earthquake requirements and seismic capabilities for eaton’s electrical distribution and control...

7/29/2019 Earthquake requirements and seismic capabilities for Eatons electrical distribution and control equipment A12501

1/24

Table of contents

Description

Part I Abstract and general overview . . . . . . . .

Part II Seismic terminology and earthquakeengineering . . . . . . . . . . . . . . . . . . . . .

Part III Seismic requirements . . . . . . . . . . . . .

Part IV Test facility and test methodology . . . . .

Part V Shared responsibilities . . . . . . . . . . . . . Part VI Typical Eaton seismic equipment

specifications . . . . . . . . . . . . . . . . . . . .

Part VII References . . . . . . . . . . . . . . . . . . . . . .

Effective August 2009Seismic White Paper SA12501SE

Earthquake requirements and seismic

capabilities for Eatons electricaldistribution and control equipment


2/242

Seismic White Paper SA12501SEEffective August 2009

Earthquake requirements and seismiccapabilities for Eatons electrical

distribution and control equipment

EATON CORPORATION www.eaton.com

About the authors

Mr. Eddie Wilkie graduated from North Carolina State University,earning a Bachelor of Science degree in Mechanical Engineering.Eddie has been employed with Eaton for 19 years. During that time,he has held a variety of engineering and management positions.Eddie has worked as a design engineer, design engineering man-ager, business operations manager and most recently as the division

engineering manager for Eatons Power Distribution OperationsAmericas. Eddie is currently responsible for coordinating Eatonselectrical equipment seismic program, which includes annual testingand equipment certification.

Mr. Frederick M. Paul has spent over 31 years in various aspects ofthe electrical industry. The last 11 have been as an application engi-neer with Eaton, covering the central valley of California, USA. Priorto Eaton, he was involved in the sales and application of electricalcontrol and automation equipment. Earlier in his career, he was vicepresident of operations for 11 years at an electrical contracting firmin Southern California, USA.

Dr. Mostafa A. Ahmed has 37 years of extensive experience instructural and mechanical design and construction of powergenerating stations. He is a fellow engineer with the WestinghouseElectric Company Nuclear Service division. He is skilled in civil andmechanical engineering practices, dynamic analysis of equipmentand structures, and finite element analyses. Dr. Ahmed is anexpert in equipment seismic qualification and seismic testing.He earned his bachelor of science degree in civil engineeringfrom Cairo University in 1971, and received his master of scienceand doctorate in structural mechanics from the University ofPittsburgh in 1981 and 1991. He is currently working in Shanghai,China, as a technical advisor for the Westinghouse On-ShoreEngineering Organization.

Mr. Nathan M. Glenn, P.E., is a practicing Mechanical Engineerspecializing in equipment qualification. He is experienced in shockand vibration testing, structural dynamics, electro-mechanicalanalysis, and design. Nathan earned his Bachelor of Science degreein Mechanical Engineering Technology from The PennsylvaniaState University, and received a Master of Science in EngineeringMechanics specializing in Explosives Engineering from New Mexico

Institute of Mining and Technology. He is a registered ProfessionalEngineer in the state of Pennsylvania. Nathan is a Senior Engineerwith Westinghouse Electric Company. Currently he is responsiblefor the qualification of nuclear power plant components andsystems. In addition to nuclear power plant equipment qualification,Nathan provides seismic certification for electrical equipment usedin building code applications.

Part I

Abstract

Eaton Corporation is a global diversified industrial manufacturerconsisting of two sectors: Industrial and Electrical. Throughoutthis document, all references to Eaton are in regards to EatonsElectrical Sector.

Beginning with qualification testing in 1985, Eaton has led theindustry in seismic certification of electrical equipment for use infacilities across the continental United States. Eaton was the firstelectrical equipment supplier to employ seismic simulation testingfor equipment seismic certification.

For more than 20 years, Eaton has had a comprehensive programfocused on designing and manufacturing electrical distributionand control equipment capable of meeting and exceeding theseismic load requirements of the Uniform Building Code (UBC),the California Building Code (CBC), and the Building Officials andCode Administrators (BOCA) International, Inc. The entire pro-gram has been updated to demonstrate compliance with the 2006International Code Council (ICC) and the 2006 International Building

Code (IBC) unified seismic requirements. This also includesthe 2007 CBC.

