Not long ago and late in the year 2021, I became curious just how cheaply a water filter could be built at a time of both skyrocketing prices and scarcity of rafting gear. To be sure, there are plenty of comparable prebuilt and pricey water filters out there, but it occurred to me the process of building a functional water filter from basic parts could prove insightful for some and maybe curate confidence in others pondering just how exactly to create clean drinking water on river trips for a lot of people. Considering the necessity of doing this on longer ones, and that healthy/safety is paramount to enjoying them, I put this together to hopefully go beyond just anecdotes about best practices to craft some form of a knowledge base. As a heads-up this is only really about disinfecting water for drinking from a relatively pristine surface source, and not addressing or acknowledging other potential impairments as previously discussed.
Power & Pump:
The first piece of equipment supporting this project, albeit unknowingly at the time, was a handy little lithium power station as well as a versatile solar panels on sale for the holidays. From there, the next piece procured was through another deal found on eBay; a new Whale submersible pump kit that was nicely prewired with a 12v cigarette lighter port and some other cool accessories for only $40. Score! There were other pumps out there I was also considering; like this powerful one from Joolca, as well as the Whale Babyfoot pump as a potentially a battery-free option or backup.
![Image]()
Filtration:
Next came a lot of hemming and hawing about filter selection. I deemed it important to have a two-stage filter, since removing larger particles first increases overall efficiency and flow by helping avoid saturating the small pores that must keep the nasty biologicals out. Here’s an excellent opportunity to connect this to the importance of flocculation before filtration either by adding a couple dashes of (completely safe) alum or some PACL (Water Wizard) and letting it sit for a bit…as even the clearest water has ample suspended particulates to impede your filtering and silty rivers will do so with the quickness. For this stage, I decided on a pleated design (folded for max surface area) made from polyester. There are less expensive cellulose (paper) options too, but those can become a growth medium for microbes and are prone to tear over time. As an aside, from personal experience using the spendy Millipore ones in the lab for tangential flow ultrafiltration, I learned cellulose filters can actually be stored/preserved using a weak phosphoric acid solution. But for this purpose, let's just concede these relatively inexpensive pleated poly sed filters as consumable items for every trip and to always have two new ones on hand (one to use, one spare). After pouring over some boring technical data, I settled on the 5-micron nominal (more here about nominal vs. absolute) size for the best particle retention distribution relative to flow and pressure drop (ΔP), additionally accounting for the limit of the Whale pump to “push” water. So the general rule of thumb is to make your filter pore sizes iteratively smaller and ideally decrease them by no more than an order of magnitude at each step.
A bit more thought went into the 2nd filter stage since it needed to balance critical exclusion of microorganisms while maintaining a decent flow. There are many opinions out there as well, all which have different strengths and weaknesses…but I was already set on a Stersyl (ceramic) design, because they progressively exclude large particulates at the surface to smaller ones as water sinuously transits the depth (3d) of the material, can be scrubbed and backflushed for reuse, and are impregnated with silver to prevent unwanted microbial growth. Here’s a bit more info about ceramic filters, but the brass tacks is I decided against more complex offerings with multiple stages or mediums (carbon, resin, etc) because of their impact on pressure/flow. Keeping this in mind, I also considered a single candle against a multi-candle design with a larger surface area. Again, surface area of a filter relates to the amount of void available for water to pass through; therefore more area equates to more pores, and likely less resistance (back pressure) along with increased capture. To those curious, basic math can perhaps articulate this point more clearly by calculating and comparing the surface area of one large cylinder versus six smaller ones within the same volume. Anyhow, I ended up on a Doulton Rio 2000 multi-candle filter for the above reasons and admittedly again because I found an unbeatable deal on eBay for a brand new (deadstock) filter set including the manifold and scour brush for only $70.
![Image]()
Housing & Fiddly Bits:
I probably spent far more time than necessary looking at double Big Blue filter housings, as one seems almost good as another because the standard sizes. At first, I thought it would be aesthetically pleasing to have transparent housings to see everything at work. But those were oddly expensive and then came across this one for a cool $81 that has two O-rings per stage, thinking that two were probably better than just one in preventing leaks. Though I did buy extra upper and lower o-rings just in case of a failure from over torquing, etc. Didn't much care for the white bracket it came with though, as it was heavy and unwieldy; so disassembled it all, added a brass connector between the housings for strength, and screwed them into a nice light piece of wood that had sealed with spar urethane.
