Model PCB Thermals with Greater Accuracy

Feb 1, 2008 12:00 PM
By Byron Blackmore, Electronics Cooling Engineering Supervisor, Flomerics, Marlborough, Mass.

By dividing PCB layers into arrays of small cells and calculating the thermal conductivity of each cell, a thermal simulation tool generates a more accurate thermal profile and accounts for local concentrations of copper on the board.

Electronics companies at the leading edge of performance are being forced to address board-level thermal requirements at the earliest stages of design. Printed circuit boards (PCBs) constitute the primary area where mechanical engineers can influence the thermal design at the conceptual design phase. So, the ability to accurately predict the thermal performance of the PCB early in the development process is becoming more critical than ever before.

A key limitation of tools designed to simulate the performance of PCBs early in the design process has been their inability to take into account the effects on thermal conductivity of localized concentrations of copper. But a recent improvement in these design tools gives designers the ability to model the effects of copper concentration on thermal conductivity in board-level thermal simulations.

The latest generation of PCB design tools makes it possible to model copper concentration at whatever level of detail is desired, even to the point of modeling each individual trace. This approach has been demonstrated to substantially improve the accuracy of upfront PCB thermal simulation. In turn, this greater accuracy can help improve time to market, and reduce engineering and manufacturing costs.

Increasing Importance of Board Design

In the fairly recent past, board-level thermal simulation was not considered to be a critical part of the mechanical design flow. Power-dissipation levels were low, and as long as temperature and airflow guidelines were met, thermal issues were usually relatively easy to resolve.

In this scenario, the task of the mechanical engineer was simple: ensure that the chassis housed each PCB within sufficient airflow at the right temperature. Thermal management was usually addressed at the time the chassis was designed, normally by adding fans and cooling vents.

The thermal design was typically based on a system-level simulation. The chassis designed by this method normally had a relatively long shelf life, typically three to five years. Once the thermal solution was fixed, it was expected to last through several generations of boards. The mechanical and PCB design processes were largely decoupled, enabling mechanical and PCB designers to work independently.

As Moore's Law marched on, mechanical and PCB designers found themselves having to interact more frequently. This was driven by the fact that PCB power dissipation for many designs was crossing an important threshold beyond which thermal compliance came into question. This threshold is linked to the ability of the PCB itself to act as a heatsink.

As a large flat surface, the PCB is very effective in transferring heat from components to the air. However, like any heatsink, a limit exists. A common rule of thumb to describe the limit comes from a simple heat-transfer calculation, which is an expression of the heatsinking ability of a PCB.

Given that 100°C is a common maximum temperature of components, the power that a PCB can dissipate is estimated to be:

Q = h × SA × (T1 - T2),

where Q is power dissipation expressed in watts (W), h is the heat-transfer coefficient expressed as W/(°C in²), SA is the surface area expressed as in², T1 is the maximum component temperature in °C and T2 is the air temperature in °C.

The heat-transfer coefficient is largely dependant on the airflow speed, although there is no simple equation to describe the relationship between these two variables. The relationship between heat-transfer coefficient and airflow speed can vary depending on flow regime (laminar, transitional or turbulent) and the geometry of the heat-dissipating object (flat plate, cylinder, etc.). For a PCB sitting in airflow at 20°C and 200 linear ft/min, the maximum power dissipation is 1.8 W/in².

Simulation Process Flow

Electronic manufacturers are beginning to address these problems by paying more attention to thermal design at the board level. Often, when designing a new board for an existing enclosure, electrical engineers are simulating the board alone to identify hot spots. Problems identified at this stage often can be addressed by layout changes that can be made nearly without cost at this stage of the process. Board-level simulation tools are usually much easier for electronic engineers to use, because they are designed around tools they already use, such as functional block diagrams and physical layouts.

In a typical board-level thermal simulation process flow, the systems architect will develop the initial concept design by creating a functional block diagram. The hardware design engineer then drives the first physical layout directly from the block diagram.

At an early stage in the design process, long before the mechanical engineer gets involved, the electrical engineer can use board-level simulation to evaluate the new board design in an existing system. A 3-D computational fluid dynamics (CFD) solver predicts airflow and temperature for both sides of the board.

Often the designer will identify hot spots, and thus, cooling management can be considered from the earliest stages of the design process. Changes made to the functional block diagram are reflected instantly in the physical layout and thermal representations.