Timer for solar powered pump

[.RC.]

I have a solar pump here I want to put a timer on.   My reasons are the pump only needs to pump a few hours a day to put water into a tank.  Currently the pump runs when there is enough sun and at lower light levels the pump spins but pumps no water as it is spinning too slow.

So I was thinking a timer to run the pump for only a few hours each day in the middle of the day would be the best solution.

The pump controller has signal inputs to start/stop the pump, they work the pump runs when the terminals are open, when the terminals are closed the pump stops.

So I need a device that has a timer, battery so it keeps the time at night, and the battery needs to be recharged by the panel during the day.

The panel is 200W with something like 45V max.

I have googled and searched without success so far, google seems to not be as good as it used to be for obscure stuff.

I did find this though https://www.ebay.com.au/itm/126012759513

I may be able to find a 12V rail on the solar pump controller.

[themadhippy]

So I was thinking a timer to run the pump for only a few hours each day in the middle of the day would be the best solution.

Hows about a different  approach and  only turn the pump on when the solar output  is over a preset threshold

[.RC.]

I had thought up that idea, but had no idea how to implement it.  I am also not sure of the voltage of the panels under load.

No load the panel voltage is going to be high regardless.   As soon as the load goes on the panel voltage drops.  So you could end up with a feedback loop of panel high, so pump switches on.  It draws the pane voltage down, switches off, which causes the panel to go high again and so on.

I have only got into solar power the past year or so and only in pumping water and this solar renewable energy (we have run windmills up to this point) is more tricky to set up then when you have constant 24hr mains power that can run a pump as a constant speed all the time.

[Kim Christensen]

No load the panel voltage is going to be high regardless.   As soon as the load goes on the panel voltage drops.  So you could end up with a feedback loop of panel high, so pump switches on.  It draws the pane voltage down, switches off, which causes the panel to go high again and so on.

Have a circuit that monitors the voltage across the pump motor when it's running. If the voltage is too low (After a few seconds run time), meaning the pump is turning too slow, turn pump off and set a timer (30mins, hour, whatever) and turn the pump back on again once the timer expires. Rinse repeat. No need for clocks or batteries.

[gnuarm]

What is the problem with your current arrangement?  Does this hurt the pump in any way?

[NiHaoMike]

I had thought up that idea, but had no idea how to implement it.  I am also not sure of the voltage of the panels under load.

No load the panel voltage is going to be high regardless.   As soon as the load goes on the panel voltage drops.  So you could end up with a feedback loop of panel high, so pump switches on.  It draws the pane voltage down, switches off, which causes the panel to go high again and so on.

Have a normally on dummy load that's turned off when the voltage gets sufficiently high? Then it would have its own hysteresis.

[.RC.]

What is the problem with your current arrangement?  Does this hurt the pump in any way?

At the moment the pump runs all the time, when there is no need to as the tank just overflows.  Wearing the pump out prematurely.

Have a circuit that monitors the voltage across the pump motor when it's running. If the voltage is too low (After a few seconds run time), meaning the pump is turning too slow, turn pump off and set a timer (30mins, hour, whatever) and turn the pump back on again once the timer expires. Rinse repeat. No need for clocks or batteries.

That is possible, but remember these solar pumps are brushless DC motors kind of like a 3 phase AC motor, but rather than a sine wave it is a PWM square wave. Or at least that is the way I understand it works.  Measuring the voltage may not be as straight forward, plus it complicates matters.

Have a normally on dummy load that's turned off when the voltage gets sufficiently high? Then it would have its own hysteresis.

Yes that is possible a 100W load would do it, but again I think it complicates matters.  I am sort of wanting to go the off the shelf kiss method.  if there is a $20 or less timer out there, off the shelf just wire and go, it is just simple.

[gnuarm]

What is the problem with your current arrangement?  Does this hurt the pump in any way?

At the moment the pump runs all the time, when there is no need to as the tank just overflows.  Wearing the pump out prematurely.

Ok, fair enough.  But, you never said anything about that... or I missed it.  But, you still did not answer the question.  I will assume it does not harm the pump to run without moving water. 

This is an easy problem to solve, depending on your level of experience with electronics.   Restating your problem to:  "I want my pump to only run when the tank is not full." 

You will need a level sensor in the tank, which can be as simply as a float connected to a switch.  I have seen these in cisterns and they are very common.  I don't know if they handle the power line voltage, but I would want to use 12V power to be on the safe side. 

So you need a 12V transformer to supply power.  This runs through the tank switch to a relay.  The relay switches the power to the pump.  Now, the pump will not run unless the water level in the tank is not full. 

If you don't want the pump to run when the solar level is too low, you will need a light sensor for that, which can operate another relay, wired so both need to close to provide power to the pump. 

Is that clear enough? 

[themadhippy]

12v  time clock with a battery back up  it is then ,assuming you can get a 12v source to it every so often ,maybe something like https://www.bladespowergeneration.co.uk/7-day-timer-12v-dc

[.RC.]

