New BruControl Automated Home Brewery

[ooob44]

Background

Several years ago I built a Kal-clone HERMS system following his design pretty closely. It has worked fairly well but I think I'm ready for the next step to streamline brewday even more. One of my frustrations is having to re-arrange tubes several times through the course of the brewday such as going from heating strike water, to filling mash tun, to filling the boil kettle, etc... I want to be able to automatically route water/wort where it needs to go with minimal interaction.

What this is: 240V 30A automated HERMS control panel wiring diagram and piping layout utilizing BruControl for all automation. The system uses 5500W heating elements in the HLT and Boil kettles, two 120V chugger pumps, automated ball valves to route water/wort/CIP water, proportional ball valves to regulate sparging rate and wort cooling rate, pressure sensors to monitor rough volume of water/wort in kettles, and RTD temperature sensors for control loop feedback and general temperature awareness.

Control Panel

Black wires and terminals are Line 1 AC, red wires and terminals are Line 2 AC, grey wires and terminals are Neutral, blue wires and terminals are 24V DC, orange wires and terminals are 5V DC, and purple wires and terminals are either 3.3V DC or analog voltage/current. Dashed lines are DC common; the color only references the corresponding DC voltage of the circuit it's used for. All DC commons should be connected together through the UniShield.

The thickest lines are 10 AWG, the middle thickness lines are 18 AWG, and the thinest lines are 22 AWG.

Two and three level terminals are used throughout. Black dotted arrow lines show terminals that are connected together with plug-in bridges.

All sensors interface through the BruControl MEGA Format UniShield, specifically an Adafruit Grand Central M4 Express. The UniShield allows me to save space by using slim-format din rail relays rather than the bulky Electronics Salon relays that are popular in designs. It also can be powered by 24V and creates a 5V output so I only need a single DC power supply (24V, 20A) for the entire system.

The DC section was originally going to be in its own panel. Since I'm still renting and have limited space, my thinking was it's easier to position, mount, and later move two smaller panels than it would be a single larger, heavier one. There would be connections for AC power from the AC panel and there would be Element Enable/Control and Pump Control signals going to the AC panel. The length of the components on each din rail (both AC and DC) is comparable (less than 13in) so there wouldn't need to be much if any panel layout changes to move all four din rails into a single panel.

My main goal when it came to panel layout and wiring is to minimize the amount of wires that have to cross and be routed around the box. My drawings attempt to show actual position of components on the din rail, albeit size of components not to scale. I tried to keep components that get wired together as close as possible on the din rail. For components that are connected between top and bottom din rails, I tried to vertically align/position them as close as possible.

I have an external GFCI breaker (from a SPA panel) that I plan to reuse. I've tried searching for din rail mounted GFCI breakers but have only been able to find ones that trip at 30mA (for machine safety) rather than the recommended 4-6mA (for human safety). Does anyone know of any din rail mounted GFCI breakers that trip at 4-6mA?

The element enable signals are connected through hardware interlock relays such that if somehow both elements get enabled at the same time, neither contactor will be energized. I've seen designs where a single "enable" relay is used and another signal connected to a DPDT relay controls whether it gets routed to HLT or Boil kettle. I chose my design instead because I didn't like the idea of having a bit in the software always be in either "HLT" mode or "Boil" mode.

I decided to connect the computer running BruControl to the Grand Central hardwired through a dedicated network switch. My wireless router is a floor above the brewery system and I've read a lot of posts complaining about getting wireless components connected and troubles with updating firmware over wifi. There will be an RJ-45 port on the side of the DC panel that I can plug my laptop into on brewday.

The DC side is powered off the black L1 breaker while the pump and heating element control AC are powered off the red L2 breaker. This will prevent a pump or heating element relay from tripping the breaker and killing the whole DC section.

The two terminals immediately to the right of the UniShield convert the 4-20mA current feedback signal from the proportional valves to a voltage signal for the Grand Central to read. The three terminals to the right of those are a simple voltage divider to bring the 5V signal of the Pressure Sensors to 3.3V.

