Automated Garden Watering system

Each summer I take off a couple weeks, but my garden keeps growing, so I want to  set up an automatic watering system. Last year I bought some inexpensive solar pumps for the rain barrels. When I switch on the pump it runs until the battery drops to a certain level. The pump runs again when the battery charges up to whatever level is set. The problem is that the solar panel is too small to charge the batteries up enough and they just don't work well. My co-worker suggested that I use a valve since my garden is downhill from the rain barrel. I have documented my project so far, but I have a few questions. How can I set up a simple timer to switch the valve, or maybe I should ditch the RC circuit and control the valve entirely with the micro controller.
I purchased this “off the shelf” irrigation valve. The spec sheet had no useful info and I soon discovered it is marked with 24VAC. My plan is to have a solar powered system to supply rainwater from a barrel to a garden. I did some research on-line and found that it has been done. I measured the solenoid at 43Ω.
I connected the solenoid to bench power to see what the current draw is at 12 VDC, .28 A
Release below 3 VDC .07A
Hold 5VDC .118A

One discussion on-line gave Two ideas of how to make a simple circuit.
The idea is to activate the solenoid with 12VDC and then hold it at a lower voltage.
The first schematic has a capacitor in series with the dropping resistor. When the switch is open, the cap charges to 12v. When the switch is engaged, the cap discharges, actuating the solenoid. The solenoid is held at a lower voltage, determined by the ratio of R1 to the coil.
The second schematic has a capacitor in parallel with the resistor. When the switch is closed, the discharged capacitor behaves like a short and delivers 12V to energize the solenoid. The capacitor begins to charge and all the DC current goes through the resistor to hold the solenoid at a lower voltage.
I prefer the second circuit. The capacitor in the first circuit could have some leakage, not good for a battery supply.

I will need a 7V drop across R1. The solenoid resistance is 43Ω. 43Ω/5V = 8.6, 8.6*7V = 60Ω, 43Ω/60Ω = .71, .71*12V = 8.6V drop.  I have a 62Ω resistor, so 43/62 = .69, .69*12 = 8.3V drop or a holding voltage of 3.7.
Now to build the circuit. I need the dropping resistor, terminals for the supply and the solenoid and a capacitor, but what value? I’ll try a couple. Add switches to audition a couple caps, and to open/close the circuit.
When I hook it up to bench power I get 4.8V hold at .116A. It works down to 9VDC at 86mA with 3.6V hold. Now to look at the capacitors.
The 470uF capacitor gives 30mS from 12V to 4V. The 2200uF gives about 90mS. The 470uF works, the 2200uF gives a stronger kick and since I’ll have battery supply, I like the 2200uF.

Now I need a long delay one-shot or something, a timer to control the valve. I like simplicity, and low power consumption. I’ll use an Arduino UNO. I could use pwm, but I may change the timer to something simpler.
My simple code to have the valve open for 5 minutes every day. I use the on-board LED to monitor if it works. I set the off time to 55 minutes and left it overnight, but it looks like when the pc went to sleep, nothing registered in the serial monitor. I’ve had it running this morning and it’s gone through a couple cycles. The problem with the delay command is that it does just that and it makes it hard to monitor what is happening. I may put some more effort into the code, but I think it will work as is. It would be nice to have a timer display and a knob to set the on/off time.

http://www.gardeners.com/buy/solar-irrigation-kit/36-517.html  not available ?
http://www.irrigatia.com/irrigation-kits
https://www.newegg.com/Product/Product.aspx?Item=9SIA08C4G51602&ignorebbr=1&nm_mc=KNC-GoogleMKP-PC&cm_mmc=KNC-GoogleMKP-PC-_-pla-_-HI+-+Pumps+%26+Accessories-_-9SIA08C4G51602&gclid=CNSwxKLihtUCFQePaQodAWIJNw&gclsrc=aw.ds

perhaps some ideas to borrow from commercial units...

JR

The Flowerhouse and Irrigatia look like the same thing, and that's what I bought last year.  The way they operate is dependent on the level of battery charge, so if I charge the batteries up fully before installing them, they work o.k. They don't work o.k. now. The more I read about automated systems the more convinced I became that I could build a better system from discrete components.

