Ask John, Part 2 Why a Correct Reading Doesn’t Always Mean a Correct Answer

Ever so often, I like to just open the mailbag and answer what's in there. No theme picked in advance, no lesson plan, just real questions from real people about their real pools. And that's most of what this job looks like day to day. Not some grand chemistry lecture, somebody standing at their own equipment pad, looking at a test strip, wondering why the number in front of them doesn't match what they were told it should say. Answering that one pool at a time. Today, eight different people sent me eight different versions of that exact moment. Different pools, different setups, same basic question underneath all of them. What's true here, and what do I do about it?

That's what today is, so let's get into it. I've been getting quite some questions about the music I used for the podcast. And I have a bit of a history with it, so you gotta start hearing some new music on the show, just to mix it up a bit. I wanted to say a word about where it comes from before I just throw it out there. I wrote and produced these tracks myself back when I was a working nightclub and rave DJ in the techno and trance scene here in the United States, mostly through the 90s and the early 2000s, on the West Coast in Seattle.

The one you're familiar with is called Signal, and I composed it back in about 1995. That's a part of my life that doesn't come up here much, but the music's mine, made a long time before any of this podcast existed, and using it felt better than pulling something out of a stock library. I've had requests for it, so I'm also putting together a channel on SoundCloud with the full versions for anyone who wants more than what you hear at the beginning of the episode. I'll share that link once it's up. And it's not all old material either. I still dabble in music production between pool visits, and a lot more of it during the off-season.

So some of what you'll hear going forward is actually new music. Okay. Now that we got that out of the way, let's go back to the swimming pools, and our first question today is from Mike, and it's a good one to start on. He has a plaster pool with a salt system with an infinity edge, and for two years running he's been getting these sharp little calcium crystals on the plaster walls. One time it was bad enough that his kids actually cut themselves on them. He had it acid washed and buffed, and then it went quiet for two years. Now, it's back, and his numbers aren't really that bad at all.

pH is at 8.0, alkalinity is at 75, calcium hardness is at 280, CYA at 20, and salt around 3200. His water temperature is about 82 degrees Fahrenheit or about 28 degrees Celsius. And the app he's using told him that his LSI was a positive 0.04 right when he tested it. Basically dead on balanced. So he asked me, if the number is that close to perfect, why is this scale still showing up on my walls?

The first thing worth clearing up is that 280 parts per million calcium isn't low for a plaster pool. It's comfortably mid-range. So this was never a calcium leaching out of the plaster problem in the first place. What he's looking at is scale forming on top of it, which is a completely different problem with a completely different cause. LSI isn't a single snapshot the way most people treat it. It's a kinetic thing, more like a running total than a light switch he's just flipped on or off. Mike tests roughly once a week, and in the days between tests, his infinity edge and his salt cell are both pushing pH up at the same time.

The edge is off-gassing carbon dioxide on one side, and the cell is doing its own local pH thing on the other. And by his own account, LSI on that pool has climbed as high as 0.4 during that stretch between corrections, which isn't really a rounding error. It's a real sustained oversaturation sitting there depositing calcium carbonate the entire time. And then, by the time he tests it again, it's already settled back down near balanced, like nothing ever even happened. Which is really the whole lesson here. A snapshot doesn't undo an excursion that already ran its course. It also explains those two quiet years because that buffing and acid wash left a genuinely smooth surface, and smooth surfaces just don't give new calcium carbonate crystals anywhere easy to grab onto.

No nucleation site, no foothold. So this year's crystals aren't a new problem. They're the same old problem, finding its opening again, the first real foothold since the surface got smoothed out. So the fix isn't calcium. And it isn't even chasing a tighter target LSI. It's correction cadence. Test more often, correct sooner, and if this keeps happening, a continuous acid feed or a real pH controller solves it permanently instead of just chasing it by hand every week.