In terms of tubing, fittings, and other fiddly bits for assembly, I’m not going to go into much details here beyond the basics. Besides, these are parts available at a local hardware/aquarium/brew store and so don’t want to suggest slavishly looking for stuff with ample substitutes (brass, steel, nylon, etc). But keep in mind it’s important to matching up the bore of your (food-grade) hoses to the pump barb (10-13mm) as well as the barb on the fitting you get for the 1” NPT inlet/outlet on the filter housing. Since I just used the hose that came with the Whale RV kit, that made things easier and went with whatever fittings had a 1” male NPT for the housing inlet and a reducing barb to fit that hose. The outlet side of the filter housing was a bit more complicated for me, as needed an elbow and another connector to thread on my UV unit. I guess my last suggestion is to avoid wrenching on any fitting really tight (esp if plastic) with tools, instead just wrap any fitting threads well with (blue roll) teflon tape to help prevent leaks and tighten as much as possible with a strong hand.
![Image]()
![Image]()
Disinfection/Sterilizing:
Here there’s a bit of the “secret sauce” to any water filter as the final polishing step to ensure it’s free of pathogens. Although the foolish might argue this last step is unnecessary because they don’t believe viruses exist or heed the insight from smug experts/scientists/doctors. Regardless, they are obviously quite real, ubiquitous to natural waters (notwithstanding the ongoing global COVID-19 Pandemic), and remain an existential threat to health on a river as well (Case in Point: the Grand Canyon is currently briming with norovirus outbreaks). Their tiny sizes (<0.1 micron) make them particularly pernicious since they can pass right through standard filtration for biologicals, unless you’re going down smaller (which I promise produces an untenably slow flow). Virus are also highly transmissible, so its good to equally consider the role of proper hygene in keeping water clean for drinking, with common sense stuff like hand washing and conscientious handling of storage container to prevent cross-contamination, etc etc.
The most idiot-proof strategy to disinfect water and render viruses inert is to add the EPA-prescribed amount of bleach or another halide like iodine to chemically oxidize their dna/rna thus preventing them from infecting you. Another more involved and interesting method is to utilize UV light to photo oxidize that same genetic material. Either way will also destroy any microorganisms still present and/or their reproductive capacity via the same pathway of denaturing organic acids and proteins. However, I absolutely would not recommend skipping filtration entirely in favor of only a UV or bleach unless you’re simply trying to make water clean enough for say doing dishes or a handwashing station (which even then would still flocc first). The reasoning being that 1) the reactive component of bleach (free chlorine, which starts degrading as soon as a bottle is opened, so always buy a fresh one before a trip) can become neutralized by interacting first with suspended materials or solutes present and 2) turbid water could scatter UV light to reduce its disinfection capacity.
I should probably now predicate all this info on the whether you're inclined to follow the rambling advice of a true world-class idiot or someone accredited by education, as these methods can present a burn or blinding hazard if not performed with a modicum of awareness. Because I detest both the taste of chlorine and dangerous disinformation, my preference is using UV in a safely directed manner and dose. But that leans on 3 earned degrees and a relevant research background, which has further motivated me towards a novel sterilizer solution that’s low power (12v) and (ruggedly) LED-based. Once import restriction relax a bit, I’m hoping to get more of them manufactured(by a secret lab in China) and make them available to other science-friendly folks on here. But in the meantime, there are a few other options available that provide the required UV intensity/dose for the flow rates allowed by this setup for less than $100 that am happy to share to those that ask via DM. Although have decided it’s my prerogative to not openly share such heady details with the cranks and cretins that spent this pandemic breathlessly wailing on here against science, masks, public health mandates, or the kind of prodigiously published researchers and luminaries I hold in the highest regard (like Dr. Fauci).
