CSI calc - out of balance?

[polyvue]

Are the calcuations that were documented in 2007 for CSI still valid?
See viewtopic.php?p=14664#p14664

The result differs from my LSI calculation (the CSI tends to be .05 - .25 lower than mine.)
If the CSI calculation is correct, additions of MA to lower pH have consistently pushed the saturation index into corrosive territory… what to do?

In the example below, I acted to lower pH with MA from 7.7 to 7.4
Per CSI calculation, this would result in -0.60 CSI
To achieve a CSI near zero, it seems the right action would have been to increase pH to 8.1
Am I missing something? Thanks.

INPUTS

Temp. °F 48
pH 7.7
TA 100
CH 275
CYA 50
NaCl 2600
Borates 5

OUTPUTS

Temp. °C 8.89
extra NaCl 2278.80
CarbAlk 82.05
Ionic strength 0.05

CSI -0.31

 

[JasonLion]

CSI is an approximation, however it is a more accurate approximation than LSI. LSI simplifies things in several ways that make it easier to calculate by hand with just a few very small tables. CSI uses some math that isn't easy to do by hand to get more accurate results. But neither one is really showing exactly what the plaster will do.

As a general rule, you want the PH to go up as the water gets colder. With a water temperature of 48 F you want the PH to be about 0.3 higher than you would keep it at swimming temperatures (85 F). This will normally happen "automatically", the PH shifting on it's own as the temperature goes down.

You don't generally want CSI (or LSI) to be at exactly 0.0. Small, or even medium, CSI changes really don't make any difference. You can let CSI vary a fair bit without causing any problems. Secondly, if you have either a heater (of any kind) or a SWG you want CSI to be a little below zero (it will be higher inside the heater and the SWG).

I get the same CSI results, given your numbers, as you do. By the by, my Pool Calculator will calculate CSI for you, so there isn't any need to do all that math.

 

[chem geek]

With your numbers using a pH of 7.7, The Pool Calculator gives -0.31 and the slightly more accurate calculation in my spreadsheet gives -0.32, but the LSI calculation gives -0.29 which is not 0.05 to 0.25 lower (though sometimes it does differ, but usually not by more than about 0.1). Where are you getting your LSI calculation? There are quite a few formulas on the web for that, but the normal ones use the formula given on this page, though often using tables derived from these formulas.

What source are you using for your LSI calculation? CSI is theoretically better than the LSI mostly because the LSI doesn't have the right temperature formula nor TDS as they use logarithms while the real relationships are more complex (inverse absolute temperature and a formula using the square root of ionic strength). Some newer formulas that Wojtowicz refers to result in a saturation index that is around 0.15 higher than the CSI that I use, though I have the Wojtowicz option in my spreadsheet. I chose to stick with numbers that are close to the Taylor Watergram and that use the standard thermodynamic values from CODATA.

Richard

 

[polyvue]

As a general rule, you want the PH to go up as the water gets colder. With a water temperature of 48 F you want the PH to be about 0.3 higher than you would keep it at swimming temperatures (85 F). This will normally happen "automatically", the PH shifting on it's own as the temperature goes down.

Thanks for that information; it’s entirely new to me. You suggest that this is a temperature issue and an examination of my logs tells me you’re right. If I had read here that it was desirable to maintain a higher pH to accommodate colder water I have since forgotten it. Nor was I aware that a CSI -0.31 was within normal limits. Confusingly, the pool calculator insists that such water has “Potential to become corrosive to plaster.”

I get the same CSI results, given your numbers, as you do. By the by, my Pool Calculator will calculate CSI for you, so there isn't any need to do all that math.

I quite realize that the Pool Calculator is a convenient vehicle for determining CSI but for me there is compelling need to do the math. Assuredly, I would not have found this forum had I fully accepted the confident assertions made by those who know a lot more than I do about pools, chemistry and best practices. Granted, some questions are purely academic (though not this one.) But I’ve enjoyed the exploration and I hope you’re not too tired of my questions!

 

[polyvue]

What source are you using for your LSI calculation? CSI is theoretically better than the LSI mostly because the LSI doesn't have the right temperature formula nor TDS as they use logarithms while the real relationships are more complex (inverse absolute temperature and a formula using the square root of ionic strength). Some newer formulas that Wojtowicz refers to result in a saturation index that is around 0.15 higher than the CSI that I use, though I have the Wojtowicz option in my spreadsheet. I chose to stick with numbers that are close to the Taylor Watergram and that use the standard thermodynamic values from CODATA.