Andrew's question is a good follow-up on that one. And I don't think that's an accident. He runs a salt pool with his pH sitting around 8 to 8.2, and his own calculator tells him that his LSI is almost perfect. So he asks me straight up, is there any real point bringing that pH down, or am I just chasing a number for its own sake? LSI and disinfection are two completely different questions, and a good score on one tells you literally nothing about the other. LSI is entirely about calcium carbonate saturation, whether your water wants to deposit scale or dissolve it, and it has nothing to do with how much chlorine is actually working.

That second question is controlled almost entirely by pH, because pH decides the split between hypochlorous acid, which is the active sanitizing form of chlorine, and hypochlorite ion, which mostly just sits there. At a pH of 7.5, you're looking at roughly half your free chlorine is in the active form. Climb that pH to 8.0, and that drops to somewhere around a fifth. Push it to 8.2, and it's lower still. So Andrew can have a genuinely well-balanced pool from a scale and surface protection standpoint and still be sanitizing at a fraction of what the same chlorine reading could be doing for him at the correct pH. That's two completely different report cards, and a great grade on one doesn't mean anything about the other.

And the pH creep itself isn't a symptom of anything broken. Salt cells push pH up naturally as a straightforward byproduct of the electrolysis reaction, hydrogen gas and hydroxide both coming off the plates inside the cell, and every salt pool deals with this to some degree. It's expected, not a fault.

The fix doesn't touch his calcium or his alkalinity at all, just bring pH back into the 7.4 to 7.6 range, and if scale genuinely isn't a concern for him, he can even run a hair tighter at 7.3 to 7.5. A balanced LSI answers exactly one question. It was never claiming to answer the other one.

Brad's question is the deepest one in this episode, and it actually exposed something in how this show has been doing the math, so I want to just own that plainly before I get into his numbers. He has a vinyl liner salt pool, and he plugged in what looked like completely reasonable center of range values, pH of 7.5, water around 85 degrees Fahrenheit or about 29 Celsius, alkaline at E70, calcium at 200, CYA at 25, TDS around 4000, which is completely normal for a salt pool, and his calculator spit back a negative 0.53, nowhere close to balanced. So his question was simple. Which of these numbers does he actually need to move?

There are two real corrections, most simplified LSI explanations skip entirely, and Brad's numbers walked right into both of them. Cyanuric acid quietly eats into your effective alkalinity, since only carbonate alkalinity actually drives calcium carbonate saturation, and CYA ties up something like a third of its own PPM value as a separate non-participating form, cyanurate alkalinity. At CYA of 25, that's roughly 8 parts per million getting pulled out of a 70 part per million alkalinity reading before it ever does any real LSI work. Small on its own, but it adds up. The bigger driver of Brad's gap is the TDS constant. The standard number most LSI formulas use for total dissolved solids assumes water sitting around a thousand parts per million, which is a non-salt pool assumption.

And a real salt pool's TDS from the salt alone often runs 3000 to 4000 or even higher. The relationship between TDS and LSI doesn't stay flat as TDS climbs either. It steepens. So plug a salt pool's real TDS into a formula built around a 1,000 part per million standard, and you'll get a number that looks a lot worse than the water actually is. So which numbers should Brad actually move? Calcium hardness is the first move, running it toward the top of vinyl's real range of 150 to 250. That's legitimate headroom and not gaming anything. Alkalinity is the first move in the standard adjustment order, so push that toward the top of the salt range, about 80 parts per million, before you touch pH at all.

And pH itself, if it's drifted outside 7.4 to 7.6 while someone tries to compensate for a scary looking number, needs to come back. Because pushing pH to cover for something else is overriding one number instead of correctly using the headroom you actually have in another. A modestly negative LSI matters a lot less on a vinyl liner than it does on plaster, since vinyl isn't cementitious and it doesn't etch. The real risk from aggressive water on a vinyl pool shows up in the metal equipment and fittings, not the liner itself.

