Power supply units (PSUs) are never as up-to-date as CPUs, but they are technologically very detailed and attractive.

In what follows , we look at the internal structure of the PSU. An examination of the internal components and how they work reveals some interesting points about the lesser known part of the processing world and its tasks.

Name parts

Many computer components have names that require technical interpretation and understanding (such as a solid-state drive or SSD ) to understand their operations. As for the power supply, such an interpretation is incorrect and upon hearing its name, we know that we are facing a unit tasked with supplying electricity.

To better examine the internal anatomy of the power supply, there is nothing better than analyzing an existing sample and its components. For this, a Cooler Master G650M power supply was investigated that had a fairly normal design. The specification of the source is not the same as in most ordinary products of the market body. Of course, the PSU Coolmaster has a special ability that sets it apart from its competitors.

The power supply for this article uses the ATX 12V v2..31 form factor.

As a result, it falls into the category of ordinary products and is applicable to most PCs today.

Power supplies form a variety of factors, including smaller dimensions or specific designs for various equipment. On the one hand, not all products meet the industry standard. For example, you might find a power supply of the same length and width as another, but of a different thickness, to a similar standard.

Naming and labeling of power supplies is generally done with a view to maximizing their power supply. For example, the case study in this article is capable of supplying a maximum of 4 watts. Here’s a look at the meaning of this maximum power supply. Lower power supplies are also on the market, and you don’t necessarily need to buy high power supplies for all applications. Conventional parts in the market have the power supply of between 2 and 2 watts.

Large, professional power supplies are generally housed in metal boxes that make them heavy

Power supplies, such as the Cooler Master sample in the present article, are generally housed in metal boxes that come in a variety of colorless or black metal designs. The same metal box also increases the weight of the PSU. Laptops almost always have a power supply that sits externally next to the machine and is often made of plastic. Of course, the internal structure of the laptop’s power supply is not much different from the one mentioned above.

Most power supplies on desktop PCs come with a mains key and a fan to cool the components. Of course, some of them may be available without these parts. In addition, not all of them are supplied in metal boxes with multiple holes.

Metal boxes are rarely used to produce samples on servers.

After opening the metal box of the power supply, we come across some interesting and interesting parts that we will explain below.

Why do we need a power supply?

Before examining the internal components of the power supply, it is more important to examine the reasons for its use in computers. Have you ever asked yourself why we don’t connect the computer directly to the mains? In fact, the needs of modern PCs for power are different from the kind of power provided by urban outlets.

The diagram below shows the conventional power supply of the main grid. The X vector represents the time in milliseconds and the Y vector represents the voltage in units of volts. To define the voltage, just imagine the energy difference between the two points.

If the voltage is applied to a conductive material (such as a metal wire), the difference in the energy level causes the electrons in the material to move from the higher surface to the lower surface. As you know, electrons are a constituent of the atom, and metals also have a large number of free electrons. The current of the electron is known as the current intensity in the electricity concept and is measured in units of amps.

To better define and define the flow of electricity it can be likened to the flow of water in a hose.

Voltage corresponds to the pressure applied to the hose and the water flow rate to the intensity of the current. Any obstacle along the hose path can also be matched with electrical resistance.

The diagram above shows that the mains power fluctuates over time, which is why it is called AC power. The frequency of urban electricity flows in different regions of the world is different. In some logic, the current fluctuates 2 times per second and reaches a maximum of 1 or 2 volts, while in other regions we see a current of 1 or 2 volts.

The need for a power supply in personal computers stems from the alternate nature of urban electricity. In fact, computers need constant voltage that never changes during operation. In addition, the current intensity should be much lower.

In fact, the power flow diagram for the computer is as shown below.

The need for a PC to power is not a constant voltage. Of course, the difference in the flow they require is so low that it is hard to see. The PC requires voltages of +1 and +1, as well as smaller currents of +1 and +1.5. Since these numbers are constant, the new current is called direct current or DC. Finally, one of the main tasks of the power supply is seen in converting AC current to DC.

After first understanding the reason for using a power supply on a PC, it is time to examine the internal components. After you open the shield box, you will see an image similar to the one below. If you are unfamiliar with specialized hardware and electrical processes, avoid unplugging the power supply and manipulating components.

 Many internal components store electricity, which in some cases is extremely dangerous.

As mentioned, power supplies have a similar internal structure and pattern. Although the model and manufacturer of the internal components are different, you will still find the same structure in the different PSUs. The main mains connection to the power supply can be seen above and to the left of the image above. The current flows clockwise in the image above and exits the power supply at the bottom left of the image (colored wires).

