Residual Alkalinity, how important?

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.

nilo

Supporting Member
HBT Supporter
Joined
Aug 18, 2009
Messages
749
Reaction score
43
Location
Seattle
First, I think there should be a room here just for water chemistry talking, seriously. Yooper, what you think?:D

How important is to match the residual alkalinity with the beer style?
I'm new to water chemistry here , and like many I guess, very frustrated.
John Palmer stated (I just heard him saying this on Jamil's show from 2009 podcast) that we should have residual alkalinity as follow, to yield good beer:
-60 to 0 =>Pale, golden, yellow and light ales
0 to 120 => Pale Amber, dark amber, copper color
120 to 250 => Brown ales, porters and stouts

I understand that the most important part of adjusting our brewing water is
-make sure your mash PH is between 5.2 and 5.4
-try to set your chloride to sulfate ratio to your preferences
-reduce minerals that comes in huge quantities for those with very hard water.

What about RA? Should we be adding salts to match Palmer's suggestions? Is this outdated and RA has proven to not be as important as previously thought?
 
Hopefully not very important. There's no way I can get my water up there in ra, without blowing up my ph levels. Unless i'm missing something.

_
 
The major problem I see with RA is people trying to apply it to mash when it is really intended to describe water so that people can compare one water supply to another. I think it can be helpful in coming up with something in a particular style if you tune (roughly) your water to resemble at least approximately the RA of the region in which the beer originated. Now having said that I'll note that I have 6 profiles for Burton which range in RA from slightly positive to about -75. So which do I choose for Burton? Generally one that puts the RA in the middle.

If you do succeed in matching RA (and sulfate and chloride as well) then you are on the way but you must know what to do with that water. The outstanding example is perhaps Munich where the water has relatively hefty RA. They brewed Dunkles with it without treatment but for Helles they decarbonated it. At least that's the story. I have a chart at http://www.pbase.com/agamid/image/57446374 that you can use to find the RA of various brewing cities and upon which you can plot the water you have available to see which city it resembles (in this sense). I just looked at it and see that Munich's RA is only about 50. I've posted many times that it is, IMO, pointless to try to establish a Munich RA only to turn around and undo what you just did (i.e. decarbonate the water).

Part of the problem with trying to establish a particular RA in water is that it can't be done easily. RA comes from calcium bicarbonate. You can't call Fisher and order calcium bicarbonate. You have to make it the way nature did and that's by reacting CO2 with limestone. If you try to do it by dissolving chalk with lactic or sulfuric or hydrochloric acid you get something with the RA you wanted by the overall water profile is not nearly as close a match as you would have gotten if you did it the "right" way.

Another factor to consider is that the brewers of yore lived with the RA they had to live with. It is entirely possible to brew a better Pils, for example, by augmenting calcium chloride such that the RA is much more negative that it would be for Pilsen.

A few years back people discoverd Kolbach's paper (you can see a translation and www.wetnewf.org) and got all excited about RA as the path to perfect brewing water. It got to be a solution in search of a problem to some extent. All in all I'd say that being aware of RA is important as it does seem to give an indication of expected mash pH shift relative to DI water (though, as you will see if you read Kolbach's paper the shift he calculated is really intended to describe knockout wort) but I believe adhering to the 3 points you enumerated is more important than matching RA. If you get the pH right the rest is a matter of tweaking this ion or that for taste.

Lot's of opinion there. Hope others respond.
 
The problem with RA is, that it focuses too much on the water. While I think RA is a useful means for expressing the water's ability to change mash pH it is not the whole story. You'll need to take mash thickness into account since both the amount of water and its RA matter for changing mash pH since the pH buffer in the mash is established by the grain and not the water. At mash pH the water's buffer capacity is pretty much exhausted and even for very high alkalinities the buffer strength of water is nowhere near that of the malt.

Maybe RA*R were R is the mash thickness is a better indicator for how much the water will change mash pH. But that may lead people to think that thick mashes are the way to compensate for high RA which should not be done.

I think some guidelines on RA are reasonable and better than no attention to water at all.

