CYA Testing Update

[mas985]

I thought I would give those who are interested an update on some testing that I have been doing regarding CYA levels and chlorine.

First a summary of the testing that I did in this thread.

- Running my spa with the SWG at CYA levels from 45-80 ppm showed no change in chlorine production.
- Extinction rates seemed to improve better than what theory predicted for higher CYA levels.
- Unfortunately, I could not get consistent results from some of the extinction tests that I did.

So after the testing I did early in the summer, I lost exclusive use of the pool and spa since the kids were off from school .

About 3 weeks ago, I decided to raise the CYA level in the pool since at least some of the testing showed an improvement with higher CYA levels. At the time I thought my CYA level was around 45 ppm since that was what the last test showed me. I added 5 lbs of CYA which should have brought the level up to about 75 ppm but when I tested it again after 2 weeks, it showed only 50 ppm. Even after 3 weeks, it was still 50 ppm so I added another 5 lbs which brought the CYA up to 80 ppm. During the summer, I lost about 25 ppm of CYA somewhere. Still a mystery.

Anyway, after I raised the CYA in the pool to 80 ppm, I noticed that the chlorine level changed from 2.5 ppm to over 6 ppm. I have since dropped the SWG setting from 60% to 35% and the Chlorine level still increased to 8 ppm. Some of the increase was due to using the solar cover more often so I really couldn’t conclude anything from that change.

So I thought about an easy test I could run that would show how well different levels of CYA would protect chlorine without extraneous factors getting involved with the test. Since I am now at 80 ppm of CYA, I thought about dilution. Dilution would accomplish several things. It would reduce the CYA and chlorine levels by the dilution ratio while still keeping the chlorine to CYA ratio constant. According the Chemgeek formulas, this should keep a fairly constant level of HOCL independent of dilution ratio. It would also expose the samples to same sun exposure on the same day. So I used three 5 gallon buckets side by side with tap water for dilution, yes distilled water would have been better but I thought I would try this first:

First bucket: No Dilution CYA = 80 ppm, CL = 8 ppm
Second Bucket: 1:2 Dilution, CYA = 60 ppm, CL = 6 ppm
Third Bucket: 1:1 Dilution, CYA = 40 ppm, CL = 4 ppm

I let them sit in the sun for a full day and then tested the chlorine again.

First bucket: No Dilution CYA = 80 ppm, CL = 6.5 ppm a 1.5 ppm loss (19% extinction)
Second Bucket: 1:2 Dilution, CYA = 60 ppm, CL = 3.5 ppm a 2.5 ppm loss (42% extinction)
Third Bucket: 1:1 Dilution, CYA = 40 ppm, CL = 0.8 ppm a 3.2 ppm loss (80% extinction)

So even though the first bucket had the highest chlorine level, it still lost the smallest amount of absolute chlorine. If true, this is clearly a significant economic benefit to running higher CYA levels.

This means that you could run an 80 ppm CYA pool with half the chlorine cost as a 40 ppm CYA pool even though the residual is twice as high. However, the buckets are only 12" deep so the results for a 5' pool may be different.

I think the biggest revelation, unless I am missing something, is that this should be true for bleach run pools as well which kind of goes against current thinking.

I for one am pretty convinced that running at higher CYA is better but I am curious about levels exceeding 80 ppm. At some point there will be diminishing returns and possibly adverse effects. It may be possible to curve fit the data, one of the reasons I chose three buckets, to see if there is some point where the increased gain is so small it just doesn't matter that much.

If I have time, I may try it again with smaller containers and use distilled water instead of tap water.

Can anyone identify any flaws in the test or conclusions?

Also, would anyone else be willing to do the same test to verify the results.

 

[JasonLion]

CYA levels over about 90 quickly become impractical due to testing considerations: it is difficult to measure CYA at 100 or higher, it is difficult to measure PH if FC is too high. So I see CYA of 90 as a practical upper limit.

 

[mas985]

Thats a good point, testing would become a problem. One other thing I just remembered reading about was that high CYA levels have been known to be very corrosive to plaster pools and can shorten the life of the plaster.

 

[chem geek]

The plaster study is shown here and seems to indicate that levels below 100 ppm would show minimal degradation though if you look at the curves it seems to me that it's still at a risky level. Probably 80 ppm would be better and as Jason said there are other reasons for testing CYA and pH (with high FC) that make this a practical limit. Also, if the pool were to ever develop algae, it would be very hard to combat at high CYA levels -- even 80 ppm takes a rather high FC level for shocking.

