In his ubiquitous manuscript, How to Brew, John Palmer states:
“It is a good idea to remove the hot break (or the break in general) from the wort before fermenting. The hot break consists of various proteins and fatty acids which can cause off-flavors, although a moderate amount of hot break can go unnoticed in most beers.” (Ch. 9, Sec. 1)
Despite stating that “moderate” amounts of break material can go unnoticed, the primary point of this statement appears to be that hot break in the fermentor is generally not a good thing. Many homebrewers go to great lengths to separate this gooey gunk from their wort with concerns of imparting undesired off-flavors, developing a stuck fermentation, or ending up with a hazy beer.
Then there are those those few rogue (lazy) brewers who couldn’t give two shits about kettle trub making it into their fermentors, claiming their beers attenuate just as well, taste just as good, and end up just as bright as beers fermented without trub. These folks annoy some of us, what with their flippant remarks about how we’re working too hard for something that has minimal, if any, impact on the finished beer and only results in the unnecessary expenditure of time, effort, and money…
Okay, I’m sort of one of these guys. It’s not that I’m against geeking out on equipment, I just appreciate simplifying the process while maintaining a high quality finished product. For more cynical folks like me, claims that kettle trub making it to the fermentor leads to lower quality beer are suspect, particularly when those claims aren’t necessarily backed up by any real-world evidence (that I’ve been able to find). As such, we’re left wondering: does it even matter?
In one study, researchers measured the impact of kettle trub on levels of isoamyl acetate (banana) and ethyl acetate (nail polish remover), splitting the same wort into 3 fermentors with the following trub volumes:
Low trub: .13 mg/ml
Moderate trub: 1.7 mg/ml
High trub: 15.9 mg/ml
To put this in a more understandable context, let’s consider what this would look like for a typical 5 gallon batch of beer:
Low trub: 0.1 oz (essentially nothing)
Moderate trub: 1.13 oz (probably close to what homebrewers who care can achieve)
High trub: 10.56 oz (quite a bit)
They found the wort with the most trub produced a beer with significantly lower levels of the aforementioned ester compounds. Huh?! We’re talking 106 times more trub than the beer in the low trub condition, equivalent to nearly 3/4 lb of kettle sludge in a 5 gallon batch. With all of my skepticism, not even I expected this. Fascinating.
Still, this is only a tiny slice of a huge pie and many questions remain: does kettle trub have an impact on clarity, flavor, or head retention? How do differing levels compare in regards to fermentation lag, vigor, and completion time? Do I really need to worry about this shit? These are the questions I aimed to at least attempt to provide some answers to in this exBEERiment.
To compare the impact different amounts of kettle trub have on myriad aspects of 2 beers made from the same wort and fermented with the same yeast.
A 10 gallon batch of 1.046 OG Cream Ale was brewed on May 4, 2014:
13.5 lbs US 2-Row Malt
4.0 lbs Flaked Maize
8.0 oz Acidulated Malt
8.0 oz Honey Malt
6 IBU Summer (Australia) @ first wort hop (14 g)
99 g Summer @ flameout (15 minute steep)
1 Whirlfloc Tab @ 15 minutes
I mashed at 150°F for an hour, collected the wort, and started the boil as usual.
My decision to use 100% Australian Summer hops had mostly to do with the fact I had purchased 4 ounces a few months prior from Farmhouse Brewing Supply and thought “light melon and apricot” would compliment this style well, as non-traditional as it may be. While designing the beer, I figured it would also be cool to forgo all boil additions, settling on doing only FWH and flameout additions, the latter of which amounted to nearly 3.5 ounces.
Once the boil was complete and the wort was chilled to 72°F (my groundwater is warm this time of year), the first carboy was filled halfway with excessive amounts of kettle trub intentionally added; a long spoon was used to coax trub from the bottom of the kettle out of the ball valve. I then covered the kettle and let it sit for about 15 minutes to allow the break material and hop matter to settle out, propping the front end up with a towel so that it settled on the opposite side of the valve.
