Friday, September 26, 2014

Sour-Worted Soma

One of my main homebrewing projects/focuses for the next year is developing ways to get a reliable and repeatable tartness or light acidity into beers which I ferment 'clean' (meaning without Brettanomyces or bacteria, only Saccharomyces). My basic goal for doing this is to ferment a 'clean' beer in a normal ale time frame while still achieving a refreshing tartness. I am definitely not interested in using this method to provide all the acidity in a beer that is fully sour. For these beers I will still use mixed fermentations, where the flavor development during a slow acidic fermentation will result in what I think is a much better beer than relying only on quick pre-primary souring. I talked about more about my motivations for pre-primary souring and methods like sour worting in this post.

I've frequently used pre-primary souring methods in my saisons to add a tart edge to an otherwise normal beer. I find this helps the refreshing aspects of the beer and, in this style, that refreshing character is an important part of what makes the beer. Usually I use acidulated malt out of ease, but I've done some sour mashes with good results. This beer is based off a recipe which I initially sour mashed, but this time I wanted to try sour-worting (which I learned of from the Mad Fermentationist). This technique involves souring wort with Lactobacillus and then re-boiling this soured wort and fermenting it without bacteria. In my case I decided to only sour a portion of the wort which I pulled off a previous batch and I would blend back in during the boil of this beer.

The soured portion after adding it back in the boil
A couple weeks before this brew I brewed a similar saison which ended up going into an oak barrel, and I took the opportunity to produce some extra wort to pull off for souring. I decided I would leave the batch at room temperature as I had time to wait and while lacto may thrive at elevated temperatures, it certainly works fine at ambient temps as well. There was a small snag, that I didn't have any Lactobacillus around. So I used the dregs for The Bruery's Hottenroth, a beer almost completely fermented with Lactobacillus. Unfortunately I didn't get a chance to make a starter with that and while it did add a bit of acidity, it took some time to get going. Fortunately I was able to pick up a pack of the Wyeast seasonal Lactobacillus brevis. So after about a week and a half of mild souring I added a the lacto, which dropped the pH down to 3.5 over the next week. At that point I was happy with the acidity of the sour-worted portion and I was ready for the brew day. I've made sure to leave some of the wyeast lacto around so that I'm prepared for future sour worting trials and I don't have to go through quite as much extended pre-souring.

Target OG: 1.052
Volume: 7.5 gallons
IBU: ~25-30, though I didn't end up calculating it for this batch
FG: ~1.002-1.001
ABV: 6.7%

86.4% Doehnel malt #25 (1.5 L light pilsner), more about Doehnel malts here
9.1% Doehnel malt #27 (6-7 L Light Munich)
4.5% Acidulated malt (Weyermann)

85g Czech Saaz pellets, 3.6%aa
29g Hallertauer Mittelfrueh pellets, 4.8% aa
57g Tettnang pellets, 4.3% aa

1/2 tsp Wyeast yeast nutrients
My standard Wyeast 3724 and 3711 blend (more about blending here)
Dregs from The Burery's Hottenroth
Wyeast 5223 Lactobacillus brevis
Yeast Bay Lochristi Brett blend

Salts and other:
8 g CaSO4, 6 g CaCl2 in 6.5 gallons of mash water (+120 ppm Ca2+, +181 ppm SO42-, +79 ppm Cl-)
5 g CaSO4, 2 g CaCL2 in 4 gallons of sparge water (+101 ppm Ca2+, +184 ppm SO42-, +43 ppm Cl-)
Note: Victoria's base water is very low in minerals, so the final mineral concentrations are basically the same as the added concentrations.
1 tab whirlfloc

The Mash
Mash Plan:
146F (63.3 C) for 50 minutes, 5-10 minutes of recirculating with heat to raise the temp to 154 (67.8 C) for 20 minutes. 1/2 of the acidulated malt will be mashed with the rest of the grist and 1/2 will be added to the mash with 10-15 minutes left. I overshot the 154 by a bit and ended up spending the last 20 minutes between 156 and 158 F (68.9-70 C). This shouldn't really be a problem as the bulk of my conversion happened during that first 50 minutes and the second step is to finish everything up quickly (and it would probably happen more quickly at 156-158F/68.9-70 C). So it should not have a huge influence but it was definitely not my goal.