If I had a long enough hose (don't we all wish that?) I could send you irrigation water from either my dehumidifier collection tank, or from my reverse osmosis water filter bypass, both output water usable for irrigation. I dump well over a gallon a day down the drain in summer weather (more in winter). Maybe catch some more summer humidity dripping from my air conditioner.

I wonder if there isn't some old school natural way... like tons of straw or pine needles around the plants to retain moisture...  Maybe tenting them with clear plastic (like a pseudo greenhouse) to retain moisture.

If you want to throw a micro at it, why not include moisture sensors in the dirt so you only water when and as much as needed?

If you have to hold the solenoid open for water flow , you may want to fill a secondary reservoir that drains slowly to reduce the open time and battery drain, I don't think I've ever seen a latching solenoid valve that stays open and/or stays closed. Only requiring power to change state.

but amazon has... https://www.amazon.com/s/?ie=UTF8&keywords=latching+solenoid+valve&tag=googhydr-20&index=aps&hvadid=204189281708&hvpos=1t2&hvnetw=g&hvrand=7616892077834879553&hvpone=&hvptwo=&hvqmt=e&hvdev=c&hvdvcmdl=&hvlocint=&hvlocphy=9013894&hvtargid=kwd-21728566&ref=pd_sl_qgcs0uhby_e_p20



JR

PS: Yes I've used that trick before with cap in parallel with resistor to provide the higher pull in current, and lower hold current for relays to conserve power supply draw. 

Considering all the people growing pot at home, there must be a treasure-trove of water/nutrient plumbing and timers.

Maybe not what you want, since pot usually needs Grow Light with far bigger demands than any battery, so milli-power valves are not needed.

JohnRoberts said:

... I don't think I've ever seen a latching solenoid valve that stays open and/or stays closed. Only requiring power to change state.

Click to expand...

In industry, this is generally done with motorized valves, with limit switches that can also show the state of the valve on a panel somewhere, open or closed.

Warning: Useless trivia ahead:

They do have latching relays, either two-coil that toggle back and forth between states, or the kind that have permanent magnets in them. These are clever, the magnet isn't strong enough to pull the armature in, but strong enough to hold it in. A plus current on the coil adds to the magnet force and pulls it in if it is not already in, and minus current on the coil nulls out the magnet force, and it drops out.

PS: Yes I've used that trick before with cap in parallel with resistor to provide the higher pull in current, and lower hold current for relays to conserve power supply draw.

Click to expand...

These days, how about PWM? No series resistor dissipation loss.

Gene

JohnRoberts said:

.... I don't think I've ever seen a latching solenoid valve that stays open and/or stays closed. Only requiring power to change state.

Click to expand...

There are. 

One of my customers used to use them on their water fountains and we designed the controller for it. I still have left over controller boards and components.

Irrigation systems use valves with a Venturi valve. They require a voltage pulse to open and one to close. They also need at least 0,5 bar water pressure to open and close. Generally, they need only a few milliamps current. It's the water pressure that opens the valve itself. Some bigger valves need >1 bar.

The ones I have here are 24 VAC too.  I simply use an intermediary 5V relay and  keep the 24 VAC circuits (that have long wires)  completely separate, to protect from lightning and electrostatic voltages.

Guys, I already posted a link to a page full of latching solenoids....

JR

And there is a latching solenoid available for the valve I bought, but it costs more than the non-latching.  One issue with the latching solenoid, if it opens and then the power is lost, the water reservoir will drain.  That first link to the Gardeners system looks like a simple one. A solar panel connected to a pump. You control time and duration by placing the solar panel behind a shade, when the sun hits it, the motor runs, in my case the valve opens. If I had consistent sun, I would go for that, but I can't rely on the sun here (near Seattle). I did some current tests, and the Arduino with the RC circuit draws 170mA, Arduino with PWM  @ 75% on  draws 104mA.  PWM I hear the solenoid, I'm not sure if that chatter is harder on the life of the device than DC.  I did a search here for similar posts and found Two. One was about moisture sensors. I'm not sure moisture sensors are needed  as I plan to water a little bit every day, like I feed my cats weather they tell me to or not, I know it needs to happen. I also can not put down a layer of straw or anything to reduce evaporation because that would be a nice environment for pill bugs. They eat my sprouts, or follow up after the slugs, so the environment needs to dry out occasionally.

