Call for collaboration

[Jpadie]

Hi all

I'm about to have to quarantine in the UK for 14 days. So I thought that was a good opportunity to advance a project I've had in mind for a while. I'm calling for collaborators if anyone is interested.

The project is a hobby level pool management system. The features would be

pH measurement
pH dosing
ORP measurement
Liquid chlorine dosing and
SWCG control
SWCG (the whole thing other than the cell).

Outside of the cell I'm targeting a price point roughly as follows

Peristaltic pumps *3: £15ea
ORP sensor: £25
pH sensor: £15
Depth sensors for the acid/base/chlorine tanks: £5ea
Components and board fab: £25
Toroidal transformer 230vA. No idea of price.

Probably also need a flow switch and maybe an optional pressure transducer. Because why not.

So roughly £120 + transformer + cell

These are one-off prices. I'm quite sure the price point could be pushed way down in some quantity.

I've got replacement plates for my 1440cm2 cell which are costing £175 delivered incl tax. Against the manufacturer replacement cost of £1100

Additionally there would be

* Design files for housing
* Web/mobile interface design and coding
* Firmware coding
* Sample Alexa skill to query and control the interface vocally.

My intent is to make this entirely open source. So providing all files in usable and modifiable formats.

I've got the time available to get a long way into the design of all this in early September, if not finish it. but I'd love collaborators and people to help answer questions about the electrolysis chemistry, UX and UI design etc. And if anyone has microelectronics expertise, to collaborate on the design of the circuitry.

I can also add control of standard vfd devices too (either by modbus or the usual four switch method), if that's thought useful.

I'm not trying to disrupt the market at all. But this might be something that the tinkerers amongst us might enjoy building. And that way you could make it yours.

When I say hobby grade, I suspect that it will function and be as safe as the stuff I've currently got. I've taken that to bits and it's a very basic design. The fundamentals are all ok but these days we can be much more efficient at.much lower energy costs.

Any interest let me know. Looking forward to chatting.

Justin

 

[cowboycasey]

Here are some like minded people that might get into something like this

Does anyone know of a wifi-enabled filter pressure gauge?

Hello everyone, At the moment, my pump and filter are still inoperative and I just filter my pool with my pool vacuum robot once a day... But as I gear up to install my new pump and get my old filter to work again, I need to replace its broken pressure gauge. I plan to run my pump on low 24/7...

www.troublefreepool.com

 

[Jpadie]

as a progress report, I've (nearly) finished a proof of concept that does the

chemical dosing
chemical empty measurements
temperature measurements
ph measurements
ORP measurements.

all works on the bench. I have to print up a holder for the probes to be screwed into a standard 63mm union then I can test properly.

If I have any spare pins on my repurposed circuit board I may include the VfD controller too. Unlikely, though, as this board is pretty full up.

I do need the VFD control before I go into quarantine so if I cannot rig it up then I will create another quick board to do the necessary. the concept is straightforward.

assuming all that works, all that is left is the SWCG (which is v easy, I think) and maybe a pressure transducer or two (my pool uses two pressure gauges, for example). No reason why the system could not turn on/off pool lights too. I'm not sure how most people do this.

then comes making the software and UX more usable. and getting the chemistry right.

 

[cowboycasey]

That is a lot in 4 days, WOW ... is this going to be something that the average person can do or something you are going to sell?

 

[cmc0619]

I think the consensus here seems to be that ORP isn't a reliable indicator of Chlorine levels in the presence of CYA. That could be a problem. The rest looks cool though! You could also add a water level sensor. When I was doing my DIY controller I remember reading a thread about a few people using wireless transmitters to push that data to their pad. It'd be a neat idea to just pickup a SWG cell from wherever and hook it to your own power supply and control it that way without all the extras and branding.

 

[Jpadie]

That is a lot in 4 days, WOW ... is this going to be something that the average person can do or something you are going to sell?

absolutely not going to sell it. just publish the designs and how-tos and let everyone have at it! hopefully people will build on it and make it snazzier. give back to the community.

Once I get the circuit board neat and tidy I may get some extras made and people are welcome to those.

I'm constantly frustrated that a number of industries dress things up as complicated and expensive; whereas inside there's virtually nothing of true complexity. So, although this probably isn't for the general public as it requires soldering skills; hobbyists and tinkerers will be able to build it in a couple of hours. Perhaps a secondary market might spring up of people willing to build these for friends and colleagues.

if they prove any good ...

 

[Jpadie]

I think the consensus here seems to be that ORP isn't a reliable indicator of Chlorine levels in the presence of CYA. That could be a problem

i can see that. I don't use any CYA. seems counterintuitive with a chlorine generator. i guess a small amount wouldn't harm though. i'm currently intending that the system do orp measurements at night; so that sunshine does not alter arbitrarily the redox.

