Electronics?

HomeBrewTalk.com - Beer, Wine, Mead, & Cider Brewing Discussion Community.

Help Support Homebrew Talk:

socommaker

Member
Joined
Sep 4, 2009
Messages
11
Reaction score
0
Location
AZ
So I am interested into have more control of the fermentation process of making beer. I have seen many home made projects that incorporate electronics, but rarely are the the DIY methods talked about how those were done.

My interest in this is from wanting to start taking some of the variables out that can alter the taste throughout fermentation. Like temperature of primary, secondary, conditioning all the way to the pour. Not to concerned with the actual brewing yet as I am still green in number of beers and still using extract most of the time.

My thoughts are building a fermentation cabinet, which I have seen posts and have ideas. But looking for some help with temperature sensors on the carboy, in the cabinet and ambient temp. I am pretty handy and am looking at using the Arduino with a wifi or ethernet shield to generate a web page that will log all that information so that I can duplicate conditions in the future if that batch turns out excellent. Another thought I am having is using a similar platform with the Arduino creating a LCD display over a tap that says what beer by name, serving temp, and amount left.

Again I have seen bits and pieces on here, just looking for some help with a total build out. Also, fairly new to the board so don't chew my head off if there is a thread somewhere dedicated to this and I missed. Thanks for your help and I will be happy to post my progress with what I find.
 
OP
socommaker

socommaker

Member
Joined
Sep 4, 2009
Messages
11
Reaction score
0
Location
AZ
Two great links. Was looking for more of a DIY, this takes the fun out of the build. I just finished up a stir plate. Sorry no pics or reviews yet on that, still waiting for the stir bar in the mail. But I was hoping to get more solder iron time in and play with electronics. Used to do it as a kid and forgot how much fun burning stuff and melting stuff was till that project. Then when it all works out, the sence of accomplishment is great. Of course the SWMBO was like, "Oooo and what does that do again? Don't they sell those already made?" ARG :off:

Anyways, I think the kegbot has promise, seems like they are working on shield for the arduino. Maybe I will see if I can offer some help in there comments. Can't wait to see what comes.

So far so good, but still would like to get some more input and maybe links to builds that others have done. Again my interest is just in temp sensing and volume dispensing and sending it to a database or website for record keeping. I am good with PHP and MySQL so it not the coding, just the electronics or hardware part.

Enjoy!
 

kegtoe

Well-Known Member
Lifetime Supporter
Joined
Sep 11, 2009
Messages
1,210
Reaction score
25
Location
Stevens Point
id get your brewing techniques squared away before you worry about a fermentation chamber, yikes. I brew 5 gal every two weeks for the last 10 months and my favorite i have brewed 5 times with little variation in taste even through the different seasons.
 

MrShake

Well-Known Member
Joined
Jul 23, 2006
Messages
1,313
Reaction score
14
Two great links. Was looking for more of a DIY, this takes the fun out of the build. I just finished up a stir plate. Sorry no pics or reviews yet on that, still waiting for the stir bar in the mail. But I was hoping to get more solder iron time in and play with electronics. Used to do it as a kid and forgot how much fun burning stuff and melting stuff was till that project. Then when it all works out, the sence of accomplishment is great. Of course the SWMBO was like, "Oooo and what does that do again? Don't they sell those already made?" ARG :off:

Anyways, I think the kegbot has promise, seems like they are working on shield for the arduino. Maybe I will see if I can offer some help in there comments. Can't wait to see what comes.

So far so good, but still would like to get some more input and maybe links to builds that others have done. Again my interest is just in temp sensing and volume dispensing and sending it to a database or website for record keeping. I am good with PHP and MySQL so it not the coding, just the electronics or hardware part.

Enjoy!
I'd suggest the brewtroller/fermtroller again. If your looking to DIY, thats your best bet. Good community to help. You can even order the board bare and solder it up yourself. Even after the controller is soldered, there's still the need to DIY fabricate an enclosure, wire up relays and LCD readouts and temp sensors. Its a DIYer's dream, yet within reach of a total newb!
 

bodhi314

Well-Known Member
Joined
Aug 2, 2009
Messages
75
Reaction score
1
Location
Corvallis
That touch controller is sweet - also interested what this can be used to control and how difficult writing program would be?

