RapidPure Filters Review

Transcription

1 RapidPure Filters Review Some filters leave a bad taste while others do an insufficient job of treating the water. The RapidPure filter changes the game and pushes the boundaries of water treatment. by Roger Caffin :00:00-06 Introduction and Background Filling a water bottle - but is it safe to drink? It isn't. The last time the National Parks people went through this valley, they took out about 500 wild pigs. Plus cows. Everything comes in cycles, and means for water treatment are no exception. We have had chemicals and pumped filters, UV and micro-tube filters, each of which has had problems. An extremely brief and probably highly biased review follows; the opinions expressed are the author's only. Please note that I am not bothering to mention anything which does not meet at least some EPA requirements - 'kitchen filters' generally do not. The chemicals - iodine, chlorine, chlorine dioxide for instance, often either smelt or tasted a bit off (the electrically powered MSR MIOX generator was utterly appalling), or required rather careful handling in the field, and many people simply do not like the idea of using chemicals anyhow. A major problem with chemical treatment is that it can take up to 4 hours (in cold water) for them to work on protozoa like Giardia L, and it is questionable whether they work at all on Cryptosporidium. In addition, many of the chemicals are not all that stable, so you have to keep buying fresh ones. Finally, some of the by-products of the chemicals are not all that safe for human consumption.

2 The pumped filters were all slow, even the classic Katadyn Hiker which I used for many years, and most of them blocked up fairly quickly. That meant you probably had to carry a spare filter element on any long trips. In addition, with one exception, pumped filters only ever blocked bacteria and protozoa (which are much larger) - they did not block viruses. And did I (we) get sick once from viruses in the water! (Bed-ridden for 24 hours, much of that unconscious.) The rather heavy Hiker was far and away the best of the lot, but even then I once managed to block one in the middle of an alpine trip, which was a bit inconvenient. Far worse was the Sweetwater pumped filter which blocked quickly and then sprayed water all over you: there was an over-pressure relief valve in the body which was very easy to trigger. The MSR Hyperflow, based on a mass of fine 'hollow fibers was much heralded, but turned out to be a fast route to failure as well: it blocked up and backflushing did little to fix that. MSR revised the design, but that did not help much: another failure. I could go on (and on), but I won't. I will just say that apart from the robust but heavy Hiker, no pumped filter I have ever tested has been anything more than a dismal failure. UV-C treatment works, but you are relying on batteries for the operation. The problem is that the only useful source of UV-C radiation at present is a Phillips germicidal lamp which draws a lot of power. Using AA Lithium batteries is (mostly) fine, but the very nice Steripen Adventurer Opti (my current favorite) uses CR123 batteries, which can be variable. Most of the rechargeable batteries (RCR123) were not able to supply the 1 amp needed, although the blue Tenergy ones could when new and if carefully maintained. Many of the non-rechargeable CR123 batteries sold cheaply on the web simply cannot supply that amount of current, and that has caused some public relations problems. Good brand-name CR123 batteries do work well, but are more expensive. (You get what you pay for.) There have been a few other brands of UV systems come and go on the market, but their design and performance has been lacking or worse (like a serious safety hazard). What we want are UV-C LEDs, but they are a 'future product', and will remain so for some years to come. Photo from A Survey of Water Hazards and Water Treatment Methods. More recently several companies have taken the 'hollow fiber' idea as originally developed for kidney dialysis and put them into filters which, while they need still some pressure to operate, do not need pumping. These typically come as 'straws' (which you suck through), squeeze bottles (more sucking), and bag systems. The latter use a bag of water hung up high, a longish hose down to a filter element to create some pressure, the filter itself, and a short hose off into a collection bottle. The big name in this area is Sawyer. Most of these will block bacteria and protozoa, but not viruses, although Sawyer does offer one unit which filters down to 0.02 microns and which passes EPA requirements. The problems with all these have been the speed of operation: the filtration rate is slow, even with regular cleaning by back-flushing. RapidPure Now we have a new competitor in the market: RapidPure, an off-shoot of Atlas Air and Water (Lakeland, MN). They have taken an idea originally used by General Ecology, with their First Need filter (unfortunately rather heavy), and also by Pure Hydration of the UK with their Aqua Pure Traveller systems (squeeze bottle). The idea is that you create a filter which is uses a special chemical treatment of the filter fabric which causes the surface to be charged, so that all the bits and pieces in the water are attracted to it. Some call this electro adhesion: it uses a form of electrostatic adhesion. The filter does not have to be very fine: I will hazard a guess here and suggest that the holes could easily be 10 microns or larger. Actually, I don't think that the filter has 'holes' per se: it's more of a labyrinth. Anyhow, what happens is that all the viruses, bacteria and small protozoa are attracted to the surface of the filter as the water goes through it, and they stick there. RapidPure claim: Electropositive fibrous media has high particle removal efficiency as well as 50 to 250 times greater dirt holding capacity than the competitors. RapidPure filters are also infused with silver. Silver infusion is a self-sanitizing agent that kills bacteria, preventing growth of captured bacteria.

