Least expensive means of controlling Intellichlor?

[Dirk]

I actually made a couple corrections in my earlier post, because I had written it wrong. But I still think you can do what you need without adding anything, electrically or physically.

You should be bracketing around your SWG runtime, in case the timer gets out of sync with the pump. I'll use an example to explain.

Let's say your SWG's minimum flow requirement needs 1500 RPM to get a green flow LED.

And let's say you need 7 hours and 15 minutes of SWG runtime to get the perfect amount of chlorine each day.

- Program your Intermatic timer to run the SWG for 8 hours a day, from 9:00AM to 5:00PM.
- Program your IntelliFlo to schedule a run for 10 hours a day at 1200 RPM, from 8:00AM to 6:00PM ("bracketing" around the Intermatic runtime).
- Create a second IntelliFlo schedule for 7.25 hours a day at 1600 RPM, from 9:30AM to 4:45PM.

Your pump will run for 10 hours.
Your SWG will power on for 8 hours, within the pump's runtime.
Your pump will ramp up to 1600 RPM for 7.25 hours a day, which will trigger the SWG to produce chlorine for 7.25 hours, within the Intermatic's runtime.

You can fine tune the chlorine production by adjusting up or down that second IntelliFlo schedule, in five minute increments more or less. Even after you get the perfect runtime figured out, you'll then need to adjust for the seasons: more runtime for the middle of summer, less for spring and fall, etc. But you'd do all that by adjusting just that second IntelliFlo schedule.

The Intermatic timer keeps the SWG from firing outside of the IntelliFlo runtime. You've got an hour of pump runtime on either side of the Intermatic runtime, in case the Intermatic gets out of time sync with the IntelliFlo. And the SWG chlorine production (the 1600 RPM run, which triggers the SWG's flow requirement) runs within both the pump runtime and the Intermatic runtime.

So a power failure will trash the Intermatic runtime, but that's a problem you've currently got, so the scheme above doesn't make that any better or worse. The correct way to address that is to install a secondary flow switch, or a power-sensing relay, which are not hard to do. Or you can just run your IntelliFlo 24-7 at a very low RPM, so that no matter when the out-of-sync Intermatic decides to run your SWG, you'll always have water flowing through it.

But either way, the scheme above will work with your current equipment, at least until you decide to go with some sort of home automation solution, or a more advanced pool automation controller. Now if you want remote control, that's a whole 'nother can'o'worms.

 

[JoyfulNoise]

I actually made a couple corrections in my earlier post, because I had written it wrong. But I still think you can do what you need without adding anything, electrically or physically.

You should be bracketing around your SWG runtime, in case the timer gets out of sync with the pump. I'll use an example to explain.

Let's say your SWG's minimum flow requirement needs 1500 RPM to get a green flow LED.

And let's say you need 7 hours and 15 minutes of SWG runtime to get the perfect amount of chlorine each day.

- Program your Intermatic timer to run the SWG for 8 hours a day, from 9:00AM to 5:00PM.
- Program your IntelliFlo to schedule a run for 10 hours a day at 1200 RPM, from 8:00AM to 6:00PM ("bracketing" around the Intermatic runtime).
- Create a second IntelliFlo schedule for 7.25 hours a day at 1600 RPM, from 9:30AM to 4:45PM.

Your pump will run for 10 hours.
Your SWG will power on for 8 hours, within the pump's runtime.
Your pump will ramp up to 1600 RPM for 7.25 hours a day, which will trigger the SWG to produce chlorine for 7.25 hours, within the Intermatic's runtime.

You can fine tune the chlorine production by adjusting up or down that second IntelliFlo schedule, in five minute increments more or less. Even after you get the perfect runtime figured out, you'll then need to adjust for the seasons: more runtime for the middle of summer, less for spring and fall, etc. But you'd do all that by adjusting just that second IntelliFlo schedule.

The Intermatic timer keeps the SWG from firing outside of the IntelliFlo runtime. You've got an hour of pump runtime on either side of the Intermatic runtime, in case the Intermatic gets out of time sync with the IntelliFlo. And the SWG chlorine production (the 1600 RPM run, which triggers the SWG's flow requirement) runs within both the pump runtime and the Intermatic runtime.

So a power failure will trash the Intermatic runtime, but that's a problem you've currently got, so the scheme above doesn't make that any better or worse. The correct way to address that is to install a secondary flow switch, or a power-sensing relay, which are not hard to do. Or you can just run your IntelliFlo 24-7 at a very low RPM, so that no matter when the out-of-sync Intermatic decides to run your SWG, you'll always have water flowing through it.

