Transition from vegetal flavor during first crack

Discuss roast levels and profiles for espresso, equipment for roasting coffee.
User avatar
drgary
Team HB
Posts: 14297
Joined: 14 years ago

#1: Post by drgary »

In a thread about a roasting spec mentioned during Anthony Douglas's winning presentation at the 2022 WBC championship, Neal Wilson mentioned a vegetal flavor that suddenly changes at about 10°F past first crack start - on his equipment.

This was the originating thread.

Anthony Douglas WBC Roast Profile

I've probably been tasting this and not recognizing it in my roasts. But here's the quote where he mentioned it and my follow-up question started a discussion worth a topic of its own.
N3Roaster wrote:I've done this a few times as part of an exploration into certain flavor changes that happen early between cracks. My lab roaster lets me pull enough to cup from the trier while maintaining a rate of change steady at a bit under 1F°/min. There is a particular flavor which I personally find objectionable (my preference here is not universal) that never goes away until 10F° above the start of first crack regardless of anything else in the roast profile.
Neal, what is that flavor?

Otherwise, on my propane, drum roaster, I'm able to set a low ROR at 1C if I don't have excessive heat when charging. There's a thread here somewhere, where several of us tried stretching the roast with minimal temperature rise at the start of 1C to emphasize fruity notes.
Gary
LMWDP#308

What I WOULD do for a good cup of coffee!

N3Roaster
Posts: 116
Joined: 12 years ago

#2: Post by N3Roaster »

drgary wrote:Neal, what is that flavor?
It's somewhere in the family of vegetals. I call it a raw coffee flavor. Some coffees hide it better than others, but once you train on it the flavor becomes difficult to ignore. What's truly remarkable is that this flavor component undergoes a rapid transition from present to absent that is strongly connected to maximum temperature of the coffee to the point that it can be used to work out equivalencies between different machines within the light roast range. I've been able to narrow that down to ±0.2°F and it's happening there regardless of whether that's this sort of stretched out roast or any more common profile shape (I thought this was weird and unexpected given that this is not how flavors in coffee generally behave, but it also turned out to be fully repeatable on blind and randomized evaluations across a range of very different coffees and very different roast profiles prior to that point). If anybody is interested in replicating that work, I'd recommend starting with a neutral washed central appropriate for use as a base for flavored coffees and really stretching things out. The lack of other distracting flavor elements makes this transition quite dramatic and once you're picking it out in a coffee like that it's easier to identify it in more interesting coffees.
★ Helpful

Milligan
Supporter ❤
Posts: 1470
Joined: 2 years ago

#3: Post by Milligan replying to N3Roaster »

The coffee in question went through an 80 hour anaerobic fermentation. Did you find the note to change a bit from washed to natural (I figure it is reaching to assume you may have tried it on a typically more expensive process like anaerobic fermentation)? I'm wondering if a heavily fermented coffee would no longer exhibit that vegetal note and instead have a sweet fermented note? I know smelling a raw washed "good" coffee vs a raw heavily fermented coffee has a drastically different aroma. The washed coffees typically smell of hay/grass while the fermented could smell sweet and winey/fruity even before being roasted.

N3Roaster
Posts: 116
Joined: 12 years ago

#4: Post by N3Roaster »

The flavor that I'm referring to is always the same and was identically expressed in naturals. You are, however, right that I haven't repeated this with an anaerobic, but I'll make a note to at least do a small trial the next time I have one, and more to the point I haven't tasted the exact coffee under discussion. I can, however, say that with better and more interesting coffees one might still reasonably choose something at these lower ending temperatures. It won't produce something that I'm likely to enjoy, but there are certainly people who can identify that flavor and do not perceive it as objectionable so if there are other flavors well expressed there, it can be a reasonable choice. That said, those other flavors tend to still be present and can often even be strengthened going a few degrees hotter.

GDM528
Posts: 830
Joined: 2 years ago

#5: Post by GDM528 »

N3Roaster wrote:I've done this a few times as part of an exploration into certain flavor changes that happen early between cracks. My lab roaster lets me pull enough to cup from the trier while maintaining a rate of change steady at a bit under 1F°/min. There is a particular flavor which I personally find objectionable (my preference here is not universal) that never goes away until 10F° above the start of first crack regardless of anything else in the roast profile.
Interesting... 10F above first crack, for how long?

I think I may be experiencing the same effect, and I've been experimenting with various ways to work around it: overshooting and ramping back down to first-crack temperature mitigates but doesn't completely eliminate the off-note. Fortunately for me, that harshness seems to disappear in milk drinks.

N3Roaster
Posts: 116
Joined: 12 years ago

#6: Post by N3Roaster »

GDM528 wrote:Interesting... 10F above first crack, for how long?
Doesn't matter. The change I'm talking about is effectively instant and just happens at that temperature regardless of whether things are really stretched out (it's easier to identify the flavor or its absence this way since it'll mute a lot of other flavors) or if things are still progressing quickly (as mentioned before, this was unexpected, I was looking for a time/temperature relationship and was very surprised to find that the time didn't matter here). In a progressive roasting exercise all the cups below that temperature would have that note and all the cups above that temperature do not. Important to note that's as measured on my machines. For context on my machines I'm getting start of first crack at 380°F, this temperature we're discussing now at 390°F, and start of 2nd crack at 430°F. If the first and second crack temperatures aren't the same you'd likely find this to happen at a different temperature, but it should still be in the neighborhood of 1/5th between cracks.
★ Helpful

Milligan
Supporter ❤
Posts: 1470
Joined: 2 years ago

#7: Post by Milligan replying to N3Roaster »

This is an interesting observation. That would mean that the note is strictly temperature dependent and not necessarily roast level dependent. A faster RoR to 10F above first crack would give a lighter roast than a slow RoR to 10F above first crack. Have you taken roast level (agtron/roastvision) measurements around where this is observed to be present vs disappear?

N3Roaster
Posts: 116
Joined: 12 years ago

#8: Post by N3Roaster »

That's correct. You should be able to get different agtron readings for the event. Unfortunately, the way that I collected the samples for tasting doesn't get quite enough coffee to run through my degree of roast analyzer and in production I tend to find benefit to going a little hotter still (I think the lowest I've got right now is ending at 404F and something closer to 420F is where my personal preference tends to put things when I'm not going into 2nd crack) so it's not convenient for me to get a spread of tested roast level measurements. It's something to keep in mind if I have the opportunity to repeat this on a larger machine.

Marcelnl
Posts: 3827
Joined: 10 years ago

#9: Post by Marcelnl »

My experience is similar, yet it never dawned upon me that there is a temperatire threshold involved, in my roaster FC sits around 198 'C (388F) and anything below 204-5'C ish (399F) leaves me with a 'chlorofyl-like' note I do not like, for a while now I finish at 206-207'C (402F) and it's gone at that temp. I tend to try race my roasts to the finish in order to preserve as many flavor compounds as possible to maintain complexity.
LMWDP #483

GDM528
Posts: 830
Joined: 2 years ago

#10: Post by GDM528 »

Any theories on the chemistry at work here: Trigonelline, Malic acid, etc.?