Beer fridge, sensor on FV or in free air?

[Twostage]

PM me if you are going to build something and I can help more (with code and simpler hardware).

Just skim read the build thread and it looks really good. I've picked up a few of the pi mags but never bought them as I didn't really have a need/project.

I've also seen a thing on the web where you reprogram an stc+ and it can do staged, day by day temp ramp ups and downs (like in your device).

I monitor my brews progress by its temperature rather than how many days so it isn't either forced into the next stage or left too long. In general the process is :-
Set a starting temp, fermentation causes the wort to go above this for 2-4 days then cools down to starting temp.
I ramp up a degree and the worts temp ramps up more for a few days then falls back.
Ramp up a degree again, wort again goes up but then typically stabilises.
Cold crash.

So the question is could the pi be programmed to move to the next stage based on a drop of the wort temp rather than a number of days ?

(sorry if this has been answered further in your thread but I've only read the first few pages).

 

[Robbo100]

Robbo, that's what I said.... In as few words as possible!! Haha. But really, that's right on the money.

Not wanting to be pedantic (actually, that is a lie, I love being a pedant), but what you said was completely opposite to what I said. It was predominantly your comment that made me add the technical detail to back-up my claim :lol:

Getting the wort to stabilize right after making and while the yeast is at its most active. My view you'd want the sensor in the air. Get the environment stable. Once it slows down, switch to the sensor on the FV. Just my thought. But then again depends on what your fermenting.

Whereas I have suggested that you should use the sensor on the FV initially, so that the target temperature is reached as rapidly as possible, and then you move the sensor to the air, to maintain as stable an environment as possible.

Click here to see video of me

:lol:

 

[Robbo100]

So the question is could the pi be programmed to move to the next stage based on a drop of the wort temp rather than a number of days ?

(sorry if this has been answered further in your thread but I've only read the first few pages).

You can program the Pi do do anything you want. It is written in Python and can use the sensor data to react in what ever way you want.

Although, if you are regulating the outside environment, how will the wort generate a temperature change for you to detect?

 

[Robbo100]

pffffttttt, a real engineer would have done some maths, maybe used some Laplace transforms and arrived at an optimum solution.

He/she would then be overruled by the site operator though, who would just hit auto tune on the PID and wonder why it doesn't work.

I do like the idea of controlling based on the FV temp but controlling the FV-air deltaT rather than heat/cooling inputs. Are you doing that with a Rasbery Pi?

Ha ha. I was genuinely going to use a full on PID, but it wasn't worth the effort. My control theory days are well behind me, so I couldn't be bothered with learning it again and coding it all.

Essentially I am controlling the fridge air deltaT, by turning on and off the fridge or heater at the required times.

And, yes, it is using a Raspberry Pi.

 

[Robbo100]

By the way, for anyone still awake after reading this thread...

Here is another thread from this forum when I uploaded a graph of the controller parameters: http://www.thehomebrewforum.co.uk/showpost.php?p=424097&postcount=49

 

[Twostage]

You can program the Pi do do anything you want. It is written in Python and can use the sensor data to react in what ever way you want.

Although, if you are regulating the outside environment, how will the wort generate a temperature change for you to detect?

Because its alive (said with a mad professor voice). The fridge temp is essentially the base temp, the lowest the wort can get to but whilst it is active it generates enough temp to be measurable, about 1.5 degrees during fermentation and a fraction of a degree in post fermentation activity.

I control the fridge temp using the STC probe in the air but I monitor both the wort and air temp separately. I use a temper2 USB with two temp probes attached to a laptop.

Below is a graph from one of my recent brews, blue line is wort, orange is fridge. It follows my double ramp then crash method :-

I'll read your thread properly now to see what I would need.

Its a long time since I've written any code more complicated than an excel function :-)

 

[Twostage]

Been a while since my last post but I haven't been IDLE ( in joke for Python programmers ).

