Berkey’s gone from most shelves. I break down gravity filters, pump systems, and what I’d trust for a family when municipal water stops — with real flow rates.
We may earn a commission when you buy through these links, at no extra cost to you. As an Amazon Associate I earn from qualifying purchases.
I was there doing emergency preparedness electrical work — restoring power to neighborhoods that had been dark for days. The water treatment plant was running on backup generators, barely. Boil water advisory was in effect for the whole county, and half the people I talked to were filling bottles from their garden hose like nothing happened.
Top PickOne guy had a Berkey on his kitchen counter. Said he had been using it for years. Smart move. Except the flow rate had slowed to a trickle and he had no idea when he last cleaned the elements. He was probably still getting decent filtration, but he did not actually know — and that is the problem with water systems. They do not blow up when they stop working. A membrane degrades silently. Carbon exhausts without any warning. An O-ring develops a hairline crack you cannot see. You keep drinking, thinking you are fine.
That trip changed how I think about water prep. Not which filter has the best spec sheet, but which setup keeps working when I cannot order replacement parts off Amazon in two days.
Before talking about filters, it helps to understand what you are filtering out. Most prepper content skips this part and jumps straight to products, but knowing the threats determines which filter you actually need — because no single filter handles everything.
| Contaminant | Examples | What Removes It | Found Where |
|---|---|---|---|
| Bacteria | E. coli, Salmonella, Cholera | Most filters (ceramic, hollow fiber, RO) | Any untreated surface water |
| Protozoa | Giardia, Cryptosporidium | Most filters (0.2 micron or smaller) | Streams, lakes, wells near agriculture |
| Viruses | Hepatitis A, Norovirus, Rotavirus | RO, UV, chemical treatment (chlorine dioxide), purifiers like Grayl/MSR Guardian | Sewage-contaminated water, flood runoff |
| Heavy Metals | Lead, arsenic, mercury | RO, ion exchange, some carbon blocks, some gravity filters (ProOne) | Old plumbing, industrial runoff, mining areas |
| Chemicals | Pesticides, herbicides, PFAS | Activated carbon (partial), RO (most effective) | Agricultural areas, industrial zones, some municipal water |
| Dissolved Solids / Salt | TDS, brackish water, minerals | RO only | Coastal areas, brackish wells, mineral-heavy groundwater |
The critical takeaway: most camping and backpacking filters do NOT remove viruses or chemicals. They handle bacteria and protozoa only. If you are filtering flood water, urban runoff (see our water purification methods guide), or anything downstream of agriculture or industry, you need either RO or a multi-stage setup with chemical treatment. Making a dangerous assumption about what your filter catches is worse than having no filter at all, because at least then you know to boil it.
I am going to break this down simply because the water filtration world is full of overcomplicated marketing. There are really only three approaches, and the smart play is knowing which one fits which situation.
RO is the heavy hitter. It strips out dissolved solids, heavy metals, PFAS, fluoride — contaminants that gravity filters and carbon cannot touch. If you are dealing with agricultural runoff or industrial contamination in your water source, RO is really your only option short of distillation.
The catch is that RO needs pressure, electricity, and replacement membranes. It also wastes water — some systems dump 3 to 5 gallons for every gallon of clean water produced. At home on city water, who cares. At the cabin filling from a well during a power outage? That waste ratio matters a lot. I have looked into putting one at the cabin but I keep coming back to the same problem: if the well pump is down, the RO is just a paperweight.
This is where my head is at for most scenarios. Slow? Yes. Simple? Extremely. No electricity, no pumps, no pressure. A ceramic element physically blocks bacteria and protozoa because they are too large to pass through the pores. Carbon handles chlorine, taste, and organic compounds. Gravity does all the work.
The gap with gravity systems is viruses. Ceramic pores are not small enough to catch them. That is why I always keep chlorine dioxide tablets as a backstop — a few bucks of chemical backup covers that gap with almost zero weight — Aquamira drops if you like liquid ClO2, or Potable Aqua ClO2 tablets if you want sealed tabs that can’t leak in a pack.
