Another water pump automation project (suggestions ?)

[Dannyx]

Re: Another water pump automation project (suggestions ?)

how are the tanks vented?

Through the top lid. The inlet pipes at the top are not caulked in place, so there's plenty of room around it for air to get it. Since this is not for drinking, it doesn't matter too much if nasties get inside too, though it may eventually get to a point where the dirt clogs up the pipework.....

[stj]

Re: Another water pump automation project (suggestions ?)

how are the tanks vented?

[Dannyx]

Re: Another water pump automation project (suggestions ?)

Those water tanks at 1,000L/275gal are large weight up to 1,060kg/2,332lbs.
But they have a small 6" fill cap lid which makes putting sensor and fill pipe inside a hassle.
.. uh make sure the stand is good for a ton, the legs might sink into the ground and the tank fall over.

I noticed the two tanks are connected together at their outlet, but a closed valve is between them for some reason. It would be simpler with them connected together.

Did you finish the plan for the pump logic? An ESP8266 is really good and what I'd use for something small and low cost like this, with a webpage. Before, I thought you were doing something industrial in a plant where the tanks and pumps are much bigger and a PLC is standard use.

Yeah, I'm not sure how the chap was planning to get those floats inside the tanks, since I knew from the very start, just by looking at them, that they only have a very small lid, unless maybe he enlarged the holes some more....

Too bad he already bought the floats, since I would've proposed the hall sensor variant. I'm sure there exist pre-made devices, not just the bare sensor.

[keeney123]

Re: Another water pump automation project (suggestions ?)

I once encountered a magnetic level sensor of the same description in an ironing station. It was used to sense when the water level in the tank dropped too low or if the tank was removed entirely. It had a magnet stuck inside a piece of plastic inside the tank that went up and down with the water level. Water had gotten into the plastic shell and corroded the magnet which no longer "worked". The reed switch that sat on the outside wall of the machine near the tank no longer closed, so the station would always think the tank is missing and wouldn't run. I couldn't get the magnet out without breaking the tank open, so the only solution was to replace the whole tank and that got it going again.

In the case of these large tanks, which I assume are fairly beefy plastic, a stronger magnet would probably be required to trigger a reed switch on the outside...which would also have to be water-proof.

Hall Effect switches have been used in crank shaft position sensors for years. Many millions and millions of cars and trucks have them. Although there may be a problem in very few vehicles, I would say their success rate is around 99% for several years. Unlike your pump a crank shaft sensor has to turn on and off several times a minute. These sensors can work in dirty environments successfully. I would say there has to be many ways to successfully enclose a natural magnet in a float so as not to compromise it.
The tank you talk about replacing tells me whomever designed the float and the magnet inside most likely needs to go back to design school because components fail and tank lids need to be taken off if only for cleaning the tank inside, tanks get dirty. I sure would not use a reed switch and magnet to close it on the outside of the tank. One job i had while I was attending College was as a Security Alarm installer. The magnet and reed switch are used in that industry. They fail quite a bit of the time. You need to look up Hall effect switches. They will acquire a small amount of current from the magnet when it comes in line with it. I would say that the hall effect switch would be places at the top of the water line where water pressure is not great. It actually could be places above the water line. The top of the float in the tank could also be place above the water line. So even when the float went down it would still be above the water line. One could actually place the Hall effect switch inside the water tank in this case. Depending on plastic of the tank one could both screw into the tank and also have a plastic adhesive on the mounding bracket for the Hall Effect switch The Hall Effect switch then could be mounted with stainless steel screw and nut on to the bracket so it could be removed easily. Of course, one has to have a removable Lid on the tank to do this. On then could drill a hole in the lid so the wires to the hall effect switch can exist. The wires would be sealed with food grade silicone caulk where they exist the tank. If the tank does not have a lid one could make a lid for it. Simply mark a circle around the tank near the top on the side of the tank. Use a plunge cutting tool and follow around the marked circle. Then buy food grade gasket maker. Line the cut top with this gasket maker. Then put the top lid on after the appropriate time and wait for a little more time then secure the top lid with stainless steel screws.
If cutting the whole top off is too much, then just cut a section off big enough for a human to get in the tank and clean it. Use stainless steel hinges to secure it to the top and a side hinge. use the gasket maker to fill the space where cut and then tape over the gasket maker.

