Fermentation System Design - Looking for Feedback!

Hi everyone,

I’m piecing together an expandable cooling fermentation system using a BrewPi Spark, SS Brewtech conicals and a homemade cooler/glycol unit. I have 90% of my parts on hand, and am about ready to start building it, but first wanted to get some feedback on my proposed schematic and make sure I wasn’t missing anything or endangering myself. For now, I’ll just have one fermenter until the new BrewPi firmware is available, but I am building for the future and look forward to the day when I can get the 2nd conical and use the BrewPi Spark to control both! :smile:

I built an electric brewery a couple years ago, so I have some basic electrical knowledge, but I still like to get input from others with more experience before starting to wire things up and turn the power on. I’m new to setting up a BrewPi Spark in general, but also to wiring SSR’s to it and building a system like this.

At a high-level, I envision the system working like this. The cooler will be a reservoir for a cold (~25F) glycol/water mixture and will be chilled by an A/C unit evaporation coil (this is not in the diagram, but assume it exists). The A/C unit will be controlled by a separate temp control (see in the diagram below). Inside the cooler will be the evaporation coil and two pond pumps (one for each fermenter). Assuming the switch is on for the pump, each pump will be automatically turned off/on by the BrewPi based on the temp reading by the one-wire temp probe sensor in each fermenter. As the temperature rises beyond the set temperature on the BrewPi for that vessel, the pump and indicator light on the control panel will be turned on. Overall, its a fairly simple system, but I have a few specific questions to confirm before I move forward with building it. I’ve provided my diagram below, and the major questions I have are:

  1. In general, does anything in the diagram look out of place or look like it will not work?

  2. Should I assume that the power will come from the BrewPi Spark, to the SSRs, then through the switch/light and then off to the pump? I have no experience wiring anything up like this, so wanted to check.

  3. For the pumps, can I just cut the plugs off and wire them directly off the lights?

Here are the switches - http://www.amazon.com/gp/product/B0054UFAI8?psc=1&redirect=true&ref_=oh_aui_detailpage_o03_s00

Here are the lights - http://www.amazon.com/gp/product/B00HU06MWI?psc=1&redirect=true&ref_=oh_aui_detailpage_o04_s00

Here are the SSRs - http://www.amazon.com/gp/product/B009AQNXT0?psc=1&redirect=true&ref_=oh_aui_detailpage_o09_s00

Here is the pond pump I bought: http://www.amazon.com/gp/product/B0018X2XT4?psc=1&redirect=true&ref_=oh_aui_detailpage_o09_s00


After submitting that, I went through my diagram once more on my own and realized that I had the SSRs hooked up wrong. I was thinking of them as a conduit, instead of a switch. I’ve revised my diagram to only run the hot line through the SSR, and the controls are coming from the BrewPi Spark. All ground wires now go directly to the lights and then on to the pumps.

Are you sure you mean “ground”?

There are three wires in the mains supply, “Live” (or “Hot” in the US), “Neutral” and “Earth” (or “Ground” in the US). Live and neutral provide the power, and earth (or ground) is for safety.

I have built my mains control into an electrical back-box, and I decided to wire the output to a standard electrical socket. Then I can plug any mains device into the socket instead of hardwiring it to the relay (or SSR). The live wire goes through the relay and then to the live pin. The neutral wire goes directly to the neutral pin and the earth wire goes directly to the earth pin. If you have a metal box you should probably also connect earth to the box itself.

Just curious, is your switch intended to always cut power, so that the pumps will be off regardless of the state of the SSR?

Ah, I believe you are correct. I should have the neutral wire from the main power (wired off of a socket that the BrewPi and RPi plug into in my case) going to the lights/pumps instead of the ground wire. I attached a new diagram with this in place and also now have the ground running from the main power cable to the ground bolt on the box. Thanks!

To answer your question, yes I plan on the BrewPi sending a signal to the SSR no matter what, but the switch controlling whether or not the light/pump get powers. Think of it as a power override for each pump. See any issues with that design?

That looks better.

Do the pumps have a 2-wire cable, or a 3-wire cable? If it’s 3-wire you will have to connect the earth wire in the pump cable to the earth point.

And about the switches, I was just checking that the implementation matched your requirements/expectations.


The pump is three wire and will actually be in a crawl space. I am going to put the fermenters in my office and then run all the coolant lines and pump wiring through the wall and floor so they are not visible (except for the soft tubing going to the fermenters directly). It will be about an 8-10 foot distance from the control panel to the cooler so I plan to get some three-wire cord so I can run from the control box down to where the pumps are, and tie it in that way.

One additional modification I was thinking of here is to omit the receptable/socket all together and just strip the wires for the BrewPi and RPi and wire them directly to the main power. I know that is not what you did, but it seems like it may be a simpler route since I only have two devices to wire in. Any objections to that?

I think if you can avoid running pipes and wires through crawlspaces you should. It sounds like a lot more work than is necessary, although I think I can visualise why you might want to do this.

Should you perhaps put the switches next to the pumps? There would be nothing worse than discovering a problem with the pipes/pumps and have to run 10 feet to the switches to turn them off.

Regarding hard-wiring the power to the Pi and BrewPi, I expect there are local laws about it, but in my opinion a dual-outlet receptacle for these devices would be better. At least you could swap out power supplies easily, for example. One thing I forgot to mention is that my power box has a fuse in the live supply line. Immediately where the cable enters the box I installed a 20x5mm 10A fuse. This protects the unswitched outlet for Pi power, and the downstream power to the relay-controlled outlets. It means if I plug in a big load which blows the fuse then the Pi stops working, but I think that’s not unreasonable.

In my typical fermentations my heater’s duty cycle is about 5%, out of a total time between 14 and 30 days, mostly needed during the rise to a diacetyl rest. For a saison that duty cycle might be 10+% because of a desire to ferment at 75F or higher.

But many people, even those making saisons, just employ a “free rise” on the temp thanks to the exothermic energy of fermentation. So I am curious whether that’s your plan and you’re not planning for dual-stage (cool vs. hot) control?

Hi Dan,

For now I am planning to go with the ‘free rise’ approach and cool-treat only. I’m doing this for two reasons, first simplicity of design, and second, my house stays about 73F constant through the year, so I think achieving 75F through exothermic energy of fermentation is not unattainable. If however it doesn’t workout, I will rework my design and add a thermal reservoir and solenoids to the system to switch between heat/cool for each tank.

Looking at the diagram, would it matter at all to have the Switch/Light prior to the SSR’s? Then you don’t have any power running to the SSR’s when not in use and you don’t run the chance to have an issue there when out of use? I’m guessing it doesn’t matter really but any thoughts on this would help me.

The reason for this question is my build is going to be very similar to yours except added heat function and the placement of the Switch/Light in the wiring.

Last question, does it matter what switch or light you use? I have seen a lot looking at people’s builds but I am just using this to control fermenting.