Depending on one's locale, arguably one of the most challenging problems to deal with in making a home setup is finding, setting up, and maintaining a water filtration system. While it's 'known' what ideal water is (~150 ppm TDS, ~50 ppm alkalinity, ~100 ppm general hardness), getting there can be expensive and/or inconvenient ... and even this hardness level, while considered ideal for espresso taste and minimizing scale in brew boilers, will cause scale buildup in steam boilers where heated to much higher temperatures.

I could go on, but the issues at hand are well-documented (if slightly outdated ). You can find a number of water filtration solutions on this very site, such as classic salt-charged resin-bed softeners like Chris' Coffee's commercial softening system and even more classic DVA softener that's bundled with many commercial espresso machines (including the Speedster! ). The trouble with these is that (a) they replace calcium in your water with sodium, which is generally considered to be not-too-great for coffee extraction--not that it likely matters given the ion exchange concentrations, and (b) they don't reduce alkalinity at all, meaning that you're stuck with what you've got there (see the Water FAQ for more on the relationship between alkalinity and scale buildup). Another typical solution is an RO system, but these are typically expensive to maintain (membrane replacement cost), and remove far too much from the water to be ideal for coffee. There are a few solutions out there, such as a calcite filter and a bypass system, but these have issues of their own (which I don't care to get into here).

One potentially awesome setup for a pourover user that's fairly new would be to combine a ZeroWater filter with Cirqua's AB Formula packets to produce "ideal" coffee for your espresso machine (and flush your steam boiler with unformulated water). However, this option isn't viable for those looking to plumb in, and is arguably rather expensive given the cost of the packets and ZeroWater filters.

I've been rather idly looking for a solution that's superior to my current setup, which is simply a DVA softener and a high-grade 0.5 micron carbon filter. Some time ago, Mark Hoy brought up the possibility of the Everpure ESO softening filter system, which has a fixed bypass valve in it to bleed back some dissolved minerals in the water. Armchair engineers that we are, we criticized the fixed limitation of that bypass valve, and eventually a guy from Everpure, Roy Parker, posted a blurb about Everpure Claris.

The system is intriguing--it has an adjustable bypass, and uses a bed of hydrogenated resin (rather than salted resin) to soften the water by reducing the alkalinity as well as the minerals, solving (at least on paper) most of the problems with more traditional softening setups. Of course, there are weaknesses to this system as well, in particular that it's a proprietary system with potentially expensive replacement filters.

A couple of forum members have expressed interest in the Claris system, and I believe Mark (networkcrasher) has bought one already. A few weeks ago my friend Paul (pjones) contacted Everpure directly and wound up getting in touch with the Everpure product line manager, Tony Fischer. He mentioned that the two of us were interested in the system and figuring out what its capabilities are, and Tony wound up offering to send us a pair of Claris filters free of charge (and to keep) in exchange for providing feedback about the system--how well it works on our setups. Tony's notion of feedback seemed to be as simple as answering the question, "did it do what you wanted?" However, I'd like to delve somewhat deeper than that requirement in this thread, as well as invite other adopters to post their thoughts here. Also as usual please feel free to ask any questions you have, and I'll try to answer them as best I can.

Before I get any further, I want to disclose my relationship with Everpure (if it's not clear already), which is simply, as I say, that of receiving the product free to evaluate and keep; I'd also like to thank Tony for the generous offer, as I probably would have bought the system to try in any case.

I received the Claris unit today, and after harvesting several gallons of water from my current setup for future tests, I installed the Claris system and tested it with a rudimentary TDS meter and a GH ('general'/calcium hardness) and KH (carbonate hardness a.k.a. alkalinity) test.

The Claris filter head has an adjustable bypass with 7 presets (numbered from 0 to 6). You have to flush a good bit of water through the system after changing the bypass setting before it normalizes (on a TDS meter), and this is probably aggravated by the fact that Everpure sent me the XL filter, which is really huge--seems about the size of an oxygen tank for SCUBA diving (!). The Claris manual provides a table of recommended settings and estimated capacity based on the alkalinity of your tap water:


Image Source: Everpure Claris Manual; I am not 100% sure this isn't subject to change.

I performed my water tests on my tap water, on the Claris filter at settings from 0 to 5, and on my Cirqua-formulated distilled water. I've compiled my results into a Google Document which you can access here. The data is subject to change, of course. I am working on getting a dedicated pH test kit to augment my test setup, and I will re-do the tests over the next few weeks to get a feel for how consistent it is across any inlet water changes I may experience.

As you can probably see, the granularity of my tests doesn't really give a very good sense of how much the bypass changes, as the steps seem to be smaller than the 17.9 ppm granularity of the tests. However, you can get a pretty good idea from the tests that the filter seems to reduce calcium hardness quite a bit more than alkalinity, although it reduces both by a lot, and gives me enough of a reduction to safely run about 35 ppm of mineral hardness (at level 4 or 5) with about 50 ppm of alkalinity, which should not scale my brew boiler (and cause very minimal scale in the steam boiler). At the 0 setting, I got for virtually no alkalinity or mineral hardness--both titrated in a single drop. It looks like you could close the bypass, flush a gallon or three of water, and fill your steam boiler at 0, and get all the benefits of RO water.

