Sunday, 24 July 2016

Brew Day: Old World Saisons w/ Whole-Leaf Hops

Once I'd made the De Ranke-inspired beers and the C19th-inspired IPA, I found myself left with a lot of whole-leaf hops.  What to do with them?  Make some saisons, of course!  (Some bitters too, but that's another post.)

I've blogged before about Yvan de Baets' description of old saisons as "either sour or very bitter ... with bitterness obtained by the use of a massive amount of hops low in alpha acid".  The hops I had left over fit the bill nicely: some East Kent Goldings (3.6% AA), some Bramling Cross (3.3% AA), and some Hallertau Mittelfruh (2.9% AA).

Bière de garde

For the first beer, I wanted to make a sort of bière de garde, in the literal sense of a beer intended for ageing.  To this end, I planned a recipe with a slightly higher gravity than I'd usually aim for in a saison, 1.054.  (I bet there was a time when that O.G. would have been lower than 90% of the saisons brewed in the U.S!)  I went with a simple grist of 90% pilsner malt and 10% wheat, since the main focus was going to be the hops and the fermentation character, though I also did an extended three-hour boil to add a bit of colour and complexity.

In his essay, Yvan mentions hopping rates of between 5 and 8 grams per litre, with a third of this added at the end of the boil.  Assuming this was based on the volume of wort in the kettle after the boil, that gave me a range of 110g to 176g on my system, and I decided to go for the upper end of that scale with 180g.  That meant I needed a bittering addition of 120g and a late boil addition of 60g.  I decided to split the bittering into roughly 2/3 Bramling Cross and 1/3 Hallertau Mittelfruh, with reverse proportions for the late addition.  That gave me 57.8 predicted IBUs, or just about a 1:1 BU:GU ratio.  Higher than a lot of saisons, but not all that different from beers I've brewed before!

Primary fermentation was carried out by a blend of saison yeasts: Wyeast 3726 and The Yeast Bay's Saison Blend II.  After this was completed, I transferred the beer to a three-gallon carboy and added some random brett strains I had been storing in the fridge (the C1 and C3 strains isolated from a bottle of Cantillon Iris by Dmitri at BKYeast) , along with the dregs of a beer brewed by someone at Omega Yeast Labs, which was dosed with the brettanomyces strain from their C2C blend.  I'll let it sit for at least a few months before bottling at the end of the summer: its likely that packaging this and other beers will be contingent on finding enough heavy bottles, as I am starting to run short again.

Spelt Saisons

For the rest, I planned to make three variations on my basic spelt saison recipe, one for each hop variety.  As I mentioned in that earlier post, I think the fuller mouthfeel from the spelt helps prevent the bitterness from overwhelming the beer.  Although I was making an extra gallon to accommodate for wort lost to the whole hops, I did not vary the amount of spelt in the recipe, which meant that with a predicted O.G. of 1.046 the base was 82.6% pilsner and 17.4% unmalted spelt.  I varied the base malt slightly for some of the batches.  Details on that below.

Once again, I wanted to really push the bitterness while also getting a good hop character and mouthfeel, so I went with three roughly equal additions at 60, 30, and 2 minutes left in the boil, aiming for a BU:GU ratio of about 1:1.  Even by the standards of the beers I've been brewing lately with these whole hops, that was a lot of vegetable matter relative to the O.G. of the wort!

Here is a sketch of each beer.  For more details, I suggest you look at the post on the basic recipe linked above.

Mittelfruh Saison: This was the simplest of the three, with no modifications to the grist.  It was fermented with a blend of Wyeast 3724 and Wyeast 3726.  This one has been a little disappointing so far.  It has a slightly odd soapy taste that I can't get over.  Unless I forgot to rinse out a fermentor or bottling bucket, it must have something to do with using such a large volume of Hallertau Mittelftuh in the beer.  I've used those same hops in other beers too, but usually as an aroma hop later in the boil, and I haven't seen this same soapiness.  I'm hoping it will age out as the hops fade a bit.

EKG Saison: For this I substituted approximately 30% Golden Promise for some of the pilsner.  It was fermented with a mix of The Yeast Bay's Saison Blend II and Wyeast 3726.  This has turned out to be one of my favourite homebrews to date.  It has a striking bitterness, without being at all harsh or astringent, and an earthy and citrusy hop-character that I find very appealing.  J said it reminded her of Taras Boulba, and I can see what she means.  Its not that they taste the same, but the overall character is very similar: dry, bitter, hoppy, and very drinkable.

Bramling Cross Saison:   I didn't get round to brewing this before I left for England.  Blame the summer heat, and a sense of fatigue from brewing over the past few months.  I'll probably do this batch when I get back, substituting about 15% Vienna for some of the pilsner.  Fermentation will go one of two ways: I'll either keep it clean and keg it (only because I want to use whatever heavy bottles I can amass for other batches), or add brettanomyces and let it sit for a while.
Blogger Tricks

Sunday, 10 July 2016

Brew Day: Nineteenth Century IPA v.2

The recipes inspired by De Ranke XX Bitter described in the last post weren't the only reason I had my eye out for some fresh whole-leaf hops this year.  The IPA in the photo opposite, based on a nineteenth century, is one of my favourite home-brews from the last few years, and I've been thinking about a re-brew since I opened my first bottle at the end of the summer.  Of course, in keeping with the tradition, I had to wait for the new season's hops, which seem to arrive in the US almost six months after they were picked!

I decided to use whole leaf hops this time, both because I presume that was what the Burton brewers used, and because I hoped they might have a freshness that would have been lost in the pelletizing process.  This is certainly what the brewers at De Ranke think: in the article I linked to in my last post, Nino says "By using only hop flowers, we ensure the hop oils are never oxidised so we never get any harsh bitterness in our beers".  I used pellets in the first version of the beer, which had a distinctive mouthfeel that I associate with large volumes of kettle hops, so I'm hoping I'll only improve things by switching to leaf rather than pellet hops.

