Can you expound a bit or point me to a good source on that?
I'll do both!
In mEq units for all components involved, RA is expressed as follows:
RA = Total_Alkalinity -(Ca/3.5 + Mg/7)
In the 1950's (or before) German brewing scientist Paul Kolbach was determining the degree of pH drop exhibited from the onset of mash to the terminus of knockout (post boil and cooling). Based upon the malt(s) he used, he derived the divisors of 3.5 and 7 seen in the equation above. He defined RA as that portion of initial alkalinity which survives through to knockout, when the mash water (and sparge water) have extant calcium and magnesium ions.
Somehow this got contorted initially into a presumption of mash exclusive conditions, totally ignoring that it was intended only for knockout.
Then the divisors of 3.5 and 7 got elevated to the status of unquestioned and definitive truth. And thus it stood until:
In about 2016 a research chemist named Roger Barth and his graduate student assistant Rameez Zaman undertook to duplicate and verify Kolbach's work, while limiting it exclusively to the mash only (such as how that is what everyone does, for reasons totally unknown, and in defiance of the actual Kolbach 'knockout pH' in the presence of calcium and magnesium ions research).
They chose three base malts. Pilsner, Pale, and Munich. They concentrated upon researching Calcium only, excluding magnesium. And they observed during the mash only. What they found through repeated study was that the 3.5 Kolbach divisor for calcium spanned from 7.2 to 14.2 (from memory here) within the mash depending upon which of the three malts they were mashing at the time. So in respect to the mEq (milliequivelents) equation, it becomes:
RA = Total_Alkalinity -[Ca/(~7 to ~14) + Mg/(?)]
Who knows what malts and unmalted grains other than the specific lots of their three chosen representatives might turn up with for their denominator values. All that Barth and Zaman proved was that given lots of given malts did indeed have repeatable denominators (just as Kolbach discovered, albeit that he was (1) looking at knockout, and not the mash, and (2) he perhaps limited his research to a single malt type and species and lot). Again, off the top of my head, I believe they assigned a calcium divisor of ~14.2 to their Pilsner malt, a divisor of ~7.2 to their pale malt, and a divisor of ~12.3 to their Munich malt. Mind you that these are specific lots analyzed at certain ages and your lots and malt species will likely not be in line with theirs. And a few months later, theirs may no longer be in line with theirs... And perhaps if they chose a different crush their denominator values would have been different... Etc...
Thus RA is and will always be an unknown and contrived derivative conception that can no longer stand due to what were presumed fixed value divisors being found to be quite wildly variable. Barth and Zaman's published and peer reviewed research paper on this matter is titled "Influence of Strike Water Alkalinity and Hardness on Mash pH". It must be purchased.