Lauren's question takes the same idea Brad just walked us through and points it somewhere completely different, at a dosing calculator instead of a balance number. Lauren is a customer of mine here in Kansas City, a good one. She runs a salt pool, 33,000 gallons, or 125,000 liters, and she uses the Orenda calculator to figure out how much myriatic acid to add when her pH drifts. Last time, sitting at a pH of 7.8 and targeting 7.6, the calculator told her just under 15 ounces or 440 milliliters. And she's learned through repetition that number consistently underdelivers. So she compensates by adding more acid than the calculator says. Except the one time she did that most recently, she overshot, landing at 7.2 instead of 7.6.

And her question, fairly, is why a calculator can be wrong in both directions depending on when she uses it. It's not a bad calculator. It's a static tool being asked to solve a moving problem. A salt cell pushes pH upward the entire time it's cycling as a straightforward byproduct of electrolysis. We've been through this. Hydrogen gas and hydroxide are both coming off the plates of the cell continuously, and any acid dosing calculator, orendas included, is solving a simple before and after problem. Get the water from this number to that number, assuming nothing else touches it in between. A salt pool isn't holding still during that window.

If the cell keeps running while the acid settles in, it's working against the correction in real time, regenerating some of the exact alkaline pH the acid just knocked down. That's why the prescribed dose consistently comes up short for her, and why adding more than the number on the screen was the correct instinct. The overshoot's the same mechanism working in reverse. Once you've calibrated yourself to routinely needing more than the calculator says, and you build that extra amount into your habit, the one time the cell happens to be producing less during that window, there's nothing left pushing back against the larger dose. So the correction that usually just barely gets you there overshoots straight through the target.

And there's a second thing happening here too, and it's physical, not chemical. Pouring a concentrated dose of acid into one spot creates a strongly acidic pocket right at the point of contact. That pocket off-gasses carbon dioxide rapidly before it ever fully mixes into the rest of the pool's buffering capacity. So some of the acid's punch gets spent locally in a reaction that partially reverses once it dilutes into the bulk water, instead of doing the full even work against total alkalinity the calculator assumed it would do. Diluting the dose in a bucket first, then adding it slowly along a return line with the pump circulating gets real results a lot closer to what the math predicted.

Treat the calculator's number as a starting point, not a finish line. Dose it and then let it fully circulate. Retest before adding anything else instead of just trusting the first number to land exactly where it says it will. On assault system specifically, time your acid corrections for when the cell is off or running low, rather than mid-cycle, so you're not fighting live production while you're trying to fix the same thing it's undoing. Even a perfect formula can't see whether your cell's running. That part's still on you, not the math. And in a case where you overshoot to 7.2 when you're aiming for 7.4, so what? No harm done, and Henry's law is gonna fix your mistake in about two hours anyway.

Before we get to the next question, a quick word on where the rest of this stuff lives. Every month I put together something called the Standard. It's a monthly newsletter and it goes out to podcast listeners and the folks in our actual field lab in the same send. No separate versions, no watered down list. Five sections every issue. The first issue went out last week. I hope you've had a chance to read it. Let me tell you what you're going to see in there. I'll include a short personal note on whatever's actually on my mind that month in the From the Chemist section. The science section is a real chemistry deep dive, the kind of thing we do here on the show just in writing.

The technology section covers new equipment, testing gear, anything worth knowing about before you buy it. Cool side is lighter seasonal stuff, the kind of thing you'd actually want to read in August versus January. And they're usually a story or a tidbit from working with someone's real pool. Some of them are going to be pretty interesting. And the on the show section recaps whatever we've covered on the podcast in that last month. So if you miss an episode, you're not completely lost. It goes out once a month, last week of the month, and that's it. No daily emails, no funnel, nothing to unsubscribe from in frustration three weeks later.

If you want it, there's a sign up right on the podcast page at poolsscientific.com, and it's free. And since we're on a break from the questions anyway, let me tell you about something else we're doing because a couple of these questions today came directly out of it. The chemist circle is the paid side of what we do here. Two personalized chemistry submissions a month, and I mean personalized. You send me your actual numbers, your actual pool, and get a real response written for that water, your pool, not a generic chart or a percentage score from a machine that's never seen your equipment. Most of the questions in this episode came from members submitting exactly that kind of data, which is really the whole point.

