. Second, anything woven tighter than 50 mesh behaves as a solid under ultrasound; openings larger than a quarter inch are what counts as open. Third, neither parts nor basket may rest on the tank bottom, because they reflect energy back into the transducer; a Branson-authored technical article puts the clearance at one inch.

One more that gets ignored: elastomers and non-rigid plastics absorb ultrasonic energy, so be careful with fixturing. Clamping a part in a silicone pad is roughly the same as putting acoustic foam next to it.

Six live units, grouped by what you are cleaning

All six ASINs below were opened as /dp/ pages this round and parsed for a product title and an add-to-cart entry (2026-11-13). Prices came back empty from this location, so none are quoted; check the order page.

Group one, small parts at high density. The Elma Elmasonic EP10 (B06XDJYDWY) runs 37 kHz, 0.2 gal, no heat. Two real strengths: the 30 W figure is specifically effective ultrasonic power, so the basis is unambiguous, and it ships with sweep (frequency modulation) plus pulse (20% extra power), the former aimed at standing-wave hot spots. Real defects: no heater at all, with heating power listed as 0 in the official table, so it will not touch congealed grease; and a 0.76 L tank means a 10 cm square board only fits at an angle.

Group two, cheap, small, unusually explicit. The VEVOR 1.3L (B083KHP4B8) is 40 kHz, 60 W, one 60 W transducer, internal 155 x 140 x 65 mm, timer 0-30 minutes. Its strength is publishing the tank dimensions so you can check the arithmetic. Real defects: no heat; the timer stops at 30 minutes with no continuous mode, so long cycles need restarting; and the 60 W figure never says whether it is effective power or input, with its own "1 x 60W=60W" wording suggesting a transducer rating, which is not the same basis as Elma's effective power.

Group three, heat for grease. The CREWORKS 2L (B075FTFD3R) is 40 kHz, 60 W ultrasonic, 100 W heat, 304 stainless tank, thermostat 68-176 F (20-80 C). Strength: power is split into two labelled fields, and heat and ultrasonics can be set independently. Real defects: only "2 L" with no internal dimensions anywhere, so the capacity basis cannot be recomputed; and the thermostat floor is 20 C, so in a cold room it can only add heat, never remove it.

Group four, large tank. The CREWORKS 6L (B0CGX7CDGL) is 180 W, 40 kHz, 1.6 gal, with a drain valve, same 20-80 C thermostat. Strength: the drain valve, so you do not tip the whole machine to change the bath. Real defects: density is 29.7 W/L, essentially the same as the 2 L model, so "a bigger tank cleans harder" is an illusion; internal dimensions are again absent.

Two more deserve their own paragraph. The EIWEI CD-E3 3L is dual-frequency and its listing is EIWEI CD-E3 3L; it lists "2 high-quality 60W ultrasonic transducer", and does publish both the internal 24 x 14 x 10 cm and a "theoretical capacity of 3L". Its real defect is the vaguest power basis of the six: two 60 W transducers are neither 60 W nor necessarily 120 W of effective power, and the listing has no effective-power field, so W/L can only be bracketed between 35.7 (on 3.36 L geometric) and 40.0 (on the 3 L rating). The VEVOR 6L runs 40 kHz with a 6 L tank, a degassing mode and two power settings, but the listing never states an ultrasonic wattage anywhere, just 40 kHz and 6 L. You cannot tell whether it is a 60 W machine or a 180 W machine, which makes it the largest information gap here.

Troubleshooting: when the bath stops biting

These five checks come straight from manufacturer documents, ordered by likelihood and cost.

First, degas a freshly filled bath. Branson's technical article is specific: run the unit five to ten minutes to drive dissolved gas out, and only repeat it after the bath is newly filled.

Second, check whether parts or basket are touching the tank bottom. Both must clear it, by about an inch.

Third, do the loading arithmetic. Past 70% of the tank cross-section, the parts block their own sound field.

Fourth, inspect the basket. Anything tighter than 50 mesh acts as a solid wall.

Fifth, consider whether the chemistry is spent. Manufacturers warn that solutions wear out and that suspended soil settling on the tank floor suppresses activity; changing the bath usually beats adding time.

Two manufacturer prohibitions: IPA and mineral acids

The reflex for cleaning a circuit board is to pour in isopropyl alcohol. Branson is explicit: "Never use solvents in a small benchtop cleaner. It is neither safe nor environmentally responsible." The reasoning is that solvents vaporise fast and can collect underneath the unit, where electrical components can ignite them. The same passage names mineral acids and bleach as damaging to the machine.

