• Coffee Smells
  • Author avatar
    Erik Johnson

Coffee Smells

Developmental events can be experienced through smelling coffee as it roasts. It takes a lot of time, but eventually you can recognize smells which help you to know the development of the roast while it’s in the roaster. And being able to do that is the key to being a good roaster.  

I came upon the desire to use and recognize smells for the benefit of the roast because of two things. First, when I started roasting it was at home with a popcorn popper. For those of you who have used a popcorn popper to roast, you understand what I’m describing when I say it makes roasting a transparent process. Entirely transparent. I encourage everyone to go to a thrift store, buy a Westbend Poppery II (the kind that every suburban home had in the late 80’s early 90’s with the orangish-transparent lid). Second, I once watched Joe Marrocco roast at Cafe Imports, and for the first time I saw someone smell the coffee in the trier. Mind blowing.

When I had gotten my first job as a roaster, I immediately grieved the loss of my popcorn popper. I no longer had the direct experience of bean development. I had to link the developmental events that I knew by observation with a thermometer and watch. But that day I saw Joe roast, it made me think that I should start smelling the roast again. Up until that point I had only seen people use the trier to look at the beans. I find looking helpful, but at that point I wasn’t gaining any addition info by looking at them. Smelling on the other hand was an entire set of information that I hadn’t been utilizing since I huddled over my ½ cup of green beans spinning away beneath my eyes, nose and ears in the popcorn popper.

When I went into work the next morning, the first thing I did was begin to smell every batch, sniffing the beans throughout the roast. It was certainly informational, but it was so much categorized information that it didn’t do me much good; other than confirming that there was so much being communicated, none of which I could understand. It made sense that I should try to limit the information coming into my snout and combine it with things I did know. I knew I like the Peru we had at the time. I knew how I like to roast it; what temp and time and what method (how I categorized profiles in my head). Starting with that, I began to smell.

I smelled probably every 10-30 seconds. I had already figured out that the amount of information communicated through smells is exponentially greater post First Crack than pre. So in the Drying Phase I would pull the trier less than I would once the Maillard Reaction began.

As the months went by I had worked my way through all the roasts and had a fairly good sense of consistencies; general olfactory observations of roasting coffee. But something that I began to utilize immediately on my roasts were developmental events that were only being noticed by smell.

These events were occurring consistently across all coffees, every roast. As the coffee gets further into First Crack, the longer it is in the Developmental Stage, there are these unmistakable bursts that the beans emit every few degrees. Just by holding the trier close to your face, without even smelling, you can feel the bursts on your lips and nose. When you take in the smell you can sense the difference in the beans. There is a different smell at 405F than at 408F.

Now, I can recognize specific smells that tell me about the roast. when I am writing a profile, or throwing a batch without roast logging software, I smell constantly to gain knowledge about the roast. The time, temp and color are large developmental markers, and smells are minute. So for instance, when I get into the large category of Development, I use my sense of smell to figure out the details.
  • Author avatar
    Erik Johnson