But either way, the scheme above will work with your current equipment, at least until you decide to go with some sort of home automation solution, or a more advanced pool automation controller. Now if you want remote control, that's a whole 'nother can'o'worms.

I have always found that adjusting pump run time for chlorine production always has the bigger and more measurable effect. I almost always leave my SWG running at 40-50% and then just add or subtract 30mins of pump run time to change chlorine levels. I suppose adding more pump run time is probably more costly than bumping up the SWG output (pumps use more energy) but it’s a lot easier in my opinion.

Of course, I have an EasyTouch system so either methodology works.

 

[sbcpool]

My flow switch is triggered even at 1000 RPM, so I don't have to go through that exercise of having different speeds. My only concern with long pump run times is exorbitant electricity prices in the afternoon, which can run $0.65/kWh. Small changes add up. Even an extra 30 minutes of pump run time could represent an extra $10 a month. Ideally the pump is as closely coordinated with the SWG as possible.

I think a relay triggered by the pump running is my best economical bet. Now to find a relay that can handle up to 10A but be triggered by as little as 300 mA.

 

[Dirk]

Now to find a relay that can handle up to 10A but be triggered by as little as 300 mA.

It's been written about here, but I'm not sure what to suggest for a search. @ajw22, do you recall a thread about a pump current sensor?

Edit: maybe I do. I just searched for "pump current sensor" and a bunch of threads popped up...

 

[ajw22]

It's been written about here, but I'm not sure what to suggest for a search. @ajw22, do you recall a thread about a pump current sensor?

Edit: maybe I do. I just searched for "pump current sensor" and a bunch of threads popped up...

Here are a bunch of threads to review...

Google search results

www.troublefreepool.com

Current Sensing Relay for SWG with Variable Speed pump

First, thanks to all of the posters who make this such a valuable forum! I've only recently joined, but hope to add value over time! There are several threads regarding using a current-sensing relay to avoid the problem of a SWG running when the pump is NOT running. Recently, the flow sensor...

www.troublefreepool.com

Current Sensing Relay for Aquarite SWG with a Tristar VS Pump

I began to have problems maintaining steady chlorine levels with my one-year old pool. After doing a bit of investigation, I found that every morning, the salt level on the Aquarite's display panel was reading 800 ppm - too low for the SWG to work. Some hours later, the average would get high...

www.troublefreepool.com

 

[sbcpool]

Here are a bunch of threads to review...

Thank you for the references. It occurs to me that I won't be happy with a relay since it will activate the SWG any time the pump is running rather than sticking to a fixed schedule. I really need some timer that controls both pump and SWG so they can never get out of sync or have issues after a power outage.

It seems like the DIY devices are made to work with an existing controller like IntelliTouch, although I'm not sure what the point of that is. It's not clear to me if they will work completely on their own connected to nothing but the pump and SWG directly.

 

[ajw22]

Thank you for the references. It occurs to me that I won't be happy with a relay since it will activate the SWG any time the pump is running rather than sticking to a fixed schedule.

Why would you not want to generate some chlorine anytime your pump is running?

I really need some timer that controls both pump and SWG so they can never get out of sync or have issues after a power outage.

That is what a current sensing relay will do.

The pump control panel has the clock timer and will turn on both the pump and through the relay the SWG.

It seems like the DIY devices are made to work with an existing controller like IntelliTouch, although I'm not sure what the point of that is. It's not clear to me if they will work completely on their own connected to nothing but the pump and SWG directly.

I don't understand what you are musing about.

An IntelliTouch system can control a pump and SWG without any DIY devices.

Can you be more specific?

 

[sbcpool]

Can you be more specific?

I don't want to generate chlorine every time the pump runs because:

1. Spa mode
2. My chlorine demand is stable and addressed by a fixed schedule. I often run the pump at other times while cleaning, for extra skimming on windy days, swim parties, etc.

As for the DIY controllers: I am not willing to pay the exorbitant prices of commercial automation systems, so I am looking into the DIY systems posted here. They appear to be designed to work with commercial automation instead of in place of one of those systems.

 

[ajw22]

As for the DIY controllers: I am not willing to pay the exorbitant prices of commercial automation systems, so I am looking into the DIY systems posted here. They appear to be designed to work with commercial automation instead of in place of one of those systems.

I think the node-js controller will run and control equipment without any commercial automation system.

@rstrouse can confirm.

