Water chemistry book by Palmer and Kaminski

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.

monkeyman1000

Well-Known Member
Joined
Nov 10, 2013
Messages
380
Reaction score
66
Hello, new guy on here. I was wondering if someone could help me with some questions I had about the new water chemistry book. It is a great read and very informative and well worth the money by the way.
1) The Buffering Capacity graphs created by deLange- The two graphs show the buffering capacity from -20 to -100 mEq/pHkg and -25 to -150 mEq/pHkg. Is the buffering capacity more at -100 and -150 or on the other end of -25 and -20.
2) Palmer gives some great examples to work thru in chapter 7. Earlier in the book he introduces the concept of Z Alk and Z RA but from what I can tell in the examples, Kolbach's RA is used. When we are figuring out our own water would it be better to use the Z Alk and Z RA?
3) With regards to checking mash pH deLange shows that time is definitely a factor. When should the mash pH be checked ( 30 minutes after adding the grains for example) and is their enough time to make an adjustment at that point? Or do you make changes to the next time you brew the same recipe and try to dial it in over successive attempts?
Thanks in advance for the help and again, this is really a book that should be in every brewer's library.
 
1) The Buffering Capacity graphs created by deLange- The two graphs show the buffering capacity from -20 to -100 mEq/pHkg and -25 to -150 mEq/pHkg. Is the buffering capacity more at -100 and -150 or on the other end of -25 and -20.

The magnitudes of the numbers are the buffering capacities. Thus if the number from the graph is the -100 mEq/kg-pH that means it takes 100 mEq to move the pH of a mash containing 1 kg of the malt 1 pH unit. The reason for the minus sign is that adding protons (acid) results in a decrease in pH. Or put another way the curves in Fig. 20 represent the slopes of the curves in Fig. 19 which are all decreasing with increasing pH for all values of pH.

2) Palmer gives some great examples to work thru in chapter 7. Earlier in the book he introduces the concept of Z Alk and Z RA but from what I can tell in the examples, Kolbach's RA is used. When we are figuring out our own water would it be better to use the Z Alk and Z RA?

On p170 he uses the 'conventional' RA to work an example in which he wants to raise the 'conventional' RA to 150 and starting on p171 he raises the zRA to 150 and gets different answers. This isn't unsurprising as RA is defined as RA = alk - (Ca/3.5 - Mg/7). There are 3 possible values of RA here

RA(4.3) = alk(4.3) - (Ca/3.5 - Mg/7)
RA(4.5) = alk(4.3) - (Ca/3.5 - Mg/7)
RA(pHz) = alk(5.4) - (Ca/3.5 - Mg/7)

where pHz = 5.4, the desired mash alkalinity. 4.3 is the value which makes the alkalinity the M or total alkalinity and 4.5 the pH for the ISO method of alkalinity specification. In the two examples he gives (without specifying end point, i.e. 4.3 or 4.5, for the first) he shoots for the same RA value, 150. Assuming that some level of RA is desired the target for RA(pHz) should be adjusted down as RA referenced to a higher pH (5.4 > 4.5) will be smaller than the RA referenced to a lower pH. I tried to get John away from such heavy reliance on RA but he is still of the opinion that it is the key to all this and it is his book.

As I don't think there is much basis for an RA of 140 or 150 or 160 for this particular beer I don't think it matters much which of the two RA's you use. If you know something about the malts then you won't be calculating RA anyway but will be approaching the problem via the proton deficit method (which is hinted at in the book but not fully developed because there wasn't time). Using the proton deficit method you would calculate the proton deficit of the water with respect to it's pH and pHz. This is obtained from Ct = 1.94 (which he calculates on p172) multiplied by the number of protons required to lower the pH of 1 mmol of Ct from 9 to pHz. From the graph on p127 this is -.1 - -1.04 = 0.94. This results in a proton deficit of 0.94*1.94 = 1.82 mEq/L. To determine how much alkalinity is necessary you would have to multiply by the total amount of water to determine the water's deficit and add to that the deficit of each base malt and light colored specialty malt and subtract the surfeits of any highly colored malts. If the sum of all these is a negative number (a proton surfeit) then you would have to add that much base to absorb those extra protons thus bringing you to pHz. I think the main flaw here is trying to push you to a particular RA when in fact you need to be considering the effects of the malts. There may be a correlation between beer color and malt spec but it is a loose one.


3) With regards to checking mash pH deLange shows that time is definitely a factor. When should the mash pH be checked ( 30 minutes after adding the grains for example) and is their enough time to make an adjustment at that point? Or do you make changes to the next time you brew the same recipe and try to dial it in over successive attempts?

Initially take as many measurements as you can and write them down with the times you took them. Plot these out. It won't take you long to be able to 'lead' your pH meter readings by which I mean have a pretty good idea as to where they are going to level out before they actually do.

Most of the change should have taken place in the first 20 minutes or so. I'd say that unless things are drastically off you should defer adjustment until the next brew. Making a test mash with a portion of your grist (well mixed) and some of the water you intend to mash with should prevent things being drastically off.
 
