Why does my roast color look so uneven? Home coffee roasting issues…

A 5 minute conversation downstairs on roast color at Sweet Maria’s shop ends up as a 26 minute video on roasting dry-process coffee and uneven surface color. Is that a good thing?

This started as a simple conversation between Chris, Erik and myself. They had roasted the same dry process Ethiopian coffee, one on the Aillio Bullet, the other on the FreshRoast 800. They looked so different! One was very even in color, the other had lighter and darker beans.

We wanted to turn the conversation into a short video … but as you pull apart the assumptions and issues with roasted coffee color appearance, it’s quite a tangle! I ended up with a 26 min. video out of this, breaking out the influence of roast level (dark roasts always look more even), a focus on the variability of dry process aka natural coffees, the presence of chaff, and pretty much everything else. I end up looking at a fancy 90.4 scoring Gesha natural that was probably the most troublesome roast of the whole video (I scored it 87, but the defect cup from the light beans scored 82 for me!). 

The process wasn’t neat here. But it made me look at my assumptions…

  • Is a roast with bean-to bean color variation a “bad roast?”
  • Does uneven color mean uneven roasting? What does uneven roasting mean anyway?
  • And shouldn’t we evaluate coffee by taste in the end, not “eye-cupping” it (which sounds impossibly painful anyway. )

The chapter is not closed on all this topic….

#homeroasting #greencoffee #roastingcoffee #roastedcoffeecolor #quakercoffee #unripecoffeefruit #coffeedefects #coffeecupping 

Sweet Maria’s offers green coffee and coffee roasters for the home. Our sister business CoffeeShrub serves small roast businesses.

8 Responses

  1. I found this one insightful and entertaining. I too eat my roasted coffee beans post roast but not typically the obvious quakers. Now I will be trying them too. Dry process are the most challenging type in my roaster (a home built machine) and I would love to understand what makes them tend to roast darker than planned (when roasting like a wet process), sometimes quickly bridge 1st to 2nd crack, and occasionally run away on me.

    1. Moisture content is one thing that can make a coffee behave very differently. Some of the Ethiopias have lower moisture content of 8.5 or 9% compared to other origins that typically arrive with 10.5% or so

    2. Could it have something to do with the the seeds drying with the fruit still attached?

    3. Hi Greg, that’s certainly part of the equation! Dry process coffee coloring when ‘green’ (unroasted) varies quite a bit within a single lot. A lot of that comes from the fruit drying intact, though the darker color is often the silver skin, and not the coffee itself. There are other factors that play a role too, which Tom touches on in the video. One of the main takeaways for me is that uneven roast color by no means equals a bad tasting coffee. It can be par for the course with light roasted, dry process coffees, which happen to be some of my favorites.

      Glad you found the video interesting!


  2. Nice video. I watched it because I am getting varying results of the Ethiopian Guji Gerba (no longer listed at Coffee Shrub) despite multiple roast attempts across a Bullet, Roest, and Artisan 3e. I did a safe medium roast on the 3e which was ok. My Bullet roasts are unbalanced with a lot of variation in the roast color across beans.

    1. Wow you have a real arsenal of roasters there… I did find gerba to be a bit challenging too. I like the cup, but in terms of roast color it seemed to have a lot of variation. When we looked at this new crop selections, I kept that in mind when I was in Ethiopia since a good coffee selection for SM is about good cup, but also about it not being a PITA to roast! I think the super high altitude of that coffee and the varieties planted in the area have something to do with it.

  3. Is the SR540 temp display a reliable way to determine roast level? For example, if the display reads 426 and I stop the roast is it City? Or do I need to “hack” the machine with a thermocouple inserted into the spinning beans?

    1. Hey Jason, thanks for your question.

      There is a temperature difference between the readings on the panel and actual bean temp. That’s not unusual for a lot of small roasters with built-in sensors. But that does not mean they aren’t useful. It’s just a matter of finding the numeric benchmarks that line up with the roast level you’re happy with.

      You can add a thermocouple, but I believe it involves drilling a hole in the lid (I’ve not tried myself). I know there’s a lot of information out there on adding thermometry to read bean temp, and Artisan (freeware roasting software provider) supports the FreshRoast. We tend to steer clear of advertising the mods, as we don’t want to take responsibility for any damage that may occur. But I will say that lots of people have modded this roaster, so you should be able to locate a vid or post about this online 🙂

      We do sell thermocouples as well – HERE.

      Hope this helps!


Leave a Reply

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

This site uses Akismet to reduce spam. Learn how your comment data is processed.