Product safety and the heat sink - dilemma of minimizing radiated emissions and maximizing thermal cooling

Abstract

Microprocessors are designed with tens of thousands or millions of transistors that potentially can generate a significant amount of heat. Cooling of the package has become a challenge. Heat sinks are generally provided for the purpose of thermal dissipation, which helps keep components from becoming too hot. Destruction of the component may result. Designing heat sinks for optimal cooling within the constraints of size, space and airflow has become a challenge. In addition to being an efficient thermal radiator, a metal heat sink begins to appear as an efficient radiator of RF energy at higher operating frequencies. During normal operation or maintenance of a product, the heat sink may become exposed to users or service personnel. The temperature of the heat sink must be low enough as not to cause a burn injury if accidentally touched. Product safety standards require a label be affixed to the heat sink when a certain temperature level is exceeded, or if the heat sink is at voltage potential to prevent electric shock. In addition, the device must be kept cool enough as not to exceed the glass transition temperature (T/sub g/), or melting point of the printed wiring board material (PWB). Printed wiring boards, when exposed to high temperatures, may start to discolor, delaminate or even ignite causing serious safety consideration. To satisfy areas of concern related to heat sinks (heat dissipation and product safety while minimizing the propagation of RF energy), designers must maximize thermal cooling. To minimize the total amount of radiated emissions, the heat sink must be physically small based on the wavelength of the highest generated frequency internal to the component. Conversely, to maximize cooling, the heat sink must be physically large. The designer must be cognizant of all concerns, and select or design a heat sink that meets operational requirements.