ORP Control

[lembi]

since instead it's reading an ORP proxy that is affected by other oxidizers in the water such as dissolved oxygen and by the hydrogen gas from the SWG.

Hello Richard,
can you explain this a little closer?

We are developing a web-based PoolController, dosing liqiud chlorine based on ORP readings. This works pretty well. But we still have "problems" with ORP readings if an SWG is used. It works for about 30-40% of the pools "normal" and for the rest it works ... let's say something between "bad" and "unusable".

"bad" means ... if you try to hold an ORP Level near or over 750mV you need FC levels above or over 2.5 .. 3.5ppm (0 CYA)
"unusable" means, that in some cases the ORP level breaks down to values between -200mV and +200mV and there is no more stable reading possible. The value is kind of bouncing in a range of +/-200mV within minutes.

I've already talked to a chemist about that and in his opinion hydrogen should not be the problem. He said that hydrogen only poorly dissolves in water, and if so, it should be oxidized by the HOCl...?! Well i'm not a chemist ...and maybe you are more experienced in this field. Do you have any further Information about that?

What i don't really understand: it is working in some pools and in some others not. It seems that pools with an "oversized" SWG (i.e. 4000-6000 gallons - 40g/hr SWG) work pretty well. (780-800mV+ at about 0.5 to 0.8ppm FC - 0 CYA). The problems seem to appear only in some pools with "normal" sized cells.
In each case, where the ORP value is bouncing or extremly low a small amout of calcium hypochlorit or sodium hypochlorit, added to the water, is enought to get back a high ORP value. So it looks like something is oxidized then. Do you know what's happening here?

Regards,
Alex

 

[chem geek]

Re: Total Control - ORP vs. Chlorine PPM - Confused

Welcome to TFP!

I'm not a chemist either, but can tell you from what I've read and from basic knowledge. SWG pools with "normal" sized cells will be running a lot longer (i.e. have a % ontime closer to 100%) and that means they will be generating hydrogen gas bubbles for a lot longer and of course generating chlorine gas a lot longer. Of course those with oversized cells should be generating the same amount of hydrogen gas over time, but doing so more quickly (just as with chlorine) but this may mean that more hydrogen gas bubbles outgas from the water and have less time to dissolve. Basically at the higher output, the bubbles may coalesce more and outgas more. With lower output, they remain smaller and distinct longer and therefore can dissolve better. Also, running longer gives less opportunity for the pool to outgas the hydrogen gas so it will be at a steady-state level longer (though if that were the only factor then those pools with oversized SWG systems should see the problem show up periodically some time after the SWG is on).

Do you notice any correlation with flow rate through the SWG? If the problem is worse at faster flow rates then the hydrogen gas bubbles would be smaller and would likely linger longer in the pool before being able to outgas so the water may get more saturated. The same would be true if there was more of a problem in pools with returns that were pointed downward rather than upward. Finally if there were more of a problem in pools with still water compared to those with aeration features (waterfalls, spillovers, fountains, etc.) then that would also be consistent with some sort of gas transfer.

As to exactly why SWG pools are worse off with ORP I don't believe is known definitively. There is speculation that the hydrogen gas generation is the culprit. Henry's Law constant for hydrogen gas is 0.00078 M/atm which is fairly comparable to nitrogen gas at 0.00061 M/atm so it is true that it does not dissolve well in water but some will dissolve and the gas transfer rate out of a pool is not particularly fast. So water can get over-saturated with hydrogen gas just as it most definitely does with carbon dioxide gas (pools are intentionally over-carbonated to provide pH buffering and to protect plaster surfaces by saturating the water with calcium carbonate). By the way, at 1 atm of the hydrogen gas bubbles, 0.00078 M (moles/liter) would be 1.56 ppm (mg/L) so if the water got fully saturated from the SWG, then that's how much aqueous hydrogen there could be (it won't get fully saturated, but I'm just giving you some perspective on worst case).