Eaton recognized that the most direct and proven method ofassuring seismic performance of electrical equipment is throughsimulation testing via triaxial or biaxial shake tables. Representativeconfigurations for each of Eatons product lines were designedand built for seismic testing. Considerable attention was given toselecting test units that conservatively represented the entire familyof products being certified.

Test units were initially subjected to independent 0.2g resonantsearches in each of the three principal axes prior to being subjectedto a series of seismic simulation tests. The test assemblies wereproven to meet or exceed the seismic performance requirementsand remain operational immediately after the seismic event. Thispaper provides a summary of the efforts that were involved in the

achievement of this objective.

Background

Although the need for seismic-capable electrical equipment isknown, there is a lack of understanding of how to comply withcurrent code requirements. The 2006 International Building Code(IBC) and the 2007 California Building Code (CBC) both emphasizebuilding design requirements with limited information for seismiccertifications of equipment. Electrical equipment and distributionsystem components are treated as non-structural attachments tothe building.

Since seismic testing contains many special terms and formulations,this paper begins with the basics of seismic terminology andearthquake engineering, then proceeds with addressing the specificfactors involved with meeting the requirements of the IBC and CBC.

The 2006 IBC seismic requirements, along with associated codesderived from the IBC, will be addressed, explaining how they relateto Eatons previous and current test programs. The most stringentrequirements of these codes (IBC and CBC) will be presented asthey apply to electrical distribution and control equipment andwill be combined to formulate a single reference for purposesof evaluation.

To properly define the acceptability of the equipment to thespecified codes, it is necessary to present the equipment seismicrequirements and the equipment seismic capability data on thesame technical basis. For this purpose, the use of the responsespectrum concept will be introduced. To simplify the application forthe user, the seismic capability of all of Eatons equipment has beenestablished to the same basic levels and requirements.

The equipment is considered acceptable, or granted a seismic

certificate, if it can withstand the seismic event and perform itsfunction immediately afterward. Eaton participates in a cooperativeeffort with the user, building designer, and installer to ensurethat the equipment is mounted properly to a foundation that canwithstand the effects produced by an earthquake.


3/24

Seismic White Paper SA12501SEffective August 2

Earthquake requirements and seismiccapabilities for Eatons electricaldistribution and control equipment


General overview

Eatons electrical distribution and control equipment has undergoneseismic simulation tests and meets or exceeds performancerequirements as identified in the 2006 IBC1 and the 2007 CBC2. Itis important to note that Eaton has tested its equipment using themost typical mounting methods. All Eaton floor-mounted equipmenthas been seismically tested as free-standing units, with no lateralsupports at the top that are affixed to adjacent walls or structures.This allows users to either secure it from the base alone or in com-binations of base and top lateral supports. It must be recognizedthat equipment tested with top lateral supports is not certified to thesame levels as free-standing items. Eaton has shown by simulationtesting that equipment seismic capabilities are reduced by morethan a factor of two when mounted at the base only as compared tosecuring with lateral supports located at the top of the equipment.Complete and proper certification of equipment to achieve maxi-mum flexibility for the user must include testing of equipment asstand-alone items. Eaton equipment that is certified to higher levelswhen installed with top lateral supports is specifically indicated onEatons seismic certificates.

Eaton is highly experienced in the design, manufacture, and seismiccertification of electrical distribution equipment to meet the mostrigorous seismic standards. As new products are developed, or

existing products are modified, Eaton continues to verify the seismicacceptability of nearly all lines of electrical equipment for applica-tions requiring certification to the IBC and CBC.

Over a period of more than 20 years, over 100 different assemblies,representing many product lines, have been successfully testedand verified to seismic levels higher than the maximum seismicrequirements specified in the IBC and CBC. The equipment main-tained structural integrity and demonstrated the ability to functionimmediately after the seismic simulation tests. This achievement, anindustry first, is consistent with the Eaton commitment to producethe most reliable equipment that exceeds both present and futurerequirements. Testing was performed on simulation tablesat Wyle Test Laboratory in Huntsville, Alabama, along with theformer Westinghouse Advanced Energy Systems Division inPittsburgh, Pennsylvania.

The general concepts for seismic test methodology, ANSI/IEEEStandard 344-1987, and the applicable procedures from ANSI/IEEE

C37.81Guide for Seismic Qualification of Class 1E Metal-EnclosedPower Switchgear Assemblieswere consulted.

