Estimate water profile from water softener?

Homebrew Talk - Beer, Wine, Mead, & Cider Brewing Discussion Forum

Help Support Homebrew Talk - Beer, Wine, Mead, & Cider Brewing Discussion Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Nateo

Well-Known Member
Joined
Jul 22, 2010
Messages
2,050
Reaction score
46
Location
Bennett Springs
Hey guys,
I understand the basics of how a water softener work. Na+ replaces Mg++ and Ca++.

I got a C- in chemistry, so give me a break for the dumb question, but if I know the water profile (how much Ca and Mg my water has before the softener), how would I estimate the water profile post-softener?

I wanna say just add up the Ca and Mg, and multiply by 2, but I think that's wrong.

If the softener strips the bicarbonate of its Ca, where does the CO3 part go?
 
The Ca and Mg are replaced on a milliequivalent for milliequivalent basis. Ion exchange softeners are very efficient at removing nearly all divalent ions such as Ca, Mg, Fe, and Mn. So, you can use a program like the water adjustment sheet in Bru'n Water to estimate the new Na concentration.

Input your water profile information and note the milliequivalents of your cations. Then zero out the Ca and Mg concentrations and increase the Na concentration until you calculate the same number of cation milliequivalents. That is the approximate Na concentration produced by the water softener.

The concentrations of the anions are unchanged by the cationic exchange resin in the water softener. That means that the carbonate or bicarbonate concentrations are unchanged by the softening. The alkalinity is unchanged. Although the Ca concentration, the carbonate and bicarbonate are still readily paired with cations such as Na. Don't forget about sodium bicarbonate!

By the way, I saw the water profile you posted. Darn hard water!
 
Divide calcium as mg/L by 20 or as ppm as CaCO3 by 50. This is the number of milliequivalents of calcium in the source. Divide Mg as mg/L by 12.15 or as ppm as CaCO3 by 50. This is the number of mEq/L magnesium. Add these two together. That's the total number of mEq/L calcium and magnesium removed and replace by sodium. Multiply that number by 23 to get the mg/L sodium in the processed water. Ca++ and Mg++ go to close to 0 to be replaced by Na+ as calculated above. All other ion stay more or less the same (except that cations such as iron, strontium etc will be replaced with sodium).
 
Thanks for the quick responses guys! I just moved to a new place that has less-than-ideal water, so I'm having to re-figure all my water chemistry. The water actually tastes good, it's just problematic for brewing.

A quick google search told me that ppm and mg/l are close enough to not worry about it. Is that correct?

My Ca is 74ppm, Mg is 47ppm. 74/20 = 3.7, 47/12.15 = 3.87. Ca+Mg = 7.57. 7.57*23=174.

So my softened water has 0 Ca and Mg and 174 Na.

Is that correct?

I'm thinking of doing a 50/50 mix of softened and unsoftened water, which would sort of balance out my water decently. My unsoftened water only has 2ppm Na to begin with, so a 50/50 mix would give me 37ppm Ca, 23.5ppm Mg, and 88ppm Na, 6 SO4 and 4 Cl. The sulfate is still really low, so I don't think I should be worried about harshness from the Na, but should I be? I almost only brew low-hopped malty beers, so I thought the Na would add some nice sweetness

The bicarbonate would still be an issue, but I was planning on knocking that out with phosphoric acid.

The other thing I wanted to get a GH/KH kit and try was Kai's-via-AJ's method of slaked lime magnesium/carbonate reduction. My water is so low in everything except bicarb and Mg I think I could get some decent water that way. I don't think many homebrewers go that route. Besides Kai's website, I haven't found much info on that.

I know I could just buy RO water (which is a pain because I live out in the boonies now) or buy an RO unit, but that just feels like cheating to me, for some reason.
 
Yes, 1 ppm is, to all practical purposes, the same as 1 mg/L but where you have to be careful in this context is that hardness can be reported "as calcium carbonate" in which case they almost always use "ppm" whereas if it is reported as the ion it is almost always "mg/L". If you see the words "as CaCO3" then you are sure.

Assuming that the calcium and magnesium numbers are mg/L as the respective ions then the numbers look right. As the exchangers aren't 100% efficient you will still have a mg/L or so of each of Ca++ and Mg++ and, correspondingly less Na+.

The blend should split the differences but why not just brew with the water as is and blend with RO if you need to control alkalinity. Remember that if you do that with phosphoric (or any other acid) you are swapping each bicarbonate ion removed for phosphate ions. Unless the alkalinity is out of sight this is pretty decent water.

