We all know that the power supply is one of the important elements of a computer. A low-quality model can quickly fail, taking away the rest of the components. Let’s find out how the components used in the PSU affect the reliability and stable operation of the PC.
When assembling, some users do not pay due attention to the choice of the power supply unit, but buy it “for delivery.” Others deliberately try to save money and choose budget models, even for powerful computers. Let’s figure out what it is fraught with.
Expensive and cheap power supplies differ not only in the quality of the radio components used, but also in the circuitry used in them. That is, decisions that determine by what principle the conversion of alternating mains voltage into low constant output voltages will occur. Namely, how will their stabilization and smoothing be carried out, according to what algorithm will the protection systems work against short circuit, overcurrent, overheating, overvoltage and undervoltage, and others.
The power supply generates voltages for all PC components. The difference in the circuitry implementation of this process in budget and premium models lies in the principle of stabilizing the generated voltages.
What is stabilization? This is the ability of the power supply unit to maintain the output voltage unchanged under the influence of various destabilizing factors: when the current consumption, temperature, mains voltage and nominal parameters of radio components (when they age) change.
Budget blocks have group voltage stabilization – that is, only one stabilizer for two lines of 5 and 12 V. Voltages are literally interconnected. This is dangerous because with an increase in the current consumption along one of the lines, the voltage on it will begin to sag, and the power supply unit will increase the pressure on both lines to compensate. As a result, an excess over the nominal value will occur on the second line.
Note. The 3.3V line is individually stabilized in any case.
For example, the higher the current consumption along the 5 V line and the greater the drawdown on it, the stronger the increase in voltage on the 12 V line – up to 13, or even up to 14 V. Having received such a generous “vigor” from the power supply, the components may fail.
In expensive power supplies , separate stabilization is used, in which each of the lines has its own independent converter. Load fluctuations for any of the lines do not affect the voltage of the others.
Load characteristics of PSU with separate and group stabilization
In inexpensive models, not all the necessary protections are implemented. Either they exist, but they do not work perfectly. Due to the combination of the 5 and 12 V lines, these power supplies have a common power overload protection. The latter is triggered not by the set values of the maximum current in the lines, as is done in expensive power supplies, but by exceeding the total power. Specifically for the 5 V line, the pickup current may be too high. A situation may arise when the components actually begin to burn out, and the power supply unit considers this to be simply an increased load.
Premium PSUs have overcurrent protection (OCP) in addition to overpower protection. Moreover, an individual trigger value is set for each line. According to international standards, one conductor of electrical current (the wire from the power supply to the accessories) should not transmit more than 240 watts of power. Since the most powerful line (12V) can exceed this threshold many times, manufacturers divide each line into several groups. All of them must be equipped with their own overcurrent protection. But some vendors are tricky – they use only one protection scheme, but with an increased response threshold, connecting the outputs of all groups to it.
Since we are talking about defenses, let’s briefly list the main ones.
OCP (Over Current Protection) – overcurrent protection, which was discussed above.
OPP (Over Pover Protection) – overload protection for the total power of all channels.
OTP (Over Temperature Protection) – overheating protection.
OVP (Over Voltage Protection) and UVP (Under Voltage Protection) – protection against over and under voltage, respectively.
SCP is a conventional fuse at the input of the power supply.
In cheap PSUs, as a rule, only four of the six protections listed above are used – these are OPP, OVP / UVP and SCP.
The reliability of the power supply and the quality of the generated voltages directly depend on the components used in the design. The most common radioelements in power supplies are, of course, capacitors. In budget models, they put aluminum electrolytic ones. Their distinctive features are: low cost, low reliability, short service life and rather average performance characteristics.
More expensive PSUs use polymer capacitors. But not everywhere, but only in critical areas of the electrical circuit. The “polymers” are much better in terms of reliability, and the operating parameters are significantly superior to “electrolytes”.
The moment has come to understand the device of capacitors in more detail. Let’s find out how their quality affects the formation of supply voltages.
The aluminum electrolytic capacitor has a large capacity in a relatively small size. The production cost is low, so this type is inexpensive and very popular.
Aluminum electrolytic capacitor design
Structurally, it consists of two strips of aluminum foil, between which paper impregnated with electrolyte is placed. The whole structure is rolled into a tight roll and packed in a sealed metal case. The dielectric is alumina on the surface of the foil, which acts as a positive plate (anode). Oxide is formed by the interaction of the electrolyte with the surface when an electric current flows, therefore its thickness is very small – due to this, a large capacitor capacity is achieved. The cathode is an electrolyte that is in electrical contact with the entire surface of the unoxidized plate, connected to the negative terminal.
In addition to aluminum, there are other types of electrolytic capacitors – for example, tantalum and niobium. The dielectric layer in them is formed by the oxide of these metals, so they are more expensive to manufacture.
The design of polymer capacitors is similar to that of aluminum electrolytic capacitors . The difference is that they use conductive polymers as an electrolyte. The latter are in a solid state: a dielectric oxide layer is created not on the plate, but on the surface of the conductive polymer layer.
Design of polymer and hybrid capacitors
A liquid electrolyte can be combined with solid conductive polymers – such capacitors are called hybrid capacitors .
Currently, four types of polymer capacitors are produced, three of which (SP-Cap, POSCAP, OS-CON) have a solid conductive polymer as an electrolyte and differ from each other only in the material of the plates. The fourth type is hybrid (Hybrid).
Types of polymer capacitors
Any polymer capacitor is better in performance than even the highest quality electrolytic capacitor. We’ll talk about this in more detail in the next section.
Speaking of terminology, it is worth noting that it is wrong to separate polymer and hybrid capacitors from aluminum electrolytic ones. In fact, they all belong to the same group – electrolytic. But in technical jargon there is a traditional division into “electrolytes” and “polymers”, and we will use it for convenience.