The equipment was subjected to the following vibration excitationand seismic simulation testing:

1. Initial resonance searches in all three principal directions inthe frequency range of 1 to 50 Hz, using sine sweep motionat the base of the test units, with a sweep rate of 1.0 octaveper minute. Peak acceleration of the sine wave was designedaround 0.2g. (Some resonance sweep tests have beenconducted up to 100 Hz.)

2. Seismic simulation testing using 30-second-long randommultifrequency inputs imposed simultaneously and

measured at the base of the test cabinets in all threeprincipal directions. The base acceleration levels wereincreased further to encompass the combined coderequirements, and additional testing was performed todemonstrate margin beyond code requirements.

Nearly all of Eatons electrical assemblies have been tested andwere found acceptable when evaluated to IBC and CBC seismicrequirements. Eaton continues to lead the industry in usingsimulation testing to ensure conformance of electrical distributionand control equipment to the most current codes. (See Part III.)

Establishing the equipment seismic capability is only the first stepA seismically qualified mounting base with anchors or welds isrequired to hold the equipment safely to the supporting structure.(See Part V.)

Part II

Seismic terminology and earthquake engineering

Earthquakes occur in most every region around the world.3 (SeeFigure 1.) As reported by the U.S. Geological Survey, in 2007 alon55 earthquakes greater than 6.0 were recorded around the world.These resulted in 681 reported deaths and widespread damage tostructures, buildings, and equipment with damage estimates in thbillions of dollars.

The May 12th, 2008 earthquake in Chengdu, China, exceeded 6.9and caused a major tragedy of more than 69,000 deaths andresulted in several thousands of people missing. The problemswere further compounded due to the delays in restoring powerand service to the affected areas.

To restore function of emergency management facilities as quicklyas possible, public officials have revised building codes to mandatimproved seismic design. This includes not only buildings, but alsothe electrical and mechanical equipment contained therein, as weas machinery necessary for safe occupancy and normal operation

Eaton has taken the unique step of performing seismic simulationtests on various lines of distribution and control products. Users cabe sure that Eatons electrical equipment has been designed andtested to exceed the requirements as identified by the IBC and CBCFor purposes of understanding, it is important to review a few of thbasic principles of earthquake engineering. From an analytical per-spective, it is not easy to quantify the severity of an earthquake.Most news reports refer to the magnitude of the earthquake in

terms of the open-ended Richter scale. Although most people haveheard of the Richter scale, the understanding is limited. The originadefinition is:4

Richter magnitude is M, where M = log10(A)

Where A is equal to the trace amplitude (in microns) of aWood-Anderson Seismograph having magnification of 2800,natural period of 0.8 seconds, a damping coefficient of 80%,located on firm ground, at a distance of 62.5 miles (100 km)from the earthquake epicenter.

Since 1000 microns are equal to 1.0 millimeter, and log10 (1000)is equal to 3, M could then be redefined as:

M = 3 + log10(trace amplitude in mm)

1 International Code Council, International Building Code.

2 California Building Standards Commission, California Building Code.

3 Newmark, Fundamentals of Earthquake Engineering, p. 252.

4 Ibid., p. 217.


4/244





Table 1. Relationship of Earthquake Magnitude toOther Parameters

EarthquakeMagnitude M(Richter Scale)

Maximum7GroundAcceleration(% g)

Duration of7StrongMotion(Seconds)

Length of8Fault Slip(Miles)

Equivalent9Energy(Tons of TNT)

8.5 50 73 530 70 million

8 50 43 190 13 million

7.5 45 30 70 2.2 million

7 37 24 25 400,000

6.5 29 18 9 70,000

6 22 12 5 13,000

5.5 15 6 3 2200

5 9 2 2 400

Table 2. Modified Mercalli Intensity Scale (abridged andRewritten by C. F. Richter10)

Intensity11 Definition1 Not felt. Marginal and long period of large earthquakes.

2 Felt by persons at rest , on upper f loors, or favorably placed.

3 Felt indoors. Hanging objects swing. Vibration like passing of light trucks.Duration estimated. May not be recognized as an earthquake.

4 Hanging objects swing. Vibration like passing of heavy trucks; orsensation of a jolt like a heavy ball striking the walls. Standing motor carsrock. Windows, dishes, doors rattle. Glasses clink. Crockery clashes. In theupper range of 4, wooden walls and frames crack.

5 Felt outdoors; direction estimated. Sleepers awakened. Liquids disturbed,some spilled. Small unstable objects displaced or upset. Doors swing, close,open. Shutters, pictures move. Pendulum clocks start, stop, change rate.

