Automated Testing

[Katodude]

Correct. You get me something that will read my CYA at 70 ppm, CH at 800 ppm, and FC at 30 ppm (when a SLAM is necessary), then back up the truck.

Thats probably going to be a while. This happens one step at a time. Right now there are a bunch of here that have the Atlas Scientific Industrial pH probe installed in their systems. It works great. I test mine every few months and calibrate as necessary. It is very accurate. And now I dont have to tell the difference between shades of red. Turns out my pH is much more stable than I thought. Its not cheap, but its not expensive either at $200.

The next step is a true FC probe. Those used to be about $1000, but we are starting to see models come out in the $300 range. Non have been tested so far, but it is only a matter of time. Again not cheap, but not crazy expensive.

Now these are not drop in your pool, battery powered self contained fully automated systems. They are probes that need to be installed in your plumbing (usually through a bypass), and connected to a control system. That will come eventually. But right now this is relegated to hobbiests that dont mind spending the time and money to get a level of control they dont need, but just want.

I also still wonder why someone has not come up with a phone app that will look at a pH comparator block and look down the CYA tube and tell you the result. I would think the cameras on phones have enough of whatever it takes to get the right info, and some basic AI to read the results. Isnt that what all these other devices do?

 

[Orion7319]

I also still wonder why someone has not come up with a phone app that will look at a pH comparator block and look down the CYA tube and tell you the result. I would think the cameras on phones have enough of whatever it takes to get the right info, and some basic AI to read the results. Isnt that what all these other devices do?

I’ve just seen apps like that for certain test strips. Never tried any of them myself.

 

[sean.a.hyde]

I went down the Atlas Scientific ORP, pH and temp probe path a couple years ago. Spent $$ for a system that essentially ended up being a clock (periodic drift during the day/night). And that's for readings in the "shade" under my opaque Autocover, using isolated supplies and isolated I2C.

My next project on the back burner would be to essentially create automated titration. A valve that siphons off water and then adds the powder, and then adds the reagent a drop at a time and watches for a color change.
I need to figure out the best way of dispensing the powder (while it is stored in a sealed way so it doesn't oxidize/spoil), and then counting titration drops.
The non-chlorine tests are easier as long as the reagents can survive being outside in the shade.
Probably, with a controlled light source, I can also do CYA.
I don't think there is a titration test for pH (?) so I'd either use an off-the-shelf probe, or hope that the controlled light source didn't drift too much.
I can handle the camera and valves.
All of that is doable, but not commercially viable... just a project I would play with. This year is busy. Maybe next year.

 

[Orion7319]

I went down the Atlas Scientific ORP, pH and temp probe path a couple years ago. Spent $$ for a system that essentially ended up being a clock (periodic drift during the day/night). And that's for readings in the "shade" under my opaque Autocover, using isolated supplies and isolated I2C.

My next project on the back burner would be to essentially create automated titration. A valve that siphons off water and then adds the powder, and then adds the reagent a drop at a time and watches for a color change.
I need to figure out the best way of dispensing the powder (while it is stored in a sealed way so it doesn't oxidize/spoil), and then counting titration drops.
The non-chlorine tests are easier as long as the reagents can survive being outside in the shade.
Probably, with a controlled light source, I can also do CYA.
I don't think there is a titration test for pH (?) so I'd either use an off-the-shelf probe, or hope that the controlled light source didn't drift too much.
I can handle the camera and valves.
All of that is doable, but not commercially viable... just a project I would play with. This year is busy. Maybe next year.

FYI. The DPD powder for the Chlorine tests seems to be very sensitive to hot and cold and easily goes bad if not kept at room temps. Before I realized this I was storing my test kit in a non climate controlled shed.

 

[jliddil]

As an ex-bench chemist I like the drop test. Got a brochure for the waterguru and $75 off. While technology is cool I'm not an early adopter. I did get w Warrior SI pool robot and it rocks. Automated testing will come along but I think it is still iffy. Keep in mind those on TFP vs the majority of pool owners who don't give a care about pool chemistry and let the pool person come once a week. I've have friend who can't be bothered with testing themselves. The go to the local store to get "testing". They are hit or miss. I took sample in to get tested. Distilled water. Had the tester freaking out until I had to admit the joke.