Your methodology was very clever and reasonable. Assuming a pH of 7.5, the disinfecting chlorine concentrations in the three buckets are all 0.043 so essentially identical and all equivalent to 0.09 ppm FC with no CYA. You should test your tap water for chlorine -- I suspect it probably has Combined Chlorine (CC), specifically monochloramine, at around 1 ppm but the key thing to find out is what it measures for Free Chlorine (FC) as that is what you were testing over time in your test. Even accounting for any extra FC added with dilution, that would only make the result even more striking.

What I find very interesting is how this effect happens even at a 12" deep bucket.

The loss of chlorine comes from UV breakdown of hypochlorous acid, hypochlorite ion, and some or all of the chlorinated cyanurate species (i.e. chlorine attached to CYA). In addition, all of these, as well as CYA, have a UV blocking effect, though the effect with hypochlorous acid and hypochlorite ion results in their destruction while the effect with CYA does not (i.e. CYA absorbs the UV and converts it to heat and possible retransmission at a higher wavelength). The industry graph implies that there is breakdown of the chlorinated cyanurates, but at a slower rate -- possibly depending on where the photon hits the molecule since the ring portion is more like CYA absorption while the chlorine-nitrogen bond may be more like hypochlorous acid but at a lower probability of breakdown.

There is a shielding from the hypochlorous acid and hypochlorite ion themselves, but I calculate that is negligible given their very low concentrations. I'll plug in your numbers in my spreadsheet model for the CYA absorption to see what I come up with. The strange thing about it, though, is that such a large effect is seen in the one foot bucket. If this were just CYA absorption of UV, then a pool should see an even more dramatic effect as the lower depths would be even more shielded. Instead, I think your numbers are more consistent with the reports from pools as a whole (Janet's pool, for example), but I'll check on that to see if that's so. If it is, then maybe there is something else going on.

I've got to go to bed now, but will try and run some numbers and models this weekend to see if I can figure something out. Very surprising and rather strong results, however. Very, very different than what the industry normally says happens in terms of "diminishing returns". In the commercial/public pools, they talk about how anything above around 10 ppm CYA doesn't seem to help reduce chlorine demand at all, but in that situation much of the chlorine demand is due to bather load (i.e. oxidation of organics and ammonia) rather than mostly from sunlight. So once there is some CYA to protect chlorine from sunlight at least somewhat, then it gets used up so quickly by oxidation of organics and ammonia that more CYA doesn't matter. In a residential pool, and certainly in a bucket of pool water, we don't see that effect so the pure sunlight protection dominates.

There is another possibility that needs to be eliminated or reduced: there may be chlorine demand from the tap water. That is, the tap water may contain organics that combine with chlorine or other components that the chlorine will oxidize. So a better test would have another set of similar buckets that are covered or in shade to not be exposed to the sun, but still contain the same proportions of dilution. That way, if there is chlorine loss in those covered buckets, then that amount of loss is due to something in the water, not to sunlight. Or you could rerun the same experiment at night, assuming the tap water doesn't change in character in the interim.

One more thing to check is the pH in each bucket. If the tap water changes the pH, then that will affect the chlorine life. It turns out that the half-life of hypochlorite ion (with no CYA shielding and at low concentrations or shallow depths) is about 20 minutes while that of hypochlorous acid is about 2 hours 10 minutes while a 50/50 mix at a pH of 7.5 is about 35 minutes. So lower pH is expected to not only have more disinfecting chlorine, but to have the chlorine last longer as well and this effect is amplified by about a factor of 2 when CYA is present due to CYA's buffering of hypochlorous acid (so hypochlorite ion concentration varies about twice as fast vs. pH compared to not having CYA).

You mentioned a "1:2 dilution" but if that was one part tap water with 2 parts pool water, then that would dilute by 2/3 * 80 = 53.33 whereas I think you wanted "1:3 dilution" which would give you 3/4 * 80 = 60. So which did you actually do?

Richard

 

[duraleigh]

Somehow, I overlooked this thread but am fascinated by the results.

Mark, I agree with Richard that your test method was genius in it's simplicity. I am appreciative of all the work ya'll do in this area. This is real, practical benefit to pool owners and the knowledge gained can be easily applied.

