Solder Reflow is accomplished using an equipment known as the solder reflow oven. Reflow ovens employ two major techniques to expose the board assembly to the necessary temperature profile, namely, Infrared (IR) Reflow and Convection Reflow, which are not necessarily exclusive of each other. In fact, many modern ovens utilize both techniques to achieve excellent reflow results.

IR reflow involves the transfer of thermal energy from infrared lamps to the board assembly. The board assembly is heated by IR reflow primarily by line-of-sight surface heat absorption and because of this, variations in the density of the board can result in 'hot spots' (or localized areas with significantly higher temperatures) on the board during IR reflow. As such, some components experience higher stress levels than others on the board even if they're subjected to the same IR reflow conditions.

Convection reflow transfers heat to the board assembly by blowing heated air around it.  Convection reflow provides a more uniform heat distribution to the circuit assembly compared to IR reflow.

There is actually a third method for solder reflow, although it is used to a much lesser extent today than the first two techniques mentioned.  Vapor Phase Reflow, as it is called, transfers heat to the board assembly by boiling inert fluorocarbon liquid and enveloping the board with its resulting vapors. Its major drawback is that its reflow temperature depends on the boiling temperature of the liquid used.

A solder reflow process follows an optimized temperature profile to prevent the board from experiencing unrealistically high thermal stresses while it is undergoing reflow.  A typical reflow temperature profile would consist of the following steps: 

1) Preheat, which consists of gradually ramping up the temperature to the preheat zone temperature at which the solvents will be evaporated from the solder paste;

2) Flux Activation, which consists of bringing the dehydrated solder paste to a temperature at which it is chemically activated, allowing it to react with and remove surface oxides and contaminants;

3) Actual Reflow, which consists of ramping up the temperature to the point at which the solder alloy content of the solder paste melts, causing the solder to sufficiently wet the interconnection surfaces of both the SMD's and the board and form the required solder fillet between the two; the peak reflow temperature should be significantly higher than the solder alloy's melting point to ensure good wetting, but not so high that damage to the components is caused;

and 4) Cooldown, which consists of ramping down the temperature at optimum speed (fast enough to form small grains that lead to higher fatigue resistance, but slow enough to prevent thermo-mechanical damage to the components) until the solder becomes solid again, forming good metallurgical bonds between the components and the board.

The reflow temperatures required by Pb-free board assemblies are higher than those required by non-Pb-free boards, mainly because Pb-free solders generally have higher melting temperatures than Pb-Sn solders.  As such, the optimization of the reflow profile is more critical in Pb-free assemblies with regard to preventing the occurrence of package cracking in the surface mount components on the board.