How Improved Packaging and Control Technologies Advance the Case for DC-DC ‘Micropower’ Modules

Power system designers commonly face a decision early in a new development over whether to realize the main DC-DC conversion stage as a discrete circuit design with multiple active and passive components, or to use an integrated module, thus implementing the same function with a single part.

Traditionally, the pay-off and drawbacks have appeared simple: the module simplifies the board layout and occupies a smaller footprint. The discrete design, however, will have a lower bill-of-materials cost, and gives the designer greater flexibility to tailor the circuit’s performance to the needs of the application.

Recent advances in packaging and controller technology, however, have somewhat shifted the argument: in industrial and automotive applications (with an input ≤70V) in particular, the integrated configuration of a module offers advantages in terms of emissions and efficiency that are beyond the reach of any discrete design.

The question, therefore, has now become whether the flexibility and cost advantages of a discrete design on their own are sufficient to outweigh the superior performance available from a module.

Basic operation of a ‘micropower’ module
The micropower module, a device type pioneered by Linear Technology, steps down a higher DC voltage to a low-voltage output, feeding a load switch to power 5V or 3.3V circuitry or a multiple-output PMIC (power management IC). A typical application circuit is shown in Figure 1.

Fig. 1: in a micropower module’s typical application circuit, a 12V, 24V, 36V or 48V input must be stepped down to a level suitable for logic devices

The micropower module integrates all the components necessary to realize this converter circuit (see Figure 2). This dramatically simplifies the system’s circuit design and board layout compared to the equivalent discrete design. It also reduces component count, which tends to increase system reliability, and reduces the size of the board footprint occupied by the power system. An important secondary benefit of using a micropower module is that the behavior of the converter circuit is precisely characterized by the module’s manufacturer and is comprehensively documented in the module’s datasheet.

Fig. 2: a micropower module from Micrel, the MIC28304, integrates a DC-DC controller, MOSFETs, inductors, capacitors and resistors (click here for larger image)

In the past, the benefits of micropower modules have been available only to designers working within a narrow operating voltage range – typically with inputs up to 24V. This might be adequate for applications with a well regulated power bus, but industrial and automotive applications require a more rugged power-handling capability. In the presence of electrically noisy systems such as electric motors, a nominal 36V or 48V bus might be subject to power spikes producing transient voltages of up to 70V. Industrial and automotive products’ loads also typically peak higher than those in, for instance, telecoms, networking or computing equipment.

To encapsulate a complete, high-voltage circuit in a package, however, creates a problem: heat. This has in the past limited the range of applications for which a micropower module is suitable, but packaging and control innovations have opened up new possibilities for designers of industrial and automotive systems to benefit from a module’s integration and performance.

Pathways for heat dissipation
Earlier micropower modules for low-power applications could be safely accommodated on conventional PCB or ball grid array (BGA) substrates without breaching their operating temperature limits. The higher-power requirements of industrial and automotive applications, however, call for a more effective means of dissipating heat from the package.

In response, Micrel has implemented a new copper lead-frame structure for its high-voltage MIC28304, the first time this packaging style has been used in a micropower module. Copper combines high thermal conductivity with low on-resistance, thus helping to accelerate the flow of heat out of the package while reducing the amount of heat generated by resistive losses.

In the MIC28304, this copper lead-frame structure has enabled Micrel to produce a complete DC-DC converter circuit with a 3A maximum current capability, a broad input range of 4.5-70V, and output voltages adjustable down to 0.8V with a guaranteed accuracy of ±1%. Housed in a miniature 12mm x 12mm x 3mm 64-pin package, it is rated for operation across a junction temperature range of -40°C to +125°C, and normally requires no additional heat sink or other cooling device. The device’s 70V peak voltage capability provides ample headroom for transient voltages, even in high-voltage applications such as automotive 48V systems.

Performance advantages of integrated design
The copper lead-frame structure, then allows Micrel to pack multiple high-voltage components extremely tightly inside a single package, without suffering from thermal overload. This high-density design, obviously, keeps the device’s footprint to a minimum.

