Buttered Popcorn or English Yeast?
Your taproom is three days from releasing a 5.0% pale ale, fermented with an English ale yeast and dry-hopped with Citra. The marketing team's menu copy reads "stone fruit, toffee, and fresh citrus."
At the pre-release panel, the bright-tank sample is warm (about 60°F) and shows a distinct buttery, popcorn-like aroma with a slick, coating mouthfeel. A second sample, pulled the next morning and tasted colder, smells sharply of cut green apple. The brewer shrugs: "That's just the English yeast - it's supposed to be fruity." Gravity has been stable at 1.012 for four days and the beer is already carbonated. Forty kegs are committed to accounts.
1. Sort these aromas into intended yeast character versus fermentation faults, and describe the test that separates them.
Read the worked answer
Yeast can make fruit, but the descriptors here split cleanly:
| Aroma | What it is | Positive or fault |
|---|---|---|
| Buttery, popcorn, slick mouthfeel | Diacetyl (vicinal diketones) | Fault - a fermentation timing problem, not a yeast signature |
| Cut green apple, cidery, grassy | Acetaldehyde | Fault at this level - an incomplete-fermentation/young-beer marker |
| Stone fruit, banana, pear drop | Esters (e.g. isoamyl acetate) | Intended - stable, and part of the English character |
| Toffee | Caramel malt, not yeast at all | Menu copy overreach unless the grain bill supports it |
The separating test is behavior, not vocabulary:
- Warm it. Diacetyl reads louder as the sample warms (your 60°F sample is the loud one for a reason) and can increase over time as its precursor converts. Esters stay put.
- Hold it warm. A forced test - warm a sample to roughly 140-160°F for 15-20 minutes, cool, and re-smell - drives precursor to diacetyl; if the buttery note jumps, you have a real diacetyl problem rather than a sensory trick.
- Give it time. Acetaldehyde is reabsorbed by live yeast during conditioning; if it drops over a few days, that confirms young beer rather than a spoilage organism.
- The slickness matters more than the smell. A coating, buttery mouthfeel is a stronger diacetyl signal than aroma alone.
2. What do you do about the release date, the 40 committed kegs, and the menu copy?
Read the worked answer
Hold the release. Three days is not enough to fix a beer that is both buttery and green-apple young, and the beer is already carbonated - which means the window for a warm rest is closing.
Brewing side:
- Warm the tank back up a few degrees and rouse gently to keep yeast in contact, so it can reduce diacetyl precursor and reabsorb acetaldehyde. A warm rest is the standard fix; carbonating and crashing early is what created this.
- Retest daily with the forced warm test. Release when the buttery note is gone, not when the calendar says so.
Accounts side:
- Tell all 40 accounts now, in plain language: date slipping, reason, new date. A late honest call is recoverable; 40 kegs of buttered popcorn in the market is not.
- Don't ship "drink it fast" as a workaround - diacetyl can grow in the keg if precursor was packaged with it.
Menu copy:
- Delete "toffee" unless the grain bill actually delivers it, and never write a fault into the copy. If diacetyl is even borderline at release, do not describe the beer as "buttery" or "creamy" to make it sound intentional - it removes your ability to fix the next batch.
3. If the beer ships as-is and goes on draft across 40 accounts, what happens to these aromas over four to six weeks, and how does that change your decision?
Read the worked answer
They don't move in the same direction, and that's the trap.
- Diacetyl can increase if precursor was packaged with the beer - the conversion continues in the keg, and warm storage at account level accelerates it. A beer that was borderline at the brewery becomes a butter bomb on tap in week three, far from your control.
- Acetaldehyde usually declines as it's reduced or bound, so the green-apple note may soften. Shipping now trades a fault you can still measure for one you can't.
- Hop aroma (the Citra) fades fastest - that's the one thing the menu copy actually promises, and it will be the first thing gone.
Decision logic: the fault that will grow in the market is diacetyl, and it grows in exactly the accounts you can't visit. The fault that shrinks is the one you'd be gambling on. That asymmetry is why a borderline beer should be held at the brewery rather than shipped "drink it fast": the downside isn't a mediocre pint, it's 40 accounts deciding your pale ale is buttery.
Also flag the process cause so it doesn't repeat: carbonating before the yeast finished its cleanup is the root of both notes.