6 Felt by all. Many frightened and run outdoors. Persons walk unsteadily.Windows, dishes, glassware broken. Knickknacks, books, and so on, offshelves. Pictures off walls. Furniture moved or overturned. Weak plasterand masonry D cracked. Small bells ring (church, school). Trees, bushesshaken visibly, or heard to rustle.

7 Diff icult to stand. Not iced by drivers of motor cars. Hanging objectsquiver. Furniture broken. Damage to masonry D including cracks. Weak

chimneys broken at roof line. Fall of plaster, loose bricks, stones, tiles,cornices, unbraced parapets, and architectural ornaments. Some cracksin masonry C. Waves on ponds; water turbid with mud. Small slides andcaving in along sand or gravel banks. Large bells ring. Concrete irrigationditches damaged.

8 Steering of motor cars affected. Damage to masonry C; partial collapse.Some damage to masonry B; none to masonry A. Fall of stucco and somemasonry walls. Twisting, fall of chimneys, factory stacks, monuments,towers, elevated tanks. Frame houses moved on foundations if not bolteddown; loose panel walls thrown out. Decayed piling broken off. Branchesbroken from trees. Changes in flow or temperature of springs and wells.Cracks in wet ground and on steep slopes.

9 General panic. Masonry D destroyed; masonry C heavily damaged, some-times with complete collapse; masonry B seriously damaged. Generaldamage to foundations. Frame structures, if not bolted, shifted offfoundations. Frames cracked. Conspicuous cracks in ground. In alleviatedareas, sand and mud ejected, earthquake fountains, sand craters.

10 Most masonry and frame structures destroyed with their foundations.

Some well-built wooden structures and bridges destroyed. Seriousdamage to dams, dikes, embankments. Large landslides. Water thrownon banks of canals, rivers, lakes, and so forth. Sand and mud shiftedhorizontally on beaches and flat land. Rails bent slightly.

11 Rails bent greatly. Underground pipelines completely out of service.

12 Damage nearly total. Large rock masses displaced. Lines of sight andlevel distorted. Objects thrown into the air.

To determine the magnitude of an earthquake in terms of RichterScale M, one could proceed as follows:

1. Measure the amplitude of the Wood-Anderson Seismographin millimeters at a location 62.5 miles from the earthquakesepicenter.

2. Take common logarithm of the amplitude.

3. Add 3 to it.

The result is the magnitude (M) of an earthquake on theRichter scale.

ote:N It is important to note that for a change of one unit in the Richter scale,M means a change of 10 in the amplitude of the motion of the earthquake.

To relate the magnitude (M) to the energy radiated by the earth-quake, Gutenberg and Richter developed the following relationship5:

log10(Es) = 11.8 + 1.5 M

Where Es is the seismic energyof the earthquake and M is theRichter magnitude.

A one megaton bomb releases about 5x1022 ergs. If all of the energycould be converted into seismic energy (typically only about 2%would be), it would correspond to a magnitude 7.3 earthquake6.

Table 1 relates the earthquake magnitude to other relevant param-eters. A change of one unit in the magnitude M means an increaseof 1.5 in the right-hand side of the equation, resulting in a change of32 in the total energy of the earthquake, Es, since log10 (32) = 1.5.

Although the magnitude of the earthquake is a direct measure of itsseverity, there are a number of difficulties in using it for equipmentdesign. Specifically, they include:

1. Maximum displacement alone does not provide necessaryinformation about the frequency content of the motion. Dueto the effects of amplification, equipment is most susceptibleto damage when the earthquake motion contains the equip-ments inherent natural frequencies.

2. Maximum displacement is not necessarily a good measureof the total amount of energy the equipment is subjected toduring the earthquake. Velocity is a better indicator ofthe energy.

3. Maximum displacement is not a good measure of the forcethe equipment will experience. Acceleration shows a bettercorrelation to the resultant seismic forces on the equipmentmounted inside buildings or structures.

4. Seismographs are normally tuned to frequencies in the 1to 2 Hz range. This is adequate for measuring the magnitudeof the earthquake, but does not provide accurate informationabout the frequencies typically translated to buildingsand equipment.

