found drama

get oblique

Aroxa off-flavor training notes

by Rob Friesel

During our 2017-2018 BJCP study group, we talked a few times about doing an off-flavor training. We’d pool our money, pick up a spiking kit, and spend one session working through a few of the common off-flavors. For better or worse, we didn’t manage to get around to that.

Meanwhile, a study group formed for the Certified Cicerone exam. While I wasn’t able to commit to that group in addition to the BJCP group, there was an open-ended invitation to attend some of those sessions. Luckily, I caught wind of one of their off-flavor training sessions and got myself on the list.

Who couldn’t use some bad beer education?


We would use Amstel Light as our control beer, and spike with an Aroxa kit available through the Cicerone program. We would have six total off-flavors to train on (3 fermentation related and 3 handling related1). For the first phase, we would be blind2 to the off-flavors. After taking notes and assigning some guesses, we would find out which off-flavors were in front of us. Third, we would find out which samples had been dosed with which spikes. Lastly, after confirming, we would go back through again, re-tasting and discussing along with some supporting materials.

Blind Phase

After calibrating our palates against the control beer, we received 2 oz. each of the six spiked beers. Each glass was marked with a small colored dot. We performed our initial blind evaluations in silence.

Dot Aroma Flavor Conclusion
  lightstruck? “Old Heineken” — catty stale/papery — back to lightstruck (“Old Heineken”) — some skunk lightstruck (or oxidation?)
  diacetyl? vinyl? vinegar! vinegar! acid! acetic acetic (infection)
  butterscotch? “artificial sweet” buttery — “artificial butter” VDK? diacetyl
  plastic? harsh? papery (but… not?) — can’t place no guess
  cooked vegetable? (DMS?) later: like the veggie blew off? subdued & (again) papery? — some vegetal? DMS
  rose? fruity? (estery?) fruity (again) ester

I was very confident in two of my conclusions (lightstruck and acetic), and fairly confident in another two (diacetyl and DMS). I was less confident in the “ester” guess — but only because it seemed like an odd thing to include.3 And then of course that last one where I simply had no idea.

Match Phase

After everyone confirmed that they’d made their assessments, the list was revealed to us:

  • Dimethyl sulfide (DMS)
  • Diacetyl
  • Acetaldehyde
  • Trans-2-nonenal (T2N)
  • Lightstruck
  • Infection

With the knowledge of which specific off-flavors to look for, we went back through the samples to quickly reassess them. Which conclusions would we stand by? Which would we adjust?

So with that list, I went back through and re-matched as follows:

Dot Original Conclusion Adjusted Conclusion
  Lightstruck Lightstruck
  Acetic (infection) Infection
  Diacetyl Diacetyl
  no guess T2N
  Ester Acetaldehyde

Reconciliation Phase

This was the big reveal — where the secret identity of each dot’s off-flavor was unveiled and we could see which we’d gotten right, and which had thrown us off.

Dot Original Conclusion Adjusted Conclusion Actual Off-Flavor
  Lightstruck Lightstruck Lightstruck
  Acetic (infection) Infection Infection
  Diacetyl Diacetyl Diacetyl
  no guess T2N Acetaldehyde
  Ester Acetaldehyde DMS


We watched the video that went along with the off-flavor kit, reconciling our hits and our misses now that we had the complete information. During that discussion phase, I took down some notes to help improve my performance in future evaluations and judging.


Guessed it blind. And with a high degree of confidence. I knew this one immediately and well.


Guessed blind. We had an interesting discussion about how most of us called it diacetyl at first. What was most interesting about this fact was how the video mentioned an overlap between the diacetyl and the acetic elements — that those two off-flavors often occur together in an infected beer because the microbiological agents are producing both of those compounds. Either way, the acetic flavors here were strong for sure.


Guessed blind. Which was reassuring for me, because I’d concluded last year that I was “not exactly diacetyl blind, but certainly diacetyl near-sighted.” I probably do have a slightly above average detection threshold, but not as unusually high as I thought it might be. Regardless, to my palate, diacetyl comes across more like butterscotch than butter — and the impression is that it’s from a 100% artificial product. Or as one of my friends summed it up “like a sugar-free Werther’s.”


While I was relieved about not being diacetyl-blind, I may as well have been blind to acetaldehyde. I didn’t guess this one at all during the blind phase. And this is an off-flavor that I’ve historically not been able to detect when others were calling it out — and granted that was in part due to the fact that I didn’t have a good reference for it. During this training, my impressions of acetaldehyde were of plastic and clay — Play-Doh (of all things) and latex paint were the strongest impressions. I did get some of that “raw pumpkin” that other people talk about though. By and large, it was mostly “latex paint” in the aroma and some “raw pumpkin” on palate.

And maybe that’s why I always had trouble picking it up before. It’s so often cited as “green apple” and that is just not at all what I got from this.


I managed to call this out, but on the wrong sample. To my palate, I did indeed get that papery, wet cardboard impression. And the dull flavors. And/but/so… I want to know how I managed to get “cooked vegetable” here during the blind phase. (The dull flavors, I guess?)

Dimethyl Sulfide

Again: called out during the blind phase, but on the wrong sample. To my palate, I picked it up as cold cooked vegetables; corn was in there for sure, but it seemed less specific overall.

  1. Although I didn’t know that at the time. []
  2. ”Semi-blind” is probably a more accurate statement here. If I recall correctly, the regular study group members had some foreknowledge of the off-flavors that would appear in the kit. []
  3. I realize that most lagers should have very low-to-no esters, and that excessive esters can ruin a beer of nearly any style — but they’re acceptable enough in most ale styles, and necessary for certain Belgian and German wheat styles. It also seemed like, if any ester was going to be in the kit, it would be isoamyl acetate, and not something with a rose-like quality. []

About Rob Friesel

Software engineer by day, science fiction writer by night. Author of The PhantomJS Cookbook and a short story in Please Do Not Remove. View all posts by Rob Friesel →

Leave a Reply

Your email address will not be published. Required fields are marked *