Chlorine output from SWG

G

Guest

Cliff_s,
I have to agree with Richard after studying his material and from my emperical experience with my customers pools. In addition, I have seen too many pools with SWGs and low CYA and the drastic redution in cell output in EVERY case I have seen (usually from percentages in the 70-90% range) down to around 30% or even lower can't be account for by just the effect of CYA keeping the chlorine from being degraded by UV. Pool size does not seem to play a major part in this either! Some of the larger pools I have seen have not been able to hold a FC level until the CYA was increased but I attribute this to the fact that the amount of chlorine produced by a cell is finite and these pools were already at the upper end of pool size they were designed for. My own pool is very small and when I was first started up my CYA was only at 30 ppm and my generator was at 70% and was only maintaining a 3 ppm FC. Now I am running at 8% with a CYA at about 70 ppm to maintain 4 ppm FC. I do not believe that this drastic difference can be accoounted for simply by the UV inhibiting effects of CYA. In both cases pump run time has been 8 hours during sunlight. I do have 50 ppm borates in my water which does decrease sanitizer demand but even before I put them in I was running at 20% cell output once the CYA was raised.
 

cliff_s

Well-known member
Mar 28, 2007
93
I agree that more actual lab work is needed to find out what is happening
in the salt cell itself.

I have used C02 fro pH control for about a year now. Yes, it has been successful.
I have confirmed that it will negate the additional pH elevation caused by the cell
operation. I am quite amazed at how little C02 addition it takes to lower the pH.
This of course will not control the TA, so some acid has to still be used. The
real problem with C02 is to try and keep it in contact with the water as long
as possible, after all it is a gas. The only way I know is to make smaller
bubbles(which isn't easy).

If I was more ambitious I would investigate the lowering of the Calcium
levels using the plating properties of the salt cell. It should be possible
by reversing the water flow and current at the same time. Sort of like back flushing
a sand filter. I know it will work because I had a EcoMatic non-self cleaning
system and when you cleaned the cell there was a large amount of calcium
that ended up in the bottom of the cleaning jug.

The only way to confirm(or deny) that CYA affects chlorine generation is to
measure the output of a cell with constant current and varying the CYA level.

The only experiments I have tried is to find out where the pool will naturally find
equilibrium with TA and pH. Fill water has a great effect on this along with the
type of surface the pool has. The limestone in cement for some time after the pool is constructed
will react to the pool water chemistry. If you try to move the parameters very far from this
original level you will constantly fight the battle of the pool trying to restore its original equilibrium.
Wouldn't it be better to remove the offenders in the water(calcium and others) rather than
trying to keep neutralizing them.

This whole subject would make a good Phd. thesis, maybe some day someone will rise to the
challenge.

Cliff s
 
G

Guest

cliff_s said:
The only experiments I have tried is to find out where the pool will naturally find
equilibrium with TA and pH. Fill water has a great effect on this along with the
type of surface the pool has. The limestone in cement for some time after the pool is constructed
will react to the pool water chemistry. If you try to move the parameters very far from this
original level you will constantly fight the battle of the pool trying to restore its original equilibrium.
Wouldn't it be better to remove the offenders in the water(calcium and others) rather than
trying to keep neutralizing them.

This whole subject would make a good Phd. thesis, maybe some day someone will rise to the
challenge.

Cliff s
Calcium is necessary in the water if one has a plaster pool. If you lower the calcium then calcium from the plaster finish will start to dissolve in the water. However, many people with vinyl and fiberglass pools have already done this experiment and run extremely low calcium levels with no apparent ill effects (however, there is some conjecture that with a fiberglass pool too low a calciuml level could cause cobalt to leach from the gelcoat or might increase the incidence of staining, which fiberglass is prone to.
 

cliff_s

Well-known member
Mar 28, 2007
93
Ran into a couple of things since my last post. NewScientist.com had an article on
something called Oxychlorine. Here is a partial quote,"by exposing purified water to
sodium chloride and then using electrolysis, various oxychlorine ions are formed.
These kill microbes and viruses, but are present in much lower amounts than bleach".
also,"Oxychlorine ions are the key ingredient, rapidly piercing the walls of free-living microbes and
killing them". It would appear as though this is the same process as our salt cells. It
also seems as though we are generating more than chlorine. Certainly a subject for more study.

Another was a company called Calsaway. They have a large truck which has some filtering
and as they say, patented equipment, for cleaning pool water. They claim they can clean a algae
green pool to crystal clear in 2 to 3 hours. On this truck they also have a process to remove
calcium carbonate. They claim on a average 15,000 gal. pool they remove more than 500 lbs of
calcium at the same time. This eliminates the bath tub ring and deposits around the pool.