Thanks for the links! I really don’t recall my source for the LSI calculation; it was acquired months before I built my pool as part of a knowledge base. I likely pulled it from an industry journal or pool school web site and it is clearly derivative, consisting of factor tables that feed a simple equation. I’ve used it in conjunction with the Taylor Watergram, which has often (but not always) indicated a lower SI than the LSI sum using tables. During the first year after building my pool I relied upon it to guide treatment of the pool water. Now it’s more of an exercise.

My question is primarily concerned with whether my efforts to date to suppress pH have been unproductive or potentially damaging. When I have used MA to bring pH down from 7.9 or 8.1 to a more reasonable 7.5/7.6 it often results in pushing an already slightly negative LSI into corrosive territory. Using the improved model for Calcite SI the effect is even more pronounced. Adding acid to adjust pH from 7.7 to 7.3/7.4 (not an unreasonable thing to do, you might agree) the CSI would plummet to between -0.6 and -0.8.

By the way, I utilized the simpler calculation that inputs chloride measurement in lieu of TDS. Were I to be able to derive a value for TDS to use as input in the alternative calculation, would that produce a more qualitatively useful result?

 

[chem geek]

As the temperature of water drops, the pH naturally rises due to the chemical equilibriums involved and this tends to keep the saturation index fairly constant (only dropping a little). So you normally should not fight this and should let the pH rise as the water gets colder. My pool is now at around 50F and the pH has risen from around 7.6 to something between 7.8 and 8.0 and I'm just leaving it there. The saturation index dropped by a little less than 0.1 unit. If I had tried to adjust the pH to keep it the same, the saturation index would have dropped by around 0.3 units.

So I would not add acid to lower the pH as you have been doing. If you were to do this in the long run, then you are risking plaster degradation, though at colder temperatures everything takes a lot longer (i.e. chemical reactions slow down). Nevertheless, why are you taking such chances? It is so easy to maintain an appropriate saturation index by having sufficient calcium in your pool -- why aren't you doing that? At least having the saturation index be at or above -0.3 -- I usually target -0.1 though -0.2 would be OK. At -0.3, there is half the concentration of calcium carbonate (as a product of concentrations) than there is at saturation. In practice, plaster degradation happens at lower pH -- the saturation index does not predict rates but only whether plaster degradation (or scaling in the other direction) is possible, not how quickly it occurs. At a saturation index very close to 0, neither degradation nor scaling are possible, though there is still a slow "exchange" of calcium carbonate between the water and plaster, but it is equal back and forth.

[EDIT] For SWG pools, it is usually best to target a lower CSI, usually around -0.2 rather than 0, to prevent scaling in the salt cell. The use of 50 ppm Borates also helps prevent such scaling by buffering the pH to prevent it rising as much at one of the cell plates. [END-EDIT]

When you use my chloride input in the calculation, it automatically adds in the additional TDS coming from other components such as calcium and bicarbonate. So either way is fine to use. It's usually very hard to know the accurate TDS. Usually it is just estimated from conductivity, but you need to know or make assumptions about the composition of the water and not just assume it is all sodium chloride salt. My calculations account for all of this, but I do assume that the calcium level is created by adding calcium chloride. This works well for most water, but for certain hard water where the calcium may have come from calcium carbonate and that pH lowering came from compressed carbon dioxide, then there may not be as much chloride as I assume. Nevertheless, such errors are very small because it takes large movements in the TDS number to affect the saturation index. You have to almost triple TDS from 500 to 1400 to have the saturation index drop by 0.1 units though doubling from 1500 to 3000 results in another 0.1 drop but that would bring you to saltwater chlorine generator (SWG) pool levels.

As for your older LSI calculation, I would throw that away as it isn't even the normal or typical LSI formula used (apparently).