3 Does this idea work? Yes, and independent lab testing has verified this for the different companies in this area. For RapidPure the figures are impressive, to say the least. For viruses the filtration is reported as being % effective, while the EPA requirement is 99.99%. For bacteria the reported figure is again %, which is the same as the EPA requirement. For protozoa the figure is %, while the EPA requirement is 99.9%. Why can't they get 100% reduction you might ask? Well, the scientific reality is that they cannot claim 100% without testing millions of gallons of water (us scientists are a bunch of pedants), but the lab test report did say 'No species were detected in the filter effluent for the duplicate samples analyzed.' Yes, that does mean that not one bacterium or one virus got through. Flow Rate and Pressure But if the holes are so large, do you need lots of pressure to force the water through the filter? No, and that is what RapidPure are offering: low-pressure but 'completely' reliable filtration. We cannot avoid making some comparisons here. Sawyer quotes their throughput figures at 20 psig (138 kpa), which is equivalent to 45' or 13.7 m of water. That is a high pressure - about 5 stories high. In some cases you can get that pressure, or try to get close, by squeezing the bag the water is in: the Sawyer Squeeze system. You will understand that using a suspended bag to provide the pressure is going to be a little slower, unless you have a very high cliff and a very long hose. The RapidPure flow rate figures are quoted at 2 psig (13.8 kpa) or 4.5' (1.37 m) of water pressure: one tenth of the Sawyer pressure. This is very easily achieved just by hanging the bag from a branch. I will add here that RapidPure also sell fittings for buckets, for use in the third world. Of course, the next two questions are how do the filtration rates really compare, and how quickly does the flow rate drop as liters of water go through. Before addressing those questions, we will take a look at the different products in the RapidPure range. Filter Elements The picture here shows the various filter elements. They come in different sizes and with different fitting for different applications. All of the filters use the same filter 'fabric'. Note that some of the filters are still in their wrappings. In all cases the water goes 'into' the filter from the outside. For dimensions, use the floor boards: they are 4.0" wide. Filter A This unit is designed to go into a bucket or a bag. The two white washers at the bottom provide the seal. A hose gets pushed over the spigot at the bottom. The bucket is suitable for fixed camps (large groups, Boy Scouts, etc) or the third world. The bag is discussed below. The filter element is the same as G below. RapidPure also sell this fitting without the filter: you put the filter at the end of a hose below the bucket. Filters B & C

