My garage (probably like many) is very hot in summer, and terribly cold in winter. Far from perfect brewing conditions. I was inspired by this bloke to do something about it. I’ve now made my own temperature controlled brewing fridge.
My solution to the problem is a control system using 1-wire components, monitored by an old laptop. Cooling is provided by the fridge, and heating by a brew heat pad.
1-wire is a series of products from Maxim that use a single pin for power and serial communications. There are a bunch of reasons that I chose 1-wire – the top one was how easy it is to add new components to the “network” of devices.
The part lists includes:
|8 Channel I/O (4 Relay Version)||4 SPDT relays and 4 opto-isolated inputs from hobbyboards.com|
|DS9490R 1-wire USB adapter||Used to connect the 1-wire network to a PC|
|3x DS18S20 Digital Thermometer||0.5C accuracy from -10C to +85C, 9-bit thermometer resolution|
|2x DS18B20 Digital Thermometer||0.5C accuracy from -10C to +85C, 9 to 12-bit thermometer resolution|
|3x JQC-3FF Relay||Subminiature high power relays|
|PC||An old Windows box to run the control software.|
|Plastic boxes||To hide the electronics in|
I had originally built my own I/O board. However something didn’t work – I think it was when I misinterpreted the pin layouts of the relays.
So instead I went with the I/O board. Unfortunately the relays on it are only rated to 1A, and the fridge would require more than that. I was able to re-use the relays I had bought for my I/O board – and this time I triple-checked the pin layout before etching the board. In the picture below, the relays are hidden underneath the green and yellow electrical tape. I included one of the digital temperature sensors to measure the garage temperature.
The 1-wire components are linked using bog-standard Cat5 cables. To get the cable into the fridge, I picked up a surface mount keystone box, and glued it to the fridge door.
Inside the fridge, there are 2 temperature sensors. 1 measures the ambient temperature of the fridge, and the second measures the wort temperature. The wort sensor is attached to the fermenter with a little CPU heat transfer paste, and is insulated from the air in the fridge with some foam covered in bubble-wrap (this can be seen in the first image).
Maxim provide SDKs for a number of languages. I spend my days writing .NET applications, so I went that way.
Currently the heating pad power is squished through the door seal. I need to make another hole somewhere so that I can get power inside while maintain a decent seal. I intend to use the fridge as a fridge when I’m not brewing.
I also have quite a few plans for new features in the control software, including an easy way to archive previous brew data, to a web services interface so I can monitor and control the system remotely.
The previous screen shot shows the first attempt of making beer (an American Pale Ale) in the fridge. Initially I started without the heating pad, instead using tubs of hot water when the wort temperature dropped. The software stopped processing once, and on another time a hardware issue cause false readings when heat transfer past grounded the pins on a temperature sensor. While a steady temperature was not maintained for this brew, the experiment was a success – it tasted like beer and will be bottled soon. Only a few small changes need to be made before I start the next brew.