Inoculating a Sour Beer

Today I brewed a final batch of beer with the Ardennes yeast I grew up a few weeks ago.  This time I’m aiming for something sour, and using a mix of different techniques to achieve this, so I thought I’d take the opportunity to post some general thoughts on various methods of inoculating and fermenting sour beers.

As with any fermentation, the way to gain some control over the final product is by controlling the environment in which the fermentation takes place.  This can be done in at least the following related ways:

• By controlling the sugars and starches in the food-stuff to be fermented.
• By controlling the kind and population of organisms doing the fermentation (by killing anything already in the foodstuff and pitching new organisms in controlled amounts, or by encouraging some organisms already present).
• By creating an environment that will favour some organisms over others (through temperature, pH, isomerized alpha and beta acids, salt-levels, etc.)

All of these basic methods can be seen in the various ways people brew sour beers.  The character and sourness of these beers will be determined by the course of their fermentation, and we control this (to some extent) with a variety of techniques: by controlling the sugars and starches in the wort, by pitching organisms at particular points during the fermentation or by controlling their cell-counts in the wort, and by controlling the environment in which the fermentation takes place so that it favours some of these organisms over others.

For instance, there are various ways to ensure you end up with a decently sour beer.  One is to pitch the bugs early, either in a blend like Roeselare or separately from the yeast, so that they can grow up sufficient numbers to out compete the sacch for at least some of the sugars in the wort.  The extreme version of this is the fermentation of a berliner weisse, where you pitch far more lactobacillus than sacch (~5:1 ratio), or pitch the lactobacillus a few days before pitching the sacch so that it doesn’t have to compete at all.  But even if you don’t pitch a large population of bugs and brett, you can still give them an advantage by adding them along with the sacch, and some home brewers like Michael Tonsmeire recommend this to ensure decently sour beer. 

Another technique (which can be used in conjunction with the previous one) is to deliberately brew a highly unfermentable wort (e.g. with a turbid mash, or by adding starches) so that the bugs have more sugars to consume once their numbers have grown.  This will favour slower growing organisms like pedio and brett, which will work together to break down and ferment all of the complex sugars left in the wort. 

Jamil Zainasheff used to recommend mashing at a higher temp, and then fermenting the beer with a clean yeast before adding bugs and brett.  Vinnie Cilurzo does something similar at Russian River, adding different organisms at different points in the fermentation to control the final product.  Some home brewers complain that this results in a final beer that is not very sour---especially if you’re relying on commercial blends like Roeselare for your secondary pitch, since these have low cell counts.   One way round this is to pitch either larger populations and/or more aggressive organisms.  Ron Jeffries at Jolly Pumpkin has described how he mashes low to ensure a highly fermentable wort, which is first fermented by his regular sacch strain (related to Wyeast 3522) in open stainless steel fermenters before being transferred to barrels for a secondary fermentation by the resident critters.  Since JP have aggressive bugs, the resulting beer still sours and can be packaged relatively quickly.

(I should also mention barrel-aging as a related technique: the micro-oxidation through the barrel walls seems to be an example of an environment that favours certain organisms in a way that is difficult to replicate without a barrel---but this is outside my experience).

I’ve used all of these techniques before, with varying levels of success.  My first sours were made with a single packet of Roeselare (I describe one of them in this post). I’ve also made several berliner weisses by co-pitching large amount of lacto and yeast (again, see previous post), and I have several batches currently fermenting that began with deliberately starchy worts. 

In general, though, I tend to prefer the Jolly Pumpkin technique, even if it results in slightly less sour beers.  I am particularly cautious about pitching brett too early in a mixed fermentation.  I once added a full pouch of Wyeast Brett Brux to a 3 gallon Orval clone before primary fermentation was completed, and the result, at only a few months old, was a very brett driven beer: lots of barnyard aromas, horsey, almost fecal, and not at all what I wanted.  I looked for advice around the web, and asked questions in the comments for this post on Michael Tonsmeire’s site, all of which confirmed my suspicions: the dominant character was probably partly a result of the strain, and partly a result of pitching so much brettanomyces early in fermentation (Wyeast packs have a relatively high cell count). The key seems to be pitching a lower numbers of brett and bugs so that they don’t out-compete the sacch in the early portion on fermentation, but are able to grow up sufficient numbers to make an earlier contribution and consume more of the sugars in the wort.  (Although brettanomyces should be able to make a contribution whenever you add it, since it can metabolise other compounds besides sugars in the beer.  I like the results I’ve got from adding it to very dry beers at bottling.)