 

[rstrouse]

Yes nodejs-poolController will perform all those functions without any Pentair controller.

 

[sbcpool]

Yes nodejs-poolController will perform all those functions without any Pentair controller.

OK. I think I was just confused after reading the excellent write-up about using it with existing Pentair automation. I'm not clear about connecting it directly to individual pieces of equipment. Do you need a USB-RS485 converter for each piece of equipment or is it more flexible in that all the equipment can be on the same line and gets addressed separately?

 

[Dirk]

I won't be happy with a relay since it will activate the SWG any time the pump is running

The current sensing relay would be wired in series with whatever timer you are, or will be using. Both would have to be "in the green" for the SWG to get power. The current sensing relay would not be used to schedule the SWG, it would only be used to keep the SWG from running when the pump wasn't. You would use a secondary flow switch in the same way.

 

[jmastron]

Thank you for the references. It occurs to me that I won't be happy with a relay since it will activate the SWG any time the pump is running rather than sticking to a fixed schedule. I really need some timer that controls both pump and SWG so they can never get out of sync or have issues after a power outage.

It seems like the DIY devices are made to work with an existing controller like IntelliTouch, although I'm not sure what the point of that is. It's not clear to me if they will work completely on their own connected to nothing but the pump and SWG directly.

Note that *many* people (me included) use separate timers as the primary coordination (in my case the IntelliFlo onboard schedule and a WiFi plug for the SWCG), with the flow switch as the backup in case things get out of sync. In my case, the WiFi timer always has the right time after a power outage, while the IntelliFlo is supposed to have an internal clock battery but it has never kept time so gets out of sync. We don't have many power interruptions here, so I just go reset it when that happens.

A current sensing relay in series with a timer is a nice third safety, but IMO is not really necessary. Some SWCG manuals (at least one of the CircuPool lines) even allow using the flow switch as primary with a variable speed pump (i.e. say just to connect the SWCG to always on) -- I wouldn't do that, however.

 

[sbcpool]

Note that *many* people (me included) use separate timers as the primary coordination (in my case the IntelliFlo onboard schedule and a WiFi plug for the SWCG), with the flow switch as the backup in case things get out of sync. In my case, the WiFi timer always has the right time after a power outage, while the IntelliFlo is supposed to have an internal clock battery but it has never kept time so gets out of sync.

I just don't like how badly the clock in the IntelliFlo drifts. It's surprisingly bad even when compared with the mechanical Intermatic. Within a couple of weeks it's off by more than a minute. That's 50-cent Alibaba clock level of accuracy. I have $20 digital watches that drift by less than 10 seconds a month.

 

[jmastron]

I just don't like how badly the clock in the IntelliFlo drifts. It's surprisingly bad even when compared with the mechanical Intermatic. Within a couple of weeks it's off by more than a minute. That's 50-cent Alibaba clock level of accuracy. I have $20 digital watches that drift by less than 10 seconds a month.

Agree with that 100%. My $20 Timex watch was last adjusted in March, and is 11 seconds ahead right now. I don't know why other clocks can't be that accurate. The $200 clock radio in our bathroom gains at least a minute a month (and is a pain to adjust), and my IntelliFlo is right now 5 minutes ahead and I'm pretty sure I adjusted it within the last couple months.

I just make use a 15 minute buffer after the pump starts/before it stops to run the SWCG.

 

[randytsuch]

I just skimmed the posts from this thread, so apologize in advance if this was covered.

You can click on njsPC for SWG in my signature to see how I setup and connected a pi3 to my Hayward SWG. Its my understanding that Hayward makes the Intellichlor too, so 485 interface is the same, or at least really similar.

Its not set and forget, you have to leave the pi connected to the SWG at all times. If you disconnect the pi, the SWG will revert back to the manual (knob) setting. My SWG lights a remote LED when the 485 is connected.

It was really not hard if you're ok playing with Pi's. Probably even easier with a USB to 485 interface, but I had already bought the one I used, so I just went with it.

On using time instead of percent. I was going to do that. But I figured out it gets a little tricky. The cell wants to run 1/2 the time at positive polarity, and 1/2 the time at negative. If you stop it in the middle of a cycle, then you're going to mess up that balance. You also need to be careful when you stop it, unless you run it at 100%. At any other percent, it runs on a 3 hour cycle, So at 50%, its on for 1.5 hrs and off for 1.5 hrs.
After I learned about the duty cycle and cell balancing, I decided to add a pi using njsPC to control the SWG, it works very well.

