Let’s take a look on how to tame power quality problems

Effort to minimize power quality problems in our increasing microprocessor-based system starts by defining power quality and understanding how to recognize it.

Exactly what is power quality?  IEEE 1100, IEEE Recommended Practices for Powering and Grounding Sensitive Electronic Equipment known as the Emerald Book, defines it as powering and grounding sensitive electronics equipment in a manner that is suitable to the operation of that equipment.  Another definition could be the degree to which the power system affects the operation of electrical equipment.  The user generally grades power quality on the number and effects of problem they have associated power system, real or otherwise.  The more equipment outages, erratic behavior, or damage, the worse the power quality. 

Are all power disturbances real?  The answer is no.  This is not to say that we don’t have power quality problems, but it is easy sometimes for the layman or even the experienced to blame the power system for erratic behavior or failures.  It is also sometimes easy for a vendor, when put on spot, to blame bad power for their machine’s failure to operate properly.  In many cases a careful investigation is needed to verify that is was really a power quality problem.

What is good power quality?  Most users would associate an ideal sine wave with good power.  What they want all the time.  Unfortunately, multiple exposures of power system both inside and outside of the facility to transient, short, and long term events affects power quality. 

Sags, Swells, Surges, and other Bad Actors

Now that we have defined good power quality and have some grip on its boundaries, we need to define some of the common disturbances that can create bad power quality.  One of the problems is that a number of different definitions exist for some event, which may lead to some confusion.  Following are some of more common power disturbances:

·   Sag: Also known as dip, a sag is a reduction in RMS voltage at power frequency for a duration of a half a cycle to a few seconds (IEEE 1100). 

·   Swell: An increase in RMS voltage at power frequency from a half a cycle to a few seconds.

·   Outage, transient, or dropout: A complete loss for a period of time.

·   Impulse, transient, or surge: A sub-cycle disturbance- a sharp, brief discontinuity of the waveform of either polarity that may be additive or subtractive to the normal waveform. 

·   Oscillatory transient or ring wave: a sub-cycle disturbance that rings or has the appearance of a decaying sine wave. 

·   Notch: A switching or other disturbance of the normal power voltage waveform lasting less than a half cycle.  Initially opposite in polarity to the normal waveform.

·   Under-voltage: A sustained low voltage, this is often called brownout, though sometimes incorrectly.  A brownout is an under-voltage, but not all under-voltages are brownouts.  A brownout is a sustained under-voltage caused by a utility on purpose in time of heavy load. 

·   Noise: Unwanted electrical signal in a circuit that produces unwanted effects. 

·   Harmonics: Multiple of the fundamental frequency that exist in a power system due to nonlinear device operation  

Noise and Harmonics

Power disturbances can originate from both inside and outside of a facility.  They can also be due to normal and abnormal activity in the power system.

Normal activities that cause power disturbance are switching transients, utility capacitor bank switching operation, starting and shutting down of load equipment, arc welders, furnaces, adjustable-speed drives, fluorescent lights, and the operation of nonlinear electrical equipment.  Common abnormal activities are faults, fault clearing, and lightning.  A poor grounding system can contribute substantially to bad power quality.

Harmonics are an area of growing concern.  We are increasingly installing equipment with nonlinear switch mode power supplies, other nonlinear devices, and variable-speed drives.  Many facilities no longer have pure sine wave power and this can cause detrimental effects.

One of the most touted problems is overheating of the neutral on three-phase system, four-wire circuits feeding computer loads.  Harmonics also can cause overheating of transformers, premature over-current device operation, metering problems, malfunction of computer and sensitive electronic control boards, high crest factor, and voltage distortion problems.  For three-phase circuits, the odd harmonics (3,5,7,…) can cause problems because they are negative sequence (opposite in rotation sequence) to the fundamental.

The third and odd multiple of three (9^th, 15^th, etc) commonly called the “Triplen” harmonics, cause problems because they are in phase with the fundamental- rather than cancel in the neutral, they add up, leading to potential overheating of the neutral if it’s undersized.  Harmonics are discussed in ANSI/IEEE standard 519, “IEEE Guide for Harmonic Control and Reactive Compensations of Static Power Converters”.

Noise in AC distribution system comes from high-frequency components that are impressed on the 60 Hz wave.  These can come from transients with high-frequency components, from nonlinear devices, and from devices that operate at high frequencies.  High-frequency components tend to damp out quickly due to the inductance in the wiring.

