Carbonation: Carbonation Levels and Priming Sugar

Lately I've been having a spate of beers that I've felt were overcarbonated. For some styles, especially American Ales, higher carbonation is okay. Too much carbonation can make other styles taste harsh and hide the malt character, especially british-style ales and ales with lots of crystal malts or malt complexity. In particular, my English IPA, Arrogant Bastard clone and English-style Apple Ale all ended up with enough carbonation to taste unpleasantly harsh. Several of my beers lately have also been gushers and require long refrigeration periods to avoid gushing.

Taste and mouth feel aren't the only problems. Last night while grabbing a few beers out of the closet I dropped one of my English IPAs. It only dropped 6 inches to the ground, but it EXPLODED. Blood and broken glass everywhere, lots of freaking out by the wife. I ended up with several small cuts on my calves and one long gash that exposed some subcutaneous tissue. Fortunately it was shallow and I closed it with some butterfly bandages and got a tetanus booster and it'll probably be okay, but it could easily have been worse.

I'm only going into disgusting details to stress the point: beer and glass is no joke! Watch your carbonation levels and be careful with your glass carboys unless you want a major hospital visit!

Anyway...

CO₂ is measured in what is called 'volumes'. A volume is defined as a 1:1 ratio of CO₂ and liquid, measured at 1 atomsphere of pressure. For example, one gallon of beer with one gallon of CO2 dissolved into it would have 1 volume of CO2 dissolved in it.


I use BeerSmith for both the priming sugar calculation and the recommended carbonation levels. Every style definition in BeerSmith has a carbonation range. For example, the English IPA style recommends 2.2 - 2.7 volumes of carbonation.

The first thing I considered was that I always calculate my priming sugar before measuring my final bottling volume. I look at the side of the keg and try to estimate how much beer is in there and I use that volume in the priming sugar calculation. The problem is that I always lose some of that to trub and yeast, so my volume is lower and my carbonation is higher than predicted.

Here's an example. With my English IPA (#10) I thought I had 11 quarts, because that's how much went into the fermentor. I used 73 gm for 11 quarts to create 2.6 volumes. The actual volume racked into the bottling bucket was 10 quarts. 73 grams of corn sugar added to 10 quarts actually creates almost 2.8 volumes, way higher than I intended!

The solution I've adopted to avoid this is to rack first, measure carefully using the markings on the bottling bucket and THEN calculate and mix the priming sugar. Problem solved, right?

Not really. 2.6 volumes is way too damn high for an English IPA to begin with. I didn't know this at the time, as I didn't realize that English IPAs have a more delicate flavor and need a lower carbonation than American IPAs to bring out that flavor. I started looking around at various resources on the web and realized that BeerSmith is full of shit!

Here are some carbonation levels for an English IPA from various sources:

• BeerSmith - 2.2 - 2.7
• The standard 3/4 cup per 5 gallons method - ~2.3
• Beer Recipator (http://hbd.org/cgi-bin/recipator/recipator/carbonation.html) : 1.5 - 2.3
• Screwy Brewer (http://www.thescrewybrewer.com/p/brewing-tools-formulas.html#bpc) : 1.5 - 2.3
• BYO (http://www.byo.com/resources/carbonation) : 1.5 - 2.2 (note: this is for British Ales, without regard to style, so I imagine IPA would be at the upper end)
• TastyBrew (http://www.tastybrew.com/calculators/carbonation.html) : 1.5 - 2.3
• How To Brew (http://www.howtobrew.com/section1/chapter11-4.html) : 1.5 - 2.0

In retrospect, I would probably carbonate at 2.0 volumes or less.

In summary, here are the things to watch out for when carbonating:

