Bubble Catcher Watches Your Booze Burp

Making your own booze involves a lot of sitting around waiting for things to happen, like waiting for the fermentation process to finish so you can get on with bottling and drinking it. That involves watching the bubbles in the airlock: once the frequency of the bubbles falls below a certain level, your hooch is ready for the next step.

[Waldy45] decided to automate this process by building a bubble catcher that measures the frequency of bubbles passing through the airlock. He did this using an optocoupler, a combination of LED and light sensor that changes resistance when something passes between them. You can’t see it in the image, but the horseshoe-shaped optocoupler is slotted around the thin neck in the bubble tube to sense when a bubble passes through.

The optocoupler is connected to an Arduino, running a bit of code that generates an interrupt when the optocoupler is triggered. At the moment, this just outputs an average time between bubbles to the serial port, but [Waldy45] is looking to add an ESP8266 to wirelessly connect the Arduino and contact him when the bubble frequency falls, indicating that the booze is ready for bottling.

We’ve seen a couple of over the top beer breweries before (here and here), but none of them have automated the actual fermentation stage, so something like this would definitely be an addition. Cheers!

39 thoughts on “Bubble Catcher Watches Your Booze Burp

  1. This is based on a false premise.

    There is way more to fermenting alcohol than CO2 production. Using bubble frequency is not a reliable determiner. Without taking a specific gravity measurement (more reliably, multiple over a period of days) you can’t be sure fermentation is actually done.
    Bottling before fermentation leads to inconsistent results, ‘gushers’ and the threat of bottle bombs.

    Don’t use bubbles, use a hydrometer.

        1. If you bottle the brew while it’s still bubbling, you get suspended dead yeast and a thick sediment in the bottle. You have to let it sit a couple days anyways unless you like cloudy beer that jumps out of the bottle similiar to the mentos effect.

          That’s why carbonation was considered a flaw in champagne originally – it blew up the sediment in the wine bottle. Then the champagne method was developed where they freeze the neck of the bottle in brine and blow the cork off to expel the yeast.

        2. >>” with their flatulence carbonating the end product.”

          Yes, and if you add more sugar when they’ve already got enough sugar, you’re going to get a gusher or possibly bottle-bombs. You only need 0.003 gravity points to carbonate most styles of beer.
          Also do yourself a favor and add your sugar to the whole batch to ensure uniform carbonation. If you absolutely must carbonate by the bottle instead of in bulk, get some carbonating tabs (such as Coopers).

          Autolysis is greatly exaggerated at the home brew scale. Unless you’re abusing your beer’s temperature it’s not really a problem. Wine, cider and mead are different due to their low nutrients, but for beer it’s generally a non-issue.
          The reason the big guys do it is the hydrostatic pressure at the bottom of a 2 story fermenter really stresses the yeast.

          Champagne was cloudy for most of it’s history. Exploding champagne bottles is a reflection of bottle technology and absolute insane carbonating pressures (4-6 bar; 45-75 psi). The punt (cone at the bottom) adds a tremendous amount of strength.

          1. “The reason the big guys do it ”

            Do what? I’m not entirely sure what you’re talking about.

            In practical experience, when you bottle beer when it’s just stopped bubbling (the 1 minute per bubble rule), you still have lots of suspended yeast and you get a large amount of sediment in the bottles, which then gets churned up by the bubbles and turns the beer very cloudy and very foamy. The suspended yeast also has an off taste and a mouth-feel that resembles wet toilet paper on the tongue.

            So just let it sit for a couple days extra, even an extra week is just fine, and siphon the beer off into a clean vat. That also makes it easy to mix the priming sugar or malt extract in the whole batch, so you get even distribution in the bottles and no explosive accidents.

          2. @dax

            Apologies, I got comments mixed up a bit. “it” being the reason breweries use secondary and bright tanks is not just beer clarity. Hydrostatic pressure increases stress and on yeast leading to more prevalent autolysis. In small scale (5-20 gal) there simply isn’t enough beer to cause this.
            >> The suspended yeast has an off flavor.

            Unless you’re drinking the dregs you’re not going to get this. This is why most bottle conditioned beer is decanted before drinking.Some styles are deliberately cloudy and taste just fine.
            My main pick is the suggestion by [lupus] and others that prolonged storage on the Lees negatively affects flavor. In beer it does not. Autolysis is an exaggerated phenomena in home brewing that frequently gets blamed instead of poor brewing technique or ingredients.

          3. “Unless you’re drinking the dregs you’re not going to get this. This is why most bottle conditioned beer is decanted before drinking.”