For hiking and overlanding, these are the move. Light, fast, and you can drink straight from a stream. But most of them only handle bacteria and protozoa — not viruses, not chemicals. They are a backup or a field tool, not a long-term solution by themselves.
| Filter | Type | Bacteria | Viruses | Metals | Chemicals | Certification | Price |
|---|---|---|---|---|---|---|---|
| SimPure Y7P-BW | Countertop RO | Yes | Yes (UV) | Yes | Yes | NSF/ANSI 58 | ~$400 |
| iSpring RCC7AK | Under-sink RO | Yes | Yes | Yes | Yes | NSF 58 | ~$220 |
| APEC ROES-50 | Under-sink RO | Yes | Yes | Yes | Yes | WQA Gold Seal | ~$200 |
| ITEHIL Portable RO | Battery RO | Yes | Yes | Yes | Yes | Mfr claims | ~$250 |
| ProOne Big+ | Gravity | Yes | No | 99.9% lead | Partial | IAPMO | ~$280 |
| British Berkefeld | Gravity | Yes | No | Some | Partial | NSF (cysts/bacteria) | ~$250 |
| Alexapure Pro | Gravity | Yes | No | 96.4% lead | Partial | Mfr claims | ~$250 |
| Sawyer Squeeze | Portable | Yes | No | No | No | EPA registered | ~$35 |
| Grayl GeoPress | Bottle purifier | Yes | Yes | Yes | Yes | NSF P231 + P244 | ~$90 |
| MSR Guardian | Pump purifier | Yes | Yes | No | No | NSF P248 | ~$350 |
| LifeSaver Jerrycan | Vehicle filter | Yes | Yes | No | No | Tested to NSF P231 | ~$330 |
Note: “Mfr claims” means the manufacturer publishes filtration data but it has not been independently verified by NSF, WQA, IAPMO, or another third-party lab. That does not mean the filter does not work — it means you are relying on their testing rather than an outside audit.
I am not running a lab. I do not have access to an electron microscope or a water testing facility. What I can do is cross-reference manufacturer claims against third-party certifications, read the actual test reports when they are publicly available, and combine that with hands-on field use where I have it. For products I have not personally used, I say so — I would rather be honest about that than pretend every product on this page has been through my personal gauntlet. The certifications column in the table above is probably the most important thing on this page, because it tells you which claims have been checked by someone other than the company selling you the filter.
Full disclosure: I do not have an RO system installed yet. I have been going back and forth on putting one at the cabin in PA, and I have spent way too many hours researching which ones make sense for someone who is not on city water. Here is where I have landed.
This is the one I keep coming back to for the cabin. Five stages including UV, which adds virus kill that most countertop RO units skip. The thing that sold me is the power draw — only 30 watts. My Jackery and a small solar panel could run this all day without breaking a sweat. It is NSF/ANSI 58 certified, so the filtration claims have been independently verified, not just pulled from a marketing deck.
The waste ratio is 4:1, meaning 4 gallons of clean water for every 1 gallon of waste. That is actually good for this price range — some budget units run 1:4 the other way. I would still collect the waste water for washing dishes or watering the garden. No reason to dump it.
If I were putting an RO in my house instead of the cabin, this would be it. Six-stage under-sink system with a remineralization filter that adds calcium and magnesium back after the membrane strips everything out. Best waste ratio I have found in the under-sink category at 1:2. NSF 58 certified. iSpring has been around long enough that you can find replacement filters everywhere, which matters when you are stocking spares.
The remineralization stage is not just about health — it makes the water taste normal again. Straight RO water tastes flat and weird. Around $220 installed. Annual filter cost runs $80-100.
Budget option. Made in the USA, WQA Gold Seal certified, been the top-selling under-sink RO on Amazon for years. No remineralization (add mineral drops separately), slower output at 50 GPD, but it just works. About $200. Stock two years of filter sets at $50-60 per year — that is cheap insurance to keep the system running through a supply chain disruption.