Piping Diagram

On the piping diagram, valves with a circle are plain 2-way valves. Valves with a square are proportional valves. Valves with an "L" coming out of them are manual hand-controlled valves. Those are used for anything coming into the system or going out of the system. Green circles with a "T" are RTD temperature probes.

Although the piping diagram currently shows 13 2-way valves and 5 RTD probes, the DC drawing shows connections for 16 valves and 8 RTD probes. I wanted to reserve the space and connections in case the number of valves and RTD probes increased. I will likely trim down the number of relays, terminals, and external connections in the panel when I build it to save costs.

My biggest frustration on brewday is taking the kettles down and cleaning them by hand. When designing the piping, I wanted CIP capability to add water to the kettles, circulate with PBW, and flush out of the system. I've seen others use a Brew-In-A-Bag mesh bag in their Mash Tun for ease of removing grain and cleaning and I plan to do that as well. I took inspiration for the piping layout from several designs I've found online especially this post from the founder of BrewPi: https://www.homebrewtalk.com/forum/threads/automated-closed-system-herms-layout.490798/

The piping layout doesn't necessarily represent physical layout; it's laid out for ease of visualizing flow. I will probably have the water line near the bottom by the pumps instead of above the kettles. That will take a lot more thought and planning to simplify the physical piping paths.

----------------------------------------------

I haven't worked on anything BruControl related yet. I wanted to lock in the hardware design first to know exactly what I had to control before setting up the scripts.

I welcome any and all thoughts, comments, suggestions, criticisms, or ideas for improvements!

 

[TexasWine]

I have zero comments on the electrical side of things, but I've made a career reviewing reviewing process flow diagrams and P&IDs, so this was kind of fun

Here is the LINK to my recommendations. This arrangement reduces the number of MOVs (motor operated valves) by one, but there are some enhancements, like ability to underlet, flush the HERMS coil when sparging (reduces brewhouse losses), flush the lauter piping with sparge water (further reducing losses), whirlpool the BK, use spray balls for CIP with both fresh water and pumped PBW.

When I hear 'CIP' it involves spray balls located at the top of the vessel. If not, you could eliminate some piping and one valve I've added. Also, it looks like you missed a valve where the cold water in connects to the BK. I would also recommend adding a valve to dump the exchanger backflush. I wouldn't want that going into the rest of the system.

 

[sedgington]

I have zero comments on the electrical side of things, but I've made a career reviewing reviewing process flow diagrams and P&IDs, so this was kind of fun

Here is the LINK to my recommendations. This arrangement reduces the number of MOVs (motor operated valves) by one, but there are some enhancements, like ability to underlet, flush the HERMS coil when sparging (reduces brewhouse losses), flush the lauter piping with sparge water (further reducing losses), whirlpool the BK, use spray balls for CIP with both fresh water and pumped PBW.

When I hear 'CIP' it involves spray balls located at the top of the vessel. If not, you could eliminate some piping and one valve I've added. Also, it looks like you missed a valve where the cold water in connects to the BK. I would also recommend adding a valve to dump the exchanger backflush. I wouldn't want that going into the rest of the system.

Very interested to see your recommendations because I am also thinking about doing exactly this kind of setup. However, wheb I follow your link I get this:

Could you please repost your link?

 

[sedgington]

Background

Several years ago I built a Kal-clone HERMS system following his design pretty closely. It has worked fairly well but I think I'm ready for the next step to streamline brewday even more. One of my frustrations is having to re-arrange tubes several times through the course of the brewday such as going from heating strike water, to filling mash tun, to filling the boil kettle, etc... I want to be able to automatically route water/wort where it needs to go with minimal interaction.

What this is: 240V 30A automated HERMS control panel wiring diagram and piping layout utilizing BruControl for all automation. The system uses 5500W heating elements in the HLT and Boil kettles, two 120V chugger pumps, automated ball valves to route water/wort/CIP water, proportional ball valves to regulate sparging rate and wort cooling rate, pressure sensors to monitor rough volume of water/wort in kettles, and RTD temperature sensors for control loop feedback and general temperature awareness.