Figure 1. Seismicity of the Earth, 19611967

5 Wiegel, Earthquake Engineering, p. 31 (Ref. 8).

6 Newmark, Fundamentals of Earthquake Engineering, p. 252 (Ref. 7).

7 Wiegel, Earthquake Engineering, p. 79, Table 4.3 (Ref. 8).

8 Wiegel, Earthquake Engineering, p. 77, Table 4.1 (Ref. 8).


10 Newmark, Fundamentals of Earthquake Engineering, Appendix 2 (Ref. 7).

11 Intensity is frequently represented by Roman numerals.


5/24




To eliminate many verbal repetitions in the original scale, the follow-ing convention has been adopted. Each effect is named at that levelof intensity at which it first appears frequently and characteristically.Each effect may be found less strongly, or in fewer instances, at thenext lower grade of intensity; more strongly or more often at thenext higher grade. A few effects are named at two successivelevels to indicate a more gradual increase.

Masonry A, B, C, and D. To avoid ambiguity of language, the qualityof masonry, brick or otherwise, is specified by the followinglettering (which has no connection with the conventionalClass A, B, C construction).

Masonry A. Good workmanship, mortar, and design; reinforced,especially laterally, and bound together by using steel, concrete,and so forth; designed to resist lateral forces.

Masonry B. Good workmanship and mortar; reinforced, but notdesigned to resist lateral forces.

Masonry C. Ordinary workmanship and mortar; no extremeweaknesses like failing to tie in at corners, but neither reinforcednor designed against horizontal forces.

Masonry D. Weak materials, such as adobe; poor mortar;low standards of workmanship; weak horizontally.

In addition to the magnitude of the earthquake, which measures theamount of energy released, another parameter, the intensity, is usedto measure the local destructiveness of earthquakes. Therefore,one earthquake will have a single magnitude, but a number of differ-ent intensities, depending on the location of the observers.12Mostintensity scales are based on personal and subjective observations,including scary feeling and the ability (or inability) to remainstanding, as well as the sorts of property damage that occurred.

Although quantitative and based on actual damage effects, a reviewof Table 2, the Modified Mercalli (mm) scale13, reveals that it is toosubjective for use in electrical equipment design and qualification.Despite its limitations, the intensitycan be quite useful in areaswhere there are no seismic instruments available to record theearthquake, and it may provide the only consistent way to interpretthe diaries and other written accounts of historical earthquakes.14

When available, the most accurate descriptions of actual earth-quake motions are the time history records. A time history recordis simply a graphical recording of the earthquake motion (it can bein terms of displacement, velocity, or acceleration) as a function oftime. Figures 2 and 3 illustrate time history records for two differentearthquakes.15

Although accurate and readily available soon after the event, thesrecords are not ideal for translating requirements to equipmentdesign and seismic certification. For example, note that theEl Centro, California, earthquake acceleration magnitude reaches0.3g several times, and is consistently above 0.1g, while the maxmum peak-to-peak displacement is about 30 cm. For the MexicoCity earthquake, the acceleration magnitude is generally about0.01g, with one peak at about 0.02gonly about 10% of the

El Centro acceleration levels.The maximum peak-to-peak destruction displacement of the MexCity earthquake was 60 cm, or about twice the El Centro displacement value. Noting the difference in the time scale, one immediarealizes that the Mexico City earthquake motions are characterizeby much lower frequencies than the El Centro event.

Figure 2. El Centro, California, Earthquake of May 18, 1940,NS Component16

Because acceleration is a function of displacement times the squaof the natural circular frequency (for sinusoidal motions), the domnant frequencies of the El Centro earthquake are four to five timethose of the Mexico City earthquake. This illustrates that while thtime history is very accurate for any one earthquake, it is difficult use as the basis for generalizations about other earthquakes.

0 5 10 15 20 25 30

Time, seconds

South

North Displacement

20cm10

0

10

South

North Acceleration

0.3 g0.2 g0.1 g

0-0.1 g-0.2 g-0.3 g

South

North Velocity

20cm sec-1

0

20

40


13 Ibid., Appendix 2.

14 Ibid., p. 218.

15 Ibid., p. 227.

16 Ibid., p. 227.


6/246





Finally, the last concern in using time history can be illustrated asfollows. Figure 4 shows the time history records (both displacementand acceleration forms) for a shake table test run.

The question arises related to the adequacy of the time history toaccurately represent the earthquakes shown in Figures 2 and 3.Because the acceleration record for the test contains several peaksin the range of 2 to 3g, there are enough peak accelerations andinertial forces to satisfy the requirements. Similarly, the peak-to-peakdisplacements are in the range of 60 to 75 cm for the test, whichwould appear to be sufficient to meet even the Mexico Citydisplacement levels.