Of course, some big manufacturers who want to cut cost like Ford Motor Company will design something that is almost impossible to replace without pulling the engine. I believe a friend of mine had a 7.3L Diesel sensor that went bad, and it cost over $700 to replace it. Luckily the truck was still under warranty.

[redwire]

Re: Another water pump automation project (suggestions ?)

Those water tanks at 1,000L/275gal are large weight up to 1,060kg/2,332lbs.
But they have a small 6" fill cap lid which makes putting sensor and fill pipe inside a hassle.
.. uh make sure the stand is good for a ton, the legs might sink into the ground and the tank fall over.

I noticed the two tanks are connected together at their outlet, but a closed valve is between them for some reason. It would be simpler with them connected together.

Did you finish the plan for the pump logic? An ESP8266 is really good and what I'd use for something small and low cost like this, with a webpage. Before, I thought you were doing something industrial in a plant where the tanks and pumps are much bigger and a PLC is standard use.

[Dannyx]

Re: Another water pump automation project (suggestions ?)

Very useful information. I think one could also use hall effect sensors. Using a float on the side of the tank that has a magnetic material inside of it and the hall effect sensor on the outside. One has to be careful with the potting. one needs a potting material that is high enough resistance as to not affect the circuitry and has to be such as to not shrink when it cures.

I once encountered a magnetic level sensor of the same description in an ironing station. It was used to sense when the water level in the tank dropped too low or if the tank was removed entirely. It had a magnet stuck inside a piece of plastic inside the tank that went up and down with the water level. Water had gotten into the plastic shell and corroded the magnet which no longer "worked". The reed switch that sat on the outside wall of the machine near the tank no longer closed, so the station would always think the tank is missing and wouldn't run. I couldn't get the magnet out without breaking the tank open, so the only solution was to replace the whole tank and that got it going again.

In the case of these large tanks, which I assume are fairly beefy plastic, a stronger magnet would probably be required to trigger a reed switch on the outside...which would also have to be water-proof.

[keeney123]

Re: Another water pump automation project (suggestions ?)

With a plastic tank, capacitive liquid level sensors that mount outside on the tank are popular.
XKC-Y26/WS-03 can mount on a plastic pipe (with bracket) OP could use that on a vertical run and not bother sensing the tanks at all.
XKC-Y25/WS-02 flat on the tank outside. Example use Ford Fiesta low coolant light but it's the WS-03 for some reason.
Or available with entire controller. That Ali store is a bit expensive compared to others.

But OP's tanks are outdoors in the rain, so the sensors would get wet. I think some of them are potted/waterproof.

many youtube vids: https://www.youtube.com/results?search_query=xkc+sensor

Very useful information. I think one could also use hall effect sensors. Using a float on the side of the tank that has a magnetic material inside of it and the hall effect sensor on the outside. One has to be careful with the potting. one needs a potting material that is high enough resistance as to not affect the circuitry and has to be such as to not shrink when it cures.

[redwire]

Re: Another water pump automation project (suggestions ?)

With a plastic tank, capacitive liquid level sensors that mount outside on the tank are popular.
XKC-Y26/WS-03 can mount on a plastic pipe (with bracket) OP could use that on a vertical run and not bother sensing the tanks at all.
XKC-Y25/WS-02 flat on the tank outside. Example use Ford Fiesta low coolant light but it's the WS-03 for some reason.
Or available with entire controller. That Ali store is a bit expensive compared to others.

But OP's tanks are outdoors in the rain, so the sensors would get wet. I think some of them are potted/waterproof.

many youtube vids: https://www.youtube.com/results?search_query=xkc+sensor

[sam_sam_sam]

Re: Another water pump automation project (suggestions ?)

I have been following this post for awhile I am very interested in what solution you find for this

I briefly work for a company that used those type of tanks and was trying to use them with float switches but continuously had over flow issues and it was because of the way they implemented the setup and did not want to change it so I did not try to find a solution for there issues with that setup

I do not know if you can find a ( type of switch ) ( if can not find the exact one you are looking for you could add this and a small regulating valve for what ever cutoff height that you might want ) that is used in a washer machine water level detector and do something like this I am just thinking out loud this was one thing I thought of and wanted to try to solve there issues with there setup

[redwire]

Re: Another water pump automation project (suggestions ?)