Consider the main parameters by which capacitors differ.
Electrical capacitance is the ability of capacitor plates to accumulate electrical charge. Measured in Farads (F) or fractions (μF, nF, pF). The value is usually indicated on the case.
Rated voltage – the value at which the operating parameters of the capacitor are maintained throughout the entire service life.
The maximum permissible operating temperature is also usually indicated on the housing.
Examples of designations for the maximum permissible temperature
The dependence of the life of capacitors on temperature
ESR (Equivalen Series Resistance, translated as “equivalent series resistance”) consists of the sum of the active resistances of the plates, leads, electrolyte and contact connections of the plates with the leads. It is parasitic, that is, harmful. The electrolyte has the greatest influence on the ESR value. A real capacitor can be schematically represented as a series connection of a parasitic resistance R and an ideal capacitor C:
This resistance leads to losses during both charging and discharging of the capacitor. Thus, the quality of smoothing the voltages generated by the power supply deteriorates. In addition, when the current passes, heat is generated, that is, the capacitor heats up. We conclude: the lower the ESR, the better the capacitor.
ESI or ESL (Equivalen Series Inductance) is also parasitic. It arises from the imperfect design of the capacitors and consists of the sum of the inductances of the plates and leads.
Spiral wound capacitors are of great importance ESI (ESL). Considering this parameter, we represent the real capacitor as a series connection of the parasitic inductance L and the ideal capacitor C:
At a low frequency of the pulse current passing through the capacitor, the inductive reactance will be very small and will not affect the operation. But as the frequency increases, the inductive reactance will also increase. At frequencies over several hundred kilohertz, the electrolytic capacitor will completely cease to perform its functions.
Thus, the equivalent circuit of the capacitor, taking into account all the physical imperfections of the design, is as follows:
In addition to the above parameters, a parasitic resistance R leakage has been added . It characterizes the leakage current between the capacitor plates due to imperfections in the dielectric material.
Having described the equivalent circuit as the sum of the resistances of all its active and reactive elements, we obtain the complex resistance Z, also called impedance . The lower the impedance of the capacitor, the better it is.
Dependence of the impedance of an electrolytic capacitor on frequency
It can be seen from the graph that the impedance in the low-frequency region is determined by the capacitance of an ideal capacitor, in the mid-frequency region it is limited by parasitic ESR, and as the frequency increases further, the impedance is increasingly influenced by the inductive reactance of the parasitic ESL.
TKE (temperature coefficient of capacitance) characterizes the relative change in capacitance with a change in temperature. This is a harmful phenomenon, frequency-setting circuits are especially critical to it. When the temperature of the operating device or the environment changes, the temperature of the capacitor also changes, and the frequency begins to “float”.
Comparative TKE of polymer and ceramic capacitors
DC-bias ( DC bias effect) characterizes the dependence of the capacitance on the applied voltage. For example, if the voltage across the MLCC capacitor (see graph below) is increased to its maximum value, the capacitance may decrease by 65% of the nominal value.
Comparative dependence of capacitance on voltage
Every self-respecting capacitor must keep the capacitance constant. As you can see, polymer copes with this task perfectly.
Advantages of polymer capacitors
We figured out the device, now let’s find out what it all means in practice.
Compared to conventional electrolytic capacitors, polymer capacitors have a lower ESR, respectively, and a lower impedance. When using the former in the power supply smoothing filter, the charge accumulated from the source and delivered to the load will be greater, the smoothing of the output voltage ripples is better, and the heating is much less.
Influence on the quality of voltage smoothing of electrolytic and polymer capacitors
The reliability of polymer capacitors is an order of magnitude higher than that of aluminum electrolytic ones. In the latter, liquid electrolyte often dries out, especially if they are not correctly placed in the device. For example, in the immediate vicinity of hot cooling radiators. The increased temperature not only promotes faster drying, but also reduces the life of the electrolytes. It also leads to swelling – a violation of the tightness of the case by breaking the safety notches.
Bloated aluminum electrolytic capacitor (left)
The drying effect leads to a decrease in the capacitance of the capacitor and an increase in the ESR. The power supply will definitely not say thanks for this, but sending components to heaven can easily.
There can be no drying out in polymer capacitors – they use a solid conductive layer. But operating at elevated voltages can also cause the housing to swell and rupture.
Bloated polymer capacitor
“Polymers” are capable of self-healing upon local breakdown of the oxide layer. When exposed to a large short-circuit current, a strong heating of the conductive polymer occurs at a local point. The molecular chain in the defect zone is destroyed. As a result, a dielectric layer is formed that insulates the breakdown site.
Local breakdown self-healing processes
In aluminum electrolytic capacitors, such a breakdown will grow like an avalanche. This will rupture the case and damage the entire power supply.
Summing up, let’s compare the performance parameters of the considered types of capacitors.
Comparative assessment of the parameters of capacitors of various technologies
Choose your power supply as carefully as you choose other important components of your computer: processor, graphics card, or motherboard. Be sure to pay attention to the presence of all types of protections, manufacturers indicate them in the description.
Before buying, study the reviews , from them you can determine what type of capacitors is used in a particular unit. The use of polymers, albeit in part, has a positive effect on the reliability and durability of the power supply unit.
Polymer capacitors in the Corsair HX1200 PSU
Thermaltake Toughpower GF1 1000W Polymer Capacitors
Thermaltake Toughpower GF1 1000W – TT Premium Edition Power Supply [PS-TPD-1000FNFAGE-1]
We repeat once again – you should not save on the power supply . As Baron Rothschild said: “We are not rich enough to buy cheap things.”