 

[BlackScimitar]

My mistake. Sutro uses optical detection.

pHin and Pentair use ORP. WaterGuru uses DPD pads. They all use sensors to “read” the color changes or voltage changes and they all, so far, do a terrible job of it. We’ve had several users on TFP return their Water Guru’s because the FC was several ppm off. I doubt Sutro does any better. Garbage data in, garbage results out. Increasing the amount of garbage data doesn’t make the results more accurate. These sensors will never be able to deliver on their promises at the price-point consumers will expect. Manufacturing corners will have to be cut and that will drive failure and inaccuracy.

And let’s be honest, they don’t make money by selling you the WiFi testing toy. They make money selling you the chemical service and overpriced chemical packs. It’s nothing more than the pool store sales philosophy with a “tech” edge to it.

I too have been around the IoT world and worked for a Fortune 50 tech company (one of the original tech behemoths) for over 15 years. Same story over and over again, we will revolutionize the world in the next 5 years … same results - businesses still prefer the same mainframe technology from the 1980’s to run their zero-downtime-required businesses. I personally could care less if my refrigerator can connect to the internet and control my landscape lighting and tell me what the Dow Jones is doing. It’s primary job is to keep my food from spoiling. And I bet in the grand scheme of things, there are way more people out there that care about their pools just being a sanitary bucket of water to swim in on a hot day rather than some tricked out piece of tech that can be controlled from the International Space Station generating reams of output data for some AI to analyze.

But hey, no need to mind me, go buy a Sutro and see how utterly unconnected and underwhelming it all is.

And as for Elon Musk - I’ll not think so poorly of him when he pays me back for the self- driving Tesla that rear ended and destroyed my wife’s Toyota while it was parked in my kids school parking lot …

"businesses still prefer the same mainframe technology from the 1980’s to run their zero-downtime-required businesses."
Well, not really that reason... IBM heavily discounts Mainframe MIPS time and hardware maintenance because they know companies a) cannot get Cobol programmers to port to distributed and b) getting staff and resources to migrate MF apps to a distributed environment is just not worth the effort. It's actually cheaper to continue to pay IBM than to hire staff and commit time to code new distributed stuff. IBM sells it as "mission critical" benefits but the reality is just a numbers game. Reliability on MF or distributed with larger scale horizontal systems is pretty much on par now a days. MF guys will disagree though
Sorry to hear about your wife's car; same thing happened to me but got rear ended by some testosterone fueled 19 year old driving a Diesel truck ... thankfully I was driving a 4Runner ...

 

[joboo7777]

"businesses still prefer the same mainframe technology from the 1980’s to run their zero-downtime-required businesses."
Well, not really that reason... IBM heavily discounts Mainframe MIPS time and hardware maintenance because they know companies a) cannot get Cobol programmers to port to distributed and b) getting staff and resources to migrate MF apps to a distributed environment is just not worth the effort. It's actually cheaper to continue to pay IBM than to hire staff and commit time to code new distributed stuff. IBM sells it as "mission critical" benefits but the reality is just a numbers game. Reliability on MF or distributed with larger scale horizontal systems is pretty much on par now a days. MF guys will disagree though
Sorry to hear about your wife's car; same thing happened to me but got rear ended by some testosterone fueled 19 year old driving a Diesel truck ... thankfully I was driving a 4Runner ...

Still have an AS400 running critical business functions at my place of work. And that puppy is not going anywhere.

 

[JoyfulNoise]

Still have an AS400 running critical business functions at my place of work. And that puppy is not going anywhere.

Is that one of the dung-powered versions or did it get the upgrade kit to coal? Careful, fly-ash is not good for the hard drives … the hamsters in the wheel that runs them get sick from the fumes.

 

[joboo7777]

I decided to try the Sutro device. The setup took me about 10 minutes which includes setting up the Hub, installing the battery and cartridge and placing the device in the pool. The mobile app guided me through the entire process which was pretty well scripted. They do advise you to give it 48hrs for the readings to normalize which in my case only took about 24hr.

the device performs testing of PH, FC, and ALK 3 times a day and temp every hour. For me it’s around 10am, 7pm, and 2am. Not sure how Sutro decides this frequency and You cannot initiate a test yourself.

I have compared the results against my exact iDip, and my Hayward PH probe. I’ve also validated result against Taylor drop test for all you die hard drop testers out there!

