Evaluating power quality for mission critical facilities

Mission critical and data center facilities that use sensitive power electronics have small tolerances for power source deviations. Power quality and the mitigation of problems affecting power quality are very important to the continuous operation of the facility

By Kenneth Kutsmeda, PE, LEED AP, Jacobs, Philadelphia September 25, 2019

Learning objectives

  • Learn about what causes power source deviations (poor power quality). 
  • Understand the impacts of power quality on facilities like mission critical and data centers. 
  • Review different mitigation techniques that will help power quality.

IEEE Standard 1100 defines power quality as “the concept of powering and grounding sensitive equipment in a matter that is suitable to the operation of that equipment.” Power quality can be summarized as the compatibility between what comes out of an electric outlet and the load that is plugged into it,” according to Alexandra Von Meier in the bookElectric power systems: a conceptual introduction. The electrical load or equipment that is connected to an electrical circuit has a certain power tolerance range that is required to allow it to function properly. Any power source deviations outside that power tolerance range can affect the operation of that equipment.  

Power source deviations or disturbances most commonly include the following: 

  • Interruptions. 
  • Voltage variations. 
  • Waveform variations. 

Some equipment — motors, heaters and incandescent lighting, for example — have a wider power tolerance range and can accept larger deviations in power (poor power quality). Reduction in voltage may make the motor operate at slower speeds or the incandescent light operate with less lumens. Other equipment, such as power electronics and information technology equipment have a much smaller power tolerance range. Slight deviations in power can cause this type of equipment to malfunction, fail prematurely or not operate at all. 

Many businesses and facilities rely on sensitive power electronics and IT equipment to properly operate their industries. As this trend continues to grow more important, having good power quality and keeping the power source within the tolerance range of the equipment becomes paramount. Poor quality power can cause malfunctions and failures that could severely affect the operation of the facility 

This is particularly true with mission critical type facilities that relay on the power electronics for safety or data centers where the entire facility is made up of IT equipment 

Interruptions 

The most commonly recognized power source deviation is the interruption of power or power outage. This is when there is a break in the supply of power and the load or equipment becomes de-energized. Obviously, the total loss of power is outside the tolerance range for any piece of electrical equipment causing that piece of equipment not to operate. Depending on the type of outage, the interruption can last a few seconds or potentially for multiple days.  

Another issue with interruptions is the unexpected shutdown of equipment. Older facilities and processing facilities have operating systems that must go through a shutdown sequence to make sure the running processes have correctly terminated before turning off. Sudden shutdowns can stop processes midway, causing computational issues and/or physical damage. Newer facilities like cloud data centers provide redundancy at the software level not the hardware level so that processes continue to function during an unexpected outage.  

Interruptions can be caused by events external to the facility such as damage to utility transformer or transmission line. Interruptions also can be caused by internal eventsuch as a fault that trips open a protection device, damage to distribution equipment/wiring and required maintenance. 

Voltage variations 

Another type of power source deviation is voltage variations. Voltage variations are brief instances where the voltage increases or decreases beyond the normal level. If the voltage increases or decreases outside the voltage tolerance range of the equipment it will affect the operation of that equipment.  

The following are different types of voltage variations that can occur. 

  • Voltage sag: Decrease or reduction in root mean square voltage below nominal voltage, typically lasting from a cycle to a few seconds. 
  • Voltage swell: Increase or escalation in RMS voltage above nominal voltage, typically lasting from a cycle to a few seconds. 
  • Voltage flicker: Random or repetitive variations in RMS voltage. 
  • Voltage spikes/surges: High increases of voltage for very short periods of time. 
  • Undervoltage/overvoltage: Small decreases and increases in voltage for longer periods of time. 

Voltage sags are the most common power disturbance. A voltage sag occurs when the RMS voltage decreases to 10% to 90% below the nominal voltage for less than a couple of seconds. Voltage sags are typically the result of an abrupt increase in load such as large motors starting or electric heaters turning on. They also can be caused by short circuits and faults that draw large amounts of current or rapid increases in impedance caused by loose connections.  

Voltage swells are the exact opposite of voltage sags. A voltage swell occurs when the RMS voltage increases to 10% to 80% above the nominal voltage for less than a couple of seconds. Voltage swells typically are caused when there is an abrupt reduction in load on a circuit. Voltage swells also can be caused by a damaged or loose neutral connection. Majority of voltage sags and swells are caused or generated by events occurring internal to the facility.  

Voltage flicker occurs when there’s random or repetitive variations in RMS voltage between 90% and 110% nominal voltage. Voltage flicker is most noticeable with lighting. The variation in voltage makes the lights flicker or have flashes of brightness. Voltage flicker is caused by machinery or motors with rapid fluctuations in load such as large motors during startup, machines that use static frequency converters and elevators. Voltage flickering also can be caused by loose connections. 

Voltage spikes/surges are both when high increases of voltage occur for very short periods of time. Spikes last 3 nanoseconds or less, while surges last more than 3 nanoseconds. Internally voltage spikes/surges can be caused by short circuits, tripped circuit breakers and by buildup of static electricity that suddenly discharges. Externally, voltage spikes/surges are caused by lightning strikes, damaged power lines and utility power outages.  

