The Influence of the Operating Temperature on the Service Life and Reliability of Switching Power SuppliesMay 21, 2019 by Stefan Bergstein
This article highlights EMTRON electronic GmbH contact cooling that heat dissipation must be ensured via suitable heat sinks/housing surface/baseplate.
The question of whether to use a transformer power supply or a switch-mode power supply is no longer discussed today. In addition to superior technical properties, switched-mode power supplies impress with low costs, low weight and little space requirement. Certain requirements such as PFC (Power Factor Correction > 0.9 and sinusoidal mains current) can only be met with switched-mode power supplies. The question remains, what are the greatest influences on the service life and reliability of a switched-mode power supply?
Significant technical advantages of high-quality switching power supplies include:
- Regulated output voltage without additional downstream regulator
- High efficiency, low losses independent of input voltage
- Wide range input (e.g. 95 - 264 V) possible with little effort
- High short-term overload capacity
Where there is light, there must be a shadow. Hf interference on the supply lines, as well as radiated interference, may be disadvantageous, as well as possibly shorter service life and lower reliability compared to conventional power supplies due to the majority of components required.
Operating Temperature is a Significant Influence
The internal operating temperature has the greatest exponential influence on service life and reliability: the service life decreases by half per 9 °C temperature rise.
Decades at low temperatures can quickly turn into a few years or even months at high temperatures. Every degree less counts! Although it is possible to achieve a longer service life at high temperatures by using highly developed components, the cost-benefit factor must be considered when using such special components. The use for standard applications is excluded in many cases for reasons of cost-effectiveness.
Provided that a switching power supply has been professionally developed, the manufacturer first determines the required installation and cooling conditions. However, if the user operates the application in a thermally critical situation, e.g. due to inadequate ventilation or heat dissipation, he impedes the cooling of the unit. This leads to overheating and thus to a reduction of service life or even to failure. Although all high-quality switched-mode power supplies contain an over-temperature sensor that switches off when its response temperature is exceeded, this does not prevent long operation just below the switch-OFF threshold. This drastically reduces service life.
Maximum permissible operating temperatures are specified for all components. However, this only means that the function is guaranteed at maximum temperature. However, this has a considerable influence on the service life, which is always subject to the “halving per 9 degrees law”.
For automotive applications, for example, components are available for operating temperatures well above 100 °C, but it should not be ignored that a car only has a service life of a few thousand active hours, during which the maximum temperatures only occur temporarily.
High Efficiency for a Long Service Life
What options does a manufacturer have to achieve a long service life and reliability? The higher the efficiency, the lower the losses and thus the lower the self-heating. This entails a certain amount of extra work, which may justifiably result in a higher price. The most critical components in any conventional switching power supply are the electrolytic capacitors because they contain an electrolytic liquid that escapes more or less quickly through the seal in the event of overheating.
Electrolytic capacitors belong to the most important and largest components of the power supply. The temptation is great to save money on electrolytic capacitors, i.e. to overload them or to choose inexperienced and therefore low-cost suppliers. The manufacturers we represent only use high-quality capacitors from well-known brands, because a cheap, unsafe component source does not meet our responsibility and understanding of quality.
Since the production of these components requires decades of experience the number of competent manufacturers is limited and their prices are marginally increased. Also the desire for small dimensions and in particular low overall height, restricts the available volume for electrolytic capacitors. Low, wide electrolytic capacitors have a shorter service life because they provide a larger sealing surface and shorter ways for the electrolyte to evaporate. SMD electrolytic capacitors are exposed to extreme temperatures during reflow soldering and thus lose some of their service life. Towards the end of the electrolytic capacitor’s service life, they become exponentially hotter and hotter. As a rule, they finally “smash through” and blow the electrolyte into the environment via safety seals.
In the comparative test of switching power supplies, the measurement of the surface temperature of all electrolytic capacitors on the roof is one of the first and most important measures. In order to ensure a diligent approach, the technical data of the electrolytic carbon manufacturer should be obtained, indicating the service life above temperature. In many cases, power supply manufacturers also provide technical reports with information on component temperatures.
However, temperature limits apply to all components of a switched-mode power supply, especially to all insulation materials. The insulation properties of conventional PE films decrease noticeably as the temperature rises and drastically at high frequencies. According to the electrolytic capacitors, plastic film capacitors are at risk at temperatures above 100 °C. The power ferrites of inductive components show rapidly increasing losses at temperatures above 100 °C, which can lead to saturation with subsequent destruction of active components.