Of course, using whole hops in the brewery presents a number of challenges, which are conveniently listed on the De Ranke website:
  • When the quality of this expensive material drops, it can lead to off-flavours in the beer, which is why we work closely with a local hop farmer to ensure we always get hops that meet our high standards.
  • The annual price fluctuates and depends on the harvest.
  • The hops must be refrigerated to maintain quality all year long (which makes them even more expensive)
  • Hops can only be bought once a year, just after harvest. It requires good planning for the entire brewing season to avoid hop shortage or surplus.
  • Whole-leaf hops require lots of cleaning at the brewery. Hop cones tend to stick to everything during the boiling of the wort, which results in more work for the brewers because the hops must be manually removed, as opposed to breweries that use hop pellets or extracts, which requires far less work to remove.
This fits my experience: whole hops are messy, they take up a lot of space in the freezer, soak up a lot of beer, and make siphoning an absolute nightmare if you don't have some kind of filter or false bottom in place.  And as you can see below, these old IPA recipes use a lot of hops...

I brewed a series of beers with these hops (including the De Ranke-inspired beers from the last post, and some others that post about later), and I gradually learnt how to adjust my brew day to accommodate them.  The biggest problem is the sheer volume of hop matter in the kettle, especially when (as with these recipes) you're using an exorbitant amount of hops to begin with.  I deliberately brewed a whole gallon of extra beer, in anticipation of leaving a lot behind with the hops in the kettle.  (Even when I used pellet hops last year, I lost some volume in the IPA batch because I didn't make adjustments.)

The other problem is racking from the kettle.  If you have a fancy kettle, with a ball-valve, screen, or even false-bottom, it might not be an issue.  I rack with an auto-siphon, which quickly gets clogged by the whole hops, even if I cover the end with some kind of filter.  This meant that I had to pour the wort manually from the kettle, which is neither safe nor particularly sanitary (since I inevitably end up making a mess).  Frankly, its the kind of dumb s#@t I used to do when I first started brewing, and which I swore off entirely after having surgery to fix an inguinal hernia a few years ago.  Still, here I was again, precariously balanced with a heavy kettle as wort splashed off the whole hops over the kitchen floor.  No fun at all.

Since I get a lot of trub from my BIAB process, I generally like to whirlpool and leave a good amount of wort in the kettle.  Obviously, pouring the whole thing makes this impossible, so instead of doing a whirlpool I poured the wort into a keg as soon as it had cooled, then let this sit for a while until the trub and hop matter had dropped out of suspension.  At this point, I racked the much cleaner wort into the final fermentation vessel, oxygenated, and pitched the yeast.

Anyway, back to the beer itself.  Besides changing to whole hops, I didn't make any real modifications from the recipe I linked to above: 100% pilsner malt, elevated levels of sulphate (though not quite to the levels reported for historic Burton waters).  This time I pitched Wyeast 1318 for the primary fermentation, and the dregs from a beer fermented with Wyeast Brettanomyces Clausenii for the secondary. 

Last time the beer was all but undrinkable for the better part of eight months, so I'm not expecting to be drinking it any time soon.  I'm hoping the whole hops will soften the bitterness slightly, which might mean I'm drinking this one earlier than the previous version.  For now, its sitting in a carboy in my brew closet.

Sunday, 26 June 2016

Brew Day: XX Bitter

I think that De Ranke's XX Bitter is for me what Saison Dupont is for many American saison drinkers: one of the beers that made me fall in love with the style and start brewing it seriously at home.  That partly reflects the indirect route I took to these beers, first becoming excited about them after reading the descriptions of old saisons in Yvan de Baets' essay in Farmhouse Ales.  That essay helped me to see a pattern in the kind of Belgian beers I enjoyed and the kinds I wasn't so keen on.  And XX Bitter (along with beers from Brasserie de la Senne) came closest to how I imagined the bitter versions of those beers.

Because of that, I've been meaning to brew a batch directly inspired by XX Bitter for quite a while.  Luckily, Nino and Guido are very open about how they brew their beers, making it a principle to "offer our customers insight in our brewery, the brewing process and the materials used, with nothing to hide".  This article contains almost everything you need to know in order to brew a beer along these lines, and there is also plenty of information on the De Ranke website.  Nino was kind enough to answer a few extra questions for me as well.

Here are the main takeaways I got from all of this:

  1. The recipe is 100% pilsner malt.
  2. Whole hops are essential.  The brewers are very serious about this part of their process, stating that they "use exorbitant amounts of hop flowers, which results in unmatched complexity and mouthfeel".  This certainly fits my own experience of their beers.
  3. There are only two additions of hops in the boil.  The first is of Brewer's Gold, added with 75 minutes left in the boil, aiming for 60 IBUs.  The second is with Hallertau Mittelfruh, at two minutes left in the boil.  Nino recommended that I add a minimum of 1.5g/l, and I ended up using closer to 2g/l.
  4. The brewery uses a dried Fermentis yeast.  Discussion with other homebrewers online left me fairly certain that this was Safbrew T-58.  Nino simply suggested that I "select a yeast strain that is quite neutral so the hops will shine through better". They also "use very little cooling during the brewing process", which suggests to me that the yeast is allowed to free-rise after a certain point.
  5. The beer finishes very dry, and is given a relatively long conditioning period at fairly high temperatures.  From the article: "We condition at higher temperatures than a lot of breweries, mostly at 15°C. We also allow for 4 weeks of conditioning after primary fermentation which is longer than a lot of other breweries. This gives us a really dry beer.”
That is more than enough information to base a homebrew recipe on, though I had to make a few changes to what I brewed.

I wanted to use the freshest whole hops possible, so I waited until Hops Direct announced that their new European crop had arrived, and picked my hops from there.  Unfortunately they did not have any Brewers Gold this year.  I ordered some Bramling Cross instead, since they have a similar oil profile and flavour description to Brewers Gold (plus I knew I could use them in some other recipes).  The only problem was that the AA% was surprisingly low, at only 3.3%.  This meant I had to use A LOT of hops to get to 60 IBUs.  I thought about supplementing with pellets instead, but decided that went against the spirit of the thing: if I had enough whole hops, why not use them and see how things turned out.  I adapted my water profile to emphasise this bitterness, going for around 150ppm calcium sulfate to around 50ppm calcium chloride.

