We have a water softener that handles all household needs, and an RO system (Kinetico K5) for drinking. I have been using BH recipe water for a while, but would like to plumb in the machine (to play with PI and, well... mixing water gets old). So, I added a remineralizer cartridge to the K5. No one at Kinetico seems to have any real idea about what is coming out of that thing, other than total hardness (mainly mg, some Ca) should be below 50. Test strips show total hardness is less than the value I get for the recipe water, but really hard to tell with those strips.

Has anyone had success plumbing in remineralized RO water?

This is the reason I will be using a water recipe when I get to AZ. Mix valve would mean I would end up with chlorides hence I will be using a a pair of 6 gallon tanks and mix my own water that should last a numbest of months on each tank.

redbone wrote:Have yet to find a remineralizer that is consistent. Issue with them is due to how they function. Water passes by a mineral media during flow but when there is no flow the water in contact with the media increases in it's mineral percentage. First flow of water is much higher due to long contact time and subsequent water is much lower due to low contact time. Best bet for greater mineral consistency would be to use water post a carbon filter and use a mix valve to blend with R.O. water for desired mineral content.

Have you tried the Kinetico? The K5 runs a slow flow (more or less a trickle) of water from the RO membrane to the remineralizer. The RM's output fills a holding tank at the same slow rate. Once the tank is full (takes a few hours), RO flow stops until tank water is used and the system calls for water again. From what I can tell, that slow flow into a good size tank buffers the output against mineral spikes. There will be some spike in mineral content each time the first water hits the RM cartridge, but any water you take out of the tap will be a mixture of new (trickling in) and old. The result is that variation in mineral content is low.

At least, that is the claim. I do know that I have yet to see a high mineral level. In fact, it may be too low. Repeated testing shows somewhere between 25 and 50 ppm total hardness so mineralization is both low and fairly constant. What I have not tried is to start and stop the process a bunch of times while refilling an empty tank. That might yield more variation. I am also unsure as to whether my RO/RM "recipe" is doing as well as the BH water. The difference is not huge but I am no expert. Too bad it is so hard to A/B that sort of thing for espresso.

One interesting note, the RO/RM water tastes better on its own than the BH recipe. That always seemed a bit brackish.

Never used a Kinetico system. If my incoming water was hard enough I'd put together a non R.O. filtration system consisting of a sediment, carbon bloc and polishing fine filter then mix with R.O. water. K.D.F. can also be used a prefilter.

redbone wrote:If my incoming water was hard enough I'd put together a non R.O. filtration system .... then mix with R.O. water.

My criteria for a system like this would take into account the chloride, or silica, or any other undesirable that may have dictated the use of RO filtration. In the case of chloride, for example, I'd like the tap water's total alkalinity (mg/L as CaCO3) to be at least 4x the chloride ion concentration, that way if I blend back the alkalinity to 40 - 50 mg/L I would expect chloride ion down below the 10 - 13 mg/L neighborhood.

So, I've been using RO/RM water for a while and will try to give impressions. Overall, the taste of the espresso seems similar to the BH SCA water I used before (I am not an experienced taster). My wife uses RO/RM from the hot water tap for tea (yes, I know), and clearly prefers it to the BH water. In fact, she complains about "funny taste" if I switch back to the BH water. This is not a result of salts concentrating in the steam boiler, as that is flushed almost daily. It may be because the RO/RM water has a clean sweet taste on its own. In comparison, the SCA water tastes brackish.

As to taste of the coffee, I have said I detect little difference but I do think I slightly prefer the RO/RM to the BH water. I find myself adding a few percent of Evian to the Sychronica tank, however. Not sure if that is because I have a real need for a bit more mineral content, or if it is a religious thing. I will try to get a better measure of minerals than the test strips are giving me.

There is an interesting complicating factor in deciding which water to use, at least for me. As I age, my iron gut has become more sensitive to what I take in. I drink mainly milk coffees (cortados and cappuccinos), so there is some buffering of the coffee acids by the milk. I find that cortados cause a bit more upset than cappucinos, and that makes sense given that more milk in the caps yields more acid buffering.

Drinking the BH recipe, you are getting quite a bit of bicarbonate buffering with every cup. When I switched to the RO/RM water, the buffer was deleted and I may have been more prone to gi upset. So, I did an experiment and switched back to the BH recipe buffer content but with the RO/RM mineral content. We'll see, though I find I do not care for the taste of the hybrid as much as the straight RO/RM.