A podcast episode can teach you the principle, the circle applies it to your specific pool. Most of the damage I see to pools every year doesn't happen in July. It happens in the off season when the pool's closed and nobody's testing anything. That slow drift in pH or alkalinity has months, not days, to turn into etching, scale, or a stain nobody notices until spring when the cover opens again. By the time it's visible, it's already been happening for quite a while. The circle doesn't stop just because the swim season does. If anything, the off-season is exactly when a set of eyes on your numbers matter the most, because it's the one stretch of the year nobody's watching the water at all.

We still have some slots available for the founding rate of$49 a month with a discount for a full year. And once you're in at that rate, it's locked permanently for as long as you stay a member. It's not a promotional rate that creeps up in six months. It's capped and the first hundred members, period, get it. And once that's full, it moves to standard pricing and it won't come back down. This isn't insurance, and I'm not going to pretend that it is. Think of it more like a regular doctor's visit instead of an emergency room. You're not paying for a crisis to happen, you're paying so the crisis doesn't happen in the first place, or gets caught early enough that it's an$8 fix instead of a$10,000 one.

It's honestly the cheapest way to actually protect that five or six figure investment in your backyard that I know of. And if that sounds like something you or your pool could use, especially heading into the months when nobody's looking, it's at poolsscientific.com/slash chemist circle. Alright, back to the questions. Anthony's question moves us somewhere completely different, off of water balance and into something more like a jurisdiction dispute.

He runs a Jandy Aquapure salt cell, oversized for his 10,000 gallon pool, or about 38,000 liters. The owner's manual tells him to run CYA between 50 and 75 parts per million, never letting it drop below 50. This shows own ceiling is 50. So which one's right? Is the manufacturer's number real chemistry? And does running CYA higher actually extend the cell's life the way people on Reddit keep claiming? Neither one of these numbers is wrong. They were never trying to answer the same question in the first place. The manufacturer's range exists for one reason, and that's to protect the hardware. It has to guarantee a cell can reliably produce enough chlorine across every installation it might ever get bolted onto, undersized pools, heavy sun, heavy bather load, all of it.

That's a duty cycle guarantee, not chemistry. The 7.5% rule exists to protect disinfection. How much of your chlorine is actually active and biologically effective at a given CYA level, that's a water chemistry and public health question. Two completely different disciplines, two completely correct answers, and it's basically a coincidence whenever they happen to land in the same neighborhood. Here, they don't. The manufacturer's floor is exactly this show's ceiling. Now, let's look at the Reddit claim that higher CYA extends cell life. That's a real mechanism underneath that. It's just an incomplete picture. It only holds if you keep your free chlorine target fixed regardless of CYA, instead of scaling it with the 7.5% ratio, the way this show teaches.

Scale free chlorine correctly as CYA climbs, and you cancel out most of the supposed savings, because now the cell has to produce more total chlorine to hit that proportionally higher target anyway. Anthony's already keeping his CYA around 35, holding free chlorine steady at 3 to 4 and running an oversized cell. His own instinct that lower cell output, correct chemistry, and real maintenance are what actually extend a cell's life, not a bigger CYA number. That's exactly right. So in this case, you have two rule books doing two completely different jobs. One protects the machine, one protects the water. The manufacturer is just using both versions of what CYA actually stands for.

A chemist circle member, Mitch, asked a question that has almost the same shape as Anthony's, just in a different domain. Somebody's marketing material says one thing and the mechanism says something else completely. He runs a deliberately zero CYA pool, full sun, and is using unstabilized liquid chlorine. That setup's worth pausing on for a second because it means neither of the two jobs CYA actually does is happening in his water at all. CYA protects chlorine two separate ways, not one. It forms a loose, reversible bond with hypochlorous acid, tucking a reserve of it away where sunlight can't reach it, and the free CYA molecule itself also absorbs a slice of that same UV band directly before it ever reaches whatever chlorine isn't bound up.