You could also add a water level sensor. When I was doing my DIY controller I remember reading a thread about a few people using wireless transmitters to push that data to their pad.

i don't immediately see a way to do that with my current plan as you'd then have quite high currents (and potentially voltages) near the water line. possibly a separate module that can run on batteries or solar and can be put inside a skimmer cover could work. coupled with a cheap irrigation valve to fill the pool via the return jets. that's very easy. probably use some float switches for the skimmer module. it would only need to measure once a day ( assuming the flow rate through the irrigation valve were known); so a decent lipo battery might well last a year or so between recharges.

worth thinking about!

 

[Katodude]

So I see that @cmc0619 has already responded. I would suggest you read his DIY pool controller thread. My pool controller is based on his design. I have also added pH sensing but not acid dosing (easy to do at a later date if I decide). My signature has the link to my thread.

Both of ours are based on a Rpi with Sequent microsystem relay (and ADC converter) hats. The code is NodeRed based. We also have incorporated Nodejs-poolController to use as a link to control a VS pump. We are happy to share anything on our design. Or if you just want to bounce ideas happy to help.

 

[our_hummel_abode]

Hi all

I'm about to have to quarantine in the UK for 14 days. So I thought that was a good opportunity to advance a project I've had in mind for a while. I'm calling for collaborators if anyone is interested.

The project is a hobby level pool management system. The features would be

pH measurement
pH dosing
ORP measurement
Liquid chlorine dosing and
SWCG control
SWCG (the whole thing other than the cell).

Outside of the cell I'm targeting a price point roughly as follows

Peristaltic pumps *3: £15ea
ORP sensor: £25
pH sensor: £15
Depth sensors for the acid/base/chlorine tanks: £5ea
Components and board fab: £25
Toroidal transformer 230vA. No idea of price.

Probably also need a flow switch and maybe an optional pressure transducer. Because why not.

So roughly £120 + transformer + cell

These are one-off prices. I'm quite sure the price point could be pushed way down in some quantity.

I've got replacement plates for my 1440cm2 cell which are costing £175 delivered incl tax. Against the manufacturer replacement cost of £1100

Additionally there would be

* Design files for housing
* Web/mobile interface design and coding
* Firmware coding
* Sample Alexa skill to query and control the interface vocally.

My intent is to make this entirely open source. So providing all files in usable and modifiable formats.

I've got the time available to get a long way into the design of all this in early September, if not finish it. but I'd love collaborators and people to help answer questions about the electrolysis chemistry, UX and UI design etc. And if anyone has microelectronics expertise, to collaborate on the design of the circuitry.

I can also add control of standard vfd devices too (either by modbus or the usual four switch method), if that's thought useful.

I'm not trying to disrupt the market at all. But this might be something that the tinkerers amongst us might enjoy building. And that way you could make it yours.

When I say hobby grade, I suspect that it will function and be as safe as the stuff I've currently got. I've taken that to bits and it's a very basic design. The fundamentals are all ok but these days we can be much more efficient at.much lower energy costs.

Any interest let me know. Looking forward to chatting.

Justin

I've been pursuing this too, though nowhere as ardently as you. I'm excited to follow along and contribute where I can. If the thread deals with pool automation - especially of the open-source variety, I'm interested!

I've experimented with automated chlorine dosing (10% bleach) and encountered a couple of problems that maybe you or other readers would have tested solutions for: Bleach dissolves or corrodes a lot of construction parts I've tried, doing so within weeks if not days.

I haven't found an epoxy that holds up, even ones the manufacturer thought could withstand bleach. Hot glue seemed to survive better. Valves and motors (that survive) I'm still trying to find too.

Anyone have components to suggest that have been tested for their ability to survive both the chemicals and the elements (UV, etc.) long-term?

 

[Dirk]

If you watch enough YouTubes, you'll stumble on the commercial for a UV-activated glue. The glue material is plastic. Perhaps that can do the trick.

Pentair's IntellipH sports a pump that is holding up to acid and the elements. They do require periodic parts replacement (the tube thingies in the pump). I thought I read that the pump is a readily available part via a different brand, which might circumvent a good deal of Pentair's mark up.

 

[Jpadie]

Where is the Cl- exposure? In the design I'm planning the dosing is done with inert aquarium hose (Cristal in French). I helped some elderly chaps avoid getting ripped off by a pool.company near me and when I looked at their system the hose was discoloured and lacquered up slightly on the inside. After five years of continued use. 2 EUR to buy new hose.

If you're concerned about the entry port for the hose, I'm planning to use a cable gland made from ABS. If that doesn't work then I may make one in nylon.