![Image]()
Filter setup with a store-bought UV unit
![Image]()
Same setup with my (redacted) UV-LED unit and the shrink wrapped 12V DC driver
Final Thoughts:
I clearly did not reinvent the wheel on this whole effort, and it was mostly guided by an informed evaluation of cost & benefit. Accordingly, I am well aware of the overlap to companies selling existing products that utilize some of these components since are proven robust and reliable. It’s kinda interesting to me though that prices of these discrete pieces did not increase over the course of this project (actually with the filter housings, most went down), while the cost of some turnkey product built from them appear to have jumped significantly. Therefore, I figure folks might now-more-than-ever appreciate a lower cost Do-It-Yourself approach along with some (hopefully) clear explanations in consort to other posts I’ve made about considerations for creating drinking water from natural sources. But if this all seems only that much more daunting or confusing, I can appreciate the comfort from buying something prebuilt. Peace of mind tends to come at considerable expense though, so it’s worth stating that the fundamental processes of water filtration do not vary so wildly that its worth a premium for any specific brand. There is also the less tangible benefit of DIY in that it instills familiarity of how everything works, so diagnosing problems say out there on a river trip becomes easier than with a blackbox product.
Here's a few pics of what everything ended up looking like all packed up, so this doesn't seem all so imaginary. Everything fit perfectly into this svelte Pelican case (19x14x8), although am sure it would go into a ammo can as well. It's pictured here with the store-bought UV setup (w/ a more delicate mercury bulb) and some foam pieces added for cushioning to hopefully prevent a big drop/shock from breaking it, but if that happens can just swap in UV-LED unit.
![Image]()
![Image]()
![Image]()
As a parting word of advice, once assembled I suggest testing your own whole setup at home first, and consider timing how long it takes to run a gallon or two of distilled water through it to get a general idea of flow rate (super helpful for later identifying when to clean/replace filters), and then of course disassemble/dry everything for storage when not in use. So while I built this whole shebang expressly for bottom dollar….I will acknowledge to already owning a fancy Katadyn Expedition water filter that has now been relegated to a backup system that could still be plumbed to either UV unit if needed.
Disclaimer:
Please feel free to disregard and dismiss any/all this advice. It makes absolutely no different to me, as neither care about credit or pretty criticism. Who knows, maybe the rubes and right-wingers savor more their precious freedumbs to drink water deliciously untreated and have never once gotten sick (GFY). Whereas other low IQ ilk might feel it’s more of a personal decision whether to swallow all this and the underpinning germ theory, because otherwise would require acknowledging Science a wonderful thing that serves us all sans any bias or political agenda. Then again, perhaps I’m just some preachy apostate or one of those "Sheeple" peddling assertions based on unassailable evidence from credible sources proven to protect public health….baaaaaaaah I digress.
Power & Pump:
The first piece of equipment supporting this project, albeit unknowingly at the time, was a handy little lithium power station as well as a versatile solar panels on sale for the holidays. From there, the next piece procured was through another deal found on eBay; a new Whale submersible pump kit that was nicely prewired with a 12v cigarette lighter port and some other cool accessories for only $40. Score! There were other pumps out there I was also considering; like this powerful one from Joolca, as well as the Whale Babyfoot pump as a potentially a battery-free option or backup.
Filtration:
Next came a lot of hemming and hawing about filter selection. I deemed it important to have a two-stage filter, since removing larger particles first increases overall efficiency and flow by helping avoid saturating the small pores that must keep the nasty biologicals out. Here’s an excellent opportunity to connect this to the importance of flocculation before filtration either by adding a couple dashes of (completely safe) alum or some PACL (Water Wizard) and letting it sit for a bit…as even the clearest water has ample suspended particulates to impede your filtering and silty rivers will do so with the quickness. For this stage, I decided on a pleated design (folded for max surface area) made from polyester. There are less expensive cellulose (paper) options too, but those can become a growth medium for microbes and are prone to tear over time. As an aside, from personal experience using the spendy Millipore ones in the lab for tangential flow ultrafiltration, I learned cellulose filters can actually be stored/preserved using a weak phosphoric acid solution. But for this purpose, let's just concede these relatively inexpensive pleated poly sed filters as consumable items for every trip and to always have two new ones on hand (one to use, one spare). After pouring over some boring technical data, I settled on the 5-micron nominal (more here about nominal vs. absolute) size for the best particle retention distribution relative to flow and pressure drop (ΔP), additionally accounting for the limit of the Whale pump to “push” water. So the general rule of thumb is to make your filter pore sizes iteratively smaller and ideally decrease them by no more than an order of magnitude at each step.
