My job is taking me to Los Alamos, NM in about a month. I've lived there every summer for the last 3 years, though I have never had a machine that needed plumbing while I was there.

Anyway, I have heard from others, and experienced it myself, that the tap has high concentrations of silica. I am not sure exactly what form it is in the water (colloidal or what not), but there is enough of it that it shortens the lives of hot water heaters everywhere. My experience with it was with my La Pavoni last summer. splashed water flash boiled against the side of the machine and scaled it with silica... I know this because the citric acid wouldn't touch it--silica is, I believe, negatively charged and itself acidic in water, so it doesn't respond to limescale removers.

And that's my fear--getting scale in my boiler that won't come out. I am considering 2 options: buy a Flojet, accumulator, and calcite filter/other remineralization filter, or buy an inline filtration system. I have been looking around for something suitable for a few days, with bites but no luck. Silica seems to be a pretty nasty thing in water, with much more complex chemical interactions... seems to have a knack for fouling membranes in RO systems, especially.

I was just wondering if anybody out there has experience with filtering high-silica water for home use, especially in terms of for an espresso machine.

Thanks in advance.

Thanks for the interesting question; new one for me.

It seems like you might be out of luck. Carbon filters don't work, nor does regular (cation) ion exchange. I don't know if anion ion exchange is used on drinking water, but the process described doesn't sound promising. Your best bet may be an RO system along with much more frequent cleaning and replacement of the filters. This isn't much good, but better than nothing.

Yeah, I actually read that thread from one of my Google searches. I didn't immediately give up, but that's the impression that I've gotten from the first time one of my coworkers talked to me.

From my further reading, anionic ion exchange does work for getting silicon that actually is dissolved in water, but may not catch the colloidal, suspended silica. For that you combine it with RO, as near as I can tell. The one I have seen is the Maxcap RO/DI system from SpectraPure. It's a filtration system designed for filling aquariums. Supposedly it would give you ultrapure water, and I can then toss on a calcite filter and have decent espresso.

My question is, will this be economically feasible vs. buying a $200 flojet + accessories setup, and then the year over year cost of buying 5 gallons of RO water as needed, which I would probably also use as drinking water? I am not sure how fast I could go through 5 gallons, but in a 3 person family, I could see it being 2 per week... maybe. On the other hand, those RO membranes run up to $70 a pop, plus $10-30 apiece for the other 4 fliters--it starts to get kind of depressing.

I will call the SpectraPure people and ask more. I'll post back here later next week, I suppose. Thanks, Jim.

This is an observation I've made and I've no idea if it's true, but it seems to be.

There is a lot of marketing in the water filter field and not so many manufacturers so many of the web sites dealing with filters are marketing machines offering excellent technical/sales support/ marketing materials in exchange for a rather significant mark-up. If you can read between the lines and figure out what they're selling someplace like http://www.thepurchaseadvantage.com or http://www.kfswater.com might have the same pieces for significantly less money and much worse delivery times.

I purchased my filters from The Purchase Advantage and while the prices were very good, the delivery time was weeks so it's not the place to go if you're in a hurry. It's a division of Siemens so no worries about being a scam, but it seems like it's a toy for one of their salesman. It's far from the only non-service/good pricing filter site but it worked for me.

Ira

Ira, thanks. It may turn out that a high-quality RO setup will be the best thing, as Jim suggests. I can put it after my ion exchange water softener, which might help extend the life of the membrane somewhat, and then put a calcite filter or two downstream of the RO unit. I will definitely be looking for a unit that has a constant TDS meter, and I will need to get some type of silica level test kit--I have seen these.

In any case, I need to find someone with more silica expertise to ask about this, specifically how often I'll have to replace the membrane, which is usually quite expensive--if this is going to call for a replacement more than every 8 months or so, I can imagine it's probably cheaper to just buy a few 5 gallon tanks of bottled water and top them off with RO water at the grocery store.

Jim Schulman wrote:In reverse osmosis, some water is forced through a membrane impermiable to minerals which are washed away by the proportion of water not passing the membrane. RO should produce virtually pure water (below 1 mg/l total solids); however, mineral removal is compromised when the amount of waste water is reduced. For instance, supermarket RO vendomats in very hard water areas may put out as much as 50 mg/l hardness and alkalinity levels. Home reverse osmosis systems are generally self-cleaning, but need to be installed and set up by specialists. They are more expensive to buy and operate than ion exchange softeners (see below).

My emphasis. From The Insanely Long Water FAQ.

Jim has me a little concerned about my idea of buying RO water from the supermarket in Los Alamos. I certainly am not sure I would trust them to set the product:waste ratio properly, especially in light of the price of water in our desert. Here in Norman, RO water at Wal-Mart is $0.33/gallon. I can imagine that the price in Los Alamos will be comparable, if not significantly higher. In light of that, I have on one hand serious concerns about the quality of water from supermarket RO water, and it will also be expensive. I can imagine that, considering I want to set this up for drinking water as well as Vetrano, that I will be using 3-5 gallons every day. Maybe not quite that much, but I expect my initial guess of 10 gallons per week was quite low. Regardless, that gets expensive. Even at an optimistic $0.33/gal and 3 gal/day, that will be $365. More realistic estimations are much pricier.

It begins to look as though, even if I had to replace all filters (a given) AND my membrane every year, that it would quite quickly become cost-effective to go with an under-sink RO unit.

I talked to a number of different water treatment folks today. The first guy was from a local place in Santa Fe, NM. He was optimistic about RO taking out silica, though he didn't seem like he was very knowledgeable about it. He took my digits and promised to consult his literature to see if he could get more details and get back to me. His prices, including installation, were pretty good, although I forgot to ask him about any form of remineralization for coffee. I suppose I can always add a calcite filter myself (I have the housings on hand from my current setup).