Another problem is AC noise coupled into or conducted in the DC side.  Bad grounding can cause AC and lightning transients to enter the DC side with detrimental effects.  Noise can affect communication circuits through bad grounding techniques and exposure to noise sources.  IEEE standard 58, “IEEE Guide for the installation of electrical Equipment to Minimize Electrical Noise Input to Controllers from External Sources”, covers this subject extensively.

Band-Aid approach

As in many other control-related areas, the tendency with power quality problems is to take Band-Aid approach: we fix what is biting us at the time.  But because power systems are interconnected and extend into every corner of a facility, it’s important to evaluate the system as a whole.  We need to understand our equipment sensitivities, locate the problem, and uncover the sources, 

Our power systems are connected to external sources as well as interconnected within the facility, so power quality problems from both inside and outside can propagate throughout our system.  The effect of propagation depends on the size and duration of the disturbance and the propagation path from the source to the equipment under concern.  Since our systems are constantly changing, a Band-Aid or partial solution today may be inadequate tomorrow or may just shift the problem elsewhere.

Do an Audit

A power quality audit takes a systematic, comprehensive look at the power supply, the distribution system including grounding, the disturbance spectrum, and the equipment you have connected to the power system.  The purpose of this audit is to identify current, potential, and future vulnerabilities to power disturbances and to provide the basis for a comprehensive protection scheme.  It typically includes monitoring the power supply and distribution system over a period of time with a power monitor to get a baseline.  Some companies install power –monitoring equipment as part of their system for automated or semi-automated collection of this data. 

 A power quality audit should include a comprehensive and systematic visual and drawing inspection of the power distribution system and the grounding system.  If available, historical data can be collected from any maintenance management system.  Your utility may also be able to provide pertinent information.  

If you don’t have the expertise internally, or if you use external firm, make sure you check them out, as doing a good power quality audit requires a high level of expertise in electrical system and power quality and not all firms that advertise this service can meet this criterion. Remember, a process plant is in a constant state of flux with equipment being removed, upgraded, replaced, and added.  Future plans must be taken into account when the audit is done.  If not, you may install more or less protection that you need, or the wrong protection altogether.

Surge & The source of surge

Surges can be defined as a sub-cycle disturbance of an AC waveform that is evidenced by a brief, sharp discontinuity of the waveform.  It is also known as a transient or impulse.  Surges or transients, while defined for AC, can also appear in DC power circuits and digital communication circuits.  Surges can cause damages to equipment and the power system as well as cause equipment reset and erratic behavior.  Surges can also be cumulative, that is, a series of surges can each do a little damage until a failure occurs.

Surges can originate from inside or outside the facility.  Some sources of surge inside your facility are internal faults, power switching, ground transients, or caused by welding machines.  Surges can also come from outside sources including power line faults, utility capacitor bank switching operation, lightning, and load switching by transmission dispatch.  Lightning and ground disturbance also can enter a facility via paths other than the utility power lines, such as lightning strikes within the facility and thundercloud movement.  Phone or communication lines may also propagate surges.

In general electromechanical devices can withstand voltage surges until a dielectric breakdown occurs.  The sensitive electronic equipment, on the other hand, typically will have their operation upset before a hard failure occurs.  Electronic device susceptibility to surges is often related to smaller geometries, line widths, and component spacing in devices and circuits.

The move to switch-mode power supplies also has opened a window of exposure because the older, linear power supplies were basically low-pass filters, while a switch-mode power supply can act as a high-pass filter.  The move to high-speed digital circuits with 1’s and 0’s has made such equipment susceptible to reset or erratic behavior due to reference voltage and ground potential variation.

Surges can be normal or common mode.  Normal mode, also called transverse-mode, surges appear between any two power or signal wires.  Common-mode surges generally appear equally and in phase on each power or signal line to ground.  The mode of the surge can determine what type of protection is required.  Like many things, in order to provide a comprehensive and standard protection scheme, we need to be able to categorize surges.   ANSI/IEEE Standard C62.41-1991, “Recommended Practices for Surge Voltages in Low-Voltage AC Power Circuits, “breaks down a facility into three categories or areas and defines different type of surge waves with different characteristics for each area.  Three areas are: 1) service entrance and outside; 2) mid-building, feeders and short branch circuits; 3) and long branch circuits and receptacles.  The impedance in your wiring system can also have a natural effect of reducing propagating surges. 

Note that once inside the building or facility, the voltage test limit in the standard is 6kV, the voltage above which you will have insulation breakdown and flashover in low-voltage power distribution circuits.

