Mass-produced UPSs first appeared on the market in the 1970s, essentially to meet the needs of large computer systems users by maintaining continuity and quality in the supply of electrical power. They then evolved to adapt to the vast increase in the number of sensitive loads and their diversification, due to the explosion in digital technology. As a result, UPS systems were modified to meet the needs of applications such as mini and then microcomputers, industrial processes, instrumentation, telecommunications, and other sensitive electronic applications.

Adaptation to markets undergoing such rapid change resulted in numerous technological innovations and an extension of power ratings at both lower and higher power extremes, to meet the respective needs of microcomputers as well as the vital applications of digital telecommunication systems. As a result, UPS terminology today covers very diverse products ranging from a few hundred VA up to several million VA.

During this process of adaptation and change, UPS topologies became more diverse, depending on the application, its criticality and power level. At the same time, product descriptions were often confusing and even misleading to consumers.

The Original "On-line" UPS

Back in the 1970s, the term "on-line" UPS commonly referred to a UPS topology where AC input went through an input rectifier/charger that supported an inverter and charged batteries which supplied backup energy in the event of a utility power outage. An inverter provided power continuously with a consistently high level of quality (frequency, voltage, etc. within tight tolerances) while a static switch (static bypass) capable of transferring the load to bypass power without a break in the supply of power, assured that the load would receive power directly from the utility in the event of UPS fault.

The standardization bodies observed that the term "on-line”, which taken literally means "on utility power", does not represent the true situation in this topology because the load is supplied by the inverter and not directly by the AC input. The term was nonetheless fairly rapidly construed to mean a UPS supplying a load continuously via an inverter connected in series with the AC utility.

Enter the " Off-line " UPS

In the 1980s, the types of loads and the range of power ratings increased substantially and "off-line" UPSs were developed, the term "off-line" being simply the opposite of "on-line". This term is used to describe a topology in which the inverter is not connected in series with the AC input, but rather in parallel, in a passive standby configuration. It does not operate continuously, but only when the AC-input supply voltage goes outside tolerances.

This topology typically includes a filter with loosely defined functionality sometimes misrepresented as voltage-regulation. It does not include a static switch (static bypass), which can result in load switching times that are unacceptable for many sensitive applications.

Again, the standardization bodies observed that the term "off-line", which taken literally means "not on utility power", does not represent the true situation in this topology. The load is primarily supplied directly with AC-input power supplied from the utility with the inverter operating only sporadically in the event of a problem with the utility power. Users nonetheless have adopted this topology and the term "off-line", primarily for low-power UPS ( £ 2 kVA).

Other terms

In the 1990s, further techniques were developed. The term "line-interactive" is used for UPSs implementing reversible inverters. However, the uncertainty created by the many versions of topologies exposed consumers to abusive tactics. For example, some UPSs were termed "in-line" and in some cases, the term "on-line" was utilized in a very misleading manner.

IEC 62040-3: Clear definition of UPS topologies. Over time, this situation created ambiguities and even outright fraud that hurt consumers and penalized reputable manufacturers. The need to establish a standard with clearly defined terms became unavoidable and that is why in 1999, an IEC (International Electrotechnical Commission) working group drafted a standard on the types of UPS and methods used to measure their performance. CENELEC (European standardization committee) adopted the contents of the IEC standard.

These efforts produced IEC 62040-3 standard and its European equivalent EN 62040-3, which clearly define the standardized types of UPS and their performance. These types of UPSs are presented here, with the advantages and disadvantages of each type and their field of application.

Definitions

Standards IEC 62040-3 and EN 62040-3 distinguish the following three types of UPS topologies:

• passive standby;
• line-interactive;
• double-conversion.

These terms refer to UPS operation with respect to utility power, i.e. the distribution system upstream of the UPS. The standards define the following terms for input power:

• primary power: power normally continuously available which is usually supplied by an electrical utility company, but sometimes by the user's own generation plant.
• standby power: power intended to replace primary power in the event of primary power failure.
• Practically speaking, a UPS has one or two inputs:
• the normal AC input (sometimes called Input 1) is supplied with primary power;
• the bypass AC input (sometimes called Input 2), when it exists, may also be supplied with primary power or, where possible, with back up power (for instance a separate cable from the same main low-voltage switchboard).

Passive-standby UPS

The inverter is connected in parallel and acts simply to backup utility power (Fig1).

This topology is a compromise between an acceptable level of protection against disturbances and cost. Essentially an implementation of the "off-line" topology mentioned earlier, the standards suggest that the term "passive standby" is a more accurate description of the operating principle.

Normal mode

• The load is supplied with utility AC-input supply, generally via a filter/conditioner that eliminates certain disturbances and can sometimes provide voltage regulation. The standards do not mention this filter and speak simply of a "UPS switch". They do stipulate that " additional devices may be incorporated to provide power conditioning, e.g. ferro-resonant or automatic tap changing transformers.”
• The inverter is on passive standby.

Stored-energy mode

• When the AC-input supply voltage goes outside the specified tolerances or fails, the battery and inverter ensure power continuity to the load with a very short switching time (generally < 10 ms). While the standards do not mention a specific transfer time, they do stipulate " the load [is] transferred to the inverter directly or via the UPS switch (which may be electronic or electro-mechanical).”
• The UPS continues to operate on battery power for the duration of the backup time or until the AC-input voltage returns to specified tolerances, at which point the UPS returns to its normal mode.