It may also be true that the factor of 1/3.5 that is applied to the calcium hardness is not necessarily a constant either, which changes the actual alkalinity neutralization that you are getting from calcium and magnesium.

Kai
 
On that same note as Wildwest; that range for the dark beers seems so excessive. I've recently brewed several stouts and didn't even approach the low end of that range (120) and my mash pH is in the 5.4-5.6 range. Then again, stouts can vary widely on the roasted and crystal malts and I tend to be on the conservative side there.

Using the HtB calculator, pH strips, and the recommendations there had me using 100% relatively alkaline water and then adding chalk and/or baking soda. Now I cut it with RO and add CaCl and the mash pH is ~5.5. It's no wonder my dark beers sucked.
 
To put some perspective on the mash thickness aspect of this I modeled a mash consisting of 1 kg malt in 2 L water as a pH 5.2 phosphate buffer with about 10 grams of phosphate (the nominal phosphate content of 1 kg of malt). Diluting that mash by doubling the water (i.e. to 4L/kg) with low ion water would only cause the pH to shift to 5.21. A buffer is supposed to hold its pH under dilution and stress and this shows that such a strong buffer does resist dilution well (with RO or DI water). But if the water is alkaline to the extent of 80 ppm as CaCO3 doubling the water would raise the pH to 5.59. A more reasonable dilution to 3L of 80 ppm water per kg would result in a pH shift to 5.34. So even though there is way way more phosphate in the malt than alkali in the water that alkali does stress the buffer and pull pH around a bit.

Real mash is much more complicated than my model but the general concept is the same. Just further evidence that "Alkalinity = Bad". Note that if the alkalinity in the water is from calcium bicarbonate that the calcium will offset the effect of the alkalinity somewhat (the effective alkalinity would be 80*(2.5/3.5) = 57 and the pH shift to about 5.31 (assuming the 3.5 ratio is correct which I too doubt is the case in many situations - 3.5 is just too round a number).
 
So even though there is way way more phosphate in the malt than alkali in the water that alkali does stress the buffer and pull pH around a bit.

The way I see it, and have it implemented in my spreadsheet, is that mash can be seen as a strongly buffered system and that any acid or alkalinity added just moved the pH based on the amount that was added and the amount of buffer substance that is present in the mash. Being buffered doesn't necessarily mean that mash is completely resistant to pH changes. This model makes it much easier to think about mash pH and predict changes since the math involved is a simple linear function.

The substance that is providing the buffer doesn't matter all that much. I wonder how well a phosphate buffer can be used as a model given that the mash pH pretty much between two of its pKas.

Kai
 
The mash is pretty much midway between the pK's of phosphate which is why it isn't a very good buffer and why it is so susceptible to drift from dilution and alkali. This is why the Five Star 5.2 isn't effective as we have noted many times. But if you get mash to pH 5.2 and there is phosphate in it (and there is) then the phosphate will be distributed such that the dibasic form is only about 1% of the total and the monobasic form about 99%. This is not a good buffer design, certainly, but it does have buffering capacity as is demonstrated when you dilute or stress it with acid or base. The fault with the model is that it does not consider other buffering systems which are also present. I assume phosphate to be the dominant one simply because there is so much of it in mash but certainly other acids in various stages of dissociation would be present too if dark malts or especially sauermalz is in use.

The action definitely isn't linear but I think you can approximate it by a linear function if the amount of stress isn't too great. For the 3 L/kg vs 2 L/kg model the slope is 0.0017 pH/ppm as CaCO3 at low levels but reduces to 0.0004 pH/ppm as CaCO3 at 400 ppm. Now 400 ppm is pretty high. At 100 ppm the slope is 0.00133 which isn't that much of a change for 0.0017
 
Here are two scenarios about PH and RA for your imput.

This is a water report, soft water:
Ca=8
Na=2
Mg=1
SO4=1
Cl=4
Total hardness CaCO3=24
Alkalinity as CaCO3=20
Bicarbonate HCO3=25PH=7.8

note. calculation from EZ water calculator 2.0:


1)Brew a recipe for a ligh ale with about 10lb grist, no crystal nor roasted malts, SRM=3, mash water =3.8gals

Starting mash PH is high(5.6). Adding 3g of CaCl2 and 2g of epsom salt and 2g of lactic acid will make my mash PH =5.36/ RA would be -200.
Why adding this instead of just lactic acid? To build some minimum Ca, Cl/SO4 ratio, getting PH down as a plus.
What about -200, is that ok, although much lower than the range suggested by Palmer (-60 to 0)?