Control Panel

Black wires and terminals are Line 1 AC, red wires and terminals are Line 2 AC, grey wires and terminals are Neutral, blue wires and terminals are 24V DC, orange wires and terminals are 5V DC, and purple wires and terminals are either 3.3V DC or analog voltage/current. Dashed lines are DC common; the color only references the corresponding DC voltage of the circuit it's used for. All DC commons should be connected together through the UniShield.

The thickest lines are 10 AWG, the middle thickness lines are 18 AWG, and the thinest lines are 22 AWG.

Two and three level terminals are used throughout. Black dotted arrow lines show terminals that are connected together with plug-in bridges.

All sensors interface through the BruControl MEGA Format UniShield, specifically an Adafruit Grand Central M4 Express. The UniShield allows me to save space by using slim-format din rail relays rather than the bulky Electronics Salon relays that are popular in designs. It also can be powered by 24V and creates a 5V output so I only need a single DC power supply (24V, 20A) for the entire system.

The DC section was originally going to be in its own panel. Since I'm still renting and have limited space, my thinking was it's easier to position, mount, and later move two smaller panels than it would be a single larger, heavier one. There would be connections for AC power from the AC panel and there would be Element Enable/Control and Pump Control signals going to the AC panel. The length of the components on each din rail (both AC and DC) is comparable (less than 13in) so there wouldn't need to be much if any panel layout changes to move all four din rails into a single panel.

My main goal when it came to panel layout and wiring is to minimize the amount of wires that have to cross and be routed around the box. My drawings attempt to show actual position of components on the din rail, albeit size of components not to scale. I tried to keep components that get wired together as close as possible on the din rail. For components that are connected between top and bottom din rails, I tried to vertically align/position them as close as possible.

I have an external GFCI breaker (from a SPA panel) that I plan to reuse. I've tried searching for din rail mounted GFCI breakers but have only been able to find ones that trip at 30mA (for machine safety) rather than the recommended 4-6mA (for human safety). Does anyone know of any din rail mounted GFCI breakers that trip at 4-6mA?

The element enable signals are connected through hardware interlock relays such that if somehow both elements get enabled at the same time, neither contactor will be energized. I've seen designs where a single "enable" relay is used and another signal connected to a DPDT relay controls whether it gets routed to HLT or Boil kettle. I chose my design instead because I didn't like the idea of having a bit in the software always be in either "HLT" mode or "Boil" mode.

I decided to connect the computer running BruControl to the Grand Central hardwired through a dedicated network switch. My wireless router is a floor above the brewery system and I've read a lot of posts complaining about getting wireless components connected and troubles with updating firmware over wifi. There will be an RJ-45 port on the side of the DC panel that I can plug my laptop into on brewday.

The DC side is powered off the black L1 breaker while the pump and heating element control AC are powered off the red L2 breaker. This will prevent a pump or heating element relay from tripping the breaker and killing the whole DC section.

The two terminals immediately to the right of the UniShield convert the 4-20mA current feedback signal from the proportional valves to a voltage signal for the Grand Central to read. The three terminals to the right of those are a simple voltage divider to bring the 5V signal of the Pressure Sensors to 3.3V.

Piping Diagram

On the piping diagram, valves with a circle are plain 2-way valves. Valves with a square are proportional valves. Valves with an "L" coming out of them are manual hand-controlled valves. Those are used for anything coming into the system or going out of the system. Green circles with a "T" are RTD temperature probes.

Although the piping diagram currently shows 13 2-way valves and 5 RTD probes, the DC drawing shows connections for 16 valves and 8 RTD probes. I wanted to reserve the space and connections in case the number of valves and RTD probes increased. I will likely trim down the number of relays, terminals, and external connections in the panel when I build it to save costs.

My biggest frustration on brewday is taking the kettles down and cleaning them by hand. When designing the piping, I wanted CIP capability to add water to the kettles, circulate with PBW, and flush out of the system. I've seen others use a Brew-In-A-Bag mesh bag in their Mash Tun for ease of removing grain and cleaning and I plan to do that as well. I took inspiration for the piping layout from several designs I've found online especially this post from the founder of BrewPi: https://www.homebrewtalk.com/forum/threads/automated-closed-system-herms-layout.490798/

The piping layout doesn't necessarily represent physical layout; it's laid out for ease of visualizing flow. I will probably have the water line near the bottom by the pumps instead of above the kettles. That will take a lot more thought and planning to simplify the physical piping paths.

