Auto Pouring Kegerator

Homebrew Talk - Beer, Wine, Mead, & Cider Brewing Discussion Forum

Help Support Homebrew Talk - Beer, Wine, Mead, & Cider Brewing Discussion Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Automater11

Member
Joined
Feb 1, 2021
Messages
11
Reaction score
27
Hey everyone,

Got pretty close to finishing my COVID project, so I wanted to share my Kreezer creation and a Iittle about it. My initial goal was to create a self pouring Kegerator with a minifridge. As I went through design iterations, I decided the bigger the better and landed on using a 7 cu ft kreezer that would be able to hold 4 sixtels.

This Kegerator is fully automatic in the sense that it will pour any of the 4 taps (I only have 3 currently) through google home or an app on the phone.

I used an arduino Mega for the brains of the automation and I also used an arduino uno and thermistor to control the temperature of the freezer (think of it as a DIY inkbird)

I have a L brackets locket in place that holds the pint glass. The L brackets tilt ~45° for when the beer is poured. This tilting is completed with a small servo.

I use a distance sensor that is right beneath the tap to monitor how much beer is in the glass and when it should stop. (In the video attached, I had the pour based on time as the sensor was not completed yet, also I'm aware the tilt wasn't long enough and we got some foam issues)

Another Servo pushes the tap handle to pour the beer. Tap handles allow for the Kegerator to be fully functional even if the automation were to fail.

A gearmotor is used to move the plank the taps are connected to left and right in order to line up the correct tap over the glass.

Also inside the kreezer, I have put weight scales under each keg so that I can view how many beers are left on my phone through the app.

This was my first build and I think it turned out pretty great so far! Still some more work to be done to perfect but let me know what you guys think!

Video File showing automation - https://drive.google.com/file/d/1MNZHjYwXTwwaazMPSrKvtynU2wMgbg2E/view?usp=sharing

P.S. I live in an apartment so some of the construction was difficult but we managed to get it all done. And yes, the Kreezer is in my closet, the girlfriend wouldn't allow it out.
 

Attachments

  • 20210201_152600.jpg
    20210201_152600.jpg
    993.9 KB · Views: 228
  • 20210201_152548.jpg
    20210201_152548.jpg
    1.1 MB · Views: 249
  • VideoCapture_20210201-153134.jpg
    VideoCapture_20210201-153134.jpg
    196.3 KB · Views: 251
That's pretty awesome!

How are you measuring the volume poured? Are you timing, or do you have flow meters?

It's all actually based on the height level of the beer. A sensor is constantly measuring how much distance is between itself and the object in front of it. So it starts at the bottom of the cup and it monitors the level as it starts to fill up. Once it gets to about 3.5 inches away from the sensor, the beer should shut off.

This is actually a problem for me at the moment though, since on first pours I'm getting ~50% foam (I think from temperature issues), you'll probably know once you stop pouring if there is a large foam presence the beer level will actually keep rising even if there is a large part of foam, so this can cause overflowing issues. The main solution is to get a better pour, ~15% head would be ideal to curb this.
 
It's all actually based on the height level of the beer. A sensor is constantly measuring how much distance is between itself and the object in front of it. So it starts at the bottom of the cup and it monitors the level as it starts to fill up. Once it gets to about 3.5 inches away from the sensor, the beer should shut off.

Ah, got it. I missed that you mentioned the distance sensor.

Assuming that the foam isn't due to your beer lines not being balanced, could you combine distance sensor with a SwissFlow SF800? The flow meter would tell you how much beer you've poured and the distance sensor could tell you the status of the beer's head. You could pour a fixed volume with the flow meters, sense the height of the foam's rise and fall, and then top off with another quantity once the head height is stable and the foam has settled.

I use the SF800s with my Raspberry Pints install and they are very accurate. Since they are measuring the liquid in the line, you'll get an accurate volume measurement which takes the volume of liquid in the foam into account. This isn't true if you have foam in the lines, obviously, but then you have temperature or line length issues.
 
Ah, got it. I missed that you mentioned the distance sensor.