Advantages

• simple design.
• low cost
• small size

Disadvantages

• no real isolation of the load from the upstream distribution system.
• long switching time. Without continuous inverter operation or a true static switch, the time needed to transfer the load to the inverter is relatively long. Though acceptable for certain applications (e.g. stand-alone desktop computers), this level of performance is not appropriate for large and complex groups of sensitive loads (e.g. large computer centers, telephone exchanges).
• no regulation of the output voltage.
• no regulation of the output frequency, which depends on the AC-input and can cause problems in the case of operation with generators or other less stable inputs.

Applications (300–1200VA single-phase)

• Small office and home office – PCs, Modems, Gateways. Practically speaking, due to its disadvantages, this UPS topology is used only for low power ratings.

Line-interactive UPS

• The inverter is connected in parallel and provides backup to utility power. It also charges the battery. Through its reversible operation, it interacts with utility power (Fig. 2)
• The terms “Boost/Buck”, “AVR” (automatic voltage regulation) and “Delta Conversion” topologies belong to the line-interactive category.

The standards define three operating modes.

Normal mode

• The load is supplied with “conditioned input power” via a parallel connection of the UPS inverter with the AC input. The inverter provides output voltage conditioning and/or battery charging.
• The output frequency is dependent upon the AC input frequency.

Stored-energy mode

• When the AC-input supply voltage exceeds UPS preset tolerances or fails, the inverter and battery maintain continuity of power to load. The switch (e.g. a static switch) disconnects the AC input supply to prevent backfeed from the inverter.
• The UPS runs in stored-energy mode for the duration of the available backup time or until the AC-input returns to UPS design tolerances, at which point the UPS returns to normal mode of operation.

Bypass mode

This type of UPS may include a maintenance bypass. In the event of a UPS internal malfunction, the load may be transferred to the bypass input via the maintenance bypass.

Advantages

• the cost may be lower than a double-conversion UPS of equal power rating
• simple design
• low cost
• small size

Disadvantages

• no isolation of the load from the upstream distribution system.
• no regulation of the output frequency, which depends on the AC input supply.
• poor protection against spikes and overvoltage
• output-voltage conditioning is mediocre because the inverter is not connected in series with the AC input. The standard speaks of "conditioned power" via a parallel connection of the AC input and the UPS inverter. Conditioning is however limited due to the reversible operation of the inverter.
• Decreased efficiency when operating with non linear loads, or when the AC input voltage is different from the output voltage

Applications (500–3000VA single-phase (possibly up to 10kVA)

• Enterprise networking
• PC servers, workstations and storage
• Unix servers
• Network equipment

This topology is poorly suited for sensitive loads with medium to high power ratings because it does not provide frequency regulation. For this reason, it is almost never used at such ratings.

Double-conversion UPS

The inverter is connected in series between the AC input and the load. Power for the load flows continuously through the inverter (Fig. 3)

The standards define three operating modes.

Normal mode

The load is continuously supplied via the rectifier/charger - inverter combination that performs a double conversion of AC to DC and back to AC again, hence the name of the topology.

Stored-energy mode

• When the AC-input supply voltage exceeds UPS preset tolerances or fails, the inverter and battery continue to support load power.
• The UPS runs in stored-energy mode for the duration of the stored-energy time or until the AC-input supply returns to preset tolerances, at which point the UPS returns to normal mode.
• Bypass mode
• This type of UPS is generally equipped with a static bypass (often called a static switch).

Advantages

• continuous protection of load from the inverter, whether it originates from the AC-input supply or the battery.
• isolation of the load from the upstream distribution system eliminates transmission of any upstream fluctuations such as spikes and over voltages to the load.
• very wide input-voltage tolerances and precise regulation of the output voltage.
• precise regulation of the output frequency and possibility of the UPS operation as a frequency converter (if configured for this purpose) by disabling the static switch.
• superior performance levels under steady-state and transient conditions.
• instantaneous transfer to stored-energy mode in the event of an AC-input supply failure.
• no-break transfer to bypass mode.
• manual bypass designed to facilitate maintenance.

Disadvantages

• higher price, generally offset in total cost of ownership by the numerous benefits.

Applications (10kVA to 800kVA single & three-phase)

• Critical Systems
• Enterprise Networks
• Data centers – Co location, ISPs
• Telecom: central offices, base stations
• Industrial: process control, manufacturing plants

This is the most complete topology in terms of load protection, regulation possibilities and performance levels. While it is essentially the "on-line" topology discussed earlier, the standards recommend use of the term "double conversion" as a much more accurate description of the operating principle.

This topology enables no-break in operation during load transfers from normal mode to the bypass mode and back, using the static switch. It also ensures total independence of the output voltage and frequency with respect to the input voltage and frequency.

Conclusion

The IEC 62040-3 and its European equivalent EN 62040-3 standard have put an end to a difficult situation for customers by defining three types of UPSs and the methods used to measure their performance levels.

The three basic UPS types are defined by their topology and operation. The standards recommend use of the terms below as they more accurately describe the respective UPS operating principles.

 

UPS Topology	Blackout protection	Voltage regulation	Noise isolation	Frequency regulation	Harmonic isolation
Passive standby	X
Line-interactive	X	X	X
Double-conversion	X	X	X	X	X