Another water pump automation project (suggestions ?)

[Dannyx]

Re: Another water pump automation project (suggestions ?)

Yeah, my bad: I promised a diagram and forgot to add it Here it is....in all its MS Paint glory

The principle of operation: when either float switch drops down (so that tank is low on water), it activates the pump AND the respective solenoid valve to fill that particular tank, independent of the other. Obviously, we don't want to run the pump with both valves closed. Sounds simple and I would've done it without any micro at all, using just relays, but I got hung up on the client's requirement to be able to inquire the status of either tank and switch on the pump on demand (along with the respective valve of course).

Now when I say "status", in this implementation, it can only know whether the float is up (one contact closed) down (other contact closed) or in-between (neither contact closed) - it won't have any way of knowing the actual level, so it's just a dumb DPST switch there and there's no need to obsess over that too much. It's the user interface I'm most interested in, plus it's an opportunity to learn how to expand such a system in the future, for a remote pump, like I tried to do last time.

[stj]

Re: Another water pump automation project (suggestions ?)

the large nucleo boards can run a web server - there is even example code for it online to communicate with a phone browser
but why??
and whats to control??
the pump fills the tanks till the floats stop the pump - job done.
no need for a fucking computer - it's not made by mercedes

[keeney123]

Re: Another water pump automation project (suggestions ?)

My suggestion would be to show a pictorial of what you have and where you want to go. Leave spaces in the diagram to fill in what you may not know at a later time. Put things in block diagram. With such a picture one can then work on each part of the diagram. Then people on the site who may be better at certain parts of the block diagram can help you with each part of their expertise. That would be my suggestion to You.

[Dannyx]

Good day folks. Some of you may recall that a while ago I was struggling to put together what I considered to be a very basic automation for a water supply system. It didn't work out the way I intended and I learned my lesson to do it properly from the start, rather than try to reinvent the wheel and patch something together myself from scratch, so I'll stick to commercial alternatives for future projects and such a project just came in today !

This one is easier simply because this time the pump and level sensors are in the same place, so they don't require a remote connection. The client DOES require remote control and monitoring though. I'll let the doodle speak for itself, instead of trying to describe it too much: two separate tanks, one pump for both, 2 valves, 2 float switches - make sense ?

Not wanting to DIY it and fail again, I began searching for OTS solutions, which brings me to the actual topic: I'd appreciate suggestions from people who happen to work in the field and/or have more experience with industrial automation than I do

Several things I thought of:

- some sort of PLC (Siemens, Schneider). The most professional approach, but also the most pricey and requires the most work, because I'd need to come up with some sort of web server, which is the the hardest bit. I HAVE played with PLCs before at a very basic level, but never with more advanced features like control over ethernet.

-a GPRS/4G solution like a gate opener: easiest to implement out of the box, plus it only requires a SIM card to get it online. Not the most visually appealing to use - no GUI: the most it can do is return SMS messages when the status of the floats flips and the pump turns on/off. Many of these don't have more than 2-3 outputs, which is a bit close for comfort and leaves no room for further expansion. They may also not support any IF-THEN functions, so I may not be able to map input 1 to outputs 1 AND 2 simultaneously...

-Shelly Pro 4PM: this one almost won, but not quite. Cheap, app is ready to go out of the box, has 4 inputs and 4 outputs, has ethernet, is DIN-rail compatible. What killed it is that, according to the manual, there's no isolation. The inputs are referenced to the "hot" wire - no low voltage control - so I'd have to pass 230v through the float sensors underwater and in humid environment ??? Yikes ! I WAS thinking of taking it apart and getting it down to the bare-minimum, which I'm sure is like a transistor/optocoupler+resistor divider jobbie, dropping the 230v to pull the respective pins of the microcontroller low/high, but that's not 100% guaranteed....

Any more ideas ? Thanks.