How gravity filters work
A gravity filter is two stacked chambers with filter elements between them. You pour water into the top, gravity pulls it through the elements over one to two hours, and filtered water collects in the bottom. No plumbing, no power, no water pressure — which is exactly why they appeal to renters, off-grid households, caravanners and anyone treating rainwater or bore water without a pump.
That slow flow is a genuine engineering advantage, not just a convenience compromise. Most filtration media work better with more contact time, and a plumbed under-sink filter gets a fraction of a second while a gravity element gets minutes. It is the reason gravity systems can use dense composite or ceramic elements that would choke a pressurised line, and why the category can credibly claim performance a jug filter cannot.
The elements themselves are usually one of three types: ceramic (a fired porous shell, often around 0.2–0.5 micron, sometimes silver-impregnated to inhibit bacterial growth in the pores), carbon composite (proprietary blended blocks, the format used by the best-known brands), or multi-stage stacks combining ceramic, carbon and ion-exchange resin in one cartridge.
What the technology can and can’t do
Judged on physics rather than branding, gravity filtration is capable of a lot — and structurally incapable of some things regardless of how good the element is.
| Target | Capable? | Why |
|---|---|---|
| Chlorine, taste, odour | Yes | Carbon plus long contact time is ideal for this |
| Sediment, turbidity | Yes | Sub-micron ceramic handles it well |
| Cryptosporidium, Giardia | Yes, if absolute rated | Requires an absolute 1 micron or finer rating — nominal is not enough |
| Bacteria | Plausible with fine ceramic | Depends entirely on verified pore size, not claims |
| Lead, heavy metals | Possible with the right media | Needs specific adsorptive media — and verification |
| Chloramine | Weak unless catalytic carbon | Standard carbon handles chloramine poorly, even with contact time |
| Fluoride | No, unless a dedicated add-on element | Carbon and ceramic do not remove fluoride |
| Dissolved salts, hardness, nitrates | No | Requires RO or ion exchange — not what these elements do |
| Viruses | Generally no | Too small for typical pore sizes; a specific claim needing specific proof |
The honest summary is that gravity filtration is a strong particulate and organic technology and a weak dissolved-solids one. That is not a flaw — it is the same trade-off every carbon-based system makes. The problem in this category is not the physics. It is verification.
The verification problem
Almost every filter technology has a certification path: NSF/ANSI 42 for taste and odour, 53 for health contaminants, 58 for reverse osmosis, 401 for emerging compounds, P473 for PFAS. Certification means an independent laboratory with no financial stake confirmed the product does what its maker says.
Large parts of the gravity segment sit outside that system. Instead, the standard practice is self-commissioned testing: the manufacturer pays a laboratory — often a genuinely accredited one — to run its product against a contaminant list of its own choosing, then publishes the results. Those numbers can be perfectly accurate and still not be equivalent to certification, because the manufacturer controlled what was tested, under what conditions, and what got published. Nobody audits the ongoing production, and nobody can pull the listing if a later batch performs differently.
The distinction to hold onto: “tested by an independent lab” is not the same as “independently certified”. The first describes who held the pipette. The second describes who chose the test, verified the result and remains accountable for it. Certification bodies publish searchable databases — if a product is genuinely listed, you can look it up yourself under the exact product name.
A case study in what that costs
The best-known gravity brand internationally, Berkey (made by New Millennium Concepts Ltd), illustrates the consequences of operating outside certification — and it is worth understanding because the same structure applies across much of the segment.
The company’s longstanding position has been that NSF certification is optional and poorly suited to gravity elements, and that its own third-party laboratory testing is more rigorous than the standards require. The consequences of that stance have accumulated: California and Iowa restricted sale of the systems because their state rules require independent third-party certification; a consumer class action filed in late 2022 alleges the elements do not perform as advertised; and in August 2023 the US EPA issued a stop-sale order, classifying the elements as pesticide devices under FIFRA because of their silver content. The company has litigated against the EPA over that classification, with one case closed and one still live as of early 2026. In November 2025 it introduced a replacement element line with NSF certification pending — which, if completed, would be the first in the family to carry independent validation.
Two clarifications, because this story is routinely distorted in both directions. The EPA action was a regulatory classification dispute about silver, not a finding that the filters are unsafe or ineffective — and it was a stop-sale, not a recall. Equally, the absence of certification is not proof the elements underperform; it means no independent body has verified either way. Both of those statements are true at once, and anyone telling you only one of them is selling something.
What this means in Australia
There is a regulatory gap here worth knowing about. WaterMark — Australia’s mandatory certification scheme for plumbing products — does not apply to gravity filters, because they are not connected to the plumbing. A benchtop gravity unit sits entirely outside the scheme that governs every under-sink and whole-home filter sold here.
That is not a scandal; it is a sensible scope decision, since WaterMark is about plumbing safety rather than contaminant performance anyway. But it means the one mandatory checkpoint most Australians know about simply isn’t operating in this aisle. Performance claims for gravity filters rest entirely on voluntary certification — and Australian practice has been converging on NSF/ANSI standards for exactly that purpose, with AS 3497:2021 dropping its reference to the older AS/NZS 4348 protocol in favour of NSF/ANSI certification for performance claims.
The practical upshot for an Australian buyer: NSF/ANSI listings are the only verification layer available in this category, and a meaningful share of budget gravity benchtops sold here carry none at all — only brand claims. Check the specific product name in NSF’s public database rather than trusting a certification logo on a product page, and treat “lab tested” language as the beginning of a question, not the end of one.
Our verdict on the evidence
The technology is sound; the verification culture is the problem. Gravity filtration’s long contact time is a real advantage, and there is nothing implausible about a well-made ceramic or carbon composite element removing chlorine, sediment and protozoan cysts effectively. For rainwater, bore water, off-grid living or rentals where plumbing isn’t an option, it is often the most practical technology available.
What you cannot do in much of this category is verify that the specific element in front of you performs as claimed — and that matters more here than in most, because these filters are frequently bought precisely for emergency, untreated or contaminated-supply use, where being wrong has consequences. Our position: gravity filters are a legitimate choice, and within the category, choose one whose performance claims you can look up in an independent certification database under the product’s exact name. Where that isn’t available, treat the published numbers as the manufacturer’s best case rather than a verified floor, and size your expectations accordingly — especially for anything you are relying on for microbiological safety.
Primary sources: US EPA stop-sale order and FIFRA classification of gravity filter elements (August 2023) and subsequent litigation reporting; consumer class action filings (2022–); California and Iowa third-party certification requirements; Australian Building Codes Board WaterMark scheme scope; Standards Australia AS 3497:2021 and its removal of AS/NZS 4348 references in favour of NSF/ANSI performance certification; NSW Health guidance on absolute 1 micron filtration for Cryptosporidium and Giardia. Reviewed August 2026. FilterOut sells nothing and takes no commissions — see who to trust for filter advice.
FAQ
Are gravity water filters any good?
The technology is sound — slow flow gives long contact time, which suits carbon and fine ceramic media well. They handle chlorine, sediment and, if absolute rated at 1 micron or finer, protozoan cysts. They do not remove fluoride, hardness, nitrates or dissolved salts.
Do gravity filters need WaterMark certification in Australia?
No. WaterMark applies to plumbed-in products, so benchtop gravity filters fall outside it entirely. That makes voluntary NSF/ANSI certification the only independent verification available — and many budget gravity units carry none.
Is “independently lab tested” the same as certified?
No. Self-commissioned testing means the manufacturer chose the lab, the contaminants, the conditions and what to publish. Certification means an independent body set the protocol, verified the result, audits production and can withdraw the listing.