- Aerospace and Defense
- Battery Chargers
- Broadcast Transmission
- CO2 Laser Exciter
- Internet of Things (IoT)
- ISM 5-6 GHz
- Motor Drives
- Plasma RF Generator
- Point-to-Point Radio
- Power Conditioning
- RF Heating
- RF Test and Measurement
- Solar Power
- Wind Power
- Wireless Infrastructure
Uninterruptible Power Supply (UPS)
Supplier Storefronts for Application: Uninterruptible Power Supply (UPS)
Access the latest products and information from our suppliers
Richardson RFPD provides design assistance for almost any power component designed into a UPS system. Our Energy Management and Power Conversion engineering team has over twenty years experience with single-phase and 3-phase power supply designs, including sophisticated UPS systems. We supply most components and sub-systems needed for UPS circuit designs. Our engineers can help you select the right filter capacitors, power factor correction (PFC) capacitors, snubber capacitors, current sensors, heatsinks, rectifiers, MOSFETs, IGBTs, ultracapacitors, power resistors, or gate drivers to best fit your circuit design requirements.
|Table of Contents:
Overview of Uninterruptible Power Supplies (UPS)
The main purpose of a basic back-up power system, or a truly "uninterruptible" power supply (UPS) system, is to keep the power on during utility power outages. These systems thus enable all types of electrical equipment to keep working, even when the incoming utility electrical power cannot supply the proper current and/or voltage needed. There are three basic types of back-up power systems: off-line (or “standby”), line-interactive, and in-line (double-conversion). In-line (double-conversion) is the true uninterruptible power supply (UPS) system and is considered to be the best type, because it provides the highest protection against all types of utility power anomalies (e.g., full outages, transients, surges, droops, brown-outs).
Off-line/Standby Power Supply System
When using an off-line (or standby) system, the power being used normally inside the facility is usually derived directly from the incoming utility power feed–until power fails. After power failure, a standby power generation system is switched-in (automatically) to continue supplying electrical power.
Line-Interactive Power Supply System
With line-interactive systems, a power inverter works in parallel with conditioned-input AC power (from the utility company) to supply power to the load (boosting or bucking), and only handles the full load power when the AC input power fails in various manners.
In-Line (Double Conversion) - The True UPS System
This is arguably the best type of back-up power system for any critical business that needs to have uninterrupted power, 100% of the time: data centers, banks, stock exchanges, hospitals. Here are the reasons why this type is called the true uninterruptible power supply (UPS):
In addition to backing-up a complete power failure, a true UPS system prevents all power disturbances (sags, surges, spikes, transients, harmonic distortion, and even noise distortion) from affecting the performance, stability, and life of the electronic systems under power -- and the vital data passing through them. UPS systems are designed to handle 100% of the rated load, 100% of the time. True UPS systems incorporate one or more energy storage systems and provide a complete isolation between the utility company and the load.
The remainder of this page is dedicated to the true UPS system: the In-Line (Double Conversion) type.
The main functions of a UPS system can include input filtering, harmonic filtering, power factor correction, AC-DC conversion, DC Link (i.e., short-term energy storage), longer term energy storage (i.e., back-up power), stored energy system controls, power transfer switching, DC-AC inverter circuit, snubbering, current sensors, and output filtering.
Important specifications to consider when designing a large UPS system are the required kilo-volt-amp output rating, kilowatt rating (load), and the input voltage range. 3-phase UPS systems are output-rated in kilo-volt-amperes (kVA). Note that the kVA rating is not the same as the power drain (in kilowatts, kW) of the load. The kW rating is less than or equal to the kVA rating. The main performance specifications for 3-phase UPS systems include runtime (during utility power failure) at half load, and runtime at full load. Note also that one of the back-up power sources (stored energy) could indeed be a gas-powered generator.
|View the Full-Size UPS Block Diagram|
click the image
to view the block diagram
Input Electro-Magnetic Interference/Radio Frequency Interference (EMI/RFI) Filter
Energy from electrical signals (including energy that we do not want) is virtually everywhere: on the power wires coming in from the utility company and promulgating through the air too. UPS circuits have to deal with unwanted energy (e.g., electro-magnetic interference, radio frequency interference, distortion) riding along on the input power lines and filter out the unwanted energy from the desired, relatively clean 3-phase electrical power signals. This is called EMI/RFI filtering. Making sure to reduce and/or eliminate electrical energy from storms, power transients, and other unwanted sources (nearby power generating equipment, high-power radio transmitters, etc.) is the job of the EMI/RFI filter. Also, the high-powered UPS system itself could potentially conduct unwanted energy, generated by high switching currents, back out onto the input power lines. The EMI/RFI filter works to filter-out this unwanted energy as well. Both X2 and Y2 filter-type capacitors are needed.