I know of no reason to have any calcium carbonate in any pool water. Some parts of the country are
lucky and the fill water has none. Here in Phoenix we have plenty.

Cliff s
 

Buggsw

LifeTime Supporter
Apr 22, 2007
925
Arizona
I wonder if Calsaway is expensive. I'm in Phoenix and just drained and refilled my pool 2 months ago and already have too much calcium hardness.
 

mas985

TFP Expert
LifeTime Supporter
May 3, 2007
12,375
Pleasanton, CA
I was going to wait until I repeated some the results that I have but given this topic is pretty heated at the moment, I thought I would give the results that I currently have. These are preliminary so don't flame on me. I need to repeat them to make sure I did not make any errors.

I have been doing an experiment in my SPA with different CYA levels. The spa is a bit easier to control the CYA levels as well as shorten the run time for the SWG to measure chlorine (24 min.).

Because I use my pool water as a starting point here are the starting numbers:
45 ppm CYA
2.5 ppm Cl
7.4-7.5 PH
110 ppm TA (yes I know it is a bit high but I am having trouble lowering it because my fill water is at 250)
225 ppm CH
3000 ppm salt (test strips say 3400).

The spa is 700 gallons
SWG run time for spa is 24 min.
At 1.45 lbs per day, the SWG should add 4 ppm by my calculations however it only adds about 2.5 ppm. I have not figured out why yet.

So here are steps that I am following.

Step 1 Run the spa in spillover to replace most of the water. SWG is off.
Step 2 Add enough CYA for 80 ppm or skip this step for 45 ppm.
Step 3 Test chlorine level. Pool is about 2.5 ppm so that is what the spa is.
Step 4 Run the spa in filter mode with SWG on for 24 min.
Step 5 Test chlorine level and CYA level.

I performed these tests in early evening after the sun has gone down.

So far I have done one test at 45 ppm CYA and one at 80 ppm CYA and both seem to be raising the chlorine level about the same amount, +2.5 ppm. I cannot tell the difference of < +-0.25 ppm in chlorine with the test kit so there might be a slight difference but that would probably be less than 10%.

However, what I did notice was a very large increase in chlorine retention over the next few days. I let the water sit in the spa after each test for a few days in direct sunlight.

At 45 ppm, the chlorine levels dropped by 50% per day or from 5 ppm to 2.5ppm and then down to a bit over 1 ppm the second day.

At 80 ppm, I hardly noticed a drop the first day (~4 ppm) and the second day the chlorine level was ~3.5 ppm . This is a retention of about 85% per day.

My plan was to repeat the tests next week to see if I get the same results.

I also ran a test with my pool cover on and off for a week each (I did the above tests during the week it was on). The pool is at a CYA of about 45 ppm and my pool cover is clear plastic. With the cover on at the end of the week, the chlorine was at a level of about 2.5 ppm. Without the cover, the chlorine level did not seem to drop much maybe 2.0 ppm. So the clear cover does not seem to help much for chlorine degradation.

Any suggestions on test modifiers?
 

JasonLion

TFP Expert
Platinum Supporter
LifeTime Supporter
May 7, 2007
37,880
Silver Spring, MD
This is facinating. The diminishing returns of chlorine lifetime with additional CYA chart that everyone uses must be completely wrong if your data is correct. The only theory I have heard that could have modified that chart involved deep bodies of water, and I assume your spa isn't all that deep.

Your two CYA levels seem like reasonable test points. Perhaps you could increase the cell run time so you could more reliably measure small differences in cell efficency.

Great work!
 

chem geek

TFP Expert
LifeTime Supporter
Mar 28, 2007
12,082
San Rafael, CA USA
I started a discussion a short time ago about alternative explanations for CYA preventing chlorine degradation since some users reported a significant improvement at higher CYA levels, far higher than found in the normal chart. This link is to that thread in the Pool Forum which I've copied over here, at least for my main posts (I don't want to copy other's posts without permission). One alternate explanation involves having CYA itself "shield" lower depths of water, but this effect should be seen more in pools than in a spa which I presume is no more than about 4 feet deep. I'll also have to play around with the chlorine vs. chlorinated isocyanurate half-lives to see if that helps explain what is being seen. What is certainly interesting was the data I found that higher chlorine levels themselves tend to shield lower depths from the UV so apparently not every UV "event" affecting chlorine breaks it apart permanently.