Richard

 

[JasonLion]

By and large, tracking your CSI isn't especially important. If you stick to chemical levels within the recommended ranges, there is no need to think about what your CSI is. This works because a fairly wide range of CSI values are all acceptable in practice. I tend to aim for -0.2 as a target, but expect that to vary (say within a range of -0.4 and +0.2). Even noticeably higher and lower CSI levels are usually fine for relatively short periods, several days at least in most cases. (Typically PH tends to rise over time, which will raise CSI, and CSI levels inside a heater or SWG will be higher than the levels in the bulk pool water, which is why my target level is negative.)

The most common situation where tracking CSI becomes important is for people in arid climates with high CH levels in their fill water. In this situation it is valuable to allow CH levels, to get very high, to avoid having to drain and refill the pool more than once a year. When dealing with such a situation it becomes important to adjust your PH and TA levels to compensate for your elevated CH. CSI is a good way to estimate the changes to your target levels required in this situation.

 

[polyvue]

As the temperature of water drops, the pH naturally rises due to the chemical equilibriums involved and this tends to keep the saturation index fairly constant (only dropping a little). So you normally should not fight this and should let the pH rise as the water gets colder.

Pragmatically, that's my takeaway from this discussion.

By and large, tracking your CSI isn't especially important. If you stick to chemical levels within the recommended ranges, there is no need to think about what your CSI is.

You don't mean it! :-D If it's not very important then I wouldn't bother to measure it. I seem to recall that some smart person :idea: somewhere authored an application in Java script and HTML that provides a mechanism for tracking CSI. Moreover, it was precisely this: sticking to chemical levels within the recommended ranges, that brought me to the question at hand. Note this rather didactic passage:

Adjust your pH to 7.5-7.6 and not any lower. Monitor your pH and when it climbs to 7.8 add acid to lower it back to 7.5-7.6 (This is also IMPORTANT!)

From the article, Water Balance for SWGs, its content hardly conflicts with other sage advice dispensed in Pool School, and elsewhere. It's no good to say that these levels are only "suggested", the advice to maintain pH within the recommended range seems to fall somewhere between a prescription and a commandment. What I have learned from this is that the SI is a rather important tool that can be used to predict or validate the usefulness of the so-called recommended ranges. Is this the wrong lesson? I won't be dumping any more acid in the pool trying to fight off a measured pH of 8.0 because I now have an improved understanding of the relationship between temperature and pH level and a useful tool called the Pool Calculator that provides corroborative information in the form of a calculated CSI. To have stuck with NSPI/APSP's dogmatic 7.4-7.6 seems to me, in the light of this discussion, to have been the more foolhardy choice. If I may dare recommend some modification to Pool School material, it would be to include at least a paragraph in articles and a starred footnote in tables that explains or notes the temperature/pH relationship so that readers can benefit from information that took me three years to come by.

ABCs of Pool Water Chemistry
Water Balance for SWGs
Recommended Levels
The Pool Calculator

You have to almost triple TDS from 500 to 1400 to have the saturation index drop by 0.1 units though doubling from 1500 to 3000 results in another 0.1 drop but that would bring you to saltwater chlorine generator (SWG) pool levels.

Oh, how I wish I'd never heard of TDS! It may be my undoing, but at this point I'm unwilling to turn back from the journey already begun. It matters. It doesn't matter. It sometime matters. Wide disagreement over how to report its constituents, differing views as to the seriousness of its effects on potable water, swimming pools, waterways, wildlife and health. TDS, like phosphates and sulfates, may rightfully be relegated to Advanced Topics, well outside the purvue of the average pool owner but it's not purely academic, unless the entire notion of measuring calcium carbonate saturation is bogus. And I don't think anyone here is suggesting that it is. Perhaps one day I can (short of sending a sample off for gravimetric testing - that would be too easy!) compile a definitive report listing every known element in my pool water. Then I'll have to figure out what to do with it. But that's an issue for another day. Thanks for your contribution to my understanding, as always.

Greg

 

[JasonLion]

There are many different levels of detail. If you really want to get into the chemistry there is lots and lots to learn. But if you just want to maintain your pool with as little effort as is practical then there are all kinds of things you can ignore. Also, there are many different ways of doing the chemistry, all of which have different minor advantages and disadvantages, but all of which work most of the time.

In the Water Balance for SWGs article the reasons for keeping PH in that range have nothing at all to do with CSI. Those "rules" are about getting into a routine where you expend minimum effort and still keep the PH in perfect balance.