The last time I made a makeshift chlorine doser I just drilled through the lid of the can and fed the hose through. A 20kg can is used up in a couple of weeks so there is not too much that's outgassed through the small diameter left around the hose.

There was no prejudice to the cheap dosing pump I used. It's not exposed to chlorine.

Im using sodium hypochlorite (12%) by the way. Which is not quite the same as bleach in France (eau de javel). Typically bleach has only half the available chlorine (3-5%). I've never checked but this could just be dilution rather than a different chemical. Is this what you mean by bleach?

In my experiments to date chlorine doesn't seem to prejudice any of the plastics I've printed with. But the whole point of my designs has been to use minimal cost parts. Hence aquarium hose, 3d printed unions, cheap peristaltic pumps etc. If something wears out our perishes then it's replacement is easily and cheaply available.

 

[Jpadie]

I cannot try out my salt water generator idea yet as the electrodes won't arrive until I am in quarantine. But I have got a bunch of n channel MOSFETs in my post bag yesterday. So that's exciting.

Nice things. Quite expensive at 25c each but they handle up to 80A at 40v with an ultra low internal resistance and a switching voltage under 3.3v. Which makes the directly compatible with 32bit processors.

If they are handling, say, 13Amps through the cell electrodes they should only need to dissipate an additional 2.5°C. which does not need heat sinking.

Fingers crossed!

Now I need to think about gate drivers to keep all the fets as N Channel and still make the h bridge work


A question for the community: my SWCG has nine electrode plates. Which bothers me as I like even symmetry. I cannot easily tell how the electrodes are joined as the joints are potted. Presumably so the manufacturer could use brass or steel clamping bolts rather than titanium.

Since there is an odd number should I assume that there is one plate in the middle nominally acting as an anode and the two farthest plates are nominally cathodes? The intermediate six plates then act as passive charge passers to increase the total electrolytic area?

I did try to measure the resistances between each plate but for highly anomalous readings. Sufficient that I thought salt water may have got trapped inside the potting compound. Before I build the new cell I will of course ream out all the potting compound to get a proper look

 

[cowboycasey]

I have one that I can get good pictures of, it has 3 electrodes.. Not sure what reading you want but if you let me know I have a voltage tester

 

[Katodude]

Oh, I missed the part where you were making an DIY SWG. Now that is really cool. You might want to read this thread.

The Arduino - Intex SWG

Intex - inexpensive SWG unit with limited control. Arduino - an open-source electronics prototyping platform based on flexible, easy-to-use hardware and software. It's intended for artists, designers, hobbyists, and anyone interested in creating interactive objects or environments. Yank out the...

www.troublefreepool.com

It would be very interesting to see how this works out.

 

[cmc0619]

I've experimented with automated chlorine dosing (10% bleach) and encountered a couple of problems that maybe you or other readers would have tested solutions for: Bleach dissolves or corrodes a lot of construction parts I've tried, doing so within weeks if not days.

I haven't found an epoxy that holds up, even ones the manufacturer thought could withstand bleach. Hot glue seemed to survive better. Valves and motors (that survive) I'm still trying to find too.

Anyone have components to suggest that have been tested for their ability to survive both the chemicals and the elements (UV, etc.) long-term?

Most people around here, myself included pre-SWG days, use a Stenner peristaltic pump. It keeps the chlorine in an non reactive UV protected hose and can be controlled via automation or on a built in timer, depending on model. The output just goes into a tee into the plumbing and holds up very well. Search Stenner on here and you'll see the various projects.

 

[Jpadie]

That's interesting. How are the electrodes connected? And are they titanium with ruthenium-iridium coating?

On mine I was trying to see whether there were five connected in parallel as the cathode and four as anode. Or whether the outliers were cathodes and the middle an anode. And if the latter were all the cathodes connected? In which case the resistance between two cathodes should be close to zero (you'd need to take measurements on the edge of the plate where there is no coating).

I've seen setups where the electrodes are connected and those where they are not. and some where they are connected in parallel and even some connected in series. I wonder whether it actually matters. possibly the core determining factor is only charge density and charge area.

 

[Jpadie]

It would be very interesting to see how this works out.

thanks for the link to that thread. will add to my reading list this evening.

for my design, I don't want people to feel that they have to butcher anything second hand or pre-existing. I'm hoping that for less than $50 they can build everything new themselves (except the salt cell - and for this I want to allow parameterisation so that the controller can be used with any cell).

so similar end goal but perhaps a different path?

