Heat Transfer Methods Involved In Roasting

Heat is a form of energy in transit. Heat moves from higher concentration to lower concentration. When two objects come into contact with each other, heat will flow between them until the two objects reach thermal equilibrium or are separated and are no longer in conctact.

There are three main forms of heat transfer that we’re concerned with in drum roasting; convection, conduction, and radiation.


Convection is the dominant form of heat transfer throughout the roast and is a key driver of coffee body and flavor formation.  Convection heat is heat moving through a fluid such as a liquid or a gas. The heat movement can be natural, in which case warmer fluid rises and cooler fluid drops. Convection can also be forced using a fan or a pump. In drum roasting, we have forced convection moving heat from the heat source (burner, heating element, heated metal, heated beans, the drum surface, etc) towards the exhaust. We also have natural convection with liquids inside the bean such as water, air, etc.

We can manipulate forced convection in a drum roaster through heat application at the burners as well as air movement through the drum itself via fan speed, air dampers, etc, depending on the design of the particular roaster.

Some things to keep in mind with air control:

  1. Air flow has a dehydrating effect.
  2. Water is necessary for forming better sugars later in the roast
  3. Water has a high heat capacity and therefore has both an insulative effect and a heat-sink effect

As with everything in roasting, balance is key and experimentation is the best way to find what roast profile your particular beans need.  Remember when experimenting; change only one variable at a time so you know what change had a particular effect.  In the case of playing with airflow and sweetness, maintain time and temperature as normal for a given profile, but adjust airflow during one phase only and see what the outcome is.  For example, if you normally have a 5:4:3 roast (5 minutes drying phase, 4 minutes maillard phase, 3 minutes roast-development phase) then maintain a 5:4:3 profile, with the same temperature curve as normal, but during the maillard phase, test using less air, then do another roast exactly the same except using more air in the maillard phase.  Then cup and compare the two roasts and see which turns out better.

In the case of roasting coffee, convection transfers heat from the source to the coffee bean, where conduction occurs.


Conduction heat is transferred through direct, physical contact.  Heat travels from higher concentration to lower concentration and can occur with bean-to-fluid contact (air), bean-to-bean, bean-to-drum, etc.  A key area of concern regarding conduction is the extent or intensity of the pre-heat.  Think of a hotter drop temperature as more stored energy in the system (the metal of the drum, the backplate, the air, etc).  As soon as beans come into contact with heated matter, they begin absorbing heat energy. The metal in the roaster is of special concern early in the roast because if it’s too hot, it can cause roast defects.

Symptoms of too much conductive heat include tipping, uneven roasting, and mottled, scorched beans.  In a drum roaster, conductive heat is difficult to directly control. For example, cutting back heat application at the burners does not have an immediate effect on the temperature of the drum. There is some lag time between control changes and their effect on conduction.


Radiated heat is the most complicated of the three forms of heat transfer and the most difficult to measure or control.  Radiated heat travels at the speed of light and unlike convection or conduction, radiation requires no medium to carry it.  Instead, it is emitted as electromagnetic waves. All surfaces within the roaster produce thermal radiation as do the beans themselves.  Because of the difficulty in controlling or even measuring radiated heat, it’s best to simply be aware of its presence and its ability to affect total heat.

None of the heat transfer forms are independent of one-another.  The air will conduct heat to both the beans and the drum while at the same time the pre-heated drum will conduct heat to the beans, etc. While it’s difficult to independently control any single form of heat transfer, it is very useful to know which method is dominant at which stage of the roast.

According to Terry Davis, the following shows which heat transfer method is typically dominant during the three roast phases1:

  1. Drum, Air, Bean
  2. Air, Drum, Bean
  3. Bean, Drum, Air


At a minimum, you want to measure the environment temp and the temperature of the bean mass.  To get the air temperature, we want a temperature probe extended into the airspace of the drum itself.  We don’t want it too close to any metal surface and we don’t want it too close to the bean mass itself.  For the bean temperature reading, we want a temperature probe placed so that the probe is always in contact with beans as they roll around the drum.  Having temperature measured from these two locations give us some insight into what is happening inside the drum and the momentum we have remaining, i.e. are we currently increasing environment temperature, or are we on course to maintain current environment temperature or are we losing environment temperature?  Knowing this helps the operator to reach a desired time and temperature target consistently. For more on the momentum of a roast, read about the rate of change (or rate of rise).

See also: How To: Avoid The Flick And Crash


In a drum roaster, the bean temperature is the result of the environment temperature, which is the result of the burners, airflow, and heat radiated from hot surfaces inside the roaster and at later stages of the roast, the beans themselves.  Roasting based solely on bean temperature is reactive.  The optimal system controls the roast based on the rate of change in both environment and bean temperatures as well as the difference between the two (a larger difference means more potential energy to do work).

In the roast development stage, just after first crack, the beans become exothermic meaning ongoing chemical reactions are producing heat as a byproduct and that heat is of course added to the total heat energy of the system.  Therefore it is important to be aware of and accommodate for this new source of heat.  There is infinitely more heat available at the end of the roast, which makes it very easy to lose control and overshoot your target roast.  Remember your roast benchmarks as you go!

Some things to remember with heat transfer:

  1. The amount of moisture present in green beans influences the speed of roasting
  2. Higher airflow may require higher heat energy to avoid heat loss

Knowing the three methods of heat transfer in a drum roaster and having accurate and consistent temperature readings of the environment and the bean mass give us better control of the roast and allow us to more accurately hit time and temperature targets.  And knowing is half the battle.

1. Terry Davis, “The Heat is On, A Roaster’s Guide To Heat Transfer,” Roast Magazine May/June 2009

  Continue Learning: Looking for more articles about roasting? Check out this page for more!

Value 4 Value

  If you found this content useful, please consider supporting my work. I charge no set fee or price for providing this. You can help keep information like this openly accessible by matching the value you received in the content; value 4 value.

Ko-fi / Bitcoin Wallet: 32SW9kcAsJdZvQKBazhLUZBSD9YS8DDqe8

Learn more

Michael C. Wright

Michael is a licensed Q Grader, licensed Q Processor Pro, an Authorized SCA Trainer (AST), and most recently, a graduate with a degree in horticulture and a concentration in horticultural business management. He has over ten years experience in the coffee industry operating on both the supply and demand sides of the value chain.