Subscribed. Prost:mug:
 

kladue

Senior Member
Lifetime Supporter
Joined
Dec 28, 2006
Messages
2,318
Reaction score
60
Location
Turner, Oregon
I guess the programming would depend on functions and display features you want. The hardware is easy, the programming is time consuming as you decide to add features as you go. I have built an application in Java for touch screen brewing recipe and system control, many many hours of coding and testing with an input simulator to create a working application for my system.
 
OP
socommaker

socommaker

Member
Joined
Sep 4, 2009
Messages
11
Reaction score
0
Location
AZ
Why not start with a more robust platform like this http://cgi.ebay.com/Samsung-S3C2440-ARM9-Board-3-5-TFT-LCD-Touch-Screen-/390075115551?cmd=ViewItem&pt=LH_DefaultDomain_0&hash=item5ad249e81f and add 1 wire temp sensors and a pulse output flow meter for serving measurement.
This thing looks really sweet :rockin:. That's what I would aspire to build.

The coding isn't so tuff for me. If I had to have an order of hobbies, programing would be numero uno, then brewing, then DIY. So this hole project would boil down to using all three hobbies. I feel I could be led into a never ending loop and more than likely this is way more than I can chew. I think I might just start with getting one of the hardware pieces already done up and just work on the programing of a few items and not include everything I want on the first build out.

Again thanks for the wonderful feedback and suggestions. :mug:
 

CodeRage

Death by Magumba!
Joined
Aug 22, 2007
Messages
2,209
Reaction score
68
Location
Melbourne, Fl
kladue, I've got one of those coming in the mail with hopes of installing Android and making some easy to use interface libs using their SDK. Should be a fun project. Did you ever order one?
 

Ethos

Member
Joined
Mar 9, 2010
Messages
7
Reaction score
0
Location
Oregon
I built something pretty similar to what you are talking about. You can read about it here. You will need to scroll down a bit for the writeup on the hardware. You can see what it is outputting if you go here, but keep in mind that it is a work in progress.

The blue lines on the graph are readings from a photointerruptor that is rigged to detect airlock activity. It doesn't work very well yet. I didn't get the sensors hooked up until after this batch was in the fermentor for awhile so I missed the more interesting data.

Right now it is set up to record every 5 seconds. Every 15 minutes it uploads a batch of measurements to a server where they get imported into sql by a cron job. I am using google's chart api to render the graph. I don't recommend setting the data resolution to highest as it will probably take awhile to render.

I actually just got a batch of free samples from Maxim so I will probably start reporting ambient temperature and storing configuration on an eeprom after I bottle this beer on Monday.

Let me know if you have any specific questions about how to hook everything up.
 

kladue

Senior Member
Lifetime Supporter
Joined
Dec 28, 2006
Messages
2,318
Reaction score
60
Location
Turner, Oregon
After firing the panel PC for testing today and having hardware/software issues it looks like it is time to migrate to a newer platform. Did some loop testing/tuning from laptop with new boiler installed, but water tank developed a leak in a solder joint and now I have to tear it out to repair before resuming testing.
 

Sawdustguy

Well-Known Member
Joined
Mar 2, 2009
Messages
2,701
Reaction score
43
Location
Manorville, New York
kladue, I've got one of those coming in the mail with hopes of installing Android and making some easy to use interface libs using their SDK. Should be a fun project. Did you ever order one?
I had a feeling you were going to buy one of those. :D Ya think that will be the next brewtroller platform.;)

At that price I may order one also. As it turns out my nephew teaches a course on mobile application development for Sacred Heart University. Maybe he can teach Uncle Guy a little. He's drank enough of my beer over the years. Payback time has arrived.:p
 

gromitdj

Well-Known Member
Joined
Feb 7, 2008
Messages
1,546
Reaction score
528
Location
Orange County
If I could make a suggestion, I would recommend that you get an arduino & some sensors and start messing with it. That is what I did to accomplish exactly what you are trying to do (minus the serving aspect). I started with an Arduino Mega knockoff and 3 DS18B20 sensors and wrote the code to control a fridge for fermentation with no programming experience.

If you are interested, I will start a separate build thread with more details. Although, I am not finished yet because I want to find a way to make it permanent without devoting the Arduino to the project. I want to be able to use it for other projects and besides, it seems wasteful with 45 I/O pins not being used on the thing.

