Trying to design control board for many solenoids that can be programmed

Thread Starter

chris6785

Joined Jul 29, 2019
14
What I am trying to design is for a prototype for a simple project, simple in the terms that not much is needed and all should be fairly cheap.

I am trying to design (and see what it would entail and if it's plausible) a circuit board that can be produced in VOLUME or a board that can control between 20-150 solenoids each and I need to be able to program them with a fast and responsive language and if its possible, have an analog knob that can speed up or slow down the program. If this isn't possible that's fine, but it will make the programming process WAY WAY EASIER!!! I code in many languages so that part isn't a problem at all.

So here is a simple breakdown:

I need to design a board that can control a variety of solenoids, I need to be able to write a program for each solenoid OR have a board that would allow me to program them all as a whole preferably. Another words I would like the program to be able to assign each solenoid a number, class name, function name, etc... and the program will call on these functions to close a specified solenoid for a specific amount of time defined by each program that it runs. And to make the programming a bit easier and would be beneficial for the end product as well if I were able to incorporate a speed control knob so the program speed can be adjusted which in code is easy but not sure how to incorporate the button circuit into that. I have done a ton of programming, but have never programmed "hardware" from scratch so I am not sure what it entails. I have friends that can code all the way down to assembly but would like to be able to just write up a program in C or C++ that way it is more responsive and quick to make changes to. I need to know what the board would need to accomplish this (such as components, circuitry, memory (ex: type, size, printed on board or attached at different facility, etc...), solenoid connectors, power supply plug adapter for the board, and am trying to figure out specifics so I am able to communicate with manufacturers about what it is that i need produced. I would like to know what kind of power supply would work for "ALL" of the different amounts of variations of solenoids (another words I would like to get a power supply that would be sufficient for 20 solenoids and would still be enough for 150 of them as well, the solenoids I was looking at had a power draw of 3w so I would imagine it wouldn't need much, but also need power for all the rest of the board components if any. These solenoids are only going to be having water pass through them at lower pressures, if I were to guess nothing over 15-20 psi and that is shooting pretty high, but would rather get ones rated for more so I have some wiggle room in case there is a variable I am not considering at this early stage of the project. I don't know the exact pressure yet because I haven't built the prototype yet but they need to be FAST AND RESPONSIVE and have a high number of cycles they are rated for, I want them to be good quality and have the longest life possible.......within reason of course as far as affordability goes. But I would like to know how I would need to accommodate if I were to go up in pressure, would I need bigger solenoids, or is it just more power being delivered to the solenoids. All these solenoids are going to be on a solid sheet of metal where they will be stationary and the sheet metal will be anywhere between 18 inches or so up to 8-12 feet long and they will be bunched up, if possible, so they will all be very close to one another. They will be controlling water flowing through nozzles that are very close together, and if one fails it would be much easier to know which one if they were in the order they were being used. So they will be permanently attached to a piece of sheet metal another words, and don't need them water proof, they will not be near water unless line bursts or cracks. Them being as quiet as possible, also within reason for affordability is also extremely important!!! 100 solenoids going off at the same time is going to be loud I would think. The solenoids I am wanting to get would need to be able to be purchased in China due to the fact that I am going to be buying thousands of them so I would like to know what kind of solenoid would be appropriate for such a project. I also need the solenoids to be open by default, not closed, that is very important!!!!

I'm not sure if all of this can be incorporated onto a small and versatile, good quality board for an affordable price or if numerous kinds of boards need to be manufactured for each individual function of this process, but I would imagine it would be cheaper and easier to just manufacture one board due to less shipping, manufacturers to deal with, but also many be cheaper so have each board have one function, not all functions on one board.....heck if I know, that's where your guys' experience and know how comes in, but I am trying to just understand what it would entail and if affordability of manufacturing is plausible.

So in a nutshell, i just need to figure out how to make (MANUFACTURE, NOT MAKE AT HOME) a circuit board that can be if need be manufactured in quantity and made with quality that will last many many years, need to know how to power it all and have all use same power supply for cost effectiveness, And I need to know how I would go about programming a ROM or a memory module, and will need one that can be reprogrammed anytime it needed to be with a simple laptop because I would like to put many programs on each, and have them run according to a digital selector that is weather resistant if that would even be possible affordably, so I'm not sure how to integrate being able to communicate with it and pretty much just send it new programs that can be run, make changes, remove, add, edit, change times it runs certain programs, add more programs on top of what is already there, and anything in between.