4 These are the 'same' filter housings, but in different sizes. They are the 'Explorer' size, and are for use with bags. Unit B has an extension ring at the top and a 4.5" filter. Unit C was delivered with the same extension ring, but I removed it to see how it worked. A simple push and twist does the mating. It has a shorter 2.5" filter element in it. The black line visible at the junction in the housing is an O-ring for sealing. Weights are as follows: Filter B complete: 250 g (8.82 oz) Filter C complete: 163 g (5.75 oz) Filter C complete, wet: ~200 g (~7 oz) Filter B housing: 185 g (6.53 oz) Filter C housing: 125 g (4.41 oz) Filter B extension ring: 60 g (2.12 oz) Filter element 4.5", white: 64 g 2.26 oz) Filter element 4.5", grey: 51 g (1.80 oz) (see below re difference) Filter element 2.5", grey: 38 g (1.34 oz) Unit C has commercial screw/hose adapters at both ends for hoses; unit B also had two but I borrowed one adapter for use during testing. They are labeled '1/4', which I think means '1/4 BSP'. (I have yet to discover any form of logic behind the BSP sizes.) The hose fittings are about 10 mm clear of the barb. Teflon tape is needed on the screw parts for a good seal. Filters G & F These are the actual filter elements which go inside the housings B & C. You will notice two black O-rings at the spigots at the top of each one: that spigot goes into a socket at the base of the housing. You can see these O-rings in B & C if you look closely. The weights are given above. Filter D This is a 'Scout' housing with a 3.5" filter element inside it. It too has an extension ring at the top: if you remove that ring to get a lighter weight you need a 2.5" filter element. The 2.5" filter claims a flow rate of 1.2 L/minute. I do not know what the claimed filter rate for the 3.5" size is. The weights are as follows: Filter E Filter D complete, long: 90 g (3.17 oz) Filter D complete, long, wet: ~120 g (~4.2 oz) Filter D complete, short: 73 g (2.58 oz) Filter D housing with ring: 73 g (2.58 oz) Filter D housing, short: 56 g (1.98 oz) Filter D extension ring: 16 g (0.56 oz) Filter element 3.5", white, wet: ~30 g (~1 oz) Filter element 2.5", grey, dry: 17 g (0.60 oz) This is the housing which goes inside a bottle, which is shown later. You will note that there are holes up the side of the housing to let the water in, and the bottom end is not covered anyhow. Some of the literature calls this a Pioneer filter. Further details are given below. Filters H & I These are 4.5" and 3.5" filter elements: the right sizes for the Scout but with fittings for other uses. The screw fitting for example will fit a Seychelle bottle, replacing the Seychelle filter. That said, the filter media is all the same. Filter J This is a replacement filter for the Scout and the bottle unit E. It is size 2.5", which is supposed to have a flow rate of 1.2 L/minute. White and Grey filters I quote from RapidPure: "All RapidPure filters are made of the same media. The only difference in our filters is length and width." and "They are the same with one exception. The white ones have a 10 micron pre wrap. The grey ones have no pre wraps." If the water has lots of floaties, the 10 micron pre-wrap may be useful at preventing actual blockage. If you want more flow, consider a large filter element. Filter 'Systems': Bottle and Bag

5 This is the RapidPure drink bottle: 750 ml capacity with a 2.5" filter, weighing 250 g (8.8 oz) empty for the lot. The label on the bottle says 'BPA free'. The cap flips up when you depress the blue catch seen at the top, and the spout will pivot out. The spout has a silicone rubber sheath - presumably to prevent any plastic taste. When you suck water out of the bottle, a little one-way valve in the lid lets air back in. A cute and rather smart feature of the bottle is the silicone rubber 'boot' at the bottom of the bottle. This protects the bottle itself from getting scratched or 'dinged' by rocks, and also stops the bottle from sliding around a bit. The bottle is fairly rigid: it is not a 'squeeze' bottle by any means. On the other hand, I suspect it would take a little bit of rough treatment. While RapidPure supply silicone rubber or clear PVC beverage tubing for use with the bag, the tubing inside this bottle is a bit different. It is not flexible, so fitting it to the bottle or the filter requires heat. RapidPure suggest wrapping the bare hose in a wet paper towel and heating the lot in a microwave oven. I would have thought that just immersing the end of the hose in boiling water for a bit might have worked just as well (be careful). Being stiff, the hose does not let the filter rattle around inside the bottle. To replace the filter element inside the plastic housing is also tricky. There are two tabs at the bottom of the housing to prevent the filter from falling out. Well, that makes sure the filter can't let any dirty water bypass the filter into the hose anyhow. You have to squeeze the filter housing in the right way to get the tabs clear of the filter: the instructions give more details.

6 The picture here shows a filter unit below the RapidPure bag, but that's not quite how you would use it in the field. First of all, RapidPure provide two hoses: 600 mm and 1370 mm (4.5') long. The long hose should go between the bag and the filter to get as high a pressure as possible, while the short hose goes into your clean water container. I have used the short hose between the two bits just for the photo. The hoses are quite large diameter, and I can't help thinking that a thinner and lighter hose might be suitable for the smaller sizes of filters. At 1370 mm (4.5') the longer hose does give you the 2 psig operating pressure RapidPure use for their claims. That is not all that long, and it should be fairly easy to hang this hose high enough from a tree for this to work. In my testing I actually used a slightly shorter hose, purely for convenience. Second, the top of the bag is shown open in the photo. (Those are surgical forceps hooking it to the roof beam.) The top of the bag is a standard roll-top dry-bag arrangement, with D-rings at each end of the closure. You fill the bag, roll the top down, then tie the lot to a high branch via the D-rings. Weights are as follows: Bag with RapidConnect hose adapter: 183 g (6.46 oz) Hose clear PVC 1370 mm w tube clip: 95 g (3.35 oz) Hose clear PVC 600 mm: 42 g (1.48 oz) Filter unit w extension ring and & 4.5" filter element: 250 g (8.82 oz) Total: 570 g (20.1 oz) One would have to say that this is not super light - but it is very robust. (Actually, the bag looks a bit like an Ortlieb one.) A Sawyer 64 oz Squeeze bag with lid weighs under 40 g (1.41 oz), so clearly RapidPure need to work on their bag weights for the light-weight market. I suspect that being so new to the outdoors gear market, they still have a bit of learning to do here. Performance Testing

