Last month, we visited John Nanci, the Chocolate Alchemist in Oregon. It was a lot of fun and we learned a ton about his process, ingredients, and recommendations!
One of our favorite parts of the visit was roasting a batch of cocoa beans with him. Check out our previous posts on roasting here. We picked out the beans after sticking our heads in all of his big barrels of beans and smelling the wonderful scent of raw beans over and over again. I wish I could share smell through this post, because it’s incredible how different (and delicious) each barrel smells. We decided on the Venezuelan Carupano Corona, 2014 harvest, which has a savory, almost spicy scent.
In the meantime, John had turned on his homemade roaster to start heating it up.
John Nanci’s homemade roaster
This incredible device has two coupling thermometers which show the temperature inside the cylindrical drum of beans and outside the drum, where the heat originates. He filled the canister with about 5 pounds of beans and when the roaster hit about 400 degrees, he put on his heat-protective gloves and lowered the drum into the roaster. He closed the top and we started to chat.
Let me set the scene… we’re in an open garage/workshop with a misty rain keeping the humidity high, though temperatures were likely in the mid-50s. Everything in the workshop smells like a part of the chocolate-making process: from raw beans to the brownie smell of roasting to the almost syrupy smell of undeodorized cocoa butter.
As as chatted, every once in a while, John would pause, waft some of the rising hot air from the roaster over toward him, and comment on the smell. He glanced, every once in a while, at the coupling thermometers to gauge the difference between the temperature inside the bean canister and outside in the roaster, but the majority of his conclusions about the progress of the roast happened through his nose. He got excited when the smell seemed to waft over to us suddenly, letting us know that this is typically the peak of the roasting process. Around that time, we lowered the temperature and eventually he turned the roaster off entirely as they continued to roast. And he pointed out again when the smell shifted from our noses to the back of our throat. That was when he recommended taking them out. He waited even a few more minutes before pulling out the drum, dumping it onto his cooling table, and aiming a fan at the beans.
He emphasized that, unlike coffee, cocoa beans have more forgiving roast potential. In other words, if you leave them in too long, they are less likely to become disgustingly over-roasted than coffee beans. In fact, he made sure to point out that it is difficult to over-roast cocoa beans.
To be honest, this whole process was like watching a magician at work. His enthusiasm was contagious and Richard and I understood the intent but couldn’t necessarily recreate the magic in our own noses and throats.
That’s why we decided to attempt to match the roast. So, we brought home those roasted beans as well as a few pounds of the same beans, unroasted. And over the weekend, we brought out a bottle of wine and the two batches of beans, and did our best to recreate the process. And – believe it or not – we’re not quite as good as the Alchemist himself!
Matching the roast tools
We set the oven to 400 F and prepared to follow our noses. A few caveats before we get started:
- Unfortunately, we were both recovering from a cold, so our sense of smell wasn’t quite up to snuff.
- We used our relatively old oven, not a self-engineered roaster.
- The temperature in our apartment was in the high 60s and not at all humid, compared to John Nanci’s colder, humid garage.
In other words, we may have been doomed from the start! That said, we did take some of John’s advice very seriously, including the following seemingly logical advice:
- Stir the beans regularly. That could avoid “tipping.” Tipping is when the part of the bean touching the hot pan will roast faster (and potentially burn) than the rest of the bean. It develops an uneven roast and could add extra toasting flavors to the chocolate.
- His other brilliant advice wasn’t possible this time. He suggested doubling the pans, so there’s a more even distribution of heat on the bottom of the beans. However, we have exactly two pans and used them both for this roast, so we’ll need to try this next time.
This process seemed to happen at full speed, as I was taking notes, flipping beans, checking the clock, sniffing to the point of hyperventilating, tasting hot beans, tasting pre-roasted beans, and hand-winnowing as we went. Whew! Here’s the run down…
We flipped the beans after 5 minutes and at 10 minutes, we started hearing the snapping in the oven and the smell of brownies pervaded the apartment. The taste of the beans at that point was still quite raw and chalky, but the cocoa mass felt softer than a fully raw bean.
We dropped the temperature to 250 at 10 minutes and by 12 minutes, it smelled like dark brownies and we started to get the sense in the back of our throats. We reasoned that they couldn’t possibly be done yet, and took John Nanci’s words to heart… it’s very difficult to over-roast cocoa beans.
At 14 minutes, we flipped them again and at 20 minutes tasted a second time. This time, they tasted bland, almost nutty, without much flavor development.
At 22 minutes, we pulled them out and did a full flip of the beans with a spatula rather that stirring them around in the oven (Richard advised me that I wasn’t flipping quite right, so this would be a more robust flipping system). We compared the taste to John’s beans at this point (starting to get giddy eating so many beans) and noted that ours tasted chocolatey and rich but the texture still felt raw – hard and not crunchy yet.
At 26 minutes we pulled out a really bad bean that tasted underfermented; not particularly helpful in our comparison. A minute later, we found a good one that tasted pretty toasty and nutty. We compared it to John Nanci’s beans and noted that his had more flavor at the end, almost caramelly.
At 28 minutes, we pulled out the tray to flip and put it back in 2 minutes later. As I flipped, Richard tasted and at exactly 31 minutes, we pulled out all the beans determining them definitely done, if not overdone!
We quickly used Richard’s brilliant newly engineered cooling system for about 15 minutes until they felt very cool.
Roasted bean cooler
And the result – our beans definitely taste different than John Nanci’s beans. Ours taste a little over-roasted and slightly bitter at the end, while John’s beans have that caramel finish. Whew, we’ll try again next time!
Any suggestions from the audience on how you train your nose for the perfect roast?