It also offers another benefit: the traces connecting one component to another are extremely short, which means they produce almost no electro-magnetic radiation. Compliance with electro-magnetic compatibility (EMC) regulations is a common bugbear for the OEM designer, and can sometimes lead to costly delays in development projects as designers are forced to reduce emissions, or add shielding to radiating components, at a late stage in a product’s design.

The risk of EMC problems is acute in designs using discrete power components, since long traces connecting switching devices act as antennas. The MIC28304 shows the advantage of a modular architecture for designers with responsibility for EMC compliance, generating emissions at levels that pass tests such as CISPR22 Class B (EN55022 Class B in Europe) by a wide margin (see Figure 3).

Fig. 3: radiated emissions from the MIC28304 fall far inside the limits specified by global EMC standards such as CISPR22 Class B

In addition, the MIC28304 provides the designer with the freedom to configure the switching frequency to any value between 200kHz and 600kHz. The designer can therefore select the frequency that can be most easily filtered, or that avoids a sensitive frequency. In automotive systems, for instance, power system designers are commonly required to avoid AM band radio frequencies.

A module also gives the system designer an optimized implementation of the features inherent in the converter’s controller. In the MIC28304, for instance, the controller includes Micrel’s Hyper Speed™ Control technology, which provides for very rapid response to transients and changes in load. Figure 4 shows the minuscule voltage ripple produced under extreme load stress, jumping from a 0A to a 10A output. The fast and stable response of the module makes it ideal for use in end products with rapidly varying loads.

Fig. 4: the duration of the voltage ripple produced by the MIC28304 under a 1-3A load swing can be measured in nanoseconds.

The controller also implements HyperLight Load® technology. Switching converters typically reach peak efficiency at full load, but are markedly less efficient when operating at less than 50% of peak load capability. HyperLight Load is a Micrel technology for improving converter efficiency at low loads, automatically detecting low-load conditions and initiating discontinuous-mode operation (in which the inductor current is allowed to fall to 0A). In discontinuous mode, the bias current of most of the controller’s circuits is reduced, and the normal operation of the PWM switching circuit is suspended. This dramatically reduces the current drawn by the module’s internal circuitry, and enables it to achieve high efficiency in light-load applications.

At full load, the MIC28304 is helped by the relatively low switching frequencies at which it is designed to operate. In general in switch-mode converters, there is a trade-off between size and efficiency: the faster the switching frequency, the smaller the inductor required to support the controller. But since the higher frequency produces more switching operations, the switching losses are correspondingly higher, thus lowering efficiency.

In the MIC28304, both the board footprint and profile are already small, because of the module’s optimized internal layout and packaging; the extra space saving offered by fractionally reducing the size of the inductor is therefore negligible. Instead, Micrel has optimized for efficiency, configuring the module’s design so that it operates at a modest frequency to moderate switching losses, leading to high efficiency of more than 90% across a broad output range of <0.5A to >3A (see Figure 5).

Fig. 5: the MIC28304 reaches its peak efficiency across a broad output range when operating at 275kHz

More benefits than size and simplicity
Going into a development, the benefits of size and simplicity are often uppermost in the thoughts of the power system designer when considering a micropower module for converter outputs of 5V, 3.3V or lower.

The discrete alternative offers cost and flexibility advantages over the micropower module. As this article has shown, however, the module provides performance benefits that are simply not available in a discrete design, in particular a low level of radiated emissions. In deciding between a module and a discrete solution, then, performance as well as size and simplicity should be brought into consideration in order to make the best choice.

About the Author

Mr. Mendoza has 14 years high tech experience in the semiconductor industry. He is currently based in Micrel headquarters in San Jose, where he is responsible for Automotive and Industrial Product Marketing of Micrel, Power Management, MOSFET Drives and LED Driver Solutions.

source:http://www.powerpulse.net/powerViews.php?pv_id=81&page=1

Comments are closed.