Below are the results after a week of use:

PH: has been within .1 of my PH probe. I currently maintain 7.7 and usually the Sutro is 7.6. There are a few readings at 7.5 and some at 7.7 but for the most part it has been consistent.

Tenp: always within 1 degree of my Hayward temp sensor

FC: Sutro readings are consistently 1ppm higher than the actual. I’m usually around 3ppm but the Sutro has me at 4ppm. They claim to be within .2 ppm accuracy but given the results I’m not sure that statement is accurate.

Alk: instead of a single number, Sutro provides a range. 31-50, 51-70 are the ones that have been provided during the 1 week testing period. Taylor has me at 60, exact iDip has me at 55-60. I was hoping for a more precise reading from Sutro or at least a smaller range that is more in line with the standard 10% variance that Taylor provides.

I did reach out to Sutro to discuss the 1ppm difference in free chlorine readings, they responded to my inquiry within 6 hours and attempted to re calibrate the device. The next day the results were still the same so I provided them with an update and they responded within 8 hours informing me they would send me a new cartridge. We’ll see if that helps.

Some folks had issues with the device constantly staying in front of the skimmer. I did not have this issue as the device seems to float around the pool without favoring a particular area.

I definitely had concerns going in to the evaluation given the experience of others from the forum. It’s not perfect but not a bad attempt at automated testing.

Not sure I will be keeping the device given the $30 a month price tag. If they add CH and CYA I think I might jump on board.

 

[IceShadow]

I’m enjoying my WaterGuru. It’s within 1ppm of FC (and reads low if anything which is fine) and within 0.1 pH as well. I still check it every week or so against the Taylor kit, but it warns me when I need to add FC every couple of days and when I need to adjust the pH (just once so far this season, today).

 

[MyAZPool]

I went down the Atlas Scientific ORP, pH and temp probe path a couple years ago. Spent $$ for a system that essentially ended up being a clock (periodic drift during the day/night). And that's for readings in the "shade" under my opaque Autocover, using isolated supplies and isolated I2C.

My next project on the back burner would be to essentially create automated titration. A valve that siphons off water and then adds the powder, and then adds the reagent a drop at a time and watches for a color change.
I need to figure out the best way of dispensing the powder (while it is stored in a sealed way so it doesn't oxidize/spoil), and then counting titration drops.
The non-chlorine tests are easier as long as the reagents can survive being outside in the shade.
Probably, with a controlled light source, I can also do CYA.
I don't think there is a titration test for pH (?) so I'd either use an off-the-shelf probe, or hope that the controlled light source didn't drift too much.
I can handle the camera and valves.
All of that is doable, but not commercially viable... just a project I would play with. This year is busy. Maybe next year.

@sean.a.hyde
I'm not really going to address or comment on the whole debate here regarding the use of technology versus titration methods for pool water testing. I understand both sides of the argument and I only know what is working well for me presently. I use both. Technology for day-to-day monitoring and titration as a backup occasionally to insure that the technology has not led me astray as evidenced by my pool math logs.

I would like to address your comments regarding the Atlas Scientific probes/sensors however and see if I can offer up some points for consideration.

1. As @Katodude pointed out earlier, there are a handful of us here that are using the Atlas Scientific probes/sensors. As far as I'm aware, all with good success. That is not to say that there have not been "speed bumps" along the way, because in my case there were. But after negotiating some pitfalls that were due to my own erroneous assumptions and getting some very helpful assistance from other TFP members, I can honestly say that I am very pleased with the results now.

2. Personally, I am currently using six Atlas Scientific sensor/probes.
ORP
pH/Water Temp
EC
Air Temperature
Filter Pressure Transducer
Humidity/Dewpoint

Do I need all of these for pool chemical management? Heck no. I just like to play with the technology and see what I can make work for me (or not lol).

My ORP readings are merely an indicator for me because of my higher CYA levels (80ppm). But when I look at either my chemical management dashpanel or my Grafana dashboard and if I were to see an unusually low ORP value, it just tells me that I might want to double check my FC level (yes, using the Taylor) and to check my IntelliChlor output level. That is all I use it for really. That said, I agree with @Katodude , that at some point those reliable amperometric free chlorine sensors are going to get better and cheaper and then I'll add one to my sensor manifold. If and when that occurs, maybe it might be possible through chemical management software to tie SWCG output to the FC sensor readings.