Say Mark (mas985), I live in San Rafael and work on the Peninsula so we live pretty close. Great experimenting, by the way. The only modification I would suggest, if you didn't already do this, is that when you have the SWG on to see if it's output rate changes at different CYA, then that should be done either at night or with an opaque cover on the spa, just to eliminate the UV losses. However, the relatively short time pretty much makes your results very solid -- I'm just being picky. It does seem from your experiment that certainly in your situation the CYA increase leads to greater protection of chlorine and not to greater efficiency of chlorine generation. Certainly, if there are losses from sunlight then one could indeed lower SWG output at higher CYA levels since less would be destroyed from the sun AT THE SAME FC level. The trick, of course, is to see if having a higher FC level for the higher CYA level still makes the overall "absolute" chlorine loss less because only that will allow for a lower SWG output. So after you do your experiment at the same FC level with different CYA levels, then you can repeat the experiment using constant disinfecting chlorine levels, say the 0.05 column in this chart.
 

chem geek

TFP Expert
LifeTime Supporter
Mar 28, 2007
12,082
San Rafael, CA USA
Cliff,

Oxychlorine is chlorine dioxide and is another sanitizer that water sanitation systems use (along with chlorine, ozone and UV). It can be manufactured from electrolysis of a brine (saturated salt) solution (though usually chlorine gas and hypochlorite result from that -- special conditions are needed to get chlorine dioxide) or from sodium chlorite, NaClO2 (also typically via electrolysis) or from sodium chlorate, NaClO3. I'm not sure what some claims are to oxychlorine ions and quite frankly negative ions don't do well killing pathogens since the cell walls of pathogens are negatively charged. Chlorine dioxide is neutral (just like hypochlorous acid) and is the true disinfectant. Chlorine dioxide is more reactive than chlorine, but that also means it doesn't last as long so doesn't provide a good residual. I don't know if it breaks down in sunlight. [EDIT] I found a source that says that chlorine dioxide breaks down VERY rapidly in sunlight with a half-life of less than 20 seconds (compared to 15-30 minutes for hypochlorous acid without CYA). I also found sources saying that chlorine dioxide is effective against protozoan cysts such as Giardia and Cyrptosporidium for which chlorine is not effective (in practice). [END-EDIT]

As for calcium carbonate, you absolutely want to have a saturation level of calcium carbonate in the pool water if such water is exposed to plaster/gunite/grout otherwise such surfaces will dissolve (corrode) into the water. It is true that too much calcium or too much carbonate are a problem that can lead to scaling so for certain well water high in both (CH and TA) their solution is a good one, but you can lower CH yourself using a water filter (ion exchange resin). I'm not sure how they remove the carbonate except possibly through acid and aeration.

Richard
 

mas985

TFP Expert
LifeTime Supporter
May 3, 2007
12,375
Pleasanton, CA
First, thanks for the feedback.

Next, to be clear, I did the tests after sunset so I assume there is little UV if no loss during the test.

Also, for pool cover test, I would have thought that even a clear solar cover would block most of the UV light so I am wondering if other wavelengths break down chlorine just as well. Even though no one swam in the pool during the test, it could also have been organics in the pool which reduced the chlorine level but this is difficult to control.

Next, I thought a bit about chlorine residuals and retention. If you model the chlorine residual and SWG as an infinite sum, since it adds the same chlorine each day, then one should expect the residual to have the relationship, assuming the only chlorine loses are due to sunlight, chlorine is added at night and the residual measured in the morning:

CLr = CLa + CLa * f + CLa * f^2 + CLa * f^3 + ....................

CLr = Measured chlorine residual
CLa = Chlorine added each day
f = fractional retention 0.5 or 0.85

The infinite sum reduces to

CLr = CLa / (1 - f)

So let's say you add 1 ppm per day. Then at 50% retention for the 45 ppm case, the residual should be about 2 ppm. But in the 80 ppm case, the residual is over 6.5 ppm. An over 3x increase in residual for < 2x increase in retention.

So if the tests are correct, this would easily explain the large increases that some see at 80 ppm.
 

chem geek

TFP Expert
LifeTime Supporter
Mar 28, 2007
12,082
San Rafael, CA USA
The continuous breakdown of chlorine over time is modeled in one of two ways depending on whether the chlorine level is maintained or is left to decline on its own.

If left to decline, then this is a differential equation:

dC/dt = -kC

as a first-order rate reaction where the rate of degradation of chlorine is proportional to its concentration. Experiments with chlorine degradation by sunlight and with chlorine degradation from other processes, including oxidation, show that this is the case (chlorine breakdown from "self" destruction is a second-order process, but is very slow at the low concentrations in the pool -- it's more noticeable in chlorinating liquid).