I looked at the logic for the pump and fill valves and it's a bit weird.
It's like each tank has an empty/fill cycle, once empty the pump runs until it fills up.
But if you're filling one tank, and the other is 1/2 full, why not top it up too? Why not fill both until the FULL float switches trip on both tanks?

I still think you need a latch somewhere to stop the logic flipping out with noisy float switches. It's standard in PLC motor controls to use a latch function.

If both tanks were connected to make one tank, it would simplify things a lot.

[keeney123]

Re: Another water pump automation project (suggestions ?)

In this case, we're not going DIY though, since the guy already bought (and possibly installed) the float switches I showed.

One possible issue I see with your idea is that it only works for small-ish tanks. If the tank is very deep, at least some part of the mechanism would have to be impractically long to reach the top (since you can't install a microswitch at the bottom of the tank underwater)

The beauty of this is that it is hung from the top. So, one can make it as long or short as one want to. The bottom of the outside tube does not need to go to the bottom of the tank. It just needs to go as far down as one wants the level of water to turn on.

The microswitch is installed at the top of the tank.

I have not shown the side walls of the tank this is just the float system installed at the top of the tank. The tank can be as long as you want it.

[stj]

Re: Another water pump automation project (suggestions ?)

so sealed, but hopefully vented tanks and they are outdoors and exposed.
that adds to the problems

[Dannyx]

Re: Another water pump automation project (suggestions ?)

is it not possible to add a balancing pipe between the tanks ? then it should only need 1 pump and float .

That's how I would've done it from the very start, TBH: connect the two together at the bottom. Come to think of it, now that I look at it better, I think the same effect could be achieved by simply opening the bottom valves at the same time with no additional modifications....

[petehall347]

Re: Another water pump automation project (suggestions ?)

is it not possible to add a balancing pipe between the tanks ? then it should only need 1 pump and float .

[petehall347]

Re: Another water pump automation project (suggestions ?)

look possibly a cubic metre .so 1000 litres

[Dannyx]

Re: Another water pump automation project (suggestions ?)

About what size are the tanks?

THIS big. A common sight around here. I think they're 100L each. The valves shall go where those manual valves are, either before or after them. You can see how the fill pipes run to the top.

Did you go with a smaller relay for the latching aspect?

No, because I can't. Like I said: I'd need some way to UN-latch it and he only got 2 floats. I'd need 4 to do this, the way I picture it at least....

[redwire]

Re: Another water pump automation project (suggestions ?)

About what size are the tanks? As I mentioned, as long as the incoming water cannot pour down and hit the float sensors, or if there is not a lot of swirl and turbulence, they can work.
They are designed for sump pump applications, which do not have this problem. I hope this guy does not keep buying the wrong stuff.

Did you go with a smaller relay for the latching aspect?

[Dannyx]

Re: Another water pump automation project (suggestions ?)

In this case, we're not going DIY though, since the guy already bought (and possibly installed) the float switches I showed.

One possible issue I see with your idea is that it only works for small-ish tanks. If the tank is very deep, at least some part of the mechanism would have to be impractically long to reach the top (since you can't install a microswitch at the bottom of the tank underwater)

[keeney123]

Re: Another water pump automation project (suggestions ?)

Here is one of my simple designs for using a float and limit switch. Maybe it will trigger an idea for you. You will have to rotate it clockwise.

[Dannyx]

Re: Another water pump automation project (suggestions ?)

here's a question.
is the water on the pump inlet under *any* type of presure at all?
can it free-flow slowly through the pump when it's off????

No, I don't think so: even though I haven't actually seen the setup myself, the guy told me the water inlet sits above the tank, so it's never submerged.

From OP's pic, it's a submersible pump in the well, with solenoids to divert flow to either of the two tanks. Usually there is a backflow valve to prevent siphoning/water running back down into the well.

Correct. Said check valve may have already been installed by the guy.