BeaverBarber wrote:I'm not sure you if realize this, but you just made my point exactly...thank you. Like I said, you need to purchase one sophisticated piece of machinery to ferment in celsius to get as accurate as you can with a simple dial on a Johnson Controller in Fahrenheit.
..."If your controller is reading out in whole degrees F (i.e. 67 F not 67.0 F) like the Ranco that MoreBeer! sells, that means that the temperature could be anywhere between 66.5 F and 67.5 F (~19.2 C - ~19.7 C.) Now take the same controller reading 19 C. That could be between 18.5 C and 19.5 C (65.3 F - 67.1 F). The actual difference between those is quite small +-0.5 F vs +-0.9 F"...
...and it's still not as accurate. I close my case your honor.
Ahh quotes taken out of context... My point was that other factors in the system outweigh the difference in *precision* between controlling in C or F. If you care about that level of difference in precision you probably both need to ditch your controller and your freezer/fridge, use a controller with a more sophisticated control algorithm than these bang-bang controllers, and use a cooling system can either cycle fast or apply a variable level of cooling. And even then, after all that work, when you've got your error down to +-0.5F, you probably still won't notice a difference in the final beer as our "imprecise" controller using C with and error of +-2.3F.
As for your Johnson Temperature controller has a fixed differential of 3.5 F (2C). In addition that things entirely analog. It's not controlling based on temperature, it's controlling based off internal voltages in its circuits. The fact that the dial is in F is purely an artifact of the scale that's printed on the dial. You could print a new scale on it in C and it would work EXACTLY the same and control the the same *precision*.
I suppose we'll have to "agree to disagree" here.