Monday, 23 August 2010

Thornbridge: Bottle-Conditioning for the 21st Century

There's no question that Thornbridge are one of the most exciting breweries in the UK today. While they stop short of actually using hummingbird's tears as an ingredient, they push the envelope both with what they can get out of conventional ingredients, and what they can do with more unconventional ones. They rarely miss the mark.

Thornbridge recently moved over to putting live beer in half-litre bottles. On bottles of Jaipur, it uses the phrase "bottle-conditioned", although this has caused a bit of confusion among consumers - if it's bottle-conditioned, why isn't there any sediment? Being an inquisitive sort, I called Kelly Ryan at Thornbridge. The short answer is that there isn't any visible sediment because they don't re-prime the bottles. The long answer is a bit, well, longer.

When Thornbridge moved over to their new brewery, they had a new level of control available to them. The closed cylindro-conical fermenters meant that they could allow the beer to carbonate naturally under the pressure of its own fermentation. The same fermenters allowed them to take dead yeast out of contact with their still-fermenting beer. Centrifuges meant that they could control how much yeast was left in the finished beer, removing the need for filtration. The conclusion to the long answer is that they don't reprime their bottles because they don't need to. Not only is there enough natural carbonation in the beer, there is also enough yeast left in suspension to keep things ticking over nicely when they bottle it. The fermentation continues slowly, maintaining carbonation and scavenging oxygen.

For my part, I've often thought that bottle-fermentation can actually substantially reduce the character of a beer - it's as though that vigorous secondary fermentation actually scavenges some of the flavour compounds from the beer. Or maybe if the brewery filters the beer before repriming (a very common practice), a lot of the character of the beer is removed, and never gets put back in the second time around.

So what do we make of all this? Is this live beer? Bottle- or brewery-conditioned? Does it even matter to the ordinary drinker? My take on it is that if 21st century brewing and hygiene technology allow this to happen, and the beer tastes better for it, then great.

Either way, what I find really impressive is that Thornbridge have decided to ditch 40,000 Jaipur back labels that say "bottle-conditioned” and replace them with ones that say “May contain sediment”. Not only do they care passionately about the beer, they also care passionately about getting the right message on the label. The devil, as ever, is in the details.


  1. WOW 40,000 labels, thats some commitment! well done Thornbridge!

  2. Do the new labels have a little tankard logo and the words "CAMRA Says We'll Have To Have A Bit Of A Think About This One"?

  3. Hmm, will CAMRA accept this as RAIB? And, if they don't, will Thornbridge be bothered?

  4. Good for you Thornbridge! I completely agree with The Beer Nut the CAMRA "Real Ale" thing has always bothered me:

    It the drink physically exists then it must be 'real' and if it's ale then it's 'ale' - so 'real ale' should probably just mean 'not a lager'...

    Any what's the difference between 'real ale' & 'ale'? Is 'ale' fake?

  5. This comment has been removed by the author.

  6. There's no debate to be had. It's quite clearly brewery-conditioned, not bottle-conditioned. And there's nothing wrong with that.

    Is reprinting 40,000 labels impressive? Sounds more like common sense. It would only take one dick to report them to Trading Standards and they'd be in trouble.

  7. Barm, the beer is fermenting in the bottle ("there is also enough yeast left in suspension to keep things ticking over nicely when they bottle it. The fermentation continues slowly"). Surely that makes it bottle-conditioned?

  8. But does it matter that it's 'bottle conditioned' (or not). Some bottled conditioned ales are no better for it, others are, surely it should be up to the brewery to decide if a little yeast in the bottle makes their beer better rather than _having_ to add it to obtain a CAMRA badge?

  9. "But does it matter that it's 'bottle conditioned' (or not)."

    Well, to some CAMRA members that is Holy Writ that determines whether it is Good Beer or chemical fizz ;-)

  10. Funnily enough, I've been writing something on the same lines; Beer Technology

    I think it answers "would they care?"

    I still think it IS bottle conditioned.

  11. The elephant in the room here is that all bottled drinks, with or without yeast in the bottle, undergo changes in the bottle over time: Mouton Rothschild doesn't contain yeast in the bottle, but a 15-year-old bottle will taste very different to a 1-year-old one. The "contains live yeast" argument is pretty irrelevant unless, like champagne, you want the carbonation to take place in the bottle - and even with champagne they take the yeast out before they sell it to the customer.