A bit more thought went into the 2nd filter stage since it needed to balance critical exclusion of microorganisms while maintaining a decent flow. There are many opinions out there as well, all which have different strengths and weaknesses…but I was already set on a Stersyl (ceramic) design, because they progressively exclude large particulates at the surface to smaller ones as water sinuously transits the depth (3d) of the material, can be scrubbed and backflushed for reuse, and are impregnated with silver to prevent unwanted microbial growth. Here’s a bit more info about ceramic filters, but the brass tacks is I decided against more complex offerings with multiple stages or mediums (carbon, resin, etc) because of their impact on pressure/flow. Keeping this in mind, I also considered a single candle against a multi-candle design with a larger surface area. Again, surface area of a filter relates to the amount of void available for water to pass through; therefore more area equates to more pores, and likely less resistance (back pressure) along with increased capture. To those curious, basic math can perhaps articulate this point more clearly by calculating and comparing the surface area of one large cylinder versus six smaller ones within the same volume. Anyhow, I ended up on a Doulton Rio 2000 multi-candle filter for the above reasons and admittedly again because I found an unbeatable deal on eBay for a brand new (deadstock) filter set including the manifold and scour brush for only $70.
Housing & Fiddly Bits:
I probably spent far more time than necessary looking at double Big Blue filter housings, as one seems almost good as another because the standard sizes. At first, I thought it would be aesthetically pleasing to have transparent housings to see everything at work. But those were oddly expensive and then came across this one for a cool $81 that has two O-rings per stage, thinking that two were probably better than just one in preventing leaks. Though I did buy extra upper and lower o-rings just in case of a failure from over torquing, etc. Didn't much care for the white bracket it came with though, as it was heavy and unwieldy; so disassembled it all, added a brass connector between the housings for strength, and screwed them into a nice light piece of wood that had sealed with spar urethane.
In terms of tubing, fittings, and other fiddly bits for assembly, I’m not going to go into much details here beyond the basics. Besides, these are parts available at a local hardware/aquarium/brew store and so don’t want to suggest slavishly looking for stuff with ample substitutes (brass, steel, nylon, etc). But keep in mind it’s important to matching up the bore of your (food-grade) hoses to the pump barb (10-13mm) as well as the barb on the fitting you get for the 1” NPT inlet/outlet on the filter housing. Since I just used the hose that came with the Whale RV kit, that made things easier and went with whatever fittings had a 1” male NPT for the housing inlet and a reducing barb to fit that hose. The outlet side of the filter housing was a bit more complicated for me, as needed an elbow and another connector to thread on my UV unit. I guess my last suggestion is to avoid wrenching on any fitting really tight (esp if plastic) with tools, instead just wrap any fitting threads well with (blue roll) teflon tape to help prevent leaks and tighten as much as possible with a strong hand.
Disinfection/Sterilizing:
Here there’s a bit of the “secret sauce” to any water filter as the final polishing step to ensure it’s free of pathogens. Although the foolish might argue this last step is unnecessary because they don’t believe viruses exist or heed the insight from smug experts/scientists/doctors. Regardless, they are obviously quite real, ubiquitous to natural waters (notwithstanding the ongoing global COVID-19 Pandemic), and remain an existential threat to health on a river as well (Case in Point: the Grand Canyon is currently briming with norovirus outbreaks). Their tiny sizes (<0.1 micron) make them particularly pernicious since they can pass right through standard filtration for biologicals, unless you’re going down smaller (which I promise produces an untenably slow flow). Virus are also highly transmissible, so its good to equally consider the role of proper hygene in keeping water clean for drinking, with common sense stuff like hand washing and conscientious handling of storage container to prevent cross-contamination, etc etc.