To summarize, my informal impression is that RO/RM water makes coffee that tastes fine. However, it lacks any buffer content and that may let coffee cause more issues n the gi tract. In contrast, the BH SCA recipe may or may not make equally good coffee (as opposed to my neutral pH RO/RM water), but may make coffee that is easier on the digestive tract. Has anyone noticed a beneficial gi effect of using buffered water to make coffee?

Finally got around to testing hardness with a Hatch 5B total hardness test kit. This gives results similar to the test strips, but a bit more accurate. I diluted the tiration reagent 1:1 with pure RO water but, sadly, the titration is still very coarse. My RO/RM water is about 1.5 grains/gal total hardness (CaCO3 equivalent). Most of that is from Mg2+ with just a bit of calcium for taste. Kinetico tells me the amount of Ca2+ is very small so let's ignore it for present purposes. Therefore, we are comparing the RO/RM water at about 1.5 grains/gal with the BH/SCA water at just under 5 grains/gal.

The RO/RM water is too soft. My taste buds have been telling me to add a bit of Evian to each tank, but that defeats the purpose of all this (taking water straight from the tap). To plumb this system in, I would want to add more mineral to the RO/RM water. Fortunately, the Kinetico K5 system that I use allows insertion of an additional RM cartridge. That would bring the mineral content to about 3 grains/gal, closer to the hardness of the BH recipe water. Whether it is close enough I can't say but I will try it at some point. For now, Evian.

My goal is to have remineralized RO water plumbed to the machine. That would be convenient and would let me play with my ECM's preinfusion feature. The disadvantage is that the RO/RM water is what it is. It is not useful to those who want to experiment with water recipes.

Thanks to Homeburrero who corrected my grain to Ca+2 and Mg+2 conversions.

Thanks for testing and reporting back.

One thing I think is worth clarifying . . .

cunim wrote:The BH recipes cluster at about 20 mg/l, all Mg2+.

Those BH recipes (https://baristahustle.com/blog/diy-wate ... o-bottles/ ) cause a little confusion because they use only magnesium for their hardness mineral and they report in units of mg/L of the magnesium ion.** So when comparing to units of general hardness, which always use some measure of chemical equivalence (gpg, °dH, french degrees, mg/L CaCO3 equivalent, etc) you need to do some converting. Here's how you convert mg/L of hardness minerals as ion to mg/L CaCO₃ equivalent:

Multiply the [Mg⁺²] ion mg/L by 4.1 to get magnesium hardness in mg/L CaCO₃ equivalents
Multiply the [Ca⁺²] ion mg/L by 2.5 to get calcium hardness in mg/L CaCO₃ equivalents
Add the magnesium hardness to calcium hardness to get total hardness

So for the BH recipes with zero calcium, a value of 20 mg/L magnesium ion comes out to a general hardness of about 82 mg/L of total hardness as CaCO₃, or just under 5 grains per gallon.

The BH article also reports bicarbonate ion as [HCO₃⁻] mg/L instead of using chemical equivalence units. For that you can multiply their [HCO₃⁻] ion by 0.82 to get bicarbonate alkalinity in CaCO₃ equivalents. So their 50 mg/L bicarbonate comes out to 41 mg/L bicarbonate alkalinity as CaCO3 - right at the long recommended SCA target.

You can find more conversion factors in this post: Good references on water treatment for coffee/espresso


**[Edit addition] If you visit the newer version of their recipes, DIY Water Recipes Redux, posted in July 2019, they fixed this confusion and started quantifying their GH and KH in units of CaCO3 equivalent.

homeburrero wrote:Thanks for testing and reporting back.

So for the BH recipes with zero calcium, a value of 20 mg/L magnesium ion comes out to a general hardness of about 82 mg/L of total hardness as CaCO₃, or just under 5 grains per gallon.

The BH article also reports bicarbonate ion as [HCO₃⁻] mg/L instead of using chemical equivalence units. For that you can multiply their [HCO₃⁻] ion by 0.82 to get bicarbonate alkalinity in CaCO₃ equivalents. So their 50 mg/L bicarbonate comes out to 41 mg/L bicarbonate alkalinity as CaCO3 - right at the long recommended SCA target.

You can find more conversion factors in this post: Good references on water treatment for coffee/espresso

Thanks for correcting my mistakes. I have edited the original post so that people are not misled.