Manufacturing facilities or processing plants many times are more spread out than the “building” description in the standard and may have a number of separately derived services.  This spreading out is generally physical or geographical and opens the system to transients that originate from lightning or large-scale ground disturbances.  This can make providing a comprehensive protection scheme much more difficult, particularly for remote areas such as tank farms and pipeline stations.

Solutions 1: Dedicated Circuits

Now that we have a grip on what surges are and where they come from, how do we protect against them? There is variety of protection devices and schemes available, ranging from fairly inexpensive to expensive.  Obviously, the money spent on protection must be weight against the cost of the equipment and the cost of any outages that might result from not having the protection.  Risk or consequences and probabilities may also be used as defining factor.

One means of providing some protection is to put sensitive equipment on dedicated circuits.  This is common but relatively expensive method.  It provides some isolation from other equipment and common connection and uses the impedance in the wiring to reduce the exposure to internal transients.  It is not however, a cure-all as some surges can and will propagate down dedicated circuits.

Solutions 2: Surge Suppressors

Probably the lease expensive device to protect against surges is the transient-voltage surge suppressor (TVSS).  ANSI/IEEE Standard C62.41-1991 provides a basis for the selection of voltage and current tests to be applied in evaluating surge withstanding capability of equipment connected to utility power system, primarily residential, commercial, and light industrial facilities.  This standard provides the basis for testing of TVSSs.

ANCI/IEEE C62.45, “Guide on Surge Testing for Equipment Connected to Low-Voltage AC power Equipment” provides guidance on testing methods and safety aspects for testing equipment connected to low voltage (1000volts and less).  Other ANSI/IEEE C62.xx series standards cover a range of surge-related areas.  UL Standard 1449-1998 (second edition), “UL Standard for Transient Voltage Surge Suppressors” covers the UL requirement for listing of permanently connected, cord-connected, and direct plug-in TVSSs.

The tests performed for the listing requirement assigns a “Measured Limiting Voltage” or suppression voltage rating that places the TVSS in a UL voltage class.  In addition, to the safety requirements, there is an optional test for effectiveness and reliability of the TVSS.  This adjunct testing was requested of UL by Federal government and consists of three grades (A, B, and C, wit A the best) that indicate endurance; three classes (1, 2, and 3 with 1 being the best) that specify the let-thru voltage; and two modes (1 or 2 with 1 meaning ground not contaminated) that specify whether the ground is contaminated or not during operation.

Be careful of statements like “designed to meet” or “designed per”

You want the TVSS to be UL-listed or approved by some other national recognized testing laboratory (NTRL), and you need to understand what listing or approval means.  If additional testing is advertised, it should be done by an independent third party (independent is sometimes hard to determine). Independent testing should have its basis in ANSI/IEEE C62.41.

TVSSs come in many shapes and forms, some effective, some almost useless.  These are essentially three types:  ones that shunt the surge to the neutral and/or ground, ones that block the surge (series type) and hybrids of the two.

The shunt types use three basic technologies: metal oxide varistors (MOV), silicon avalanche diodes (SAD), gas discharge tubes, and combinations thereof.  Basically, MOVs are inexpensive, relatively fast, and appear by themselves in cheap TVSSs, but have limited energy-handling capability.  MOVs are generally notorious for failure after repeated surges, are typically fused, and should have a failure light or indicator.  If not, you may think you have the protection when you don’t. They may also come with counter reading indicating the end of life cycle.

Series TVSSs are generally an indictor or choke in series with the power line combined with capacitors in parallel. TVSSs may also contain active electronic element to provide protection.

TVSSs have two basic purposes: to protect the system and attached equipment, and to keep the equipment operating satisfactorily during a surge.   These purposes can be in conflict in the design of a TVSS.  A shunt-type TVSS may provide protection but can not always prevent the reset or erratic behavior of equipment due to contamination of ground line or high neutral voltage.  The series types block the surge but the surge may find a lower-impedance path and cause damage elsewhere.

The hybrid TVSS is a combination of series and shunt elements that tries to take the better of the two methods to provide both protection and continued operation of equipment. TVSSs also come with bells and whistles such as a light or buzzer to indicate failure or surge strikes, fuse protection of MOVs, reset, and noise filters.

Experience proved that a single TVSS will not always provide adequate protection.  This gave rise to the cascade protection scheme, sometimes called the zone, fortress, or staged method.  In this scheme, two or three levels of TVSS protection are provided: at the service entrance, at the distribution panel, and at the load equipment, with each level designed for a finer degree of protection as you approach the equipment.  Depending on the risk, the protection at the equipment may involve other methods such as a power conditioner, isolation transformers, or uninterruptible power supplies (UPS). 