Collecting runnings
Boil Plan:
90 minute boil. After 30 minutes of boil the sour portion (now at pH = 3.5) will poured into the wort. A small portion of the soured portion will be left behind and final wort will be added back to the sour carboy to allow the beer to continue to sour with lacto in the primary. 3 hop additions: 42 g Tettnang and 56 g Saaz with 20 minutes left in the boil, 15 g Tettnang and 15 g Saaz with 5 minutes left in the boil and 14 g Saaz and 28 g Hallertau at flame out for a 15 minute hop stand.

About to finish the boil
After blending in the pH = 3.5 soured portion to my pH = 5.1 wort, the pH dropped to 4.2. It smelled nicely lactic and tasted good at this point. You do have to be careful with too low of a pH inhibiting Saccharomyces (I don't know what a typical threshold value would be), but I didn't notice any problems like that with this batch.

I used a blend of 3724 and 3711 (~80% 3724 and ~20% 3711, ~255 billion cells total) into the two carboys - one clean and the live lacto carboy with wort added back in (more on how I blend yeasts here). I pitched the blend at 68 F (20 C) and steadily to 78 F (25.6 C) over the first 2 days. I held the temperature steady at 78 (25.6 C) for the next 4 days and then at day 6 I raised the temp to 80 (26.7 C) until the beer reached it's terminal gravity, at which point I turned off the aquarium heater I was using to keep it warm and let the temperature fall down to the upper 60's (~20 C), which is my room temperature.
The bottled beer

I bottled the beer on August 12th. I am reasonably happy with it and find that it is improving with a bit more time in the bottle. I'll give it full tasting notes in the next week or two.

Sunday, September 21, 2014

Yeast growing and blending

Many homebrewers out there have a good idea how to grow yeast into pitchable quantities from smack packs or tubes and then pitch this into wort, so this will mostly be focused on how I grow up vials/tubes of yeast into multiple vial/tube equivalents of known cell counts as well as how I do my yeast blending and why I do it that way.

Generally when I am blending yeasts where the blend ratio is important to me I am brewing saisons. But there could be other reasons out there that folks might want to blend yeasts and this works just as well. I don't apply this approach to adding Brettanomyces to beers. When I am brewing mixed fermentations with brett I usually just add a (relatively) small amount of cells (either after primary or when I add my primary yeast) and don't worry much about pitching rates. Okay, so on to the first step, growing the yeast.

Sanitizing jars
I get packs/vials of the individual yeast and grow them up on a stirplate, usually in about 2 L increments because that is convenient sizing for me. I do this whether or not I will end up blending yeast. According to the Mr. Malty pitching rate calculator, 1 fresh vial/pack (~100 billion cells) into a 2L starter on a stirplate yields about 300 billion cells. When the starter is done I cold crash it, decant the starter wort (it is important here to use a magnet to hold the stirbar out of the way so it doesn't disrupt your settled yeast cake), pour in boiled cooled water to stir up the yeast cake until I get a roughly homogenous slurry and then split the yeast equally into 3 mason jars. I now have 3 equivalent vials/packs from the one I started with (which I why I chose the 2 L size to begin with).

Filled with boiling water to cool
With these jars I will now repeat the process to get either the quantity of yeast I'll want to pitch for a future batch or to split again into 3 jars of approximately 100 billion cells each. Sometimes I'll split into more or fewer jars and/or grow at different numbers of cells, but these are probably the most frequent ways I do it. When I have this process on the go for multiple different yeasts at once, as I usually do, it is important to keep good notes of what is going on. I keep a piece of paper as a log for all the yeasts on the go of when I made starters and how many cells I expected them to produce. And each jar gets a label of what is in it, when it was added to the jar, how many cells I think are there, and what generation of cells it is (I treat the vial/pack as generation 0, the cells I've grown from the first starter I call generation one, and cells I grow from a generation one jar I call generation 2, and so on).

