Controlling Fermenter temperature for a micro-brewery

Hello there all you wonderful fellow nerds (or geeks, or neither if that offends you)!

I am the co-founder of Sanctum Brewing Co in Pomona, CA and I have a project that I want help with. Actually need help with. Here's the nutshell version:

1) Use an arduino to control: A) a pump, which is connected to our glycol (food grade coolant) system. The glycol is kept at around 26-35 degrees Fahrenheit. the pump recirculates the glycol around a fermenting beer to maintain its temp. B) Valves. With the glycol, the goal is to be able to open a valve for only the fermenters which need chilling. This means if three tanks were hooked up, Arduino would need to know all their temps, know when to turn on the pump, and when to open which valve. C) As I just mentioned, the unit would need to keep track of the temps of each tank, and know how to turn on the pump, open the valve for a given tank, and keep the pump running if one valve needs to close but another needs to remain open.

2) Use some sort of touch screen/lcd/encoder/button based interface to be able to change parameters quickly without the use of a laptop. I want to be able to select a tank, and change the temperature at which it should be controlled to.

3) Provide information at a glance as a default setting. Meaning: I want to be able to see the temperatures for each tank on a common screen/area.

Treat my like a 5 year old when responding. I have zero practical experience with this, but am willing to learn, so I don't know what most of you would probably consider common knowledge.

Okay, if this project sounds simple, please advise as to what to do. I'm very new to this, but fairly well-read on the subject of microcontrollers. I think I'll need some relays (or at least one for the pump). and I'd prefer to make ball valves that open using a servo myself. Solenoid valves are extremely cheap, but they're not designed to be opened for very long. And the ones that are designed to be opened long aren't cheap. Sometimes, when taking a tank from say, 67 down to 34, the valve needs to be opened for 24 plus hours. Hence, why I think it would be easier to just buy a decent $20 servo with metal gears and jerry-rig it to a brass home-depot valve (which I already have in the system, so...free to me!).

My thoughts on the interface for controlling the temps are all over the place and where I need the most help. I think I could get by just setting up the arduino to do the tasks I've laid out, but programming a screen and all that seems way, way, way over my head. I don't know anything about how they work or what they are capable of and what research I've done has been more "look how cool this is," more than "here's how to make something cool" with touchscreens.

Could I get the Android sheeld and use an old phone? Should I get a TFT and program it? Would it be possible to do a really simple LCD readout that just has a few buttons/encoders to change/select parameters? Again, this is the biggest mystery to me.

How hard is it to get a screen to change the cooling target in this setup?

I need help! I am ordering parts as I type this. I have settled on the MEGA as a safegurad for future programming needs (and if the screen takes up tons of pins we'll still have plenty leftover). I will also need 3 temp sensors (one for each tank). I will start there and see if I can get a working prototype and then add on the screen/control surface as you all advise!

Thanks so much. Any help will be rewarded with what beer I can get to you (if you're into that), if not I can at least give you a shirt or glass or something as an additional incentive to being a nice and wonderful helper to me!

What you are trying to do sounds feasible, but with all of the sensors, wiring, outputs, and user interface, it's going to be a BIG PROJECT. And, it sounds like it could be expensive if something goes wrong...

Like any project, take it one step at a time. You've got valves to control, temperatures to monitor, etc., and all of those things can be developed and tested independently before putting everything together.

There are example projects for reading temperature, controlling solenoids, using an LCD, etc. None of them will do exactly what you need for your project, but they should get you started and you'll just have to modify the hardware and software to meet your specific needs.

And if you are new to programming, write, compile, and test one or two lines of code at a time. The most common mistake beginning programmers make is trying to write the whole program at once. Then when it doesn't work, they can never find the problems. Of course, professional programmers write more than one or two lines of code at a time, but NOBODY writes the whole program before testing. There's an art to this... You can't just start at the top and work down because the compiler needs to see a "complete" program.

It might be useful and interesting to monitor the temperatures without any of the automation, so you might want to start with that.... Just a thought.

Could I get the Android sheeld and use an old phone? Should I get a TFT and program it? Would it be possible to do a really simple LCD readout that just has a few buttons/encoders to change/select parameters? Again, this is the biggest mystery to me.

I've never done anything wireless, or anything with a touch screen, but wireless with a phone interface is probably the most complicated. And, I'm not sure I'd want to rely on that as your only interface. A simple LCD with regular pushbuttons is probably the simplest (but on the hardware-side its more to wire-up than a touch screen that serves two purposes).

Hence, why I think it would be easier to just buy a decent $20 servo with metal gears and jerry-rig it to a brass home-depot valve (which I already have in the system, so...free to me!).

It's probably not "easier" than buying something that's already built and already works. You can buy motorized valves, but they are expensive (solenoid valves are expensive too).

I really think you are going to have your hands full with this project, so I'd say buy whatever you don't absolutely have to build yourself. (And, I'd be kinda' surprised if you can turn a valve with a $20 servo.)

For an "industrial" project like this, I'd buy and build (at least) two of everything. If something dies/breaks you don't want to be shut-down while you wait for parts to arrive, or while you assemble a new circuit board, or troubleshoot the old one, etc.

And, make sure everything plugs-in... You don't want to be un-soldering/re-soldering if you have to replace a servo motor.

Build a prototype system that is small enough to fit on your desk complete with tanks, pumps, sensors etc. Get that working reliably before upscaling.