I gently opened the valve to clear any leftover trub then transferred 5.25 gallons of very clear wort to the second carboy, Non-Truby. I finally finished filling Truby, making sure to transfer as much kettle trub as possible. I’ve never left such clear wort in my kettle before.
Both carboys were placed in the fermentation freezer and allowed to chill to my target pitching temp of 62°F. The difference was pretty remarkable after only an hour in the cool chamber.
It took about 4 hours for the wort in each carboy to reach pitching temp, at which point the trub had compacted into a more solid mass. As hard as I tried to keep the trub out of Non-Truby, there remained a small amount of material that settled out, still significantly less than what was sitting at the bottom of Truby.
I rehydrated 1 sachet per carboy of Danstar Nottingham yeast in about 250 mL of 90°F water, allowing it to proof for 20 minutes. It was 68°F by pitching time.
I’ve never actually used this yeast, I tend to stick to liquid varieties, but I’d heard it works really well at cooler temperatures and I’m always game for trying something new. Moreover, since each sachet had the same date on it, I figured this would potentially increase the equality between each batch in terms of pitch rate compared to trying to split a starter from a single vial of yeast.
Not much was happening 12 hours post-pitch, I understand dry yeast tends to lag a bit longer than liquid, even when rehydrated.
Just a few hours later, 18 since pitch, both beers were waking up.
One thing I noticed was Non-Truby had slightly less krausen development (look toward the right side) while Truby was completely covered.
The difference was very slight and both beers were ripping along nicely by 36 hours post-pitch.
A few hours went by where both airlocks were blurping at the same exact rate, as if they had been purposefully synchronized. I felt like I’d seen a shooting star. Anyway…
After 3 days fermenting at 62°F, both beers were showing signs of reduced activity, so I bumped the ambient temperature of the chamber up to 70°F to ensure complete attenuation. The beers were sitting at 67°F the next evening and the krausen had dropped on both.
The difference in material at the bottom of each beer was pretty drastic.
If you look closely in the photo above, you’ll notice 2 fairly distinct layers in the carboy to the right. I’ve always assumed the yeast covering the greenish bottom layer of trub would limit any flavor or aroma impact it might have on the finished beer, I’m curious how true this is. Another very subtle difference was noticed on closer inspection of the top of each beer, with Truby appearing to have slightly more particulate matter floating around.
By the following morning, the beers were closer to 70°F and looked similar to the night before. The airlocks on each were totally inactive.
I tested the FG of both beers 5 and 8 days after pitching.
On both occasions, the beers read the same (precise photos of hydrometers are tough to take). I tasted each and detected no diacetyl, so I dropped the ambient temp in the chamber to 30°F to cold crash.
I was interested to find exactly zero difference in clarity between Truby and Non-Truby. In fact, they looked, smelled, and tasted the same to me. The Australian Summer hops were coming through with definite hints of apricot and sweet melon.
By the evening of the 11th day since pitch, both beers were sitting right at 32°F and dropping clear.
The only real difference seemed to be in the amount of stuff sitting atop each beer.
Kegging commenced on the 13th day after the beers were made. Both beers were removed from my fermentation chamber and left to settle for a few minutes.
Some slight differences between the beers became more evident in this brighter environment.
The trub at the bottom of Truby was obviously thicker than the creamy cake at the bottom of Non-Truby.
The stratification again suggests the hop matter and hot break from the kettle settled before the yeast.
As had been the case for days prior, the top of Truby had noticeably more particulate floating on the top, while Non-Truby had almost none. I was developing some expectations as to how the finished beers might look at this point. The beers were kegged using one of my favorite homebrewing tools, the sterile siphon starter.
The yeast cakes in each carboy looked very different once the beer was removed.
The leftover trub in Truby resembled something like a neuronal network of sorts, while Non-Truby was much smoother with interesting bubble-like formations. The swipes on each cake occurred post-racking during the removal of the siphon; I was sure not to transfer any muck from either batch to the kegs.
I stole a sample of each beer before kegging to see if there were any differences when experienced from a smaller glass.
Both seemed to be dropping bright at a relatively similar rate, though I did perceive some subtle differences in clarity.
