Roast level is more a matter of physical changes vs. color. Development is a different matter altogether.

Before the end of First Crack = Cinnamon
End of First Crack = City
Just before 2nd Crack = Full City
Just into 2nd Crack = Vienna
Well into 2nd Crack = French

Obviously a photo can't tell us these things that only you could have witnessed.

The focus required for this type of roasting is easy to underestimate. Maintaining a steady & effective stir with the whisk is important. Constructive criticism here is that the whisking/stirling was not done at a steady or efficient pace, based on the pics.
Use weighed batches, and keep roasting the same coffee until you get good at it. After a while you see that a specific amount of coffee at a specific temperature stirred in a specific way for a specific length of time yields a specific result. Get a laser thermometer, and record the stats of each roast, along with what you thought of the result in the cup.

I've had a friend remove a quarter cup of beans every thirty seconds after the middle of first crack until the middle of second crack, and each quarter cup was sampled to determine which I liked best. This method is hardly imperial for the stovetop roaster, but since I can roast just over a pound per batch, it's proven worthwhile.

Hey Endless Cycles,

I've recently sent a vintage La Cimbali two group Lever machine from 1970 to Asheville from Eugene, Oregon. It's a beautiful machine, and I have a lot of work into it. I'd love to hear if it's being used well. It's soon to be installed at a swank new bar called Sovereign Remedies on the corner of Walnut & Market St. I've never been to Asheville, and never met the owner of the bar.

endlesscycles wrote:Roast level is more a matter of physical changes vs. color. Development is a different matter altogether.

• Before the end of First Crack = Cinnamon
  End of First Crack = City
  Just before 2nd Crack = Full City
  Just into 2nd Crack = Vienna
  Well into 2nd Crack = French

Obviously a photo can't tell us these things that only you could have witnessed.

Marshall makes a good point, however the situation is a little more complicated than he paints it. If you roast too hot and fast, the outside of the bean will color before the inside of the bean cooks. If you roast too cool and slow, you'll dry out and bake the bean relative to its finish color.

As you've surmised though, color is a very important diagnostic. If it weren't Agtron wouldn't make money.

The nice thing about roasting in an open roaster is your access to sights (including color), sounds and smells. In terms of finish level, besides color, you want to keep an eye on how much skin remains in the crease, the amount of oil which appears on the beans post-roast, and how quickly it develops.

After I've established the bean temperatures for roast milestones (like crack temps) with a "sample roast," I rely very heavily on RoR and other real time, temperatures and temp related relationships. It allows me to repeat times for every interval consistently; anticipate certain events without obvious sensory information in order to adjust power and airflow in advance; hit exact drop temps, etc.

It's improved my roasting in many ways, but sometimes I miss the purely sensory approach.

Rich