DOC

# Anatomy of Switching Power Supplies

By Yvonne White,2014-12-02 21:21
8 views 0
Anatomy of Switching Power Supplies

Anatomy of Switching Power Supplies

Author: Armando Mtz. R

Type: Tutorials Last Updated: junio, 2009

ITNL

Introduction

Power supplies used on the PC are based on a technology called “switching mode” and thus are also known as SMPS, Switching Mode Power Supplies (DC-DC converter is another nickname for switching mode power supplies). In this tutorial we will explain you how switching power supplies work and we will provide a journey into the PC power supply showing you its main components and what they do.

We have already published a Power Supply Tutorial, where we dealt with

form factors, how to calculate the power supply nominal power rating and also explained the basic power supply specs. In the present tutorial we go a step further, explaining what is inside the box, what are the power supply

main components, how to identify them and what they do.

There are two basic power supply designs: linear and switching.

Linear power supplies work by getting the 127 V or 220 V from the power grid and lowering it to a lower value (e.g. 12 V) using a transformer. This

lower voltage is still AC. Then rectification is done by a set of diodes, transforming this AC voltage into pulsating voltage (number 3 on Figures 1 and 2). The next step is filtering, which is done by an electrolytic capacitor, transforming this pulsating voltage into almost DC (number 4 on Figures 1 and 2). The DC obtained after the capacitor oscillates a little bit (this oscillation is called ripple), so a voltage regulating stage is necessary, done by a zener diode or by a voltage regulator integrated circuit. After this stage the output is true DC voltage (number 5 on Figures 1 and 2).

click to enlarge

Figure 1: Block diagram for a standard linear power supply design.

click to enlarge

Figure 2: Waveforms found on a linear power supply.

Although linear power supplies work very well for several low-power

applications cordless phones and video games consoles are two

applications that come in mind , when high power is needed, linear power

supplies can be literally very big for the task.

The size of the transformer and the capacitance (and thus the size) of the electrolytic capacitor are inversely proportional to the frequency of the input

AC voltage: the lower the AC voltage frequency, the bigger the size of those components and vice-versa. Since linear power supplies still use the 60 Hz (or 50 Hz, depending on the country) frequency from the power grid which

is a very low frequency , the transformer and the capacitor are very big.

Also, the higher the current (i.e. the power) demanded by the circuit fed by the power supply, the bigger the transformer is.

Building a linear power supply for the PC would be insane, since it would be very big and very heavy. The solution was to use the high-frequency

switching approach.

On high-frequency switching power supplies, the input voltage has its frequency increased before going into the transformer (50-60 KHz are

typical values). With input voltage frequency increased, the transformer and

the electrolytic capacitor can be very small. This is the kind of power supply used on the PC and several other electronic equipments, like VCRs. Keep in mind that “switching” is a short for “high-frequency switching”, having

nothing to do whether the power supply has an on/off switch or not…

The power supply used on the PC uses an even better approach: it is a closed loop system. The circuit that controls the switching transistor gets feedback from the power supply outputs, increasing or decreasing the duty cycle of the voltage applied to the transformer according to the PC consumption (this approach is called PWM, Pulse Width Modulation). So the power supply readjusts itself depending on the consumption of the device

connected to it. When your PC isn’t consuming a lot of power, the power supply readjusts itself to deliver less current, making the transformer and all

other components to dissipate less power i.e. less heat is generated.

On linear power supplies, the power supply is set to deliver its maximum power, even if the circuit that is connected to it isn’t pulling a lot of current. The result is that all components are working at their full capacity, even if it isn’t necessary. The result is the generation of a greater heat.

Switching Power Supply Diagram

On Figures 3 and 4 you can see the block diagram of a switching power supply with PWM feedback used on PCs. On Figure 3 we show the block

diagram of a power supply without PFC (Power Factor Correction) circuit

used by cheap power supplies and on Figure 4 we show the block diagram

of a power supply with active PFC circuit, which is used by high-end power

supplies.

click to enlarge

Figure 3: Block diagram for a switching power supply design with PWM (no

PFC).

click to enlarge

Figure 4: Block diagram for a switching power supply design with PWM and

active PFC.

You can see what is the difference between a power supply with active PFC and one without this circuit by comparing Figures 3 and 4. As you can see, power supplies with active PFC don’t have a 110/220 V switch and also don’t

have a voltage doubler circuit, but of course they have the active PFC that we will talk more about later.

