Sunday, 22 May 2016

Bière de Coupage: Kettle Sours

Since writing the first post in this series, I've been slowly making my way through copies of the correspondence pages from early volumes of Le Petit Journal du Brasseur (provided by Dave Janssen of Hors Catégorie Brewing).  One common question about bière de coupage is whether there might be any shortcuts that would allow a brewer to produce them more quickly or cheaply than by buying or ageing beers for blending.  The answer to questions framed in this way is usually a resounding 'no': if you want anything of the complexity and character of an aged beer, you have to use a well-made beer of that sort. But the authors do often make suggestions about how brewers could accentuate or imitate certain elements of blended beers (e.g. by adding acetic acid), and its in that spirit that I planned the beers for this post.

Most of my attempts at making bière de coupage have involved using aged sour beers, usually drawn from a solera, because I was looking for both tartness and complexity.  But if the only thing you want is to add a bit of tartness to a beer, aged sours are not a necessity: all you need is a beer with a clean and pleasant acidity.  An increasingly popular and reliable way of achieving this is by a process of kettle-souring, i.e. pitching lactic acid producing bacteria into the wort in the kettle, boiling to kill them off once the desired level of acidity has been reached, and then fermenting the beer with a regular strain of saccharomyces.

From my experience, I'd say that beers made in this way tend to be a bit one-dimensional and boring by themselves, and I often find commercial versions to be overly acidic.  But these are all characteristics that make such beers excellent candidates for a certain kind of blending with clean beers: where the acid component can provide a controllable degree of tartness, the clean component can provide the further array of flavours we associate with beer, which are often suppressed (in the case of yeast-derived flavours) or overwhelmed in a straight-forward kettle sour.  This is essentially what Off Color does for their gose Troublesome, and what New Belgium does for their wheat beer Snapshot.

With that in mind, I had the idea of brewing a large batch of kettle sour, and storing it in a keg for regular blending.  My thought was that this would allow me to measure out different volumes of the acid beer as required: a few ounces if I just wanted to a small amount of tartness to a beer, or a gallon or more if I wanted some prominent acidity.

Kettle Sour

Grist: Pilsner (80%), Wheat Malt (20%)
Hops: EKG
Yeast: Safale US-05
O.G.: 1.042
IBUs: 20
ABV: 4.4%


I made a litre starter for the Omega lactobacillus, and let this sit (without agitation) for 24 hours before pitching.  Since each half of the brew-day would be relatively short, I managed to fit them in across consecutive week nights.  On the first day I made the wort, doing a BIAB mash with as much of the total liquor as I could fit in my kettle, then liquoring down after mashout to reach my full boil volume.  At that point I raised the temperature up to just below boiling, then quickly cooled it to around 95°F before pitching the lactobacillus starter.  I wrapped it up in some insulation and a sleeping bag (the same stuff I use if I want to hold my mash-temperature for more than an hour) and left it to sit until the next evening.

The Omega blend performed as advertised, taking the wort down to 3.33 in around 18 hours.  Once I'd confirmed this drop in pH, the rest of the brew-day was quite straightforward: just a normal boil, cooling the wort, and pitching the yeast.  I was a little worried that the low pH would inhibit the US-05, but when I checked the next day there was visible activity after about 14 hours.

Once the beer had fermented out, I transferred it to a five gallon keg, added finings, and put it on my balcony overnight (the 'Chicago winter cold-crash') before transferring to a second keg the next day.  Its been sitting there under pressure ever since: whenever I need some for blending, I just attach the keg to some tubing and measure out the proportion I need.

The beer came out as expected: clean, lemony sourness, without much else going on.  My idea was to use it in varying proportions to add some degree of acidity to different beers, from a tart acidic bite at the end of an otherwise normal beer, to a pronounced but balanced sourness.


As of writing this post I have used this batch in three different beers, each with varying proportions of acid beer blended in.  For the first I made a dry stout, since I've been brewing them a lot anyway, and I thought it would do well with a small percentage of acid beer.  The second was a wheat saison, to which I was hoping to add a more noticeable acidity.  And third, a wheat beer, which would be the most acidic of the lot.