2)Brew a recipe for a stout with 13.8lb grist, 2lb crystal and 3/4lb roasted, SRM=36, mash water=5gals

Starting PH is low (4.90). SO4 and Cl are almost nothing as well as Chloride. Gotta pick something to add that rise PH right, baking soda or chalk?
If adding baking soda only, would need 10g
If adding chalk only, would need 30g :drunk:

I assume this much chalk would make my beer taste like drywall:cross:

A combination of both, like 8g baking soda and and 2g chalk would bring my PH up to 5.20 (good).
My Ca would also rise to 50.
At this point, sulfate is still nothing, do I care? probably not right, otherwise I wold have to add somehitng that woud lower my PH again.
Finally, the RA after the 8g of baking soda and 2g of chalk would be 287. Good?

Edit: For the stout, before Ajdelange hit me with a hammer :) , should one with a good PH meter check the mash PH first, then decide if adding chalk or bicarbonate is necessary to increase PH?
 
2)Brew a recipe for a stout with 13.8lb grist, 2lb crystal and 3/4lb roasted, SRM=36, mash water=5gals

I recently updated what %roasted means in my mash pH prediction based on SRM and %roasted. It should be the percentage of color that is coming from roasted malts. Before it said % of roasted malts among the specialty malts. That can make a significant difference in the mash pH.

With that I get a distilled water pH estimate of 5.1, for which you would need only 1 g baking soda and 5g undissolved chalk to bring the pH to ~5.25.

Kai
 
I recently updated what %roasted means in my mash pH prediction based on SRM and %roasted. It should be the percentage of color that is coming from roasted malts. Before it said % of roasted malts among the specialty malts. That can make a significant difference in the mash pH.

With that I get a distilled water pH estimate of 5.1, for which you would need only 1 g baking soda and 5g undissolved chalk to bring the pH to ~5.25.

Kai

Thanks. I downloaded your updated file.
 
I recently used a more analytical approach to derive a SRM->grist pH formula and with that approach it was more apparent that you want to use the %of SRM that is coming from the roasted malts. A good approximation for %SRM in this case is %MCU (Malt color units), which can be calculated from the grist more easily.

Kai
 
Indeed, RA is an imperfect predictor of mashing performance. That performance is a result of carbonate and phosphate buffering systems that at times compliment or compete.

But given that, in the small pH range that brewers typically deal with it, does provide some guidance. As Kai has illustrated, RA is completely incapable of providing pH prediction as the mash pH falls below 5.1 or 5.2. Further reducing RA into negative values by increasing hardness results in the mash pH only falling to the low 5 range and not falling any further. The only way to get the mash pH to drop further is through acid addition. I'm not sure why anyone would want to take pH this low, but I offer it only as an example of the non-linearity of RA in practice.

The other component that complicates the issue is the attempt to tie RA with beer color. As Kai again illustrated, acidity vs grain color has odd relationships. As a graph on my Bru'n Water website illustrates, the amount of acidity of a grist can have wide variation as the beer color increases. But, there is still a broad correlation that the acidity provided by the grist does generally increase with increasing color. That does indicate that RA can GENERALLY be correlated to beer color since the grist needs more alkalinity (increased RA) to counteract the increasing acidity. Its just that it can be imprecise.
 
Here are two scenarios about PH and RA for your imput.

This is a water report, soft water:
Ca=8
Na=2
Mg=1
SO4=1
Cl=4
Total hardness CaCO3=24
Alkalinity as CaCO3=20
Bicarbonate HCO3=25PH=7.8

note. calculation from EZ water calculator 2.0:

To get this level of bicarbonate (and balance the profile) your alkalinity would have to be 45.


