Edit 2/7/2025: One week since a half-inch of rain washed away the remaining salt on the roads … and no sign of salt in the water yet. TDS (total dissolved solids) readings for properly aged (i.e., room-temperature) water samples are steady at 210 ppm, plus or minus some single digits.
Recall that, as of my last post, my road-salt-in-drinking-water experiment was floundering. My tap water was showing far more variation in measured total dissolved solids (TDS) than seemed reasonable.
Turns out, that’s because a) my tap water is cold, b) temperature strongly affects the conductivity of water, c) this $6 meter measures and adjusts for temperature,
d) extremely slowly. And e) I’m not exactly a patient person.
I didn’t wait anywhere near long enough for the meter to adjust to my tap water temperature. And going forward, I’m not going to stand around for a quarter-hour holding this meter in a glass of water, waiting for the temperature adjustment to reach equilibrium.
The solution is simple. I have to let the glass of tap water sit for a couple of hours, and come up to room temperature. Then measure TDS. Once I do that, these “well-aged” water samples all provide consistent readings for parts-per-million total dissolved solids.
Properly measured, my tap water TDS has been around 210-215 ppm TDS for the past three days. A little higher than the 170 ppm I expected based on “10 grains of hardness” of the water. But definitely in the ballpark. And seemingly stable.
Hey, maybe I’m not crazy. It does, in fact, take about a week for water to pass through the Fairfax County drinking water system.
The presence of a stable, measurable baseline is important for this experiment.
And yet, as I go day after day without an increase in TDS, I begin to wonder whether I just imagined the salty-tasting tap water of winters past.
I expect road salt runoff to produce a big upswing in my tap water TDS, Wednesday-ish of this week, best guess. That’s based on last Friday’s half-inch of rain washing (almost) all the remaining salt off the roads. And my vague memory that the salt taste showed up on-order-of a week after road salting.
FWIW, I finally found confirmation that it takes about a week for water to move through my local water distribution network. When Fairfax flushes the water mains, they change disinfectant chemicals. Depending on where you are in the system, those chemicals may take up to a week to show up in your tap, and a week to go away (reference).
Depending on your usage patterns and location within the distribution system, it could take up to a week for your drinking water to transition from combined to free chlorine at the beginning of the flushing program, or from free chlorine to combined chlorine at the conclusion of the flushing program.
The upshot is that a) we may still be a few days away from salt showing up in my tap water, and b) while it has taken me a while to figure out how to use my $6 TDS meter, there’s no harm done.
So far, properly measured, the TDS in my tap water has remained steady at around 210-215 ppm. If a flush of road salt passed through the system, that ought to stand out pretty sharply against that steady background rate.
The full story
- This meter measures water’s electrical conductivity.
- That conductivity is increased by ions in the water.
- Such ions are generated when minerals and salts dissolve in water.
- Thus, the meter can infer the amount of ions in the water, from the water’s conductivity.
- It then translates that into something the user can understand, such as parts-per-million total dissolved solids (TDS) or salinity. Depending on the end-use market that is being targeted.
Source: Mettler Toledo white paper, “Reducing Measurement Error in Conductivity Readings”. Annotations in red are mine.
- But water temperature strongly affects conductivity. A 9F decrease in water temperature creates a more-than-10% reduction in water conductivity.
- Hence, this measurement typically requires temperature correction. The goal is to measure the water’s conductivity, adjusted to some standard water temperature.
- And this $6 TDS meter includes that temperature correction via a built-in thermometer (and presumably a look-up table on a chip, or something).
- But the meter is excruciatingly slow about doing that.
I finally got the bright idea of sticking this meter in a glass of ice water and see how long it took to display a temperature of 0 C.
I gave up, it took so long. I got tired of holding the meter in the ice water. I’m guessing it would eventually get there, but it would take five or ten minutes to do so.
In any case, that adjustment is so slow that what I interpreted as the meter reading “setting down” to a final value, in just a few seconds, was nothing of the sort.
And that’s what tripped me up. With incomplete temperature adjustment, cold water registers as “cleaner” water (lower TDS), owing to the lower conductivity of cold water.
Conclusion: Never rule out operator error
On the on hand, I could blame the meter for being so slow to adjust to different temperatures.
On the other hand, it’s up to the meter operator to use it correctly. Or spend the big bucks on one that works faster.
In any case, for $6, I got a very smart meter. Smart enough to do the temperature correction for me.
But the hardware? That’s still the best that $6 can buy. It’s fine, as far as I can tell, but there’s no expectation that $6 bought me some kind of heirloom-quality super-tool.
And, as it turns out, what I got for $6 is a meter that works, but takes forever to settle to a final reading, owing to the glacial pace of adjustment of its internal temperature sensor.
Which I consider fair, for $6. That it works at all is kind of a miracle. That was unkind. What I should have said is “more than fair”.
Now that I know that the temperature correction takes forever to register, all I need to do is let my tap water samples warm up to room temperature.
And poof, what seemed like a ridiculously inconsistent meter turns out to be … pretty consistent.
Well worth the $6.
I probably need to buy some distilled water, for another buck or two. Not to test the meter, but to rinse it after I’m done. By device design and by common acclaim, I get the impression that I’m never supposed to let anything touch the electrodes but water. Which precludes wiping the electrodes dry, in any fashion. But, I think that if I just let the little electrodes air-dry, after tap water, I risk “poisoning” the electrode surfaces over time with calcium carbonate deposits, a.k.a., water spots. This, by analogy to premature dulling of un-dried razor blades by the thickness of water spots (Post #1699).
Distilled water, by contrast, leaves nothing behind when it evaporates. So you don’t dry them, you rinse them with pure water and then allow them to air-dry.
Otherwise, the experiment is now on track. I have documented a stable baseline of around 215-225 ppm dissolved solids in my (room-temperature) tap water.
I just need to give it a few days for the road salt to work its way from the Potomac River to my water faucet.