  12. Andy it's not just about the beer, it's the whole package (literally in this case)

    Beer Nut, Curmudgeon - I think that it's live beer, but as Barm points out, mainly brewery conditioned. It's still awesome though.

    Baron O - I think you're right that it's up to the brewery to do the best for their beer, although a surprising number don't. And there's always the argument that a stable, filtered product that tastes the same every time is more acceptable to the majority of consumers than something that is as potentially fragile and variable as live beer.

    Dave - I think they're nice complementary pieces, cheers. I think Kelly is also about to post something about their changes, and their future plans.

    Zyth - I'm not sure that it's really an elephant, but certainly another facet to the argument. I think the approach of having enough live yeast to keep things ticking over is a sensible one, but it doesn't guarantee anything. And of course, the effects of time are a personal preference - I can count on the fingers of both hands the number of beers that I'd like to drink at older than about a year or 18 months in the bottle. For me, fresher beer is usually better - but that's an argument for another post.

  13. Hmm, I'm not sure I'm convinced either way about bottled beer.

    On the one hand, yes, Bottle Conditioned sounds natural and instils the feeling of goodness.

    On the other hand Filtered and Pasteurised is basically evil death if you are a Camra member.

    Questions, at what point does a beer stop evolving once bottled? At what point in its life-cycle was it actually bottled, and did that bottling stop its evolution?

    Hmmm! I know from my brewing that one beer can reach a peak condition in as little as 1 week after bottling but others can take months to reach the point where all the malts and hops have combined into a whole. Different for every style.

  14. those were actually the same hand!! ;)
    I think the real second hand is the questions!

  15. Sounds like the way forward to me. I look forward to trying the beers. Well done Thornbridge!

  16. Zak, Thornbridge teach this knowledge to the De Dolle brewery for their awesome Stille Nacht range.So many of their awesome beers have contained sediment.

  17. To me the 'real' (not literally :~) elephant in the room is that CAMRA's definition of RAIB bizarrely excludes Thornbridge (seemingly specifically because the bulk of condition in the beer is created in the CCFV before it reaches the bottle) yet includes the bulk of the big name RAIBs which are in supermarket & winning Champ Bottled Beer of Britain awards are actually sterile-filtered before being reseeded with fresh yeast (& possibly also primed with some form of sugar).

    Why is this important? Well it's not necessarily - I've drunk beers that have been through all sorts of processing & come out the other end perfectly fine (I'm told US craft-brewing veteran, Anchor's beers are all pasteurised, yet I love them when I drink them). I'm not sitting on the fence here - by and large my beers of choice are cask & unfiltered, I would agree with Zak - generally fresh is best, but I'd go further & say generally unfiltered is best too.

    What confuses me is CAMRA's acceptance of practices in bottled beer production that they would abhor in a cask beer.

    Sadly this (correct) re-definition of Thornbridge's bottled beer as not bottle conditioned might put some people off drinking it, yet, to me at least, they're actually *more* natural, being unfiltered than many of the ones with the little CAMRA logo on them.

  18. pdntc - it's a fair point, although I'd argue against the idea of all filtered and pasteurised beer being bad. It's just not as great as it might be.

    Paul - if you haven't then they really are worth a try.

    Thomas - are you saying that Thornbridge have been teaching De Dolle this technique? Have De Dolle bought new brew kit recently?

    MicMac - I saw your comments on Adrians blog about this very topic yesterday. I agree with you that it's potentially problematic for Thornbridge, but at he same time, the beer is indisputably a live beer, and one that has been been far less messed about with than most. And that is, indisputably, why it tastes so alive - because it still is!

  19. The definitive process in bottle conditioning is the prodcution of CO2 (condition) in the bottle. If there is no fermentation in the bottle it can't be called bottle conditioned. If there is live yeast albeit invisible live yeast and even a trace of sugar then there will be an increase in CO2. This may not be perceptible which is why they have done the right thing by changing the labels to keep the customers satisfied.

    It's not really bottle conditioning for the 21st century the Belgians have been doing it for decades it's just that they reseed and prime to raise CO2 levels and protect the flavour from premature staling.

