Biollante;78288 said:
By definition when we reduce the “hardness” of the water, we are reducing the conductivity, total dissolved solids (TDS) and osmotic pressure.
By definition, Total Dissolved Solids is a measure of all dissolved molecules and ions as well as colloidal suspensions within an aqueous solution while Total Hardness is a measure of the divalent metallic ionic Molarity present in the solution. Since we are merely trading out divalent metallic ions for monovalent metallic ions while maintaining charge balance, I do not understand the assumption that TDS reduces with TH in the case of a salt-based DI unit.
In the example I used in post #4, the conductivity went from 0.000134-[SUP]mho[/SUP]⁄[SUB]cm[/SUB] to 0.0000625-[SUP]mho[/SUP]⁄[SUB]cm[/SUB] (134-[SUP]µS[/SUP]⁄[SUB]cm[/SUB] to 62.5-[SUP]µS[/SUP]⁄[SUB]cm[/SUB]), TDS went from 86-ppm to 40-ppm and osmotic pressure from 0.040038-bar to 0.039538-bar (assuming 25C).
For the same reason you provided the Morton chart, we can only make assumptions as to the starting and ending characteristics of the problem that might not bear fruit on the poster's situation. The OP hasn't even stated whether or not the supply is municiple or from a well. The only information we can glean lies within the OP's profile, which places the location as Edmonton and that the OP is a De-water tech (what is that, exactly?) because the question was posed in the simplest of forms without any background whatsoever.
In this case, we are exchanging Ca[SUP]2+[/SUP] and Mg[SUP]2+[/SUP] ions for two Na[SUP]-[/SUP] ions based again on assumptions, not knowing if Iron or any of the other TH components are present, but since we are dealing with a salt-based DI then it is fairly safe to assume that only the readily-formed chlorides within the exchange medium are affecting but the calcium and magnesium components, with the preponderence most likely to the calcium. But as we are dealing with a salt-based DI, then the salt most likely used is Potassium as many manufacturers choose this over Sodium chloride in order to decrease Sodium exposure for the sensitive.
Calcium has a thermal conductivity of 201 Wm[SUP]-1[/SUP]K[SUP]-1[/SUP].
Potassium has a thermal conductivity of 102.5 Wm[SUP]-1[/SUP]K[SUP]-1[/SUP].
So if we are indeed exchanging for Potassium then conductivity would drop a bit as Potassium has only about 80% of Calcium's ionization energy.
But...but if the DI is Sodium based, then we're looking at a thermal conductivity of 142 Wm[SUP]-1[/SUP]K[SUP]-1[/SUP], a 40% increase in conductivity to more than negate the difference in ionization energies between Sodium and Calcium.
Additionally, you provided an example based on an ambient of 25C, a reasonable assumption based on what is probably the majority of cases with community or planted aquaria. There are, for example, many soft water fish that prefer much higher temps. My main room has a heavily-planted, CO[SUB]2[/SUB]-enriched tank full of a dozen
Dicrossus filamentosus, a half-dozen
Otocinclus affinis and a dozen-and-a-half
Corydoras habrosus at 31C and about 2dH (50-50 mix of tap and RO). Conductivity generally increases with temperature, and in a lot of cases actually increases with a drop in Molarity. As you can see in this
simplistic chart from Wikipedia, the difference between 25C and 31C at lower Molarities can account for a 17% increase in conductivity were the solution straight Potassium chloride.
The two biggest obstacles to breeding softwater fish outside of lighting are tanin concentrations and osmotic regulation. Too much tannin will harden the egg membrane to the point of trapping the emergent fry within, and an increase in osmotic pressures due to salt content will essentially "dessicate" the egg as water moves from the egg to the environment.
Either TFH or FAMA (I'm thinking FAMA) ran an article in the 1998-1999 time frame on the effects of salt-based DI exchange on different localities which was both quite revealing and very much more detailed. I will try to track down the specific article, but I can't garantee success in the search.
So, on the basis of so many assumptions as to the starting and ending conditions I'm just not comfortable with the idea of such a broad generalization even though it applies to your own specific case...