Bottle Bombs and Other Catastrophes

Ah, the infamous bottle bomb.  I'd heard about them, but never experienced them.  All good things must come to an end, I guess.

I experienced my first one a week or so after bottling Hefeweis City (see Hefeweis City, Part I and Hefeweis City, Part II).  Nothing major; I had gone down to my laundry room/brewery to check on things...and saw a puddle under the case.  Never a good sign.  Fearing the worst, I opened the case.  Glass and beer everywhere, but much to my relief only one broken bottle.  I checked the other case, and found no further victims.  I cleaned up, happy I'd only lost a single bottle.

A few days later, I was downstairs washing bottles and I heard it--a dull thump.  It sounded like something had fallen over in my brewing cabinet.  I opened the doors, and lo and behold...beer was dribbling out of the case.  This one made much more of a mess, despite the bottle breaking cleanly into only a few pieces.

While I waited for all the cardboard to dry, I sat down to think about causes.

What could have caused this?  The proximate cause is overcarbing, of course--carbonation causes more pressure than the bottle can stand.  But why this batch?  Why these particular bottles?

The answers I came up with:

• Bottled too early.  Only a week had passed since the beer went into the fermenter.  Fermentation was early and ended early, with very low, stable gravity readings by the end of the week.  I could have been wrong about fermentation being over, though.  If the beer still had residual fermentables when I bottled it, that could cause bottle bombs.
• Too much priming sugar.  I used a full 5oz package of corn sugar.  This was higher than the style guidelines, according to BeerSmith (BeerSmith recommends 3.61oz of corn sugar), but certainly not more than a bottle should be able to handle.  Lots of homebrewers use a full 5oz every time without issues.
• Weak bottles.  Several cases of bottles were donated by a friend and are reportedly decades old.  Some still had old homebrew in them (cleaning them was not fun).  Perhaps age and/or holding beer for so long had weakened them.  I've never heard of it, but Hamlet was right when he said "there is more in Heaven and Earth [...] than is dreampt of in your philosophy".  It would explain why only one had gone off.
• Poorly-Mixed Priming Sugar. If some bottles had more sugar than the others, they would over-carb even if the total amount of priming sugar was correct.  Weak bottles might exacerbate it, but too much sugar will bust even a strong bottle, I suspect.

It took me a while to figure out, but I think the last was the cause.  My standard procedure is to boil up a solution of priming sugar and water, cool it to room temperature, and add it to the bottom of the bottling bucket, trusting to the agitation from racking to mix it thoroughly.  It works for my dad, and it worked for my first few brews.

However, a recent hop experiment (5 1-gallon batches of IPA, each with a different flavor/aroma hop) gave me mixed results in several areas, including carbonation.  At the time I chalked it up to the difficulty of small measurements and doing things on a different scale than I was used to, but now I'm not so sure.  The first several I tried turned out flat, but others were well-carbonated.  It doesn't appear to be a function of time, either: carbed and uncarbed have come up interspersed.  However, I think it might relate to the position of the bottle in the case (and thus to the order they were brewed in).  If most of the priming sugar went into either the early or late bottles, it would explain the differing carbonation levels.

I had assumed that racking one gallon over just didn't give the priming sugar enough time to mix.  With the hefe, however, I racked a full five gallons.  But it's the only answer that fits.  If I bottled too early, all of the bottles would be dangerously carbonated.  The first test beer was flat, which disproves that theory.  Same thing with too much priming sugar--I should be seeing overcarbonation on every beer and multiple bottle bombs.  Weak bottles would explain why I had two bottle bombs and no more, but it doesn't explain why the test beer was flat after three weeks.  An inconsistent level of priming sugar, however, explains all of the symptoms.

After a little research on HomebrewTalk.com, several members recommend stirring the beer (gently!  We don't want to oxygenate the beer) after racking it to the bottling bucket.  I think I will have to try that in the future.  The IPA's were an experiment, and successful in their goal despite inconsistent carbonation, but the hefe was supposed to be something I and my friend could show off.