Looking back at the main board of the power supply, we can see the big difference with the connectors on the motherboard . The electronic connections on the PSU board are deeper and wider, and are designed to allow more current to flow.

The most obvious point in the design of the flow board was the gap that created two distinct parts of the board.

The existence of two distinct components in the power supply board connections indicates that this equipment necessarily requires a distinct classifications, known as primary and secondary. The first part of the task is to initialize the input voltage so that the high-efficiency voltage can be changed from the urban input. The second or second section manages change activities and subsequent processes.

Smoothing

PSU’s main activity on power supply is not just switching from AC to DC or reducing voltage. In fact, the power supply smoothes the input voltage. Since many electrical appliances in homes, offices, and business environments are connected to municipal electricity, their activity causes a fluctuation in current. Electrical appliances are constantly switched on and off and emit electromagnetic signals. The oscillation at the input stream not only makes it difficult to synchronize the urban current, possibly damaging the internal components of the PSU as well.

The power supply investigated in this article uses two transient filtering steps. The first filter is applied to the inlet socket and uses three capacitors to filter. In fact, three capacitors can be considered as accelerators for fluctuations in the input voltage. The second step is more complex filtering, but it performs almost the same operation.

The main task of the power supply is to convert the current and reduce the voltage

The yellow blocks you see in the pictures are capacitors and the green circles covered with copper wires are inductors (they are also known as “choke” when used as inductors). The inductors hold the electrical energy in the magnetic field, but the same field pushes back the supply voltage. Finally, a sharp change in voltage causes the magnetic field to counteract the reverse pressure.

The two small disks you see in the image are also capacitors and beneath them (in a black plastic case) is the metal oxide varistor (MOV). These components also help create resistance oscillations against the input voltage. For a better understanding of the difference between flow filter circuits, read this link .

The public stream filtering section sees the most action to reduce costs. Cheaper feeds use less filtering, and cheaper samples have no filtering part, which ultimately makes them less suitable.

Voltage change

As mentioned above, the main task of the power supply is to change the AC voltage from more than 5V to DC voltage of 5, 5, and 1.5V. During this change, the AC voltage first changes to DC. The Master Cooler power supply uses a piece called a rectifier. In the image below, the rectifier is the same black piece that is attached to the other metal piece (heatsink).

Rectifiers are one of the common goals of manufacturers to reduce costs. Cheaper components have poor voltage conversion activity and sometimes generate more heat. However, after the AC voltage, eg 5V (which is the maximum voltage of 5V) to the rectifier, we will see the DC voltage output equal.

After converting the voltage to DC, it is time to reduce it.

The next piece in the existing sample is called active power factor correction converter or APFC. The existing circuit regulates the electrical current in the unit. In fact, because the circuit consists of multiple components with the approach of maintaining and releasing energy in complex approaches, the final output power may be less than required; therefore, it needs adjustment.

Conventional power supplies use passive converters instead of APFCs, which have the same performance as the above component. Passive components, though less effective, are suitable for low power supplies.

You can see the APFC segment in the image above. The large cylinder parts to the left of the image are capacitors that hold the regulated current.

The flow is then transferred to the next phase of the PSU process chain.

The piece behind the APFC is known as the Pulse Width Modulation or PWM. This unit receives the DC voltage and utilizes multiple field-effect transistors to cut the voltage at a very high rate. In fact, the above component converts the DC voltage to AC again. The current conversion is done because the voltage reducer in the power supply is a transformer. These components use electromagnetic induction and a set of two coils (one of which has more rings than the other) to reduce the voltage. Finally, the transformers only work with AC voltage.

The AC voltage frequency (the rate of change measured in units of Hz) has a large effect on the efficiency of the transformer (the higher the frequency, the better), thus the frequency of 1 to 2 Hz coming from the mains to a frequency of between 2 and 2 Changes to 4,000 Hz.

The higher the efficiency of the transformer, the greater the capability to reduce it.

Ultra-fast DC voltage switching can also be called a switched mode power supply or SMPS.

In the picture below, you will see three transformers. The largest of them produce only 4 volts of output. In other power supplies, large transformers supply all the voltage needed. The next largest piece generates five volts of output, which we will describe in the next sections.

The smallest transformer also plays an insulating role for the PWM circuit. This third component removes the PWM circuit and also prevents interference with other voltages in the PSU.

Different power supplies have different ways of making the required current, isolating the PWM circuit and other activities.

The use of components is highly dependent on budget constraints and the need to source. However, irrespective of the part used, the power supply must be converted back to DC after receiving the voltage from the transformer.