I'm left to wonder at this point--with such low alkalinity and mineral hardness, where is the ~250 ppm TDS in my water coming from? I'll have to call Tony later this week and ask. Finally, one parting note that raises my eyebrow: The water tastes clearly superior to tap water, naturally because it removes the dominating notes of chlorine in the water. However, it doesn't seem to taste quite as good as simple carbon-filtered tap water (using my aforementioned 0.5 micron carbon filter), and most certainly doesn't taste as crisp and nice (IMO) as the Cirqua-formulated water. Whether this bears out in coffee tasting is yet another question entirely. Questions, questions... I have so many.

shadowfax wrote:. . . As you can probably see, the granularity of my tests doesn't really give a very good sense of how much the bypass changes, as the steps seem to be smaller than the 17.9 ppm granularity of the tests.

If you are using the liquid test kits with a 5 ml test tube, I believe you can cut the granularity in half by using a 10 ml test tube and doubling the number of drops.

Yes, I've noticed this before (IIRC Jim mentioned it on the water FAQ). A fool and his money... I've ordered a pair of 10 mL graduated test tubes and a 10 mL pipet. Yesterday got me really, really fed up with doing these tests using the equipment that comes with the Aquarium Pharmaceuticals test kits. Why they give you those stupid plastic covers rather than a rubber stopper that seals properly, I will never know.

Thanks, Nicholas. I'll be following with great interest as we plan our own plumb-in.

Great work. One reason I never updated the Water FAQ is the memory of doing the titrations. Better you than me

The Claris filter looks like a sodium softener on the cation side, since it cuts hardness without cutting TDS, but it looks like the anion side may be an OH exchange. An ion exchange "purifier" like the zero water exchanges H+ and OH- ions and cannot be recharged on either side.

I'm surprised how much hardness you are getting from the Cirqua water. It looks like they go for the perfect brewing water and let the scale build up. In a cafe with well serviced machines, I suppose using this water would require regular descaling as part of the maintenance. I always thought they got some of their extra TDS by adding non-scaling salts.

I was just going on the press release information when I describe the Claris resin:

Claris Press Release wrote:Scale build-up can be avoided by removing the scale minerals, removing the carbonates and/or by lowering the pH. The Everpure Claris uses the latter two. The Claris resin is loaded with hydrogen (H+). The hydrogen ions dissolve the carbonates (CO32- and HCO32-) , and the resin then removes the calcium and magnesium ions (Ca2+ and Mg2+). This process also lowers the pH.

Am I misunderstanding this? It sounds like the resin is loaded with hydrogen (H+) ions, which apparently both dissolve the the carbonates (alkanity), converting them into dissolved CO2, and then removes the calcium/magnesium ions. That does not sound like dual-stage (+ and then - or vice versa) ion exchange to me. Of course, their bit of literature does little to clarify the point. For quick reference, here's their description of the '5-stage' filtration:

Claris Press Release wrote:

1. PRE-FILTRATION
   Pre-filtration media at the entrance to the cartridge retains coarse particles such as dirt and sediment.
2. CARBONATE HARDNESS ADJUSTMENT - The ion-selective filter medium reduces the levels of carbonate hardness while retaining all other minerals needed for good hot beverages. In combination with the DUOBLEND bypass technology it ensures a precise and consistent adjustment of carbonate hardness.
3. FINE FILTRATION - Behind the carbonate hardness adjustment process a fine filter compresses the filter bed and retains fine particles.
4. 5-MICRON PARTICLE FILTER MEMBRANE - All of the water passing through the CLARIS filter system flows through a final 5 micron particle filter membrane.
5. HIGH EFFICIENCY ACTIVATED CARBON ABSORPTION MEDIA - A high efficient activated carbon absorption media at the filter outlet

As you say, the Cirqua water doesn't seem to be the type of "set it and forget it" water that you'd want to put in your steam boiler, though it should be near-scale-free in a brew boiler if you have a double-boiler setup. I'd say it's certainly worth noting that if you use that you'll want to flush your steam boiler with near-zero TDS water every month or every couple of weeks.

Also, as a matter of reference, although I don't have a pH test kit right now, I added an equilibrium pH (pHeq) field to my measurements from yesterday, which again you can find as my published Google Document here. I realize it's formatted rather uselessly for spreadsheet wizards, and I'll try to have another go at it when I have a chance to collect some more data.

Maybe I misread it, or am missing something, but it seemed your measurements on the Claris show a drop in both GH and KH, but no drop in TDS. The zerowater system, for instance is a dual ion exchange that substitutes 2H+ and OH- for Ca++ and HCO3-, and thereby eliminates all minerals. The Claris also seems to reduce both kinds of hardness, which requires both anion and cation exchanges; but it's not eliminating TDS, so it must be substituting other minerals, no matter what their PR says.

My tap water was 318 ppm TDS as measured yesterday. The claris measurements ranged from 230-260 ppm TDS (between 0 and 5 on the bypass valve setting). So, it's a TDS reduction, but is not really commensurate with either the mineral or carbonate hardness reductions. Whatever the system is doing, it doesn't seem as simple as a typical salt-based ion exchange setup. I guess my main point is not really that there's not some kind of exchange going on, but rather that the exchange is not using salt ions. This is based on my informal conversation with Tony, and it's on the top of my list for clarification.

On the other hand, IIRC, I also got a small TDS reduction with my old DVA softener + carbon filter setup. I forgot to test that water, but I harvested plenty of it for testing before dismantling it, so I'll post an update with those numbers as well.

Nicholas,

I just installed the Claris system Monday. I opted for the M(edium) cartridge and it's plenty big - I can imagine your XL is huge.



What method are you using for he GH and KH measurements? I'm using an HM Digital TDS meter and getting nearly identical TDS readings as you are.

Thanks for posting your findings,
George