- X2 Capacitors are specifically designed to be placed from line to line (see block diagram).
- Y2 Capacitors are specifically designed to be placed from line to earth ground (see block diagram).
EMI/RFI Filter Module Solutions
In addition to the discrete component EMI/RFI solution of X2 and Y2 capacitors, we offer a "module" -based solution, consisting of complete, self-contained EMI/RFI filter modules from Spectrum Advanced Specialty Products. Both single-phase and 3-phase "power line filters" are available in many different configurations (e.g., single-stage, dual-stage, delta-connection, wye-connection).
AC Harmonic Filter
When a 3-phase UPS system drives non-linear loads (like variable frequency drives, DC motor speed controls, or a combination/mix of different types of machines), and this is typically the case in practice, it is very important to reduce the total harmonic distortion (i.e., current and voltage waveforms at frequencies that are multiples of the fundamental grid frequency) which occurs as a result. These harmonics must be removed, as the UPS system must not corrupt either the utility/grid or a generator (if the UPS is running on back-up generator power) with harmonic currents and voltages. These harmonic (non-sinusoidal) waveforms can cause damage, especially on a long-term basis, to a utility's network elements (e.g., transformers), to the UPS system itself, or to the back-up generator. The total harmonic distortion, or THD, of a signal is a measurement of the harmonic distortion present. It is defined as the ratio of the sum of the powers of all harmonic components to the power of the fundamental frequency.
To protect against a high THD measurement, either a passive or an active AC Harmonic Filter is used. Such a filter removes much of the distortion, bringing the THD measurement down. A large, unprotected 3-phase UPS with a 6-pulse rectifier can create heavy harmonic feedback of 30-40% THD (and higher). It can cause grid voltage distortion or generator malfunction. The Harmonic Filter typically mitigates harmonics to less than 5%. Any power capacitors used in the harmonic filter need to be able to handle relatively large ripple currents at the various harmonic frequencies. Note that the Harmonic Filter also helps with Power Factor Correction – since, as we eliminate harmonic distortion, we bring the circuit closer to the desired unity PF measurement of 1.0 (see the Power Factor and Power Factor Correction (PFC) section below).
The first step in converting the input 3-phase AC power is rectification. This stage is typically accomplished with six large rectification diodes (a.k.a. "rectifiers"), working together in a bridge-rectifier configuration. Each rectifier needs to be able to withstand the peak voltage and peak current specified for the circuit. When rectifier diodes are used, this is called an "uncontrolled bridge" application.
Alternatively, 6 SCRs (or "Thyristors") can be used in a 3-phase, fully-controlled bridge rectifier circuit. This is generally considered a better design, because the resulting voltage and current waveforms can be controlled so as to be more sinusoidal and, thus, harmonic currents can be reduced.
Whether diodes or SCRs are used, there is typically an "in-rush limiter" (most likely a power resistor) employed at the output of the bridge rectifier. This will help to limit the surge current experienced when power is restored or when the load experiences a transient.
Power Factor and Power Factor Correction (PFC)
The Power Factor (PF) that any AC circuit presents to its source (usually the utility company) is a measure of the ratio of real power (in kW) used by the entire loaded circuit to the total power (a.k.a. "apparent" power), measured as voltage times current (volt-amperes, VA), provided at the input to the circuit.
total power = resistive power + reactive power
kVA (kilovolt-amps) = kW (kilowatts) + kVAR (kilovolt-amps-reactive)
[Mathematically: PF can thus be looked at as simply the magnitude of the cosine and the phase angle between the voltage and current waveforms at the input to the circuit - a number between 0 and 1.]
A utility company (or any power generator system) would like to see a PF as near to 1.0 as possible. In other words, the utility wants to supply the amount of voltage-current product (voltage x current) that is needed by the customer (to perform the work required by the load), and no more. If the PF is low, say 0.8, then the utility needs to supply about 25% more voltage-current product (voltage x current) than it needs to in order to drive the actual load.
If a load is completely resistive, then the current and voltage waveforms are in phase, and the power factor is exactly 1.0. However, most loads are either somewhat inductive (current lags the voltage) or somewhat capacitive (current leads the voltage) and therefore present a "reactive" load to the power source... and the power factor is therefore less than 1.0 in almost all cases.
