Industrial Automation in the Grow Room - Why or Why not?

[merlin44]

This will be the discussion topic for grow room automation and the automation class that I intend to present.

My specific interest is to introduce the concept of industrial automation and to discuss the pros and cons of implementing these systems to control our grow room equipment.

I will be putting on a multi-session class in automation when I have completed my preparations and have a few folks willing to give it a try. There will be no upfront cost so long as you already have a Windows PC with an internet connection.

@doobienoobie @missiles @covertgrower @macgyver_stoner @BudBrother @dbrn32
@organicgrower4m @Bogleg @hoppiefrog @zparkie2 @piper @chrisgrillo @myfriendis410
@tdubwilly @nick @imSICKkid @tcowen1971 @mattybear

[merlin44]

Technically speaking, most of us already have automation in our grow rooms. We have timers to turn lights on and off. We have thermostats to turn fans/AC on and off, etc, etc.

The difference between this sort of passive, ad hoc automation and automation of the sort that I am referring to is that a “truly” automated system is likely to be programmable and may or may not be expandable. All devices are controlled from a central unit, the processor, and there is a common time base (ie Clock).

So from here on out, automation is referring to going the next step beyond using timers and thermostats. There are dedicated grow room control systems out there and they are perfect for many, if not most folks. Even over kill for most.

My focus here will be on the process of designing and programming a custom control system that is specifically designed to meet your needs.

This is not as complicated as it may seem at first glance. It may not cost as much as you might think. For less than $800 you could have some really nice hardware. Could even get by for a bit less than $400. It just depends upon what your needs and expectations are.

[merlin44]

The two main arguments against custom control systems (industrial automation) is cost and complexity.

Both are valid arguments. I hope to pull the curtain back and reveal that the complication is manageable and the costs my be recoverable with improved efficiency.

The big arguments in favor of using industrial automation equipment are reliability and flexibility.

[merlin44]

At the core of a our control system will be our Programmable Logic Controller (PLC). The programming resides in the PLC and our equipment (that which we intend to control) will be connected (via wires) to the PLC input and output (I/O) points.

[merlin44]

Just to be clear, this IS NOT the class.

I am just trying to get the conversation started so folks can determine if this something that they want to look into.

[merlin44]

There are fundamentally three parts or phases to implementing a custom control system.

The first is to determine what you need the system to do. All of you know what you want to do at the basic level. You want to turn lights on and off at the correct times each day. You may also want to turn fans on and off with the lights or you might opt for using temperature to control the fans.

The next step is to determine what hardware you must have to implement your design.

The final step is to program your system to make the hardware control your lights, fans, etc.

We intend to cover all of these topics over a few weeks time.

Example:
My little veg tent has a small space heater in it to insure that we don’t get too cold. Initially I used the temperature setting on the space heater but that proved to be a disaster! The accuracy of the thermostat on the heater is so poor that the temperature spent more time in the high 80’s than the desired high 70’s and while the heater was running, the A/C was running. Arrghhhhh…

The answer was to add proper temperature (and humidity) sensing and let the PLC determine which equipment is on or off based upon a single temperature input rather than letting each device do what ever it wants to do independently.

Now, when the lights are on and the temperature reaches 82 F, turn on exhaust. If it then reaches 83 F, turn off exhaust and turn on AC. When temp reaches 75 F (going down) turn off AC or Exhaust. If temp reaches 72 F going down, turn on the heater until the temperature reaches 78 F.

Problem solved! Temperature locked in and there is provision for if it gets out of control.
I have not done this (yet? LOL) but it is certainly possible to have the system send email notifications if certain conditions occur. An example might be high temperature warning.

Normally, you would design your system to do the bare minimum to start. Then as you watch the system and how it performs, you might see an opportunity to improve things. You might, alternatively, decide that the basic operation is all you need or want. It is entirely up to you.

Your imagination and the needs of your plants are the only limitations on what you can do with one of these systems.

I should have posted a couple of photos earlier, here are photos of my control system and user interface panel:

Wiring is a bit of a mess as I am in the process of an upgrade

This is the main screen for an overview of the system

This screen allows for UV light setpoint changes

[piper]

Whats a “Windows PC” ???

apt purge $(dpkg -l | awk ‘/microsoft/{ print $2 }’) && format c: /fs:NTFS

[merlin44]

Unfortunately the PLC programming software for this (and most automation systems) is only available for Windows. Nothing for UNIX, Apple or Android LOL

[piper]

Well, Apple is based on BSD (Unix), Android is based on Linux, Windows stole from both and still uses BSD tcp/ip stack

I moved off of windows in the 90’s (windows 95) and moved to linux

[piper]

There is actually more automation in linux than what windows offers

Most of your tv’s, your tv setbox, refrigerator’s, cars, use linux

Why is Linux Trending?