I've now got all the kit together, most of it tested, learned to program the Raspberry Pi and sorted out most of the routines to do the temperature measuring and control. The first time you switch a relay on by typing in some code brings you more pleasure than it should really 😀

I realised last night that what I have done so far (and planning to do) would apply just as well to controlling our ancient central heating system. So I'm doing two projects in parallel.

 

[Robbo100]

You have an enormous box surrounding your house that you can turn on to make it colder?!

;-)

 

[Twostage]

You have an enormous box surrounding your house that you can turn on to make it colder?!

;-)

😀😀😀😀

Nice one.

Hot water tank = FV ( or boiler)
House temp = fermentation chamber.

My initial thought was to emulate the wifi thermostat (which I wired into the boiler) I.e. Heating on at 6:30 until it's 21 degrees then off at 8:00 during the week but I've realised that the controller can be smarter than that if you program it yourself. Quick example - the missus wants the heating on to dry the clothes on the radiator so you have to put the thermostat on a temp where it won't turn off and hold it for a few hours. What you could do is put it into drier mode and it ignores the temp for a couple of hours and is just on.

 

[Robbo100]

@Twostage,

You have so many options, and it is a really interesting project.

If the radiator you want to use for drying cloths is closest to the boiler, you could even use some solenoid actuated valves to stop all the hot water going round the rest of the system, and only allow it to go to the one radiator.

You can also directly measure the water temperature going into your hot water tank to make sure it isn't too hot (and turn off the heat if it is), or measure the heating temperature for the same reason. Loads of options!

 

[Twostage]

@Twostage,

You have so many options, and it is a really interesting project.

If the radiator you want to use for drying cloths is closest to the boiler, you could even use some solenoid actuated valves to stop all the hot water going round the rest of the system, and only allow it to go to the one radiator.

You can also directly measure the water temperature going into your hot water tank to make sure it isn't too hot (and turn off the heat if it is), or measure the heating temperature for the same reason. Loads of options!

The heating system is old and the radiators have really stiff manual valves. I've been thinking for ages about replacing them with TRV valves but as you say solenoid ones would be a better option now.

My main design goal for this project is for the control system to be humanised and simplified as opposed to having to have rigid time based programming. So you will tell it what you want and it figures out the settings. Another goal is economy. We've learned from the repeated times we've ran out of oil that our hot water stays hot for well over 24 hours as its triple insulated so why should I reheat it twice a day ?

 

[Robbo100]

A while back, I toyed with the idea of getting some of ->these<-, ripping our the internal electronics and replacing it with an Arduino with RF or wifi transmitter.

I was then going to get some similar Arduino units, fitted with temperature sensors, powered by installing them (safely) in the light sockets) with a suitable transformer.

I was then going to gather all the information centrally, to have a far better control (room by room) of the heating in my house. Why heat the utility room when you spend all night in the lounge!

Then I realised I was best not to bother because:

a) It would probably be quite costly
b) getting it all installed would be quite tricky and take quite a lot of effort
c) the "adhoc" power solution in the lighting circuit would invalidate my household insurance if it ever caused a fire
d) my wife would kill me on all 3 of the above.

 

[Twostage]

Doing my first test run now of the Raspberry Pi based brew controller.

A lot of the algorithms and ideas are shamelessly ripped from Robbo100's code.

I re-wrote quite a bit of my original code having spent quite a bit of time learning to program Python natively rather than hacking code together.

This first test just ramps an FV full of water to 18, holds it there for an hour, ramps it down to 5, holds for an hour then back up to 18.

The ramp stages take account of the temperature of the FV contents to accelerate reaching the target similar to Robbos' code by increasing/decreasing the target air temp and recalculating as it goes.

Hold stages just keep the fridge at the target temperature.

The big difference to Robbos' code is the fermentation stage which is like a hold stage but it monitors the FV temp and looks for a rise and drop which signifies fermentation is over and then moves to the next stage.