Efficient power supply for critical loads with minimal PCB space

In distributed power systems where multiple DC converters are used, the challenge for designers is to keep these components as small as possible to save space on the board.


Big data servers and applications such as machine learning, artificial intelligence (AI), 5G cells, Internet of Things (IoT), and enterprise computing often require high-performance specialized integrated circuits (ASICs), directly programmable gate arrays (FPGAs), graphics cards, and processors that need high currents at low voltages and high power density in compact housings. To ensure the integrity of the power supply of the entire system, distributed power management systems are used that bring DC power sources directly to the load point (POL), i.e. high-performance processors. There can be many such DC converters on one board, so the problem that designers face is to keep these devices as small as possible to save space on the board. At the same time, they must meet performance, latency, temperature, efficiency and reliability requirements while simplifying the design process and keeping costs low.

The solution to these problems is to combine high-performance semiconductor and passive components with advanced housing technologies to achieve a higher level of system integration. It has been shown that compared to other technologies available today, this allows for a smaller footprint with a lower profile and at the same time improved heat dissipation. At the same time, an integrated approach reduces design costs, including inventory management, and development time.

The article discusses the need to create distributed power networks and the role of POL power supply devices. Next, a class of DC Point Load (POL) Converters from TDK Corporation that uses advanced housing technology to achieve the required performance characteristics is presented. The article also discusses their essential features and shows how designers can apply them to successfully meet load point power (POL) requirements.

Why use point-of-load DC converters (POL) as power sources

Computers, servers, and other digital devices increasingly make use of directly programmable gate arrays (FPGAs), specialized integrated circuits (ASICs), and other advanced integrated circuits that require multiple supply voltages that are not available from system power. Additionally, they require that the tensions be listed in the correct, ordered sequence with minimal latency. System power supplies generally provide several DC voltages such as 1, 3.3, and 5V. A typical directly programmable gate array (FPGA) requires voltages in the range 1.2 to 2.5V (Figure 1).

Figure 1: A typical Direct Programmable Gate Array (FPGA) requires different voltages for specific functions in the processor. The presented processor uses eight dedicated power inputs using three different voltages. (Image credit: Art Pini)

At a minimum, a Direct Programmable Gate Array (FPGA) requires separate power for the core section and the I / O section. The directly programmable gate array (FPGA) has a 1.2V core section and a 2.5V input-output section. In addition, it requires six other power levels for the auxiliary circuits. It is obvious that having seven power sources in close proximity to the gate arrays puts a strain on the PCB design. Heat dissipation must also be taken into account, which makes power sources small and efficient.

The full article You can find here: https://tek.info.pl

The article presents the mechanism of formation of the pad cratering defect leading to detachment, explains the differences between pad cratering and delamination, and presents methods of preventing this unfavorable phenomenon.
Companies providing EMS services - assembling electronics, PCB manufacturers or producing cable harnesses, fulfil the orders of their clients. Then, the selection of components and assembly method...
The quality of cable harnesses not only plays a key role in the process of device production but also is pivotal for the reputation of the producer’s brand. That’s why great importance is attached to make sure that the companies which produce them are specialised.
Sensitive electronic components are exposed to numerous risks: excessive moisture, electrostatic discharge, and UV radiation. To increase the durability of components, manufacturers use ...