A bit of warning though. This thing has already cost me about twice what a Love Controller would have. But with a Love Controller, I wouldn't have learned to program the Arduino and I couldn't say that I built it.
 

kladue

Senior Member
Lifetime Supporter
Joined
Dec 28, 2006
Messages
2,318
Reaction score
60
Location
Turner, Oregon
Not to disparage the Arduino platform, it seems like bringing a moped to a motorcycle race, yes it can accomplish the task if you spend a lot of time working to make it do the things that it was not meant to do. Might as well start with a more capable platform that will be easier to connect to the world out of the box, and interfaces with I/O hardware built by the robot builder crowd.
 

Sawdustguy

Well-Known Member
Joined
Mar 2, 2009
Messages
2,701
Reaction score
43
Location
Manorville, New York
If I could make a suggestion, I would recommend that you get an arduino & some sensors and start messing with it. That is what I did to accomplish exactly what you are trying to do (minus the serving aspect). I started with an Arduino Mega knockoff and 3 DS18B20 sensors and wrote the code to control a fridge for fermentation with no programming experience.

If you are interested, I will start a separate build thread with more details. Although, I am not finished yet because I want to find a way to make it permanent without devoting the Arduino to the project. I want to be able to use it for other projects and besides, it seems wasteful with 45 I/O pins not being used on the thing.

A bit of warning though. This thing has already cost me about twice what a Love Controller would have. But with a Love Controller, I wouldn't have learned to program the Arduino and I couldn't say that I built it.
I already have a Brewtroller which is Sanguino based (Arduino). The guys over at the Brewtroller site are looking to step up to the next level with an embedded unix platform (ARM9) to support Modbus modules via RS485.
 

gromitdj

Well-Known Member
Joined
Feb 7, 2008
Messages
1,546
Reaction score
528
Location
Orange County
Currently the sanguino is based on the atmega 644p chip. My mega is using the 1280. More pins, more memory. The brewtroller is an excellent option. My recommendation was for the OP, I thought he was looking for more of a DIY aspect.

Kladue, what you are doing is in a differant league. I am not ready to even attempt something so ambitious, The arduino is an excellent learning tool for me, and I would recommend it as such.
 

kladue

Senior Member
Lifetime Supporter
Joined
Dec 28, 2006
Messages
2,318
Reaction score
60
Location
Turner, Oregon
Yes the system I have is in a different league, but it would be easier to start with a more capable platform able to run multiple programming languages and operating systems. You would not need to build something that does everything but it would be better to have a platform that would be capable of accomodating enhancements without having to change hardware as you want to add features. Possibly basic building blocks could be developed for this platform that would make program development for various uses easier.
 
OP
socommaker

socommaker

Member
Joined
Sep 4, 2009
Messages
11
Reaction score
0
Location
AZ
More power = bigger system! I like your thinking. Start with a big rig for the microcontroler, then build the brewing system around that. Hmmm... Don't think I can get approval for that. Was pulling teeth today, just to get a fridge for kegs. Only spent $40 bucks, re-enameling it as I type (black on black, used to be that puke yellow color). I will have to figure out how to post pics and start a new thread to show it off.

I will take advice any which way I can get it. I almost went with the brewtroller today, but pulled back. Started looking at doing something from scratch again. Damn, it just seems daunting. But I like the idea of it being what I want and not over powering.
 

Sawdustguy

Well-Known Member
Joined
Mar 2, 2009
Messages
2,701
Reaction score
43
Location
Manorville, New York
Currently the sanguino is based on the atmega 644p chip. My mega is using the 1280. More pins, more memory. The brewtroller is an excellent option. My recommendation was for the OP, I thought he was looking for more of a DIY aspect.

