top of page

How To Achieve The Best Sample Roasting Curve

Green coffee quality analysis implies good sample roasting skills. It is a mix of theory and experience and every variable changes according to the environment conditions, altitude, roasting machine and coffee beans.

Nowadays, roasters and producers are more connected in order to understand how coffee should be roasted based on the process, variety, farm conditions and more. At Dinamica Coffee we know how sample roasting influences green coffee purchase decisions and so we want to share with you the key parameters to achieve the best sample roasting curve.

Peter Engelhardt roasting samples

Why Is It Important To Establish A Roasting Curve?

Peter Engelhardt, our quality director, explains that sample roasting is about finding the best roast profile for a coffee bean in order to perceive all its qualities and defects and judge fairly every sample he gets.

Getting coffees from different farms, different varieties, processing methods and more, requireS an independent analysis and this is the case for sample roasting. The cupping results from each sample help us provide feedback to the producers and decide whether to buy their coffee.

He explains that many coffees are being analysed throughout the year, from different varieties to processing methods and different regions. And in order to roast each sample, there are some key variables to consider in order to roast. For instance, some varieties such as Pacamara and Maragogype are less dense and bigger in size than Caturra and Bourbon, therefore, they would require lower charge temperature and a longer roasting curve in order to achieve better caramelization during the roast, more on that later.

Analysing green bean samples before roasting

Moisture Content And Density

According to Peter, moisture content and density measurements are the basis to approach a green coffee sample and decide how to roast it. Denser beans normally require higher charge temperature which have a lot of correlation with the bean size (e.g Caturra and Bourbon being smaller in size and more dense than Maragogype and Pacamara).Regarding moisture content, it should be between 10% - 12% to be able to roast.

Measuring moisture content of green bean samples before roasting

The Process

First of all, accessing each sample for physical analysis and defects discharge is a key, this should be done in order to get an uniform roast. We don’t want to ruin the sample due to a physical defect that might burn during the process.

Peter explains that his workflow consists of measuring moisture content and density and based on this he organises what to roast first. Normally he roasts softer beans (lower density) first and then hard beans (higher density). Softer beans’ cellular structure is less solid than hard beans, therefore, heat transfer for both types of beans should be managed differently. Normally, hard beans manage heat transfer better but also require higher charge temperature and soft beans would require lower charge temperature to avoid over roasting. For soft beans Peter starts with 120°C / 248°F and for hard beans with 150°C / 302°F.

At our SCA certified lab we use four Probat barrel sample roasters and we try to keep our variables very simple as we normally use all the drums at the same time to be more efficient. We roast 140 grs per batch and keep gas pressure constant at 70%, the only variable we modify along the roasting process is airflow.

Probat sample roaster at Dinamica's certified SCA lab


Airflow in roasting is heat being managed through convection. This is one of the two ways that heat impacts the bean during roasting. The other one is conduction, meaning heat transfer through the metal of the drum. With our sample roaster we focus on managing airflow without modifying gas pressure. This helps to achieve consistency more easily with the four drums by adjusting only one variable.

  • At the beginning of the roast, airflow is kept at 20% to allow moisture release from the bean during the first minute.

  • Once the bean has reached its turning point at 90°C/194°F and temperature starts to increase, at minute 6 when the yellowing phase begins, we open airflow to 50% in order to extend this phase up to 3 minutes before first crack.

  • Finally, crack begins at minute 9 at 200°C / 392ºF so we open airflow to 80% in order to clean the beans from chaff and give clarity to the cup. The caramelization phase lasts up to 1’30’’ for hard beans and 1’20’’ for soft beans.

Peter explains that when airflow is too high from the beginning, it might delay the roast and create baked notes in the cup. Airflow is a crucial variable and contributes to clarity in the cup when managed right. It is up to each machine’s system and how you want to play with the different variables influencing the roast.

Sample roasting contributes to the buyer’s purchase decisions and understanding the main variables influencing the roast and keeping them constant according to each coffee and roasting machine is key to achieve the best curve that will allow you to evaluate and judge each coffee the best way. Ultimately it is the producer’s work and you want to perceive and understand it in order to make decisions and provide the right feedback.

Recent Posts

See All


bottom of page