Automatic salt addition calculations?

[BrunDog]

Hi peeps,

I have been using some of the different water calculators (Beersmith, EZ Water, Brewers Friend, Brun water.). Are there any worthwhile that automatically calculate the amount of each salt to match the desired profile? Beersmith is the only one as far as I can tell but it is severely lacking in many other ways.

If not, is there a documented "best" method other than manually working it out?

-BD

 

[ajdelange]

Hi peeps,

If not, is there a documented "best" method other than manually working it out?

It is wise to put 'best' in quotes because there a various definitions of that term. In my own work I minimize the sum of the weighted squares of the logs of the ratios of the realized ion concentrations to the target (profile) concentrations using Excel's Solver to find the salt additions that give the minimum. Solver is, once you tell it what to do, automatic but it does essentially what you do manually, that is change salt addition amounts accepting any change that reduces the concentration errors.

Looking for mmse in p(concentration ratio) leads to uniform percentage concentration errors rather than uniform mg/L concentration errors. User selectable weights allows you to focus on one ion's (say sulfate's) error at the expense of other ions.

Conceptually you can use Solver with any Excel based spreadsheet except Brun Water in which the cells are locked so Solver can't change them.

 

[BrunDog]

Thanks AJ. I assume you are referring to Goal Seek? I use that regularly, but wasn't sure what to seek for.

-BD

 

[ajdelange]

No. It is similar but much more powerful. It is made by a company called Frontline Systems and usually included with the Excel distribution. You may have to install it. See http://www.excel-easy.com/data-analysis/solver.html. You do have to tell it what sort of optimum you want it to solve for. It needs a command of the general form of "Set cell A1 to some value or to the minimum value by changing the values of cells B1, B2... subject to the constraint that B1 > 0, B2 > 0..." A1 then needs to reflect the overall error and can be something like
A1 = sum ( (desired_ion_concentration - ion_concentration_realized_by_the_current_salt_additions)^2)
or, as I mentioned in the last post
A1 = sum ( p(desired_ion_concentration - ion_concentration_realized_by_the_current_salt_additions)^2)
where the p operator is p(•) -log(•) i.e. the same operator as in pH

 

[BrunDog]

Got it. I now recall using it years ago and just assumed it went the way of the dodo. Thanks for the help once again!!

-BD

 

[ajdelange]

It did disappear for a while, that is Microsoft stopped including it with Excel, at least for the Mac. Frontline was smart enough to make it available free to any Excel purchaser through download from their website and eventually they and Microsoft got back into bed with each other. Excel is Microsoft's only good product and Solver is a big part of what makes it so.