Tetra, Eheim and Fluval sell quiet without a number
Tetra's Whisper pages explain how the design flattens sound wave frequencies. They give no decibel figure for any size. Eheim's manual keeps noise to the level of a whisper, again with no figure, and Fluval's 07 canister series claims to be up to 25 percent quieter than the previous generation, a comparison with itself that converts to nothing. OASE we could not fully check, and where we could, we found nothing.
On the other side stand hygger, AQQA, Pawfly, HITOP and Uniclife. hygger prints a figure on every product page; the other four print theirs on Amazon storefronts. The result is a table with the century-old aquarium names missing from every row.
A missing number does not prove a pump is loud. Still, a category that has sold silence for decades without measuring it deserves a raised eyebrow.
The litter box and massage gun guides ran into the same pattern, the best-known brand selling a trademark instead of a figure. Pet water fountains split the same way too: the pet fountain guide turned up no figure from Catit or PetSafe, while newer brands print numbers like 20 dB and 23 dB with no distance attached.
Why is an aquarium air pump loud?
Because most of them drive a rubber diaphragm with an electromagnet. The magnet pushes the diaphragm back and forth, the whole assembly vibrates, and the vibration grows as the rubber stiffens with age.
Piezoelectric pumps throw all of that out. A thin ceramic plate flexes when voltage crosses it, and that flexing moves the air: no motor, no diaphragm, almost nothing in motion. The claims follow the mechanism, with hygger's MiniPro Clip at a 33 dB ceiling and Pawfly's mini at 30 dB, both drawing around 1.5 W.
What you give up is airflow.
A few hundred millilitres a minute confines a piezo pump to tanks of roughly 15 gallons and under. For a nano tank beside a bed, it is the technology to buy. For a stocked 75 gallon, it is not an option at all.
Piston pumps sit at the loud end
At the far end from piezo sits the piston pump, and the maker of the one here is blunt about it. hygger's Submarine page gives 39 to 48 dB, then adds "Not suitable for bedroom use".
Pistons vibrate hardest of the three designs, and that vibration travels into whatever the pump stands on. A brand warning buyers off a product on that product's own page is the most credible sentence in this survey, which is how the Submarine got its slot: the loud option that admits it.
Is there a quiet aquarium filter with a decibel rating?
One, and it is hygger's: the AquaFall at below 30 dB, with hygger's internal 4-in-1 filter under 35 as the only other figure. Filters hum too, and on the established brands' pages that hum is described only in adjectives.
The AquaFall's claim arrives with its mechanism. Its pump sits apart from the filter body, so the box cannot act as a sounding board. That leaves this guide with one ranked filter and a plain note for anyone shopping for a canister from Eheim or Fluval: you are buying on reputation, since neither brand gives you a number to hold it to.
The aquarium filter guide cast a wider net. It found figures on hang-on-back and internal filters from hygger and AQQA, still none from any canister maker it checked, and one filter figure it singles out as the most useful, AQQA's about 30 dB with the outlet under water against about 50 dB above it.
Read each claim by its shape
No claim in this category comes with a test standard or a measuring distance. What the claims do have is a shape, and the shape tells you something.
Uniclife publishes a pair, 25 dB at low flow and a stated maximum of 45, along with an admission that its measuring equipment is imperfect. hygger publishes a ceiling for each product, plus one range with a warning attached. Pawfly stamps under 45 dB across its whole standard line. AQQA manages two different figures on one page, and both are quoted here. Pawfly's under 45 dB and Uniclife's stated maximum of 45 look alike, yet with no distance or standard behind either they cannot sit side by side, because ratings with no stated test conditions do not line up.
A pair beats a ceiling, a ceiling beats a bare number, and a warning beats all three.
