Yes, yes you can.

In my previous post I outlined what I was hoping to acheive with this project. I have now completed the mark II revision and will soon have some results to taste.

The process was reasonably simple with a little electronic magic.

The first part of the project was to build a cabinet to hold the fermentation carboy. This is a 30 litre container that is used to hold the beer mixture while the yeast chows down on the fermentable sugars to produce alcohol. The cabinet is basically just a 6mm MDF box with another 6mm MDF box a little smaller on the inside. This creates a 1 inch void on each side that I filled with marine expanding foam.

Voila! A big ice-box basically.

That was the easy part.
OK onto the electronics and hardware.
In order to control the temperature, you are going to need two components.
1. a method to control the temperature
2. a method to affect change on the temperature

Part 1 was reasonably simple in theory, however in practice it wasnt so easy to acheive. I purchased 2 controllers before I found one that worked as I wanted, so that was 3 all up.
The first one that i got was a kit and allowed me to control a heater and a cooler and to acheive a pre-set temperature, however it has a few serious drawbacks. The first was that while there some degree of hysteresis (the ability to have a dead zone) the unit basically could only go from heating to cooling or cooling to heating. There was no zone where if the pre-set temperature was reached where the unit would turn off the heater or cooler.
This is a bigger deal than it seems.
You dont want your system to drive up or down to the required temperature and then when it gets there to drive back the other way. basically it would heat until it reached the set temp, then cool until it wasnt at the set temp then repeat the process adnausium.
This is not ver efficient.
Secondly, there was no easy way to set the temperature. It had a potentiometer and you dialed up a resistance basically. Not very user friendly at all.

These two things made this controller unusable for this application.

The second false start that I had at the controller involded a PID based digital controller. This was very close to what I needed however in excitement to buy something I didnt think hard enough about what i was buying. This meant that I ended up with a controller that while it could handle heating and cooling circuits, you had to choise which one you were going to use. it couldnt control heating and cooling at the same time.

Third time is a charm!
I now have the right controller and it is a beauty. You dial up a temp that you want to maintain and it switches one of two relays based on if it needs to heat or cool to get to that temp. Hysteresis is varyable for either the heating or cooling side. I have it set to 0.2 degrees C either side of the set temp. This means that if your set temp is 10 degrees C and the item that you are measuing is at 10.0 deg C then nothing is turned on. If the item gets to 10.2 deg C then the cooler is engaged until the temp gets down to 10.0 deg C again.  The reverse is also true.

So, that is the controller taken care of, but how am I going to heat or cool?
For this I decided to use a peltier device.
A peltier is a solid state device that heats on one side and cools on the other when a voltage is applied. Simply reverse the polarith and the heating and cooling sides reverse as well.

With a big computer heatsink on the top of it, and a 50mm x 50mm x 100mm aluminium bar on the bottom of it and then another computer heatsink on the bottom of that I fashioned a basic heating and cooling element.

So, there is the active hardware side taken care of.

From here on in it was relatively simple with the exception of a couple of electronic tricks.
I had to be able to handle reversing the polarity of the power to the peltier. In order to do this i needed to create a little bit of a franken-relay situation due to the way that the temp controller was made.

The controller uses a common earth on the output relay, so you only have 3 terminals. The heating circuit joins terminal 1 and 2, while cooling joins terminal 1 and 3. Unfortunately that meant that I had to make my own way of powering the peltier and ended up using the controller to through 2 double through double pole relays. In this way the common ground didnt matter and I used the relays to swap the polarity.

The tricky bit was being able to power the computer fans as they have a set polarity that they work with so that the blads only spin in one direction. This one was easy to fix as well as it only meant using a bridge rectifier to rectify the output to the peltier so that it was always in one polarity.

OK so does it work?

I have had a lot of feedback from variuos forums that the peltier wont work. Well, i can tell you that it does.
The first batch that went through the fermentinator was my test batch with a fermentation temp of 22 degrees C. It was heating most of the time as it was winter at that time., and the temps were within 0.2 deg C of 22 degrees C the whole week.

The second batch was a pilsener and it needed a low temp. It spent the first week at 15 degrees C and handled that very very easily, however I decided to go for broke and gave it 10 deg C for the remaining two weeks. It was able to keep up…. just.
When i say just, it was perfectly fine, but the tolerance was increased to 1 degree C. Lowest temp 10.0 deg C, highest 11.1 dec C.

I am happy with that considering that the ambient temps got up to about 26 or 27 degrees C.
Sure, that is not going to be able to cool 23 litres of wort to a consistent 10 deg C when it is 40 deg C ambient, but really, that would be unreasonable.

I can always add another peltier and heatsink module in there as well, that will make it easier.