What is interesting is that if you add some hypochlorite, the problem goes away. Adding hypochlorite raises the pH so you should first see if just raising the pH also resolves the problem. It is true that chlorine should oxidize aqueous hydrogen gas though I do not know the rate for that reaction. It doesn't make much sense that chlorine already in the water wouldn't react with the hydrogen but adding additional chlorine would except of course locally where you add the chlorine since it's higher in concentration (so would get oxidized faster) there and higher in pH. Does it matter where you add the chlorine in the pool for resolving the problem? Does it work to add it away from places where the water would soon return to the circulation system to the ORP system?

ORP measurements are very dependent on the integrity of their membranes. One obvious question: where did you put your ORP system? I assume you didn't put it after the SWG and instead have it somewhere in the circulation system before the SWG, correct?

By the way, everyone's ORP system seems to behave quite differently not only in the absolute ORP reading for a fixed FC level at a given pH with no CYA, but also the mV per doubling of FC also differs and of course none of it correlates with theory (Nerst Equation) since the doubling rates imply less than one electron for the half reaction. So what does your ORP vs. FC standard look like (i.e. for a few spread out FC levels)? Generally speaking, ORP works OK for process control using a setpoint, but isn't so good for determining the FC level even holding other parameters fixed. One usually needs to validate using a test kit and occasionally adjust the setpoint.

 

[lembi]

Re: Total Control - ORP vs. Chlorine PPM - Confused

Thanks

>>and that means they will be generating hydrogen gas bubbles for a lot longer and of course generating chlorine gas a lot longer
I also thought about that. But i think the proportions of generated hydrogen and generated Cl² should be always the same? No matter if the cell is running 1hr/day or 10hr/day. So it maybe should affect the ORP readings the same way all over the day?


>>Basically at the higher output, the bubbles may coalesce more and outgas more. With lower output, they remain smaller and distinct longer and therefore can dissolve better.
hm... this is perhaps a possibility. But to get back to "my" chemist: He did a calculation for that last year (I have no numbers) but he said that it would take such an amount of free chlorine to oxidize the hydrogen that there wouldn't be any FC left in the water when hydrogen would really dissolve.


>>Do you notice any correlation with flow rate through the SWG?
No, not really. Some of the people have a Speck Eco Touch installed (similar to the Pentair intelliFlow) and there is no visible (or assignable) difference if the flow rate is at about 4m³/h (1 gallon) or at 12..15m³/h (3-4 gallons/hr)


>> Finally if there were more of a problem in pools with still water compared to those with aeration features (waterfalls, spillovers, fountains, etc.) then that would also be consistent with some sort of gas transfer.
Unfortunately, most of the pools in Germany don't have any of these gadgets ... there is only one guy (with this "ORP problem" ) who has a hydro-flow system installed and if he starts it, ORP level is in deed rising (reproducible). But there is only this single pool, so I can not say whether this is a general behaviour or something like an artifact.


>> Adding hypochlorite raises the pH so you should first see if just raising the pH also resolves the problem.
No, rising the pH only, lowers the ORP level (as expected)... if you already have ORP levels at about 100mV you don't really see any more changes in ORP, but if the ORP is at a more normal level you can see it falling if you only rise pH with sodium hydrogen carbonate (NaHCO3).


>>What is interesting is that if you add some hypochlorite, the problem goes away.
yes, interesting ...this is a typical example of what we see here in this case:

http://www.lembi.de/pd/shareX_54635204.jpg

Pool: 32m³ of water (8460 gallons), pH 7.0. CYA 0, TA normally between 20 and 40 ppm, Cell production rate: 11g/hr (a small one)
The green bar above the graph shows the working cell. SWG was running nearly all over the night and you can't see a real change in ORP level. He had a DPD reading of about 5.5ppm FC in the morning (way too much), opened up the cover at 12:00 (sunny day, water at about 25°C ... 77°F). You can see the ORP falling from that point till it completely collapses (60..100mV). It starts rising again in the afternoon, when UV intensity is weaker, but if you do "nothing" in most of the times you are not able to raise it back to 600..700mV only using the SWG. DPD reading at 8:00pm was 0.8ppm. He added 80 gram CH at about 9 pm which should raise FC (HOCl) level for about 1.5 ppm. Not too much, but enough to get back an ORP level of 830-850 which is absolutely OK/normal for that FC levels and clean water. pH of course also reacts and goes up for a short period, but the dosing control brings it back down to the setpoint within 45..60 minutes.
As I mentioned previously, this is absolutely reproducible (not only in this basin).