Assuming that the foam isn't due to your beer lines not being balanced, could you combine distance sensor with a SwissFlow SF800? The flow meter would tell you how much beer you've poured and the distance sensor could tell you the status of the beer's head. You could pour a fixed volume with the flow meters, sense the height of the foam's rise and fall, and then top off with another quantity once the head height is stable and the foam has settled.

I use the SF800s with my Raspberry Pints install and they are very accurate. Since they are measuring the liquid in the line, you'll get an accurate volume measurement which takes the volume of liquid in the foam into account. This isn't true if you have foam in the lines, obviously, but then you have temperature or line length issues.

That's not a bad idea! Could also use it as a redundant safety for overpouring if the sensor goes awry, currently I just use a timer (like it shouldn't take more than 8 seconds to pour the beer) but that also means I could be spilling for a couple of seconds if it's quicker.
 
Almost certainly a "first pour" syndrome, but too-low line resistance can magnify the classic problem, as bubbles beget bubbles.
Also, that drop height isn't helping.

fwiw, I've been running a six-pack of sf800s with RaspberryPints (classic) since 2014. They are indeed ridiculously accurate given the application,...

Cheers!
 
Almost certainly a "first pour" syndrome, but too-low line resistance can magnify the classic problem, as bubbles beget bubbles.
Also, that drop height isn't helping.

fwiw, I've been running a six-pack of sf800s with RaspberryPints (classic) since 2014. They are indeed ridiculously accurate given the application,...

Cheers!

Yeah, when I measure just the temperature of the first 2 fl ozs which is 50-80% foam (you can here it happen in the video, right as the tap opens, there's a whisp of the foam starting) its about 42-45°, so that's definitely not great. If I get rid of that first foam, the rest of the pour is more even. So that leads me to believe its the first pour syndrome. Driving me nuts to find a solution to it (that's why I have a second and third fan blowing directly onto the shanks)

As far as a balanced system, I've scoured posts and posts to make sure I'm accurate. I'm 1000ft above sea level, with a amber bock and blue moon. I have 8' of 3/16" tubing, 2nd immediate pour beer is 38°. I have my PSI set at 10-11 psi which should result in 2.5 vol CO2 (I think blue moon should be at 16 but I get a shooting beer when the PSI is that high). I'm open to any suggestions, the blue moon will always give me 50-70% foam pour after pour, the amber bock settles down after 2 pours and I can get 15% head.
 
If you haven't already, give the only beer line length calculator worth using a look. I'd say your line for the 11 psi is shorter than optimal - and definitely too short for 16 psi. And I'm guessing at higher altitude it might be good to go an extra 10-20% longer and see if that helps keep the carbonation in the beer.

fwiw, I keep a wheat beer at 15 psi while my ales are typically at 11, and I have half again as much line on the wheat as the others...

Cheers!
 
If you haven't already, give the only beer line length calculator worth using a look. I'd say your line for the 11 psi is shorter than optimal - and definitely too short for 16 psi. And I'm guessing at higher altitude it might be good to go an extra 10-20% longer and see if that helps keep the carbonation in the beer.

fwiw, I keep a wheat beer at 15 psi while my ales are typically at 11, and I have half again as much line on the wheat as the others...

Cheers!

Yeah I've messed around with that calculator, maybe I have the wrong pints/second rate because using 7 seconds per pint at 11 PSI, it equates to 5ft so I would assume 8 would be well enough, unless that flow rate is way to fast?

And I actually just got a response back from blue moon saying they recommend 14-16 PSI to keep carbonated.
 
Diplomacy in action! :D

fwiw, I allow 10 seconds for a 12 ounce pour. And have no first-pour issues - pretty much every pour pic I've ever put up on HBT was a first pour :)

Cheers!
 
i honestly have no idea what this thread is about, but i tried to folow along as best i could...and thought of this, only in robot version...

1612231178085.png
 
Hands down the best use of Google Assistant I've ever seen! I might actually consent to having a smart speaker in my house if it could do this.
 
