exBEERiment | Yeast Pitch Rate: Ale vs. Lager in a Kölsch

Author: Malcolm Frazer

We often read/hear that yeast and fermentation are paramount to producing a successful beer. While there are myriad components of fermentation, a primary one homebrewers seem to focus on is yeast starters, with numerous stories of massively improved beer following the use of this method. There are a few primary reasons commonly cited for the importance of propagating yeast in a starter, namely to “wake up” the yeast, thus improving overall vitality, and to increase the amount of cells to the pitch rates determined to be most appropriate for a particular style and specific gravity.

But what impact does pitch rate ultimately have on the resultant beer. Purportedly, specific gravity, style of beer, and even the yeast strain matter. Given the results of other pitch rate xBmts and anecdotal reports from homebrewers, I became curious about the merits of so called proper pitch rates, particularly whether others and myself would even be capable of distinguishing between the same beer fermented with different amounts of yeast. The time had come to put this one to the test!

| PURPOSE |

To evaluate the differences between a split batch of the same wort fermented with the same yeast strain, half at ale pitch rates and the other half at lager pitch rates.

| METHOD |

As a lover of Kölsch, a style in which fermentation is touted to be rather important, I figured it’d be a great beer to use for this xBmt! I designed an 11 gallon batch of typical gravity for the style that would be split post-boil into separate fermentors, pitched with different volumetric measures of slurry, then otherwise treated similarly. I chose to use slurry rather than fresh yeast for this xBmt for a few reasons, namely the relative precision of popular yeast calculators based on actual evidence, plus I was able to score some WLP029 slurry from a local brewery.

Kölsch Style Ale – “It’s Kölschy”

Fermentables

Hops

Yeast

Given my beer’s expected OG and batch size, Mr. Malty projected it would would require ~180 billion cells, I manipulated the figures slightly to get to an easily measured volume of 90 mL. A few days prior to brewing, I made a rather large starter that would result in plenty of yeast, making it easier to accurately measure out the slurry, as well as enable me to harvest some for future use.

I chilled the starter the night before brewing with plans to decant the spent beer and split the slurry into their respective amounts prior to pitching.

I performed a single-infusion mash in my nearly decade old converted Rubbermaid MLT, nailing my target temp of 148°F .

As a believer in the massive importance of water chemistry, I always adjust my brewing liquor to achieve a particular profile. On this day, Bru’n Water helped me hit my target mash pH on the nose.

After 60 minutes, I collected the sweet wort and transferred it to the boil kettle with my little helper.

Despite my uncertainty regarding boil length and its impact on DMS, I’ve always boiled wort consisting primarily of Pilsner malt for 90 minutes and wasn’t about to change things up for this batch, as this was a competition beer. I’m certainly curious about the outcomes of future boil length xBmts!

Following the boil, I proceeded to chill the wort to 65°F in about 15 minutes using my whirlpool chiller, which is about the best I can do given my summer groundwater temps.

I then split the wort evenly between separate 6 gallon glass carboys, moving the hose back and forth to ensure both received similar amounts of kettle trub. The carboys were loaded into my chest freezer to drop a few more degrees to my target pitching temp of 58°F, after which I aerated both for the same amount of time with pure oxygen.

While I tend to prefer my Kölsch very sessionable with an OG around 1.043, I hit my target for this batch of 1.047.

I decanted most of the starter beer then swirled with vigor to homogenize it as much as possible. Using a cute little graduated flask, I measured out 90 mL and 180 mL of slurry for the ale and lager pitch rate batches, respectively, then pitched.

I don’t typically see activity for 18-24 hours when using Kölsch yeast, something I assume is due to the lower pitch and fermentation temperatures. This was certainly the case for these beers.

Work life made it so that I was unable to observe the differences between the batches at first sign of activity, but when I came home a couple days later, both were happily chugging away. I suspected the lager pitch rate batch would have gotten off to a quicker start due to being hit with twice as much yeast, though even if it did, results from prior xBmts indicate differences in lag don’t appear to have a lasting impact.

After 3 days of fermentation, I raised the temp in the chamber to encourage complete attenuation and clean up any potential fermentation byproducts, namely diacetyl. Plus, I’m impatient.

A couple days later, 11 since brewing, I took an initial FG measurement and was happy to see both beers had reached my target of 1.009!

Both beers were kegged, fined with gelatin, then allowed to lager at 35°F for 3 weeks while carbonating.

If I may say so, the beer was rather appealing when it came time for sampling!

| RESULTS |

Once again, Pittsburgh area homebrewers, members of TRASH and TRUB homebrew clubs, brewery owners, and beer enthusiasts came through, your support is beyond amazing! Special thanks to Grist House Brewery and The Brew Gentlemen for hosting the xBmt tastings as well as participating. I actually got lucky with the timing on this one, as an AHA rally was being hosted by The Brew Gentlemen, providing me with a crowd full of willing participants!