My other main concern was about the yeast.  Nino states that the beer gets very dry, but from what I could learn from other homebrewers, it seems that T-58 is not a particularly attenuative strain.  With that in mind, I did everything I could to make a fermentable wort, including a long low mash rest at around 146°F (followed by a shorter one at 154°F).  Also, though I was using dried yeast, I gave the wort plenty of oxygen before pitching.

My predicted O.G. was 1.054, but both batches ended up higher than intended, between 1.058 and 1.060, perhaps due to the longer than usual mash rests.  This probably won't help with attenuation, but I decided to just let things be rather than adding water to get closer to my intended O.G.  After pitching the yeast, I kept the beer in the mid 60s for the first 24 hours or so, and allowed it to free-rise after that.

The batch brewed with T-58 was still at 1.020 after two weeks of fermentation.  I transferred it to a second keg and let it sit at room temperature with a spunding valve, and after four weeks it was down to 1.012.  I'd hoped to get it down to at least 1.008, so I was a little disappointed with where it ended up, but I decided to just got ahead and package it rather trying to start another fermentation with a different strain.

It's been in the bottle for about three weeks at this point, and while its a nice beer, it doesn't come close to XX Bitter.  First, the bitterness just isn't as pronounced as it is in the original.   Its definitely there, but doesn't have the lingering quality that I love in De Ranke's beer.  I don't know if this is because the beer didn't dry out enough, because I used lower AA hops, or something else entirely.  The fuller body also makes it less drinkable than XX bitter.  That aside, its a nice beer: aromatically complex, with both the yeast and the hops making their presence known.  But its just not what I was going for, and disappointing for that reason.

I think if I brew this again, I'll give up on the idea of making a 'clone' and use a yeast strain that I'm more familiar with.  I actually made some other beers along these lines using these same whole hops, but they're a subject for another post.  In the meantime, I plan to set the rest of this batch cold-condition in the fridge while I'm away in England, and I'm sure we'll have no problem finishing it off once we get back.

But wait, there's more...

As Nino mentions in that article, there was a period between 1994 and 1998 when the beers at De Ranke were brewed with yeast from the Rodenbach brewery.  Yvan de Baets has a very evocative description of that beer, and the flavour profile associated with old saisons, in his essay:
It is often said that sourness and bitterness do not go well together in beer but, because [saison] was a beer that had matured for a long time, the bitterness decreased, permitting the equilibrated development of the sour and vinous flavours of the beer. We had evidence of this until several years ago when the excellent XX Bitter, a heavily hopped beer from the De Ranke brewery in Wevelgem, was still fermented with yeast from the Rodenbach brewery in Roeselare. This yeast is in fact a mix of diverse yeasts, some of which are of the Brettanomyces strain, and of lactic bacteria. When the beer was young, bitterness dominated, balanced by a light tartness. As is aged, the bitterness diminished, giving way to a more pronounced and slightly vinous tartness. The balance of this beer was always perfect. It certainly came close to old saison beers.
With plenty of whole hops left over after I formulated the recipe for the clean batch, of course I had to give this one a try as well.  The recipe was exactly the same, but instead of using T-58 I pitched a packet of Wyeast's Roeselare blend, which is intended to imitate the yeast at Rodenbach.  As with the beer described above, I tried to make sure this one finished pretty dry, giving it a long, low mash rest and plenty of oxygen. I'm hoping this will help the beer attenuate fairly low after a month or two, so that I can package it while its still quite fresh, and see it develop in the bottle.  If that isn't possible (and as of posting this, its looking unlikely), I'll leave it to develop in a carboy over the summer.

Sunday, 19 June 2016

Basic Spelt Saison Recipe

If you've been reading this blog for a while, you might have noticed that I stopped posting recipes at some point.  This was a conscious decision rather than laziness, but it didn't come from any desire to be secretive about my home-brew.  The real reason is that I just don't think that precise recipes are all that important, especially for the kinds of beer I make, and I think posting them can be a bit of a distraction. (If I were making a lot of stouts or amber ales, I might feel differently.) For me the real interest is in the thought process behind the recipe, rather than the numbers on the page.

However, when the team behind the wiki at Milk the Funk asked for some home-brew recipes from regular posters, I decided to provide one that I brew quite regularly.  And having taken the time to write out the details of my process, I figured I might as well turn that material into a post that could serve as a sort of complement to the "A Typical Brew Day" post at the top of the blog (which needs to be updated), using it to describe some of the thinking behind the recipe.  Hopefully people won't find the 'sloppiness' of my approach too horrifying!  As you'll see, there are ways in which I am a very imprecise brewer...

The recipe I provided is for a Basic Spelt Saison, i.e. a dry, bitter, hoppy beer with a reasonably low ABV.  If you want some pedigree, a version of this beer scored 43.5 at MCAB this year, and took bronze in its category.  I have made this recipe, or variations on it, at least ten times, and it probably has as good a claim as any to be my 'standard' saison recipe.  That already shows you something that I think is important in the way I think about home-brew (and that is important background to the details of my brew day): repetition.  For any recipe I'm serious about, I'll rebrew it time after time after time.

I learnt how important this is from making bread.  I'm by no means an expert baker, though I can usually throw together a decentish loaf from any recipe, and know a reasonable amount about the techniques and processes involved in baking.  But I have been making the same sourdough loaf, or variations on it, 1-3 times a week for at least the past five years.  They don't always come out great, but when I put my mind to it, I can make a good loaf, and when they don't come out well I usually know what's gone wrong.  Some of that came from reading a lot about bread-making, but most of it is from making the same loaf over and over again.