Diplomacy in action! :D

fwiw, I allow 10 seconds for a 12 ounce pour. And have no first-pour issues - pretty much every pour pic I've ever put up on HBT was a first pour :)

Cheers!
Well I just ordered 100Ft of Tubing so I'm gonna try 15ft for the 15psi and 10 ft for the 11 psi and see how that works! Should slow down the pour
 
Impressive engineering. My only question...WHY?

Seems very Rube Goldbergy. Don’t you still have to go to the Keezer to “load” a glass?
 
Impressive engineering. My only question...WHY?

Seems very Rube Goldbergy. Don’t you still have to go to the Keezer to “load” a glass?

Obviously it doesn't save much effort. It's mostly for the cool factor and the fact that I wanted a project to work on while stuck in my apartment for months. I could have built a standard Kegerator in a week, but this took 200+ hours to design/engineer/build. Something fun to work on instead of watching netflix.
 
Hey everyone,

Got pretty close to finishing my COVID project, so I wanted to share my Kreezer creation and a Iittle about it. My initial goal was to create a self pouring Kegerator with a minifridge. As I went through design iterations, I decided the bigger the better and landed on using a 7 cu ft kreezer that would be able to hold 4 sixtels.

This Kegerator is fully automatic in the sense that it will pour any of the 4 taps (I only have 3 currently) through google home or an app on the phone.

I used an arduino Mega for the brains of the automation and I also used an arduino uno and thermistor to control the temperature of the freezer (think of it as a DIY inkbird)

I have a L brackets locket in place that holds the pint glass. The L brackets tilt ~45° for when the beer is poured. This tilting is completed with a small servo.

I use a distance sensor that is right beneath the tap to monitor how much beer is in the glass and when it should stop. (In the video attached, I had the pour based on time as the sensor was not completed yet, also I'm aware the tilt wasn't long enough and we got some foam issues)

Another Servo pushes the tap handle to pour the beer. Tap handles allow for the Kegerator to be fully functional even if the automation were to fail.

A gearmotor is used to move the plank the taps are connected to left and right in order to line up the correct tap over the glass.

Also inside the kreezer, I have put weight scales under each keg so that I can view how many beers are left on my phone through the app.

This was my first build and I think it turned out pretty great so far! Still some more work to be done to perfect but let me know what you guys think!

Video File showing automation - https://drive.google.com/file/d/1MNZHjYwXTwwaazMPSrKvtynU2wMgbg2E/view?usp=sharing

P.S. I live in an apartment so some of the construction was difficult but we managed to get it all done. And yes, the Kreezer is in my closet, the girlfriend wouldn't allow it out.
I think you did a fine job, yes better than watching Netflix. something to keep in mind you won't see many of them. good job. :bigmug:
 
Search the forum for EVA Barrier tubing (loved by @day_trippr and others) or BevSeal Ultra (my preference, but harder to work with than EVA)
Before I make the switch, I'm still using BevSeal Ultra 235 line, 19' per tap in an upright keezer. Why do you prefer it over EVABarrier line?
 
Why do you prefer it over EVABarrier line?

Price and length. I picked up a 500' roll of 0.25" ID BevSeal Ultra 235 for $104 from Installation Parts Supply. Although that was only half the price of the 4mm ID EVA Barrier per foot, I could only find EVA Barrier in thirty-ish foot lengths at the time. Since I was building a long draw draught system, this would have meant a lot of short cuts (or a lot of John Guest connectors) increasing the price per foot even more.

I had about 225' left of the 500' roll after the beer-line build, but I use it for everything cold-side (pressure transfers, filtering, gas lines, long lines from lagering freezer to keezer for no-lift kegging, etc.) as it is so easy to clean and sanitize. I'm down to 130' left after cutting various lengths for transfers.

Obviously, the beer flavor is great. I can have a light lager sit in the beer line for a week with no issues at first pour.
 
@duncan.brown Thanks for the reply!
That's surely the right line for long draughts like that.

I agree, the Ultra 235 line performs very well for beer dispensing and is easy to clean. I see some discoloration from use over the years, but it's quite minimal, and the beer from the line is as good as from the keg.

It's the handling of the wide stiff springy coils and the large bending radius that are always a challenge.
 
Back
Top