I came up with some rough percentages before brewing, but also took the time to test them after the beers had fermented out.  One thing I'll say is that this time around it was very clear to me that the perceived acidity increases once a beer is cooled and carbonated.  I tested these blends at rooms temperature, and in every case the final beer ended up a bit more acidic than I'd anticipated.

The proportions for each blend were as follows:

Dry Stout: 5% acid beer.
Wheat Saison: 12% acid beer
Wheat Beer: 33% acid beer.

Dry Stout

A dry stout seemed like an obvious candidate for this kind of blending, and as I mentioned in the first post of this series, there is considerable historical precedent , which is nicely captured in this rather poetic description of "first-class Irish stout" from the 1920s:
There is something at unspeakably seductive and evasive of true description about a first-class Irish stout, It is extraordinarily full and round, mellow and succulent. Yet is it bitter - but that somehow you don't notice. Behind it and enriching the whole lies that soupcon of strange lactic-like sub-acidity. 
The author continues in the same vein, suggesting that its this "strange lactic-like sub-acidity" that is the key to a great stout:
To the mind of the writer it is the will-o'-the-wisp sub-acidity that does the trick in Irish stout. Take it away and you've little left but a black, heavy, dry, but very soft and full mild ale with a lot of hops in it — nothing very characteristic or outstanding. Curious that no one has succeeded in fathoming and grasping that extraordinary suggestion of a rare old vintage wine — something lactic it exposes to us — hidden away in the chocolate-coloured depths. In the export variety you get too much of this sub-acid touch and consequently too little of the limpid polished fullness. But the home consumption product has a veritable perfection of nicety of balance in this respect : it is indeed a wonderful work of the Art of Brewing.
Of course, the beer being described was probably much stronger than the one I brewed, and the acidity likely came from a beer that had been 'vatted', rather than a pasteurized kettle sour.  Still, the basic idea is the same: some small amount of sour beer blended in to add a slight but noticeable acidity to the finish.

Grist: Golden Promise (57.6%), Torrified Wheat (17.4%), US 2 Row (11.5%), Roasted Barley (7.7%), Chocolate Malt (3.8%), Dark Crystal (1.9%)
Hops: EKG
Yeast: Wyeast 1318
O.G.: 1.034
IBUs: 33.2
ABV: 3.1%
Kettle Sour: 3-5%

I'm still working on this recipe, but I felt like the beer came out rather well.  The torrified wheat gives the beer a fuller mouthfeel that belies its very low ABV.  At test-blending the addition of 5% kettle sour seemed to mainly accentuate the bitterness while adding a slightly tart snap at the end of the palate.  However, once the beer was cooled the acidity became more prominent, giving the beer a more noticeable tartness.  I enjoyed it a lot, and the keg didn't last long (why would it, at 3.1% ABV?), but I felt there was room for improvement.  I wanted a little more roastiness and bite, so for the re-brew I increased the O.G. and bitterness, along with the percentage of roast malts, and added a small late addition of Challenger.  I also reduced the kettle sour down to about 3% of the blend.

Tart Saison

For the next beer, I wanted to try making a simple and drinkable saison with a touch of tartness.  To this end, I came up with a recipe for a wheat saison, fermenting it with The Yeast Bay's Saison Blend II and their Lochristi brettanomyces blend.  After primary fermentation was complete I blended the beer with around 12% of the kettle sour (proportions were roughly measured for this one), and allowed it to sit for another month while the brettanomyces and saison yeasts continued to ferment the residual sugars in both components of the blend.  Once it had reached a gravity I was happy with, I packaged it in heavy bottles with priming sugar and allowed it to continue fermenting for another three weeks.

Grist: Pilsner (75.5%), Wheat Malt (18.9%), Torrified Wheat (5.7%)
Hops: Saphir, Bramling Cross, Sterling
Yeast: Yeast Bay Saison Blend II, Lochristi blend
O.G.: 1.038
IBUs: 23.8
ABV: 4.4%
Kettle Sour: ~12%

The final beer is almost exactly what I was hoping for.  The aroma is very fruit-forward, with a mix of citrus from the saison blend and the distinctive aromas of the Lochristi strains.  The beer itself is crisp, tart, and very easy to drink.  Not tremendously complex, but with enough going on to keep you interested.  I think the tartness works particularly well with the more fruity yeast flavours, and I don't know if I'd like this beer as much if I'd used more phenolic yeast strains.  At first I felt the beer was too tart, but as its aged the edges have smoothed out a little and I'm enjoying the slight sharpness.  Still, I'd like to experiment with smaller amounts of acid beer for a more subtle tartness in the finish.