Decarbonating by lime treatment or boiling is a lot more trouble, IMO, than just diluting with RO, is a bit iffy (i.e. you don't get the same result every time) and really requires the use of a pH meter to do properly.


Guess I don't understand why RO is cheating any more than ion exchange.
 
Just for reference, here is the water profile Nate posted on the AHA forum.

Sodium, Na 2
Potassium, K < 1
Calcium, Ca 74
Magnesium, Mg 47
Total Hardness, CaCO3 381
Nitrate, NO3-N 0.5 (SAFE)
Sulfate, SO4-S 2 (6ppm)
Chloride, Cl 4
Carbonate, CO3 < 1
Bicarbonate, HCO3 440
Total Alkalinity, CaCO3 360

Darn hard and alkaline. Extremely suitable for lime softening excepting that he should not be following Kai's or AJ's advice on the methodology. The high Mg content should also be addressed by performing an Excess Lime treatment. The main difference from what Kai and AJ recommend is that the pH MUST be increased to at least 11 in order to precipitate the Mg along with the Ca.

Although AJ and I appreciate the simplicity of using our own RO systems, that is an investment in the hundreds of dollars. That compares to the investment of about $6 for the lime needed to treat thousands of gallons of water, and the cost of an acid and a pH meter. RO is nice, but in this case a much cheaper alternative exists if Nate doesn't want to go that route.
 
Readers will have to decide themselves whether to use my method or not so lets just be clear on what my method is. It is a modification of the traditional 'split' treatment.

1. Calculate how much lime is needed to drop all the calcium and magnesium
2. Add this much to the whole volume of water (this is a difference WRT the split method).
3. Check pH. Continue to add lime until pH is 11 or a little over. This will cause both CaCO3 and Mg(OH)2 to precipitate.
4. Decant off the precipitate.
5. Add a bit of chalk to the clear water and get it into suspension. Just a smidgeon - it's only there to provide nucleation sites.
6. Bubble air or CO2 through the water while monitoring pH. Eventually it will get back into the eights or even lower depending on how long you are willing to wait. Be prepared to wait quite a while. When the pH is below 8 you are finished.

Where this differs from the traditional split treatment is in the way the excess Ca(OH)2 remaining aftger Step 4 is dealt with. Depending on how much lime you added the pH may fall back as the CaCO3 precipitates. If it does you haven't added as much as you need to get all (or as much as possible) of the magnesium. So if you've done it right the pH should be near 11 at the completion of precipitation and there will be excess Ca(OH)2 which is a base and must be neutralized if you don't want it contributing to the alkalinity of the water. You undertook this process to reduce alkalinity. In the traditional split treatment one third of the water is held back (the lime is added to 2/3) and the acid used to neutralize the (OH)- ions is the bicarbonate ion (HCO3-) in the reserved 1/3. The problem is, of course, that you don't get the magnesium (or much of it) in that 1/3. This is, in many cases, fine but if you want to get as much as you can then you'd have to use my method or something like it. In my method we're using carbonic acid, a stronger acid than bicarbonate ion, to do the neutralizing:

Ca++ + 2(OH)- + H2CO3 ----> CaCO3 + 2H2O

One can, of course, use any acid (well obviously nitric and prussic wouldn't be good choices) bearing in mind that the neutralized hydroxyls are being replaced by the anion of the acid

Ca++ + 2(OH)- + H2SO4 --> 2H2O + Ca++ + SO4--

In a case such as the water in question the sulfate and chloride are low and use of sulfuric or hydrochloric may be reasonable approaches. Of course carbonic acid is free if you take it from the air. Your cost will be time.

If ever I saw a case for an RO system, this gentleman's case is it. I do, however, appreciate that peoples budgets and approaches.... are different. One can also learn a lot of water chemistry by following the method I've outlined and trying to understand it.
 
AJ, that's odd that you say this water is better suited to RO since it has very low permanent hardness and very high temporary hardness. Lime softening will produce a ending Ca content in the 30 ppm range and Mg content in the 10 ppm range. Pretty good result in my opinion. The only thing not adequately addressed is the alkalinity. That is what acid is for.

PS: for a water this hard, it is best to feed an RO system with the water from the ion-exchange softener. That will greatly extend the life of the membrane.
 
Sorry, I should've posted the whole water report first so you'd know what I was talking about. Thanks Martin for reposting that for me.

I completely agree that buying an RO unit or RO water would be the easiest and most convenient (though you could make the same argument for buying beer vs making beer), but I like tinkering, and although I don't know much about chemistry it's something I want to learn more about.