I know its a lot, but I know basics but must remember I'm not making this at home and need to know just how it could be done so I can start to talk to some manufacturers about getting some boards printed and get a prototype going. This project is time sensitive, and I have all the cash flow necessary to back the project in any way needed.

I want to thank anyone in advance for their input, I hope I have enough detail, but please feel free to reply and all notifications will go straight to my phone and I will respond VERY promptly.

Thanks again and can't wait to see if this is plausible!!!
 

danadak

Joined Mar 10, 2018
4,057
Not sure I understand what conditions/state turn the solenoid on and off ?

Sensor, user pushed switch....?

In other words what, in addition to turning on/off an a solenoid, does processor
have to do ?

Latency of turning on a solenoid required ? Or rep rate needed ? Do you need
display, communications, sensor conditioning, networking.....


Regards, Dana.
 
Last edited:

Alec_t

Joined Sep 17, 2013
14,280
These guys may have some hints about water-valve array control if you contact them.
I've seen the show (very impressive) and peered into the 'works'. If I recall correctly there are hundreds of solenoid valves in their system.
 
Last edited:

MaxHeadRoom

Joined Jul 18, 2013
28,617
There is an old thread on another forum here that may be of help, detailing a water fountain build.
BTW, when writing a long description, it pay's to use paragraphs etc, for easier reading! ;)
Max.
 

Picbuster

Joined Dec 2, 2013
1,047
What I am trying to design is for a prototype for a simple project, simple in the terms that not much is needed and all should be fairly cheap.

I am trying to design (and see what it would entail and if it's plausible) a circuit board that can be produced in VOLUME or a board that can control between 20-150 solenoids each and I need to be able to program them with a fast and responsive language and if its possible, have an analog knob that can speed up or slow down the program. If this isn't possible that's fine, but it will make the programming process WAY WAY EASIER!!! I code in many languages so that part isn't a problem at all.

So here is a simple breakdown:

I need to design a board that can control a variety of solenoids, I need to be able to write a program for each solenoid OR have a board that would allow me to program them all as a whole preferably. Another words I would like the program to be able to assign each solenoid a number, class name, function name, etc... and the program will call on these functions to close a specified solenoid for a specific amount of time defined by each program that it runs. And to make the programming a bit easier and would be beneficial for the end product as well if I were able to incorporate a speed control knob so the program speed can be adjusted which in code is easy but not sure how to incorporate the button circuit into that. I have done a ton of programming, but have never programmed "hardware" from scratch so I am not sure what it entails. I have friends that can code all the way down to assembly but would like to be able to just write up a program in C or C++ that way it is more responsive and quick to make changes to. I need to know what the board would need to accomplish this (such as components, circuitry, memory (ex: type, size, printed on board or attached at different facility, etc...), solenoid connectors, power supply plug adapter for the board, and am trying to figure out specifics so I am able to communicate with manufacturers about what it is that i need produced. I would like to know what kind of power supply would work for "ALL" of the different amounts of variations of solenoids (another words I would like to get a power supply that would be sufficient for 20 solenoids and would still be enough for 150 of them as well, the solenoids I was looking at had a power draw of 3w so I would imagine it wouldn't need much, but also need power for all the rest of the board components if any. These solenoids are only going to be having water pass through them at lower pressures, if I were to guess nothing over 15-20 psi and that is shooting pretty high, but would rather get ones rated for more so I have some wiggle room in case there is a variable I am not considering at this early stage of the project. I don't know the exact pressure yet because I haven't built the prototype yet but they need to be FAST AND RESPONSIVE and have a high number of cycles they are rated for, I want them to be good quality and have the longest life possible.......within reason of course as far as affordability goes. But I would like to know how I would need to accommodate if I were to go up in pressure, would I need bigger solenoids, or is it just more power being delivered to the solenoids. All these solenoids are going to be on a solid sheet of metal where they will be stationary and the sheet metal will be anywhere between 18 inches or so up to 8-12 feet long and they will be bunched up, if possible, so they will all be very close to one another. They will be controlling water flowing through nozzles that are very close together, and if one fails it would be much easier to know which one if they were in the order they were being used. So they will be permanently attached to a piece of sheet metal another words, and don't need them water proof, they will not be near water unless line bursts or cracks. Them being as quiet as possible, also within reason for affordability is also extremely important!!! 100 solenoids going off at the same time is going to be loud I would think. The solenoids I am wanting to get would need to be able to be purchased in China due to the fact that I am going to be buying thousands of them so I would like to know what kind of solenoid would be appropriate for such a project. I also need the solenoids to be open by default, not closed, that is very important!!!!