Kladue, what you are doing is in a differant league. I am not ready to even attempt something so ambitious, The arduino is an excellent learning tool for me, and I would recommend it as such.
The ATMEGA1284P-PU is pin for pin compatible with the 644P-PU. When the ATMEGA1284P-PU becomes available on a wider scale it's take the 644P-PU out and insert the 1284P-PU. Right now it is difficult to find an ATMEGA1284P-PU. The ARM9 is alot more versatile and acording to the Brewtroller dudes, it will probably be the next platform for Brewtroller because of embedded unix, expanded memory, expanded I/0 and the ability to interface to a touch screen display instead of a text display.
 

shushikiary

Well-Known Member
Joined
Apr 5, 2010
Messages
162
Reaction score
6
Location
denver
The ATMEGA1284P-PU is pin for pin compatible with the 644P-PU. When the ATMEGA1284P-PU becomes available on a wider scale it's take the 644P-PU out and insert the 1284P-PU. Right now it is difficult to find an ATMEGA1284P-PU. The ARM9 is alot more versatile and acording to the Brewtroller dudes, it will probably be the next platform for Brewtroller because of embedded unix, expanded memory, expanded I/0 and the ability to interface to a touch screen display instead of a text display.
Agreed. Though things get a lot more complicated with an ARM... at work we use dual core ARM cortex 4's for our hard drives (in the next line up the current 7200.12 drives use an ARM 9E) and the thing has a L1 cache and a ton of other crap that causes us a head ache sometimes if your cache is not set to write threw and the hardware wrote something to SRAM while you read it and got what was in the L1 cache instead of out there, not to mention parity protection and ton of other crap.

Sure it's faster and more versatile, but there are trade offs as always. I think they'd be able to pull off using one though.

Oh a side note, I almost have the code for the SD card inteface of the brew logger done ;).
 

Sawdustguy

Well-Known Member
Joined
Mar 2, 2009
Messages
2,701
Reaction score
43
Location
Manorville, New York
Agreed. Though things get a lot more complicated with an ARM... at work we use dual core ARM cortex 4's for our hard drives (in the next line up the current 7200.12 drives use an ARM 9E) and the thing has a L1 cache and a ton of other crap that causes us a head ache sometimes if your cache is not set to write threw and the hardware wrote something to SRAM while you read it and got what was in the L1 cache instead of out there, not to mention parity protection and ton of other crap.

Sure it's faster and more versatile, but there are trade offs as always. I think they'd be able to pull off using one though.

Oh a side note, I almost have the code for the SD card inteface of the brew logger done ;).
That's a feature I am looking forward to. Thank you for your writing the code. I am sure your experience will come in handy when they move up to the next platform. I don't think they have even decided on a platform yet. I think they want to wait for the new software to become stable first.
 

TriangleIL

Well-Known Member
Joined
May 6, 2010
Messages
222
Reaction score
3
Location
Chicago
Why not start with a more robust platform like this http://cgi.ebay.com/Samsung-S3C2440-ARM9-Board-3-5-TFT-LCD-Touch-Screen-/390075115551?cmd=ViewItem&pt=LH_DefaultDomain_0&hash=item5ad249e81f and add 1 wire temp sensors and a pulse output flow meter for serving measurement.
I couldn't resist, took the plunge. Kladue, what else do we need to create full automation of the brew cycle in a standard HERMS 1 tier system? I plan to automate everything and develop in C#.
 

kladue

Senior Member
Lifetime Supporter
Joined
Dec 28, 2006
Messages
2,318
Reaction score
60
Location
Turner, Oregon
Why not a 2 vessel RIMS system with voltage switching to make RIMS chamber into instant water heater for strike and sparge. With honeywell wet sensors for level, and a board or two from the robot builders you could have a fairly robust system without breaking the bank. I am rewriting the Java app that runs my system to run in C++ after spending a couple days profiling and tuning without reducing the heap much below 20MB and the native code use significantly below 40MB.
 

StreetervilleBrewer

New Member
Joined
May 20, 2010
Messages
1
Reaction score
0
Location
Chicago
Why not start with a more robust platform like this http://cgi.ebay.com/Samsung-S3C2440-ARM9-Board-3-5-TFT-LCD-Touch-Screen-/390075115551?cmd=ViewItem&pt=LH_DefaultDomain_0&hash=item5ad249e81f and add 1 wire temp sensors and a pulse output flow meter for serving measurement.
That looks pretty cool. Pretty much exactly what I had in mind.

