I took my own advice and bought the SCAA's new "Water Quality Handbook." This 56-page book is a mix of taste test results, practical advice and some hardcore chemistry.

The TDS tests may be of the greatest interest to Home-barista readers. The test methodology was blind tasting by six tasters at the SCAA Lab in Long Beach. From page 31:

In a tasting conducted by the Technical Standards Committee of the SCAA, coffee was brewed with different levels of TDS to determine if significant flavor differences existed and how much difference actually existed. ... The same coffee, grind, and brewer were used and the same standard combination of minerals was used. The only difference was the concentration of the minerals in the brewing water. The first tasting was conducted using three water samples: one contained TDS at a level of 45 mg/L, one at 150 mg/L, and one at 450 mg/L. The coffee that was brewed with 150-mg/L water was chosen as far superior by all who judged the coffee.

A second tasting was conducted using 125 mg/L, 150 mg/L, and 175 mg/L samples to determine if minor variations in water quality would have an effect on flavor and extraction. The minor changes in the TDS of water were unanimously discernable by the panel. Acid and body balances were perceived to be off at both 125mg/L and 175mg/L TDS, and the 150 mg/L TDS brew was rated superior.

Different treatment methods are discussed, with only one being flatly condemned: the cation salt exchange method commonly sold with home espresso machines. From page 37:

The flow-rate problem through the bed of coffee is further compounded if the water has been treated by a zeolite [cation] water softening system, the most common type of water softening system, and which is required to be regularly recharged with salt. Through ion exchange, this process replaces the minerals in the water—principally calcium and magnesium—with sodium. When combined with the bicarbonates already in the water, the sodium bicarbonate forms a shiny, slimy material that binds the coffee particles together and blocks the passageways through which the water would normally flow. As shown in Table 7, this extends the brewing time, thereby causing over-extraction of the flavoring material and leading to excessive astringency and bitterness. As a result, brewing coffee or espresso with softened water is not recommended. [Emphasis added]

Table 7, shows the increase in extraction time that cation exchange causes in seven different cities, with the hard-water regions of Dallas and Los Angeles showing the greatest impact (over 40%).

Finally, I'm only reporting here. I have never used a cation system and have no personal basis for judging them.

The Handbook is available from the SCAA Store. http://www.scaa.org

Whoa so these people are telling that
espresso machines SHOULD NOT use soften water????????????????????????????
I'm I understanding this??????
I DO NOT agree with that
And I speak after opening and seeing
what hard water does inside of boilers and components
for 10 years of repairing espresso machine and brewing equipment
but this new philosophy and assumption
will make even more business for
REPAIR companies

They did not say it should not be treated, Stefano. Much of the handbook is devoted to different methods of treating water to prevent scale. The warning is specifically against the "softening" method.

ok
for sure that there are other method,
removing minerals and then re-mineralization back some of the good ones that one of them
we have a commercial customer that spent $4000 for a water system

but to go as far as saying that SOFTEN water is not recommended in MY opinion is a stretch
and a lot of people might not get this in the right meaning,

soften water is still much recommended over any hard water
unless what ever you might gain in very very fine tasting bud extreme
is worth a lot of money in repairing or and parts for your machine
just my opinion

stefano65 wrote:soften water is still much recommended over any hard water
unless what ever you might gain in very very fine tasting bud extreme
is worth a lot of money in repairing or and parts for your machine
just my opinion

Well, the options are broader than that. For all my pourover machines, I used bottled water that was processed by RO and remineralized, which is essentially what the better coffee bars in hard water regions do. There are also compact semi-commercial treatment cartridges that can be used at home for plumbing in that avoid salt-based softening.

Marshall,

Does the book provide details on the hardness and alkalinity levels of the "standard combination of minerals" the SCAA used?

I'm using a cation system here, and I've thought about giving the Everpure Claris system a try. But I'm at a bit of a loss to reconcile the SCAA's 150 ppm TDS recommendation with tests I've run on my water and my past experiences manually blending water. A ZeroWater TDS meter indicates my water has a TDS of about 140-150 ppm. Top-quality Hach testing kits show the hardness to be in the 140-150 ppm range, and the alkalinity is also in the 140-150 ppm range. I don't know why the two figures don't add up to the total TDS, but they don't. When I test water from the cation system, it still has a TDS in the 140-150 ppm range, and alkalinity still in the 140-150 ppm range, but hardness virtually zero. Given that the calcium and magnesium have been replaced by sodium ions, the respective concentrations still don't add up to the TDS figure.

Back when I ran my GS/3 from the reservoir, I produced RO using a ZeroWater pitcher and blended in enough tap water to bring the hardness up to 70 ppm, which is a figure recommended in Jim's water FAQ. As I recall, the TDS and alkalinity remained proportional. In other words, the TDS and alkalinity were both in the 70 ppm range, which makes sense given that I was simply diluting the entire mineral content of the water.