So, with that in mind, on to the recipe:


24.2% Unmalted Spelt
75.8% Pilsner


60 min - EKG - 20 IBUs
20 min - EKG - 8 IBUs
2 min - EKG 2 IBUs


Saison blend

O.G. 1.046
F.G. 1.002-4

Yes, that's it!  In a way, you don't need to know anything more, if you know your way around your own equipment.  But I'll include some commentary and suggestions , as well as some details of my process on brew-day.


First, why 24.2% of spelt and 75.8% of pilsner?  Well, after several test batches, in which I varied the proportions deliberately and precisely, I decided this was the perfect ratio...  

Just kidding.  I originally brewed a recipe with 70% pilsner and 30% spelt, the proportions provided for Blaugies Saison d'Epeautre in Farmhouse Ales, with an O.G. in the mid-1.040s, because that would give me a moderately strong (by my standards) but drinkable beer.  As I started to repeat it, I began to make some small changes.  First, a single bag of unmalted spelt from Bob's Red Mill weighs about 800 grams.  Opening two bags to get a different amount is a bit of a nuisance, especially if I'm not baking with spelt at the moment, so I decided to stick with one.  Add 2.5 kg of pilsner malt to that (an easy number to remember), and---on my equipment---you get a predicted O.G. in the range I was looking for.  That means the recipe is less than 70/30, but its close enough to not make a big difference.  My O.G. is reliably between 1.044 and 1.046.

I often vary the grist slightly, depending on what I'm going for and what grains I have on hand.  Here are some suggestions (I've listed them as percentages, but I usually round off to a convenient weight, typically 300g, 500g, 1kg, etc.):
  • Sub in 5-15% Munich or Vienna malt.  I add Vienna fairly frequently, and Munich if I'm adding other adjuncts to this base to make a darker, maltier beer.
  • Sub in 10-40% of a characterful base malt.  I've used Golden Promise, 6-row, and US 2-row, either from necessity or because I thought the flavours would work well.  
  • Add adjuncts.  The spelt gives this beer a nice full mouthfeel, which means the recipe can stand up well to relatively large amounts of sugar.  I've taken this base recipe and added a container's worth of either Candi Syrup or honey.  The latter worked particularly well. 
  • Add post-fermentation flavourings.  I've added a hibiscus tea at bottling, along with some fruity brett strains, and I thought it came out great.


The hops listed above are just a suggestion.  EKG work well, and its pretty hard to get too much bitterness from them, so I often go as high as 40 or even 45 IBUs, especially if I'm planning on letting the beer sit for a while post-fermentation.  The fuller, fluffy mouthfeel from the spelt helps the beer stand up to this bitterness, even though it finishes fairly dry.

I tend to stick to European hops, or American varieties that have some of the same characteristics, since I'm looking to both complement and accentuate the slightly savoury characteristic of the spelt with earthy, spicy, citrusy flavours. But I don't see why this recipe couldn't work with some of the North American or Southern Hemisphere hops as well.

I occasionally add a light dry-hop (1-1.5g/l), especially if I've let the beer sit for a few months during a secondary fermentation.  I'm usually going for something quite subtle here, trying to slightly accentuate existing flavours and aromas, rather than adding a new layer that screams 'HOPS!'.


I've used various blends of saison yeast for this recipe, and they all work well: just pick something that will get it fairly dry.  If I had to name one, I'd say Wyeast 3726. Recently I've been using two blends with this strain, one that is a combination of 3726 and 3724, and one that is a combination of 3726 and The Yeast Bay's Saison Blend II.

I also think this beer works well with brettanomyces.  My preference is for a more subtle brett character that emerges gradually as the beer ages, so with that in mind I prefer to pitch a small amount of brettanomyces in secondary or at bottling.  I've been using Wyeast's Brettanomyces Clausenii a lot recently (in the form of dregs from previous batches), as well as The Yeast Bay's Beersel Blend.  Based on what I've heard from a local homebrewer who works at Omega Yeast Labs, I think the brettanomyces strain in their C2C American Farmhouse blend would also work well, so you could just pitch that.

The fuller mouthfeel also means that the beer stands up well to a bit of acidity.  My preferred method for achieving this is by blending in some aged sour beer.  You may want to dial back the bitterness a bit if you're planning to do such blending.  I typically don't bother, because the hops I use rarely give a harsh bitterness, and I enjoy the changing balance between bitterness and tartness as the beer ages.


Because of the relatively large proportion of spelt in this recipe, I typically do a cereal mash. The process is quite straight-forward for Brew in a Bag, but may require some modification for other mashing regimes.  I'm still a little ambivalent about the best time to add the spelt porridge to the main mash.  Some of my recent beers made with the process below have had lower head-retention than I'm used to, and I'm still trying to work out if that is from using a more modified pilsner malt, or from including the spelt porridge in the first protein rest.
  1. Crush spelt separately to consistency of grits.  For me, that means running it through my Corona mill twice on a fairly tight setting (too tight and the mill sticks).  
  2. Bring the spelt grits to a boil in a large saucepan with a few litres of water (subtract this from the volume of your main batch, or take it directly from the liquor in the main kettle). Keep at a boil, stirring to prevent scorching, until it forms a thick porridge: usually 15-20 minutes. This stage can be done prior to brew day, with the cooled spelt porridge stored in the fridge till required.  [NB: I often throw in a handful of crushed pilsner malt as well (yes, a handful, I don't measure it).  I think the enzymes convert some of the sugars as mix passes through the conversion temperatures on its way to a boil.  Sometimes I'll let it rest for ten minutes at around 150°F first.]
  3. Heat main mash liquor and dough in with grist and spelt porridge, aiming for an initial temperature of 131F. You may need to break up the spelt porridge with your hands if you stored it before use. Keep at this temperature for around 15-20 minutes. [Optional step: you can also include an earlier rest at around 113F. This may aid with lautering and possibly increase phenolics from any brettanomyces strains.]
  4. Raise mash to around 145F. Keep at this temperature for 40-50 minutes.
  5. Raise mash to around 154F. Keep at this temperature for 20 minutes.
  6. Raise to 168F. Mash out and lauter. Top-up with water to reach your desired pre-boil volume. Proceed with boil.