The usual use of SMD components worsens the thermal situation because power diodes and transistors soldered onto the circuit board heat up the adjacent electrolytic capacitors. The usual board materials also cause losses at high frequencies.
In view of any thermal loads that may occur within the service life, an initially more expensive switched-mode power supply is usually more economical in use, as the higher-quality, temperature-resistant components installed significantly increase the overall service life of the application.
In addition, it goes without saying that every user of switched-mode power supplies should ensure optimum cooling.
Serious SNT suppliers specify a curve for the load capacity in dependence on the temperature of a defined test point on the housing. The best advice is to choose a switching power supply with a higher specified power than the required continuous power and not to use more than 75% of the rated power for a short time.
Different Cooling Concepts
Different cooling concepts are required for different applications. Primarily, it is necessary to distinguish between the physical heat dissipation concepts by convection with free or fan ventilation or by heat dissipation, so-called contact cooling, often via a base area, e.g. the upper side or the so-called baseplate cooling.
Convection cooling is classically used in applications, e.g. in switch cabinets with sufficient free air spaces. However, during installation it is important to pay attention to the mounting position and minimum distances around the power supply unit so that the required convection and thus the heat dissipation can take place. The information can be found in the data sheets and in the installation instructions.
Another large field of application for convection-cooled switching power supplies is modern LED lighting. Forced ventilation with a fan would often be inappropriate simply because of the acoustic fan noise. In many cases, the housing forms have correspondingly integrated heat sinks. In addition, the interior components are often fully encapsulated. On the one hand, this ensures high insensitivity to moisture and dust; on the other hand, and no less importantly, modern potting compounds ensure optimum heat transfer to the surrounding enclosure wall.
When implementing a switched-mode power supply in a closed system, there are often only small areas available for assembly. In addition, the air circulation in the system housing of very compact high-performance power supply units is often not sufficient to dissipate the heat loss. In this case, the requirements in the data sheets must be followed exactly. Either one dispenses with a part of the performance at increased ambient temperatures - in this case one speaks of derating - or forced ventilation by means of a fan cannot be dispensed with. The required amount of air circulation is often expressed in CFM (cubic feet per minute). In the European linguistic area one finds more frequently the data in m³/h.
A conversion results from 1 CFM = 1.699 m³/h
For high performance requirements, or with a compact design, forced ventilation is often indispensable. The manufacturers then immediately equip the power supply components with appropriate fan units. Depending on the version, the required air volume flows can be adjusted by changing the fan speed. This means that the fans rotate at a low speed and thus with reduced noise development when the power requirement or the heat development is low. If the power requirement increases, the fan speed is automatically increased, thus ensuring optimum cooling of the switched-mode power supply.
Some manufacturers also offer special functions, such as reversing the flow direction.
If the use of a fan is not possible, or if the application, e.g. in some cases in the medical field, does not allow any disturbing fan noises, other cooling concepts are possible. For contact cooling, heat dissipation must be ensured via suitable heat sinks or via a housing surface or baseplate. In addition, sufficient convection must, of course, be ensured. If convection is not possible, systems with heat pipes or liquid circuits are already in use and thus enable trouble-free operation without power restrictions.
With DC/DC converters in the higher power range, the upper side of the housing is often intended for flange-mounting an appropriate heat sink. The required size of the heat sink can be calculated from the required performance conditions and the manufacturer’s specifications. In many cases, however, suitable heat sinks are included in the Emtron electronic sales program.
A relatively modern concept for contact cooling is the so-called base-plate cooling.
The heat is transferred via a sufficiently large contact surface on the underside of the power supply unit. Due to the design, special circuit board materials and geometries are used to ensure heat transfer. The power supply unit can, for example, be screwed directly onto a correspondingly designed housing wall. In addition, it is still possible to mount the power supply unit on a specially designed heat sink based on convection or with a liquid circuit to ensure the necessary heat dissipation.
However, a sufficiently small heat transfer resistance between the heat dissipating housing part or the base plate and the heat sink must be ensured. Auxiliary means for this are for example the well-known heat conducting paste or specially offered heat conducting pads.
Depending on the application, a wide variety of requirements arise for switching power supplies.
Last but not least, the cooling concept must also be taken into account when selecting the product.
In addition to our solution-based consulting services, Emtron electronic as a specialist distributor has long-standing contacts with many manufacturers in order to select the right power supply for the respective application from our extensive product portfolio together with our customers.