The solution to the above equation is:

C = CO*e-kt

If chlorine levels are maintained, as with an SWG, then the amount of chlorine usage is simply a constant based on the differential equation, namely "-kC" with the negative sign referring to a reduction or loss of chlorine.

All of the above has been substantiated by experiments and is not in question. What is in question is two things: what happens with chlorine at different depths (i.e. does chlorine itself shield lower depths from UV) and what happens when there is CYA in the water. I found some info in the thread I referred to regarding the first question of chlorine shielding lower depths. As for CYA, I also know the amounts of each chemical species in the water. What is not known is the true half-life of the chlorinated isocyanurates (chlorine combined with CYA) though the standard graphs "imply" a certain half-life, and it is not know whether the chlorinated isocyanurates nor whether CYA itself shields lower depths of chlorine from UV.

So, another experiment that would eliminate the "depth" shielding issue would be to measure chlorine loss in a shallow pan or tub at different CYA levels. This would then show the direct chlorinated isocyanurate half-life effect. At greater depths, the shielding effects would be more important and we can use your data plus Janet's to try and come up with a reasonable model.

Richard
 

mas985

TFP Expert
LifeTime Supporter
May 3, 2007
12,375
Pleasanton, CA
Actually, I wasn't trying to model the chlorine degradation directly since I have a measurement of it. I was trying to estimate the build up of chlorine over a large number of days using the degradation measurement instead of theoretical degradation. Since the addition of chlorine is not continuous (pump run time only) and the degradation is not continuous (day time only), an infinite series should be a valid way to model the build up. However, there is a direct relationship between the term "f" I used in the series and your equation for degradation:

fn = e-k(nt) with t = 1 day of degradation, and n is the day number then 2t is two days, 3t is three days, etc.

So in the CYA = 80 ppm case, e-k(1t) = 0.85 for 1 day and e-k(2t) = 0.85 * 0.85 for 2 days, etc.
 

chem geek

TFP Expert
LifeTime Supporter
Mar 28, 2007
12,082
San Rafael, CA USA
At 5 ppm FC and 45 ppm CYA, the disinfecting chlorine (HOCl) level is 0.048 ppm (in Cl2 units). At 5 ppm and 80 ppm CYA, the disinfecting chlorine level is 0.025 ppm or about half. So even if the chlorine combined with CYA did not breakdown at all in sunlight (which is not what the "traditional" half-life curve shows, but we don't really trust that anymore), one would expect the rate of loss of chlorine to be about half the rate at 80 ppm as at 45 ppm, but you saw a greater protection than that -- roughly 30% of the loss rate. This implies that a shielding effect must be occurring so that's what I'd like to calculate.

The shielding can occur in two ways. One is from the chlorine combined with CYA and the other is from CYA by itself. I found an extinction coefficient for Cyanuric Acid of 6.283 OD220 units-ml/µmol at 220 nm which is in the same area of the spectrum where hypochlorous acid breaks down. I'll see if converting this info to appropriate units and typical depths leads to anything close to what you and Janet are seeing.

One thing is for sure from your experiment and Janet's data -- the traditional graph of diminishing returns is wrong. A key experiment will be the shallow depth experiment since that will determine how much of this protection is from shielding vs. having chlorine attached to CYA -- we know shielding is important, but it would be good to know if the chlorinated isocyanurates themselves degrade at all under UV and how long they take.

Richard
 

mas985

TFP Expert
LifeTime Supporter
May 3, 2007
12,375
Pleasanton, CA
One thing I could do at least at a CYA of 45 ppm, is let both the pool and the spa at the same chlorine levels sit for a day or two and see if the chlorine drops off at the same rate. I could even add a small bucket of pool water, 1 foot deep, next to both.

At 80 ppm, I could just test the spa and bucket alone.

Ideally you would want to measure the chlorine at different depths in the pool but I am not sure how to do that effectively so perhaps mixing the water a bit before the measurement should give a decent average CL level. I suspect that even without circulation, the water moves around due to temperature differences so measurements at various depths may not be all that conclusive anyway.
 

chem geek

TFP Expert
LifeTime Supporter
Mar 28, 2007
12,082
San Rafael, CA USA
Mark,

Yes, you are right that measuring the chlorine (without CYA) at different depths was something that I saw that one study did, but unless they "isolated" the different depths with a membrane or UV-clear plastic or something, then I would expect normal diffusion to ruin such measurements. So your "different depths" measurements will be a decent proxy and given the theory of "extinction" or "absorption" we should be able to come with a model that matches what's going on. Again, GREAT experimenting!