I just bought a new kettle, burner, MLT, digital pH meter, and some other odds and ends while I was waiting for my water report to come back, so my brew budget is kind of shot for a while, so an RO unit, if I go that route, will have to wait.

Thanks so much for you guys holding my hand through this. I was planning on using phosphoric acid to neutralize the alkalinity because of the amount of phosphate already in the malt. Is that a reasonable approach? I don't have any hydrochloric acid on hand, though I could probably find some.

I'll be getting a GH/KH test kit soon so I'll be able to do some experiments.

Thanks again for all your help. This is an interesting area of brewing that has very little info available in the homebrew world.
 
AJ, that's odd that you say this water is better suited to RO since it has very low permanent hardness and very high temporary hardness. Lime softening will produce a ending Ca content in the 30 ppm range and Mg content in the 10 ppm range. Pretty good result in my opinion. The only thing not adequately addressed is the alkalinity. That is what acid is for.

The alkalinity is huge and its hard to predict exactly what will happen with lime treatment. Done right he ought to be able to get much of the magnesium and assuming the usual 1 mEq/L rule of thumb the alkalinity and calcium down to around 50 ppm as CaCO3 each. So what I meant was, and I mentioned it because of his statement that he didn't want to do RO, if anyone would benefit from RO, he would. I certainly wouldn't want to have to deal with the lime treatment each time I brewed and I wouldn't feel comfortable without doing an analysis for each treatment.
 
EEK with that water I would just go with RO and add calcium chloride or gypsum to get the required calcium, and adjust with acid or sodium bicarbonate to get the right mash pH.

Any post-water-softener water is unsuitable for brewing due to the high sodium level. The beer will taste "salty", and the malt character will be too sweet.
 
I tried AJ's decarbonation method. I overshot the pH a bit, and hit 12. I added phosphoric to lower the pH and got it a little bit too low, right at 7. GH was 13 and KH was 6. According to Kai's calculator, that's 65ppm Ca, 17ppm Mg, and 107ppm CaCO3.

Do those numbers sound right?

I'm a little confused on the process. I found AJ's alkalinity article on his website, but on there it recommends getting up to 12 to remove magnesium hardness. On here AJ suggested 11, or just above. 12 seemed to work pretty well. I had to add some extra lime to get it to 12, but once it did, the water "broke," started clumping and cleared really quickly. I ended up using 33.8 grams of lime for 53L.

If I'm using acid to knock down the pH at the end, instead of bubbling CO2, should I still add chalk to the clear water? After I added the chalk and dropped the pH the water stayed cloudy, and didn't sediment much more. I let it sit for a while but it never got clear again. I tested the hardness and alkalinity before and after adding the last bit of chalk and neither number changed, so I'm guessing that wasn't necessary.

The numbers I ended up with (if correct) were pretty reasonable for most of the beer styles I make, although wouldn't be great for very pale beers. Any ideas on how to get more out of lime conditioning? I saw on Kai's site that he added Ca, so maybe I'll try that.
 
Guess I was working from memory - always a mistake. If the article said 12 then go with 12 though I'd think you'd drop quite a bit at 11.

A for the numbersL 1 dH = 17.86 so 13 GH implies hardness of 232 which, if all calcium, would be 93 mg/L. It is impossible to tell how much is calcium and how much magnesium but as you got to pH 12 you should have gotten most of the magnesium. As this calcium level is higher than the calcium level of the original water I am puzzled as to how it could be so high.

OTOH a KH of 6 implies an alkalinity of 107 ppm as CaCO3 - a big improvement over the original 360. I must be missing something because taking out that much alkalinity requires taking out an equivalent amount of calcium and the calcium level apparently went up.

WRT adding the chalk - no, you wouldn't need to do that if not acidifying with CO2.
 
It's possible my pH was off. I don't have any 10 calibration solution, so I just calibrated at 4 and 7. I just ordered some so I'll have it for the next time I try.

Maybe I only got to 11 and some of the hardness was still Mg. I don't know how Ca could've gone up, unless I put too much lime in, and not all of it precipitated, but I thought the KH would've been lower if that were the case. The pH stayed at 12 after it cleared, if that's important.

All in all, the process was simple enough and effective enough, and I think with a little bit of practice I could get some reasonable water this way.
 
Got some pH 10 solution to double-check my meter. It was actually spot on.

I tried AJ's method again. Added lime milk to get to 11, then slowly added more until I hit 12.1. I used 2tbsp of chalk and 2 tsp CaCl for 20gal of water. I'm still not entirely sure what the calcium deficit means, when my alkalinity drops so much further than my hardness. I thought the hardness would bottom out before the alkalinity.

