Although you can select from a wide variety of power conditioning products, no single product can solve all your power quality problems. To get the best results, you must define the problem, verify source compatibility and load compatibility, and evaluate application criteria.

One of the most commonly cited causes of electronic equipment failure is electrical power disturbances. The widespread presence of these variations in today's electrical power supply requires you to pay special attention to power quality.

You can interpret the resulting power failure problems as the difference between the quality of power supplied to sensitive electronic equipment and the quality of the power it requires to operate reliably. Using this viewpoint, you can resolve power quality problems in one of the following three ways:

  • Reduce the variations in the power supply.

  • Improve the load equipment's tolerance to these variations.

  • Insert some interface, such as power conditioning equipment, between the electrical supply and the sensitive electronic equipment.

Of course, practicality and economics will determine the extent to which you can use any of these options, and they will most likely rule out two of them. For example, there's not much you can do to help your utility reduce variations in the power it supplies to you. Not only that, you probably can't afford to have manufacturers improve their sensitive electronic equipment to cope with your site-specific problems. Taking proactive steps on your end is probably your best bet. Consider the following steps.

Step 1: Define the problem

You have several methods at your disposal for defining a power quality problem, one of which consists of performing the following activities:

  • Conduct a thorough on-site investigation, which includes inspecting all wiring and grounding systems for errors.

  • Monitor the power supply for power disturbances.

  • Investigate equipment sensitivity to power disturbances.

  • Determine the load disruption and consequential effects.

By using this method, you can define the power quality problem, develop alternate solutions, and then choose the optimal fix.

Another option is to buy power conditioning equipment that can correct the perceived power quality problems without requiring you to conduct an on-site investigation.

Although the first method appears to be the best, it may not be viable because of practical limitations. For example, on smaller installations you may not be able to justify the time and expense associated with this method. An example of this would be trying to capture infrequent disturbances, which would require monitoring over an extended period of time.

In addition, the first method tends to solve only observed problems. This may cause you to miss unobserved or potential problems in your proposed solution.

Consider the planning necessary for a new facility. There's no site to investigate, so instead of a thorough on-site investigation, you would need to implement power quality solutions to solve potential or perceived problems on a preventive basis. On the other hand, using the first method may help you avoid a more expensive power conditioning solution by correcting a relatively inexpensive wiring error, such as a loose connection or reversed neutral and ground wires.

Step 2: Verify source compatibility

You must ensure that the power conditioning device you're planning to use is compatible with its intended power source so that it will operate properly. This will also help in avoiding interference with the operation of other loads connected to the same power source.

Along with the obvious considerations like proper input voltage levels, number of phases, and frequency, you should also focus on more subtle ones, such as:

  • Tolerance to expected sags, swells, and surges

  • Limited start-up or in-rush currents to prevent voltage sags

  • Limited harmonic input current distortion

  • Limited notching

If you don't, the proposed power conditioning device may cause power disturbances for other loads in your facility. For example, while a static UPS provides interruption protection for its load, its rectifier may produce voltage notches that can cause PCs, electronic clocks, and other sensitive loads in the building to malfunction.

Step 3: Verify load compatibility

This may seem obvious to you, but to ensure load compatibility, you need to know the requirements of the intended load so you can select a power conditioning device with compatible output performance. For example, if the intended load's allowable input voltage range is +6% to -10% of nominal, you obviously wouldn't need a precision voltage regulator to maintain output voltage to within ±1%.

Certainly, you must look at fundamental operating parameters like size, voltage, frequency, and number of phases, but sizing depends on the type of conditioning device. For example, you size a TVSS device, which is connected in parallel with the protected circuit, according to the maximum expected surge current it may be required to conduct. On the other hand, you size a voltage regulator to operate and conduct the intended load's total current.

Depending upon the type of proposed conditioning device, you would size it based on the expected maximum continuous full-load current only. But you have to avoid undesirable load source interactions like those shown in the Figure on page 18. To do this, you'll need a more detailed profile of the intended load and an understanding of the proposed conditioning device's response to this load. For example, an intended load that experiences large current variations may cause a proposed conditioner's output voltage to vary significantly. You may have to oversize the conditioner or select one less affected by load current variations.

Likewise, if the intended load's current is nonsinusoidal, you again may have to oversize the conditioner or select one specifically designed to accommodate the heating effects of nonsinusoidal current without excessive output voltage distortion. Nonlinear loads with high peak currents also require special attention.

Step 4: Evaluate application criteria

Power conditioning technology can protect against various power disturbances, but to get the best protection performance from each type, you have to understand its respective application factors. Let's take a look at some conditioning devices and their respective application criteria.