A big lesson is learned by this. Don't trust industry data. Most of what I've calculated has been from independent scientific studies, but the half-life of chlorine in the presence of CYA was taken from the graphs that the PPOA (Kent Williams) and others had shown and apparently they were either wrong or something else was going on. I'll also take another look at the original Fuchs patent since that has some data in there as well -- "shallow" lab tests vs. "deeper" pool tests. The half-life of chlorine without CYA has been studied a lot, but with CYA it's only industry data that I can find. So you're really doing groundbreaking work here (well, OK, we're not saving lives, but still...).

Richard
 

chem geek

TFP Expert
LifeTime Supporter
Mar 28, 2007
12,082
San Rafael, CA USA
Mark,

An additional experiment to try after we sort out the chlorine/CYA protection from UV is to add 50 ppm Borates to the spa (and take a bucket of water from the spa) to see what difference that makes in SWG output rate (if any -- none expected) and in protection of chlorine from UV or some other factor (e.g. killing algae so chlorine loss is less -- I'm not so sure that can account for the large reduction waterbear saw since most chlorine is supposed to be consumed from UV, then organics, then algae, then bacteria in that order typically, but then that's more of what the industry says).

Richard
 

mas985

TFP Expert
LifeTime Supporter
May 3, 2007
12,375
Pleasanton, CA
Too late for that since I already have 50 ppm of borates in the pool. However, I could replace the water in the spa with fill water for 0 borates and 0 CYA to get a baseline. Then add CYA. A close to 0 test of CYA might be interesting just to see how quickly the chlorine disappears.
 

chem geek

TFP Expert
LifeTime Supporter
Mar 28, 2007
12,082
San Rafael, CA USA
That's a great idea since the more data points the better our model predictions should be. It also seems that pH should have an effect on chlorine loss since hypochlorite ion (OCl-) has higher absorption of UV and higher breakdown from UV compared to hypochlorous acid (HOCl) -- this info is from two different sources, so I'm reasonably sure this is true. In addition, the CYA protection is stronger in the lower wavelengths where HOCl absorbs more so in theory one should see greater chlorine stability from sunlight at lower pH compared to higher pH. So if you do your fresh fill of the spa to check out what happens at "normal" levels and a pH of 7.5, you could also test a lower and higher pH -- if you have time and nothing better to do :wink: . Believe me, I can go nuts with experiments and this pH one isn't critical since most people run around 7.5, but if the variation with pH is significant, that would be good to know (probably more important to measure the higher pH, say 8.0 for the experiment, than the lower since some pools are run at 7.7-7.8 to avoid the pH rise problem). Just based on the difference in half-life (not counting different CYA absorption at different wavelengths), 3 ppm FC at 30 ppm CYA should vary from a half-life of 7 hours at pH 7.0 to 6 hours at pH 7.5 to 4-1/4 hours at pH 8.0, just to give you a rough idea.

Your having a separate spa with the piping option to filter it separately from the pool has turned out to be an excellent experimental setup! I've updated this post with some more info. As soon as you get some summary results, I'll plug away at calculating a molar absorption/extinction coefficient and can then put that into my spreadsheet so I can generate a new chlorine/CYA graph for protection from sunlight with different depths of pools.
 

mas985

TFP Expert
LifeTime Supporter
May 3, 2007
12,375
Pleasanton, CA
Completed another week of tests and here are some more results.

Chlorine Production

Verified at CYA of 45 ppm and 80 ppm that they both have nearly identical chlorine production rates. 2.5 ppm over 24 min.

So it would seem to be that this is pretty conclusive that chlorine production in a SWG does not depend on CYA. I may try one more test at 100 ppm just to make sure.

Also performed a test for PH 7.1 and 8.0. Ph 7.1 production was again 2.5 ppm over 24 min but PH 8.0 production was between 1.5 and 2.0 (1.75 estimate) over the same 24 min or a 30% decrease in chlorine production. Given the results, I want to repeat this one more time to make sure.

Chlorine Protection

Verified that at 45 ppm, the water loses about 50% of the chlorine per day. Unfortunately, it got cloudy the last few days so I could not rely on a consistent test for 80 ppm. I will have to wait until the sun is a bit more constant for a repeat test at 80 ppm. As reference, the last test showed only 15% lost per day for 80 ppm.

Also, I did a side by side of pool, spa and small bucket all the same water (CYA = 45 ppm, CL start = 3.0 ppm). The pool and spa each lost about the same amount of chlorine over 24 hours ~50%. However, the bucket lost all of the chlorine or at least well below the 0.5 ppm lower limit.

I would assume that the water itself provides UV protect at larger depths so this may not be all that surprising.

The forcast is to be cloudy all week next week so I am not sure I will able to do any more meaningful protection tests.
 

Other Threads of Interest