>>One obvious question: where did you put your ORP system? I assume you didn't put it after the SWG and instead have it somewhere in the circulation system before the SWG, correct?
Yes, sure . The probes - more precisely the measuring cell - is always installed right after the filter, the SWG and the acid injection are always the last parts in the system, right before the water gets back into the pool. In most cases pH injection is the last part and SWG is before that (depending on how the users installed their system).


>> So what does your ORP vs. FC standard look like
For the pools with liquid dosing we recommend a setpoint anywhere between 770 and 800mV (0 CYA, pH7.0). This works for 90% of the pools quite well with resulting FC values between 0.6 and 1.0 (of course depending on temperature an pH value which is set to 7.0 in most cases). Sure there are some exceptions (in both directions) and some have ~740mV at 1.0ppm and some have 850mV at 1.0ppm but the majority is really between 770 and 800mV and dosage is working really reliable.


>>but isn't so good for determining the FC level even holding other parameters fixed
Yes, of course you are not able to hold a fixed FC level of - let's say - 0.6ppm. But under normal conditions it will result (on a sunny day) at a lowest value of about 0.3/0.4ppm (early afternoon) and a highest value of 0.8 / 0.9ppm in the evening. If you have guests and 20 people used the pool, it surely will result in FC levels of about 1.5 or up to 2.0ppm in the evening. But actually i think that's ok in this moment and fits the demand of more HOCl to bring the water back to a "clean condition" and reduce CC back to a lower level.
Maybe even the better way for residential pools than holding a fixed FC value of 0.4 ... 0.6 ppm (0 CYA) because the system will always raise the HOCl level to a point where (depending on the pollution of the water) a safe disinfection is guaranteed...

...as long as the measurement works . But it does - without any problems - for liquid dosing, but not for 60-70% of the pools using SWG.

 

[lembi]

Re: Total Control - ORP vs. Chlorine PPM - Confused



...well, in fact, I'm not. Our Controller is still a private project (since 4 1/2 years) and is not officially sold
We have a lot of them up and running till yet, but we still see this as a kind of "development phase".

 

[chem geek]

Re: Total Control - ORP vs. Chlorine PPM - Confused

>>and that means they will be generating hydrogen gas bubbles for a lot longer and of course generating chlorine gas a lot longer[/I]
I also thought about that. But i think the proportions of generated hydrogen and generated Cl² should be always the same? No matter if the cell is running 1hr/day or 10hr/day. So it maybe should affect the ORP readings the same way all over the day?

>>Basically at the higher output, the bubbles may coalesce more and outgas more. With lower output, they remain smaller and distinct longer and therefore can dissolve better.
hm... this is perhaps a possibility. But to get back to "my" chemist: He did a calculation for that last year (I have no numbers) but he said that it would take such an amount of free chlorine to oxidize the hydrogen that there wouldn't be any FC left in the water when hydrogen would really dissolve.

>> So what does your ORP vs. FC standard look like
For the pools with liquid dosing we recommend a setpoint anywhere between 770 and 800mV (0 CYA, pH7.0). This works for 90% of the pools quite well with resulting FC values between 0.6 and 1.0 (of course depending on temperature an pH value which is set to 7.0 in most cases). Sure there are some exceptions (in both directions) and some have ~740mV at 1.0ppm and some have 850mV at 1.0ppm but the majority is really between 770 and 800mV and dosage is working really reliable.