3. pH - As far as your statement regarding "periodic drift during the day/night". Have you considered that may only be due to the differences in pool water temperature between day and night? I have a play/lap pool (only 5.5' in the middle), so this time of year, my pool water temperature fluctuates an average of of 10-15 degrees F between night and day.

Since water temperature has a direct impact on pH readings, my pH can fluctuate an entire .1 between night and day. At this time of the year, typically this will cause my pH to drop to an average of 7.7 during the hottest part of the day when my pool water temp is nearly 90-93 degrees and rise to above 7.8 when the pool water temperature averages around 80 degrees in the early morning hours. The pH would rise even further at night but when it reaches 7.8 (my MA dosing setpoint) and is trending up, then the chemical management software that I am fortunate enough to use, slaps it back down (doses muriatic acid), as evidenced by the following graph.


As you can see, my pH decreases as the water temp increases and vice versa. So it may be entirely possible that those periodic drifts between night and day that you speak of, are merely due to water temperature fluctuations and not a problem with your Atlas probe.

Although for a different reason of course, my ORP also fluctuates between night and day as well, as you can see from the graph below. That is also normal..


Which brings me to my last point. I'm not really sure what type of software that you are using that interprets your Atlas Scientific probes/sensors, but my own opinion is that it can play a major role in how well your data is presented etc. Just a thought.

I merely wanted to offer up a different perspective that might show that using the Atlas Scientific sensors and probes provides much more than just a "clock" to some of us. I hope you have better luck with your Atlas Scientific sensors and like I said, there are a handful of us here that are having success with them and I'm pretty sure that any one of us would be happy to assist you to get to the point where you might want to be with them.

Thanks,
r.

 

[sean.a.hyde]

@sean.a.hyde
I'm not really going to address or comment on the whole debate here regarding the use of technology versus titration methods for pool water testing. I understand both sides of the argument and I only know what is working well for me presently. I use both. Technology for day-to-day monitoring and titration as a backup occasionally to insure that the technology has not led me astray as evidenced by my pool math logs.

I would like to address your comments regarding the Atlas Scientific probes/sensors however and see if I can offer up some points for consideration.

1. As @Katodude pointed out earlier, there are a handful of us here that are using the Atlas Scientific probes/sensors. As far as I'm aware, all with good success. That is not to say that there have not been "speed bumps" along the way, because in my case there were. But after negotiating some pitfalls that were due to my own erroneous assumptions and getting some very helpful assistance from other TFP members, I can honestly say that I am very pleased with the results now.

2. Personally, I am currently using six Atlas Scientific sensor/probes.
ORP
pH/Water Temp
EC
Air Temperature
Filter Pressure Transducer
Humidity/Dewpoint

Do I need all of these for pool chemical management? Heck no. I just like to play with the technology and see what I can make work for me (or not lol).

My ORP readings are merely an indicator for me because of my higher CYA levels (80ppm). But when I look at either my chemical management dashpanel or my Grafana dashboard and if I were to see an unusually low ORP value, it just tells me that I might want to double check my FC level (yes, using the Taylor) and to check my IntelliChlor output level. That is all I use it for really. That said, I agree with @Katodude , that at some point those reliable amperometric free chlorine sensors are going to get better and cheaper and then I'll add one to my sensor manifold. If and when that occurs, maybe it might be possible through chemical management software to tie SWCG output to the FC sensor readings.

3. pH - As far as your statement regarding "periodic drift during the day/night". Have you considered that may only be due to the differences in pool water temperature between day and night? I have a play/lap pool (only 5.5' in the middle), so this time of year, my pool water temperature fluctuates an average of of 10-15 degrees F between night and day.

Since water temperature has a direct impact on pH readings, my pH can fluctuate an entire .1 between night and day. At this time of the year, typically this will cause my pH to drop to an average of 7.7 during the hottest part of the day when my pool water temp is nearly 90-93 degrees and rise to above 7.8 when the pool water temperature averages around 80 degrees in the early morning hours. The pH would rise even further at night but when it reaches 7.8 (my MA dosing setpoint) and is trending up, then the chemical management software that I am fortunate enough to use, slaps it back down (doses muriatic acid), as evidenced by the following graph.
View attachment 353681

As you can see, my pH decreases as the water temp increases and vice versa. So it may be entirely possible that those periodic drifts between night and day that you speak of, are merely due to water temperature fluctuations and not a problem with your Atlas probe.