Both beers were put on 30 psi of CO2 for 36 hours, purged, then reduced to 12 psi. This is where they sat for a week while I spent a week camping at the river with family and friends, a time in which I shamelessly admit to swilling massive amounts of Heineken and Bud Light.
My wife sure has a knack for taking pretty pictures of our adorable kiddos.
I returned to two perfectly carbonated Cream Ales that had been allowed 9 days in the 38°F keezer to drop clear. A good friend and fellow camper, Ryan, was the first to taste and evaluate the beers, another friend dropped by the following day to do the same, then I finally took a couple bombers with me to the House of Pendragon taproom to share with a panel of tasters.
I feel compelled to say something here about the survey process and all that jazz. These exBEERiments were never intended to be super scientific in nature, I simply don’t have the resources, knowledge, or desire to design and perform actual experiments. While I made every effort to reduce potential bias by tasters, I absolutely understand this may have been done better. I promise to continually work toward improving the way I survey my exBEERiment tasting panels, though I still want this to be fun too. I used a simple survey for this first panel tasting and realized immediately areas that need to be reworked. In the future, I’m thinking of using an online survey platform where panel members answer questions on their mobile devices. I’m also starting to think the triangle test may not be totally necessary or even very helpful. Please, if you have any suggestions, do kindly share them with me! Without further ado…
Each taster was initially poured 3 small samples for a triangle test, the makeup of which was random for each taster. Each tasting session occurred privately and the surveys were completed by me. Once the triangle test was completed, one of the beers was removed, leaving 1 each of Truby and Non-Truby. Tasters were not informed if their response to the triangle test was correct. They were then instructed to complete the rest of the survey with the two beers left, understanding that they were in fact different from each other. Each taster was asked to indicate which of the 2 beers they thought was better in terms of appearance, aroma, flavor, and mouthfeel, and they were encouraged to provide subjective feedback as well. They were also asked to attempt to identify which beer was “experimental” (the one that had something out of the norm done to it, in this case Truby) as well as which one they preferred overall. Tasters were not made aware of what beer was in either glass until their survey was complete. In all, 6 people were included in the tasting panel.
Ryan, the first taster, was quick to accurately identify the different beer during the triangle test, which was Non-Truby, explaining it was more hazy than the other beers. He went on to report that Truby had a “crisper, sharper bite” to it, while Non-Truby was slightly smoother. When asked to indicate which of the beers was experimental, Ryan correctly chose Truby, stating his hunch the extra material in the carboy likely helped to clear the beer. Ryan’s overall impression was that both beers were “incredibly similar with the exception of clarity and that sharpness in the flavor,” stating that his slight preference was for Truby.
The next person to evaluate the beers was John, an avid homebrewer and beer lover. Like Ryan, he rather easily identified the different beer in the triangle test, which in this case was Truby, as it was “slightly though noticeably more clear” than Non-Truby. He went on to say, ” [Truby] has a little more tangy taste to it, it’s just a little more tangy up front and a little different on the back end, just mildly more bitter, but not much.” In terms of aroma and mouthfeel, he said they were almost identical. He was accurate in identifying Truby as the experimental batch, which he reported he had a slight preference for compared to Non-Truby, stating, “It’s really kind of nice.”
Aaron is a Cicerone Certified Beer Server at the House of Pendragon taproom who has also passed the written portion of the BJCP exam and currently serves on the Board of Directors for our local homebrew club, the San Joaquin Worthogs. After tasting the 3 beers presented to him for the triangle test, he was unable to accurately identify the different beer, choosing one of the 2 Non-Truby glasses instead. Perhaps it’s important to point out that the beers from here on out were being sampled from 4 oz Belgian-style taster glasses, which may have impacted perceptions of clarity. Once comparing just the 2 beers, Aaron did correctly identify the experimental batch. He reported minimal to no perceptible differences in aroma or mouthfeel, though after closer inspection stated that Truby was slightly more bright than Non-Truby. Flavor-wise, Aaron experienced Truby as being “a little more phenolic with a more biting apple-like flavor, maybe some acetaldehyde.” He was the first to report a preference for Non-Truby, explaining he appreciated its slightly less biting bitterness.
