Let me start with my question to open the solenoid on the 3/4". The valve is labelled DC12v & I guess that is to open the value when on. There are two spade connectors at the bottom and obviously no GPIO controller.
Being 12v and likely to burn out the 3-5v Pi, I am guessing I need and interface that keeps these two ranges separate like a DC-DC Power Module 25W ( SKU: DFR0205 ) whilst also needing a GPIO to control the valve. Have I got the right part to do this task as required.
Future: I am heading in small steps to creating a bus with multiple valves, but need to get my head around a single value first and kick lots of small goals and learn, but small successes. Yes it is a branched watering system that I will add things like a rain gague, humidity sensor, light sensor, soil meter both inside & out of a semi controlled environment.
Thanks for any direction of notice of similar projects in advance.
So a MOSFET acts like a solid-state switch, just send it a signal and turn it on and off. Why have a relay then as it seems to do the same thing. Or is that my answer to the ‘bus’ part of what I think I would like to know?
Yes, that is exactly what I need to do initially with a longer goal of using a bus, I guess.
So what do I use
(1) DC-DC Power Module 25W ( SKU: DFR0205 ) or
(2) a 5V 2 Channel Relay Module 10A ( SKU: CE05114 ) or
(3) a MOSFET ( SKU: CE04538 )
or are they all the same thing ?
Yes. You would use either a MOSFET (a solid state relay) or a mechanical relay - not both. They do the same job but have slightly different characteristics. For instance, the contacts for a relay are not polarity sensisitve, but the equivalent connections for a MOSFET are.
The ‘bus’ with multiple valves would usually be called a manifold - bus has a particular meaning in electronics, although it is functionally the same. You will use a bus to distribute the 12v supply to each of the relays… You will have one relay (solid state or mechanical) per solenoid.
The separate supplies are a good idea, just note that the total power consumption of the system will include your Pi, solenoids and any losses from the efficiencies of the regulators.
Opting for more overhead is recommended, accounting for 15W for the Pi about 3.5 for each of the solenoids (7 W all up) plus 10 W give or take for safety.
The DC-DC Power Module 25W (SKU: DFR0205) almost gets the job done but isnt able to toggle the solenoid with a GPIO. If you’re trying to keep one voltage going into your system using a couple regulators with enable pins will get you out of trouble, check out these ones:
Thanks Jeff, more stuff to ingest and understand, but this is helpful.
So can the Pi GPIO’s then be used as a manifold instead ? As in Pi > MOSFET > single Valve
Or do I need another piece of kit being the manifold ?
this rationale is exactly what I needed.
seems to be too many options and fewer rules which make it difficult to get my head around on first exposure. Also not knowing what I don’t know is frustrating as nothing else.
I have started chunking it down to get one thing understood ( & hopefully working ) at a time.
The total will be awesome, but over-whelming at this juncture.
The frustration level is high, but I ma sure that will pass.
‘Manifold’ is the term for a distribution of a fluid or gas - water in your case. It consists of one large pipe with a number of smaller pipes running off it, typically with a tap or solenoid near the connection point to control each one.
A bus is one large wire, or solid metal strip, with attachment points for multiple smaller wires that run off to their appropriate devices. So the two devices do a similar job in different environments.
Your project will have a manifold consisting of pipes and valves, and a bus arrangement to distribute the 12v supply to each of the relays and on to the solenoids that operate the valves.
I am aware of a car manifold and get the idea in principle.
I wish to be a little more granular and have zones without needing 80km of wiring to control it all.
Thus the direction of daisy chaining or a bus circuit with one command going to open zone 1 on the bus similar to 11.high followed after watering time by all. close 00.low, then at the next increment 21.high and so forth
The requirements to control that system will depend on the physical layout of the components. I have assumed that you need to control each valve independently - if that’s not the case the below comments may not apply.
If the solenoid valves are all reasonably close to the central control point then you can run a single controller with connections running to 6 relays connected to 6 solenoids. If the relays are close to the controller then you will need runs of 2-wire 12v cabling to each of the solenoids. If the relays are close to the solenoids then you will need runs of TTL-level cable from the controller to each of the relays - easier wiring but the maximum possible distance will be much less.
If you need to run a single set of wiring from the controller through each valve station in turn (the ‘daisy chain’ you refer to) the solution becomes much more complex. You will need a controller that can communicate in ‘daisy chain’ fashion. Your cabling will be a combination of data and power. At each valve you will need a version of the relay that uniquely responds to the controller command. This is an example for the I2C bus (but note that I2C is not practical for the distances you need): 1-Channel Signal Relay 1A SPDT I2C Mini Module - store.ncd.io.
Also, running data and power together requires a very robust data protocol.
If the valves are very widely separated then wireless control is a possibility, although the devices needed at each valve location become a little more complex.
Correct as each valve will need to be independent to keep the line pressure up as high as possible for each zone ( being mindful of the losses incurred of water pressure within a pipe run over a length of pipe )
Brilliant idea of having the distance to each solenoid as short as possible and just adding to the length of each zone run length. I can do that.
I shall investigate the TTL-cable you speak of as that is sureing up my confidence in being able to do this.
Running the daisy chain is what has scared me from the start, so keeping clear of it is now my preferred option.
The total run distance will not be more than 17m ( length of the glass house ) but compared to the monetary cost [ of pipe & power & controller wires ] together with the risk mixing up the circuits and electric shock in a wet environment verses having a central 700mm x 300mm controller area that can be made water proof and easily verified and just run 13mm water pipes as needed… no brainer really.
I can guess you have found my level of comfort with my ability, and I concur. Keeping solenoids close will also lend itself nicely to being built beyond a Proof of Concept on the dining table and simply relocated into the glass house.
Thanks for the lesson as it has helped me considerably thus far.
Hi James and Andre
Good idea using AC in these conditions. There is a very good reason for using AC in household wiring and in humid situations like what is very likely here. Most common wiring of this type is not tinned and if DC is used for any length of time you will find the positive connections will turn black with some form of oxidisation and the connections could become unreliable. This is due to the voltages developed across across dissimilar metal connections causing some migration of material (I think). Using AC with rapidly reversing voltages minimises this effect.
Although it would not apply in this case there is another problem using SSR devices with very light loads which nobody seems to have mentioned on this forum yet. It is of general interest so I will make it the subject of another post in the next couple of days.
I recently had a major amount of frustration talking to a technical person to describe something that is not in my realm of understanding (obviously) without being able to supply a photo. I learned from that !
It was a major test of my character to work around that obstacle.
The statistics you quote me below are not tangible and require a lot of work to start to understand, but that is the point of learning something new.
Thanks heaps for taking the time to reach out.
Also do you have a python script that runs subprocesses that I can use as a base for a series of timed of sensor feedback events. It is a simple question that is not addressed by Google.
No worries, thats what the forum is for!
In regards to the limitations its to prevent spam. The forum is powered through Discourse and Trust levels, to make more posts you have to increase your trust level by interacting with other topics - taking a look around the forum is also a great resource for learning how electronics work.
You can read more about trust levels here: Understanding Discourse Trust Levels | Blog
The subprocesses come under a few names, have a look at threading: