Power supplies are the part of all products, mains or battery powered, that enable and ensure proper operation in all areas of the circuit. In addition to providing operating power, some supplies can detect and in some cases correct for anomalies within complex circuits.
For quite some time now, power has been a big concern in the tech markets. With numerous companies trying to go “green,” a.k.a., environmentally friendly, as well as regulatory mandates hoping to reduce power consumption by setting efficiency standards, everyone wants to save power. Even consumers are enticed with the mirage of lower utility bills and are looking for products and devices that may possibly trim their budgets.
The power OEMs aren’t the only organizations that are focusing on curbing energy consumption. The tech community as a whole is in on the deal too. After all, you can create the most efficient, greener-than-green power supply that passes all current mandates and regulations, plus a few that no one has thought of yet. But if it’s latched to a batch of poorly designed, inefficient circuits, your efforts will be for naught.
Breaking Them Down
There are a few popular power-supply types and three viable techniques for achieving performance and efficiency nirvana. These include linear-regulated, switched-mode, and programmable supplies and power-factor correction (PFC), efficiency, and digital control. Let’s take a quick cruise through some basics and then get to where we stand now.
In most cases, people assume the power supply converts a somewhat high-level ac voltage to an equal or lower dc voltage to power a particular circuit or device. Of course, this isn’t completely accurate since some components also called power supplies convert ac voltage to ac voltage and dc to dc. For our purposes, let’s stick to a power supply being the aforementioned unit, an ac-dc converter. We’ll refer to everything else, i.e., dc-dc converters and other devices, as a power source.
Regardless of the design, be it complex, esoteric, or exotic, the front end of most supplies employs the same technique as it has since day one. The ac mains from the wall enters into a step-down or step-up transformer, which then feeds a rectifier and filtering circuit.
Depending on the application requiring dc power, the power supply can be fairly simple or complex. For experiments, hobbies, and powering up a few dc devices, a basic unipolar power supply (Fig. 1) may be sufficient.
Tried and tested over time and minimal in part counts, the basic power supply is still the backbone of many state-of-the-art components. It consists merely of a step-down ac transformer, two rectifiers, and filter capacitor.
The output voltage and current capability are determined purely by the transformer’s ratings. The rectifier diodes and filter capacitor are chosen based on these maximum ratings.
Continue on next page
For applications requiring bipolar power (Fig. 2), the same supply can be easily converted by adding two more diodes and another filter cap, creating what some may call a floating ground. A slightly better, more stable, and quieter approach would be changing the transformer (Fig. 3) to one with a center tap on the secondary, which provides the circuit’s ground.
Again, for minor tasks, basic supplies will do. But for complex digital designs, the drawbacks of these simple topologies are obvious as well as intolerable.
First, the output voltage is usually an approximation of what is actually required, i.e., 13.4 V for a required 12 V. Second, there’s no compensation for voltage and current drifts due to temperature and ac-line conditions. And third, the only way to get ripple down is to add bigger and/or more filter capacitors.
Other considerations include the need for multiple outputs, user safety and circuit protections, thermal conditions, PFC, control, and meeting agency requirements for critical factors such as efficiency.
Beyond the Basics
Going one step above basic supplies, a linear regulated power supply locks the output voltage to a precise value while also buffering the output against fluctuations of input voltage and the load. The regulator also reduces ripple and noise.
In some cases, linear regulators limit current to guard the supply from overcurrent. Another feature of linear regulated supplies is adjustable outputs, making them viable for lab and service-shop applications.
Moving to the next level, switched-mode power supplies (SMPS) operate without a large transformer. They accept an ac input voltage that gets switched on and off at high speed (approximately 50 kHz to 1 MHz) via the supply’s switching circuitry.
Input to an SMPS first passes through a rectifier, converting the ac to rough, unregulated dc. This dc then enters an inverter (chopper) circuit that converts it back to ac via a power oscillator.
Via a small output transformer operating at a frequency chosen above 20 kHz, this chopper stage couples to an output filter and rectifier where the ac is again converted to dc for the final output. For regulation, part of the output is fed back to the chopper stage through a controller circuit.
Providing numerous advantages over other types of power supplies, the SMPS finds employment in the widest range of applications, from computers to medical equipment. These advantages include high efficiency, smaller and lighter units, and lower heat levels.
Of course, these benefits come at the expense of greater circuit complexity and the need to address electromagnetic interference (EMI) and harmonic distortion via low-pass filtering and PFC, respectively.
Suitable for more specialized applications in automatic test equipment (ATE), industrial, and medical arenas, programmable power supplies enable users to remotely control output voltage, current, and control frequency via interfaces such as GPIB, USB, and RS-232.
Programmable supplies usually employ microprocessors, programming circuits, current shunts, and a variety of read-back circuits. They also pack a plethora of protection circuits, i.e., overcurrent, overvoltage, short circuit, and temperature compensation, and can output dc, ac (single or dual phase), and/or ac + dc offset.
Continue on next page
Techniques For Better Supply Performance
Regardless of the type, the power supply functions as the heart of any electronic device or system. In addition to providing and pumping operating power, the blood of the system, a power supply can also provide circuit protection for both itself and the circuits it powers; isolate itself, the system, and the user from the power mains; and enable higher efficiency in overall operation. Naturally, many of these functionalities are the result of meeting certain agency mandates.
Currently, the major concerns on power-supply designers’ plates are thermal issues, efficiency, and control. Yes, size is always a challenge. Someday in the not too distant future, some clever designer will unveil a 60-GW SMPS in suppository form. For now, let’s look at the techniques for achieving better efficiency and meeting mandates.
There seems to be some debate about the necessity of PFC in consumer electronics and other applications. One side claims that PFC circuits do not boost efficiency and cause added heat generation in the power supply, while the other swears PFC is one of the saviors of the environment in terms of power reduction and higher efficiency. While the debates continue, regulatory agencies such as EnergyStar and the European Power Supply Manufacturers Association require power-supply makers to integrate PFC into all their offerings.
Taken as the phase difference between sinusoidal voltage and current waveforms, power factor is accepted as being a ratio between real power into a load and apparent power. It’s expressed as a decimal fraction between 0 and 1 and/or a percentage derived from there.
In theory, power systems specifying a low power factor draw more current than those with a high power factor. Higher currents translate into greater energy losses, which translate into higher energy costs. Simply put, PFC boosts a power supply’s power-factor figure.
PFC circuits come in two types, which are self explanatory: active and passive. Employed in an SMPS, the active or passive correction circuit lowers the ac current’s RMS value, which in turn improves power factor while protecting against overcurrent.
In terms of effectiveness versus cost, active PFC costs more. But it can garner power-factor figures in the upper 90% range while passive versions cost less and enable power-factor values from about 80% to approximately 86%.
According to Texas Instruments, power-supply (SMPS) efficiency is calculated simply by dividing the output power by the input power. To find efficiency by percentage, use:
Efficiency (%) = (VOUT · IOUT)/(VIN · IIN) · 100
Just about everyone in the business will agree that high efficiency is a good thing for all concerned. But what exactly does that efficiency figure actually mean?
“When comparing a 75% to a 90% efficient power supply, the savings in electricity usage and wasted energy, in the form of heat, is quite significant,” says David Norton, vice president of marketing for TDK-Lambda Americas.
Continue on next page
For consumers, higher efficiency, for the moment, translates into lower electricity bills. The same advantage is significantly greater for datacenters and servers where electricity consumption is escalating rapidly. And for those going the “green” route, such as medical equipment makers and many others, higher efficiency is a welcomed plus.
Regulatory agencies like EnergyStar and 80plus are keeping their eyes on the ball. For example, TDK Lambda’s DT100-C and DT150-C series external ac-dc power supplies (Fig. 4) comply with tough energy requirements. These include EISA, CEC, and Energy Star EPS Version 2.0, Efficiency Level V. Offering models rated from 100 W to 150 W, respectively, both series feature active PFC meeting EN61000-3-2 and operate from a universal input range from 90 to 264 V ac (47 to 63 Hz).
Power supplies, like any electronic device or circuit, employ controls to keep them in line with the application they drive. Various functions such as voltage and current control and pulse-width modulation were and in some cases still are under the thumb of analog control schemes and devices.
However, since designers and OEMs are always looking for ways to trim costly and space-consuming part counts, there’s digital control. Digital control brings several advantages to the table compared to analog techniques, such as immunity to component-induced variations, the ability to perform complex control algorithms, self calibrations, and faster performance.
For SMPS, digital control is well within the grasp of designers via highly integrated and economical microcontrollers, microprocessors, and digital signal processors. Also available are digital controllers such as the Texas Instruments TMS320LF2407. According to some designers, the goal of good digital control in power supplies is to achieve a dynamic performance comparable to analog controllers.
Most power supplies employing digital control fall under the category of programmable power supplies, but that is slowly changing. TDK’s EFE series (Fig. 5) embedded front-end power supplies rely on an eight-bit MCU for full digital control of the output and to handle housekeeping routines. According to the maker, this reduces part counts by 25% and achieves a 45% smaller and 56% lighter design than comparable units.
The 300-W EFE-300 and the 400-W EFE-400 single-output supplies deliver a 133% peak power capability for 10 s and efficiencies up to 90%. With 1U profiles, the EFE-300 measures 3 by 5 by 1.34 in., and the EFE-400 logs in at 3 by 6 by 1.34 in. The EFE-300 delivers 300-W continuous power and 400-W peak for 10 s with nominal outputs of 12 V/25 A or 24 V/12.5 A. The EFE-400 outputs 400 W continuous and 530 W peak for 10 s with nominal outputs of 12 V/33.3 A or 24 V/16.7 A.