TVSS devices. Before settling on such a device, make sure you consider the following:

  • The location of the device itself, which influences the expected surge activity to which the TVSS will be subjected

  • Required energy rating or surge-current handling capability, which is a function of its location

  • Clamping or protection voltage levels, which depend on the load equipment's susceptibility; devices with lower clamping voltages aren't necessarily better

  • Modes of protection, which also depend on the susceptibility of the intended load

  • Maximum continuous operating voltage (MCOV)

  • Coordination with other TVSS devices located elsewhere on the power system

Isolation transformers. You'll want to consider the following before making a selection:

  • Required input and output voltages of the transformer

  • Voltage compensation tap configurations

  • Operating frequency

  • kVA capacity

  • Type of interwinding electrostatic shielding, which determines the level of common mode noise attenuation

  • Nonlinear load capability, which determines the degree to which the transformer won't overheat

Voltage regulators. These devices require similar considerations, but you must also address the following issues:

  • Input voltage regulation range

  • Output voltage regulation, which must be compatible with the intended load's input requirements

  • Output voltage imbalance (in the case of 3-phase systems)

  • Overload capability

  • Time response

  • Response to expected load current changes

Finally, you want to avoid undesirable load-source interactions. For example, suppose the intended load's current variations cause the proposed conditioning device's voltage to vary. This, in turn, will cause the intended load's current to vary, which can cause instability and oscillation of the proposed conditioner's output voltage.

Motor-generator (M-G) sets. Make sure you take into account these application factors:

  • Output frequency, which may be different than the input frequency

  • Frequency regulation for input voltage and output load changes

  • Stored inertial energy (ride-through) capability

  • Provisions to bypass the M-G set for maintenance

Basically, an M-G set can effectively isolate its intended load from the utility source, and vice versa. In doing so, the M-G set appears to its load as a relatively higher source impedance, as opposed to the normal low-impedance utility source. This can make load-source interactions more likely. For example, the harmonic currents of nonlinear loads are more likely to cause the generator output voltage to become distorted. Specifying a maximum allowable subtransient reactance on the alternator will help attain the best practical M-G set performance. Load current changes are also more likely to cause output voltage or frequency variations.

Offline UPS. Sometimes referred to as a standby power supply (SPS), this device operates normally with its inverter in a standby mode, so that it and the UPS' battery provide power to the load only when the UPS senses a disturbance on its normal input power source that is outside the device's input tolerance. Some UPS manufacturers have applied the term “line interactive” to describe their standby UPS systems that also regulate output voltage.

You must understand the characteristics of this type of UPS and ensure its compatibility with the intended load's requirements. For example, there will be a time delay between the beginning of a power disturbance and the transfer and operation of the load on inverter power. Transfer times for an offline UPS can be less than ¼ cycle, which has no effect on most electronic power supplies. However, some offline UPS transfer times are longer and may affect sensitive loads. Also, few offline UPS inverters generate a regulated sine wave output, but rather a square wave or quasi-square wave. Again, most electronic power supplies can tolerate these kinds of waveforms, particularly for the short time periods on inverter/battery operation.

Since an offline UPS can only correct for supply voltage or frequency variations by transferring to inverter/battery operation, you may experience excessive battery cycling and possibly shortened battery life if your site experiences frequent power disturbances outside the input tolerance of the offline UPS. Also, battery cycling can be particularly problematic at sites with standby engine-generators because they're more prone to frequency variations.

Online UPS. The two types of online UPS systems are rotary and static. A rotary UPS uses some form of an M-G set to provide uninterruptible power. The static UPS, on the other hand, has no moving parts and typically uses power semiconductors.

Static UPS systems have several topologies that provide online functionality, including line-interactive, single-conversion, and double-conversion. In the line-interactive topology, line power isn't converted into DC. Instead, it's fed directly to the critical load through an inductor or transformer. Regulation and continuous power is achieved through the use of inverter switching elements in combination with inverter magnetic components, such as inductors, linear transformers, or ferroresonant transformers.

Single-conversion UPS systems use the incoming line during normal operation to power the critical load, either through a transformer or in conjunction with some series impedance. This topology features a bi-directional inverter that filters and regulates the output voltage while maintaining battery charge.

Double-conversion UPS systems allow a break in the output while detecting a load and turning on its inverter. The transfer time typically ranges from 2 milliseconds to 4 milliseconds. The CBEMA and ITIC Curves recommend a maximum transfer time of 8.3 milliseconds.

You can apply smaller static UPS products, which are self-contained (including the battery, input, and output power connections), with similar application considerations as outlined for voltage regulators. Larger UPS applications, however, are complicated. You should retain a qualified consultant for these large applications.

The sooner you accept that you're on your own to reduce power variations in your electrical system, the sooner you can get started choosing the right power conditioning device for the job.

This article is adapted from an EPRI-PEAC Corp. paper titled “Power Conditioning Equipment: An Overview.”