Ok, fair enough.  But, you never said anything about that... or I missed it.  But, you still did not answer the question.  I will assume it does not harm the pump to run without moving water. 

This is an easy problem to solve, depending on your level of experience with electronics.   Restating your problem to:  "I want my pump to only run when the tank is not full." 

You will need a level sensor in the tank, which can be as simply as a float connected to a switch.  I have seen these in cisterns and they are very common.  I don't know if they handle the power line voltage, but I would want to use 12V power to be on the safe side. 

So you need a 12V transformer to supply power.  This runs through the tank switch to a relay.  The relay switches the power to the pump.  Now, the pump will not run unless the water level in the tank is not full. 

If you don't want the pump to run when the solar level is too low, you will need a light sensor for that, which can operate another relay, wired so both need to close to provide power to the pump. 

Is that clear enough?

The tank is 200m from the pump, so no go for the sensor.   I have a high/low sensor fitted on other installations where the pump and tank are close. Cost for this is negligible around, $20 for the sensor plus cable.

I considered a high/low sensor on the tank and wireless remote system.   Too expensive and not KISS.
I considered a pressure switch and a double float valve on tank. Simple but would require ~$1000 in parts to implement

I never thought of a light sensor.  That is something I will keep in mind.

The simplest setup I can think of is just a timer.  Run for a few hours a day. As it is checked once every four to five days anyway during times of high demand you can simply increase the duration.

The pump is harmed by not moving water as it wears the seals/bearings out and wastes electrons.

12v  time clock with a battery back up  it is then ,assuming you can get a 12v source to it every so often ,maybe something like https://www.bladespowergeneration.co.uk/7-day-timer-12v-dc

Thanks, The controller has some via's marked 12V see photo below.   I measured and there is 11V at the vias.

[beanflying]

Float valve in the tank and a pressure switch at the pump end along with a 20-80L (larger is better) pressure tank is how it is generally done. There is more fancy electronic controllers that 'in theory' avoid using a larger pressure tank but don't and KISS. The contacts on the switch also avoid issues of DC or AC switching of the fancier ones. You also need a checkvalve after the pump but before the switch to complete the fluid side.

Really high grade double acting float valve allows the level to drop 12-18" before opening (less cycles) from Alderdice but there is plastic ones available as well.



Cost on this at more like $3-400 not $1k either.

Sorry read you latest post after I posted  But this is how it is done for good reason.


Fairly recent and similar thread https://www.eevblog.com/forum/projects/switching-circuit-design-for-water-pump/msg4734056/#msg4734056

[.RC.]

In my setup the cost for the pressure system is a fair bit more involved.

The current setup also feeds multiple stock troughs from the same line.  They would all have to be upgraded to high pressure float valves.  To put a double float valve in the tank is simple enough.  But water has to flow out of the tank when the pump has zero pressure.  So you need to install a brass swing type checkvalve in the bottom of the tank so water flowing into the tank goes though the top, water coming out, through the bottom. A pumptrol reverse pressure switch is a couple of hundred dollars alone, depending on where you get it from. The plumbing fittings all add up.

I have done this before and the cost is fairly well up there.

[gnuarm]

Ok, fair enough.  But, you never said anything about that... or I missed it.  But, you still did not answer the question.  I will assume it does not harm the pump to run without moving water. 

This is an easy problem to solve, depending on your level of experience with electronics.   Restating your problem to:  "I want my pump to only run when the tank is not full." 

You will need a level sensor in the tank, which can be as simply as a float connected to a switch.  I have seen these in cisterns and they are very common.  I don't know if they handle the power line voltage, but I would want to use 12V power to be on the safe side. 

So you need a 12V transformer to supply power.  This runs through the tank switch to a relay.  The relay switches the power to the pump.  Now, the pump will not run unless the water level in the tank is not full. 

If you don't want the pump to run when the solar level is too low, you will need a light sensor for that, which can operate another relay, wired so both need to close to provide power to the pump. 

Is that clear enough?

The tank is 200m from the pump, so no go for the sensor.   

Sorry, I don't get how this has anything to do with it.  You can't run 200 m of wire? 

I have a high/low sensor fitted on other installations where the pump and tank are close. Cost for this is negligible around, $20 for the sensor plus cable.

I considered a high/low sensor on the tank and wireless remote system.   Too expensive and not KISS.

Not sure why you would need a double sensor.  It's not like you need to worry about water sloshing around in the tank.  It's just an on/off switch.  I believe I've said they are common on water tanks for drinking water.

I considered a pressure switch and a double float valve on tank. Simple but would require ~$1000 in parts to implement

  What pressure switch?  Do you mean to sense  the water level?  Don't make this complicated.

I never thought of a light sensor.  That is something I will keep in mind.

The simplest setup I can think of is just a timer.  Run for a few hours a day. As it is checked once every four to five days anyway during times of high demand you can simply increase the duration.

Not really the simplest, but it is the worst performing, running when you don't need it, which is what you have now.

The pump is harmed by not moving water as it wears the seals/bearings out and wastes electrons.

Ok, then the timer is bad, just not as bad as no timer.

