Full DIY Irrigation System!
Вставка
- Опубліковано 4 жов 2024
- Support the It Kinda Works Patreon (And help keep IKW going!!!): / itkindaworks
Join the IKW forums today and join the conversation!
itkindaworks.com/forum
It's been a project I've wanted to do for a long time, build an automated watering system for my back yard (because I cant be bothered to water my own lawn, I have far more important things to do like spend hours cutting videos together!) So here it is finally done. This is one of the largest/most expensive builds I've done totally around $80-90 for the electronics and then a whole lot more for the hosing and sprinkler heads but it's totally worth it!. Check out the build here!
Link to the code for the sprinkler system: gist.github.co...
Parts List (BOM):
1x ESP8266 prototyping board
1x AC-DC transformer (5V)
1x AC-AC transformer (24V)
3x Orbit 3/4in irrigation valves
3x 3/4in male NPT to 3/4in male GHT
3x 3/4in male NPT to 3/4in female GHT
1x Garden hose manifold
3x 5V relay modules
4x 1000uf 16V capacitors
2x tactile buttons
2x 1k resistor
various lengths of 2 core cable (for power)
various lengths of 4 core cable (for to go to valves)
Support the It Kinda Works Patreon (And help keep IKW going!!!): / itkindaworks
Subscribe for more weekly videos!
Follow me on Twitter @itkindaworksinc
![Solar Weather Station [WiFi, MQTT, Smart Home, ESP8266]](http://i.ytimg.com/vi/at7wmm9t8UE/mqdefault.jpg)
Got a question or something to add to this video? Check out the IKW forums at itkindaworks.com/forum/
Thanks for sharing. I really like the simple and straightforward way you explained every single connection including what each section of the code does. Great work
Nice setup. If this might become a permanent installation, you would benefit from some back-flow protection on your plumbing.
Appreciation for this valuable video.
Thanks for sharing this video! It's been on my to-do list as well, but your video makes it even easier to implement.
I have a similar project in the works, but it's all based off battery/solar/rain water. Some videos on my channel for it.
I got those same valves and have tested a replacement solenoid that is a latching one. Meaning instead of needing to hold the signal to keep it open/closed just pulse it. Works well.
Add a rain sensor to an extra GPIO and have it skip that day if it's raining. The sensor is maybe 3 dollars.
Oh dont worry thats already in the works ;)
I thought so.... I just wanted to sound cool.
I love it!
...or, as it is already on MQTT network, connect it to weather service to get rain status. No hardware changes needed.
Thanks for the video. This type of project has bubbled to the top of my project list.
How is this working after 2 1/2 years? What changes would you suggest? Going off to check the forum now.
Hi, I love your coding. Its very smooth. On the other hand I would not like to rely on internet time, because in the pipe some could fail. I would like to use a RTC in it, that maybe occasionally pulls the time from the net. What is your idea to code this in?
Hi, thanks for this video I need also a system like this bud for 23 valves and they gona be commandet with the ESP32 and eacht valva must have his own wireless receiver because I wane save on wires and my land is to big, do You have any recomandations ore a system like this all ready build ? Thanks again !
Hi,
This is a nice work.
Could you publish the parts list?
I have 2 comments / questions.
1. If I am correct, the relay is 5V operated and the ESP issues 3.3V on its ports. did you consider using some level converter from 3.3V to 5V using a transistor or somehow else?
I understand that this works as is, but I think that - "it kind a works....." and could be safer to implement with level converter.
2. What was the reason that you decided to put the watering program in the ESP rather than in the hub-controller, and keep the ESP more "dumb" acting as a slave operated by MQTT command coming from a master commander - the hub controller?
I think that keeping the irrigation program should be in higher layers, so that you will easier make changes in the irrigation program (Not the code).
For example disable a watering plan until further notice when you expect a rainy week, or to postpone watering until getting some sensor measurement - from another ESP connected to the home automation system.
I'd be happy to put the BOM in the description! I actually meant to before putting it out but forgot.
1. I did consider using some sort of level shifting/transistor but I've played with these relays enough to know that they need 5V to power the relay but the opto-isolated trigger works just fine on 3.3V. You're more than welcome to level shift if you want but 3.3V works great as well and I've not had any issues at all...
2. That is sort of an eventual plan. The ESP is plenty powerful enough to handle both jobs though. It's easy enough to segregate the different parts of the program which I've already sort of started on with the cycleHandler function - it's just not a fancy implementation yet. Eventually there will be a part of the program that just gets a signal and goes off and does it's thing while another part does all of the complicated scheduling. I dont see a need for 2 separate controllers in this case. I suppose the argument could be made that the more complex the software, the more likely it is to have some failure that leaves the valves open for some reason.