The Open Process Automation Journey Continues

In 1997 the NASA (space shuttle) got rid of windows for linux in it’s “hydroponics” tests and never looked back.

https://www.debian.org/News/1997/shuttle1

Sorry, didn’t mean to start a war, but software for automation is more linux than windows

I don’t use automation, but see many companies and the public do so.

So now I will keep my 2 cents to myself and let you Rock & Roll

[BudBrother]

What? @merlin44 Wow! You would think that Linux would have the market share.

[merlin44]

Some PLC based control systems have nothing more than the PLC and the machinery just does what it is supposed to do. There are likely switches and what not that the machine operator can use to “tell the system what to do”. For example there might be a switch to start/stop the process or you might have switch that tells the system to use veg vs flowering light schedule.

Other PLC based control systems have a user interface panel (UI) or a Human/Machine Interface (HMI). The UI and HMI are two terms for the same thing. A panel with a screen (a touch screen in our case) which has various “buttons”, “indicators”, etc configured or programmed into the panel. The panel is also programmed to communicate with the PLC.
There is separate software for programming the HMI.

Yet another type of system combines the PLC and HMI into a single unit. In this case the same software is used to program both the PLC logic and the HMI.

The separate PLC and HMI allows for much more flexibility and expandability. The combined (or integrated) units are typically fixed, which means that you can’t expand the I/O (ie the number of things that you can connect to the PLC).

There is, of course, a cost trade-off.

Deciding which route to take will be covered in the course material.
For some (maybe even, most) the answer may well be to stick with timers. That is a perfectly reasonable choice.

My aim is to provide folks with the information needed to make the best decision for them.

[merlin44]

There is yet another class of control hardware called a “programmable relay”.

Programmable relays are not as capable as a PLC but if all you want to do is turn things on and off at certain times or when certain events happen, this might work for you.

Programmable relays can cost from around $80 to several hundred dollars. You get what you pay for in terms of features and capability.

If I were running a single 4’ X 4’ tent or closet, perhaps a programmable relay is the answer. We won’t be discussing programmable relays in detail but many of the general principles apply, just much, much simpler.

[piper]

I feel bad now, I didn’t mean to hi-jack this thread, sometimes it is best to keep things to myself, my apologize

[macgyver_stoner]

No worries, you brought up an interesting point. It was valid.

I don’t think anyone feels it started any fights or anything.

I only see it allowing for some clarification about what is available for the average home consumer.

happy posting and discussing,

MacG

[Bogleg]

In your experience do most of the appliances we want to automate require analog power switches? Or at least the ability to return to the last settings if there is a power interrupt?

[merlin44]

If I am understanding your question, the PLC processor module has a battery which will retain program memory for about two years (on a fresh battery). When power is lost and later restored the program will restart using the values that were in memory when power failed. The battery should be replaced annually to insure proper operation.

Examples in my program are Flower vs Veg, temperature and humidity setpoints and number of days since flipping. They would all be restored to previous values on power cycle.
There are some values that you want to start with some known value rather than the previous value, that is easily setup as well.

The processor also has a “Real Time Clock” which continues to run on battery during a power outage. The processor will have the correct time when power is restored.

If this did not answer your question, let me know so we can try again.

[Bogleg]

My question is specifically about the appliances the controller will control. For example, a lot of dehumidifiers have digital controls, and when power is interrupted to them, they do not turn back on or, if they do turn back on, they turn back on with default settings that are different than how you want the dehumidifier to be configured. A lot of air conditioners don’t have remote thermistats, etc.

[merlin44]

Ah, I get the question.
That is the beauty of this sort of system. Many of the devices such as dehumidifiers and A/C units can be set to just run continuously. Setting the unit to run continuously you then use the PLC and the PLC inputs to decide when to turn power on to the specific device that you want to turn on. I selected my equipment to have a continuous mode to facilitate this sort of external control.

If the unit that you have does not have a continuous mode, it may be necessary to jumper out or bypass the internal thermostat or hygrometer to force continuous operation.

For the PLC to be able to control devices in this way, we must be able to sense temperature and humidity and get those values into the PLC for processing. In my case I use a Dwyer Instruments device which is called a temperature/humidity transmitter. This device sends a 0 - 10 volt signal to the PLC, one channel each for temperature and rHumidity. This 0 - 10 volt signal is then converted in the PLC (you must program this) to give us a temperature in F or C and relative humidity in percent (0 -100%).

The temperature/humidity transmitter cost me around $100. Barely more than a decent thermostatic switch which can only open and close at one setpoint.

We can then use these values to determine when equipment is turned on or off.

You mentioned that some devices have remote sensors, the remote sensors are not useful to us if we are using the PLC values to control the device. Having remote sensing does not cause us any difficulties so long as the device has a continuous mode.

The remote sensing input to the A/C unit or dehumidifier may be able to connect to the PLC output and control the unit in that fashion. The unit that I have does not actually turn off when the temperature is reached, it just quits cooling but the fan continues to run which use electricity with no real benefit. That is another reason why I use the PLC to control power to the A/C unit rather than use the remote sensor input.