Here's what I typically do post-boil:

Cool beer to around 65F. Oxygenate, pitch yeast, and allow to free-rise. (In the summer, I would keep it in my fermentation chamber set at 70F for 24-26 hours.) At the moment I prefer to add a small amount of brettanomyces after primary fermentation is underway, or in secondary.  This is because I'm looking for a slower development of the brett-related flavours.

Hopefully that was helpful, or at least interesting, to someone out there.

Sunday, 12 June 2016

Solera Top-Ups plus NHC Gold Medal

I had already scheduled this post about my most recent round of solera top-ups, but now it seems especially appropriate since a pale sour I blended from them won a gold medal in the sour category at NHC.  That beer had already done well in several competitions, taking a gold in MCAB and the first round of Nationals, along with a silver in the Drunk Monk Challenge.  I was pleased about the MCAB medal, but a gold at NHC is even more exciting.  I've given bottles to other home-brewers, and they all seem to have enjoyed them (and gave me helpful feedback).

I think the beer's success stands as a proof of the original concept behind my six-gallon soleras: my idea when I started them was to have a range of beers with differing flavour-profiles available for blending.  The scoresheets I've received so far have all commented on its aromatic complexity, which I attribute to the fact that its a blend.  Blending also gave me greater control over the overall acidity.  It wasn't exactly my intention, but I think part of the reason this beer has done so well is that there is still a slight perception of sweetness balancing its tartness.  I was relying on ongoing fermentation of some of the residual sugars for carbonation, and they haven't been completely fermented out yet.  This slight sweetness fills out the palate of the beer, accentuating its citrusy flavours, and it will hopefully give the beer some longevity as those sugars are gradually fermented.  You can read about each of those soleras in the links here, and about the particular components and ratios I used in the winning beer here.

That blend is quite bright and fruit-forward, with lots of citrus and pineapple, and only a light barnyard funk.  Its also rather pale, and doesn't have any of the residual earthy bitterness that I sometimes pick up on in certain gueuzes.   For this most recent round of top-ups, I decided I wanted to see if I could steer some of the soleras towards a more classical lambic profile, and to this end I changed a few key parts of the process.

As of writing this, I've only brewed top-ups for the Roeselare and ECY20 pale soleras, and I'm thinking about just leaving the Melange one as it is until after the summer.  This is mainly due to time constraints and other beers I have planned.  I varied the recipe and process for both pale soleras this time.  First of all, I switched to a simpler recipe of 65% pilsner and 35% unmalted wheat.  The wheat was very hard and more difficult to mill than either spelt or buckwheat.  I didn't do a full turbid mash, but instead used the Wyeast Lambic Mash Schedule described on p.142 of Wild Brews.  Unless I'm missing something, this is just a cereal-mash for the wheat followed by a regular mash held at higher temperatures.  I don't know how effective this will be in leaving starches in the wort, but we'll see.

The next variation was in the length of the boil.  I've been experimenting with this in a few beers recently, and have noticed a quite dramatic change in colour from extending to three hours or longer.  You can see samples from the beginning and end of each boil in the photos below.  Just to avoid any confusion: when I lengthen the boil in this way, I also liquor down so that I am starting with a larger volume of wort in the kettle at the start of the boil, and end up with the same amount of wort as I would from a regular boil.  That means that the darker colour is not just a result of a more concentrated wort.  You'll have to take my word for this, but the same amount of malt with my regular starting volumes and boil length would produce a much lighter post-boil sample.

I also increased the amount of aged hops in these batches, using 50g in one and 80g in another.  The hops were pellets purchased as 'Lambic Hops' from Farmhouse Brewing Supply, and allowed to age further in a brown paper bag in my closet.  I was a little concerned that they would add too much bitterness, as I've had that problem with other supposedly 'aged' hops that I've bought in the past.  However, although there was definitely some earthy bitterness in both beers post-primary, neither seemed overly astringent or harsh.

Primary fermentation was done by Wyeast 1318, a strain I've used fairly frequently in the past.  I also added the dregs of a bottle of gueuze for the primary as well, hoping that this will give anything alive in the bottle time to get established before each batch was added to the solera.  After primary was complete, I racked out three gallons from each solera, storing the Roeselare in a three gallon carboy (until I used it in this blend), and the ECY20 in a mix of gallon and half-gallon jugs.  I then topped each up with the three gallons of newly fermented beer.

I am planning to check on these in mid to late Autumn.  Perhaps the solera will speed along the development and fermentation of the whole, in which case I may try some blending again at that point, using anything left over from this pull as an additional component.  If they're not quite ready yet, I'll still brew a top-up, and just rack off three gallons of each into another carboy so that it can continue to mature separately.

For the red solera, I knew that I wanted to pull off four gallons: three to mature in a separate carboy for future blending, and one for a bière de coupage made with tart cherries.  I also switched recipes for this batch too, using a version of the RU55 recipe posted by Jester King.  I thought the colour was quite lovely.  in fact, I liked it so much 'veI decided to make a second batch to age independently of the solera.

Once again, I used Wyeast 1318 for primary fermentation (I think it works particular well in these beers), and will add dregs from a few home-brews and commercial beers to the smaller batch over the course of the first week of fermentation.  The larger batch will be added to the solera 'clean'.

I'm planning to blend something with these late in the year. I still haven't made an unfruited Flanders-style beer that I've been really happy with, so a lot of what I pull might end up on fruit.  The rest will be blended with either older pulls from the solera, or pulls from the separate, stronger, brown solera.

Sunday, 5 June 2016

Bière de Coupage: Boxed Lambic

In the previous two posts in this series I described my experiences making bières de coupage with both kettle-soured and mixed fermentation home-brew.  The inspiration behind all of this was the ongoing tradition of using aged beer from lambic breweries for blending, and when I conceived of this series of posts I knew that I wanted to look into making such a blend myself.

