Over the last year I've been having a peculiar issue with acetaldehyde and bottle conditioning. After extensive research--both in my print library and online--I remain flummoxed. Perhaps, the community can help.
Here is a bit of background:
I have been homebrewing for about twelve years. I rely solely on force carbonation for many of my lagers and American-style brews, but for certain classic European styles (such as Belgian ales, mixed-fermentaiton beers, and German wheat beers) and big beers I like to bottle condition--to obtain very high carbonation levels for the former example and to help promote the longevity of the beer for the latter example.
In order to really dial in my carbonation, here is how I typically bottle condition. After fermentation, I will keg the beer, chill it, and carbonate it to 2 volumes. A couple of days before the beer has finished carbonating to that level, I will make an acclimation starter consisting of 50% DME-based wort (~1.035), 50% of the base beer, and two grams of rehydrated dry yeast. After approximately 24 hours, the acclimation starter will be chilled. When it is time to bottle, I decant the acclimation starter and add the yeast slurry along with an appropriate dosage of cane sugar (which has been dissolved in water, boiled, and chilled) to provide enough sucrose to reach my final target carbonation level (usually 3.25-4.25 volumes).
I realize that my method is relatively laborious, but I use this method because it allows me to laser target my carbonation level (since I know exactly the carbonation level of my beer before bottle conditioning), reach very high carbonation levels, and take advantage of the oxygen-scavenging benefits of bottle conditioning. And, with the exception of what I will describe below, it has worked like a charm, reliably and rapidly carbonating very alcoholic and/or acidic beers.
So, here comes the rub.
About a year ago, based on positive online reviews, I started using CBC-1 as part of the above protocol. At first, I bottle conditioned a number of Belgian saisons, and everything worked like a charm and the taste was spot on. The CBC-1 led to bottles being fully carbonated in a week or less, and the lees formed a thin, compact layer on the bottom of the bottles. The primary-fermentation strain that I was using for those Belgian saisons was WLP 566 Saison II. I bottle conditioned (a) lower-alcohol, quick turnaround table saisons (~1.048 O.G., primary fermented for about ten days); (b) moderate-alcohol, moderate turnaround saisons (~1.062 O.G., primary fermented for between fourteen and twenty-one days); and (c) higher-alcohol, long turnaround bieres de garde (~1.072 O.G., primary fermented for approximately twenty-one days, and lagered for approximately two months).
Based on my success with bottle conditioning those saisons with CBC-1, I turned to some other beers.
In a line of lagers fermented with WLP 833 German Bock Lager, I decided to bottle condition a tropical stout and Baltic porter. Those beers were fermented in line with my normal lager protocol (healthy pitch of a proper amount of lager yeast, step fermented with a rising temperature over three weeks, diacetyl rested at the end), and checked for diacetyl (via a forced test), acetaldehyde, and terminal gravity before crashing and lagering. Both beers tasted as expected before and after their 14-week lagering periods. However, after bottle conditioning as per the above method with CBC-1, the beers developed a marked acetaldehyde character--underripe apple, somewhat acetic and solvent, and almost plastic-like on the finish--that completely overwhelmed the other flavors. All other lagers that I made in this series with WLP 833--German-style pilsner, rauchbier, Vienna lager, pale doppelbock, dunkles bock, schwartzbier, and Munich dunkel--were solely force carbonated and never developed acetaldehyde in the bottle.
Finally, in a line of German-style wheat beers fermented with WLP 380 Hefeweizen IV, I had a similarly bad experience. I brewed a hefeweizen, dunkelweizen, and weizenbock, again ensuring that all had proper time to primary ferment. They too were bottle conditioned using the above procedure with CBC-1. And, once again, all three batches developed acetaldehyde in the bottle. As a control, re-made the dunkelweizen and solely force carbonated it to a high (though not as high, given the limits of bottling beer force carbonated above about 3.25 volumes) level. That batch showed no acetaldehyde.
I should note that some of the batches that showed acetaldehyde were given as much as two months in the bottle. The acetaldehyde never disappeared, let alone reduced at any appreciable level.
I'm stumped about what's going on here. It appears that the bottle-conditioning step is the common factor among all of the acetaldehyde beers. However, I'm having difficulty attributing it solely to my use of CBC-1 since none of the saisons--again, all of which were bottle-conditioned using CBC-1 and the identical procedure as the lagers and wheat beers--showed the acetaldehyde fault in bottle. Also, I'm largely convinced that other possible contributors of acetaldehyde are not at play since none of the beers showed acetaldehyde before bottling and, in any event, if factors such as over oxygenation at pitching, early removal of beer from the yeast cake, and/or oxidation at packaging was the culprit, then the other beers (i.e., the purely force-carbonated beers and/or the bottle-conditioned saisons) would have been expected to show the acetaldehyde fault.
Does anyone have any guess at what's going on or have any similar experiences? If, indeed, CBC-1 is the cause, then it appears that it is throwing off acetaldehyde during bottle re-fermentation and not cleaning up after itself. If that's the case, then for some reason in the WLP 566 saisons the primary yeast seems to be cleaning up in its place, which is particularly odd in the bieres de garde since they had been lagering for two months, and, therefore, it seems strange that they would have enough viable yeast to do the clean up, never mind doing it so quickly (read: in under a week). Could there be some other factor at play--some interaction between beers fermented with WLP 833 or WLP 380 and then bottle conditioned with CBC-1 that does not arise where the primary strain is WLP 566?
Practically speaking, I will probably be turning to some other dry yeast for bottle conditioning--perhaps returning to T-58 or trying out EC-1118. But, I'm really just interested in ferreting out what's going on here (since I otherwise really like CBC-1's performance), not to mention that if this issue is not specific to CBC-1 then it might crop up again with a different strain. Also, after throwing out so much beer, I'd like to learn something concrete!
Thanks for reading, and sorry for the book-like post!