This is a very basic diagram. We didn’t include extra circuits like short-circuit

protection, stand-by circuit, power good signal generator, etc to make the

diagram simpler to understand. If you want detailed schematics, see Figure 5. If you don’t understand electronics, don’t worry. This figure is just here for the readers that want to go deeper.

click to enlarge

Figure 5: Schematics for a typical low-end ATX power supply.

You may be asking yourself where is the voltage regulation stage on the

Figures above. The PWM circuit does the voltage regulation. The input voltage is rectified before passing the switching transistors, and what they send to the transformer is square wave. So what we have on the transformer output is a square waveform, not a sine waveform. Since the waveform is already square, it is very simply to transform it into a DC voltage. So after the rectification after the transformer, the voltage is already DC. That is why some times switching power supplies are also

referred as DC-DC converters.

The loopback used to feed the PWM control circuit is in charge of making all the necessary regulation. If the output voltage is wrong, the PWM control circuit changes the duty cycle of the signal applied to the transistors in order

to correct the output. This happens when the PC power consumption increases, situation where the output voltage tends to drop, or when the PC power consumption decreases, situation where the output voltage tends to increase.

All you need to know before moving to the next page (and that you can learn from paying attention to Figures 3 and 4):

; Everything before the transformer is called “primary” and everything

after it is called “secondary”.

; Power supplies with active PFC circuit don’t have a 110 V/ 220 V

switch. They also don’t have a voltage doubler.

; On power supplies without PFC, if the 110 V / 220 V is set to 110 V,

the power supply will use a voltage doubler, in order to make the

voltage always around 220 V before the rectification bridge.

; On PC power supplies two power MOSFET transistors make the

switcher. Several different configurations can be used and we will talk

; The waveform applied to the transformer is square. Thus the

waveform found on the transformer output is square, not sine.

; The PWM control circuit which is usually an integrated circuit is

isolated from the primary thru a small transformer. Sometimes

instead of a transformer an optocoupler (a small integrated circuit

containing a LED and a phototransistor packed together) is used.

; As we mentioned, the PWM control circuit uses the power supply

outputs to control how it will drive the switching transistors. If the

output voltage is wrong, the PWM control circuit changes the

waveform applied on the switching transistors in order to correct the

output.

; On the next pages we are going to explore each one of these stages

with pictures showing where you can find them inside a power

supply.

Inside a PC Power Supply

After opening a power supply for the very first time (don’t do this with its power cord attached or you will get an electrical shock), you may find yourself quite lost trying to figure out what is what. But you will recognize at

least two things you already know: the power supply fan and some heatsinks.

click to enlarge

Figure 6: Inside a PC power supply.

But you should able to recognize very easily the components that belong to the primary and the components that belong to the secondary.

You will find one (on power supplies with a active PFC) or two (on power

supplies without PFC) big electrolytic capacitors. Find them and you will find the primary.

Usually PC power supplies have three transformers between two big heatsinks, as you can see on Figure 7. The main transformer is the biggest

one. The medium transformer is used to generate the +5VSB output and the smallest transformer is used by the PWM control circuit to isolate the secondary from the primary (this is the transformer labeled as “isolator” on Figures 3 and 4). Several power supplies instead of using a transformer as an isolator uses one or more optocouplers (they look like small integrated circuits), so on power supplies using these components you will probably find only two transformers. We will talk more about this later.

One of the heatsinks belongs to the primary and the other belongs to the secondary.

On the primary heatsink you will find the switching transistors and also the

PFC transistors and diode, if your power supply has active PFC. Some manufacturers may choose to use a separated heatsink for the active PFC components, so on power supplies with active PFC you may find two heatsinks on its primary.

On the secondary heatsink you will find several rectifiers. They look like transistors but they have two power diodes inside.

You will also find several smaller electrolytic capacitors and coils that belong to the filtering phase finding them you will find the secondary.

An easier way to find the secondary and the primary is just following the power supply wires. The output wires will be connected to the secondary while the input wires (the ones coming from the power cord) will be connected to the primary. See Figure 7.

click to enlarge

Figure 7: Locating the primary and the secondary.

Now let’s talk about the components found on each stage of the power supply.

Transient Filtering

The first stage of a PC power supply is the transient filtering. On Figure 8 you can see the schematics of the recommended transient filter for the PC power supply.

click to enlarge

Figure 8: Transient filter.