When I first came to PF about 6-7 years ago, there was a general consensus of 20-30ppm for most chlorine pools. From the knowledge ya'll are gaining, it looks like that recommendation may be moving up pretty significantly.

Thanks for your work.

 

[chem geek]

The effect of absorption is governed by the following formula:

I/I_o = e^-µl

where µ (mu) is the absorption coefficient and is a function of wavelength. "l" is the path length -- the depth of the water. So the intensity drops off exponentially, though the rate of drop-off is very dependent on the absorption coefficient. The molar extinction coefficient is related to the molar absorption coefficient as follows:

µ = ln(10) * c * ε

where "c" is the concentration in moles/liter.

One difference between the bucket of water and a pool is that the pool normally has circulation or mixing of the water. The chlorine is mostly depleted near the surface so with mixing of the water there is a more continual loss, but pools are rather deep so the bulk of the water is still protected. In a bucket of water that is still, the chlorine gets depleted near the surface but only gets replaced by diffusion from the lower depths until a steady-state gradient of concentration forms. The net effect should be less loss in still water than in circulating water (at the same depths).

Diffusion is determined by Fick's law as follows:

N = -A * D * ΔC/Δx

where N is the rate of diffusion in (e.g. moles per second), A is the area of the interface across which diffusion occurs, D is the diffusion coefficient, ΔC/Δx is the concentration gradient. A typical diffusion coefficient for most small molecules and ions in water would be around 1x10^-9 m²/s (I can't find a specific diffusion coefficient for hypochlorous acid in water, nor the effective radius of the hypochlorous acid molecule which would let me calculate the diffusion coefficient). The real calculation would be dynamic as sunlight breaks down chlorine at various depths and diffusion acts to equalize this until some steady-state is reached. That's more complicated than I plan to get into.

The reduction in chlorine concentration is solely a function of the chlorine concentration and light intensity. The standard 35 minute half-life comes from Florida sun in the summer at peak 1 PM (daylight savings time, so noon "real" time) at a pH of 7.5 and I've run the calculations in a spectral spreadsheet and come up with that number. So the rate of reduction in chlorine concentration is directly related to the intensity at each depth. The chlorine reduction follows the following first-order law:

C = C_o * e^-kt

where k is the rate constant that relates to half-life by

t_1/2 = -ln(0.5)/k = ln(2)/k

and "k" is proportional to intensity "I". So the "k" rate of chlorine loss due to sunlight varies by depth. So to find what happens overall in the bucket, I need to integrate over the depth but that's a mess even taking a single snapshot in time (assuming no diffusion nor mixing):

Integral(C dl) = Integral(C_o * e^{-kt * exp(-µl)} dl)

I don't know of a closed solution to the above "double" exponential (if anyone does, let me know). If I instead just calculate the rate differences vs. depth to just get an idea of the "instantaneous" average chlorine loss as a whole assuming perfect mixing (so more like the pool scenario than the bucket scenario), then that is the following:

Integral(I/I_o dl) from 0 to L divided by L = Integral(e^{-ln(10)*c*µ*l} dl) / L

which is just (e^{-ln(10)*c*µ*L}/(-ln(10)*c*µ) - 1/(-ln(10)*c*µ))/L
or (1/((ln(10)*c*µ*L))*(1 - e^{-ln(10)*c*µ*L})

I computed the above at various molar extinction values and varied the concentration. The net result is that the average reduction in intensity is at most reduced proportionately to the CYA concentration. So at 80 ppm there is roughly half the average intensity than at 40 ppm. Though it is true that at the lower depths the intensity gets cut back exponentially, the main loss of chlorine is at the shallower depths where the exponential looks more like a linear reduction. That is, the contribution of loss from the lower depths becomes negligible so the exponential reduction in intensity doesn't matter much when the water is well mixed so that the concentration is consistent throughout.

So at best I would predict a chlorine loss rate cut in half at 80 ppm than at 40 ppm in a body of water that was perfectly mixing to keep concentrations equal everywhere. The relative rates that you saw were:

6.5/8 = exp(^-k₁t)
k₁t = -ln( 6.5/8 ) = 0.21
0.8/4 = exp(^-k₂t)
k₂t = -ln( 0.8/4 ) = 1.6
so k₂/k₁ = 1.6/0.21 = 7.7

That's a huge difference. The only way I can get the kind of numbers you are seeing is if I assume no mixing (and for simplicity, no diffusion) and a rather high CYA molar extinction coefficient. In that case, I get faster depletion of chlorine near the surface, but the depths are protected with an exponential decline in UV intensity from the CYA shielding.