Solutions 3: Power Conditioners

Moving up in cost and degree of protection against surges, power conditioners are often used, generally at the equipment level to keep the cost down.  Common power conditioners include isolation transformers, ferroresonant transformers, and combination of TVSS or filter components and isolation transformers.

Isolation transformers provide good protection against common-mode surges but limited protection against normal-mode surges.  Typical rating for common-mode reduction for isolation transformers are 80-140db (10.000-10,000,000 reduction).  This reduction is also frequency-dependent.  Two other advantages to the isolation transformers are that it helps block noise, and you will generally be deriving your neutral closer to the equipment served, which minimizes neutral and ground problems.

One bad thing is that the transformer does not provide any voltage regulation against swells or sags.  The shielding concept in basic isolation transformers by placing an electrostatic shield between windings reduces the effect of the capacitance between windings through which transients might propagate.

A ferroresonant power conditioner uses magnetic saturation to provide conditioning, and in addition to providing surge protection also offers some voltage regulation ability.  This type of conditioner has a transformer designed so that the secondary is in magnetic saturation.  Changing the primary voltage or current does not change the magnetic flux or secondary voltage.  This type of conditioner can also reduce normal-mode impulses.  IT may, however be slow to respond to momentary overload on the secondary side.  They are generally about 90% efficient but efficiency drops with load, which can cause them to run hot and make audible noise.  Ferroresonants may also interact negatively wit other equipment downstream that has frequency elements.

Power conditioners also are available that combine transformer with other protective elements.  Some of these use isolation or other transformer with TVSS elements or filter capacitors.  These also have the advantages of driving the neutral close to the equipment, thus eliminating many of the effects of noise and transients from the grounding system.

 More on Noise & Harmonics

Noise can come from high-frequency components of many of same sources as surges.  It can also come from electromagnetic interference (EMI) introduced both by permanently installed equipment such as motors and transformers and by temporary operations such as welding.  In general, the building wiring system naturally reduces noise with its inherent impedance but some may propagate and cause problems.

Method to reduce noise includes shielding either the source or the target of the noise and filtering or isolating the noise.  Many of the devices used for surge protection have built-in noise protection or filtering. 

Common source of harmonics are switch-mode power supplies, fluorescent lamps, and variable-speed drives.   Harmonics combined with power factor-correction capacitors can cause dangerous voltage levels in a system due to resonant conditions.  Some reports have been made of overheating of the neutral in the buildings with a large number of switch-mode power supplies.  Many authors are also predicting that utilities may impose large penalties on users with high harmonic content, and user’s interest will certainly increase in this case.

Harmonics from variable-sped drives are generally minimized by installing line reactors on the drive.  These are also active types of harmonic reducers available.  Where these types of methods are not appropriate, isolation transformers, oversized components, and K-rated (harmonic-rated) transformers are used.

Power quality is not typically a simple subject related to a single area in your facility, bus a system concern.  There are many different types of protection devices and schemes out there, some smoke and mirrors and the rest valid methods. You must pick the right ones based on your system, and the risk that you are willing to assume.

Uninterruptible Power Supply (UPS)

 A UPS, as its name implies, provides power in the event of sustained outage.  It stores power in batteries or a mechanical storage element ender normal condition, then provide this power for a period of time when utility power is lost.  The supplied period of time can be from a few seconds to a few hours, though the common time period ranges from 15 minutes to an hour.  Depending on the UPS configuration, it can also provide a range of protection from power line disturbance such as surges, noise, and distortion. UPSs come in three basic forms: Static, Rotary, and Hybrid.

 Static: 

A static UPS has no moving parts.  There are basically three types of static UPS: off-line, line-interactive, and on-line.  All of these have batteries that determine the amount of power the UPS can supply after normal power is lost.  They recreate the sine wave output using one of the two basic technologies pulse width modulation (PWM) and steps (six step, 12 step, etc.).  These recreations are not perfect any may cause problem with the loads.