I let the water settle for about 8 hours, decanted, and hooked up the air stone. Aerated it for 24 hours. After that, it had dropped some more precipitate, but the pH was still at 12.0. I pulled a sample to test, and added enough phosphoric acid to drop the pH to 7.0.

Measured @ 7.0 the GH is 19 and the KH is 6. My untreated water's GH was 25 and the KH was 23. The water is cloudy again, but I'll let it settle tonight and check GH/KH again in the morning.

It actually tastes really good, but that might just be the 48ppm of Cl.
 
Retested this morning. GH is 9 and KH is 3. I think the reason my figures were wrong was that the titration tests aren't accurate when the pH is elevated.

I now assume they are calibrated at 7, so that's what I tested them at today.
 
Got some pH 10 solution to double-check my meter. It was actually spot on.

I tried AJ's method again. Added lime milk to get to 11, then slowly added more until I hit 12.1.

I went back and checked. 10.5 to 11 should be high enough. You do not want excess Ca(OH)2 as it results in extra alkalinity which, as you have seen, must be dealt with. Note that as precipitation occurs the pH will fall back (i.e. go lower). Also not that the magnesium hydroxide is a gel - i.e. the water doesn't look the same with an Mg(OH)2 precipitate in it as with a CaCO3 precipitate.

I used 2tbsp of chalk and 2 tsp CaCl for 20gal of water.

It is important to have nucleation sites but 2 tbsp of chalk seems an awful lot. Not that it should make any difference - the solution is saturated WRT chalk anyway and whatever you add should settle out.

I'm still not entirely sure what the calcium deficit means, when my alkalinity drops so much further than my hardness. I thought the hardness would bottom out before the alkalinity.

That depends. If the alkalinity is greater than the hardness then the hardness will be limiting and conversely. You added a fair amount of extra calcium with the calcium chloride addition which is the right thing to do if you want to decarbonate and don't mind the extra hardness.

I let the water settle for about 8 hours, decanted, and hooked up the air stone. Aerated it for 24 hours. After that, it had dropped some more precipitate, but the pH was still at 12.0.

The air, despite what the global warming alarmists would have you believe, contains very little carbon dioxide and so you will be there a long, long time if you depend on this to neutralize the extra lime. It is usual to use CO2 from a bottle and it still takes a long time but not nearly as long.

I pulled a sample to test, and added enough phosphoric acid to drop the pH to 7.0.

Measured @ 7.0 the GH is 19 and the KH is 6. My untreated water's GH was 25 and the KH was 23. The water is cloudy again, but I'll let it settle tonight and check GH/KH again in the morning.

You have reduced alkalinity certainly but the problem is that you don't know whether it was the lime treatment or the phosphoric acid or both which were responsible.
 
Retested this morning. GH is 9 and KH is 3. I think the reason my figures were wrong was that the titration tests aren't accurate when the pH is elevated.

I now assume they are calibrated at 7, so that's what I tested them at today.

Was phosphoric acid added to the main body of water or just the test sample?

Test kits are not calibrated for a particular pH. The hardness kits generally require a highish pH and usually incorporate a buffer to make sure that the sample is in the range the kit wants. Alkalinity does not depend on pH - it depends on alkalinity. If the alkalinity depends on pH then the reading will change with pH but alkalinity is simply the amount of acid required to bring the solution to a specified end point (around 4.3 pH).
 
I added phosphoric to the main body, after it seemed to work on the test sample. ~20ml of 10% dilution for 20 gal of water.

Besides the ions the acid gives up (phosphate) does it matter if the alkalinity drops because of acid or softening? I'm sure from the amount of slurry I have left that a good deal of something dropped. I assume it was mostly Ca, Mg and CO3.

Next time I'll try a slightly lower pH for the softening. I did notice a difference in the precipitate after I added the acid, it was much more like a gel at 7pH than the precipitate was at 12pH.

I wasn't really that concerned with the pH per se, but more with the high level of Mg. I made a Koelsch today, and hit a mash pH of 5.3 with only pils malt, without any sauermalz or additional acid. If there are flavor faults from this treatment, I thought a light psuedo-lager would be a good place to look for them.

Thanks again to everyone for all the help.
 
I tasted the first bottle of Koelsch yesterday. It wasn't quite fully carb'd but it tasted great. Nice clean bitterness, not harsh. Overall, I'm happy with the results.
 
Back
Top