No, that isn't safe to say, which more or less invalidates everything else in that paragraph.

We can always move the more technical part of the discussion to The Deep End, if needed.

TDS is used in the pool/spa industry as a proxy for water age to figure the amount of buildup of unoxidized organic material that is in the water. However, it should be obvious that keeping a pool covered and not using it except to add chlorine regularly will buildup salt and nothing else. Compare this with a high bather load pool where a lot more chlorine needs to be used to oxidize bather waste, but where the loss from sunlight is comparatively low. In this latter case, the buildup of unoxidized organics will be quite high. Hence, looking at TDS alone is really ridiculous. The rule-of-thumb for water change or dilution is much better and is based roughly on bather-hours. For spas, it is the following:

Water Replacement Interval (WRI) = (1/3) x (Spa Volume in U.S. Gallons) / (Number of Bathers per Day)

where the assumption for soak time is around 20-30 minutes. In practice for residential spas, users can usually go twice as long as the above before changing the water when using the Dichlor-then-bleach method. If using Dichlor-only, one usually needs to change the water sooner -- roughly at the time indicated above -- mostly due to CYA buildup that slows down oxidation (and sanitation). For pools, the rule in commercial/public pools is to replace 7 gallons of water per bather. Again, the assumption is some amount of swim time, probably 30 minutes. These guidelines are far better than trying to use TDS as a guide. Of course, having supplemental oxidation, such as UV or ozone, also plays a role.

The chemist's statement that "Monovalent ions may be harmless to pools while divalent ions are troublesome." is a bunch of bunk. It completely depends on the specific ions involved. For example, magnesium is far more innocuous than calcium since scale in pools forms from calcium carbonate long before it forms from magnesium. Magnesium is more of a problem with splash-out due to higher salt recrystallization pressure from magnesium chloride compared to sodium chloride. Low pH, which is monovalent hydrogen ion, can accelerate metal corrosion and can accelerate the dissolving of pool plaster. It's true that sodium is pretty innocuous, though it too contributes to conductivity of the water which at higher levels increases the rate of galvanic (metal) corrosion. Chloride is monovalent, but at higher levels not only contributes to conductivity metal corrosion, but also specifically attacks the passivity layer of stainless steel. I could go on, but I think you get the point.

As far as not knowing the ionic species in the water unless you are a chemist, again this is bunk. You have a test kit that can distinguish between some of the more important factors. You explicitly test for hydrogen ion level in the pH test. You explicitly test for calcium in the Calcium Hardness (CH) test. You explicitly test for Cyanuric Acid (CYA) in its own test. You essentially test for bicarbonate ion when you do the Total Alkalinity (TA) test and adjust it for CYA (and borates, if applicable). You explicitly test for borates (if applicable). You can also explicitly test for chloride using a salt test (though it reports as ppm sodium chloride). Unless you use dry acid or non-chlorine shock, you can assume that the sulfate level has not increased from what it was in the fill water. You do need to roughly estimate the magnesium ion level which can be based on the fill water or roughly estimated based on a portion of Total Hardness. You can pretty much assume that charge balance has sodium ion filling out the rest of what is in the water (assuming no chlorine shock is used -- that introduces potassium ion). Besides, distinguishing between sodium and magnesium isn't important anyway. I could go on, but I think you get the point.

As far as "invasive, chlorine impeding species in the water that conduct electricity", again this is bunk. It's not that there aren't chemicals that can reduce chlorine effectiveness. We already know about CYA, for example, but we measure for that and account for it. The simple ions do not interfere with chlorine effectiveness. Higher ionic strength does not interfere with chlorine effectiveness (i.e. SWG pools aren't less effective at the same FC/CYA levels than non-SWG pools). Active chlorine, hypochlorous acid, is a neutral molecule (it looks similar to water). If there is bather waste, then that can exert a chlorine demand reducing active chlorine levels, but we already account for that in our testing. For example, any ammonia introduced by bathers will result quickly in monochloramine and that shows up as Combined Chlorine (CC) rather than Free Chlorine (FC). Urea will slowly combine with chlorine to form varioius chlorourea compounds and these mostly show up as CC and most certainly do not show up as FC. Also, many of these organic compounds are uncharged and would not show up in a conductivity test. Also, it doesn't take very much of these compounds so using a conductivity (or TDS) test to try and detect them is pointless.

The statement that "50% of the TDS in the water is invasive to normal chemistry" is also bunk. Most of the TDS in the water, except soon after a complete refill, is sodium chloride salt, period. Even after a refill and adjustment, most of TDS is still salt, but contains calcium (and magnesium) and bicarbonate in addition to sodium and chloride. The amount of other chemicals, such as unoxidized organics, is much much smaller except in high bather load commercial/public pools and in such pools significant water dilution should be done. Even so, the largest components of sweat and urine get fully oxidized -- specifically ammonia and urea (though it takes longer for urea) and such oxidation results in nitrogen gas and carbon dioxide that eventually leave the system (i.e. outgas) and in water that obviously stays. So though I cannot prove it, I very much doubt that half of the TDS is unoxidized organics. The rate of increase in TDS, even in commercial/public pools, tracks almost perfectly with the predicted amount based on chlorine addition/consumption/usage (which results in sodium chloride salt), on other chemical additions (ignoring bicarbonate additions since that just outgasses as carbon dioxide eventually), and on water dilution.

The only compounds to really worry about would be those that act like CYA in binding with chlorine, but somewhat weakly such that chlorine gets released during the time of the FC test (i.e. as active chlorine, hypochlorous acid, gets used up in the test). There are such compounds (glycoluril, for example) and this probably explains the sometimes strange scenarios seen in some commercial/public pools where there is ample FC, even with no CYA, yet the pool is bacteriologically unsafe. This isn't usually seen in residential pools because the low bather load and relatively higher dilution rates (per bather load) as a result mean that such compounds don't build up.

Richard

Great stuff, Richard, and much appreciated (I also like how you placed "urea" instead of "urine" in the explanation! Doesn't sound quite so disgusting!).

I love facts, and hate opinions!