The total harmonic distortion (THD) as a percentage of the output of the UPS can be a key selection parameter.  A good UPS has a THD of typically 3-5%.  With the increase of harmonics in our system, input THD may also be a consideration.  The off-line UPS or standby power supply (SPS) supplies utility power under normal conditions and switches to battery backed-up inverter power upon power failure.  The main advantage of the off-line UPS us low cost, with other advantage including smaller size, higher efficiency, and low operating time on electronics,

Some of the disadvantages include limited surge and transient protection, and since they pass the utility power thru under normal conditions, they may also pass thru power line disturbances. They have to switch to supply back up power, which may cause problem with some loads, and you may not know if you have a problem until they fail on demand.  Battery maintenance may also be an issue,  This type of UPS is best suited for protecting small, stand alone equipment such as PCs and peripherals.

The line-interactive UPS typically has a bi-directional inverter and battery set that interacts directly with the incoming power line, often thru a shared transformer.  In the normal mode, the utility power supplies the output power and charges the batteries.  When utility power is lost, the inverter/batteries supply the needed power thru the shared transformer.  This type of UPS has a control algorithm that turns on the inverter when certain conditions are met.

The line-interactive UPS provides a good level of protection at a medium cost.  Some of the disadvantages are complex control scheme, possible cycling of the inverter due to narrow voltage or frequency window, and the possibility that you will not see a failure in the inverter/batteries until they are demanded to supply power.

In the on-line UPS, also known as a dual conversion UPS, the rectifier/inverter are online (supply power) under normal conditions.  IF utility power is lost, the battery and inverter supply power until the batteries are exhausted.  Some of the advantages of this type of UPS are continuous regulated power output, higher system reliability, and good diagnostics.  Some of the disadvantages are higher costs, lower efficiency, and larger size.  This type of UPS commonly has a maintenance bypass and a reserve power feed that can be switched online thru a static, solid-state switch if the UPS output fails.  With a reserve power feed, consideration must be given to power conditioning the reserve feed so that when operating on reserved feed, the effects of power disturbances are minimized.

The online UPS is generally the most common static UPS for highly critical, large loads in the process industries.  It would not be uncommon for this type of UPS to be used to supply a DCS, PLC, or other instrumentation systems.  In some cases, redundant arrangements of the online UPS are used.

 Rotary UPS:

The rotary UPS comes in several varieties but they typically have a motor-generator combined with a static UPS, a DC motor, or a mechanical energy storage element.  Rotary UPSs are generally more cost-effective in larger size; have high reliability; provide a clean, unity, sine- wave output; and mechanically isolate the output from input power.  This provides   complete elimination of power transients, surges, brownouts, and blackouts up to the limit of the storage element.  Rotary UPSs also can provide voltage conversion and frequencies other than 60Hz.

 One type of rotary UPS is essentially a static UPS that feeds a motor-generator.  The advantages of this arrangement are the inverter section can be made cheaper and more reliable, there is mechanical isolation between input to output power, and the output is a clean sine wave.  The inverter can be online (powering the motor all the time) or offline (powering the motor after power fails).  An alternative arrangement replaces the static UPS with a DC motor and battery set.

 Another type of rotary UPS is the flywheel UPS.  This type comes in two basic varieties.  One, the motor/flywheel/generator arrangement, stores energy in a flywheel that continues to supply rotating energy to the generator for a short period of time during an outage, allowing the system to have the same benefit as a motor-generator but also the ability to ride thru longer-duration sags.  In the second type, the flywheel storage element serves in place of the traditional UPS battery.  This type of UPS can ride through tens of seconds even hundreds of seconds of outage.

Hybrid:

The hybrid UPS typically combines a motor-generator with a standby generator.  These types of UPS are commonly used in Europe and are beginning to be used in the U.S.  The combination of a static switch and a standby generator also creates a hybrid UPS.

A common variety is the diesel generator UPS, which is a motor-generator that is also coupled to a diesel generator.  The system normally runs on motor-generator, but when the utility power fails the motor continues to turn due to inertial effects, maintaining the power while the diesel generator is started and come up to speed/frequency, it is coupled to the motor shaft thus continuing the power to the load.

Synthesizer Summary:

A power synthesizer is used for a number of reasons and there is wide variety to choose from.  Some of the primary reasons are when isolation is required, when a system must ride thru sustained sag or an outage, when other frequencies are needed, and when a clean sine wave is needed.  Power synthesizers are a higher-cost solution but provide a wide range of power quality solutions.

References

1.       IEEE Standard 1100-1992

2.       The Dranetz Field Handbook for Power Quality Analysis

3.       IEEE Standard 1346

4.       ITI (CBEMA) Curves & Application Note

5.       IEEE 581

6.       UL 1449 Adjunct Testing

7.       NEMA PE1-1992