So, if I switched to an Everpure Claris system, and set it to blend in enough tap water to produce a TDS of 150 ppm, then I would essentially be back to untreated water. That would be far too hard for the machine and would no doubt require descaling on a weekly or monthy basis. But if I dial the hardness back to 70 ppm, the TDS will only be 70 ppm, less than half the SCAA's recommendation (and no doubt I'd have to descale every 3-6 months.)

Seems to me that there's a lot of variation is water composition around the world, and the SCAA's results may not be applicable outside of the area in which their tests were conducted. At a minimum, though, it would be nice to know the composition of the water they tested.

A couple of points:

I would very much like to meet the people who can distinguish 125 TDS for 150 TDS from 175 TDS water in triangle tests, if the same hard water (diluted with distilled for the lower concentrations) were used in each test.

If cation treated water flows more slowly, it can be compensated for by setting the grind finer. The sliminess of the puck has often been noted and is even more true of pure RO water. It simply indicates the complete absence of calcium. There is a slight degradation of taste using cation softened water compared to neutral water; but it is small, certainly smaller than using RO water and a calcite cartridge, which is the only simple alternative for no-scale water treatment

The turn of phrase from your quotes, along with the tests quantities used, reminds me of the promotional literature and demonstrations done by Cirqua. I'm curious if the technical committee in question is an alias for them.

Conventional water softeners using an ion exchange (cation) system do not replace the magnesium and calcium in the water with a salt ion. The calcium and magnesium is usually bonded with some form of a carbonate molecule as the former are negatively charged (-2). This bond is not very stable and breaks under heat (inside the espresso machine) forcing them to precipitate and clog the machine. The salt molecule is comprised of two positively charged sodium atoms with one chloride molecule (-2). It is the chloride atom that bonds with the Calcium and Magnesium atoms (+2) forming a much more stable molecular bond than the carbonate molecule and does not precipitate under the heat of the espresso machine.

The new Calcium Chloride and Magnesium Chloride molecules are not removed from the water, they simply pass though the system without breaking the molecular bond and allowing the scale to form. The amount of material in the water is not changed by the cation system, only the distribution of the atoms.

There are two sodium atoms, positively charged, left over in the process. The structure of the carbonate molecule to which the calcium and magnesium were attached would determine the makeup of the newly formed molecule. Sodium Bicarbonate is a primary cleaning component in espresso machine cleaners manufactured outside of Italy. It isn't used in Italy because that bond is also not terribly strong and will precipitate into a carbonate (lime), clogging solenoid valves when the group heads depressurize. The sodium carbonate bond does keep the lime out of the boiler though, so long as it remains hydrated and under pressure.

The quality and composition of the water going into the system may have more to do with the result more than the cation softener. The softener makes no effort to filter or affect the flavor of the water, only to create a stronger chloride bond than the likely carbonate bond in the water so that it will not harm the machine. If the TDS measured is a result of the broken carbonate bonds, the improved taste as a result of the solids (to which the molecules extracted from the coffee bond) will also help to destroy the machine. It may suffice to soften the water and add non-harmful solids to the water to compensate.

This is usually the problem with reverse osmosis systems. So much is removed from the water than nothing is left for the coffee to bond and more is pulled from the coffee that isn't necessarily desirable. The new trend is to add solids to RO water to compensate, using materials that aren't harmful to the machine.

Phosphate softeners attempt to keep the Calcium and Magnesium in suspension as an alternative to Chloride, but the bond is relatively weak and will not work effectively in the temperatures espresso machine experience.

Choices are not perfect:

1. RO systems with sophisticated add-solids systems that waste water,

2. using filtration systems before a cation system that may inhibit extraneous molecules to which
the sodium ion can bond or

3. do nothing and decalcify the machine (or throw it away) every three to 24 months.

We will ask one of the three manufacturers of softeners what he thinks about all of this. I think that the root cause of differentials in extraction rates and flavor profiles are more complex than simply whether or not a cation softener is in place.

One thing is certain, IMHO, that in the country where cation softer systems dominate the espresso does not suck.

mteahan wrote:Conventional water softeners using an ion exchange (cation) system do not replace the magnesium and calcium in the water with a salt ion.

Misprint ?!? Cation softeners replace calcium with sodium (or hydrogen for the non-rechargeable ones)

Hi Marshall,

I know that you are not personally responsible for the drivel that you quote.

The passages you quote are quite honestly unbelievable to anyone who has ever designed or participated in blind tasting tests. They are undeniable horseshit, sorry. These words sound like they were written by a sponsor who sells water treatment systems, of which they have a few. I would also call attention to the fact that this organization, the SCAA, counts within its membership many companies that produce roasted coffee and coffee beverages which compare favorably only with "airplane coffee."

If this is the standard by which the SCAA should be judged, when it comes to "scientific research," then what little credence I ever gave to the mouthings from that organization will be even further reduced.

ken