I'm not sure if all of this can be incorporated onto a small and versatile, good quality board for an affordable price or if numerous kinds of boards need to be manufactured for each individual function of this process, but I would imagine it would be cheaper and easier to just manufacture one board due to less shipping, manufacturers to deal with, but also many be cheaper so have each board have one function, not all functions on one board.....heck if I know, that's where your guys' experience and know how comes in, but I am trying to just understand what it would entail and if affordability of manufacturing is plausible.

So in a nutshell, i just need to figure out how to make (MANUFACTURE, NOT MAKE AT HOME) a circuit board that can be if need be manufactured in quantity and made with quality that will last many many years, need to know how to power it all and have all use same power supply for cost effectiveness, And I need to know how I would go about programming a ROM or a memory module, and will need one that can be reprogrammed anytime it needed to be with a simple laptop because I would like to put many programs on each, and have them run according to a digital selector that is weather resistant if that would even be possible affordably, so I'm not sure how to integrate being able to communicate with it and pretty much just send it new programs that can be run, make changes, remove, add, edit, change times it runs certain programs, add more programs on top of what is already there, and anything in between.

I know its a lot, but I know basics but must remember I'm not making this at home and need to know just how it could be done so I can start to talk to some manufacturers about getting some boards printed and get a prototype going. This project is time sensitive, and I have all the cash flow necessary to back the project in any way needed.

I want to thank anyone in advance for their input, I hope I have enough detail, but please feel free to reply and all notifications will go straight to my phone and I will respond VERY promptly.

Thanks again and can't wait to see if this is plausible!!!
What you want is possible.
If you want to produce quantities then prototype and certification(UL TUV ++) is mandatory.
An estimate for prototyping + certification at least 100K ++ Dollar.
If you build 200K units the charge is little.

Production depending on design and components used.
(I only will use critical components from a certified company. cheap becomes very expensive if you don't)

We do need more information like
reaction speed solenoids and impedance.
distance between controller and solenoids.
How to address the solenoids ( network , direct)
power available (or limitations).
All error conditions ( mechanical and electrical) and how to handle.

Picbuster
 

Thread Starter

chris6785

Joined Jul 29, 2019
14
Thank you guys for all the replies. I have been busy moving forward with this project having out attorney look at patents.

As I stated to all of you, I am not educated in this field. I am trying to collaborate to understand the basics of what I would need and what it would entail.

As far as communication I figured serial, comm port, not sure a cost affective port, so I would need input on that aspect.

THANK YOU for the reply Dana, and I will address your questions as thoroughly as possible.
- as far as the solenoids go, I figured the way it would work is send an electrical current to activate the Solenoid.
-what does processor have to do. Not sure how to answer that, but I would imagine it just needs to be able to hold and interpret the data coming in, and I would like to be able to plug into it with a simple laptop and send more programs (to be held and run as they are configured to do so within the software that runs the entire process.
- networking, I don't believe so, just the data that it sends to the solenoids or the laptop, so I don't believe so.
- I'm not sure what sensor conditioning is, so I am unable to answer that.


MAX:
Thank you for your reply, I looked at that forum post you linked to, that is pretty much exactly what I'm trying to design. There will be some differences about the way it functions but here are what I would change and might change the hardware.
- The solenoids will be turning on and off on smaller displays very fast and many times per second so therefore just opening and letting water fall is not the goal I am shooting for. I believe the contrast would be affected if the water is just free falling rather then tons of small drops so you are able to see the image.
- And also the way it communicates, I would love to have stored memory that would be able to hold more then one program and will be able to play in certain order, certain times, etc... So I was hoping to build this system that is able to be minimal and able to be controlled as much as possible through just sending data to the memory of the microprocessor.
- Would like to design a very basic computer that can interpret the data and have parameters of that data (the fountain program in this case) altered in any way the user sees fit. The design concept is very similar though.

PICBUSTER:
I do understand where you are coming from when it comes to production but that's a ways down the road. I would like to come up with a similar project, such as the one that max linked in his reply, and see about improving and production from there once the system is tested on small and larger scales with all the variables delt with as they come.

I do not believe this will cost much to see if it's feasible and to produce for very small products in a smaller and controllable manner. I am not wanting to produce millions of these, but assure you my company and clients have the money to back this in any way necessary that is a necessary cost to them and myself, so money is not an issue.



