I was asked a question about the CH test and how it works so thought I'd put this into a thread in The Deep End to share with anyone interested.
The CH test is an EDTA titration with inhibitors to prevent (other) metal interference (say from copper or cadmium). EDTA binds to both magnesium and calcium. The indicator dye for total hardness is usually Eriochrome Black T, but that is not stable at the very high pH needed for the CH test (as opposed to testing for Total Hardness) so instead hydroxynaphthol blue is usually used, though Taylor uses a variant called Calcon™ which is 1-(2-Hydroxy-1-naphthylazo)-2-naphthol-4-sulfonic acid sodium salt aka Eriochrome® Blue Black R aka Mordant Black 17 aka Palatine Chrome Black 6BN whose structure is nearly identical to Eriochrome Black T except that the nitro functional group is removed (replaced with hydrogen). Calcium and magnesium ions complex with these dyes to form a red color whereas when no ions are complexed the color of the dye is blue. EDTA binds to magnesium and calcium so titrates them away from the dye.
The high pH from the "calcium buffer" (which is sodium hydroxide) is how one distinguishes calcium hardness from total hardness because a high pH of around 13 precipitates magnesium hydroxide (for the total hardness procedure the pH used is around 10 for proper EDTA titration that avoids complexing with iron and reaction with the indicator dye). This high pH can co-precipitate some calcium (making the CH reading too low) and can make the endpoint difficult to discern. Though Taylor says to add a few drops of titrant initially to help with this, technically one could add up to 95% of the titrant which one can determine by doing the procedure quickly at first to get a rough endpoint and then redo adding most of the titrant up-front. This gets the calcium down low so that relatively little co-precipitates with the magnesium. Though the initial titrant binds to magnesium as well as calcium, the sodium hydroxide (calcium buffer) that is added will precipitate the magnesium releasing some EDTA, though the EDTA binds more strongly (preferentially) to calcium than to magnesium (by about a factor of 100 when both are present in equal amounts). So in practice, the procedure prevents the co-precipitation of calcium giving a clearer endpoint and a more accurate CH reading. Adding EDTA titrant up-front also removes other metals that could react with the indicator dye.
Though calcium carbonate would be formed when the pH is raised high, apparently EDTA binds strongly enough to calcium to dissolve it whereas magnesium hydroxide (at pH 13) binds more strongly than magnesium to EDTA so does not get dissolved.