----------------------------------------------

I haven't worked on anything BruControl related yet. I wanted to lock in the hardware design first to know exactly what I had to control before setting up the scripts.

I welcome any and all thoughts, comments, suggestions, criticisms, or ideas for improvements!

Thank you so much for posting this. I have been considering using Brucontrol to do exactly the same thing, but there are so many options, it is hard to know where to start. This is an excellent primer. A question about hardware (as this seems to be another difficult to research what works and what doesn't): Will you buy all your automated valves from Brucontrol? Will your volumetric measurement based on BrunDog's model of using a pressure sensor isolated outside the tank? Will you control sparging with an Arduino Flow Meter(s) that then adjust a automatic proportional valve to prevent stuck sparging? Any details that you can share would be greatly appreciated.

 

[BrunDog]

I agree with @TexasWine... seems like a lot of valves. I'd recommend you consider 3-way valves where you can. They are slightly more expensive than 2-ways... but you will use less overall and will save money, wiring, piping, and space. In your flow path, I also don't see how you can fly sparge unless if you are flowing drained mash water back through the CFC (which will also strand wort in the HERMS loop).

This is also why I like a 2 vessel direct sparge (RIMS) approach... reduces a vessel and the valves needed because the fluid path always goes through the RIMS tube.

 

[TexasWine]

Very interested to see your recommendations because I am also thinking about doing exactly this kind of setup. However, wheb I follow your link I get this:
View attachment 655553
Could you please repost your link?

Link is fixed now. Not sure what happened.

 

[GParkins]

I agree with @BrunDog that 3-way valves are the path to reducing complexity. I'm also going to throw him another bone and say that if I knew then what I know now, I probably would have given RIMS a harder look. There is less complexity, less equipment, and significantly faster response times for changes in temperature targets to larger volumes of liquid.

 

[TexasWine]

Any updates on this?

 

[ooob44]

@TexasWine @BrunDog @GParkins thank you for your feedback! I apologize for taking a year to get back to you; between my gf quitting her job and going back to school, the covid mess, getting engaged, and buying a house... it's been a hell of a year and I wasn't able to pursue a redesign or begin construction.

I've taken your recommendations to go with RIMS and 3-way valves to heart. Like was mentioned, less equipment (cheaper) and fewer space requirements (two kettles vs three) are what appealed to me the most in a RIMS system compared to HERMS.

I plan to spend some time over the next few months researching various RIMS implementations and adapting my previous design (both flow and electrical schematics). I also have around 53 pages in the main BruControl thread to catch up on.

 

[ooob44]

This is definitely becoming one of the longest planned upgrades ever, coming up on 2 years now! Probably still have another year or 2 until I'm able to actually start building but I promise it will happen eventually. Anyway, I've revamped everything into a 2-kettle RIMS (with cold water reservoir) setup and switched to using 3-way valves which has significantly simplified things. It's basically @BrunDog 's RIMS diagram with some added paths for output, CIP, chiller, and steam condenser.

The other major change I made was to draw the control panel wiring using the AutomationDirect ProductivityOpen Arduino hardware instead of a Grand Central with UniShield. After seeing that support was in the works for it with BruControl (here, here, and here), I wanted to see how much space I could save, especially after reducing the number of valves (and hence relays) that I needed. The DC DIN rail comes out to 15.5" wide and the AC DIN rail comes out to 16.3" wide, even after adding a pH meter and Crydom SSRs. I should be able to stack the two rails on top of each other in the same panel with plenty of space to spare. All AC IO will be out the bottom of the panel and all sensors will be out the side of the panel.

I welcome further critique and discussion while I continue to research parts and create a bill of materials.

 

[ooob44]

Been spending time drawing up BruControl screens and prototyping some scripts for automating the brew day. The background of the valves will change to show the flow direction. The trends tab just has a few graphs but I'll most likely figure that out once I build the system and start using it and see exactly what's useful to graph.

 

[BrunDog]

Clean! Me likey!!