In all, 30 people participated in this xBmt, which would require 16 correct responses in order to imply statistical significance (p<0.05). Before getting into all that, a quick side story:

For every xBmt, tasters are presented 3 samples in opaque cups differentiated by colors– red, green, and blue. While Marshall sets the samples on different colored squares of construction paper, this option wasn’t available to me during one session involving 5 tasters, so I decided to attach color-marked post-it notes to the cups. A simple and effective solution, or so I thought. It wasn’t until after collecting data from these 5 participants that I realized a problem, one that arguably compromised their contributions: the post-it notes were emitting a rather overwhelming scent, thereby skewing the aromatic perception of the beer in the cup. Following tasting, several of these participants claimed both beers had a noticeable plastic-like phenol character, while there were no such comments from previous sessions. Dammit! After some consultation, we decided it’d be best to share 2 sets of results, those including the data from the post-it note tasters, as well as a set excluding this compromised data.

When analyzing the performance of all 30 participants, 16 correct responses would be required to reach statistical significance (p<0.05), yet only 15 panelists correctly identified the different beer. So close, just one away from significance even when including the post-it tasters. This failure to achieve significance suggests a moderate OG Kölsch fermented with yeast at an ale pitch rate may not be reliably distinguishable from the same beer fermented with twice the amount of yeast.

In August 2015, we began utilizing a 1-tailed binomial proportions test, a more appropriate and slightly more liberal statistic than the the 2-tailed test. Following this change, we reran all of our prior xBmt data to test for changes in significant results. This xBmt was identified as being significant (p=0.026) utilizing the new statistic when previously we said it had failed to achieve significance, even with the inclusion of the stinky post-note data.

So, what kind of impact did the stinky post-it notes have?

Excluding the data from the 5 aromatically tainted participants leaves us with a sample size of 25 and a minimum of 13 accurate responses (p<0.05) to achieve significance. Only 1 of the 5 excluded tasters correctly identified the unique beer in the triangle test, which means in this “corrected” data set, 14 tasters made the accurate selection, implying tasters were capable of reliably distinguishing between the ale pitch rate and lager pitch rate beers.

Well… hmm…

While I’m personally inclined to believe the post-it notes had a negative impact and thereby compromised the data, I suppose it’s up to each individual reader to interpret these results how they will.

Comments from some of the tasters who were correct in the triangle test suggested only slight differences in aroma and flavor, with a few claiming the ale pitch rate beer smelled more fruity and less wine like, and it was preferred by 69% of the correct respondents. Regarding flavor, the lager pitch rate beer was described as being cleaner and less fruity, taking a slight edge in terms of taster preference. General preference was essentially split down the middle with comments suggesting some panelists simply preferred a cleaner lager-like beer while others enjoyed the slightly fruitier profile of the ale pitch rate beer.

My Impressions: I was able to identify the different beer in the 3 “blind” triangle tests I took, twice by aroma alone, though I submit my performance was due in part to my involvement in the process and knowledge of the variable being tested. Truthfully, I experienced the differences between the beers as slight. The character I got from the lager pitch rate batch was akin to what I perceive in many lower OG German lagers fermented with traditional strains and temp profiles– a combo of white grape and grape nuts cereal. While this character was not as intense in the ale pitch rate beer, it was definitely there, just more in the background. Rather, the Ale Pitch beer had a very low amount of generic pear esters and more of a bready malt profile, which actually gave me the impression it was fresher despite being brewed alongside the Lager Pitch beer.

| DISCUSSION |

At Brülosophy, we believe transparency is paramount, as we want to avoid even the possibility of misleading anyone. This hasn’t been much of an issue so far, though this xBmt created a unique situation in that a small amount of data was compromised due to an extraneous variable. Prior to analyzing the data, we discussed how to approach reporting these results, and while the possibility of choosing one data set and running with it came up, it was immediately nixed in favor of sharing both sets of data. The fact one was statistically significant and the other was not made this option seemingly more poignant, as well as a bit more difficult to discuss.

Personally, I’ve chosen to accept the corrected data as being the most accurate, not just because it confirms my own opinions, but more because I experienced the smell of the cups with post-it notes and it very obviously had an impact. Assuming I’m correct, we can deduce pitch rate does appear to matter, that cell counts intended for a typical lager fermentation seem to produce a beer that is reliably distinguishable from the same beer fermented with ale pitch rates. Of course, this xBmt focused on a very light style in which flaws have little to hide behind, it’s possible differences wouldn’t have been as noticeable in a more characterful beer, hence our plans to continue exploring the effects of pitch rate of various styles under different conditions.

If you choose to accept the full data set, you might conclude pitch rate doesn’t matter very much, which isn’t totally off-base since it corroborates the results from prior pitch rate xBmts. Now, while this may be the case, I’d caution anyone against accepting it with any amount of dogma, as the evidence to this point is far from conclusive. Proceed with caution, but don’t worry too much.

As for me, I’ll continue to target slightly higher than ale pitch rates when making Kölsch or similar style beers for personal consumption and competitions, as I want any edge I can get. But when brewing for parties, casual quaffing, or I’m pressed for time, I’d have little concern pitching slightly less than ideal yeast counts given how subtle the differences are. Cheers!

Follow Brülosophy on:

FACEBOOK   |   TWITTER   |   INSTAGRAM

If you enjoy this stuff and feel compelled to support Brulosophy.com, please check out the Support Us page for details on how you can very easily do so. Thanks!