Today I decided to try something a little different, adding bugs and brett from a number of sources along with the yeast for a more mixed fermentation.  I deliberately made a fairly fermentable wort, mashing at around 149°F and using ~8% table sugar, along with ~30% wheat.  The yeast was grown up from the small pitch I cropped in this post.  Once the starter was actively fermenting, I added dregs from a bottle of Jolly Pumpkin’s Oro de Calabaza.  Since its a strong beer (~8% ABV), and the bottle was about eight months old, I don’t know how viable the dregs were.  I’m hoping that adding such a small quantity to an already fermenting starter ensured only a small growth in the population of any bugs or brett, but made sure there were some JP bugs in the wort. 

I also pitched from three other sources.  The first was a pint of apple juice inoculated a few days ago from my lactobacillus jug, and stored on my heated yogurt maker.  I don’t know what else is in this starter besides lacto: it was started about 8 months ago with a pouch of Wyeast Berliner Weisse blend that was past its sell-by date, so it may well still have both brett and sacch in it along with the lactobacillus.  My hope was that by feeding it exclusively apple juice, I would create an environment in which the lacto out-competed everything else.  I feed it more juice and yeast nutrients every few months, and store it in the fridge between feedings.

I don’t know how much contribution the lacto will make to the final beer.  A one pint starter is pretty small (I do many times that for my berliner weisses), and two days may not have been enough for the lacto to reach maximum cell numbers (I usually leave these starters for a week or two).  Moreover, although I kept the bitterness relatively low at 15 IBUs, that’s enough to inhibit the lacto, especially since I used low alpha hops which ensures plenty of other hop-related compounds are in the beer.  On the other hand, I also included a reasonable amount of wheat (~30%).  Why does that matter?  Well, I seem to remember that somewhere in this interview Brian Taylor of Goose Island makes an off-hand comment about lacto consuming sugars from wheat before anything else, so perhaps that will help it make some contribution.

The second source of bugs and brett was my “bug farm”: a gallon jug that started with some of the yeast cake from beers brewed with Roeselare and JP dregs, to which I’ve added dregs from various sour and wild beers, brettanomyces strains I’ve picked up from various people, and some of the East Coast Yeast Bug County I picked up this year.  I have no idea what of that is active in the beer now: I feed it occasionally with more wort, and right now it tastes tart and lightly funky.  Today I “pitched” from this by taking a turkey-baster full of beer from the middle of the jug, without swirling up the dregs, and adding it to the wort.  This means that I only pitched bugs and brett currently active in the wort, and their cell counts will be very low (perhaps too small to make a real contribution, although I’ve done this before along with bottle dregs and successfully inoculated batches).

Finally, I also pitched some of the starter I made from the Lochristi Brettanomyces blend I got from The Yeast Bay.  This has gradually grown up to about 1/3 gallon, which I’m going to store in a half-gallon jug.  Since the cake here is fresh, I swirled it all up and pitched a baster full of the cloudy beer into the new wort.  The cell counts should be higher here, but hopefully not so high that the brett aromas and flavours become too dominant early in the beer’s life.  I am slightly worried that by making a starter I changed changed the balance of the blend, but hopefully since there are only brett strains in it they grew at about the same pace.

I want the final beer to be a golden ale with medium sourness and a light, fruity funkiness, ready to package in 3 to 4 months.  We’ll see how things turn out: it will depend on the gravity and sourness after primary fermentation is over, so I’ll post about it again in a few weeks.

Recipe:

   

Estimated O.G.	1.056
Measured O.G.	1.049	(Low!)
Measured F.G.
ABV.
Mash:	149°F
Malt:
55.6% 2-Row
22.2% Wheat Malt
8.3% Vienna
5.6% Torrified Wheat
8.3% Table sugar
Hops:
Fuggles	60 mins	12.3 IBUs	(15g @ 4.1%)
Fuggles	15 mins	3.3 IBUs	(15g @ 4.1%)
Yeast:
Wyeast Belgian Ardennes (3522)	Lactobacillus	My ‘Bug Farm’ and Lochristi Blend