First few steps in automation

Homebrew Talk - Beer, Wine, Mead, & Cider Brewing Discussion Forum

Help Support Homebrew Talk - Beer, Wine, Mead, & Cider Brewing Discussion Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

NoNothing

Well-Known Member
Joined
Aug 1, 2008
Messages
476
Reaction score
13
Location
Chattanooga, TN
Hey guys,

Being a massive geek (about beer and many other things) automation is one of the few things that really gets me excited. Heck, Im a software engineer, I design automated systems all day long. Though that's with per-existing hardware and whatnot. I get specs and I make things work. With this whole automated brewery thing Im at a bit of a loss since I have to make the specs.

Im still a ways off from getting near automation, I need to get a single tier brew stand first, and pumps, HERMS, etc, the list goes on and on. But I want to build all of that with automation as the end goal.

So my simple question is, what kind of controllers are you guys using to interface between your inputs and outputs and your computer. Where are you buying your valves and monitors. And most importantly. To take a system from manual to automation, what's a VERY rough guesstimate of price.
 
BrewTroller is what I'm using. Although I'm not automating every aspect of my rig, the BrewTroller is capable of full automation and can be as DIY as you want it to be.
 
brewtroller looks awesome!

Though I was really looking to write my own software (being a developer by trade) and make a schnazzy touch screen interface (ala DeepSix V2.0). Im not seeing brewtroller support external control.

Sure a full blown computer is a bit much for this...oh well.
 
Do you work with Java and SQL?, how much time do you have on your hands and how much do you want to spend for hardware. I have an application written in Java that uses SQL tables for a completely automated system that might work as an example for the various tasks. The programming sounds simple but after spending roughly 25-30 hours a week for last year writing and testing code and hardware, it was not as easy as it sounded in the beginning.
 
I know SQL well and I know Java enough to get by, Im no expert, but I have about 4 years of experience in it.

I mainly develop in C#. How much I want to spend on hardware, I dont know yet to be honest. I'd love to keep it under a grand but I still have a lot to build before I get to automation. I still need a damn brewstand and a pump (next paycheck? :D).
 
That BrewTroller looks sweet ... I had been looking to do something with my Arduino ...

I want to look at the images on the BrewTroller site, but you have to register, and it currently says registration is disabled by the administrator

Bummer
 
While the Brewtroller is a nice low cost platform, it is not suited for more than simple programming and functions. It would be better in the long run to start with an ARM 9 platform with enough memory, ethernet port,USB port,232 ports, touch screen display, and SD card memory expansion that could hold tables and program. This way you can buy off the shelf I/O hardware and plug it in, not spend time building interfaces.
 
I guess it really depends on the level of equipment your looking for.

Brewtroller can Literally automate the entire brewing process from beginning to end.... a tad bit more than "simple programming and functions" I'd say.

But again, it depends on your own desires. If your looking to build the equipment, do all your own coding, and use whatever parts you wish want.. go nuts!
 
AMX for me.
I used to be a control system programmer in a past life (AMX and Crestron), and they are cheap! The programming software is free, and if you're any good with Java or any flavor of C, the syntax is a short putt.
$100 gets you a controller with (6) RS-232/422/485 ports, (12) relays, (8) IR/1-Way Serial ports, and (8) Digital I/O ports (LM34's, pressure sensors, etc).
If that's not enough, you can slave as many Axcent 3's as you want off the main controller, and get 6 more of each, for about $20 :D

Touch panels are $20-$100, and you can write web-interfaces for free.

I'm not all grain yet, but I'm in the process of writing the code for "automating" my keezer build. I have to find a way to control the chiller pump and fan, so I might as well use it to replace the Ranco too right?!? Temp probes in the trunk line and tower will control the pump and fan, and a probe in the keezer will control compressor power. All with adjustable calibration, thresholds, email/SMS alerting, and web control/monitoring. That, and it'll tie into the rest of my controlled stuff around the house (Lighting, theater, and zone audio) All for about $200...:D
 
My current plan is to use a smaller 16/32bit MCU with other support hardware to interface to lower level devices like temperature sensors, motor drivers, liquid level sensors, etc. Then that MCU will most likely interface to a small form factor PC.