Magnet the stir bar out of the way so it doesn't disrupt the yeast cake
Growing up yeast in this way has two main advantages that I see:

1) I keep a relatively good constraint on my cell counts. I am not actually counting cells with a microscope and a hemocytometer, but I trust the work behind the Mr. Malty calculator and they numbers it gives. And even if my numbers are wrong, they will be wrong consistently so this won't really be a problem. By pitching rate trials where I vary the cells/ml/degree Plato added to a given beer, I will be able to determine if, with my process, under or overpitching to a standardized value is best. And I can adjust what I am doing until I am happy with the results, regardless if that makes me 'overpitching' or 'underpitching' relative to a calculation. Bottom line: process is controllable and consistent!
Three roughly even splits of the yeast

2) I feel that the dirtiest parts of my process are chilling, where the lid of my pot is cracked open to fit my immersion chiller and therefore the cooling wort is susceptible to airborne bacteria and yeasts, and at bottling. Not specifically the act of bottling making the beer in the bottles unclean, but bottling making the neck of the carboy (which I will have to pour a yeast cake over) unclean. Growing all my yeast on stirplates does take a bit of time but it avoids the possibility of introducing and concentrating bacteria and wild yeast picked up through the entire brewing process.

So overall, I feel I am able to more cleanly grow yeast on a stirplate and it gives me good and repeatable control on cell counts, which is not something you get from a yeast cake. And having some constraint on cell counts will be necessary if you want to blend in any sort of regular and controlled way. So now that I have jars with controlled amounts of different yeasts, how do I go about blending them?

~100 billion cells each of  2nd generation WLP 007 added to jars on Sept. 18th
First you'll need to have wort around that you want to blend yeasts into, and of course those yeasts. If you have a known volume and strength of wort (going with your target numbers will work fine here for setting the blend up, but being prepared to make minor adjustments if you don't hit your targets is probably a good plan) then you can figure out how many cells you are ought to add with the Mr. Malty pitching rate calculator. Now you need to come up with a blend ratio of your yeasts. Choose a ratio based on yeasts characteristics like flavor profiles and what you are looking for in your finished beer as well as other yeast characteristics like attenuation. If you don't have one in mind or your ratio is flexible, you can choose one based on what is a convenient split for the yeast counts you have around.

Let's say you have one yeast around at 100 billion cells/jar and another at 50 billion cells/jar and you wanted to target something like 70% yeast 1 and 30% yeast 2. And let's say the Mr. malty calculator says you need 250 billion cells for your wort. Well 2 jars of 1 and 1 jar of 2 gets you right at your target total cells and is a blend of 80% 1 and 20% 2. If that is an ok blend ratio compromise for you, mix the three jars together into one of the jars (it might be helpful to pour off some liquid from the jars first so they fit fine), stir up the mixture and add it to the carboy (or you could just add the 3 jars individually).

If you run into a more complicated situation where you need to split up jars, split one blend into multiple carboys, or try the same 2 yeasts in different blend ratios in 2 different carboys, it might get a bit trickier. But definitely still manageable. It is challenging to blend cleanly by volume, but quite easy to do so by weight. The only issue is that if you over-pour, you generally can't take it back.

To work this out you will want to either know the masses of your jars when they were empty or have an empty jar of the same type around for an approximate empty mass. Let's look back at the situation above to work this out: wort that needs 250 billion cells and you want to add 70% yeast 1 and 30% yeast 2. Well, 0.7*250 = 175 billion cells of yeast 1 needed. So 1 full jar and 3/4 of a second at 100 billion cells/jar. Now if you weight the full jar and either weigh a similar empty jar or know the tare weight, you will know the mass of liquid and yeast in the jar. If you mix this up so it is roughly homogenous, you will be able to calculate the mass of 3/4 of that liquid. So now you can pour out until you have poured 3/4 of the weight into the first jar (pouring a bit less than you think, checking, and adding more if necessary is a good safeguard here). It helps to have a second pair of hands for writing down numbers and helping to remove lids, but it can definitely be done alone. Now you will need to follow the same process with yeast 2 to achieve 75 billion cells. In our example at 50 billion cells/jar that is 1.5 jars. So one full jar added to the other mixture and half of a second, with the second jar split in half as described above.