First, I needed to get my hands on some lambic.  As these advertisements from old volumes of Le Petit Journal du Brasseur show, lambic brewers have a history of selling aged lambic for blending, which continues in the present day sale of `bag-in-a-box' lambic from certain breweries.  These boxes are occasionally available from Belgian webshops that ship to the US, but the price of shipping makes them prohibitively expensive, to the point where for a while I felt that I couldn't justify the cost (they would be much more affordable if I still lived in England!).  I had all but given up on the idea, when a fellow brewer from my homebrew club offered to split the shipping on an order with me: it only saved a little bit of money, but combine that with an upcoming birthday and I had all the excuse I needed to ask for a box of lambic, along with a few examples of bière de coupage not available in the U.S.

My original plan was to purchase a single box of Oud Beersel lambic, which I would then split between two 11.4 litre (3 gallon) batches of beer.  Each box contains five litres of lambic, so subtracting about ~400ml for tasting and other analytics, that would leave 4.6 litres to divide between the two batches, or 2.3 litres each.   Although I was tempted by the idea of just adding the lambic to each batch at bottling, and relying on the fermentation of residual sugars for carbonation, this felt like too much of a risk for such an expensive batch of beer, so I instead decided that I would simply remove 2.3 litres of beer from each carboy, and replace it with lambic, giving me a blend of approximately 20% lambic to 80% home-brew.  I could then leave the beers to undergo a secondary fermentation for a few months before packaging.

The Home-Brew

The first thing to do was to make the home-brewed beers.  I settled on two recipes, one for a 'basic' saison made with 100% pilsner malt, and the other a variation on a spelt saison that I've been making a lot recently.

Spelt Saison

Grist: Pilsner (60.6%%), Unmalted Spelt (24.2%), Vienna (15.2%)
Hops: EKG
Yeast: Saison Blend (Yeast Bay Saison Blend II, Wyeast 3726)
O.G.: 1.044
IBUs: 28.3

I have a separate post scheduled about the spelt saison recipe, so I won't write much about it here.  It typically includes between 20-25% spelt, often with something like Vienna or Munich malt to give it some additional character.  In this case I went with Vienna, and backed down slightly on the IBUs.  It was fermented with a blend of The Yeast Bay's Saison Blend II, and Wyeast 3726.

Basic Saison

Grist: Pilsner (100%)
Hops: EKG
Yeast: Saison Blend (Wyeast 3724, 3726)
O.G.: 1.060 (Predicted: 1.055)
IBUs: 36.6

I've been mostly using cheap North American pilsner recently, but I decided to splash out on some Weyerman Bohemian pilsner for the basic saison.  This is supposed to be slightly under-modified compared to many other modern pilsners, and I've enjoyed its more pronounced flavour when I've used it in the past.  I put the grist through the mash-steps I use for adjunct-rich beers, with rests at 131F, 145F, and 154F.

Since I knew the beer was going to be blended with lambic, I also decided to do a longer boil, hoping to match the richer colour and perhaps some of flavours of the lambic.  In the end I settled on a three hour boil, which was shorter than the one I used for this bière de garde, but still long enough for a noticeably darker colour (relative to both the start of the boil, as shown in the picture below, but also to the same volume of wort of comparable strength made with a shorter boil).

The rest was all quite straight-forward: EKG early and mid-boil to around 36 IBUs, and a blend of Wyeast 3724 and 3726 for primary fermentation.  The original proportions were approximately 70% 3724 to 30% 3726, but this was the second or third generation, so I have no idea how that balance had changed.  I was aiming for a O.G. of around 1.055, but between the longer boil and the step-mash with an unfamiliar malt I ended up overshooting by 5 points.

The Lambic

While I was doing all of this, my friend placed his order at Belgium in a Box, and it arrived about a month after I brewed these beers.  As I said above, I only ordered a single box of Oud Beersel lambic, and had planned to split it between the two beers.  However, when my friend dropped off my order, he had added an extra box of Timmermans from his own stash!  Incredibly generous, and very exciting for me.  (Thanks Tarsicio!)  That made me re-think my plans slightly.  I knew I wouldn't have time to brew many more beers before leaving for the summer, so I decided to make an additional blend with beers I had in my brew-closet.  But first, the lambic...

I've copied out my tasting notes for each box below.


Gravity: 1.009
Tasting Notes: Nutty aroma, almost like marzipan.  Assertive, lemony sourness.  Nice but one-dimensional.  Will add a pleasant acidity.

Oud Beersel

Gravity: 1.006
Tasting Notes: More complex than the Timmermans, with a mix of bright fruitiness and light funk.  Sulphurous note that is a bit overwhelming at first.  Bit of plastic with the fruitiness.  Only lightly tart, compared to the pronounced lemony acidity of the Timmermans.  Preferred blending component.

The Blends

Since I had more lambic than I had anticipated, I let myself take larger samples from each box for tasting, gravity readings, and blending experiments.  Luckily I had just made some pulls from two of my soleras (ECY20, Roeselare), so I had some aged pale sour on hand in carboys and jugs.  The ECY20 pull has a pronounced lemony sourness, along with a bit of hay and light funk.  I decided to keep this for blending with other home-brew, as I like the acidity it adds, but that wasn't needed here.  The Roeselare pull (which has had a lot of dregs added over the year) was more funky, with a softer acidity.  I felt like this worked better with the boxed lambics. I also had five gallons of an adjunct sour that I made a few years ago.  For a long time this suffered from a strange minty/herbal taste that I attribute to 'aged' hops that added more bitterness than I expected, but that flavour was finally beginning to fade, and I found that the beer added a nice complexity in blends.  So that gave me the following components to play with:

  1. Basic Saison
  2. Spelt Saison
  3. Roeselare Solera
  4. Adjunct Sour
  5. Oud Beersel Lambic
  6. Timmermans Lambic
Blending these together was fun, but only moderately informative.  As with my previous blending session, I found that I got palate-fatigue pretty quickly, and that it was hard to make well-founded decisions between different possible blends, beyond ruling out the ones that I felt had obvious flaws.  In the end things went the same way as last time: I was testing out pre-conceived blends to see if they would work, rather than coming up with a blend on the spot.