In the image below, the large metal piece is called the Heatsink, which is used for AC to DC rectifiers. The circuit you see in the middle of the image also contains a set of Voltage Adjustment Modules (VRMs) that generate output voltages of 5 and 0.5 volts. At this point you need to become familiar with the concept of ripple.

In very ideal and impossible situations where the power supply operates at a very high efficiency, the AC voltage is converted to DC voltage at steady state.

 In reality, however, this is not possible and DC voltage is constantly changing.

The change in DC current is known as the ripple voltage, and we should minimize it in the power supply. Cooler Master does not provide voltage fluctuation information in its power supply specifications. However, reviews in the tech media report a fluctuation of up to 1.5 volts (2 mV) for the source being investigated.

The figure below shows how much the output voltage fluctuates with the voltage required. The red line is the DC voltage of 4 V and the blue line represents the actual output. Of course, the oscillation is never completely constant.

The quality of capacitors used in the power supply plays an important role in regulating the oscillation. The smaller the capacitors, the lower the capacitance used, the greater the oscillation that makes the output ideal.

If the ripple is high, the complex circuits located in other parts of the computer are likely to operate in an unstable approach.

In the sample under study, the 2 mW seems appropriate, although still far from ideal.

Regardless of the component used to generate the DC output voltage, we still need a few more circuits to reach the final output step of the PSU. The next steps are to manage the power output. In fact, we are implementing solutions in the final stages so that the demand for one voltage (out of the four voltages offered) does not affect the other voltages.

The chip you see in the image above is called the Superviser and manages the outputs from the power supply. In fact, the chip is responsible for checking the sufficient voltage and output current.

Of course, the chip does not have such a complex function, and only when the current or voltage is too low or too high does it supply power.

More expensive power supplies use digital signal processors to monitor output. The processors also perform the voltage adjustment task as needed and also send details of the power supply status to the computer. This may be more common for PCs than for ordinary users, but in advanced servers and machines it is a critical requirement.

Get started

All power supplies are equipped with huge bunches of wires coming out of the back. The number of wire assemblies and how they are connected to the main unit varies widely among the samples. Of course, all samples have standard connections.

As mentioned, voltage depends on the energy difference, so we have to have two wires for a given output. One of them is used as the main voltage and the other as a reference wire to measure the difference. The reference wire is also known as the ground wire.

Eventually the sum of the two wires creates a cycle that cuts the current out of the power supply and then returns to the power supply after reaching the power supply.

The current is generated by the wiring cycle.

Since some cycles require low volumes of current, ground wires can be used jointly in multiple cycles.

The first cycle of current transmission wires is the ATX12V version 6-pin mandatory 4-cycle cycle, which offers multiple wires for different voltages. In addition, specific systems also appear in this cycle.

One of the important wires in the ATX12V cycle is known as the 5V standby + wire. As long as the power supply is plugged in and connected to the motherboard, this cable will be active. The wire is active because the computer never actually shuts down completely.

When you send the shutdown command to the operating system, the motherboard receives the power it needs from standby connection to stay active.

There is an additional eight-pin connector for the motherboard, which offers two sets of 12V + and center wire. In addition, most power supplies have at least one PCI Express connector equipped with 2 or 8 pins. Graphics cards are capable of receiving up to 4 watts of power from the PCI motherboard.

As a result, this connection is capable of delivering even more power to today’s giant graphics cards.

Cooler Master Power Supply As discussed above, for reasons of cost savings, two PCI Express connectors use the same wires.

The difference between the eight-pin and two-pin connectors can be seen in the two ground wires. The addition of two ground wires makes it possible to reduce a high-voltage current to 4V, which would be a better power supply for very large, high-end graphics cards.

In recent years modular motors have been marketed. All new marketing claims mean that some power connections are wired to other connections that connect directly to the PSU.

As a result, using modular samples, some unused wires can be removed from the circuit to create more space inside the computer case.

Cooler Master model Like all conventional market models, it uses a completely basic connection system for modular cables.

Each connector provides 1, 5, and 1.5V voltage with two ground wires. The configuration of the wires at the end of the cable also varies depending on the device to which the cables are connected.

The SATA connector you see in the picture above is used to supply the power required for the hard drive and SSD.

 In addition, other devices, such as DVD drives, use the same connector.

The familiar look you see above is known as AMP MATE-N-LOK 1-480424-0, which is generally referred to by users and experts as Molex. Of course, Molex is the name of the top connector manufacturer. This connector offers a 1-wire with a five-volt cable and two ground wires.