David is another server at the House of Pendragon taproom who is studying to become a Cicerone Certified Beer Server. While he does not currently homebrew, he has a great appreciation for good beer. He was unable to accurately identify the different beer during the triangle test and he also incorrectly guessed which of the beers was experimental. He commented that Truby was more clear, but that Non-Truby “just has more aroma to it and it’s a little more bitter at the end, I like that.” When asked to choose which one he’d rather drink a full pint of, he chose Non-Truby.
Chris was next to complete the evaluation, he is a regular homebrewer who is known by his friends for having a rather discerning palate. Chris was able to accurately identify the different beer in the triangle test as well as the experimental beer. He reported the clarity on Truby as being better, though noted that Non-Truby appeared to have slightly better head retention. To Chris, Truby had “a more pronounced aroma and tastes crisper” than Non-Truby, which he thought had “fuller mouthfeel and a better blend of flavors.” Apparently he liked the crisper flavor of Truby over Non-Truby’s better blend of flavors, as it was the beer he said he preferred.
Jeremy was the last participant on the tasting panel, he has enjoyed drinking delicious craft beer for many years. While he was inaccurate in his attempt to identify the different beer in the triangle test, he was able to detect which of the beers was experimental. He described Non-Truby as being “a little hazier, which I like, with more subdued aroma, smoother mouthfeel, and a more firm flavor, almost like sourdough.” He experienced Truby as having a “more lively” aroma with a “crisper, lighter” flavor. Overall, he preferred Non-Truby.
A quick recap of responses from the “official” tasting panel:
- 50% (3/6) accurately identified the different glass in the triangle test
- 83% (5/6) accurately identified Truby as the experimental batch
- 50% (3/6) preferred Truby over Non-Truby
- 100% (6/6) perceived Truby as being clearer than Non-Truby
- 50% (3/6) perceived no difference in aroma while 33% (2/6) thought Truby had better aroma
- 67% (4/6) reported Non-Truby as having better flavor in general
- 33% (2/6) thought Non-Truby had better mouthfeel while the others perceived no difference
While not included on the actual tasting panel, numerous others have also compared these 2 beers. Truby has consistently been voted the clearest beer with the sharpest/crispest flavor. Interestingly, there seems to be somewhat of a split when it comes to overall preference.
My Impressions: In the multiple times I’ve sampled both of these beers, I’ve regularly come to the same conclusions, namely that Truby is clearer with more noticeable hop aroma and a perceptibly enjoyable crispness. Like some of the others who have compared these beers, I agree that Non-Truby has a smoother flavor, slightly less biting, but all things considered, I actually find myself pulling the tap handle for Truby more often than Non-Truby.
Kettle trub making it into the fermentor certainly seems to have an impact on the finished beer, the most striking and perhaps surprising being on clarity. The assumption that clearer wort in the fermentor leads to clearer beer in the end appears to be false, with all samplers agreeing that Truby was brighter than Non-Truby.
Given the subjective nature of taste preference, it behooves me to resist making any recommendation for what other brewers should do as a matter of principle. For those who tend to prefer clearer and crisper beers with potentially sharper bitterness, consider not worrying too much about the amount of kettle trub you transfer to the carboy. Alternately, those who enjoy slightly smoother bitterness and don’t mind a bit more haze in their beer may want to continue investing a little more effort in transferring only the clearest wort to their fermentor.
As with most things in this great hobby, what one brewer considers normal or even required practice may be viewed by another brewer as being totally unnecessary. Obviously, the experimental condition in this exBEERiment was purposefully overblown, as I was most interested in the impact massive amounts of trub would have compared to minimal amounts. I think this exBEERiment not only demonstrates that worrying about this issue is hugely unwarranted and that great beer can be produced regardless of the amount of trub that makes it into the fermentor, but that there is potentially some benefit to fermenting with a fair amount of break material.
If you have any questions or comments, please do not hesitate to respond to this post! I am absolutely open to any suggestions and would love to start a conversation about this topic. For now, it’s on to the next exBEERiment.