Tart Wheat Beer

For the final beer, I knew I wanted to use a larger proportion of kettle sour, so I decided to try making a wheat beer that I could serve quickly from a keg.  The base was a pretty standard wheat recipe, with lots of European hops rather than citrus and coriander.  I fermented with with Wyeast 1318---an odd choice perhaps, but it was what I had available.  I think in future I'd use a more characterful yeast here.  

I blended the beer in a keg, with two parts wheat beer to one part acid beer, letting it sit for a few weeks at room temperature and then hooking it up to the CO2 in my beer fridge.  I felt that the beer was a bit one-dimensional at blending, so at the least moment I decided to add a small dry-hop of Citra.  This ended up dramatically changing the character of the beer, which I think would have been a bit flat without it.

Grist: Pilsner (51.4%), Torrified Wheat (20.1%), Wheat Malt (14.3%), Munich (8.6%), Flaked Oats (5.7%)
Hops: Saaz Special, First Gold, Styrian Goldings, Citra
Yeast: Wyeast 1318
O.G.: 1.049
IBUs: 21
ABV: 4.8%
Kettle Sour: 33%

The photo on the left was taken quite early on, and the beer cleared up slightly over time, though it always had a bit of haze.  It was fairly tart, but I didn't think the acidity overwhelmed the rest of the beer, which seemed quite well-rounded and moreish.  I think the higher F.G. played a role here. The character from the dry-hop seemed to increase over time, and I wonder whether the acidity played a role in that.  Ultimately, it tasted like a dry-hopped kettle sour, but I found it much more drinkable than the straight-up kettle sours I've tried, which can be hard to stomach after a glass or two.  

I'll be making this recipe again this summer, using Safale T-58 in place of the Wyeast strain, and backing down on the proportion of acid beer to get something closer to a tart wit.  I think these higher proportions would work well in a gose or other adjunct-complimented sour.  Based on my experience with this beer, I think I definitely prefer a blend of clean and acid beer to a straight kettle sour.  If nothing else, it allows for more control over the acidity of the final beer.


  1. Have you been using T-58 a lot in general? What are your thoughts on it so far? I use it a lot as a primary fermenter for beers that later get my mixed culture, but curious what you think of it on it's own. How is the attenuation compared to other Belgian yeasts in particular, in your experience?

    1. I bought a few packets because it was cheaper that way, but I've only used it in one beer so far, something based on De Ranke XX Bitter. Last time I checked on that it seemed to be either stuck at 1.020, or working very slowly. I need to check on it again. If that is typical, I doubt I'll be using it much for clean beers, though I might use it like you with a mixed culture.

    2. I use very often T58 blended with Belle Saison, it gives those good flavors belgianesque of que T58 but with the monstrous attenuation of belle. I use this same blend with my mixed fermentation, I just add the brett & bugs.

      Alone T58 it's a very pour attenuator in comparative with other belgian strains but for a beer like XX Bitter I can see why this yeast would work, it would had a more robust FG than other belgian strains.

      The one time I compare it T58 with my bleng T58/Belle in a Grisette of 1.040 in two weeks the blend fermented to 1.005 and the T58 alone stoped at 1.013.

      this was my experience with the yeast, nothing scientific just anecdotal references.

      Greetings from Mexico!

    3. Thanks Eslem. That fits with my experience so far too. I packaged the beer I mentioned above last weekend, and it had made its way to 1.012. Adding Belle Saison to increase attenuation is a great idea, I may try that next time I use a packet of this yeast.

    4. I also used T-58 quite a bit as the start for mixed culture beers. For the most part that yeast has finished at 1012 for, almost religiously. Heavy or light beers, always seemed to stop right on the dot at 12. I may have to try the Belle/T58 blend as well!