I decided to stick with a variation of my original plans for the saisons.  Instead of splitting the Oud Beersel between both batches, I used it in the basic saison, and used the Timmermans for the spelt saison.  I racked out approximately 2 litres from each saison, and replaced it with an equivalent amount of lambic, for a blend of ~18% lambic to ~82% saison.

This left me with approximately 2.5 litres of each lambic (subtracting the 0.5 litres I took for sampling and testing blends).  I combined this in a five gallon carboy with 11.4 litres of pale sour racked from my Roeselare solera, along with the 2 litres of basic saison described above and approximately 2.5 litres of two year old adjunct sour that went unused in my Autumn blending.

All three blends have been sitting in carboys since blending, and each has shown some signs of fermentation. The Oud Beersel box was clearly alive and fermenting on arrival, as the bag was swollen to the point where I was worried it might burst, and continued to swell after I had let out some of the gas.   I didn't see any comparable activity from the Timmermans, but a pellicle has formed on the blend, so I think it also contained active yeast and LAB.  I haven't taken new gravity readings, but I know that I won't be able to package these beers for a while because I'm running low on heavy bottles.  I'm hoping I can bottle them before I leave for England in July, but it may have to wait until the end of the summer. 

Monday, 30 May 2016

Bière de Coupage: Aged Sours

For this next post in the series on bière de coupage, I'm going to try to summarize my experience to date of making such beers with aged sours, and talk about some recent batches made with this process.

For a while I thought that coupage was something brewers did after primary fermentation.  That's certainly true of most contemporary versions, but Dave Janssen directed me towards some material that suggests that  this wasn't the only process used by earlier brewers.  Some of the details are still unclear, but at least some of the material that Dave has passed on to me suggests that in some cases a small portion of aged beer was added at the beginning of fermentation, rather than after primary fermentation was complete.  This practice is often mentioned in connection with seasonal brewing, and its a running theme that coupage is a necessity if brewing in the summer.  This fits with the idea that the aged beer was used to steer fermentation in the right direction, by inoculating with bacteria and yeast that had proven themselves to make a positive flavour contribution, and also perhaps by lowering the pH of the wort to a point that would inhibit spoilage organisms in the early stages of fermentation.  (I do not mean to suggest that the brewers understood it in exactly these terms.)  This process is closer to using aged beer to inoculate a solera than it is to using it to contribute particular flavours to a blend.

The way I typically use aged sours is somewhere between these two practices.  My primary purpose is to modify the flavour of the young beer, adding some aromatic complexity and a degree of tartness.  But I also want the blend to slowly evolve over time, as the bacteria and yeast from the aged sour gradually take over the flavour profile.  For me, noticing how a batch gradually changes is one of the pleasures of home brewing, and I've started to think a lot about how to brew beers that will go through a positive evolution as they get older.  I'll say a bit more about that below.

When it comes to choosing base beers for this kind of blending, I generally follow this advice from Garrett Crowell at Jester King: "Beyond both components of the blend being dry/done with fermentation, there aren't any special requirements for making biere de coupage [except perhaps] a nice level of acidity in the aged beer".  Although this can be hard to predict, I think the most important thing is to have some picture of where you want the final blend to end up.  I like this general advice from Ron Jeffries, found in Brew Like a Monk, even if its a little harder to apply to blended beers:

Key in writing a great recipe [or making a great blend!] is to begin with the end.  You must begin with the finished beer.  You need a very clear vision of the beer you wish to brew.  How does it look, what color is it?  Does it glow; does it sparkle, or throw a yeasty haze?  How does the head look, its color, texture, and lacing.  Breathe in deeply through your nose.  How does it smell?  What does the aroma tell you about the beer?  [Etc.]

Sometimes I want to accentuate and the fruitier flavours of a base beer with some tartness, and compliment them with some of the sharper citrus, hay, and overripe fruit you can find in aged sours, so I look for beer that I think will pair well to this effect.  Other time I want to add a soft layer of barnyard funk and a lighter tartness to complement a more phenolic or earthy-hop-driven saison, so I'll use a blending component that I know has some of these aromas and flavours. Over time I've found myself preferring base saisons with a fairly clean fermentation profile, since I now want any brettanomyces-related funk to emerge slowly after blending.  I also don't mind a bit of bitterness, even when blending with a tart beer, since what I want is for the balance between this bitterness and the tartness of the sour to gradually shift as the beer ages.  My guide here is Yvan de Baets' description of Rodenbach-era XX Bitter: "When young, the bitterness predominated, balanced by a light tartness.  As it aged, the bitterness diminished, giving way to a more pronounced and lightly vinous tartness."

Using dry/fully-fermented beers is certainly the easiest way to make these blends, and I take the usual steps to ensure that the base beer finishes as low as possible (step mash, plenty of oxygen, healthy yeast).  My saisons can end up anywhere between 1.006 and 1.000, depending on the blend of yeast I'm using, and the aged sours are typically in a similar range.  If I want to blend straight-away without worrying about over-carbonation I try to get the saison as dry as possible.  One way to do that is to use a highly attenuative strain like Wyeast 3711, which can take a beer down to 1.000 or lower.  I'm not a big fan of the flavours from that yeast, but one technique I've used with some success is pitching a small proportion of an attenuative strain along with my primary yeast.  A dried yeast like Danstar's Belle Saison is a good candidate here, because you can weigh out a tiny pitch and seal up the packet with a vacuum-sealer.  Another option, if you're planning to let the beers age a bit before blending, is to pitch some brettanomyces along with the saison yeast.  I still do this if I know the beers are going to sit for a while, but in general I want the base beer to have a relatively clean flavour profile at blending, so that the brettanomyces can gradually make its presence felt as the beer ages, rather than dominating from the start.