Now in something more mission critical I would totally agree on having a separate simple controller that just does one job and another that handles the higher order algorithms. I have that kind of arrangement on my furnace controller for example. But given that this is a homebrew watering system project I'm willing to take the risk. The worse that happens is really that my lawn is over/under watered and I get a big water bill at some point. But I'm also home fairly often so I think I'd notice any big issues.
Hi,
Just to refine, my comments was regarding the way to design the system to work in the best way.
Regarding my first comment, I accept your reply - BUT this may not work in case of a slightly different relay - that does not, or may not always identify the 3.3V as logic high, or in case of system redesign to work using battery which can go lower than 3.3V during it's life cycle.
Regarding my second comment, I thought not only about reliability but also about the ability to allow the user better control of the watering scheduler.
By putting this functionality at the main controller rather than the ESP (even though the ESP is powerful enough to do the job), you can expose the watering control to the user, and give him a much better flexibility.
In addition this way you allow the main controller to run logic using sensors from other places in the system, and run predefined - but flexible scenarios (that does not need FW updates).
Using the main controller can also be used as a good, central infrastructure for logging the events and your system status / watering history.
I really love watching your videos - they are inspiring, and I love what you do !
The work you did is really nice - for the first phase, but I would not leave it as is for final solution.
If you plan on changing the design in the future according to my comments I will be happy to see part II of the video, with your changes :)
P.S.
What system do you use for the controller (On your raspberry pi and smartphone)? I saw both home-assistant and the apple "home" system...
Which one do you use?
Hey,
Totally understandable, it might not work with every relay but I bought a batch of roughly 100 of these a year or two ago so I've gotten fairly comfortable with them. But yes 3.3V might not be a high enough level for all relay modules or with batteries.
I see where you're coming from with the with the scheduler/added flexibility. I thought you meant 2 micro-controllers, one that handles the act of controlling the valves and such and then the ESP to communicate with the outside world and receives commands... my bad. I do plan already to add in a separate moisture sensor that will let the irrigation ESP know if it needs to run or not so obviously there will need to be some tweaking. I havent given it a ton of thought just yet but I think a hybrid of the two methods of control would be good. For example, I have been working on ways of easily storing long term information in the onboard flash memory, by doing this I could easily store a schedule/config file that can be updated remotely and without disturbing the program. It could have a schedule, moisture trigger values, etc. But it could also listen to commands from mqtt (and mqtt gets those commands from elsewhere obviously) that tell it to run. So basically two forms of control. I know this sounds like something that is overly complicated but it gives me the added benefit of not having to rely on mqtt/other networked devices for the irrigation to work properly. If for some reason the network goes down or the scheduling software doesnt work properly, it can still fall back to a built in schedule.
I would love to do a second phase of this project at some point but at the moment I done have tons of time unfortunately as the IKW day job has kicked into full gear but maybe towards the end of fall or next spring.
As for what I use in my daily life, unfortunately I use Apple HomeKit. It's far from the best and not very flexible but on apple devices it does work fairly consistently. Home assistant I felt was overly complicated in it's scripting (read - I couldnt get anything to ever work reliably because I'm mainly a c++ programmer haha). Right before I started IKW there was a great platform called EasyIoT but soon after the channel launched it moved to a full cloud platform and they stopped development on the local server software (and I'm just not a fan of sending my data out to a cloud server if I dont have to). I still tend to use the old local server install of that for logging of temps and such. That being said I have started the hunt for new home control/IoT platforms and I'm currently looking at Thingsboard.
Thanks for all of the ideas and support!
can you provide a relay - solenoid - power supply wiring schematic?
he laid it in front of our eyes, literally.
Nice video's , question where are you getting those AC-DC transformer (5V) and how are they doing ?
I got them from ebay (just searching for "5v dc transformer" or "5v power supply"). They've been working a treat for me! I usually need to add an extra 1000uf cap to the output whenever driving large/noisy loads like relays or motors but when theres nothing big happening they are great. I haven't had a single one fail on me yet (I've only had this batch for a year or two so it's not the largest sample period but still nothing DOA).
Thanks !, found it.
ItKindaWorks You need kickback diodes, not capacitors, but in two years I bet someone else has mentioned this.
The solenoid actuation and holding current should be different. Holding the solenoid at actuation current will overheat the solenoid.
So you have 5 volts going to you micro controller and the relays? How does that work with the voltage drop that each component uses up?
The devices are all wired in parallel so they all receive the same 5V power. There can be small dips/spikes when the relays are triggered but the capacitors are generally enough to keep the power smooth enough for the microcontroller.
Lawns are a disaster, there are much better alternatives.