    I don't agree that unfiltered bottle conditioning results in a less characterful beer. What bottle fermentation takes away it more than makes up for in addtional layers of complexity. For a brewer sterile filtered BC beer is a compromise between character and flavour/haze stability over long selling periods. When retailers insist that your beer should stay clear and taste fresh for 12 months this is the only practical option. It doesn't taste as good as unfiltered bottle conditioned beer in my opinion.

  20. Oh, I'm not a CAMRA member (I can drink beer without) so don't actually have any problems with filtered and Past-yur-eyesed so long as its good beer ;)

  21. So,

    if our beers are carbonating during fermentation under pressure (which they are) and we then filter our beers lightly so as not to remove all the yeast (which we do) does that make our beers bottle conditioned?

    It is an interesting question, but I am not sure how important it is. For me the key to a great bottled beer is not the source of the carbonation but the length of brewery conditioning time and the lightness of the filtration. I should also add it took us a long time to learn the importance of the latter.

    BTW Zak I completely agree on your thoughts about an enthusiastic bottle fermentation potentially scavenging flavours.

  22. oh and James = James from BrewDog

  23. Stuart - perhaps you might be able to shed some light on why the Belgians like to reprime so their beer carbonates so highly? It's question I've asked of various Belgian brewers/drinkers/experts, but I'll save my conclusion for after your thoughts.

    pdtnc - I agree - good beer is good beer, regardless of the route to get there.

    James - if I was being picky, I'd ask what your yeast cell per ml count was, and cross-reference it to Thornbridge's figures. It wouldn't give me an answer, but it would make me feel like some sort of beer scientist - a worthy goal. But as Stuart says, if there is yeast and sugar in the bottle, and CO2 is being produced, then some amount of conditioning is certainly taking place in the bottle.

  24. @James - While I'd probably prefer your method of bottling than many others, does it constitute bottle-conditioning? Only partly by the sounds of it, so I'd guess that CAMRA at least would say no, because (as with Thornridge's new bottled beers) most of the CO2 is not coming from the bottle-refermentation?

  25. James - If you could make your "light" filtration slightly lighter so that a couple of viable cells got through you could rightfully claim that your beers are BC. I wouldn’t presume of course to tell a marketer as brilliantly-gifted as yourself how to label beers. I am merely a brewer.

    Zak - I have been told by Belgian brewers that CO2 levels are high in Belgian BC beers because they are strong and the effervescence breaks up the higher residual sweetness which is generally found in strong beers. Rather like adding alcohol or citrus to sweet chocolate desserts it provides greater freshness in the palate. CO2 is also higher to provide an additional hurdle to infection in live beers. Both of these explanations make sense to me as a brewer but I am by no means an authority.

    Yeast in beer affords packaged beer protection against the effects of oxygen. Oxygen is the driving force behind the early stages of flavour staling and haze formation. Where maturation ends and staling begins and whether a beer is acceptable when hazy are determined by the consumer. Beauty is in the eye of the beholder.

  26. Great discussion guys.

    The main point is that the CAMRA definitions are outdated and really show their lack of any brewing knowledge...but we all knew that already.

    What matters and what is evident here is that brewers are trying their damndest to get the most stable and flavourful beer as possible in the package.

    We also use a similar method at Lovibonds, minus the centrifuge and minus the filtration. We aim to get enough yeast in the bottle to create a layer of 'paint' on the bottom, similar to what you'd get at the bottom of a bottle of Sierra Nevada.

    I warn you other brewers trying these methods...CAMRA may create some super secret documentation on you like this:

    CAMRA MumboJumbo

    Best line is that they are 'unable to determine if Lovibonds beer is real or not'. Classic.

    Cheers, Jeff @ Lovibonds

  27. It's very hard to get a good consulting phenomenologist these days.

  28. Stuart - that's very interesting, and makes a lot of sense, thanks. Carbonation is one of my favourite little kinks - it's something that profoundly affects how a beer is perceived, and yet seems to be an afterthought in many bottled beers.

    Jeff - that's quite a thing to have in your possession - very CIA. Or CIAle, maybe.


Sorry about the word verification - the blog was getting spammed to bits.