12v  time clock with a battery back up  it is then ,assuming you can get a 12v source to it every so often ,maybe something like https://www.bladespowergeneration.co.uk/7-day-timer-12v-dc

Thanks, The controller has some via's marked 12V see photo below.   I measured and there is 11V at the vias.

[Siwastaja]

Float switch + 200m of wire doesn't sound too bad. It can be thin, low-cost wire running low current, low voltage, if you use a relay where the pump is. This would be a simple solution which does the right thing.

[beanflying]

Sorry, I don't get how this has anything to do with it.  You can't run 200 m of wire? 

Not sure why you would need a double sensor.  It's not like you need to worry about water sloshing around in the tank.  It's just an on/off switch.  I believe I've said they are common on water tanks for drinking water.

  What pressure switch?  Do you mean to sense  the water level?  Don't make this complicated.

Dual level sensing is to keep the tank within a range to avoid the pump needing to start to often. It can be done with a single acting valve but it is not generally done in this sort of situation. Even with Electric pumps for drainage an upper and lower float system is fairly normal like below.

Running a wire 200m across a paddock with Animals is not trivial and even with it likely being low voltage it will need to be fairly deep and in some sort of sleeve. In the past we have used low pressure (RuralB not drip irrigation) polypipe for signal wires as it is fairly tough if you come across it with a shovel later on. You will need to either rip that pipe or better yet have it trenched a bit deeper so $$.

The pressure switch is part of the control system involving the double acting float valve to switch the pump on and off when flow starts. Running Troughs off that same fill line the OP has now let us know does complicate that as at any time later the pump will trigger if an Animal does some drinking causing a drop in pressure and the tank will likely just remain full.

The best option we used to set up with this sort of situation was run the pump (typical Diesel or Petrol powered but sometimes electric) on a watchdog/timer. But this leads to the tank either not being full or worst case getting dehydrated when it gets hot in Oz over summer or you finish up simply pouring water out of some overflow on the tank wasting water. Not such a big problem over 200m to walk to the tank and check then tweak the timer for Summer/Winter but plenty of farms are km apart from tank to pump.

WiFi is a solution but it is not somewhere I would bung a couple of ESP32 or a LORA solution and a booster on it to get the job done. Semi Industrial solutions are also not cheap but generally they are more protected against having issues.

Added a Random Link to a collection if Wireless systems available in Oz https://www.instrumentchoice.com.au/tank-level-meters

[gnuarm]

Float switch + 200m of wire doesn't sound too bad. It can be thin, low-cost wire running low current, low voltage, if you use a relay where the pump is. This would be a simple solution which does the right thing.

I gave this a bit more thought and realize I was mis-remembering the system  I had seen.  It was a cistern on public water that was unreliable.  The tank was at ground level and not pressurized.  The house has a pressure pump and a small pressure tank, along with a pressure sensor. 

The float/valve in the water tank was not on a wire.  It simply has some amount of hysteresis and cuts on to let water into the tank from the street.  The rest of the system is independent of this float and valve. 

But... using a float with a switch with hysteresis is a viable way to go.  No need for timers that fail anytime the power goes off or that need battery replacement on a regular basis.  It works off demand and supply.  When there is sunlight and the water level in the tank is not high enough, it comes on.  Very simple, very optimal solution.

[Circlotron]

If the pipe feeding the tank enters the tank at a point somewhat below the full level then you should be able to measure the tank water level at the pump end with a bit of creative plumbing because the water level will be identical at both ends.

[.RC.]

If the pipe feeding the tank enters the tank at a point somewhat below the full level then you should be able to measure the tank water level at the pump end with a bit of creative plumbing because the water level will be identical at both ends.

That was plan A.  I did suspect that the friction in the pipe line would mean there would increase the head a bit and sure enough, at the pump when not pumping the water level is at say point 1.   When the pump starts it will push the water up another foot or so, simply because of friction in the pipe, despite the pipe being 1 1/2".  The tank is about 35ft vertically above the pump as it is.

I am still going for a timer.

[.RC.]

Thanks for the help everyone.   I am going to try this model, but not from this seller.

https://www.amazon.com.au/Baomain-Thc15A-Digital-Programmable-Switch/dp/B01N05TP18

[beanflying]

I used a similar one for years on a 13A coffee machine which was maybe pushing the contacts a bit hard (it did eventually fail) so maybe use it to drive the pump via a more beefy one using it or the control pan.

My current solution is a Sonoff POW into a 30A contactor just to make it more bulletproof plus I can turn it off or on remotely and bypass the timer.

[.RC.]

Update, the timer turned up promptly today. https://www.aliexpress.com/item/1005005041556760.html

I did the EEV motto.  Don't turn it on, tear it apart.

I gave the green board a bit of an isoproponal bath to remove the flux.

I think it is going to be a little ripper.  I am going to power it via a LM2596HV buck converters https://www.aliexpress.com/item/32960193567.html

[.RC.]

Well it worked OK for a couple of weeks.   Then my buck converter to convert the solar panel voltage to 12V for the timer must have got a bit toasty.