Crest devotes a paragraph on its own benchtop product page to the same question, noting frequent enquiries about low-flashpoint liquids such as acetone, IPA and cyclohexane, and stating that under NFPA 79 compliance its tabletop units and tanks are not designed for Class 1 Division 1 flammable liquids or hazardous locations. The wording: "Crest does not recommend nor will it be held liable for any customer that uses flammable liquids in our ultrasonic table tops".

Two independent manufacturers, two official documents, one conclusion: clean flux with an aqueous or purpose-made cleaning solution, and keep flammable solvents out of the tank.

FAQ

Q: Is 48 kHz worth paying more for than 40 kHz?

Per the manufacturers, frequency sets bubble size and energy distribution, not a grade of cleanliness. A 48 kHz bath makes smaller, denser, gentler bubbles suited to glasses, jewellery and precision parts; 40 kHz hits harder and suits greasy hardware. For boards and delicate parts, the 40-versus-48 gap is smaller than the effect of degassing properly or staying under the 70% loading rule.

Q: Can I put plastic or aluminium parts in?

The official constraints are narrow: elastomers and non-rigid plastics absorb ultrasonic energy, so fixturing needs care, and lower frequencies collapse "with greater force", which raises risk on soft or polished surfaces. No manufacturer publishes an allowable power ceiling for aluminium, so I will not invent one. Use a higher frequency, shorten the cycle, and test on a scrap part first.

Q: Do I need a heated model?

It depends on the soil. The EP10 and the VEVOR 1.3 L have no heater, with heating power listed as 0, and they handle dust, polishing compound, salt and loose dirt; congealed grease, wax and polishing paste need temperature to soften. Branson's authored article notes that many solutions work in the 120-150 F band (about 49 to 66 C), while the thermostats on these six units generally stop at 80 C.

Q: Is a bigger tank always better?

No. What matters is part size, not total cleaning demand. When volume grows faster than power, density falls: across the two manufacturer tables here, the smallest-to-largest drop is 2.49x to 3.7x. Size the tank from your largest part first, then check that the resulting W/L is in the 25-plus range.

Q: Should I rinse after cleaning?

Yes. Ultrasonics lift soil off the surface and hold it in suspension; residue dries back onto the part as the liquid evaporates. Manufacturer procedures call for a rinse and then drying, otherwise you are plating the soil back on.

Bottom line

If you keep one sentence: a listing gives you isolated numbers, and the decision lives in the ratios between them. Divide effective ultrasonic watts by rated litres to get W/L, compare it against the 25 mark, confirm the internal dimensions clear your part, then confirm whether there is heat. Those three steps remove most of the bad matches before you order.

Everything numeric here comes from manufacturer tables and product pages. None of these machines was immersed by me, and that distinction matters: every "official" figure is attributed, and every ratio I computed shows its arithmetic so you can redo it.

For more on reading spec tables, see the 3D printing filament spec breakdown and the solder wire alloy and flux core guide - flux residue is exactly what this class of cleaner is aimed at. The other two pieces of a bench setup are the bench vise jaw width versus opening and what solder fume extraction airflow actually means.

Official sources: Elma product data sheets for Elmasonic EASY 30 H / 40 H / 60 H; the Elmasonic E Plus technical data table from Elma's laboratory and industry brochure; Crest Ultrasonics' Powersonic Benchtop Ultrasonic Cleaners page; Branson (Emerson) Ultrasonic Cleaning Primer PDF; the Branson-authored "Optimizing Your Ultrasonic Cleaner"; Zenith Ultrasonics' technical page. Product pages read 2026-11-13; ASIN availability reflects that moment.

👉 Join Xiaomi MiMo Platform: Leading AI model platform with cost-effective inference

👉 Join Aliyun AI: Top AI products with exclusive coupons for business innovation

📌 This article was AI-assisted generated and human-reviewed | TechPassive — An AI-driven content testing site focused on real tool reviews

🔗 Recommended Tools

These are carefully selected tools. Using our affiliate links supports us to keep producing quality content:

As an Amazon Associate, I earn from qualifying purchases at no extra cost to you.

☁️ DigitalOcean Cloud ⚡ Vultr VPS 🤖 QoderWork CN (Refer & Earn) ☁️ Aliyun AI Products 📚 WordPress Books 🧠 AI & LLM Books 🏗️ System Design Books 🐳 Docker & K8s Books 🐧 Linux Admin Books 🐍 Python Books 🗄️ Database Books 🚀 DevOps & SRE Books 🤖 Xiaomi MiMo Platform
← Back to Home