7 This is where it all gets interesting. The test set up is shown here. For the RapidPure filters I used a PET bottle with an adapter cap, 820 mm of 1/2" silicone hose (also supplied by RapidPure), and then the filter at the bottom. For the Sawyer filter also tested at the same time I used a Sawyer Squeeze bag, 550 mm of silicone hose (my stock), then the Sawyer Squeeze filter. Yes, I know the hose used on the Sawyer bag is shorter. Anyhow, the idea was to set the combination of bag, hose and filter into use and measure how long it took for a certain amount of water to get through. In the case of the Sawyer Squeeze I also tried squeezing the bag to see if I could get a faster flow. I will preempt all the results by saying right up front that every filter slowed down after a few liters of water had gone through. My tap water has a small amount of very fine algae in it, and the filters would probably last a bit longer with clear mountain water - but I think you will get this sort of algae from larger rivers. This means that a single 'headline' measure of time is not going to mean very much. Actual charts of flow rate are needed to make a good assessment, and you need to take into account the quality of the water you have as well. The Test Water

8 Photo from our review of the MSR Hyperflow Microfilter with Modified Cartridge The test water is known to have some contaminants in it, but I am not sure what they all are. The water comes from the tap in my kitchen, and is what is used right throughout the house except for drinking. It comes from an inlet about 2 m (6') below the surface of our fairly large dam (with wood ducks), is pumped up to the house through one filter, and then into the house through a finer filter. The filters mainly serve to block algae, not pathogens. I figure if my wife is happy to use it for all the washing (kitchen, washing machine and shower), it can't be too bad. It does have a little bit of very, fine algae in it and a little tannin, but so would any slowmoving larger river. In the photo here, the orange tub contains unfiltered tap water, while the bottle contain filtered water. There is not a lot of visible difference. Granted, a sparkling clear mountains stream will have a lot less algae, but what's the use of doing all the testing on really clear water? Measurements There was a distinct lack of automation for this work: pencil and paper were used, with a digital watch for time. For volume of water filtered I used an extremely convenient feature of the PET bottle shown in the photo. You will notice 4 distinct rings on the bottle: it turns out these are 250 ml apart. So the bottle was filled to just above the top ring, and the time the water level passed each ring was recorded. This gave 'Seconds per 1/4 L'. Multiple filter runs were done, one after the other, by refilling the PET bottle. A clamp was used on the bottom hose to stop the flow between refilling. In some of the graphs the total amount of water filtered is used on the horizontal axis. These figures are slightly wrong, as the small amount of water above the top ring is not counted in the measurement. However, there was never a lot of excess water above the top ring: maybe 50 ml. So do not ascribe any great accuracy to the data - but you won't need to. Just look at the trends instead. The term pres-corr'd is used on some of the graphs. This means the time was corrected for the drop in pressure due to the falling water level in the PET bottle, and to the nominal pressure quoted by the vendor. A linear scaling was used. That means all the RapidPure results were corrected to 4.5' or 2 psig, while the Sawyer results were corrected to 45' or 20 psig. The correction due to the change in level in the PET bottle is small, so you can ignore that if you wish. But do note the huge effect of rescaling the Sawyer results from 550 mm actual hose length to the nominal mm. This means that the pressure-corrected results for the Sawyer are really hypothetical. You should look at the 'Times per run' instead. Sawyer A single run of measurements were made on the Sawyer system just for comparison, with the 550 mm length of hose. The left hand graph shows the time recorded as the water level passed each ring, in hours:minutes:seconds format. That means it took just over 16 minutes to filter 1 L of water with that hose length. The right hand graph shows the hypothetical time to filter each 1/4 L volume if you could get the quoted pressure of 20 psi. You will see that the time taken increased steadily with the amount of water already filtered anyhow. So the last 1/4 l took longer than the first 1/4 L - which is typical. If you want to make a direct comparison with the RapidPure results, just multiply the Sawyer times by 10 (20 psi to 2 psi, linear scaling).