As I wrote, the total amount of hydrogen gas and chlorine gas (and then chlorine in water) that is produced is the same for the different cell sizes and different run times (so their output time time is the same) but the smaller cell running longer means there is less time for there to be removal of the hydrogen gas by outgassing. In theory the only difference between these scenarios would be that when the cell is running most of the time it gets closer to a steady-state concentration of some amount of over-saturation where the rate of dissolving equals the rate of outgassing. When run for shorter periods of higher output in larger cells, one would expect more of a swing up and down. Also the higher output cell may, at the same flow rate, produce larger coalesced bubbles and with their larger surface area they will not dissolve as quickly in water. So there are differences in behavior between the smaller cells and the larger cells.

Perhaps you could take a look at the output at a return at night with a light on in the pool to clearly see the hydrogen gas bubbles and see if the smaller cells produce lots of very tiny bubbles while the larger cells produce larger bubbles. That would help confirm or refute the theory of this having anything to do with aqueous molecular hydrogen.

As for what the chemist calculated, he is assuming that ALL or nearly all the hydrogen gas was dissolving, but that's not what I am assuming at all. I'm assuming that only a fraction of the hydrogen gas dissolves into the water and that the over-saturation amount in water is not high enough to noticeably affect the chlorine demand. Remember that air has only 0.0000005 mole fraction of hydrogen gas. So I'm not speculating that there is a lot of hydrogen getting into the water since even a small amount would be much much more than normal from air, but that maybe even a little has a very bad effect on ORP membranes or measurements. I'm not talking about quantities directly affecting ORP since clearly they are too low to do so, but rather something indirect. Maybe it's not even aqueous hydrogen gas but instead the presence of very tiny suspended hydrogen gas bubbles that are the problem as these could linger in the pool and make it to the ORP sensor in the pump room.

What is interesting is that you noted that the problems were triggered when the pool was uncovered and exposed to sunlight. Is that always the case? That is, are there never any problems until there is sunlight exposure on the pool? The UV in sunlight breaks down chlorine to produce chlorine and hydroxyl radicals. With no CYA in the water hypochlorous acid near the surface has a half life of 2 hours 10 minutes while hypochlorite ion has a half life of 20 minutes. At depth the half life is longer since chlorine at the surface absorbs enough photons to have them somewhat depleted at depth. At a pH of 7.5, for example, at the surface the half life is 35 minutes but in a pool with average 4.5 foot depth the half life is 51 minutes.

This post describes some of the reactions that can occur from the radical production. Note that hydrogen peroxide is produced and is then degraded by bicarbonate radical. While the hydroxyl radicals themselves are very short-lived, the lifetime of bicarbonate and carbonate radicals are longer (reaction rates about 200 times slower). So I wonder if the presence of hydrogen gas changes this radical chemistry in a way that ends up throwing off the ORP sensor. Since the sensor is in the pump room, the chemical species has to live long enough to get circulated there.

So why in your example with cell production of 11 grams/hour Cl₂ in 32 cubic meters (32,000 liters) which would be 0.34 ppm FC per hour is there no increase in ORP overnight? Is there oxidation of bather waste? If so one wouldn't expect the ORP to be so stable. Was the DPD measurement of FC at the start of the evening lower and at the end it was 5.5 ppm? That would indicate that in fact the ORP was dropping overnight from what it should have been.

Your variation in mV is consistent with what I've seen with other ORP data and just says you need to measure independently and use a setpoint, but as you point out this can work well for process control, at least when weird things don't happen.

The bottom line is that it appears that the problem is a combination of something the SWG produces that builds up more in the pool when there is a longer SWG runtime and that interacts either directly with sunlight or with something else that sunlight is producing in the water. Whatever it is it is acting as if it is a reducing agent lowering the ORP or otherwise causing the oxidizers (HOCl) to get blocked passing through the membrane into the cell. Perhaps you should buy a small saltwater chlorine generator and do experiments with your ORP sensors in a tank and also have UV light so you can experiment to see what combinations trigger a failure. If you do this to enough of an extreme, you might be able to do some analysis at the membrane or electrodes to try and figure out what is going on.