So circulation or mixing of the water may make a huge difference in what you would see and since pools generally circulate water during the day, this would make them have less strong effect compared to what you are seeing. Ironically, it would seem that turning off the pump during the day would reduce chlorine loss due to sunlight, but the price one pays for that is a reduction in chlorine concentration near the surface, thereby allowing algae and pathogens to grow there -- so obviously that's not a great solution.

If there is a way for you to repeat the experiment with some sort of mixing (swirling) of the water, then that would validate if what I said above was true. You don't want to mix too vigorously or else you'll aerate the water possibly making its pH rise. I don't know how you could easily make the water circulate or continuously mix well in the buckets.

Richard

 

[mas985]

I tested my tap water and I have 1 ppm of CC and no FC.

Also, I would have thought that heating by the sun would have caused at least some amount of convection in the water to help move it around.

One thing I did think of is that when I filled the pool last February, the water looked green until I put in a gallon of Chlorine. Could municipal water have algae? If so, that would explain some of the degradation of chlorine.

Would bottled water be free of all organics or would I have to use distilled water?

 

[chem geek]

We've smelled algae in our tap water on occasion so it certainly can happen. The monochloramine, which is what you measure as 1 ppm CC, is designed to fight bacteria, not to kill algae as it's not usually enough for that. So there may very well be algae and Dissolved Organic Compounds (DOC) in your tap water. Though distilled water is clearly best, filtered water also works well -- that's what I use when diluting (it's a activated charcoal and paper filter and removes chlorine, monochloramine, many organics, some metals, etc.). Bottled water varies a lot -- some is just repackaged municipal water! It may very well have a chlorine demand on its own. If the bottled water is from a spring, it'll probably have some minerals (e.g. calcium, magnesium and carbonate -- not a mineral, but present nevertheless) but probably not organics, so would probably be OK to use. Water from above-ground sources are the ones that tend to get the organics (from organic matter getting dropped into the water).

Richard

 

[mas985]

I think what I will try is to prep the tap water. A small amount of bleach in the tap water and left in the sun for a day should kill everything and most of the chlorine should be gone by the end of the day. I can then use that water for dilution of the pool water and there shouldn't be too much chlorine demand that way.

 

[ivyleager]

Okay, here's from someone who has run a high CYA pool (would also be nice to hear from alyad too). I did not need to add bleach as often in years past when my CYA was 80-100. I'd estimate that I'd add an extra lg jug of plain bleach 6% about every 36-48 hrs. No problems with algae, just a 'cloudy' pool 1x last year that cleared up overnight with superchlorinating. Moreover, I tried to keep FC at 5ppm, which is low based on Best Guess Guide. But, I'm one who likes to run the filter 14-24 hrs/day.

This year, my CYA at opening was 35. I also found that I had to add bleach every day to maintain residual. Pain in my rear. So, I began using pucks for about one month, including vacation. CYA back up to 60 now, but still adding bleach ALOT. I'm using pucks the remainer of the season, which is soon coming to an end.

My practical experience tells me that MY POOL CONDITIONS are such that I'm able to hold the FC longer when the CYA is 80-100.

Next year I will allow CYA to go up to 80ppm.

CaryB
24K IG 18x36 vinyl rectangular pool
sand filter

 

[chem geek]

Cary,

Thanks for the info. It is certainly consistent with what Mark is seeing.

Mark,

I've been thinking more about a way to test my hypothesis that a lack of circulation of the chlorine allows for a band of CYA near the surface to shield the chlorine below thereby explaining the higher than expected protection. If you take an eyedropper or turkey baster and slowly remove water from just below the surface to measure the FC and then do the same thing near the bottom and if the two are different take a sample from the middle depth as well, then that would tell us a lot. You should do this with the full strength (not diluted) 80 ppm pool water bucket as that should have the most stratification, if this is what is happening.

This seems easier than trying to somehow repeat the test with lots of continual mixing.