In practice, many uncorrected loaded circuits have Power Factors that are much lower than unity. Here are a few examples:
|Load Type||Typical Uncorrected|
|Induction Motor||0.70 - 0.90|
|Small Adjustable Speed Drive||0.90 - 0.98|
|0.70 - 0.80|
|0.90 - 0.95|
|Arc Welders||0.35 - 0.80|
Power Factor is so important that utility companies have regulations on how low they will allow a customer's PF. The power company may even impose fines to encourage the customer to provide proper power factor correction (PFC), adding circuitry which will improve the overall power factor until it is closer to unity. The customer's electrical bill itself is normally enough incentive for them to improve their PF. The utility measures both total power delivered (in kilovolt-ampere-hours, kVAH) and real power (in kilowatt-hours, kWH) at the meter. If the PF is below the accepted level, the power company will normally charge the subscriber for the extra power delivered, over that which would be delivered if the acceptable PF had been reached.
For single-phase circuits, it is relatively easy to provide a PFC circuit that can improve the PF to somewhere between 0.95 and unity. Either passive or active circuits can be employed, but for high-power (>1kW), active PFC circuits are generally employed. A combination of active devices, passive power components (mainly power film capacitors and an inductor), and control circuitry act together to bring the voltage and current waveforms closer to an "in-phase" condition (PF closer to unity).
With 3-phase circuits, a single-phase PFC circuit (passive or active) is generally employed on each of the 3 phases. To achieve very high power factor (0.99), a true 3-phase high frequency active PFC circuit design is currently seen as an emerging technology. These new, high-tech PFC circuits are very compact due to high frequency switching techniques. The reason it is not yet universally accepted (unlike the single phase PFC) is that there are no off-the-shelf controllers available; currently, a DSP driven design is required.
In this part of the circuit, a "power converter" is employed on each phase to control the voltage and current output supplied to the energy storage portions of the UPS (e.g., DC Link, ultracapacitors and/or batteries). If a higher voltage level is desired, a Boost Converter (or "step-up" converter) is employed to increase the voltage at the DC Link (compared to the average voltage level at the output of the rectifier/PFC stage). If a lower voltage level is desired, a Buck Converter (or "step-down" converter) is employed to decrease the voltage at the DC Link (compared to the average voltage level at the output of the rectifier/PFC stage). If the DC Link voltage is intended to vary both higher and lower than the average level at the rectifier/PFC output, then a Buck-Boost converter is employed, can supply lower and higher voltages depending on the demands of the aggregate load).
In any case, the power converter stage uses switching power semiconductors (e.g., IGBTs, MOSFETS), diodes, power passive components (e.g., capacitors, inductors, resisters), and feedback/control circuitry to step-up, step-down, or otherwise regulate the voltage and current outputs which are supplied to the next UPS stage, the DC Link and other energy storage stages.
This portion of the UPS circuit is where the regulated voltage and current waveforms are filtered (i.e. “ripple” is removed) and the energy from all three phases is stored as “DC power”. Note also that the "DC Link voltage" is also called the "bus voltage".
The DC Link is usually a large capacitor bank made up of either large aluminum electrolytic or power film capacitors, or a combination thereof. These highly specialized capacitors need to be rated to withstand high ripple currents, high temperatures (in some cases), and the high “bus voltage.” They require a high capacitance to volume ratio, high stability (electrical, mechanical, and thermal), low dissipation factor (i.e., low ESR), low ESL, very high insulation resistance, and long working life expectancy.
Richardson RFPD provides one of the most diverse DC Link capacitor product assortments in the market today. Because each inverter application is different, and because it is important to choose the right components for each circuit, we supply DC Link capacitors from Cornell Dubilier (CDE), KEMET, Kendeil, Spectrum Advanced Specialty Products, United Chemi-Con, and Vishay.
Inverter (with Gate Drivers)
The inverter is powered primarily from DC Link power. If DC Link power begins to droop, then the inverter receives power from other stored energy systems, such as ultracapacitors or batteries (or other longer term energy sources such as a flywheel). In any case, it is the inverter stage that converts DC energy back into clean, conditioned, sinusoidal 3-phase (if desired) AC power. The inverter switches at high frequencies (typically 20 - 100 kHz), and is generally made-up of 6-12 high-power semiconductor devices (e.g., IGBTs, diodes, thyristor/SCRs), or it could employ integrated inverter modules (depending on the specific application).