On the basis of amplitude alone, it appears that the shake table timehistory (Figure 4) meets both actual earthquake events (Figures 2and 3). However, there are two critical factors not yet addressed: (a)frequencies present in the required motion versus the frequenciespresent in the test motion, and (b) the inherent equipment damping.

It has already been shown that the frequency content representsa significant difference in contrasting Figures 2 and 3. Similarly,frequency is also critically important in establishing equipmentcertification. The reason is basic: each piece of electrical equipmenthas its own natural frequency that produces maximum amplification.For larger assemblies, 4 to 6 Hz is typically the minimum naturalfrequency. If the earthquake has significant 4 to 6 Hz motion, the

equipment will respond accordingly, amplifying or resonating withthe earthquake motion. If the earthquake has substantial 10 to 12Hz motion, the equipment will be too flexible to keep up with thehigher frequency, thus, it will tend to sit still or attenuate the earth-quake motion. If the earthquake has a significant amount of 1 to 2Hz motion, the equipment will rigidly follow the motion of the floor,neither amplifying nor attenuating. Figure 5 is a resonance curveand illustrates the three regions of equipment response as afunction of the ratio of the equipment natural frequency to theinput motion frequency.

Another important factor that time histories do not address isequipment damping. For simplicity, equipment damping is oftenexpressed as a ratio (C/Cc) of the actual equipment (C) dampingto that of a critically damped system (Cc). Frequently, the ratio isexpressed as the percent of critical damping. As one can see fromFigure 5, the damping property of the equipment limits the totalamplification that the equipment will experience at resonance. Withno damping, the equipment response amplification at resonanceincreases without bound. With a damping coefficient of 12.5%, theequipment response will not exceed four times the input motion,as can be seen in Figure 5.

This resonance curve is helpful in understanding the equipmentresponse to earthquakes. However, these curves, based on continu-ous sinusoidal input motions, are too conservative for representationof actual earthquakes, which have not one, but a number of differentfrequencies. Additionally, these frequencies are discontinuous,starting and stopping several times during the course of theseismic event.

Figure 3. Mexico City Earthquake of July 6, 1964, NS Component17

Figure 4. Shake Table Time Histories for Equipment Test

Figure 5. Resonance Curves for Continuous Sine Motion

0 10 20 30 40 50 60 70 80 90 100 110 120 130 140

0.02 g

0.01 g

0

-0.01 g-0.02 g

10 0-104030

20

10

0

-10

-20

-30

cm

cm sec-1

(Typically the ground displacementsare incredibly large due to small errorsin the base line for the accelerogram)

Acceleration

Displacement

Time, seconds

Velocity

Displacement(cm)

(g)Acceleration

20 Seconds

CAL2.5" Peak/25 Lines5"/Line

CAL1g/1 Line

5

4

3

2

1

0 0 1 2 3

EquipmentResponse/InputMotion

EquipmentAmplifies

InputMotion

EquipmentFollows

InputMotion

EquipmentAttenuates

InputMotion

Frequency Ratio

Equipment Natural Frequency

Input Motion Frequency

Equipment Response per Unit Input Motion as a Function of Frequency Ratio:

F= 0.25

FF

F= 0.50

FF

F= 0.125

FF

F= 1.00

FF

F = 0FF

17 Newmark, Fundamentals of Earthquake Engineering, p. 227.


7/24




Figure 7. Equipment Qualification by the Response SpectrumMethod When TRS Envelopes the RRS for All EquipmentNatural Frequencies

Figure 8. Q CurvesVibration Magnification vs. Percent ofCritical Damping

Frequency Hz

Acceleration(g)

Spectrum DipNotImportant BecauseFrequency Is Not anEquipment Natural

Frequency

Required ResponseSpectrum

(RRS)

Zero PeriodAcceleration= MaximumTable Test

Motion

Zero PeriodAcceleration= MaximumFloor Motion

Test Response Spectrum(TRS)

1 2 3 4 5 6 7 8 9 10 20 30 40 5060

70 9080 10

109876

5

4

3

2

1.00.90.80.70.6

0.5

0.4

0.3

0.2

0.1

30

40

20

10

00 5 10 1

ContinuousSine MotionQ=100/(2c/c

c)

