Yeast Propagation, Dialing in Your Brewing Water Profiles

Q
If I am starting from commercial bottle dregs for yeast harvesting, Can I just pitch the dregs into a 3-L (3-qt.) starter at 1.035–1.040 that is well fed, oxygenated, and on a stir plate; or do I need to step through multiple stages to get to my target volume, like most of the literature out there tells me to? Why can’t I just target making a single, end-point volume starter, and expect ~150 billion cells per liter as an outcome?

Mike Conant
Aptos, California

A
I love these two part questions that begin with great fundamentals and then segue into the meaning of life. You want to propagate yeast from bottles and are attracted to skipping the intermediate steps mentioned in just about every brewing text that covers yeast propagation. For the benefit of readers who are not up on these methods, I will briefly cover the basics of how a brewer with a basic home microbiology lab can propagate yeast from a bottle of beer. We need an example for this question and I am going to take the easy route and select a bottle of Sierra Nevada Pale Ale.

The first decision that we need to make is do we propagate by simply collecting and growing the yeast solids from the bottom of the bottle, or do we do a little clean up work. I selected a bottle of Sierra Pale Ale because if we really wanted to take the easy route, starting with what is on the bottom of this bottle will most likely work. Why? Because Sierra Nevada Pale Ale is conditioned with the same strain used for fermentation and because their quality control measures are extraordinary. But with all the respect brewers have for Sierra Nevada, simply propping the “dregs” from the bottom of a bottle is pretty crude. The purpose of propagation is to grow cells and whatever ends up in the prop is probably going to grow. So what is a brewer to do? Make some media and streak some plates!

Yeah, this is sounding pretty advanced but it’s really pretty simple. Making your own media at home is just a few steps away from making Jell-O! Mix up an all-purpose media such as Wallerstein Lab Nutrient (WLN), pressure cook in media bottles for 20 minutes at 15 psig (1 barg) pressure, allow to cool to about 86 ˚F (30 ˚C), and pour the media into pre-sterilized Petri dishes. Once the plates have cooled and solidified they can be used to streak out the dregs from the bottom of the bottle. This is one of the things we do at BYO Boot Camp in a conference room. Fun times!

Why do we want to plate the sediment from the bottom? Because by streaking this stuff onto plates, we can select a single colony-forming unit (CFU in microbiology speak) and propagate that one colony. Each CFU is assumed to represent a single yeast cell and proper dilutions and streaking methods help to produce plates with clear separation among colonies. Suffice to say, streaking plates provides a much better shot at growing what we want versus bringing in a bunch of riff raff into the propagation party.
OK, now what? The textbook methods you reference usually start with transferring a single colony from a plate into about 25 mL of sterilized wort. In about 48 hours, the contents of the propagation flask (usually an Erlenmeyer flask that is about twice the volume of the contents) is increased to 10X. This means that 25 mL of propagating yeast slurry is added to 225 mL of sterile wort in a 500 mL Erlenmeyer flask to yield a 250 mL volume. This method continues until there is enough yeast to pitch into wort. Yes, the propagation volume represents a significant (10%) volume and the color and flavor of the propagation wort is important to your finished beer. But that is really a separate question for another deep dive into what-about-this-wort-ism.

Now that the basic method has been established, let’s go back and explore some possible shortcuts because this process is somewhat of a pain in the neck (this is when I get to thank the yeast sponsors who generously donate to Team Mr. Wizard . . . thank you yeast sponsors, I am still waiting for those cool rugby shirts I suggested on Facebook). Idea #1 is to skip the intermediate steps and add the starter cells, either a single CFU from a plate or a shot of dregs from a beer bottle, to a volume that is ~1⁄10 the fermenter full volume (remember this is not 10% of the wort volume, but 10% of the wort + yeast volume). The real challenge to this method is transferring the starter cells into the sterile wort without bringing other organisms to the party. The idea with pure culture technique, which is what this is about, is growing cells from a single cell line, and the reality of the situation is that it is nearly impossible to transfer yeast into a sterile environment, i.e., the wort in your Erlenmeyer flask, without introducing other organisms.
So why do the textbook methods suggest incrementally increasing the volume? It’s all about competition, hurdles, and speed. While this may sound like track and field, it’s really about putting yeast into an environment where they will outcompete other critters by quickly reducing pH, gobbling up nutrients, and producing alcohol, which are all hurdles that impede the growth of other organisms that snuck past the bouncer at the door (you). As long as the selected yeast population is the super majority, non-selected bacteria and yeast strains will most likely fare poorly in this environment. “Most likely fare poorly” is not an absolute and yeast labs that do this thing day in and day out have stringent quality methods to ensure that the bouncers were effective.