Thanks,
Richard

 

[mas985]

The sun is starting to get low in the sky these days and the last test I did, I noticed that not much sun light was getting to the bottom of the bucket. So what I can do is to do what you suggest but in my spa on the sun side. I can probably reach an arms length down and on the surface and perhaps mid way. I need to get some more titrate and wait for a sunny day.

 

[mas985]

Well, I tried the deep water testing in the spa, actually it was at an arms length, but I got the same result on the surface as I did 3 feet down. So that doesn't seem to show any extra protection with depth although I am sure that there was some convection mixing going on during the day.

I'm not sure there is any easy way to test this CYA theory. A lab type test may be the only way to remove external influences.

BTW, since increasing my CYA to 80 ppm, I am now running at 20% vs 60% SWG level and a CL level of 7 ppm. But again, the sun is not as strong as over the summer and I have a clear bubble cover on the pool. It looks like I will have to wait until next July to see if there is a big difference.

 

[Guest]

I think that results speak much more than any theoretical possiblilties. I have been saying for quite some time now that SWGs work more efficiently with higher CYA levels and have seen this time and again not only in my own pool but in my customer's pools. Your reduction in generator output from 60% to 20% while maintaining FC levels is not uncommon at all! Even if you did not have the solar cover on the pool you would still be maintaining a FC level of probably around 3-4 ppm, IMHO. The cooler temps are certainly a factor but even if the temp had not changed you would have seen a decrease in cell output needed or pump run time to maintain the same FC level with the, at least, doublng of your CYA level that you did.

Bottom line is this. If increasing the chlorine level allows a lower output percentage to maintain the same FC level it really is a moot point whether this is because of more efficienct chlorine production by the cell or simply because the chlorine is not destroyed as quickly by UV light. (I still suspect that both things are going on but really have no way to prove or disprove them. However, if you compare the amount of chlorine lost in a manually chlorinated pool with the same CYA levels and determine just how much chlorine is being produced by the cell at the different output percentages I don't beleive that it will indicate that the difference is solely because of the protective effects of CYA.)

 

[Guest]

I tested my tap water and I have 1 ppm of CC and no FC.

Also, I would have thought that heating by the sun would have caused at least some amount of convection in the water to help move it around.

One thing I did think of is that when I filled the pool last February, the water looked green until I put in a gallon of Chlorine. Could municipal water have algae? If so, that would explain some of the degradation of chlorine.

Would bottled water be free of all organics or would I have to use distilled water?

The green coloration you are seeing is more likely from the chloramines used to sanitize your municipal water. It is not uncommon for water with choramines to have a greenish hue. This effect is often noticed with portable spas when they are filled and the color disappears on shocking, which breaks down the chloramines.
Remember that algae is a plant that needs sunlight so unless your municipal water is sitting somewhere exposed to light after being santized it is highly unlikely that it will contain green algae coming out of the tap. It would more likely appear cloudy instead of greenish if that were the case.

 

[mas985]

IBottom line is this. If increasing the chlorine level allows a lower output percentage to maintain the same FC level it really is a moot point whether this is because of more efficienct chlorine production by the cell or simply because the chlorine is not destroyed as quickly by UV light. (I still suspect that both things are going on but really have no way to prove or disprove them. However, if you compare the amount of chlorine lost in a manually chlorinated pool with the same CYA levels and determine just how much chlorine is being produced by the cell at the different output percentages I don't beleive that it will indicate that the difference is solely because of the protective effects of CYA.)

Well one thing I think I proved some time ago is that the SWG cell does not produce any more chlorine with higher CYA. Early this summer, I did about a dozen tests in my spa at several CYA levels and the SWG productution rate was constant over all of the CYA levels and also very close to the specified production rate in the manual. I repeated this test several times and each time production rates were the same. So I am pretty confident in ruling this one out as a cause.

So with production not responsible, I was interested to find out why the protection characteristics of CYA were more than what theory predicts and thus the additional testing that I did. I still think that it is important to understand this process in more detail because it can significantly effect the economics in pool operations and could also impact BBB operation.

I have a few more things to try before I give up on this.