When talking about all of this stuff, you need to decide if you just want to take care of your pool, or if you really want to understand the chemistry behind things. The tricky part is that knowing some of the chemistry can be very misleading. People are constantly learning a little bit and then "reasoning" from what they know to completely invalid conclusions. Pool chemistry is very complicated if you really want to know what is going on behind the various guidelines. At the same time, 99% of the time the chemistry can be extremely simple if you are happy following the suggestions and just want to have a nice trouble free pool.

Most of what appears in pool school is written for people who don't want to know the details, they just want something that works. One challenge in that kind of writing is keeping it simple and short enough that people will actually read it. The other challenge is writing something that will result in people taking the actions we recommend. It is amazing how many times we say to adjust the PH to X and people actually adjust the PH to Y, all the while believing that they are following our advice.

Suggestions for improvements to Pool School are always welcome. I don't have the slightest idea where to put the "let PH rise a little in the winter" suggestions, but I will try to work it in the next time I am making revisions (which will be in the early spring, so fairly soon now).

TDS is, to my way of looking at things, completely useless. If you don't have a CYA test or don't have a salt test, and can't get either one, then TDS could be used as a substitute. But having the CYA and salt tests is far more useful and eliminates any need to know the TDS. Essentially all of the comments about the importance of TDS levels date back to a time when CYA tests did not exist.

 

[polyvue]

When talking about all of this stuff, you need to decide if you just want to take care of your pool, or if you really want to understand the chemistry behind things.

As you may have already surmised, there's no dichotomy for me, it's a little of each....

Suggestions for improvements to Pool School are always welcome. I don't have the slightest idea where to put the "let PH rise a little in the winter" suggestions, but I will try to work it in the next time I am making revisions (which will be in the early spring, so fairly soon now).

I'm grateful. It's a bit of insider dope that adds value to some already very good articles. Though more difficult, a tight rewrite of portions of the Further Reading threads would enhance Pool School's catalog, rendering it more comprehensive. Even as it stands, the information in Pool School compares favorably with similarly named pages in those brand x forums hosted by pool supply companies, many of whom haven't updated the content or their home page since 1983. But a few are pretty good. No need to elaborate on the shortcomings of other sites where relentlessly sales-driven information pages betray not the slightest sense of embarrassment at regurgitating industry platitudes, aiming every banal suggestion toward placement of another product in the shopping cart!

TDS is, to my way of looking at things, completely useless [...] Essentially all of the comments about the importance of TDS levels date back to a time when CYA tests did not exist.

I'm not aware of its history and I was speaking in a slightly wider context than the current controversy regarding TDS testing for the purpose of corrective treatment in recreational pool water, yet since it is either measured or estimated as an input to CSI, surely it has some value. If high levels are known to impact pH, that may be another reason for assessment. Lastly, if tests for cyanuric acid and sodium chloride levels obviate measurement and anaylsis of TDS, there's a lot of pointless testing going on around the world in many different industries. But, perhaps that was your contention.

 

[chem geek]

Lastly, if tests for cyanuric acid and sodium chloride levels obviate measurement and anaylsis of TDS, there's a lot of pointless testing going on around the world in many different industries. But, perhaps that was your contention.

That is the correct conclusion -- don't bother with TDS. Technically, the saturation index needs as input something known as ionic strength. TDS is just a rough estimate to get to that, but it isn't particularly good, especially since TDS measurements aren't very good and are in the wrong units (mg/L whereas ionic strength depends on the charges on the ions themselves, not their weights). The salt level, as measured chloride converted to sodium chloride units, isn't perfect either if just translated directly to TDS in the classical LSI formulas. My formula for CSI uses as much of the real thing as possible, so if you put in salt, then I automatically add the other components from calcium to bicarbonate to CYA to borates to properly calculate the ionic strength and this is all done in The Pool Calculator as well.

Since it takes rather large changes in the ionic strength to affect the saturation index, in practice it's only the salt level that really matters as an input, above and beyond what is already known from the CH, TA, CYA and Borates readings (all of these contribute to ionic strength). The only exceptions I can see would be water that is very, very hard that has magnesium in it we aren't accounting for and water that has a lot of sulfates added to it. For magnesium, one could use a Total Hardness test and assume the difference between that and CH is magnesium. I don't think this is worth checking except for really extreme very hard water.

Richard