We say “recommended” because many power supplies specially the cheap

ones won’t have all the components shown on Figure 8. So a good way to

check whether your power supply is a good one or not is by checking if its transient filtering stage has all recommended components or not.

Its main component is called MOV (Metal Oxide Varistor) or varistor, labeled RV1 on our schematics, which is responsible for cutting voltage spikes (transients) found on the power line. This is the exact same component found on surge suppressors. The problem, though, is that cheap power

supplies don’t carry this component in order to save costs. On power supplies with a MOV, surge suppressors are useless, since they have already a surge suppressor inside them.

L1 and L2 are ferrite coils. C1 and C2 are disc capacitors, normally blue.

These capacitors are also called “Y capacitors”. C3 is a metalized polyester capacitor, normally with values like 100 nF, 470 nF or 680 nF. This capacitor is also called “X capacitor”. Some power supplies have a second X capacitor, installed in parallel with the main power line, where RV1 is on Figure 8.

X capacitor is any capacitor that has its terminals connected in parallel to the main power line. Y capacitors come in pairs, they need to be connected together in serial with the connection point between them grounded, i.e.

connected to the power supply chassis. Then they are connected in parallel to the main power line.

The transient filter not only filters the transients coming from the power line, but also prevents the noise generated by the switching transistors to go

back to the power line, which would cause interference on other electronic equipments.

Let’s see some real-world examples. Pay attention to Figure 9. Do you see something strange here? This power supply simply doesn’t have a transient

filter! This power supply is a cheap “generic” unit. If you pay attention you can see the markings on the power supply printed circuit board where the filtering components should be installed.

click to enlarge

Figure 9: This cheap “generic” power supply doesn’t even have a transient

filtering stage.

On Figure 10 you can see the transient filtering of a cheap power supply. As you can see, the MOV is missing and this power supply has only one coil (L2 is missing). On the other hand it has one extra X capacitor (placed where

RV1 is on Figure 8).

click to enlarge

Figure 10: Transient filtering on a cheap power supply.

On some power supplies the transient filter can be broke down into two separated stages, one soldered to the input power connector and the other on the power supply printed circuit board, as you can see on the power supply shown on Figures 11 and 12.

On this power supply you can find a X capacitor (replacing RV1 on Figure 8) and the first ferrite coil (L1) soldered on a small printed circuit board that is connected to the main AC power connector.

click to enlarge

Figure 11: Transient filter first stage.

On the power supply printed circuit board you can find the other components. As you can see this power supply has a MOV, even though it is placed on an unusual position, after the second coil. If you pay attention, this power supply has more than the recommended number of components, as it has all components shown on Figure 8 plus an extra X capacitor.

click to enlarge

Figure 12: Transient filter second stage.

This power supply MOV is yellow, however the most common color is dark blue.

You should also find a fuse near the transient filter (F1 on Figure 8, see also Figures 9, 10 and 12). If this fuse is blown, beware. Fuses don’t blow by themselves and a blown fuse usually indicates that one or more components

are defective. If you replace the fuse, the new one will probably blow right after you turn on your PC.

Voltage Doubler and Primary Rectifier

On power supplies without active PCF circuit you will find a voltage doubler. The voltage doubler uses two big electrolytic capacitors. So the bigger capacitors found on the power supply belongs to this stage. Like we mentioned before, the voltage doubler is only used if you are connecting

your power supply to a 127 V power grid.

click to enlarge

Figure 13: Electrolytic capacitors from the voltage doubler.

click to enlarge

Figure 14: Electrolytic capacitors from the voltage doubler removed from

the power supply.

Next to the two electrolytic capacitors you will find a rectifying bridge. This bridge can be made by four diodes or by a single component, see Figure 15.

On high-performance power supplies this rectifying bridge is connected to a heatsink.

click to enlarge

Figure 15: Rectifying bridge.

On the primary you will also find a NTC thermistor, which is a resistor that changes its resistance according to the temperature. It is used to reconfigure the power supply after it is used for a while and it is hot. NTC stands for Negative Temperature Coefficient. This component resembles a ceramic disc capacitor and is usually olive green.

Active PFC

Obviously this circuit is found only on power supplies that have active PFC. On Figure 16 you can study the typical active PFC circuit.

Report this document

For any questions or suggestions please email
cust-service@docsford.com