The BrewMonitor Lives on!

[Martin Kennedy]

Howdy all,

I'm just making my first beer in my new DIY brewing shed:

It's gonna be a gluten-free IPA around about 6-6.5%. Nom nom nom. I dusted off the BrewMonitor so it's popping out the temperature to the original PHP backend again: http://goo.gl/eYh9Cn One thing that I want to do soon is a calibration of the temperature sensor. It is currently inside the insulation I've put around the fermenter, but is still not in contact with the liquid. 

I know the liquid is currently at around 24C because I used a glass thermometer on it, but the digital sensor is currently reading approx. 21.2C. It could be an interesting problem because the reading of that sensor would probably be a function of both the temperature of the liquid and also the ambient temperature in the shed (due to dodgy insulation) (yes, this is what I think about on a Friday night now).

I'll have to pop a post up about making the shed but thought I'd give you a quick preview. Hope you're all keeping well. Drinks and/or hangout soon, depending on your locations! If you are nearby or will be home soon, then there will be some IPAs with your name on them.

M

[Vyk]

That looks class! Can't wait to taste it. GF too!!! Sweet! :-)

[Greg Jackson]

My man. That is class!

So jealous of your man shed! How much can you brew at once in it?

The temperature sensor is interesting! Spitballing to get the ball rolling. You could do a two layer temperature calibration for the system. (bear with me) 

Ideal experiment
Get a box, insulate it. Get 2 temperature sensors that work (glass for manual readings)  read box temperature and luquid temperature. Also record the digital temperature sensor reading.

Start by popping your rig in the box. You will also need a small portable heater and air con unit. Then keep the liquid at 0. (Start with ice) and vary the box temperature between 0 and 40c or something reasonable like that woth 6 or more increments across the range of temp your shed will see. Record real box temp, real liquid and digital temperature.

Then increase liquid temperature by reasonable increments (im not sure of the range of beer brewing) and repeat for at least 6 increments over the range.

Then you'll 6 datasets of 3 temperatures. Then theres a bunch of different maths you can try. If its a simple relationship you could try a linear regression to check for correlation. Theres a thing called k mode regression. You can do it with matlab, just throw the datasets at it and it spits out different functions the data could fit with their r2 values. I get a little fuzzy here tbh.

What you think?

[Martin Kennedy]

God, your logic is just so fuzzy! Cassidy's says hi!

[Michael Keane]

That is amazing Martin!!

[Greg Jackson]

I miss cassidys! Did you convince Beccy to go?

[Martin Kennedy]

Ha, no. And I was tricked into going.