As a bonus, you can get status without having to go look at the unit.

Randy

 

[Dirk]

On using time instead of percent. I was going to do that. But I figured out it gets a little tricky. The cell wants to run 1/2 the time at positive polarity, and 1/2 the time at negative. If you stop it in the middle of a cycle, then you're going to mess up that balance. You also need to be careful when you stop it, unless you run it at 100%. At any other percent, it runs on a 3 hour cycle, So at 50%, its on for 1.5 hrs and off for 1.5 hrs.

That might be confusing, if not inaccurate. So just to clarify:

When the cell is new, it switches blade polarity every two hours. After about a month it does so every three hours. I suppose by not carefully observing that "turn around" you would get more positive charge on one set of blades than the other, but I'm not sure how that is going to "mess up that balance," or why it would matter. I don't think that will affect its cleaning process. And certainly checking the blades periodically, which is recommended maintenance anyway, would reveal if your particular schedule is causing any unexpected calcium build up. Perhaps @JoyfulNoise can weigh in on that concern. If you're running your IC uninterrupted for at least six hours a day, this is a non-issue.

The manual states nothing about varying that 3-hour schedule based on output percent. Randy, where did you get that? Is that something you observed via your interface? That contradicts the manual, which warns that excessive self-cleaning is bad for the blades. Which is why they go from two hours to three hours after a month. Why would Pentair program the IC to go less than three hours? At 2% output it'd be switch polarity every few minutes? What am I missing?

And just to be super clear, I think the cycles Randy is describing have to do with the IC's self-cleaning process (blade polarity reversing process), and nothing to do with the chlorine output. The IC's chlorine production cycle is every 5 minutes. That's why I mentioned that in a previous post. If you're going to vary your chlorine production by adjusting IC runtime, you'd do so in at least 5 minute increments, or it wouldn't change your FC much if any.

Right out of the manual:

 

[randytsuch]

So I'm probably wrong, won't be the first or last time.

I'm describing how I understand my Hayward works, and assumed (I know) that the Intellichlor used a similar algorithm. Hayward uses a 3 hour cycle from reading the FAQ page here. I thought it was long, and there is no easy way to measure it since the generating light is on regardless. Then only way you know is by pushing the buttons to see voltage and current, and least on the Hayward.

I'm also assuming there is a reason they designed it to run it 1/2 the time at positive, and 1/2 the time at negative polarity, and that I should try to do the same, to maintain the "balance". It adds complexity to the SWG design, so I figured they had a reason to do it. Based on this, I would not use a control system that ignores it.

OK, I'll stop commenting on intellichlor's and how they work, sorry.

 

[ajw22]

Its my understanding that Hayward makes the Intellichlor too,

Hayward and Pentair SWG's are very different designs and have little in common except for the basic SWG technology that all the SWGs are based off of.

There are many white label clones of the Hayward Aquarite system but the Pentair IntelliChlor is not one of them.

 

[Dirk]

So I'm probably wrong, won't be the first or last time.

I'm describing how I understand my Hayward works, and assumed (I know) that the Intellichlor used a similar algorithm. Hayward uses a 3 hour cycle from reading the FAQ page here. I thought it was long, and there is no easy way to measure it since the generating light is on regardless. Then only way you know is by pushing the buttons to see voltage and current, and least on the Hayward.

I'm also assuming there is a reason they designed it to run it 1/2 the time at positive, and 1/2 the time at negative polarity, and that I should try to do the same, to maintain the "balance". It adds complexity to the SWG design, so I figured they had a reason to do it. Based on this, I would not use a control system that ignores it.

OK, I'll stop commenting on intellichlor's and how they work, sorry.

Ah, that explains the discrepancies. Yah, I have no idea how a Hayward does it's thing, I only know IntelliChlor, and most of what I explained is from the manual, so that could still be wrong. @JoyfulNoise will likely know if the calcium buildup that might happen during the positive polarity cycle of a set of SWG blades needs the same amount of time reversed to undo the build up. I'm curious about that now.

My guess is, especially if you're keeping your CSI slightly negative as recommended by TFP for SWG pools, that the self-cleaning reversal does most of it's thing right at the beginning of that opposing polarity cycle, and the rest of the cycle doesn't do much of anything. I'm basing that on the evidence I've seen in my pool: big 1/4" to 1/2" flakes of calcium lying on my pool's floor. If the self-cleaning process took three hours, I'd expect the calcium would be coming off in molecule-size pieces that would suspend in the water, not big flakes all at once lying on the floor.