The more you go through this pouring the better you'll get at it. And don't worry too much if you mess up and pour too much. As long as you've recorded the masses you poured, you can calculate what you did and you can re-create it in the future or compare it to another blend next time. It obviously gets more complex as you introduce a 3rd yeast, different blend ratios of the 2 yeasts into multiple carboys, or the same blend but into 2 carboys of unequal size (in this last case just calculate how much you need in each carboy as well as how much you need total, then make the blend for the total amount in one jar and pour by mass into the 2 carboys). If this is something you'd like to try, and it can work pretty smoothly once you've gotten used to it, I'd suggest starting with some simpler blends to get a hang of the process. And if all that seemed like too much work or it's not for you, the first parts of growing up yeast cultures are a great way to stretch one pack/vial into multiple batches while being able to have good control over pitching rates.

If you're looking for more info on blending yeasts, Kara Taylor from White Labs gave a talk at NHC in 2014 about the topic (old NHC seminars are freely available to American Hombrewers Association members). And Zymurgy had an article in the May/June 2013 issue.

Also, here is a worked example of what I did for a more complicated yeast blend involving three yeasts and three different clean (no brett or bacteria) treatments in carboys.

Saturday, September 13, 2014

Rye Spiced Saison tasting

It's been almost 2 months since the brewday for my rye spiced saison and about time I got tasting notes up. Overall I am reasonably happy with this but my memory of my first batch is that it is better than this one. The big thing I think I'll change for next time is lowering the amount of rye and lightening the malt overall. But all things considered this is a good base to work off of for next time.

Spicy rye comes through strongly. There is a strong woody spice with some pepper, and the fruity yeast character is of golden raisins and grapes with some stone fruit (like plums and a bit of peach). The ester character/fruityness is good with more grape and less stone fruit in the aroma than in the taste. The aroma overall is more balanced toward spicy/phenol character. The rye probably helps that out a lot and the rye is very forward in the aroma. Although a lot of these descriptors (like the grape and plum) might sound dubbel-like, the beer has definite differences from what you might find in a dubbel. The fruityness is a bit lighter and the spicy characteristics are different. There is a mild bubblegum aroma, but I didn't notice it until I specifically looked for it.

Pretty chestnut color, hazy. Has a thick tan head with good retention and nice lacing. Overall it is quite pretty.

Again the rye is rather forward. There is a cool combination of spices and yeast-derived esters which I think it at least partially driven by the juniper. It is mildly resinous and has a surprising bitterness in the finish. The pepper is a bit forward and adds a bit of roughness to the bitterness in the finish. The fruity character is stone fruits and grape/raisin like the aroma, but with the balance more toward stone fruit. There is a rich malt character with a strong munich character and dark rye.

Medium/light body (I think the rye adds a a bit of viscosity which increases the perception of body). There are hints of warming alcohol. The carbonation is high and pleasant. There is a light tartness in the finish, probably from the levels of acidulated malt.

I think for my tastes I should drop the rye levels a bit and also drop the pepper a bit. It is a touch rough in the finish (I think lowering the pepper may help this, and maybe also lowering the acidulated malt) and it comes across as more that the 20 or so IBUs it calculates out to be. I think I'll move forward with this recipe in two different directions:

1) I'll go for a lighter beer (in color, body and strength) using a base of pils plus a good amount of munich and a bit less rye. I'll keep the spices the same and drop the pepper a bit.
2) Keep the base and maybe make it a touch darker and brew it as a winter oriented rather than fall oriented saison. I'll probably go with more carafa or something like that. And I'll change up at least some of the spices.