 

[jonpcar]

Jpadie, hi...nice project! Here is some input.: in addition to the projects by cmc & katodude, make sure you check out segalion's project here:

RaspiPool (Pool Automation System with Raspberry pi + Home Assistant)

Hello. This is my first post here. I am working in an automation project with a raspberry pi + home assistant, and I would like to know if there are some here interested or with experience in similar projetcs. Since a few years I have a little system based on pilight, an now I am immersed in a...

www.troublefreepool.com

segalion also has a gihub respository here that is gaining a lot of momentum:

GitHub - segalion/raspipool: Swimming Pool Automation Systen with Raspberry Pi + Home Assistant

Swimming Pool Automation Systen with Raspberry Pi + Home Assistant - segalion/raspipool

github.com

I tend to agree with this comment by segalion on the github/wiki site (it definitely applies to my project): segalion/raspipool

"There are some similar projects around (mainly based on node-red and node-js systems), but, in my opinion, all of them are projects to control systems of specific (and usually closed) manufacturers, or they are projects very specifically oriented to the needs of their creators. "

There are some other really cool projects on this and other sites, some of which are general in intent, but usually still somewhat tailored for specific manufacturers. In any case...almost of these projects are being done by tinkerers...people like yourself who have the time and knowledge to advance their project. A general purpose pool automation solution is tough because it still requires some technical abilities AND lots of time to implement.

My suggestion is: don't try to go too cheap...pools cost $10Ks and the cost in time investment for all of these DIY projects is expensive...the DIY pool automation equipment is cheap in comparison to those costs.

 

[Jpadie]

all good advice @jonpcar !

I'm not really _aiming_ for cheapness. It just really is that cheap.

My thoughts:

controllers

I know rasPi's are everyone's darling but it's a massive overkill in cost and complexity for a simple project like this. Why pay $40+ when a $1 controller is more than adequate?

chem sensors
you can buy cheap sensors or expensive ones. they function in the same way. You're just paying for manufacturing robustness and a label. so long as the micro controller knows their upper/lower outputs (which can be parameterised) then they will all work. I'm assuming BNC as a standard connector. Mine are eur 15 for pH and 25 for ORP. They should last a couple of years if I remember to take them in during winter.

the control boards are a bit of a 'secret' unfortunately. Conditioning a pH signal is important due to the high impedance. This probably is about $1.50 of the build cost. ORP sensors are low impedance and need no conditioning for the average microcontroller (despite what people say ...).

temp sensors
this is a vanity but why not? waterproof ds18b20 sensors are very cheap.

salt cell controllers
the microcontroller is already there, so all that needs to happen is to measure resistance (for salinity) and current (for production optimisation) and keep track of time. Some heavy duty mosfets needed but I've got a board in front of me as I type that will handle continuous current 13A and peak of 30A. and that cost £10...

Dosing
again - you can buy expensive peristaltic pumps or cheap ones. Mine are cheap and have trundled on for a few years. some parts can be 3D printed when they wear out. Aquarium hose is cheap here in France.
Do more expensive pumps have advantages? Maybe they deliver a faster flow rate. But that's not going to matter in this scenario.


My designs don't have screens. This saves on cost and weatherproofing. I prefer everything to be controlled over the web so that I am not restricted by the memory footprint of the device in the presentation layer. This is probably where I differ from other people.

I'm also personally not a fan of node-red or home assistant or whatever. But I use mqtt extensively in all my designs. so anyone can take the mqtt messages and build node-red flows if that's what they want.

Anyway - it's all built already apart from the salt stuff. which I know is like saying that a car is built apart from the seats and motor. but it's a start. And it is all kludged from components I had in parts bins. Once the concept is proved I will design a board and have everything in one place. My constraint is to get it all on a board no bigger than 100x100 to take advantage of all the cheap fab houses in China. This may not be practical in the end; due to the heat dissipation issues from the salt cell controller. But we will see! It might well be that there needs to be a power board and an everything else board.

I've spent today writing the firmware. Then got sidetracked into a few fascinating hours researching the calculation of sunrise and sunset. 'Twas ever thus. I'll finish the first cut of the firmware tonight.

Lastly, i know I _could_ stand on the shoulders of giants and leverage work that's been done. And whilst that has its own pleasures, I would spend longer trying to work out why someone else did something a particular way than doing it myself. And the enjoyment for me is the initial build out. I'm quite certain whatever I come up with can be improved massively and I'm looking forward to comparing projects and learning after the event, then improving the design in iterative steps that way.

 

[mooseau]

Where did this end up? I had such plans too, but no time I connect my pump and booster pump for vacuum to wifi power. My raspberry pi can auto shutdown when Wattage drops. This happens when pressure gets high, but it isn't a perfect proxy for a proper pressure sensor. But saved my bacon a few times, especially when vacuum is on without the main pump.

I have a weather protected pump house, so a Pi zero can be used.