I have not used this one personally, but it is the only battery-powered portable RO that does not look like a gimmick. Dual 6,000mAh batteries, USB-C rechargeable, removes bacteria, viruses, heavy metals, and PFAS. The output is slow compared to a countertop unit, but for producing clean drinking water from a questionable source in the field, nothing else this size can match it. Around $250. If anyone has real field time with one of these, I would genuinely like to hear about it.
I will be honest: gravity filters are where my confidence is highest for long-term prep. No moving parts, no electricity, no proprietary fittings. When everything else fails, water and gravity still work.
This is what I recommend right now. IAPMO certified, 99.9%+ lead removal, 100% fluoride reduction, 5-year filter warranty. The ceramic elements are silver-infused and cleanable. I went with ProOne over Berkey because of the certification situation (more on that below), and because ProOne actually has their testing data available through IAPMO’s certification database.
Three-gallon capacity. Not the fastest — you fill the top chamber, go do something else, come back to clean water. But that is kind of the point. Set it and forget it.
The only gravity filter with actual NSF certification. The Sterasyl ceramic elements have been used by the British military and humanitarian organizations like Oxfam for decades — this is not a new product trying to find a market. The ceramic elements do not need priming, and they can be scrubbed clean and reused many times.
Where it falls short compared to ProOne: no fluoride reduction, and heavy metal removal is not as strong. But for bacteria and protozoa from surface water — creek, river, pond — the NSF certification means the claims are not just marketing copy. That matters to me. About $250.
Worth mentioning but not where I would put my first dollar. Faster flow rate than most gravity filters at about a gallon per hour, 5,000-gallon filter life, handles 200+ contaminants. The weak spot is lead removal at 96.4% versus ProOne’s 99.9%. Not bad, but if lead is in your water source, that gap matters over time. About $250. Good as a secondary unit at a different location.
I built one of these at the cabin last spring and it cost me under $80. Two food-grade 5-gallon buckets from Tractor Supply, a 10-inch ceramic candle filter element, a carbon block cartridge, some food-grade tubing, and fittings from Home Depot. Drilled the holes, set up the stages — sediment first, then ceramic, then carbon — and let gravity do the work.
Is it as refined as a ProOne? No. But every single component can be replaced with parts from a hardware store. No proprietary cartridges, no calling customer service, no waiting for shipping. If I am planning for a scenario where supply chains are broken, the DIY build is the one I trust most because I can fix it with what is available locally.
The key is staged filtration: sediment protects the ceramic (extends its life significantly), ceramic catches bacteria and protozoa, carbon handles taste and chemicals. Each stage protects the one downstream.
I carry this on every hike and every camping trip. Three ounces, $35, and it filters 100,000 gallons. I have used mine in the Pine Barrens, along the Delaware Water Gap, and at streams near the cabin in PA. I set it up as a gravity drip off a tree branch at camp — hang the pouch, let it drip into a bottle, go gather firewood. By the time I get back, I have clean water.
No virus removal. No chemical filtration. For streams and creeks in the northeastern US, that is not usually a concern — bacteria and protozoa are the primary threats. But I still keep Aquamira drops (or Potable Aqua ClO2 tablets) in my pack as a virus backstop, especially if I am pulling from stagnant water or an unfamiliar source. Total kit weight for the Sawyer plus drops: about 4 ounces.
The Grayl is the only bottle-style filter I know of that removes viruses without chemicals. Fill it from literally anything — river, puddle, you name it — press down, drink. Eight seconds for 24 ounces. The cartridge handles bacteria, viruses, heavy metals, pesticides, and microplastics.
The downside is cartridge life. You get about 65 gallons per cartridge (250 liters), which sounds like a lot until you are filling it 3-4 times a day for a family. At $25 per replacement cartridge, I keep 4 or 5 spares. That gives me roughly 1,000+ liters of virus-safe capacity for about $125 — still cheaper than a single gravity filter system. I think it is great for solo use but impractical for a group bigger than two.
