Redundant power supply overview
Redundant power supplies are used in web servers and consist of two identical power supplies controlled by an integrated circuit (IC) to provide web services. When one power supply fails, the other immediately takes over. Even after power supply separation, the two power supplies work together seamlessly. Redundant power supplies enhance the scalability of web servers. Besides web servers, disk array systems are also widely used.
RPS (Redundant Power System) power supplies serve as the external DC power supply system for some network switches.
RPS can be used as a power source for cluster servers of network switches or wired routers.
If the RPS and the receiving electrical equipment use the same AC power supply system, when an abnormality is detected in the power supply of the receiving electrical equipment, the RPS can once again provide DC power to the problematic industrial equipment, ensuring the normal operation of the industrial equipment;
If the RPS and the receiving electrical equipment use different AC power supply systems, DC power can be supplied again when the external AC power supply to the receiving electrical equipment causes problems, ensuring the normal operation of the industrial equipment.
What is a redundant power supply? What is the difference between a redundant power supply and a UPS (Uninterruptible Power Supply)?
Commonly used redundancy schemes for power supplies include volumetric redundancy, cold redundancy, N+1 backup with parallel current sharing, and hot redundancy. Volumetric redundancy refers to the power supply's maximum load capacity exceeding its actual load, but this doesn't significantly improve stability.
Redundant cold backup refers to a power supply consisting of many identical control modules. Under normal conditions, one module powers the system, and if that module fails, the backup module can immediately take over. The drawback of this method is that there is a power switching interval, which can easily lead to a voltage shortage in the operating requirements.
Parallel current sharing N+1 backup refers to a power supply composed of many identical modules, each connected in parallel via OR gate diodes, with each module simultaneously supplying power to the industrial equipment system. This type of scheme is less likely to damage the load power supply system if one power supply fails, but a short circuit at the load end can easily affect all modules. Redundant hot backup refers to a power supply composed of many modules that can operate simultaneously, but only one of them supplies power to the industrial equipment system, while the others are idle. If the main power supply fails, the backup can immediately take over, with very small output voltage fluctuations.
For systems requiring long-term, uninterrupted operation and high reliability, such as communication base stations, industrial equipment, and servers, a highly reliable power supply is generally essential. Redundant power supply design is a crucial element here, playing a key role in scalable systems. Redundant power supplies typically consist of two or more power sources. When one power source fails, the others can immediately take over, ensuring the normal operation of the industrial equipment. This is similar to the basic concept of a UPS (Uninterruptible Power Supply): when the standard operating voltage shuts off, a rechargeable battery replaces the power supply. The main difference between a redundant power supply and a UPS is that both use different power sources simultaneously, while a UPS uses one power source as a backup, automatically switching when needed.
Traditional redundant power cable connection method
Traditional redundant power supply designs consist of two or more power supplies connected in parallel to the power system bus via an OR gate, each connected to anodized diode. This allows one power supply to operate independently or multiple power supplies to operate simultaneously. If one power supply malfunctions, the unidirectional conduction of the diodes prevents damage to the power system bus output.
In specific redundant power supply system software, the current is generally quite large, capable of being tens of A. Taking into full account the power loss of the diodes themselves, Schottky diodes with low loss and very high current capacity are generally used, such as SR1620 to SR1660 (rated voltage 16A). These diodes are usually equipped with heat pipes for heat dissipation.
Traditional power supply circuits using diodes are simple, but they have inherent drawbacks: high power loss, significant overheating, the need for heat pipes, and large footprint. Because power circuits typically handle high currents, diodes are mostly in forward-conducting mode, and their power loss is not negligible. Even the lowest-loss Schottky diode has a voltage of 0.45V, and at high currents, such as 12A, it can exhibit a power loss of 5W, thus requiring special attention to heat dissipation.
Current redundant power supply solutions utilize high-power MOSFETs to replace diodes in traditional power circuits. MOSFETs can achieve on/off internal resistances of several mΩ, significantly reducing losses. In high-power applications, this not only achieves high efficiency but also saves considerable PCB footprint by eliminating the need for heat pipes and heat sinks, thus reducing heat dissipation costs in industrial equipment. When using MOSFETs in power circuits, it is advisable to have them controlled by dedicated integrated circuits (ICs).