Undervoltage or brownout occurs when the voltage drops to below 90% of nominal voltage for more than a minute. This is typically done by the utility to decrease demand and reduce load during an emergency rather than cause a power outage or blackout. Overvoltage is the opposite and occurs when voltage rises to above 110% of nominal voltage for more than a minute. Overvoltage generally occurs due to poor regulation of power or malfunctions of the electrical power distribution system. 

As indicated previously, sags do not disturb motors, heaters, incandescent and fluorescent lighting. Power electronics equipment, however, has insufficient internal energy storage to ride through sags in voltage. If the voltage sag is low enough or for a long enough duration, it could fall outside the tolerance of the equipment causing it to malfunction, fail or turn off. Voltage spikes can cause internal damage to power electronics that are not designed to withstand that kind of influx of power. 

Waveform variations 

Another type of power source deviation is waveform variations. The oscillation of voltage and current ideally follows in the form of a sinusoidal shape. Waveform variation occurs when the voltage or current waveform is altered from a sinusoidal shapeThe distortion to the voltage and current waveform is often described as harmonics. Harmonics refers to a component of the waveform that oscillates more rapid than nominal frequency.  

Harmonic have frequencies that are integer multipliers of the waveform’s fundamental frequency (60 hertz fundamentalsecond harmonic = 120 hertz; third harmonic =180 hertz; and fourth harmonic = 240 hertz). Total harmonic distortion is the summation of all the harmonic components of the voltage or current waveform compared to the fundamental component.  

Linear loads like household appliances draw current that is sinusoidal in nature and do not distort the waveforms (no harmonics). Nonlinear loads such as switchmode power supplies, variable speed drives, computers and uninterruptible power supplies draw current in high-amplitude short pulses that create significant distortion in the electrical current and voltage wave shape (harmonics). 

Harmonics or distortions in the waveform will travel back into the power source and affect other equipment connected to that same source. Most power sources can accommodate a certain level of harmonic currents, however as those harmonic currents become more significant, the following issues can occur. 

  • Overheating of electrical distribution equipment and cables. 
  • Equipment malfunctions. 
  • Higher voltages and circulating currents. 
  • Vibrations and buzzing. 
  • False tripping of protection devices. 
  • Generator failures. 
  • Increased energy losses and overheating causing component failure. 

Tolerances and limits  

One of the most widely used tools for determining the power tolerance ranges for IT equipment is the Information Technology Industry Council curve. The ITIC curve (previously the CBEMA curvewas published by the technical committee of ITIC. The curve illustrates the voltage envelope and tolerances (magnitude and duration of voltage variations) that can typically be tolerated by most IT equipment. The vertical axis represents the percentage of voltage and the horizontal axis represents time. In the middle of the curve is the “ninterruption in function region. In this area computers, servers, power distribution units, programable logic controllers and telecommunication equipment operate properly.  

Above that middle acceptable region is the “prohibited region,” which involves the equipment tolerance for excessive voltages. Voltage spikes or surges in this region can cause damage to the IT equipment. At the bottom of the curve is the “ndamage region,” which represents the equipment tolerance for reduction in voltage. Voltage sags and interruptions in this area can cause the equipment to stop functioning (shut off), but it should not cause damage to the equipment. The goal of the curve is to be a reference for determining the withstand capabilities of various IT loads for protection from power quality issues such as voltage variations. 

National standards do not currently exist for enforcing THD limits. IEEE 519-1992: IEEE Recommended Practices and Requirements for Harmonic Control in Electrical Power Systems is a useful document for understanding harmonics. The standard does make recommendations for acceptable levels of harmonic distortion: Computers and allied equipment, such as programmable controllers, frequently require ac sources that have no more than 5% harmonic voltage distortion factor [THD], with the largest single harmonic being no more than 3% of the fundamental voltage. Higher levels of harmonics result in erratic, sometimes subtle, malfunctions of the equipment that can, in some cases, have serious consequences.  

It should be noted that the harmonic values in the IEEE standard are suggestions. It is recommended that you try to keep the THD values as low as possible to ensure proper operation of equipment and extend the life expectancy of the equipment. 

Solutions  

UPSs are commonly used for improving power quality in mission critical type facilities (call enters, emergency operation centers, etc.) that rely highly on their power electronic equipment and data centers that have large amounts of IT equipment. The primary function of the UPS system is to provide backup power during a power outage or interruption.  

The backup power is provided from direct current power storage device, usually batteries or kinetic energy flywheels. For some facilities, the UPS can provide enough power/time to allow the systems to go through a proper shutdown sequence. In mission critical facilities or for longer outages, the UPS generally provides the power bridge until the facilities generators can start and supply power. 

In addition to eliminating the interruption deviation, the UPS can also help with waveform variations and voltage variationsPower that flows through a doubleconversion type UPS is also often referred to as conditioned power. The UPS rectifies (breaks down) the alternating current power into DC power and then inverts (remanufactures) the DC power back into a conditioned AC sine wave.  