Buy once, cry once. This is the only pump filter that removes viruses mechanically — no chemicals, no UV, just the pump. Self-cleaning on every stroke, so the flow rate stays consistent over the life of the cartridge. 10,000 liters per cartridge. Military and expedition teams use this in places where the water will genuinely make you sick.
At $350 and 17 ounces, it is the most expensive and heaviest portable option by a wide margin. I do not carry this hiking. But if I were building a serious overlanding kit for a trip out west, or prepping for a scenario where water sources are genuinely contaminated and not just silty creek water, this is the one I would want. You can read MSR’s testing documentation — they actually publish their methodology, which is more than most companies do.
This one I have not gotten my hands on yet, but the concept is smart: a jerrycan with a built-in filter. Fill from any freshwater source, pour clean water out the tap. 20,000-liter filter life. You are already carrying a water container on the rig, so making it do double duty as a filter eliminates a separate pump, hoses, and setup time. Mil-spec construction. The overlanding community seems to really like these, and a few guys I follow on YouTube have put serious miles on them.
My pack always has the Sawyer Squeeze, a small bottle of Aquamira drops, two Nalgene bottles, and a collapsible Platypus bladder. At the cabin I have the DIY bucket setup and a box of spare ceramic elements and carbon blocks. I keep meaning to pull the trigger on a ProOne for the house in NJ. Our survival water filter comparison has more options but have not done it yet — keep telling myself “next month.” The iSpring RCC7AK is on my shortlist for an under-sink install at home, and I am probably going to put the SimPure at the cabin once I finish wiring the solar setup there.
Spare O-rings live in a ziplock bag in my tool kit. They weigh nothing and they are the first thing to fail on a squeeze filter. I also keep one extra Sawyer as a complete backup — at $35 that is the cheapest insurance I own.
I cannot write about gravity filters without talking about Berkey. I know a lot of people have them and love them. I am not here to tell you to throw yours away. But if you are buying new today, here is what you should know:
In May 2023, the EPA classified the silver content in Black Berkey elements as a “pesticide device” under FIFRA and issued a stop-sale order. Berkey is fighting it in court, arguing that silver treatment is an industry-standard practice used by many filter manufacturers — which is technically true. ProOne, British Berkefeld, and others use silver in their ceramic elements too. The difference is that Berkey never went through EPA registration, and the EPA decided to enforce on them specifically.
In September 2026, a federal appeals court upheld the denial of Berkey’s request to block the order, so the stop-sale order still stands. Meanwhile, Berkey has never obtained NSF or ANSI certification for their filtration claims. The performance numbers on their marketing materials are from their own internal testing, not independently verified by a third-party lab. That does not mean the filters do not work — it means you are taking their word for it.
Replacement element availability is spotty. Some retailers have them, some do not, and there are reports they are no longer being manufactured. If you own a Berkey, my advice is to stockpile two extra sets of Black Berkey elements now while you still can. That is cheap insurance regardless of how the court case goes.
Nobody talks about this in prepper content and it drives me nuts. If you are in a situation where clean water is limited, you are going to be tempted to use untreated water for washing clothes and cleaning up. Bad idea, especially for anything that touches your skin directly — socks, underwear, wound dressings. Washing those in contaminated water just moves bacteria and fungi around, and then you are wearing it against your skin in heat and friction all day. That is how a minor skin issue turns into an infection that needs antibiotics you may not have.
Your water budget needs to account for hygiene, not just drinking. A gallon a day for drinking is the FEMA minimum, but realistically you need additional treated water for cooking, wound care, and personal hygiene. Our Water Storage Calculator factors in hygiene needs — plug in your group size and it will give you a real number instead of the generic “one gallon per person” that gets repeated everywhere.
Two types of activated carbon filters exist and most articles treat them as interchangeable. They are not.