 

[lembi]

Re: Total Control - ORP vs. Chlorine PPM - Confused

>>Perhaps you could take a look at the output at a return at night with a light on in the pool to clearly see the hydrogen gas bubbles and see if the smaller cells produce lots of very tiny bubbles while the larger cells produce larger bubbles. That would help confirm or refute the theory of this having anything to do with aqueous molecular hydrogen.
I will try to find out. But I have no possibility to look by myself because none of the pools is close to my location.



>>Maybe it's not even aqueous hydrogen gas but instead the presence of very tiny suspended hydrogen gas bubbles that are the problem as these could linger in the pool and make it to the ORP sensor in the pump room.
...
>>has a very bad effect on ORP membranes or measurements.
interesting points... this may would explain why replacing or cleaning the probe brings back nearly "normal" ORP readings for some hours (or in some cases 2..3 days).



>>What is interesting is that you noted that the problems were triggered when the pool was uncovered and exposed to sunlight. Is that always the case?
No it's not directly related to the sunlight. This also can occur late at night. If ORP level drops below 400..500mV it "crashes" in most cases. But it only occurs if the SWG is running and ORP level is falling anyway. If you switch off the SWG and just wait some time till ORP falls below 400mV by itself (no more HOCl in the water) it won't "crash". So there is surely a relation to a (long) runtime of the cell - maybe the bubbles you mentioned.



>>So why in your example with cell production of 11 grams/hour Cl₂ in 32 cubic meters (32,000 liters) which would be 0.34 ppm FC per hour is there no increase in ORP overnight? Is there oxidation of bather waste?
Good question ... Should not be related to bather waste, bacteria, algae or anything like that. The pool was unused the days before, as the owner told me. And always had suitable FC values. So this is, in my opinion, just related to the problem with the measurement which I can see on several pools where the ORP is "unusable" to control the SWG (similar behaviour as on the screenshot in my last post). Sometimes ORP level won't rise at all and the longer the cell runs, the lower it seems to get.



>> Perhaps you should buy a small saltwater chlorine generator and do experiments with your ORP sensors in a tank and also have UV light so you can experiment to see what combinations trigger a failure.
I've already thought about something like that. But, even if I am able to reproduce this behaviour, I have no idea how to find out what really causes the low ORP readings or how to solve it

 

[chem geek]

Well you could get some hydrogen gas and something to inject it into the water in a way to make very tiny bubbles and see if that by itself causes the same problems with an ORP sensor. Or you could do electrolysis with low chloride (use sulfate instead) and see if the problem still occurs since that would have hydrogen and oxygen so would take chlorine out of the equation at least from a generation point of view (you could still have chlorine in the water to achieve a given ORP level).

Given that cleaning the probe has it work for a while, my best guess is that it's related to suspended very small hydrogen gas bubbles that get to the probe membrane and interfere with it reading chlorine, perhaps by reducing it there or causing other problems with the membrane and or electrodes.

 

[lembi]

>>(use sulfate instead) and see if the problem still occurs since that would have hydrogen and oxygen so would take chlorine
So if I use sulfate instead of NaCl, Oxygen will be the generated disinfectant? Similar like using hydrogen peroxide?

If hydrogen peroxide is used as disinfectant you have pretty low ORP levels of about 250-300mV - and the ORP levels seems to be something like "buffered" at this level. Unlike the chlorine, you can't see the ORP rising if you increase the level of hydrogen peroxide in the water.
And maybe in some pools the level of sulfat is pretty high, if sulfuric acid is used to adjust the pH? As far as I know sulfat is one of the residual products of that?

Could this be a part of the problem?

 

[chem geek]

I meant in a separate tub test, not in a real pool. Oxygen won't be a sufficient disinfectant compared to chlorine. I meant just as an experiment, not in a real pool.

As far as sulfates being a problem, I don't know of it affecting ORP.

 

[lembi]

I understood that you meant this for testing purpose only But you said "use sulfate instead"... because SWG will produce hydrogen and oxygen then.

I don't know how this generated oxygen will react in the water. I only know: if hydrogen peroxide is used as disinfectant the ORP level is pretty low at about 250..300mV. So my thinking was: is the generated oxygen possibly behaving like the hydrogen peroxide and is maybe lowering the ORP value.