The second guy, also a local from Santa Fe, was much more pessimistic about getting Silica out. He claimed that silica removal right out of an RO unit would be pretty poor--on the order of 50%. I found this a little startling. Still, it seems like the complexity of silica's presence in water would allow there to be such ridiculous discrepancies in what people say--from what little I have gleaned from my own survey of the literature, it seems like even the basic distinction between dissolved silica and colloidal silica is not really adequate to grasp the different types of interaction it will have with water.

Anyway, this guy seemed pretty knowledgeable about the issue, especially from a local perspective. He said that Los Alamos' wells have between 50-70 mg/l silica. He told me that he was working with an engineer in Arizona to develop a practical solution to silica removal prior to RO processing. He gave me his digits (I'll give it to you if you are curious--PM me), and I gave him a call. Sam was a quite personable guy, and he said that he was looking at some specialized media that would remove colloidal silica, with the catch being that the media would need to be replaced about every 3 months (a filter replacement). He said that the price of these would depend on the price he himself gets for the media in bulk, but he anticipated in the range of $10-15. It seems that, regardless of what option I go with for RO, I can always put one of these upstream of the system, if they work out. They fit in a standard 10" filter housing.

Finally, I called SpectraPure of Tucson, AZ. They were pretty friendly, although the saleswoman was a little confused by what I said. I was inquiring about their "SilicaBuster" deionization filter, and their RO/DI solutions in general. I was curious if I could get an RO setup that used RO, then deionization, and then remineralization (basically, their "Coffee Lovers' RO unit", but with this stuck in between the RO and re-mineralization filters). In pretty short order she connected me with an engineer who was, well, an engineer (and I don't mean that in no nice way! ). He told me that the RO unit they sell generally will get rid of 98% or more of TDS, and with my levels of Silica, it may drop to 94% silica removal, but that it would still be less than 1 mg/l of silica. He said that they didn't offer any way to put deionization in between the RO and re-mineralization filters with the current products that they sell, and went on to say that DI is quite specialized and specific to aquarium solutions--that it would be vastly overkill for what I want. He told me that he thought I would be quite happy with the Coffee Lovers' system's performance for my espresso.

My slight dislike of his tone aside, I tend to believe him. And while it is expensive, the "Coffee Lovers' System" seems like it might be ideal--it would simplify the whole process of tuning the output water for a sufficient hardness to make decent espresso. June's comment in a recent H-B thread certainly recommends them after her time with the unit running on well water with whopping 500 ppm TDS (!). I feel like her numbers ought to reflect what I can expect, in light of the fact that I will have much lower TDS but large amounts of Silica.

My thinking at this point is to splurge on the up-front cost and get the SpectraPure RO unit. If I need to down the road, I can give Sam's little 10" filter a shot as well, upstream of the unit. Any comments or further suggestions are welcome, especially if you see any gaping holes in my logic. I will keep you all posted as things pan out. I'm still planning on getting a Silica test set to make sure this is working.

Like a coworker of mine told me recently in response to my silica worries, "Man, being a geek is expensive." I sarcastically responded that it was still a lot cheaper than his Volvo...

Since I wrote the FAQ, TDS meter prices have tanked, so getting a good reading of ions in the supermarket's water will be no problem. Measuring the silica in your treated water is another problem entirely.

It's very odd that there's no SOP for silica in these areas. It may be worth talking with the water board about their silica content and its control; for instance, they may allow most of it to settle out. The hardness is right in the ballpark, and they seem proud of the taste, so going through a bunch of gyrations is a pity unless absolutely necessary.

Jim, you're very right--they have great water other than the silica. It tastes perfectly fine. I am curious if remineralized RO water will be any better.

I called the utility service department for Los Alamos County this afternoon. They connected me to a nice woman who didn't know much about silica, but she managed to find a recent report that compared Los Alamos tap water to various bottled waters (to show that Los Alamos water is just as good or better). She promised to email this report to me, and said she would post it to the website--she was surprised it wasn't there already. Anyway, this report confirms my silica doomsday salesman of yesterday--71 mg/l. She mentioned that it ought to drop significantly in the next 2-4 years as they exhaust the current well and transition to some other water sources.

So, I definitely need and RO unit to get this out. It is kind of a shame. Jim, I have a question. Is a re-mineralization filter (like a calcite filter, I assume) going to bleed CO3 back into the water as well as calcium, for alkalinity? Seems like both of these (the Ca+/Mg+ and CO3-) are needed for good espresso.

I just thought I would post this for amusement. This is a comparison of Los Alamos tap water with some big-name bottled waters. It was provided to me upon request, and at this point, you must obtain it by request as well... I don't have a link.


Source: Los Alamos County Utilities Department.

Look at Fiji water. Holy crap, even more silica! I would be willing to bet that the high silica content has to do with the fact that Fiji is a volcanic island like Hawaii. As you can see, though, it's not unusual to have a noticeable amount of this stuff in any old water, and it's probably fairly harmless at such low concentrations.

So, kids--if you ever think about trying Fiji water in your machine to see if it tastes better than the tap... Don't do it!

shadowfax wrote:Is a re-mineralization filter (like a calcite filter, I assume) going to bleed CO3 back into the water as well as calcium, for alkalinity? Seems like both of these (the Ca+/Mg+ and CO3-) are needed for good espresso.

Yes, it's basically powdered limestone. A TDS meter will allow you tune it to your specs: anywhere from 50 to 100 tds will be good for both scale and taste; 100 to 150 for best taste and occasional descaling.

I'll be interested to hear how well the RO method does, and what sort of tuning and/or maintenance it requires.