            Well that’s just the probem, because when you have lots of sediments simply opening the bottle will cause the carbonation to mix the sediment back into the beer, which causes more violent bubbling, which causes more mixing and the beer jumps out. This isn’t so much of a problem for lagers because they’re chilled to low temperatures to keep the CO2 in solution, but for ales that may be drunk warm, extra yeast in the bottles just makes for extra jumpy bad tasting beer.

          4. “Some styles are deliberately cloudy and taste just fine.”

            They also use particular strains of yeast that don’t taste bad.

            For lagers the sediment in the bottle isn’t a problem because it’s chilled to keep the CO2 in solution, but with ales that are sometimes drunk warm, it just makes up for gushy yeasty beer because the bubbles mix the sediment back in.

    1. Agreed. Furthermore, it’s very easy to have leaks in the fermenter seal. You could still have an active ferment generating CO2 and it just leaks through the seal instead of the airlock.

      Personally I just leave it in the fermenter for longer. Whilst the main ferment might be over in 3 – 5 days, I leave mine in there a minimum of 2 weeks as there are secondary reactions that remove off-flavors. Yeast have dozens of metabolic pathways available to them and Sugar -> Alcohol + CO2 is only one of them.

      1. Do your self a favor and rack into a clean vessel once the yeast has flocced and sunk. Otherwise you get dead and dying yeast releasing disgusting off flavors into your beer.

        The residual yeast still floating around is healthier than the cake in the bottom of the fermenter and will do what you desire in the following week. There is a reason lagers are all racked after initial fermentation.

      1. I’ve played around with the idea of using an inertial measurement.

        So far my prototypes have been failures, but the basic idea is to suspend a relatively long U tube as a vibrating reed of sorts and make it vibrate, and measure the natural frequency of the vibration, which is related to the mass of the liquid inside the constant volume of the tube.

        I’ve tried to bend a thin tube of glass with no success.

        1. Couldn’t you use one of those alcohol sensors to sniff the exhaust of the airlock? It would require some careful calibration, and some of the alcohol would dissolve in the airlock liquid, but it seems like it would work.

          1. Or maybe put it before the airlock with a check valve between it and the airlock. I use vodka in my airlocks, so I would try to isolate the airlock from the CO2 and alcohol vapor coming from the fermentation vessel. It seems like the alcohol vapor sensor might work.

    2. I agree with this comment. Hydrometers are good, sticky ceilings and broken glass are bad. Hey, has anyone out there played with “The Beer Bug”? That thing is supposed to measure and log your beer gravity/temperature.

    3. I’ve seen this ‘hack’ before (and have also played around with it myself). I have also seen this comment before and I have to say I disagree. It is NOT based on a false premise.
      This is really a GOOD idea for a non invasive, cheap and easy way to get an idea of where your fermentation is at. BUT it does not negate the need of taking hydrometer readings.
      Fermentation DOES produce massive amounts of CO2, so there is a correlation between the rate of fermentation and the rate of ‘bubbling’. This can be used to automate the fementation process. Specifically, it can be a good indication of when active fermentation is slowing down and it is time to slowly raise temperature for diacetyl rest.
      You just need to make sure you write your code in a meaningful and failsafe way. I.e. you need to filter and/or look at trends that span for hours. You also cannot completely rely on the data, as (as previously stated) there might be interferences (like leaks and off gassing other than from active fermentation), so you need to have conditions like ‘don’t start diacetyl rest unless we’re at least 3 days into fermentation’ and ‘if not diacetyl rest is started by day 7 start it anyway’, for example.

      1. >>”It is NOT based on a false premise.”

        Using bubbles to determine when fermentation is done is indeed a false premise. You can get stuck fermentation or just really slow fermentation, in addition to the other interferences mentioned by others. These things can’t be detected without measuring SG.
        Diacetyl rests probably aren’t what [Waldy45] is doing since he’s making cider not lagers. But that’s immaterial, as those are also done according to the SG, not fermentation rate (ie, bubble frequency).

        >>”You also cannot completely rely on the data”

        That’s my point exactly. This is not a reliable way to ferment things. Save you’re effort just buy a hydrometer, they’re $9.

        1. Yeah ok. I totally agree that you should not depend on the bubbling (or rather lack of) to determine if fermentation is done. And yes, a hydrometer (or three, since they always break when you need them) should always be in your ‘arsenal’.
          But still, bubble counting ia not without merits. As I said, it can give you an idea of where your fermentation is at. This may save you a hydrometer reading or two, and can optimize your fermentation (like I said, with the diacetyl rest).
          I don’t think we really disagree here. I think we can agree, that bubble counting is NOT a substitute for specific gravity readings, but can still be a very useful tool to hint at where your fermentation is at.