• Don't use BeerSmith's carbonation recommendations, they are bullshit. Use TastyBrew or Beer Recipator. The BeerSmith sugar calculation is probably fine, though. I plan to go through and edit all of my BeerSmith style definitions to use better carbonation recommendations.
• Don't measure your priming sugar until after you've racked to your bucket. If you are taking a hydrometer sample and drinking it or throwing it out, use the volume that's left after removing the hydrometer sample for your calculations.
• Refrigerate your beer as long as possible. This not only gives you crystal clear beer, it ensures that the carbonation is properly dissolved in the beer.
• Adjust your priming sugar calculation for the temperature of the beer at bottling. Fermentation produces CO2, which stays in the beer as 'residual CO2'. This is why you often see bubbles in your beer as it passes through your racking tubes. This can be close to 1 volume, so it is significant, and the amount depends on temperature. A beer bottled at 50 F will need less sugar than one bottled at 80 F, because cold liquid holds more CO2.
• When in doubt, use less priming sugar than you think you might need, or stick to the rule of thumb '3/4 cup per 5 gallons'. This equates to 4 ounces or 113 grams of corn sugar for 5 gallons of racked beer, and will give you 2.3 volumes of CO2. This is too much for English styles and too little for some German and Belgian styles, but will work fine for most American styles.
• If you are afraid of bottle bombs (unlikely unless you carbonated to 4 or 5 volumes, did not allow fermentation to finish or used really shitty bottles), put them all in a box or heavy duty bag. You do not want to go blind or slice open an artery. And for shit's sake, don't drop any!

Infection False Alarm

I opened a two-week-old bottle of batch #11, my Hefeweizen, only to have it overflow all over the counter!

Yes, my first gusher. A 'gusher' is, as you probably guessed, the term used when your beer is overcarbonated and spills out when uncapped. To be a true gusher, you can't have shaken, dropped or frozen the beer beforehand, and it doesn't really count if it only lasts a couple of seconds. A real gusher will pour out what seems like an endless amount as time slows down and you think about all the beer you're losing.

What the hell went wrong? Carbonation was calculated perfectly. The beer was refrigerated. My sanitation was as good (read: paranoid) as ever.

I recalled that my refractometer measurement had put this beer's FG at 1.006 - pretty damn low, but I figured it was just a super active yeast. Furthermore, the yeast used in this batch was a repitch from a previous batch.

My thoughts went immediately to a gusher infection - wherein wild yeast invades your beer and starts to ferment the stuff that your normal yeast won't ferment. When this happens, your beer continues to ferment down to the point where it is nothing but alcohol and water, continuously producing CO2 until your bottles eventually explode from the pressure.

I started to freak, and assumed that I had gotten wild yeast in my WLP300 when I washed and bottled it. To test my hypothesis, I opened up a bottle of batch #12, the Pliny the Elder clone I brewed with a repitched WLP001: GUSHER.

At this point I'm running around like a chicken with its head cut off. I open one more bottle of pliny and two more bottles of hefeweizen, all gushers. I of course drank them, which left me in a less than clear state of mind. Did you expect me to commit alcohol abuse? I think not.

---

The next day, with a clear head, I start doing some experiments to find the problem. If the infection came from yeast repitching, there shouldn't be an infection in batch #15 or #13, but there would be in #14. If the infection was in the fermentor, it would be in #11 (the hefe), #12 (pliny), #14 and #15 (brewed in the same fermentors as #11 and #12). If it was in the bottling bucket, then #13 would be infection free (and thus have a normal FG) until bottled.

It was then that I discovered that my refractometer readings were off. By a lot. Normally, you can get a gravity reading of fermented wort with a refractometer if you plug it into BeerSmith along with the OG. The problem is that when double checked against a hydrometer reading, the refractometer reading was coming up 1 plato short. Consistently.

Thus it was that I thought my hefeweizen had 88% attenuation and my pliny had 93% attenuation. When I added 1 plato to their FGs and plugged in the numbers, the hefe went down to 75% and the pliny went down to 85% - both normal. Batch #13 also matched the hydrometer's reading perfectly if 1 brix was added.

Note: the refractometer reads OGs perfectly, so I'm not throwing it out. In fact, I may keep using it for FGs if it is consistently off by 1, because I can just add 1 - duh.

So, if the 'infection' didn't cause my beers to overattenuate, why did it cause gushers? Well, apparently it didn't. A pliny I put in the fridge the night before poured perfectly (and tasty, if a little grassy). A hefeweizen I poured several days later after refrigerating for a night also came out very nice (and also the palest beer I've ever managed to make!).

It turns out that because CO2 dissolves better in cold liquids, it is possible to get a gusher simply because you didn't refrigerate your beer long enough. This doesn't explain why my first bottle gushed, but it may have been inadequate mixing in the bottling bucket. I guess I'll never know.

I learned several lessons here - double check your FG with a hydrometer, even if you have a refractometer, refrigerate your beer adequately, make sure your priming sugar is 100% mixed in on bottling day, don't open your beer any earlier than two weeks bottled and RELAX. I learned these lessons the hard way so you don't have to!