9 RapidPure We have several sets of results for the RapidPure filters, as they supplied several different sizes of filter. The format used for the results is the same as for the Sawyer., but the time scales will differ! Here we have the results for the smallest of the filter elements: the Scout 2.5" unit, at 17 g bare weight (73 g complete unit). Of course, bag and hose are extra weight, but what you choose for these is up to you. There are 5 runs shown: they happened one after the next. The left hand graph shows filtration times, with a 550 mm hose, of 5 to 10 minutes per liter. Not fast, but it is an awfully small filter. And note carefully that it only took 5 L of my dam water to double the filtration time. This shows what happens when you go from the Scout 2.5" unit to the Scout 3.5" unit (bare weight ~27 g, (90 g complete unit). Again, bag and hose are extra weight. There are 3 runs shown for this filter, and the story is roughly the same for the way filtration time increases as water is filtered. However, the filtration time drops significantly, to minutes per liter, with a 550 mm hose. It would be less with a longer hose of course. This is starting to be moderately acceptable. Changing from the Scout 2.5" unit to the much wider Explorer 2.5" unit gives a lot more surface area for the filter, and if there is one thing we have learnt from the

10 Katadyn Hiker, it's that it is all about surface area. The filtration time per liter with that short 550 mm hose has dropped to under 1 minute, and the change in flow rate per liter filtered has also dropped. Yes, the Explorer 2.5" unit is heavier, at 38 g bare, but maybe it is worth it. A pity the housing is so heavy, making a complete unit weigh 163 g (without hose or bag), but even so, the performance is rather good. Finally, changing from the Explorer 2.5" unit to the longer Explorer 4.5" unit gives a whole lot more surface area again. I have to mention here that recording the times for this unit was difficult: the water level was shooting down as fast as I could read my watch and write the times down. The water shot through the filter so fast that I did 8 runs, for a total of over 8 L. Yes, the flow rate still dropped as the liters went through, but at this filtration speed, would you notice? Summary RapidPure make two claims: that their filters will block all pathogens, and that their system is much faster than the Sawyer products at far more realistic pressures. I will have to rely on the independent test lab reports for the effectiveness of the filters as I cannot do these tests, but I see no reason to not believe them. And RapidPure are not the first company to employ this sort of technology: it is 'known stuff'. That leaves the filtration rates. The smallest filter, the 2.5" Scout one, might be OK for a drink bottle where the throughput is going to be small (which is what it is sold for), but I would not recommend it for filtering larger quantities with a bag - unless you know in advance that the water will always be clear or you have a fair bit of free time. But going up in size from that one, the filters start to be very attractive, with flow rates which get quite high. For instance, going up from the 2.5" unit to the 3.5" unit only costs an extra 10 g for the filter and an extra 16 g for the extension ring. I should add here that the results also suggest that the larger units will last a lot longer as well. I am less enthused about the weights of the RapidPure bag and the RapidPure bottle. The bag is a first generation attempt and, bluntly, far too big and heavy. But I am sure readers will be able to find substitutes which are much lighter. The bottle is also a bit too heavy, at 250 g, but it is quite robust. Clearly, progress in the field of water treatment has not stopped. Citation "RapidPure Filters Review," by Roger Caffin. BackpackingLight.com (ISSN ) :00: Related Articles M True Lemon and True Lime Spotlite Review By Roger Caffin on April 16, 2013 in Gear» SpotLite Reviews Outdoor Retailer Summer Market 2010: Day 1 Lightweight Gold!

11 By Will Rietveld on August 04, 2010 in Trends» Outdoor Retailer News M Make Your Own Gear - A Thumbscrew for the Steripen Adventurer By Roger Caffin on July 07, 2009 in Techniques» Make Your Own Gear Chlorine Dioxide No-Mix Water Treatment Solution Coming to US Outdoor Market By Ryan Jordan on November 02, 2005 in Trends» Commentary COPYRIGHT BEARTOOTH MOUNTAIN PRESS LLC U.S. Library of Congress Serial Registrations: BackpackingLight.com (electronic, ISSN ) & Backpacking Light Magazine (print, ISSN ) BACKPACKING LIGHT and the FEATHER/MOUNTAIN icon are registered trademarks of the Beartooth Media Group, Inc. Mobile Version