One concern I had was, will it work with the low pressure? Yes, I tried it and got flow, enough anyway. I just need a trickle. My supply is a 55 gallon rain barrel and the greenhouse is downhill.

JohnRoberts said:

Guys, I already posted a link to a page full of latching solenoids....

JR

Click to expand...

Apologies. Missed it. Old age is kicking in.

The problem is doing this inexpensively... the rain water collection system has only gravity pressure, so in order to use an inexpensive valve & controller system the pressure must be increased with a pump.
Large diameter valves that do not drop pressure are very expensive.
I have 2  systems for 2 different gardens, one with 1000 gallons of rainwater storage and one with 600.
I use an off the shelf 6 zone controller (Rainbird or the like..I think it was $25) and a $50 Danner pond pump (I also have a koi pond, so having extra pumps around is a given), and the pump output is manifolded out to each zones standard 24 VAC valve (about $10 ea.) All drip irrigation from there...
The controllers 24VAC output for each zone is also run to a diode (I have each diode output run to a switch so I can turn off the pump in individual zones for service) and they all join together to run a pump power relay that turns on the pump when a zone is on (120 VAC pump power).
In a nutshell, when any zone is on, the pump is also on, and off when off...
There is also a water level switch in the 2nd tank in line (dampens the response time to a useful amount), so if there is no rain water the system gets filled with enough well water to function via a separate 24 VAC valve .
Also important is a sediment filter on the collection tank output.
I have all condensate and R/O backwash lines run to the collection tanks as well.
This system has worked very well for a few years now, if you want to use a DC pump (solar panel charging a battery and so on..) that would be easy, too, but as my entire property is run off PV anyways, no need to do that!

> Large diameter valves that do not drop pressure

He's not trying to move 10GPM. If I understand, he wants a small part of 55gal once a day, and it may be OK for those few gallons to take a few hours to flow. The plants are not going anywhere.

Which begs a question: why not a super-small fixed "valve" which leaks 3 gallons in 24 hours? Gives you a couple weeks away.

nielsk said:

The problem is doing this inexpensively... the rain water collection system has only gravity pressure, so in order to use an inexpensive valve & controller system the pressure must be increased with a pump.
Large diameter valves that do not drop pressure are very expensive.

Click to expand...

Got a spare toilet laying around? ;D Just the back tank part will do.

Gravity feed into the toilet float valve, and fire a solenoid to pull on the flush handle once a day. If not enough water per flush, mount the important parts in a larger tank, I'm thinking a plastic 55 gal drum is easy to work with.

Me, I'd go with the whole toilet, because it's funnier that way. And you get a free birdbath, if you leave the seat up.

Not sure how you can handle multiple zones, other than an array of multiple toilets out there. 
So what, if it looks like a bus station rest room.

Gene

Gene Pink said:

Got a spare toilet laying around? ;D Just the back tank part will do.

Gravity feed into the toilet float valve, and fire a solenoid to pull on the flush handle once a day. If not enough water per flush, mount the important parts in a larger tank, I'm thinking a plastic 55 gal drum is easy to work with.

Me, I'd go with the whole toilet, because it's funnier that way. And you get a free birdbath, if you leave the seat up.

Not sure how you can handle multiple zones, other than an array of multiple toilets out there. 
So what, if it looks like a bus station rest room.

Gene

Click to expand...

That might prove too much of a temptation for miscreants....  :

JR

Flow can be regulated by the choice of drippers...
How about a linear actuator run by the solar panel that lifts a metal ball over a given time interval, to a point where it lets it go into a cup attached to a rod attached to the toilet handle, causing the toilet to flush, then it returns to be lifted again?
Less Rube-ish (& not as much fun), but cheap & cheery, looks like a 1/4" 12 VDC motorized ball valve can be had on evilbay for around $30, another $25 for a solar charge controller that has a few timer options to control 12 VDC lighting, & use that to run the valve (& manage the battery charge)?