1)Brew a recipe for a ligh ale with about 10lb grist, no crystal nor roasted malts, SRM=3, mash water =3.8gals

SRM 3 is a dream for homebrewers. The lightest I have ever gotten was a bit over 4 using a short boil carried out with steam. Unless you dilute with rice or corn I don't think you can get that pale.

Starting mash PH is high(5.6). Adding 3g of CaCl2 and 2g of epsom salt and 2g of lactic acid will make my mash PH =5.36/ RA would be -200.
Why adding this instead of just lactic acid? To build some minimum Ca, Cl/SO4 ratio, getting PH down as a plus.
What about -200, is that ok, although much lower than the range suggested by Palmer (-60 to 0)?

Using 45 as the alkalinity the RA of this water is 39 for an anticipated upward pH shift of 0.07. Ale malts seem to have DI mash pH of around 5.6 (as you note) so you might expect 5.67 with no treatment. Adding the salts (to 5 gal) would get the RA down to -33 and the pH shift to 0.06 in the right direction so now the expected pH is about (because the pH shift thing is approximate) to 5.54.

Adding 2g of lactic acid to the water would lower its RA to -88 but that's really meaningless except in the mathematical sense since the pH would drop to 3.55 and alkalinity (from which RA is calculated) isn't really defined below pH 4.3. Blindly applying Kolbach's shift shows -0.15 from DI mash pH or 5.45. A much more reasonable approach is to add the lactic acid to the mash, consider that sauermalz is nominally 2% lactic, use this to estimate the equivalent sauermalz as 2/0.02 = 100 grams of sauermalz, note that this is 100/4540 = 2.2% of the grist and use the rule of thumb to calculate a 0.22 pH drop for a final estimated pH of 5.6 - 0.06 - 0.22 = 5.32.

Of course it doesn't make any difference whether you add the lactic acid to the mash or water. It all gets mixed in with the grain anyway but there is one huge advantage to adding it to the mash: you can do it in increments i.e. add half of it and see where the mash pH goes then add the other half if you need it. One of the advantages of sauermalz, IMO, is that the only "calculation" you need to do is 1% per 0.1 pH. Another is that it's harder to slip with sauermalz and overdo it.

2)Brew a recipe for a stout with 13.8lb grist, 2lb crystal and 3/4lb roasted, SRM=36, mash water=5gals

Starting PH is low (4.90). SO4 and Cl are almost nothing as well as Chloride. Gotta pick something to add that rise PH right, baking soda or chalk?
If adding baking soda only, would need 10g
If adding chalk only, would need 30g :drunk:
Without knowing anything about the crystal I can't WAG the mash pH but it's very unlikely it would be 4.90. Probably more like 5.3 - 5.4.

I assume this much chalk would make my beer taste like drywall:cross:
I've seen a couple of guys say this and it conjures up strange images. How do you all know what drywall tastes like? But yes, I have had reports from people that have followed this line of reasoning and gotten beers that tasted chalky.

A combination of both, like 8g baking soda and and 2g chalk would bring my PH up to 5.20 (good).
My Ca would also rise to 50.
At this point, sulfate is still nothing, do I care? probably not right, otherwise I wold have to add somehitng that woud lower my PH again.
Finally, the RA after the 8g of baking soda and 2g of chalk would be 287. Good?

Edit: For the stout, before Ajdelange hit me with a hammer :) , should one with a good PH meter check the mash PH first, then decide if adding chalk or bicarbonate is necessary to increase PH?

I just put the hammer back in the drawer. The golden rule of brewing is "Never add chalk or bicarbonate to brewing water and never add it to mash unless a reading made with a recently calibrated pH meter tells you it is necessary". All this estimating, use of spreadsheets etc. is mooted if you have a pH meter. It's fine to be informed by the spreadsheets but be guided by your pH meter.
 
SRM 3 is a dream for homebrewers. The lightest I have ever gotten was a bit over 4 using a short boil carried out with steam. Unless you dilute with rice or corn I don't think you can get that pale.
It should be noted that the SRM used for the prediction should come from the standard SRM estimation that brewing tools do. Just because we need SRM and %roasted as a proxy for the grist acidity. Obviously you can make an SRM 10 beer out of pils by boiling the crap out of it but in the mash the “SRM” was still around 3. I would have liked to use average grist color for the mash pH estimations, but that is not a number that is easily reported or is a number that brewers have in their mind when they think beer color.

