This week, we presented a scientific poster at the MBAA National Conference in Calgary on the impact of pitch rate on kveik ferments. This is a question we are constantly asked, and one which we see talked about in brewing communities a lot.
Ever heard these?
"You have to underpitch kveik" "Kveik ferments faster when underpitched" "Yeast labs supply too much kveik"
While it's known that traditional brewers do use a low pitch rate for kveik fermentation, we wondered what the real impact of this was on the fermentation kinetics and flavour profile. So we used science to find some answers.
This is really important these days, because kveik is on a hype cycle.
As more brewers use kveik, more brewers are disillusioned with this yeast that has been touted as a miracle cure-all for fermentation. This kind of disillusionment is fed by the lack of knowledge surrounding kveik.
In other words, there are many places to obtain kveik cultures, but not many places to learn how to use them.
Our aim for the next year is to convince the disillusioned to give kveik another shot, by providing actionable data to get high quality kveik ferments. Here's the first of many useful, digestible, chunks of new kveik data.
Kveik pitch rate study
This was conducted as part of a larger study, where we are trying to create a "flavour map" of all of our production yeast strains. We performed miniature ferments (50 mL centrifuge tubes fitted with airlocks) and monitored specific gravity, FAN consumption, pH change, alcohol/glycerol production, and aroma compound production (using GC-MS). All ferments were performed in triplicate.
We decided to test three pitch rates with four of our kveik products, as well as a control (Vermont Ale). The pitch rates were 1 M / mL (1 million cells / mL, 10% of a typical pitch rate), 7 M / mL (7 million cells / mL, 70% of a typical pitch rate), and 10 M / mL (10 million cells / mL, a typical pitch rate). The wort OG was 1.053.
It is important to note that this experiment was conducted at 20 ºC since it was a small part of a bigger project. The exceptional kinetics of kveik are not as obvious at this temperature. We would have loved to run the ferments at the typical >30°C Kveik temperatures, but we were limited by incubator and physical space.
Here are the fermentation curves:
As you can see from the graph, all strains fermented slower at the 1 M / mL pitch rate. So that is one myth debunked: kveik does not ferment faster at a lower pitch rate. Between 7 and 10 M / mL there was not much difference. Some of the kveiks did perform very well at the low pitch rate, especially Hornindal and Voss. Surprisingly, our control strain Vermont Ale performed admirably well with the low pitch rate, too.
We've seen a lot of chatter in brewing communities about nutrient requirements of kveik, and anecdotes suggesting overpitching can lead to an aggressively low pH. So we tested those too:
There were not clear trends in FAN (nutrient) consumption according to pitch rate, but it is worth noting that the FAN consumtion of ~180ppm is considered high for beer yeasts, and highlights the importance of getting these yeast the nutrients they need for fermentation. Low gravity worts with European/Pilsner malts may result in sluggish fermentations and require nutrient addition.
There were also no clear trends for pH. It's clear from the data that some of the strains drop pH more than the control (~4.15 for Arset vs. ~4.4 for Vermont), but pitching more yeast did not lead to excessive pH drop. Interestingly, the 10 M / mL pitch rate of Vermont resulted in a higher terminal pH.
Terminal pH can impact perception of body in beer, so this will be something to pay attention to with kveik moving forward. More to come on that.
What about flavours? Beyond fermentation working efficiently, this is what really matters to you.
There's a lot to unpack here. If you want to know what these compounds are, you can Google the name and get a sense of what they taste like. For our purposes, the ones ending in "ate" are ester, the organic acids are pretty clear, and the ones ending in "ol" are alcohols. beta-Citronellol is a terpene produced by yeast from hop precursors.
A few things jump out to us. First, there is no easily discernible trend in overall ester production according to pitch rate. Second, kveik produce more of the organic acids than the control, which is something we saw in our 2018 study. This may help explain the "creamy" character sometimes described in kveik.
Interestingly, the production of beta-citronellol by Arset and Ebbegarden was similar to that of Vermont Ale, a strain known for biotransformation capability. This data suggests Arset and Ebbegarden may be able to perform biotransformation of terpenes in a similar manner to commonly used NEIPA strains.
Okay, so there are no easy to understand trends in esters from looking at the raw data. What if we look at the flavour profile as a whole?
Using a statistical technique called PCA (principal component analysis), we can map out the overall flavour profile using the aroma compound data. On the right, we see what factors contribute to the spread in the left flavour map. We see that higher ester concentrations are strong factors in pulling a sample to the right (Dimension 1). We also see that some other compounds (ethyl decanoate, beta citronellol, isobutanol) drive Dimension 2 and provide further differentation.
This is clear in the Vermont Ale - the low pitch rate has higher isobutanol (fusel) and beta citronellol, and so appears near the top of the PCA.
Based on this analysis, it looks like there is a trend for increased aroma intensity for some of the kveiks (Voss and Hornindal) as the pitch rate decreases. However, this is not true for Arset and Ebbegarden, where the trends are less clear. So this tells us that there is not a one-size-fits-all rule for kveik pitch rate and aroma production.
Factoring in fermentation rate, fermentation specs and aroma production, we think that 7 million cells / mL strikes a good balance between the low and high pitch rate, but that the sweet spot could be somewhere between the two. We have done collaboration brews with partner breweries at around the 3 million cell / mL rate with good results and minimal impact on fermentation kinetics.
A caveat for this study is that the yeasts were produced under ideal conditions and never left the lab. The results may be different for slurry that is shipped and stored for several weeks. But if brewers want to attempt this kind of study themselves, we're all ears!
We hope this was helpful. The coles notes? Pitch rate matters less than you might think. There are clear trends for some strains (Voss and Hornindal) but not others (Arset and Ebbegarden). The more we dive in and learn about kveik, the easier it will be for brewers to have a good experience with these yeasts.