Most of the bières de coupage I've made at home have used a pale sour from one of my various soleras for the sour component of the beer.  Whenever I take a draw from a solera, I put some of it in gallon and half-gallon glass jugs to save it for this kind of blending.  In the past this has meant that I've varied the final proportions of a blend based more on convenience than tasting considerations.  If I'm planning to blend in a five gallon carboy, I'll make four gallons of saison and blend it with a one gallon jug of pale sour in the carboy.  That gives me a blend of 20% sour to 80% saison.  If I'm making a smaller batch, I'll usually make three gallons of saison, move it to a carboy, and then rack out and replace some proportion with the sour.  In some cases, this means racking out to fill a 1/2 gallon jug, and replacing with the same volume, i.e. roughly 15% sour to 85% saison.  In other cases, I've split a single half-gallon jug between two three gallon carboys, i.e. roughly 8% to 92%.

I can't always predict the final profile of the beer based on the way it tastes at blending.  One reason is that, if there is enough residual gravity left in either beer, there may be further fermentation afterwards.  Depending on the age of (and relative health of the yeast and bacteria in) each beer, the sourness will either increase rather quickly after blending, or increase more slowly in the bottle.  As with the blends made with my kettle sour, I've found that a small amount of acidic beer makes a quite noticeable difference to the perception of the final blend.  I would hesitate to go over 20-25% sour beer, unless I wanted to make something with a very pronounced acidity.  5-8% seems to be a good proportion for a background tartness, and 8-15% if you want something a little more pronounced without making the beer 'sour'.  Of course, all of this will be relative to the sourness of your acid beer, as well as your own palate.

So far I've usually let the beer sit in the carboy for at least a month before packaging, which gives it a chance to stabilize.  I have occasionally made blends at bottling, and its something I'd like to do more frequently.   The trick, of course, is getting a predictable level of carbonation after packaging, which means either ensuring that both components are as dry as possible before blending, or factoring in any further fermentation into how you calculate the amount of priming sugar.  (Jeffrey Crane's calculator is an excellent tool for this, but only if you can predict the terminal gravity of both components of the blend.)  The most interesting option---one that's inline with what many Belgian producers do---would be to rely solely on residual sugars in the blend for carbonation.  I haven't been brave enough to try this yet, but its something I'd like to do in future.  It certainly requires more patience, as the beer will need a longer period in the bottle to reach condition.  There is also the fear that the beer will over-carbonate, or fail to carbonate at all.

Old Batches

While writing this post, I took the opportunity to revist some previous batches of bière de coupage.  One recipe for which I've used this technique consistently is my buckwheat saison.  This is one of those beers where I'm looking for a more pronounced tartness to complement the fruit-forward flavours of the base, so five of the six batches I've made have had a ratio of 20% sour beer to 80% saison.  Some of the saisons underwent mixed primary fermentations, whereas other were fermented clean.

The batch I tried while writing this post was the third I made, and is almost a year old at this point.  It is very fruit-forward in its aroma and flavour, with lots of lemon up front (both in the flavour and acidity), and some more generic citrus and stone-fruit in the background.   The buckwheat still gives this one a fairly full body, along with great head retention, but the beer still finishes dry.  Its been pretty well received by most people I've given it to, and is one of my favourite batches.

The second was one of the spelt saisons I brewed a few months ago.  This one was aged with a small pitch of the Dupont Lochristi blend, and had approximately 1/4 gallon of ages sour blended in, to make up 8% of the final blend.  Both beers were fairly attenuated at blending, and I let the blend sit for a while before packaging, at which point it had reached 1.002.  I had hoped it might drop a little further in the bottle, but there's no sign of that yet.

At this point the beer has a noticeable tartness, but not one that dominates the rest of the beer.  There's is also still a slight tannic bitterness at the back end, which I rather enjoy, since it adds a bit of complexity to the finish.  The Lochristi strains are recognisable in the aroma and flavour of the beer, giving a soft fruitiness that is accentuated by the tartness.  My biggest criticism of the beer is that it is only moderately carbonated, where higher carbonation (modified by the softer mouthfeel of the spelt) would really help some of the flavours to pop out.  If I were to make this batch again, I would add the Lochristi strains along with the aged sour (or use a sour that had some of those strains), and package the beer at blending.  My hope would be that the flavour profile would gradually shift over time, and that the carbonation would steadily increase without becoming dangerous.

New Batch

While working on this series of posts, I also decided to make a bière de coupage that was a little different from the ones I've made before.  For the most part, I've tended to stick to fairly classic saison profile for the blended components, particularly in their hop profile.  This time I decided to go in a different direction, taking my basic spelt saison recipe, but using American hops for the flavour additions instead of something like EKG.  In the end, I went with Ahtanum, Amarillo, and Citra for the kettle hops (keeping additions fairly small), as well as a light dry-hop with Amarillo.

I blended in a small portion of aged sour from a pale solera as I racked the beer to secondary (probably around 5%, though I did it by eye), and let it sit for a few weeks before adding the dry-hops.  Once these had settled, I tasted the beer in preparation for bottling.  The dry-hop had given it a very perfumed aroma, with lots of citrus and tropical fruit, and a touch of funk underneath it.  This seemed promising, but I decided that the beer needed more tartness to make all of this jump out in the flavour.  With that in mind, I added an extra 1/2 gallon of aged sour to the batch at bottling (in the photo below, you can see the parts of the dislodged pellicle spiralling down through the glass jug).  I will give the beer at least three weeks before opening a bottle, and won't drink most of the batch until after I return from England at the end of the summer.  It will be interesting to see how much of the hop aroma survives at that time.

Grist: Pilsner (65.7%), Unmalted Spelt (23.9%), Vienna (10.4%)
Hops: EKG, Amarillo, Ahtanum, Citra
Yeast: Saison Blend (Yeast Bay Saison Blend II, Wyeast 3726)
O.G.: 1.046
IBUs: 28.1
ABV: 5.8%

That's it for this post.  In the next part of this series, I'll be writing about some rather special bières de coupage, made from home-brew blended with genuine lambic.