 

[chem geek]

The main reason I'm trying to figure out what's going on by modeling it is so I can confidently make some sort of recommendation of how high a CYA level will give what sort of savings and to understand what conditions make it work better or worse (such as circulation, depth of pool, etc.). Obviously experimental data trumps theory, but the theory can extend our knowledge and make potentially more accurate predictions once the model is correct. With the CYA extinction effect and with perfect circulation I can show that doubling the CYA level doubles the protection so that with a doubling of FC to have the same disinfecting chlorine level, the net result is a wash. That's better than the original conclusion which was that the lowest CYA possible had the least chlorine loss since lower CYA could have lower FC. But the only way I can (so far) figure out how the higher CYA protects the chlorine even faster than this is to assume poor circulation near the surface so that a certain layer of CYA protects chlorine below but the chlorine below does not circulate near the surface (otherwise, it would break down).

As for manual chlorination, several users, including ivyleager (in this thread) and Janet and two others from The Pool Forum, did report that they used far less chlorine at high CYA levels, though obviously to be fair one needs to increase the FC level accordingly. But the savings that was seen was far more than a linear savings -- something non-linear is going on. So I don't think this has anything to do with an SWG, especially since Mark not only tested SWG efficiency at different CYA separately, but also his chlorine loss tests included examples when the SWG was off or not involved at all (including the bucket tests, though we didn't eliminate other possibilities of contamination from the tap water). Everything we have seen so far is consistent with something about CYA protecting chlorine in a non-linear fashion and the only thing I can come up with consistent with this (i.e. no circulation near surface) doesn't seem right.

If I could get an accurate molar extinction coefficient for CYA (and the chlorinated cyanurates, if applicable) in the range of wavelengths where chlorine is broken down by sunlight (mostly 310 to 360 nm), then that could help determine if this is the true effect, though it's still a mystery as to the strong non-linearity (poor circulation notwithstanding).

Richard

 

[BC]

Sounds like a perfect candidate to conduct a Design of Experiments including specific treatment combinations of all the factors contributing to rate of FC destruction. As discussed here, sounds like several 2-way interactions btw factors would calculate out as being statistcally significant.

 

[chem geek]

I think Mark was trying to isolate to the "breakdown from sunlight" factor exclusively. His original tests included baseline results overnight to see if (without sunlight) there was any effect of CYA on SWG production and there wasn't. If he were to use distilled water, then that could eliminate some "chlorine demand" in his bucket tests, but even if the water was being treated with 1 ppm monochloramine as is typically done in many water treatment districts, it wouldn't consume very much FC (only about 0.6 to 0.7 ppm FC oxidizes 1 ppm CC).

The factor that I'd like to isolate is the "poor surface circulation" aspect of the theory. If CYA has a fairly high UV absorption, then the poor circulation in the buckets could mean that a layer near the surface of the bucket gets depleted of chlorine and the lower depths retain their chlorine level. Further chlorine depletion would then mostly be limited by diffusion of chlorine into the surface layers. If the bucket experiments were repeated with some "stirring" of the liquid in the bucket, then that would go a long way in confirming or disproving this hypothesis and would also give us guidelines with regards to pointing returns (though at a price of possibly reduced surface circulation and its effects on skimmer efficiency in removing junk on the water's surface).

Another factor to isolate would be outgassing and convection of chlorine out of the buckets. Using a "transparent to light, especially UV" quartz glass top would limit outgassing. In theory, the outgassing rates should be the same in all buckets since the hypochlorous acid concentration is the same, but if we saw some difference here, then that could mean that perhaps CYA interferes with chlorine outgassing -- I think this is unlikely, but it's something to consider and try to isolate (there's a known concept of "facilitated transport" in surface diffusion so it's possible for there to be "inhibited transport" -- certainly this is partly how the "liquid solar covers" work though that's more by having a concentrated surface layer literally one molecule thick whereas I'm talking about layers that are thicker than that).

It would also be good to independently validate the hypochlorous acid concentration by amperometric measurement or even ORP to ensure that it is indeed the same in all buckets.

The only other factor I can think of that needs to be carefully controlled is the pH since hypochlorite ion breaks down in sunlight much faster (20 minute half-life) than hypochlorous acid (130 minute half-life). Since CYA acts as a hypochlorous acid buffer, this means the hypochlorite ion concentration changes a lot with pH. Going from a pH of 7.5 to 8.0 has the hypochlorite ion concentration increase by a factor of around 2.7. By itself, it's not enough to explain Marks' results unless the pH after dilution changed that dramatically (unlikely given the buffered pool water), but it's nevertheless a factor that should be controlled.

Richard