Thank you all so much for your replies, but have been busy with this and a few other projects and moving forward on production.

I am just trying to see exactly what would be needed and a basic idea of how something such as the one max linked would take to build. Can I learn the basics of the system and be able to build a simple prototype myself not having a professional education in this field.

I am a software engineer of sorts, and deal with many aspects of my own company and many others alike, and we all work together to come up with new ideas for one another's businesses and our own. We are a team of people who build S%? T together and make money together. That's what I'm doing here, coming up with a foundation for a system that could be integrated into many different aspects of my field and products, as well as my business partners' products.

I really appreciate the help, and will be as attentive to this post as possible!

Thank you all in advance and can't wait to read the replies to see what's possible.
 

Thread Starter

chris6785

Joined Jul 29, 2019
14
PICBUSTER:
I didn't answer all your questions and will answer them now.

I need the reaction speed to be extremely quick, and also need to find what it would take to make them or buy some that will be very very quiet.

The distance between controller and solenoids will be no further the 10-15 feet, most cases 10-12 at the very most being that the end will be the furthers and get closer as they get closer to the controller.

The solenoids will be wired and will not need any network connections to my knowledge u less there is a reason to use network connections within the system. Network connection would be a future add if the fou datiib works.

There are NO power limitations either.

As far as error conditions, I would like to incorporate something simple like LEDs to the circuits, so if an electrical connection isn't being made, the LED will be out and can see where there is no connection. So use the LEDs for constant current checking and confirmation. Can move to code something and build something that could do something better for production later, again just trying to iron out the foundation of such a product.

Thank you again for your time and expertise, your time and replies are greatly appreciated!!!

We do need more information like
reaction speed solenoids and impedance.
distance between controller and solenoids.
How to address the solenoids ( network , direct)
power available (or limitations).
All error conditions ( mechanical and electrical) and how to handle.
 

DNA Robotics

Joined Jun 13, 2014
647
From that thread Max linked, using opto-couplers on LEDs would work.
A photo transistor / diode on each individually addressable LED of a Neopixel strip to turn on MOSFETs.
They are fast. Here is a Graphic equalizer light show infinity mirror that I made with them lately.


Someone on that thread said to display rows of pixels from a black & white bitmap picture. That would be perfect. After you get past the header, a bitmap is all data with the bottom left pixel first. Just make the picture as wide as the number of valves you have.
To control those Neopixels, just fill an array with the color of each pixel. Then "pixels.show()" sends it. Rinse & repeat.
 

be80be

Joined Jul 5, 2008
2,072
They use the vales on that for water falls there fast you could set 1000 up and paint by the water drops
We use them in carwash with led lights it makes what looks like lava but the company that sales them one of there guys said they been used in water falls that make it look like you painted a pic there really fast you can close by the water drop.
 

Thread Starter

chris6785

Joined Jul 29, 2019
14
They use the vales on that for water falls there fast you could set 1000 up and paint by the water drops
We use them in carwash with led lights it makes what looks like lava but the company that sales them one of there guys said they been used in water falls that make it look like you painted a pic there really fast you can close by the water drop.

Where can I find information about how the mechanism and electronic communications work? That's pretty much exactly what I am trying to design, but would like to be able to write new custom programs and upload them via laptop to individual systems.
 

Sensacell

Joined Jun 19, 2012
3,432
So far what you have posted has a bit of a 'scattershot' feeling.
I would recommend writing your own specifications document for the project.
It's super helpful as a way of gathering your thoughts and refining the ideas.

Plan on re-writing it many times.

Try to design from the customer's point of view, at the top level.
Block diagrams are essential- try to define the system at a macro level.
Define the "use model" - where is it installed, who uses it? how and when does the content change?
Concentrate on the "system modules" Break it down into function blocks, for easy design, installation, and service?
 

Thread Starter

chris6785

Joined Jul 29, 2019
14
I know exactly what I am wanting to build and how to achieve it, but am learning every day what exactly is needed to achieve that. I can build the Solenoid blocks and so on, just needing help with the electronic side and how I can put a module together that can hold the data to run the programs.

I'm assuming this is a bit hard for everybody to answer because I'm not an electrical engineer, I'm a business person and a code monkey on the side for fun and for projects myself and friends do together.

I'm sorry for not knowing the proper lingo to express what I'm needing, but am wanting to build a waterfall control system that will display images, this system and be integrated into other systems for different functions, but that is essentially what I'm attempting to build from ground up.