Circling back on your question, if you put a single CFU or shot of dregs into your 3-L (3-qt.) starter, a lot of time passes before significant pH reduction, nutrient depletion, and alcohol formation occurs. This time represents the period between bar closing and sunrise when all sorts of chaos occurs. Yeast labs do not want this sort of craziness happening . . . ever . . . and have found that incremental increases in propagation volume work pretty darn well.

A corollary to this “hurdles theory” that is so prevalent in the practical application of food microbiology, is selection. Our bottle of Sierra Nevada Pale Ale is probably clean and growing the yeast sediment from one of these bottles will probably turn out just fine. But what if you were going to grow the sediment from a great beer brewed in a garage brewery? The beer is great, we just established that, but great beer can contain bacteria that don’t cause product damage. Let’s call these bottles asymptomatic spreaders. You bring the dregs from an asymptomatic spreader into your propagation flask, encourage cell growth, and what you have is a messed-
up starter culture. This can happen with or without incremental changes in volume, and is much more likely to occur when using dregs as the cell source versus a single CFU.

Back to your original question. Can I just pitch the dregs into a 3-L starter at 1.035–1.040 that is well fed, oxygenated, and on a stir plate? This is a really great question and I sincerely hope that my brief explanation has explained why the answer is “yes, but not recommended.”

Q
A lot of recipes and water articles give mineral levels in ppms (parts per million). How do you test these figures? Or do most brewers start with reverse osmosis (RO) water and add set amounts of water salts?

Doug Milroy
Sydney, Australia

A
Thanks for the great question about water. I will jump into the middle of the pool here and try not to stray towards the deep end where the abyss of things not relevant to homebrewing lies. The crux of your question is how does one determine the starting point with water, and like most things in life there are multiple places to find information about your water. One thing for certain is that water chemistry is often inconsistent within a locale because of seasonal fluctuations and variations among sources, especially if you live in a place that draws water from surface reservoirs and underground aquifers.

For the sake of simplicity, let’s assume you have water piped into your house from a fairly consistent source, such as one of the little ponds called the Great Lakes separating the U.S. from Canada. These lakes generally fall into the humongous category of lakes, are deep, and have really consistent water. If you are lucky enough to have access to this sort of water, you can obtain water analyses from your municipal water authority and have pretty good information because the test data likely don’t vary much throughout the year.

To more fully address your question requires paddling towards the abyss. Water test methods use a variety of methods, some rather old and some rather new, to determine the composition of water. In the old days, titration methods were the norm for water testing and general buckets were defined by how water components reacted in titrations. For example, water containing calcium and magnesium ions can be titrated with EDTA (ethylenediaminetetraacetic acid) in the presence of a color indicator to determine total hardness. The total hardness bucket is then labeled in terms of mg/L of calcium carbonate. Similar methods can be used to determine alkalinity (bicarbonate and carbonate), which is also expressed in terms of calcium carbonate. Water test kits are readily available to determine hardness and alkalinity, and many of the classic brewing rules about water were based on these two values. I really don’t think you are asking about how to do water chemistry in the kitchen, but I will press on a bit farther.

More specific methods can be used to determine the concentrations of specific ions in water. For example, total hardness is a measure of calcium, magnesium, and carbonate, but brewers often want to know the individual concentrations of calcium, magnesium, and carbonate/bicarbonate. Ion selective electrodes, specific titrations methods, atomic absorption spectroscopy (AAS), and inductively coupled plasma spectroscopy (ICP) are some the most common methods to provide more specific results about water. Although AAS and ICP require specialized and relatively expensive instrumentation, the methods are widely utilized because of their reliability, wide range of measurements recorded in a single run, and speed. Water labs running lots of tests are likely using some sort of spectroscopy for analysis. OK, enough of the esoteric stuff. Most brewers, including the majority of commercial breweries (by head count, not production volume) farm this stuff out because water analysis is specialized stuff.