Thanks for reply. I've been following your advice from the water chemistry sticky for the past year and found it to be helpful but I want to get more in depth now. I move every 3-6 months for work so I should get some good practice with all this over the next few years. Cheers to you :mug:
 
I'd like to ask an additional question about the book that I can't quite reconcile. On p.68, two formulas are given for RA, one in mEq/L and a revised formula in ppm as CaCO3. Both are (summarized) Alkalinity - Ca factor + Mg factor. But later in the examples section, p.161, the RA formulas don't match up. The mEq/L formula is given as Alkalinity - Ca Factor + Mg factor, where the ppm formula is given as A - Ca - Mg (calcium and magnesium are added in all previous examples, but are subtracted in the last ppm formula). I would have believed this is a simple typographical error (though an important one), but: the American Pale Ale example uses the A - Ca - Mg formula on p. 163 and p.166, but in the Foreign Extra Stout example on p.171 the A - Ca + Mg formula is sued.

I'm operating under the assumption that this is errata in the book, as the sum of the two minerals makes more sense based on discussion earlier in the book, but wanted to make sure I'm not missing some unit conversion or other reason that the formulas used would change.
 
I'd like to ask an additional question about the book that I can't quite reconcile. On p.68, two formulas are given for RA, one in mEq/L and a revised formula in ppm as CaCO3. Both are (summarized) Alkalinity - Ca factor + Mg factor.
No. The first formula is

RA = alkalinity -[(meq/L Ca)/3.5 + (meq/L Mg)/7]

The only 'error' there is that it is usual to represent concentrations by brackets though these are concentrations in mEq/L and brackets usually represent mmol/L. Nevertheless

RA = alkalinity - ( [Ca]/3.5 + [Mg]/7 )

would look more familiar. This 'divide the calcium concentration, be it in mEq/L or ppm as CaCO3 by 3.5 and add that to the magnesium concetration divided by 7 and then subtract the sum from the alkalinity expressed in the same units as the concentrations. Thus

RA = alkalinity - [Ca]/3.5 - [Mg]/7

But later in the examples section, p.161, the RA formulas don't match up. The mEq/L formula is given as Alkalinity - Ca Factor + Mg factor, where the ppm formula is given as A - Ca - Mg (calcium and magnesium are added in all previous examples, but are subtracted in the last ppm formula).

On p 161 he has

RA = Alkalinity - ( (Ca/3.5) + (Mg/7) )

and, further down,

RA = Total Alkalinity - [Ca]/3.5 - [Mg]/3.5

These are consistent (ignoring interchange of brackets and parentheses) between them and with the earlier formulas and correct.

I would have believed this is a simple typographical error (though an important one), but: the American Pale Ale example uses the A - Ca - Mg formula on p. 163 and p.166, but in the Foreign Extra Stout example on p.171 the A - Ca + Mg formula is sued.

On p 171 he has

150 = X - (40/1.4 + 9/1.7)

and that's the same as

150 = X - 40/1.4 - 9/1.7


I'm operating under the assumption that this is errata in the book, as the sum of the two minerals makes more sense based on discussion earlier in the book, but wanted to make sure I'm not missing some unit conversion or other reason that the formulas used would change.

What you are missing is the 'Distributive Law' of algebra:

A*(B + C) = A*B +A*C

In this case A = -1
 
From the book, I understood it is possible to create a curve fit for the malt titration data (such as the cubic function on p.88).
Thus we can renounce the malt buffering capacity concept and stop trying to linearize the titration curve, which is one of the sources of errors when predicting mash pH.
Even if it was impossible to make a fit for some malts, we still could use the data for the spreadsheet (using =vlookup function).
Is it just a matter of time before we start using those titration functions for every malt instead of the malt buffering capacity?
 
From the book, I understood it is possible to create a curve fit for the malt titration data (such as the cubic function on p.88).

Seems to be.

Thus we can renounce the malt buffering capacity concept and stop trying to linearize the titration curve, which is one of the sources of errors when predicting mash pH.

We aren't abandoning the concept of malt buffering - we're just recognizing that it isn't a constant but in fact a function of pH i.e. that the titration curve is not linear.


Even if it was impossible to make a fit for some malts, we still could use the data for the spreadsheet (using =vlookup function).

It's hard to imagine a malt titration curve that would not be adequately fit by a third order expansion within the limits of pH of interest. But higher order fits are possible. Not every function is amenable to Taylor series expansion but the malt data I have so far (the three curves in the book plus one more) are nearly linear so three terms is plenty.


Is it just a matter of time before we start using those titration functions for every malt instead of the malt buffering capacity?

Somebody has to get the data and that's very time consuming. It seems to take at least a dozen measurements to get a good set of coefficients and each measurement takes about an hour including the time required to get things up to temperature and the calibration runs. What we don't know about at this point is how closely the sack of Weyermann Pils I measured last spring compares to the sack you buy this winter or how closely it compares to a sack from some other maltster. Weyermann's pneumatic pils and their floor pils are quite different. Obviously the thing to do is get the maltsters on board to the point that they do the titration and publish the coefficients along with the other per lot data on their spec sheets. Don't hold your breath.

For more details on this see http://wetnewf.org/pdfs/Brewing_articles/MBAA_FREDERIC.pptx. There's a spreadsheet too.

Shouldn't need to use vlookup to get into malt measurement data but it looks like an easy way to get into a malt coefficient data base!
 
Back
Top