The most idiot-proof strategy to disinfect water and render viruses inert is to add the EPA-prescribed amount of bleach or another halide like iodine to chemically oxidize their dna/rna thus preventing them from infecting you. Another more involved and interesting method is to utilize UV light to photo oxidize that same genetic material. Either way will also destroy any microorganisms still present and/or their reproductive capacity via the same pathway of denaturing organic acids and proteins. However, I absolutely would not recommend skipping filtration entirely in favor of only a UV or bleach unless you’re simply trying to make water clean enough for say doing dishes or a handwashing station (which even then would still flocc first). The reasoning being that 1) the reactive component of bleach (free chlorine, which starts degrading as soon as a bottle is opened, so always buy a fresh one before a trip) can become neutralized by interacting first with suspended materials or solutes present and 2) turbid water could scatter UV light to reduce its disinfection capacity.
I should probably now predicate all this info on the whether you're inclined to follow the rambling advice of a true world-class idiot or someone accredited by education, as these methods can present a burn or blinding hazard if not performed with a modicum of awareness. Because I detest both the taste of chlorine and dangerous disinformation, my preference is using UV in a safely directed manner and dose. But that leans on 3 earned degrees and a relevant research background, which has further motivated me towards a novel sterilizer solution that’s low power (12v) and (ruggedly) LED-based. Once import restriction relax a bit, I’m hoping to get more of them manufactured
Filter setup with a store-bought UV unit
Same setup with my (redacted) UV-LED unit and the shrink wrapped 12V DC driver
Final Thoughts:
I clearly did not reinvent the wheel on this whole effort, and it was mostly guided by an informed evaluation of cost & benefit. Accordingly, I am well aware of the overlap to companies selling existing products that utilize some of these components since are proven robust and reliable. It’s kinda interesting to me though that prices of these discrete pieces did not increase over the course of this project (actually with the filter housings, most went down), while the cost of some turnkey product built from them appear to have jumped significantly. Therefore, I figure folks might now-more-than-ever appreciate a lower cost Do-It-Yourself approach along with some (hopefully) clear explanations in consort to other posts I’ve made about considerations for creating drinking water from natural sources. But if this all seems only that much more daunting or confusing, I can appreciate the comfort from buying something prebuilt. Peace of mind tends to come at considerable expense though, so it’s worth stating that the fundamental processes of water filtration do not vary so wildly that its worth a premium for any specific brand. There is also the less tangible benefit of DIY in that it instills familiarity of how everything works, so diagnosing problems say out there on a river trip becomes easier than with a blackbox product.
Here's a few pics of what everything ended up looking like all packed up, so this doesn't seem all so imaginary. Everything fit perfectly into this svelte Pelican case (19x14x8), although am sure it would go into a ammo can as well. It's pictured here with the store-bought UV setup (w/ a more delicate mercury bulb) and some foam pieces added for cushioning to hopefully prevent a big drop/shock from breaking it, but if that happens can just swap in UV-LED unit.
As a parting word of advice, once assembled I suggest testing your own whole setup at home first, and consider timing how long it takes to run a gallon or two of distilled water through it to get a general idea of flow rate (super helpful for later identifying when to clean/replace filters), and then of course disassemble/dry everything for storage when not in use. So while I built this whole shebang expressly for bottom dollar….I will acknowledge to already owning a fancy Katadyn Expedition water filter that has now been relegated to a backup system that could still be plumbed to either UV unit if needed.
Disclaimer:
Please feel free to disregard and dismiss any/all this advice. It makes absolutely no different to me, as neither care about credit or pretty criticism. Who knows, maybe the rubes and right-wingers savor more their precious freedumbs to drink water deliciously untreated and have never once gotten sick (GFY). Whereas other low IQ ilk might feel it’s more of a personal decision whether to swallow all this and the underpinning germ theory, because otherwise would require acknowledging Science a wonderful thing that serves us all sans any bias or political agenda. Then again, perhaps I’m just some preachy apostate or one of those "Sheeple" peddling assertions based on unassailable evidence from credible sources proven to protect public health….baaaaaaaah I digress.