Getting back to the practical, let’s start with adjusting the water flowing from your faucet. I think it is much, much easier to do water calculations by knowing the concentration of individual ions. The three ions with the most significance to brewers are calcium, magnesium, and bicarbonate/carbonate (I lump the last two together because they are essentially the same thing, and their identity is pH dependent). I will stick to the question here and totally avoid jumping into the rabbit hole about how to adjust water. You have asked about water measurements. Assuming you don’t have super-consistent ground water, you can buy water-testing equipment that uses specific reagent kits and a device that looks like a pH meter to run your own tests, or you can send your water out to a lab. I have some pretty strong opinions about where the boundaries of a hobby lie, and personally believe that performing water tests at home and/or sending water samples to a lab are pretty extreme measures for a hobby. But if that floats your boat, go for it and be prepared to use the information that comes back. Check out the topic of residual alkalinity to help frame these results.

Another method is to start with reverse osmosis (RO) water, a blank canvas for brewers more interested in beer than burets, indicators, and millivals (milli-equivalents), and add what you want to build your own water. My bias towards this method may be showing just a wee bit here. And I am certainly not alone with my love of RO water. Check out Gordon Strong’s recipes and guidance about water. He is a preacher of the KISS (keep it simple, smarty) method of water chemistry and exclusively suggests using RO water for homebrewing.

Just to demonstrate how easy water calculations can be, I am going to do a demonstration here without any reference books and a simple calculator. The demonstration is to calculate the weight of sodium chloride required to add 25 ppm of sodium to 1 liter of RO water. Simple explanations are easiest with bullets. So grab a liter of RO water and your salt shaker!

• What is meant by 25 ppm of sodium? A part per million is a measurement that is equal to 1 mg/L. Why is it called a part per million? Well, a liter contains a million milligrams (1 liter = 1,000 grams = 1,000,000 mg), and the concentration shorthand of 1 ppm is the same as 1 mg/L.
• That was half the battle because we have our water recipe . . . we need to add 25 mg of sodium to a liter of water. Just fetch the bottle of sodium. OK, that’s not real funny since sodium metal bursts into flames when dropped into water. But, you do have that salt shaker. When sodium chloride is added to water, it completely dissociates into sodium and chloride. We just need to know how much sodium, by weight, is contained in a measure of salt. That’s actually easy to calculate by knowing the molecular weight of sodium and chloride, or by simply Googling “the mass percent of sodium in sodium chloride.” I am old school and don’t use that sort of shortcut, but do whatever works best.
• Here is where I get to show off my awesome memory of useful facts from chemistry. No joking here, if chemistry was taught using beer as a model we would have lots of smart kids who would know how to calculate the amount of salt to add to brewing water. Sodium weighs ~23 g/mole and chloride weighs ~35.5 g/mole (g/mole is also known as atomic mass units and it’s one of the properties listed on the periodic table).
• Sodium chloride (table salt) has an atomic weight of 23 + 35.5 g/mole or 58.5 g/mole and sodium represents (23/58.5) 39% of the weight of salt.
• How much salt do we need? Easy peasy . . . 25 mg of sodium divided by 0.39 (mg sodium/mg salt) = 64 mg salt per liter of water is what the brewer ordered!
• To recap using the Google method . . . 25 mg sodium divided by 39% (the mass percent of sodium in table salt) is equal to 64 mg of salt. I have been told that repetition helps memory.

You can use this same method with all of the other salts in water as long as you know the molecular weights of calcium (~40), sulfur (~32), magnesium (~24), carbon (~12), oxygen (~16), hydrogen (~1) and the forms that various salts come in. This is starting to sound like a spreadsheet and luckily for the modern brewer there are many very great water tools available out and about.

Every so often I get a question like yours Doug, where I get to jump up on my salt shaker and proselytize about using RO water. To sum this up, you can invest in water testing equipment and/or farm-out water tests (I forgot to mention that having your own equipment means getting a second set of measures to validate your home method), more or less frequently depending on the consistency of your incoming water, and blame the lab when your beer doesn’t come out perfectly because of some oddity that must be related to your water. Or you can start out with RO water and build your brewing water from the ground up. Yes this is heavy on the sarcasm, but also heavy on the sincerity of the advice. RO water makes your #1 brewing ingredient consistent.

Written by Ashton Lewis