The emergence of power modules has liberated embedded engineers from the burdensome task of power design. However, with the variety of power modules available, how should we approach selection in our daily circuit designs?
In the increasingly fierce market competition, rapid product design and development have undoubtedly become necessary conditions for leading success and quickly seizing business opportunities. Under the “whip” of project managers and the increasingly shorter project cycles, modular development, platform-based development, and solution-oriented development models have been accepted and used by more and more system designers and hardware engineers.
Figure 1: Modular Design Product – Modular Mobile Phone
Every electronic product development cannot avoid power design. If we compare a product to a person, then the power supply is like a person’s heart; the health of the heart is critical to life and death. Similarly, the quality of power design can lightly affect the performance of the entire product and heavily impact the success or failure of the entire project. Power design is characterized by its high level of expertise, long debugging cycles, and difficulty in troubleshooting, which makes it particularly frustrating for engineers.
In this context, modular power supplies have emerged as a boon for engineers. Among them, DC/DC power modules, with their compact size, excellent performance, ease of use, and low overall cost, have found widespread application in communication, networking, industrial control, railways, and other fields.
So, among the numerous models, varying parameters, and many brands of power modules, how do we filter out the suitable and cost-effective DC/DC power modules? Most people are familiar with standard selection methods, but this time we will discuss several issues that embedded system designers often struggle with when selecting DC/DC power modules.
1. “Isolated” or “Non-Isolated”?
“Does this part of the circuit really need isolation?” This is a question that every embedded system engineer will consider.
From the purpose of isolation, it can be divided into safety isolation and noise isolation. Due to the wide application of embedded systems, hardware design often encounters complex situations with multiple voltage supplies, mixed signal types, and high-speed and low-speed signals on the same board. A slight mishandling can introduce interference, which can reduce product performance or even disrupt communication, leading to system restarts or failures. At this point, isolation becomes particularly important; in embedded system design, isolated power modules are generally used to provide power to different areas of the PCB, minimizing noise interference and improving system stability.
Figure 2: Zhiyuan Electronics 3W Wide Voltage Isolated Power Module
Moreover, in embedded systems containing industrial buses, there are often challenges posed by surges, arc interference, and lightning strikes, necessitating safety isolation between the bus and the embedded system. Isolation can eliminate ground loop interference and block external adverse environmental factors from entering the core system through the bus, ensuring the safety of the core system.
2. “Performance” vs “Cost”?
“Performance” and “cost” are a pair of intertwined adversaries, putting many engineers in a dilemma, forcing them to abandon many hard-earned solutions due to cost issues. Should we prioritize performance at the expense of cost? Or should we sacrifice performance for cost? Balancing performance and cost is an age-old topic in product design.
Figure 3: The Battle of Performance and Cost
For DC/DC power modules with the same input and output voltage, output power and operating temperature range are the main factors affecting cost. The working temperature range of electronic devices is generally divided into: commercial grade (0~70℃), industrial grade (-40~85℃), automotive grade (-40~105℃), and military grade (-55~125℃). Due to different temperature grades, the requirements for materials and manufacturing processes vary, leading to significant cost differences.
If, under certain conditions of size (packaging type), the actual power usage is close to the rated power of the module, then the nominal temperature range of the module must strictly meet actual usage requirements, even with some margin. If a product with a smaller temperature range is chosen due to cost considerations, and the actual operating temperature approaches the module’s maximum temperature limit, what should be done? In this case, derating can be employed, which means selecting a product with a higher power or larger package, thus allowing for lower temperature rise, which can alleviate this conflict to some extent.
In summary, one can either choose products with a wide temperature range, which utilize power more efficiently and have smaller packaging, but are more expensive; or choose products with a general temperature range, which are cheaper but require larger power margins and packaging. Ultimately, the choice depends on the actual situation and requires comprehensive consideration.
3. How much power margin to leave? 10%? 20%? 30%?
“Design margin” is a metric that is both loved and hated. The essence of margin design is to prevent unexpected events; while it may not relate to quality, insufficient margin design poses quality risks, whereas excessive margin design can lead to increased costs.
Given the wide application of embedded systems, loads can be diverse, some being resistive loads, others inductive or capacitive loads, some loads are stable, while others fluctuate greatly; some may even experience no load, full load, sudden increases, or sudden drops in load, making it challenging to determine the power rating of the power module.
Figure 4: Wide Application of Embedded Development
In general, the size of the load current is the key factor determining power. To ensure stability and resilience against unexpected events in embedded system design, it is recommended to reserve at least a 20% design margin based on actual conditions, ensuring that the maximum power usage does not exceed 80% of the rated power of the power module. Within this power range, the performance of the power module is more fully and reliably utilized. If the margin is too large, it results in resource waste; if too small, it adversely affects temperature rise and reliability.
For loads with significant fluctuations, the design should ensure that the peak current does not exceed the maximum tolerance of the power module, and based on the frequency of load fluctuations, the design margin should be appropriately increased to enhance reliability.
4. Is a higher isolation voltage better?
No, not necessarily!
Isolation voltage is an important indicator of isolated DC/DC power modules, generally categorized into specifications of 1000VDC, 1500VDC, 2000VDC, 3000VDC, and 6000VDC, indicating the maximum voltage that the power module can withstand between the input and output terminals within a certain time (usually 1 second). The higher the isolation voltage level, the higher the requirements for the protection devices and design processes within the power module, which fundamentally means higher costs. So how do we choose the appropriate isolation voltage?
The isolation voltage of a power module should be selected based on the application scenario. In general situations, the isolation voltage requirement for the power module is not very high, but a higher isolation voltage can ensure that the module has lower leakage current, higher safety and reliability, and better EMC characteristics. Additionally, as the core components of embedded systems are highly integrated and compactly packaged, they are relatively fragile, making a common industry standard for isolation voltage typically 1500VDC or above. However, in some special industries, such as medical, outdoor communication stations, and high-voltage power, the isolation requirements for power modules are even higher.
In the current era of rapid technological advancement, fast product development has become the norm. The powerful processing capabilities of embedded systems have lessened the burden of hardware design, delegating much of the work to software. Standardized hardware platforms, standard interfaces, and rich drivers have made it possible to quickly shape products for market release. Under increasingly tight project timelines, more and more embedded system design engineers have recognized that correctly and reasonably selecting DC/DC power modules can not only eliminate the hassles of power design and debugging but also improve the overall system reliability and design level, and, more importantly, shorten the entire product development cycle.
Reply with the keyword 【Power Supply】 to view more technical topics.
Public Account Introduction
Zhiyuan Electronics Official WeChat Account, a sharing platform for 500 engineers, provides you with leading product technology and solutions in the electronics industry..