So, a quick bit about me as it's my first post. I've been brewing on the stove so far, but I'd like build a single tier automated system. I'm not 100% sure of all of the hardware (temp sensors, valves, etc) to run the system, though. Would this ARM9 device be suitible for communicating with my components or do I need an intermediate device for that? Where do you folks all go for your automation components? As you can see, I'm just getting started putting together plans. Thanks for any help!!
 

kladue

Senior Member
Lifetime Supporter
Joined
Dec 28, 2006
Messages
2,318
Reaction score
60
Location
Turner, Oregon
There are a couple directions you can go, prebuilt interface devices with USB connections, ethernet connected interfaces, or build with digital and analog interface modules that use I2C as a communications method. Temperature sensors are fairly simple with the 1 wire style being the easiest, thermistors would be second choice, thermocouples and RTD's will need an intermediate signal conditioner to interface with the analog inputs. For the flow there are pulse output flow sensors that could work with a simple R-C circuit to get an analog output, or directly connected to a digital input channel with pull up resistor and code to scale pulses into flow. It will be fairly easy to build the input/output code, then a manual operating program, but the fun begins when you go for sequenced control and control loops for temperature. For level measuring there are voltage output wet sensors that run about $120 new or you can search Ebay like I did and find 1-2 PSI dc output sensors for under $50. There is lower cost DC powered ball valves or globe valves that can be setup to do floating position control for flow control, again it will be code that takes the time to make it work correctly. Gas burner control would be fairly simple with binary operation, the same for pumping. Using opto isolators on the digital outputs will make the controller more immune to the electrical noise problems and using them to drive SSR's for all the AC loads will help prevent arc noise from relays feeding back.
 

TriangleIL

Well-Known Member
Joined
May 6, 2010
Messages
222
Reaction score
3
Location
Chicago
There are a couple directions you can go, prebuilt interface devices with USB connections, ethernet connected interfaces, or build with digital and analog interface modules that use I2C as a communications method. Temperature sensors are fairly simple with the 1 wire style being the easiest, thermistors would be second choice, thermocouples and RTD's will need an intermediate signal conditioner to interface with the analog inputs. For the flow there are pulse output flow sensors that could work with a simple R-C circuit to get an analog output, or directly connected to a digital input channel with pull up resistor and code to scale pulses into flow. It will be fairly easy to build the input/output code, then a manual operating program, but the fun begins when you go for sequenced control and control loops for temperature. For level measuring there are voltage output wet sensors that run about $120 new or you can search Ebay like I did and find 1-2 PSI dc output sensors for under $50. There is lower cost DC powered ball valves or globe valves that can be setup to do floating position control for flow control, again it will be code that takes the time to make it work correctly. Gas burner control would be fairly simple with binary operation, the same for pumping. Using opto isolators on the digital outputs will make the controller more immune to the electrical noise problems and using them to drive SSR's for all the AC loads will help prevent arc noise from relays feeding back.
Kladue,

I assume I can connect from the touchscreen device directly into this BL233 relay board? Just need to hook up the correct wires from my RS232 line to the correct solder points on the board? http://www.i2cchip.com/
 

kladue

Senior Member
Lifetime Supporter
Joined
Dec 28, 2006
Messages
2,318
Reaction score
60
Location
Turner, Oregon
Here is a I2C bus relay board that can be connected to the I2C pins on the connector plug http://www.ereshop.com/index.php?main_page=product_info&cPath=143&products_id=605. For isolated binary input here is another board for that http://www.ereshop.com/index.php?main_page=product_info&cPath=143&products_id=602. For analog inputs here is a board http://store.gravitech.us/i2c128anco.html or if you want a friendlier board with analog output http://www.bipom.com/products/us/643.html. These can be wired to share I2C bus and addressed individualy with jumpers. This is enough off the shelf I/O to handle most systems, and additional boards are available for thermocouple inputs alsohttp://www.bipom.com/products/us/319237.html(not cheap as it requires analog input board also).
 

kladue

Senior Member
Lifetime Supporter
Joined
Dec 28, 2006
Messages
2,318
Reaction score
60
Location
Turner, Oregon
After a bit more research it looks like a person could put together a control system with off the shelf hardware for flow, temperature, pressure inputs, and relay or dc output to SSR's. With basic soldering and wiring skills you could assemble a control system from the various boards and sensors out there, the schematic for complete system will still need to be built by user. The hardware is low cost per item but the complexity of the system will drive final part count and cost. The coding for I/O is not that complex, and user interface, sensor range/calibration, and control sequence code will need to be developed. The question now is what scale of system do you want to build and how much time and funds can be commited to the system build. After building a touch screen application in Java that controls a SIRIM system, the code has taken much more time than the hardware build, but it includes more features than is needed for simple process control.
 
Top