At true zero, Mitch gets none of that. Every drop of chlorine he adds is standing there completely exposed, which is exactly why he's dosing constantly instead of occasionally. And it's worth keeping in the back of your mind because it matters again in a few minutes. His actual question is that he's getting small, inconsistent, non-zero CYA readings across five different test methods, including an automated water guru monitor. And he wants to know why a number that shouldn't be moving at all keeps bouncing around. Okay, there's two things going on here. Every one of these tests, strips, photometric readers, automated monitors, work by measuring turbidity from a reaction between CYA and melamine, and that works well in the middle of the range.

But down near true zero, every method hits a real lower limit. Noise that stays roughly constant in absolute terms suddenly becomes a huge fraction of a very small true value. And because these tests are optical tests, anything else that scatters light the same way, whether it's fine particulate, trace turbidity, even differences in fill water, it can read as phantom CYA that isn't actually there. Also worth knowing, Water Guru publishes real validation data, tying its free coin and pH readings to the Taylor K1515A reference kit. And there's no equivalent published validation for CYA specifically, so that number carries less independently verified backing than the other two on the same device.

The second thing is a straight-up correction. Water gurus own support content claims CYA drops slowly from being oxidized by chlorine over time. And from a chemistry standpoint, that does not hold up. There is a real cyanuric acid hydrolase enzyme pathway that can break CYA down, but every real-world version of it has to solve one really, really hard problem first. The bacteria that produce that enzyme can't survive in chlorinated water. Commercial cell-free enzyme products do exist specifically to get around that, and products that rely on live bacteria to reduce CYA explicitly require the pool to be fully declorinated before they work at all. One published study found a cell-free enzyme's degradation rate cut in half at 4.5 parts per million chlorine, tested directly against real municipal pool water, and none of that runs in a normally sanitized pool.

Plain oxidation of CYA by chlorine directly isn't supported either. CYA's resistance to chlorine is the entire reason it works as a stabilizer in the first place. So for Mitch, his true CYA isn't moving. Five separate devices are all independently hitting their own noise floor, right at the exact value he's deliberately targeting, and single-digit bouncing at true zero isn't worth chasing. If one device ever consistently reads 20 or higher while the others sit near zero, that would be worth a second look. This isn't. Sometimes the number isn't lying to you. The measurement just wasn't built to see that low in the first place.

Jerry's question came in over on the Trouble Free Pool Forum, and it's one of the better, almost counterintuitive questions I've come across in a while. His setup is at CYA and free chlorine equilibrium already, and his question was this.

Ultraviolet breakdown of chlorine only happens right at the surface, and it's not because water blocks UV light the way a pair of sunglasses does. Plain, clear water is pretty transparent to the specific wavelength doing this damage, which sits right at around 330 nanometers. That's the sweet spot researchers identified decades ago as the most effective part of sunlight for breaking down both forms of active chlorine. If your pool were nothing but pure water, that light would travel down several feet before it lost much strength at all. But no real pool is just pure water. It's water carrying dissolved organic matter, a haze of body oils and sunscreen, occasional algae, chloramines, and even your own cyanuric acid, which absorbs a slice of that same UV band itself.

And every one of those things scatters or soaks up incoming light before it can travel very deep, and the effect compounds fast. Think of it less like a hard cutoff and more like the light losing strength on a curve, a lot of it consumed in the first few inches, and whatever survives that first pass keeps thinning out the deeper it goes. People have tested this directly, comparing chlorine loss in shallow containers against deeper containers holding the same water and at the same CYA level. The deeper water consistently held onto more chlorine, so depth itself is doing real protective work on top of whatever your CYA is doing.

The reactive zone in your own backyard pool on a real sunny day is confined to something like the top few inches, not the whole four or five feet that you're standing in. Clearer water pushes that zone slightly deeper, and hazier water pulls it tighter to the surface, but it's always shallow relative to the pool as a whole. In still water, or in a pool where the pump is off, that thin surface layer depletes and then sits there waiting on slow molecular diffusion to bring fresh chlorine up from below. The loss rate tapers off once that top layer draws down because nothing's refilling it quickly. In a circulating pool, the pump does the exact opposite job the whole time.