The sulfates itself should not have an effect on the ORP level... but if there is salt and a higher sulfat concentration in the water... the cell will generate oxygen and chlorine?

 

[chem geek]

No, if you have chloride in the water than chlorine is generated even if sulfate is present. What I was describing was replacing chloride with sulfate to remove chlorine from the equation, but you are right that the resulting ORP will be low in that case, BUT you could always add chlorine manually. That would then distinguish the problem being something related to chlorine generation in the SWG vs. the hydrogen gas generation. That's really what I was getting at.

 

[chem geek]

See Probe Failing; Current in the Water; Bonding Issue? where another user is having trouble with getting their ORP to work, but even worse than yours since theirs never works -- it almost immediately goes bad.

Have you figured anything out since you last wrote? Please keep us posted since we'd like to get to the bottom of this.

 

[lembi]

>>Have you figured anything out since you last wrote?
No unfortunately not really. We just testet some things with the one who has this hydro-flow system. ~3 weeks ago he programmed the hydro-flow to run every 6 hours for 10 minutes... Since then he had and still has more "normal" ORP readings at more "normal" FC levels. 750mV to 770mV with 1.0 - 1.5 ppm FC (no CYA), ph7.0, water temperature 32°C/90°F, 9.000 gallons pool, used by 3-5 persons daily).

Comparatively these values are pretty OK, because previously he only had something between 620 to 700mV with FC levels of 3-4ppm - same usage of the pool, same temperature and longer SWG runtime because of the low ORP reading. We just finished the prototype for a potentiostatic chlorine measurement and he will mount it this weekend. So we will have a better overview of the real chlorine level at any time. Perhaps this will help a little.

I don't really know right now why ORP levels are getting better with the hydro-flow. Either the higher water movement is enough or the injected air maybe "collects" hydrogen bubbels and brings them out of the water (if hydrogen bubbles are a problem)? ... just a guess. Unfortunately he can't completely close the air injection right now, but it makes no visible difference if the air-injection brings in a little more or less air, it work's the same way if it's adjusted to min. or max.



Alex

 

[chem geek]

Increased aeration will outgas anything that is over-saturated in the water so that includes both carbon dioxide as well as hydrogen gas. Basically, the hydrogen gas bubbles from the SWCG are pure hydrogen gas so they will slowly transfer (dissolve) into the water at full saturation level. Most of the bubbles escape to the air but if there were no outgassing at all of hydrogen from the bulk pool water then it would build up to full saturation which with Henry's Law constant of 7.8x10^-4 M/atm. The normal amount of hydrogen gas in the air is only 0.000053% so if water were to be fully saturated with hydrogen gas it would be about 1.9 million times more concentrated than normal in equilibrium with air.

Of course, there is actual outgassing between water and air, but this is fairly slow. It is sped up with aeration. Basically the Hydro-Flow injects air into the water and those air bubbles pull out the hydrogen from the water because the concentration of hydrogen in the water is higher than in those bubbles. This transfer is faster with more surface area so smaller bubbles will transfer faster. This is why the hydrogen gas is able to get into the pool in the first place because those hydrogen gas bubbles from the saltwater chlorine generator are rather small. Also, the disruption of the water's surface breaking its surface tension and increasing that surface area and mixing of the water's surface (shrinking the transport layer) also helps increase the rate of outgassing.

Now whether the Hydro-Flow is physically moving actual hydrogen gas bubbles as bubbles away from getting to the ORP as bubbles or whether it's removing any buildup of aqueous hydrogen that may more directly affect the ORP system, I do not know though I suspect it's the latter. The chemist you spoke with said that chlorine would oxidize aqueous hydrogen and while that is certainly thermodynamically favorable, I do not know the kinetics of that reaction. I would think it would be relatively fast, but one never knows. It would be interesting to have a tub of chlorinated water with hydrogen gas injected into it to see how quickly they react. Since the chlorine output from the generators results in the predicted FC rise, either the hydrogen gas bubbles are not getting into aqueous solution fast enough or they are and chlorine doesn't react very fast with it.