Although for a different reason of course, my ORP also fluctuates between night and day as well, as you can see from the graph below. That is also normal..
View attachment 353682

Which brings me to my last point. I'm not really sure what type of software that you are using that interprets your Atlas Scientific probes/sensors, but my own opinion is that it can play a major role in how well your data is presented etc. Just a thought.

I merely wanted to offer up a different perspective that might show that using the Atlas Scientific sensors and probes provides much more than just a "clock" to some of us. I hope you have better luck with your Atlas Scientific sensors and like I said, there are a handful of us here that are having success with them and I'm pretty sure that any one of us would be happy to assist you to get to the point where you might want to be with them.

Thanks,
r.

Thanks for your insight.

I guess I didn't mean to imply that the ORP sensor did "nothing". But I was looking for something that would tell me if my FC was 6 or 9, and I struggled with getting that level of output.

I'm happy to try using different software to read the probes, but I was just using the I2C interface to the isolated A->D interface that Atlas sells. I also had a temp probe and Atlas's board can use that for temp compensation of the pH. I don't remember how much it fluctuated throughout the day, but it was more than 0.1.

I'm happy to try it again when I get more time. I was all set to embed the sensors in the return line (rather than hanging out in the pool) to hopefully get more consistent results. Maybe I'll try that next year.

 

[GrandizerGo]

I definitely think there would be interest in automation products. I would settle for a dosing system for chlorine and MA that includes a flow sensor for those of us that dont have controllers. As for testing, that would be awesome if it worked or got close enough. I think though we may not be the correct market for this unless it is accurate. For example, I know about 4-5 other people who own pools and none of them own test kits. Something that is advertised as better then test strips and more convenient then pool store visits may be appealing to some. For me, it would be nice after I did testing or had a sunny/swimmer load day that I could click a button on my phone to add 1ppm of liquid chlorine.

 

[rstrouse]

You do know that what @MyAZPool is showing you does not require a controller. It does however, require a raspberry pi.

 

[MyAZPool]

@GrandizerGo
Not this kind...


But this kind.


+1 on what @rstrouse says. Controllers are now optional. I got a big ole enclosure out there full of wires, circuit boards, etc that I really don't even need now thanks to the smart guys here.

Oh well....

r.

 

[mknauss]

Not this kind...

Hmmm -- that looks pretty good ----

 

[GrandizerGo]

Thanks. I'll have to look more closely at what can be done. I have a bunch of Raspberry Pis laying around for Retro Pi, PiHole, etc. The electronics of it all is a little daunting to be honest.

 

[MyAZPool]

Thanks. I'll have to look more closely at what can be done. I have a bunch of Raspberry Pis laying around for Retro Pi, PiHole, etc. The electronics of it all is a little daunting to be honest.

Hey @GrandizerGo
When you have time, why don't you pay a visit to the nodejs-poolController lobby and have a look around.

Element

gitter.im

Several folks hanging out there there that have "homebrew" systems and that do not have the typical pool automation controllers. Lot's of really cool stuff being explored there.

Don't let the electronics intimidate you. If I was able to get through it (with lots of great help from other members here), anyone can.
So easy, even a "MyAZPool" can do it.

 

[phonedave]

I have looked at the automated testing stuff, and my biggest dislike about it is that it seems like a printer and ink model to me. The reagent based automated systems need you to buy their reagents in their packaging at their price. That is not a model that appeals to me.

My second biggest dislike is that it seems like a solution in search of a problem. I like connected devices and automation, but for me manual testing is not a huge effort. If the weather was really bad, I guess it would be beneficial not to have to go out to the pool, but then it is only taking a sample from the top layer of the pool, so if it is raining the results may be off anyway.

Sometimes automation is not worth the effort. I went to cook a few weeks back and I had to tell my wife we had to wait a bit because my grill was downloading new software and was unavailable.

 

[jkayca]

This thread was an interesting read. Obviously what we need is a Star Trek like tricorder but it seems that is a little far off.