I am sure the nitty gritty could be done from a PC but I don't have that kind of knowledge, most my experience is in the embedded world. Plan to buy a USB touchscreen kit and retrofit an existing LCD I have just laying around.
 
brewtroller looks awesome!

Though I was really looking to write my own software (being a developer by trade) and make a schnazzy touch screen interface (ala DeepSix V2.0). Im not seeing brewtroller support external control.

Sure a full blown computer is a bit much for this...oh well.

The next version of Brewtroller will be hosted on a more robust platform maybe a ARM9. I am sure Matt (the Brewtroller software guru) would welcome the help from another software stud when he get to working on the next platform. Kevin (Kladue) is also got something pretty nice. Kevin is also the resident process control genius around here. If it were not for Kevin, a lot of us would still be scratching our heads.
 
AMX for me.
I used to be a control system programmer in a past life (AMX and Crestron), and they are cheap! The programming software is free, and if you're any good with Java or any flavor of C, the syntax is a short putt.
$100 gets you a controller with (6) RS-232/422/485 ports, (12) relays, (8) IR/1-Way Serial ports, and (8) Digital I/O ports (LM34's, pressure sensors, etc).
If that's not enough, you can slave as many Axcent 3's as you want off the main controller, and get 6 more of each, for about $20 :D

Touch panels are $20-$100, and you can write web-interfaces for free.

I'm not all grain yet, but I'm in the process of writing the code for "automating" my keezer build. I have to find a way to control the chiller pump and fan, so I might as well use it to replace the Ranco too right?!? Temp probes in the trunk line and tower will control the pump and fan, and a probe in the keezer will control compressor power. All with adjustable calibration, thresholds, email/SMS alerting, and web control/monitoring. That, and it'll tie into the rest of my controlled stuff around the house (Lighting, theater, and zone audio) All for about $200...:D

I just took a look and the Axcent 3's you talk about are obsolete and the suggested replacement is $900.
 
I have built with obsolete Opto 22 G1 series boards and modules, and and older Advantek touch screen panel Pc for program execution. If some one wanted a lower cost hardware platform, the automation direct hardware would be a starting place. Build a base controller and add I/O cards as you expand your system, run the program from one of the ARM9 touch screens or a laptop. If I were to write additional hardware communication classes for different hardware I would have an application with most of the bells and whistles for automated brewing.
 
I’m still somewhat torn on what way to go. About 4 months ago when I started planning this whole thing out in my head the Labjack was going to be my device of choice. I like you am a software engineer that writes software primarily in .Net (C#/VB.Net). My main issue with the BrewTroller is I don’t want to write procedural software using C on a micro controller. I want classes, objects, external storage and a sweet GUI to drive the whole thing.

From what I understand I can use the BrewTroller with BrewTroller Remote to have the PC be the brains of the automation. I’d like some additional input from people that understand BrewTroller Remote better as I don’t think the Arduino, what the BrewTroller is built on, was designed to be a Data Acquisition Device that is driven by a PC. The Labjack is a Data Acquisition Device that will allow for direct automation from a PC to whatever electronics you’re looking to control.

My main reason for wanting to go the route of the BrewTroller is the community. There are a number of people on these boards and the BrewTroller boards that have already solved a ton of the problems your/I’m going to run into and I’d like to tap into that knowledge rather than figure this all out myself. I also haven’t ruled out the board kladue is using, I just haven’t researched it much.
 
I looked at the lab jack originally but needed the ability to handle many more points than a single unit could. I went with the Opto gear because it was cheap and had industry standard plug in interface modules for inputs and outputs. I work in industrial instrumentation and the Java and SQL programming has been an aquired skill that came from this project. The newer generation Opto22 gear has brain modules and cheap software for building a control application, but the series I am working with is mostly just I/O hardware. I have built GUI's for data input and monitoring which run on the touch screen PC that controls the hardware, and store permenant information in SQL tables. Hardware can be found on Ebay if you know what to look for and wait for the right price.
I am about to start the conversion of the Java code to C# and then rework the GUI's to run on an ARM9 platform to see if it will replace the Advantech panel PC and drive the RS 485 hardware directly.
 
