Switching circuit design for water pump

[bozboz]

I need advice on the construction of, or buy off the shelf (ebay etc) of a small circuit that can power a relay , Which will then power a pump. The set up is in a remote field and will be powered by some solar panels and some batteries. The pump, which may or not be mains fed via an inverter, is to pump water 200m to a collection of tanks for irrigation. The final tank in the series has a float switch, which will become open circuit when full. This needs to tell the pump to stop. As its 200m away, i thought the best idea was to make this part of a low voltage DC circuit and link the float switch to my switching circuit using cat5 cable (and its cheaper). So i need a small switching circuit that can sense a closed circuit (bearing in mind that a 200m pair loop of cat 5 will be about 40 ohms), and be activating a relay to switch on the pump, but then sense an open circuit and cut power to the relay and thus the pump. Bearing in mind that its a battery system with limited power and not to drain the batteries.
Hope that all makes sense,
Cheers

[sparkydog]

So, you need a device which... when it sees voltage... sends power to something else... and when it loses voltage, stops sending power? You realize you're just describing a relay (or a MOSFET), right?

You almost certainly want an SSR. I'm not sure if it's an improvement in total power draw or not to just use the signal line as part of the resistor to drop your control voltage (which is probably 3.3V or higher) to the level the SSR wants (usually around 1.2-1.5V), or if it makes more sense to use a MOSFET with the smallest current you can manage on the signal line to drive the SSR. Your SSR may or may not be able to run the pump directly; if not, you can use it to drive an EMR that controls the pump.

Be careful buying off-the-shelf stuff, though, as most of it won't be rated for use in your intended environment. OTOH, if you build something, keep in mind you'll probably have to use some form of conformal coating, because many components aren't suitable for use in Pollution Group 3 either.

That all said, more information would be helpful. What DC voltage do you have readily available? What's the voltage and current draw of the pump?

[james_s]

If I understand the issue correctly, a transistor or mosfet (if the pump is DC) and a couple resistors is all you need, it should be no problem to have the float switch on the other end of hundreds of feet of cable as only a very small current is needed.

[bozboz]

@sparkydog .....@james_s......Thanks for replying. The mosfet lowest current draw would probably be the best option as its a battery system. Am i right in thinking the mosfet would drive the SSR and its that , that would do the switching of the relay?.
I was going with the switching a relay route as it gives much more flexibility to the setup as different pumps could be swapped out easily if not powerful enough or broken, and would give me the option of switching an inverter on only when the pump was required, so not draining the batteries in standby mode.
I was thinking of a 230v AC pump to give me the power and the choice, hence the inverter
The batteries feeding the inverter would be 14v or 15.5v  in parallel, so my DC supply would be from a PSU off Ebay, at whichever current and voltage was needed between 1 - 15v.
Thanks again

[beanflying]

Couple of things to consider adding to your Pump controls is a flow or loss of prime switch if it is a truly remote system you won't be checking on regularly. Recent thread and some links in it might help explain this better https://www.eevblog.com/forum/mechanical-engineering/dry-run-protection-for-dc-solar-pump-sourcing-from-a-borehole/

Running a 'single' float switch in a remote tank with wires or shudder a wireless link  back to the pump is just not generally used. At a minimum use two float switches one for a switch off and then one set lower down the tank to trigger a start to avoid cycling the pump as often.

The more general option for filling remote tanks is a double acting float valve in either Brass or they are available in Poly to allow the tank level to drop 12-18" before flow starts again. Then you just need to run a LOCAL pressure switch or pressure LOP/ Flow switch to turn the pump on or off. Best practice is a pressure switch and small tank for the on off bit and a flow switch for protection.



The old ways are sometimes the best ways, wheels are still round at least for the time being 

[james_s]

If you're using an inverter and a 240V pump I think I would wire the float switch into the power switch of the inverter, some even have a connector on them for a remote panel that you may be able to connect to. Plug the pump into the inverter and turn the inverter on and off to control the pump. If you go that route be mindful of the surge current a motor will pull when it starts up, you need a big enough inverter to handle that without shutting down.

If possible I would probably try to use a 12 or 24VDC brushless pump and control it directly with a mosfet.

[HarryDoPECC]

As noted above, you should consider a design with some hysteresis, otherwise you will thrash that pump.  And consume more power with multiple startups and short run times - the opposite of economical running.

[MathWizard]

What about a wireless transmitter/receiver, surely someone makes off the shelf ones like that for use with arduino/MCU's

IDK how far some simple AM transmitter would go, but I could make an AM TX, that would turn on just to send a tone, that some receiver would be listening for, and then do XYZ

[james_s]

As noted above, you should consider a design with some hysteresis, otherwise you will thrash that pump.  And consume more power with multiple startups and short run times - the opposite of economical running.

Assuming the use of an off the shelf float switch, the hysteresis will be designed into the switch.

[beanflying]

As noted above, you should consider a design with some hysteresis, otherwise you will thrash that pump.  And consume more power with multiple startups and short run times - the opposite of economical running.

Assuming the use of an off the shelf float switch, the hysteresis will be designed into the switch.

Typically most industrial systems apart from simple sump pumps run at least two level probes/switches (plus alarms outside that range in some cases). The Simple answer is less starts is better and typically more reliable (better probes not so much floats). Floats get tangled for 'reasons' randomly, the second option below breaks strings or if a rigid rod is used the float fouls over time with 'gunk'. For these and chemical reasons we tended to use probe systems suitable for the application from proximity to capacitive to conductivity.

But there is a few options for single switches.

Apart from the long floating float switch like the first picture which are 'ok' a better option for tank systems with a single switch like shown in the second as you can set a large drop in level with generally less issues with floats getting tangled. Also the first one in reality tends to switch well before reaching a vertical attitude so the cable needs to be longer than you might think.

This with a 200m run to the tank is still better done with a mechanical valve and local control of the pump. Running wires in paddocks doesn't end well unless you do it properly so deep and in conduit and expensive.