If I could get some help understanding how I can learn about what would be needed, or how I can learn what will be needed for this project.

I'm not looking for a ton of easy answers, I'm wanting to put the work into it to learn about the system and how it works from start to finish, just need some help on the education side of things and that's where me coming here comes in. Seemed like you guys are beyonfmd knowledgeable and was hoping for some guidance!

Thank you all for your replies and I look forward to reading again.
 

Sensacell

Joined Jun 19, 2012
3,432
To really move forward efficiently with electronics design, you need to have a detailed specification.
If you want something cheap, fast and good, the onus of project management is on you, poorly defined ideas slow everything down and waste tons of money.

Write it all down, even if you don't know exactly what it is, write down your fantasy.
This document is the bedrock of design management.

I see lots of adjectives like "quick" - "quiet" - "many times per second" - put numbers and units to these statements, even if they are wrong, they are better than vague. You need to get this out of your head and onto paper.

Once you have this started- ask objective people to read it and provide feedback, what seems painfully obvious is often not so to others.
 

Thread Starter

chris6785

Joined Jul 29, 2019
14
To really move forward efficiently with electronics design, you need to have a detailed specification.
If you want something cheap, fast and good, the onus of project management is on you, poorly defined ideas slow everything down and waste tons of money.

Write it all down, even if you don't know exactly what it is, write down your fantasy.
This document is the bedrock of design management.

I see lots of adjectives like "quick" - "quiet" - "many times per second" - put numbers and units to these statements, even if they are wrong, they are better than vague. You need to get this out of your head and onto paper.

Once you have this started- ask objective people to read it and provide feedback, what seems painfully obvious is often not so to others.

Will do, I will get it written out as clearly as possible and post pictures of my thoughts of how the design would work as soon as I can.

Thank you for the feedback and look forward to hearing from you again!
 

djsfantasi

Joined Apr 11, 2010
9,156
I designed a general addressable device control circuit. But didn’t build it because for my specific application, there were commercial solutions available. (It was for addressable lighting devices; I ended up using neopixels).

However, the process I took may be instructive. I hope you find it so.

μProcessor Circuits
I needed 15 devices, so my address fit into a nibble. For 150 devices, you’ll need a byte.

I used a μprocessor (Arduino Nano). Three pins were used to output the address through a 74595. Adafruit has a tutorial for this chip. The three pins are Data, Clock and Latch.

A fourth pin represented the status of the addressed device (like a relay). Low meant off; High meant on. I used the 74595 Latch output to also mean that the data is ready. It was delayed by a couple of inverters in series to allow addressing to complete.

Address Bus
The addresses of the devices are sent through buffers, so that the address bus can drive all devices. The buffers I used were CMOS logic chips; with as many as 15o devices, I’d use transistor buffers. Divide the buses into “banks” and use one set of buffers per bank.

The same consideration was used for the data line.

Devices
You will need to invert bits in the address. (A) CMOS inverter chip(s) were my option. Depending on your buffer design, you may also need to invert the data and Latch lines.

The address lines were fed through AND gates. There were jumper pads which allowed you to invert/passthrough each address bit with a solder bridge. If the address is active Low, you’d invert each bit in the address before passing it on the the AND gates.

The output of the AND gate enables or disables the data line. There is one more quad input AND gate. One input is tied High. The second is the data line. Third is the Latch Pass it through a latch (an T flip flop). And the last is the Address select.

Operation
You clock the address into the 74595. You set the data line and finally assert the Latch.

The addresses device decodes the address and latches the Enable/Disable into the output.
 

be80be

Joined Jul 5, 2008
2,072
If your wanting to print with water drops ill post you a link the values we use are the way to go after all someone did the hard work.
They have power block that you control from in are case PCL.
But a micro would work just as well.
Youd use a expansion board
The thing about these valves is the contol you can drop a drop by drop.
Like this https://images.app.goo.gl/ziDuxCWC61e7fWN16
 

Thread Starter

chris6785

Joined Jul 29, 2019
14
If your wanting to print with water drops ill post you a link the values we use are the way to go after all someone did the hard work.
They have power block that you control from in are case PCL.
But a micro would work just as well.
Youd use a expansion board
The thing about these valves is the contol you can drop a drop by drop.
Like this https://images.app.goo.gl/ziDuxCWC61e7fWN16

That would work fantastic. Do you know what that technology is called. I am not apposed to buying something from somebody who has already developed it, and integrating it into the application I need it for.

How can I learn more?
 
Top