I looked at the lab jack originally but needed the ability to handle many more points than a single unit could. I went with the Opto gear because it was cheap and had industry standard plug in interface modules for inputs and outputs. I work in industrial instrumentation and the Java and SQL programming has been an aquired skill that came from this project. The newer generation Opto22 gear has brain modules and cheap software for building a control application, but the series I am working with is mostly just I/O hardware. I have built GUI's for data input and monitoring which run on the touch screen PC that controls the hardware, and store permenant information in SQL tables. Hardware can be found on Ebay if you know what to look for and wait for the right price.
I am about to start the conversion of the Java code to C# and then rework the GUI's to run on an ARM9 platform to see if it will replace the Advantech panel PC and drive the RS 485 hardware directly.

I think that's the key. With RS485 you can talk to MODBUS Modules. There are hundreds of MODBUS modules out there from many different manufacturers with functions like relay switching, rtd interfaces, A/D etc. for not too much money. By adding these modules you can grow your system easily. With embedded Unix you could choose almost any language to write your code. Another advantage of the ARM9 is the ability to support a real display instead of a simple alpha numeric display and make use of the touch screen.
 
I just took a look and the Axcent 3's you talk about are obsolete and the suggested replacement is $900.

Yes they are! That's why they are $20.00 on eBay ;) Our Cornelius kegs are obsolete too... That's why we use them! There are thousands of Axcent 3 controllers out there, and they are not going away any time soon. But, if that's a problem, you can always buy 2 or 3 spares for $60...

You need the newer master controller, which you can get for as little as $100 or so on eBay. That'll run the newest programming. The Axcent3's can be slaved off the new controllers and simply used for their IO ports.

To each his own :mug:
 
Lots of info to feed through here!

My original plan when I was sketching out my goals (to please SWMBO) was to go with the labjack like the Deep Six v2.0 system (DROOL). The brewtroller is very nice, and while I could just buy that and items to go with it, I really want to spend the time writing the tests, classes, code, everything, myself, that's half the fun for me. Automation with computers gets me going, I used to write bots for MMO's for fun, funded my first brew equipment purchase selling fake money to people, haha.

Ill dig more into these Axcent3's, the Opto kladue was talking about souncs really nice, with all the support of standards (a big win in my book).

My main goal for this month is to get a fully automated system parted out and priced, and then actually build an E-HERMS keggle with minimal overhead of parts that wont be used in the end system.
 
Automation with computers gets me going

I couldn't agree more. Please keep us posted as you go. I'm in the process of finishing off my stand and keggles right now, and hope to spend the better part of this upcoming winter working on the automation.
 
I couldn't agree more. Please keep us posted as you go. I'm in the process of finishing off my stand and keggles right now, and hope to spend the better part of this upcoming winter working on the automation.

Ill definitely be posting updates as I go.

Im starting to mock up a UI and business logic layer, that way I can do some coding without having any actual hardware and just add that as it comes.

But first things first, E-HERMS, Ive been reading all of the E-HERMS threads here over and over and over again.
 
I should clairify:
The Axcent3 is an old, slow, non-supported platform that is programmed with tht "Axcess" language. I brought it up simply because if you run out of IO on a modern controller, you can slave the ports on an Axcent3.

For the master controller, you need a system that will run "Netlinx" code (Very JAVA like, in fact, you can use JAVA if you want)

Look for anything in the "NI-" series. NI-700, NI-2000, NI-3000, NXC-ME in an NXI frame, etc. All of these run Netlinx, and are NOT $20.

Regardless of the model number, Netlinx programming is your regular vanilla IF/AND/WHILE/OR/NOT/PUSH/HOLD/RELEASE type stuff... Very C/Java-ish

If it's coding you're after, it's worth a look just for the development tools. The touch panel building software is very cool, and easy, and the programming software is a breeze once you get the hang of it.

Oh, you might want to take a look at Crestron too. I'm partial to AMX because I'm certified. Crestron does everything AMX does just as well, and just as cheap. (I have a couple Crestron controllers and touch panels I'd part with if someone's feeling code-ish)
 
Back
Top