Power outlet cabling is another area where the manufacturer has the ability to enforce cost reduction policies. Even at a higher cost, the appearance and flexibility of the wires used can be improved. Even thicker wires can be used, which are less resistant than thinner ones and produce less heat when power flows.

Internal power

We mentioned at the beginning of the article that the naming of power supplies is based on the maximum electric power they can supply.

In simple terms, power can be obtained by multiplying the voltage in the current (in watts). Although the specialized definitions certainly challenge this initial definition, it seems sufficient for the present article and a simple explanation of the performance of the PSU.

The model under consideration, like all conventional models on the market, provides information on the amount of electricity provided by each voltage line.

In the picture above, the total power available from all 4V power lines exceeds 2 watts.

If I add the sum of the other lines to my calculations, we get 2 watts. This is achieved by the fact that ordinary lines of 5 and 0.5 V are obtained by using VRMs of 1V output. In addition, the entire output voltage is extracted from a single source, ie electricity. Finally, the 2 watt rate is the maximum that indicates the power supply.

Maximum power in the power supply means that, for example, if you extract 5 watts of energy from a 5V line, only 2 watts will be available for other lines. Fortunately, most computer hardware receives a large share of its power from 4V lines.

 As a result, if you have chosen the right power supply based on your computer’s needs, you will rarely have a power outage problem.

The power supply along with the maximum power information has a label with the 80Plus Bronze label, which is recognized as an industrial efficiency rating and the manufacturer optionally observes it. Power supplies must meet a number of legal requirements to obtain the standard. The efficiency of the power supply also depends on its output volume. In other words, the current intensity extracted from multiple lines has an impact on the efficiency.

In the efficiency example, if you use the cooler master sample with a 2 watt output (half max power), one can expect from 2 to 5 percent efficiency depending on the voltage supplied from the municipal source.

Then  In such a situation, the power supply of 1 to 2 watts of electricity is extracted from the wall outlet.

Ultimately a efficiency greater than 1% does not mean that the power supply is capable of delivering more power. In fact, higher efficiency only indicates that the power provided is less wasted during multiple filtering, rectifying, switching and switching processes.

The maximum efficiency of power supplies occurs when we extract between 2 and 3 percent of the nominal load. Some manufacturers provide diagrams similar to the one below that show the efficiency rates at different loads and voltages.

Some power supplies in their specifications section are calling Single rail or Multi-rail and some have switches to switch between the two.

The term rail is a term use for a particular voltage produces by a power supply.

The current Master Cooler prototype has a single 4V rail, and all the various 4V connectors extract their current. A multi-track power supply probably has two or more systems to deliver 4V output, and there is a big difference in how the outputs are using.

Power supplies use for datacenter applications have multiple rails to minimize the output tolerance cause by a single train being stop.

The multicore power supply for the PC may also have the same structure, but in those cases only the 4V output is split into two or three outputs.

For example, the current sample extracts up to a maximum of 1 amp from a 4V line that is equal to 2 watts of power. The cheaper, multilevel version of the same power supply will probably have two 4V outputs, but each will be capable of delivering a current of 5 amps.

Finally, a PC power supply with excellent design and components does not require a 5V multilevel system.

Pricing strategy

Power supplies come in a wide range of price tags. Looking at the current product list, you can see a power supply of $ 1 and a power of 2 watts, a model with a price tag of $ 1 to $ 2, and a output of 1.2 watts. Now the question arises, what options does the higher price offer the user? Why pay more than $ 5 for a power supply?

The quality of the PSU’s internal components greatly influences its pricing

The ability to deliver more electricity is one of the obvious reasons for the increase in the price of the power supply. On the other hand, how the same high output is deliver also has a huge impact on the price of the product. The ultra-cheap models deliver only 4 amps per 1V output line.

While more expensive samples have up to three times the current intensity. Today’s mainframes and graphics processors extract the bulk of the power they need from 4V lines. Is 4 amp current sufficient for them?

Imagine buying a professional processor with 6 cores and a flagship graphics card for your PC. Both require 5 watts of maximum power.

The cheap power supply of components cannot meet their needs. On the other hand, very expensive power supplies sometimes have an unreasonable profit margin and high price. Of course, users who pay $ 3.00 or more for the GPU will certainly have no problem paying a few hundred dollars for the power supply.

The main cost of a power supply is due to the quality of its internal components. If you take a look at the beginning of this article, you will notice a small number of components in the cooler power supply.

Since all of these components have a critical impact on unit performance, paying more for higher quality components would certainly make sense.

Finally, by studying the structure and performance of the power supply unit, we discover its importance and sensitive engineering designs for its better performance. 

We now know that building a power supply that is supposedly simple to operate requires some complexity.