Richardson RFPD provides one of the most complete offerings of power semiconductor inverter devices in the industry. We offer gate drivers, diodes, IGBTs, MOSFETs, thyristor/SCRs, power IPM transistors, power semiconductor assemblies, and integrated power modules from key suppliers, such as: Dynex, EDI, Microsemi, Mitsubishi, Powerex, Semikron, and Vincotech. In addition, we can provide customized gate driver solutions under our own brand name.
To create the 3-phase sinusoidal voltage and current waveforms, the inverter devices are switched on and off rapidly. As a result, voltage and current spikes are developed and these must be carefully removed through a process called "snubbering." A snubber circuit is typically a carefully chosen capacitor/resistor combination.
The object of the snubber circuit is to eliminate the voltage transient and ringing that occurs when an inverter "switch" opens (i.e., IGBT turns off). The snubber provides an alternate path for the instantaneous current flowing through the inverter device's intrinsic leakage inductance.
Snubber circuits in UPS systems can provide any/all of these three valuable functions:
- Shape the load line of a bipolar switching transistor (IGBT) to help keep it in its safe operating area.
- Remove energy from a switching transistor and dissipate the energy in a resistor to reduce junction temperature.
- Reduce ringing to limit the peak voltage on a switching transistor or rectifying diode and to reduce EMI by reducing emissions and lowering their frequency.
To learn more about designing an RC snubber circuit, please click here.
Current/Voltage Sensors and PWM Control
To complete the inverter circuit, certain feedback mechanisms (sensors) and control circuitry (typically PWM, or pulse-width-modulation controls) are used. The current and voltage sensors make sure that the system is providing the right level of current and voltage to the load. If a level begins to drop too low, the sensor feeds this information to the control circuitry and it directs the inverter to increase the power production. Alternately, if a level begins to rise too high, the circuitry adjusts to produce a lower amount of power. If the circuitry is unable to adjust, then the control signals indicate the need for maintenance on the system.
We offer Hall Effect Current and Voltage Transducers (sensors) from Tamura Corporation. We also carry PWM control devices from CT-Concepts, Powerex, and Semikron.
The final stage of any power conversion circuit (i.e., Inverter, UPS system) is the place where relatively clean, controlled power is delivered to the "load." The load in the case of a 3-phase UPS system is typically all of the machines in one section of a factory, all of the servers and computers in a data center, or some other similar type of aggregate load. The output filter guarantees that unwanted (spurious) energy, either from the previous blocks of the UPS circuit or actually unwanted energy coming back in from the load, does not corrupt the inverted power. We feature output filter capacitors from Cornell Dubilier (CDE), KEMET, and United Chemi-Con (UCC).
Thermal Management Systems
Heat Sinks are employed mainly to increase the surface area available for heat transfer from high-power semiconductor devices (e.g., Power Rectifiers, Power Inverter Modules, IGBTs, MOSFETs, Thyristor Modules, Diode Modules), systematically reducing each device's external case temperature, as well as its internal junction temperature. We provide four different categories of heat sinks:
UPS Energy Storage Systems
The main energy storage system used in UPS systems of the past was strictly lead-acid batteries. Modern day UPS system designs are beginning to include ultracapacitors to mitigate the effects of brief power disturbances, specifically addressing the short-term ride through and bridge power requirements in mission-critical installations (data centers, hospitals, factories, and telecommunication facilities). We offer a complete line of ultracapacitor solutions from Maxwell Technologies, Inc. Maxwell's ultracapacitors have already been deployed in over 35,000 UPS systems and counting. To learn more about Maxwell ultracapacitors, please click here.
In addition to batteries and ultracapacitors, other energy storage systems employed by today's UPS systems include flywheel systems, fuel cells, superconducting magnets, and fossil-fuel micro-turbine generators.
- Snubber Capacitor Application Guide (located in our Design Resource Center)
- Powerex: Using IGBT and IPM Modules (General Considerations - Design Guide)
- More Information about Ultracapacitors
- Wikipedia: Uninterruptible Power Supply
- Mitsubishi Electric: Uninterruptible Power Supplies - 9800A Brochure
Other Related Information and References
- Allowable Harmonic Content (Mains) Standards
- Economics of Power Factor Correction in Large Facilities - Pacific Gas and Electric Co.
- Comparative Evaluation of Control Techniques for a Three-Phase, Three switch Buck-Type AC-to-DC PWM Converter System
- 35 kW Active Rectifier with Integrated Power Modules
- Galvanically Isolated 3 Phase PFC Topologies
Aron Levy: Galvanically Isolated 3 Phase PFC Topologies