    1. Why do so many “master” brewmakers here assume he’s going to just up and bottle it when bubbles stop? Sometimes we just like to know when we should go check on it. (even with a hydrometer)

      1. Because you actually might want to bottle before the fermentation stops to adhere to Reinheitsgebot – or not using priming sugar.

        Or because you’re in such a hurry that you need to know exactly when the fermentation stops, instead of just waiting the whole week, which means you’re probably going to try to bottle too early.

  2. Nice. MIght try to rebuild this with a Raspberry PI some day. Really can’t understand all the criticism here. It’s like with any technical tool, they are there to help you, not to replace your common sense.

  3. This has been tried before, as others have mentioned, and I think the result is usually that it’s a VERY rough approximation of specific gravity. The main problems are 1) the size of the CO2 bubbles vary significantly so if you try to calculate specific gravity based on volume of generated CO2 your results will likely be innacurate unless you somehow detect the size of each bubble, and 2) the optical sensor inevitably misses some bubbles and each missed bubble introduces an error that accumulates with previous errors.

    Here’s a tutorial from Sparkfun on the topic: https://www.sparkfun.com/tutorials/131

    And here’s an experiment I tried a few years ago: http://sharpk60.blogspot.com/2010/06/i-did-some-more-work-on-pressure-sensor.html

    And this is the most accurate way I’ve seen to do bubble counting which involves controlled measurement and release of CO2 with valves and pressure sensors: http://www.brewnanny.com/

  4. Wish I had seen this earlier. Thanks all for letting me know about the misinterpretation.

    Definitely do not use this as an indication of when to bottle (and to be fair I didn’t in my instructable). The instructable has now been updated to reflect this.

    My reasons for doing this were: Learning opportunity with the arduino, To inform me of the yeast lag time and record this on the brew record along with the starter method. Also after a few days of fermentation I tend to slightly increase the temperature of fermentation after a few days (When the bubbles slow down to say 1 or 2 a minute) . The purpose of this is to encourage the yeast to attenuate fully. This shouldn’t affect flavour but hopefully after a bit of experimentation, this could give me a guide to how much to inch up temperature and when . Another possible use is to alert me in the earlier stages then there is a possible stalled fermentation. Ah yes, I forgot the main reason, my fermenter is in a shed outbuilding in a brew fridge. I want to check the bubbler every now and then without going out there.

    Ken, thanks for your write up (cool work). I hadn’t seen this. What were your conclusions? Do you still use it? I haven’t tried to accurately measure co2 emissions here and I’m also not interested in early vigorous stage of fermentation so much. Having said that the 10mm optocoupler (if it remains in place) does a good job of catching every bubble (as far as I can tell) I must admit neither my cider or current steam beer fermentations are super active.

    I agree measuring specific gravity electronically is a better It’s quite a challange .. so there you go, I look forward to seeing your solutions here. Yes, you could buy a brewnany, but where’s the fun in that?

    I echo Mats sentiment. I recommend using thisa with a certain amount of common sense. This is a tool to use for my reasons given above and any others you can think of. Use a hydrometer to measure final gravity and only bottle when it is safe to do so.

    1. From my previous experiments with this, I concluded that this probably wasn’t the best path for accurate specific gravity measurement, though it works fine as a “fermentation activity” sensor. The hardest part seemed to be the early vigorous fermentation because the bubbles tend to be bigger and less regular than the later parts of the fermentation; most of the errors in the readings are introduced in the early fermentation.

      I’ve since moved on to using weight-based measurements of CO2 production which translates to specific gravity after a bit of math, but I haven’t posted my results yet.

      I agree about the brewnanny: cool device, but it’s more fun to make my own.

      1. Hi, yes I agree with you, specific gravity measurement is where its at to monitor fermentation. Be interested to see your approach. Is it like beerbug where a strain gauge is used to measure effective mass of a submerged weight? Look forward to seeing your results. I’m looking at a less scientific way using a time of flight sensor to measure hydrometer position.

        1. The weight-based thing I’ve been working on is basically like the beerbug but in reverse: I weigh the entire carboy and monitor the weight lost due to CO2 production. It works pretty well, but there are some tricky parts.

          I’m interested in hearing more about your hydrometer position measurement. At one point I experimented with measuring hydrometer position using a hall effect sensor and a tiny magnet glued to the end of the hydrometer.

          1. Interested to hear about your work too. Just coded up and arduino to take measurement from ToF device. I plan to fix a tube ( temp ) to the inside side wall of my fermentor and get the hydrometer rising in that. the tof sensor will catch a thin plate on the top of the hydrometer in the tube. I’ll set it up when I next brew and write up the results ..

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.