Granular Activated Carbon (GAC) is loose granules in a cartridge. Water finds the path of least resistance and channels through the gaps between granules, bypassing some of the carbon entirely. It works for chlorine and taste, but the filtration is inconsistent. It is also more prone to bacterial growth in stagnant water pockets between uses.
Carbon block is carbon ground to powder and compressed into a solid block. Water is forced through the entire thing. Tighter filtration, better chemical removal, more consistent results. Slower flow rate, but for the stage that sits between your ceramic filter and your glass, consistency matters more than speed.
Use carbon block for drinking water. Save GAC for whole-house pre-treatment if you need it.
RO strips everything — including the calcium and magnesium that make water taste like water. The World Health Organization has noted that most essential minerals come from food, not water, so healthy adults with a normal diet are fine drinking RO water long-term. But in a prep scenario where food variety is limited, those minerals from water start to matter more. And straight RO water just tastes wrong.
Cheapest fix: Trace Minerals ConcenTrace drops. 20-40 drops per gallon, sourced from the Great Salt Lake, 72 trace minerals, one bottle treats roughly 192 gallons. Weighs nothing. That is what I keep in my kit.
For a permanent under-sink setup, a Waterdrop MNR35 inline cartridge adds minerals automatically for about $15 every 9-12 months. Standard 1/4-inch quick connect, takes two minutes to install.
Or skip the products entirely: LMNT has published free electrolyte recipes — a pinch of salt, some cream of tartar for potassium, and magnesium citrate powder in a liter of water. Costs pennies and stores indefinitely as dry ingredients.
At home (bug-in): Under-sink RO with two years of spare filters stocked in the basement. Mineral drops for taste. ProOne gravity filter as a backup that works when the power is out and the water pressure drops.
At the cabin: DIY bucket gravity system as the primary. SimPure countertop RO once the solar is wired up. Spare ceramic elements and carbon blocks in a tote. Aquamira drops.
On the rig: Sawyer Squeeze for quick water from streams. Chlorine dioxide tablets in the glove box. LifeSaver Jerrycan is on my wish list for the roof rack. If I were heading somewhere with known industrial contamination, I would bring the ITEHIL portable RO.
On foot with a pack: Sawyer Squeeze, Aquamira drops, collapsible bladder, one extra Nalgene. Total weight under a pound. If I needed guaranteed virus protection without chemical treatment, I would swap the Sawyer for the Grayl GeoPress and carry extra cartridges.
This is the part nobody writes about because it does not sell products. But filters fail, and if you do not know the signs, you are drinking water you think is safe.
The general rule: if you cannot remember when you last replaced or cleaned a filter component, it is probably overdue. I keep a piece of tape on each filter with the install date written in Sharpie. Low tech, works perfectly.
Yes. The WHO says most minerals come from food. If you are eating normally, RO water is fine indefinitely. Add mineral drops if the flat taste bothers you, or if your diet is limited (which it might be in a prep scenario).
There is no single best. Layer your filtration: ceramic for bacteria and protozoa, carbon for chemicals and taste, chemical treatment for viruses. If one stage fails, the others still provide meaningful protection. That redundancy is the whole point.
Yes. Two food-grade buckets, a ceramic candle filter, a carbon block cartridge, some tubing, and hardware store fittings. Under $80 total. I built mine in about an hour. The advantage over a commercial system is that every component is replaceable with non-proprietary parts.
Depends on your water quality. Ceramic elements are rated anywhere from 1,000 to 6,000 gallons. Dirtier source water clogs them faster, but you can scrub them clean and keep going. When they get too thin from scrubbing, replace them. Always have at least one spare set.
Different tools. RO removes dissolved solids, heavy metals, and PFAS that gravity filters physically cannot. Gravity filters work without electricity and waste no water. If you have power and plumbing, RO wins on filtration breadth. If you need resilience with zero infrastructure, gravity wins. Best setup uses both.