It keeps delivering fresh, full strength water straight to that exact thin zone, then moves the depleted water out of the way so the next batch can take its turn. Circulation isn't protecting your chlorine from the sun. It's feeding the reaction. So a circulating pool loses more total chlorine to sunlight over time than the same pool sitting stagnant. CYA still buffers how much free chlorine is exposed at any given moment, doing the same double duty from a few minutes ago, the reserve and the direct absorption both, but none of that changes the underlying delivery mechanism. None of this is an argument for cutting your pump time to save chlorine.

That trade is a bad one on every other front. Filtration, turnover, safety, all of it. And the chlorine you'd save is so small next to what you'd lose everywhere else. Run the pump around the clock the way this show has always said, and just budget slightly more chlorine into a sunny pool than you would a shaded one. Our last question today is from Eric, also from over on Trouble Free Pool Forums, and it's the kind of question that's probably crossed every pool owner's mind at some point without them ever asking it out loud. Is it actually dangerous to pour a gallon of liquid chlorine straight into a skimmer while the pump's running?

Does that concentrated slug of chlorine cause a real problem for the pump and for the plumbing? For the scenario as he described it, this is mostly a perceived problem, not a real one. Every wetted component in that pump and filter is already rated for chlorine by necessity. It sits in chlorinated water continuously every single day, and a 30-second transient doesn't cross any threshold ordinary daily operation isn't already crossing. The total mass of chlorine that moves through a typical pump on any normal sanitizing day already exceeds what one gallon of liquid chlorine could ever deliver, even at full unmixed strength. On top of that, liquid chlorine is denser than water, so it doesn't even travel as a solid slug.

It mixes into the moving column of water as it gets pulled through the skimmer throat. The real risk lives somewhere else entirely, and it has nothing to do with the pump. And a lot of it is the basis for the old adage in the pool industry that you should never, absolutely never, pour chemicals into your skimmer. If there's a solid chlorinating tablet, trichlor or dichlor, already sitting in that same skimmer basket and concentrated liquid chlorine hits it directly, that's a genuine chemical incompatibility. It can and likely will trigger a fast localized reaction with real heat and off-gassing of chlorine right at the skimmer that your face is only a couple feet away from at the time.

That's almost certainly the real hazard underneath this old caution. It just usually gets told as a vague warning about pumps and pipes instead of what it actually is, a warning about what's sitting in the basket at the moment that you pour. And it's even more dangerous if you have a trichlor tablet feeder in line with your plumbing return line or the offline standalone type. That same slug of liquid chlorine hitting it is a different situation entirely, because the feeder's internal flow path is a lot narrower than your main plumbing line. That concentration and confinement is exactly what turns a fast, localized reaction into a real explosion risk, and it's documented, not theoretical.

People have seen extensive damage at the equipment pad with destroyed plumbing and equipment, homes with substantial damage to the wall near the chlorinator, and people have been seriously hurt this way and even killed. The chlorinator becomes a pressurized grenade, and when it goes off, it can destroy everything within 20 feet, like your face. So the rule extends past the skimmer. Never run liquid chlorine and a trichlor or dichlor feeder on the same system without keeping them fully separate, and never dose liquid chlorine anywhere near an active feeder. Keep your skimmer basket clean and single purpose for bugs and leaves. If you're hand dosing liquid chlorine, just don't put it in the skimmer.

For the sake of your face, man. So there you go. That's eight real questions from eight real people, and I genuinely appreciate every single one. If you have your own question, the Ask John page at PoolsScientific.com is always open. Next week, we're going to start looking at the actual chemistry standard sitting behind all of this, and whether it holds up the way most people assume that it does.

Rest in peace, Dolly. The Pools Scientific Podcast is an independent production by John Cooper and Pools Scientific LLC. Find us at poolsscientific.com. If this show has been useful to you, share it with someone who takes their pool seriously. New episodes are every Tuesday.