You’ll have to keep in mind that unless malt analysis sheets contain useful pH and acidity information all mash pH prediction is color based and has the same inherent limitations. May it be entering beer color or the color of individual malts, the underlying formula is pretty much the same.

Using 45 as the alkalinity the RA of this water is 39 for an anticipated upward pH shift of 0.07. Ale malts seem to have DI mash pH of around 5.6 (as you note) so you might expect 5.67 with no treatment. Adding the salts (to 5 gal) would get the RA down to -33 and the pH shift to 0.06 in the right direction so now the expected pH is about (because the pH shift thing is approximate) to 5.54.
I added the salts and acid to the 3.8 gal mash water he mentioned and I get a final pH drop of 0.12. It should be noted that acid malt tends contain 3% lactic acid by weight even though Weyermann lists it as containing at least 2% lactic acid. The two samples I tested had closer to 3% lactic acid in them.

Without knowing anything about the crystal I can't WAG the mash pH but it's very unlikely it would be 4.90. Probably more like 5.3 - 5.4.
Most of the color is likely to come from roasted malts. With 70% color from roasted I get 5.2 as the grist pH and with 90% I get 5.3, This shows how little the estimated mash pH changes when the %roasted is changed quite dramatically and which is why you may be able to eyeball it at some point. When you let the crystal malt dominate the color (40% roasted for example) you’ll start to see a much lower grist pH (5.0 for the example)

I've seen a couple of guys say this and it conjures up strange images. How do you all know what drywall tastes like? But yes, I have had reports from people that have followed this line of reasoning and gotten beers that tasted chalky.
I hear what you are saying. But I think I tasted drywall when I mimicked Burton-Upon-Trent water where I added lots of gypsum before I tasted the water. The beer came out fine, but I don’t think I’ll be making water that loaded with minerals again.


I just put the hammer back in the drawer. The golden rule of brewing is "Never add chalk or bicarbonate to brewing water and never add it to mash unless a reading made with a recently calibrated pH meter tells you it is necessary". All this estimating, use of spreadsheets etc. is mooted if you have a pH meter. It's fine to be informed by the spreadsheets but be guided by your pH meter.
I have to disagree a little here. I routinely add alkalinity to the brewing water before dough-in. I have been pretty good in predicting the amount needed based on past batches and using my water spreadsheet. I like to add everything to the brewing water or grist (in the case of Sauermalz) such that when I take the mash pH after 5-10 min I don’t have to adjust the pH. There might be cases when I venture into grist compositions that are very different from what I have been using before where I’m a bit more cautious. But I also found if the amount of salts that I’m planning to add look way too much, then they are probably too much.
But that’s all up to the individual brewer and his or her style.

Kai
 
"To get this level of bicarbonate (and balance the profile) your alkalinity would have to be 45"

Does this mean my water report from Ward is wrong?
Should I use 45 and the rest of the report is ok?
 
"To get this level of bicarbonate (and balance the profile) your alkalinity would have to be 45"

Does this mean my water report from Ward is wrong?
Should I use 45 and the rest of the report is ok?

I ran your numbers though my water spreadsheet and it reports and ion balance of -0.3%, which is as good as it gets for ion balance. But for this I selected “SO4-S mg/l” as the unit for sulfate. When using “SO4 mg/l” I get an ion balance of 1.6%. But since your sulfate is so low, it doesn’t matter much.

The bicarbonate and alkalinity seem to match as well. For 20 ppm as CaCO3 alkalinity I get 24 ppm HCO3. I neglect pH, but that seems fine at pH 7.8.

A.J. will have a more precise answer, I assume.

Kai
 
No, that was my mistake. Alkalinity depends on the pH used in the titration when it is measured. I'd been doing some bizarre stuff with end points of 3.3 and left that in my spreadsheet. If your lab had used 3.3 as the end point then my comment would have pertained. Switching to 4.3 (a more nominal number but who knows what they actually use) the profile balances fine.
 
Back
Top