Because the AC sine wave is remanufactured, it can be higher quality than the original wave, therefore eliminating any variations. The UPS system can also eliminate any voltage sags or undervoltage in the system by drawing additional power/voltage from the batteries to compensate for the decrease in voltage. The goal of the UPS is to provide uninterruptible power with a conditioned sinusoidal waveform at a constant nondeviating voltage. 

One issue the older thyristorbased UPS systems had was the amount of harmonic distortion it caused on the upstream systems. The 6-pulse thyristor-based rectifier had a 30% current THD and the 12-pulse thyristor-based rectifier had a 12% current total harmonic distortionThis caused upstream equipment such as generators to be oversized to compensate. Most doubleconversion UPS systems today are insulated gate bipolar transistorbased that have reduced the current total harmonic distortion to 3%. 

Surge protection devices are electrical devices installed to protect against voltage surges and spikes in the electrical power system. The SPD device attempts to limit the voltage supplied to an electric device by either blocking or shorting current to reduce the voltage below a certain threshold.  

Lightning arrestors also can be used to protect against voltage spikes caused by lightning. Lightning arrestors typically have a high voltage terminal and a ground terminal. During a lightning strike, the current travels along the conductor to the arrestor. At the arrestor, the surge current is then diverted through the arrestor to the ground. 

Harmonic mitigation 

Two methods for compensating for harmonics include using 200% neutrals and K-rated transformers. The theoretical maximum current the triplen harmonic can produce is 173% of the phase current. The 200% neutral is used to compensate for that additional heat caused by the harmonic current heat in the conductor. K-rated transformers are used in the same manner, to compensate or manage the additional heat generated by the harmonic currents.  

Standard transformers are not designed to handle the high harmonic currents produced by nonlinear loads. K rated transformers are designed to reduce the heating effects of harmonic currents produced by nonlinear loads. UL developed a rating system described in UL 1561 to indicate the capability of the transformer to handle harmonic loads. Typical K ratings include K4 (discharge lighting, programmable logic controllers, solid state controls), K13 (telecom/IT equipment, health care, testing equipment) and K20 (mainframe computers, solid state motor drives, operating room equipment). 

Methods for reducing harmonics in the power system include passive filters, active filters and harmonic mitigating transformers. Passive filters are based on a combination of inductors, capacitors and resistors that correct the phase current or convert nonlinear loads to linear. These types of filters do not rely on external power sources and do not contain active components such as transistors. 

Active filters are devices connected in parallel with the systems load to be corrected. These types of filters are more complex. They actively monitor the nonlinear currents and generate or inject currents opposite the harmonics to cancel the harmonic current. Because they can be connected in parallel, the active filter can be a flexible, highperformance and cost-effective solutions used to mitigate power quality issues. 

The harmonic mitigating transformer is a phaseshifting transformer that uses electromagnetic flux cancellation to treat harmonics instead of filters and capacitors. Unlike K-rated transformers that compensate for harmonics, the HMT transformer eliminates the harmonic. 

Power factor corrected  

To ensure that electronic devices did not have a significant cumulative effect on the power system, standards like International Electrotechnical Commission 61000-3-2 were established to set limits on power factor degradation and harmonic distortion introduced by power supplies. In data centers where there are large amounts of nonlinear IT equipment this standard forced the computer vendors have transition from older switched mode power supplies to power factor corrected power supply technology.  

The switched mode power supplies created distortion in the current and voltage waveforms causing high harmonicsThe goal of the PFC power supply is to make the power factor as close to one as possible, where the current waveform is proportional to the voltage waveform. The PFC uses filters and/or electronic switching elements to force the input AC current to be sinusoidal with minimal distortion and in phase with the input voltage. By minimizing the distortion of the waveforms, the PFC power supply reduces the harmonics 

Other types of facilities may use capacitor banks to correct power factor and improve power quality by offsetting inductive loads like large electrical motors. 

Other solutions 

Other solutions for managing power quality may include: 

  • Measure and locate the power quality. Two important factors in providing high power quality is determining if you have any power quality issues and what type of power quality problem exists. Once that is determined a proper plan can be derived to mitigate those power quality problems. Installing continuous power quality meters throughout the system can help detect power quality problems. The power quality meter should include the ability to capture and view waveforms, detect disturbances like voltage sags, voltage swells measure harmonic power flow and provide alarms when measurements are outside a set tolerance range. 
  • Power distribution design: Separate sensitive loads from poor power quality source. 
  • Specify variable frequency drives with line reactors or 12pulse front ends 
  • Design generators with proper pitch or alternator designs to handle power quality or even leading power factor data center loads. 
  • Select grounding systems that help reduce power quality issues. 

Power quality can mean different things to different operations. Facilities like mission critical and data center facilities with sensitive electronics and IT equipment have small tolerances for deviations in powerThese types of facilities are called upon to operate 24/7 and rely on high quality power to keep the systems operating normally. 


Author Bio: Kenneth Kutsmeda is an engineering design principal at Jacobs. For 20 years, he has been responsible for engineering, designing, and commissioning power distribution systems for mission critical facilities. He is a member of the Consulting-Specifying Engineer editorial advisory board and a 2010 40 Under 40 winner.