10 Cycles/Beat

5 Cycles/Beat

RandomVibration

X10

X6.8X5.6

X3.1

QFactor,MagnificationNumber

c/cc, Percent of Critical Damping

TypicalEarthquake

Ground Motion

Q = 1002c/c

c

Figure 6. Basic Vibration Equations18

Because of these difficulties in universally applying the time historyform, engineers have developed a method of comparing earthquakeresponse motions as a function of frequency, rather than time.This is called the acceleration response spectrum method.The acceleration response spectrum for any time history is a plot of

the maximum responses of a series of linear, single-degree, free-dom oscillators (one spring, one mass, one dashpot that can movelinearly along only one axis) mounted on a surface moving accordingto the time history being studied. Figure 6 depicts one such simpleoscillator and its basic equations of motion. Typically, the responsespectra are plotted over the 1 to 35 Hz frequency range in no lessthan 16 steps, not exceeding one-third octave. (For example, 1.0,1.26, 1.6, 2.0, 2.5, 3.2, 4.0, 5.0, 6.3, 8.0, 10.0, 12.7, 16.0, 20.0,25.4, and 32.0 Hz.)

The responses of these oscillators are easily determined in realtime, with digital computers and fast spectrum analyzers in thetest laboratory. However, the complex and difficult task of commu-nicating earthquake requirements and equipment capabilities hasbecome a routine matter of showing that the equipment capabilityresponse spectrum, as produced by shake table test, envelops theground-level seismic requirements. During this test, the response

spectrum envelops the applicable portion of the location wherethe equipment is to be installed. The applicable portion meansthat enveloping is required at all equipment frequencies. Figure 7shows a typical test responsespectrum(TRS) enveloping theapplicable portion of the required response spectrum (RRS).Note that enveloping does not occur at 4.5 Hz, which is accept-able, because this was not a resonant frequency of the equipment.Enveloping is only necessary at the natural frequencies of theequipment. This illustrates the value of the simple frequency sweeptest to identify the lowest natural frequencies and damping factorsassociated with any equipment seismic test certification program.

Figure 8 illustrates a more useful form for engineers. The peakmagnification(alignment of the equipment natural frequency withthe earthquake frequency), Q, as a function of the damping. Eachcurve represents a different type of earthquake motion. The low-

est curve for random motion (all frequencies present to an equalextent) is generally the most like the ground motion during an actualearthquake. Because most electrical equipment is mounted on arigid surface or inside another structure, the original earthquakemotion is filtered by that structure. The equipment, therefore,experiences so-called quasi-resonance effects as the structure towhich it is mounted alternately amplifies and attenuates the earth-quake motion according to its inherent characteristics. The result isthe equipment peak amplification lies somewhere between the twoextremeslower level random motion response and higher levelcontinuous sine response.

Sinusoidal Harmonic Motion

X = A sin wt = Displacement

X = dx = A w cos wt = Velocitydt

X = dx = A w2 sin wt = Accelerationdt

Where W = 2 x Frequency

Linear Single-Degree-of-FreedomOscillator

C

M X

k

18 Beer, Vector Mechanics for Engineers: Statics and Dynamics, p. 771.


8/248





CAUTION!

IT IS IMPORTANT TO VERIFY THAT THE RRS AND THE TRS AREBASED ON AND PLOTTED AT THE SAME DAMPING FACTORBEFORE MAKING THE COMPARISON.

Now that the basics of earthquake engineering have been presented,several key elements that are very useful in understanding the nature

of earthquakes, time history, the response spectrum curve (RSC), andthe potential effect on electrical enclosures can be discussed further.

Although the Richter scale M is of good use in describing earth-quake strength, it does not identify the energy content of theearthquake or its potential to damage structures and equipment.Basically, the Richter scale M is a displacement indicator ratherthan an energy or acceleration indicator.

The energy content of an earthquake can best be defined via theRSC. This curve must be carefully understood and carefully applied.It is a theoretical curve computed for application to a system orequipment. Typical curve sets are shown in Figure 9.

The only spectral acceleration magnitude that is directly related tothe earthquake event is the maximum response of rigid systems.Within earthquake engineering, it is understood that rigid systemsare those with no resonance frequencies below 33 Hz and areconsidered to have a zero period of acceleration. Because a rigidsystem will not amplify the motion of the earthquake, its maximumresponse acceleration is equal to the maximum acceleration of theearthquake time history. As a result, the part of the RSC at thehigher frequencies, referred to as zero period acceleration(ZPA),directly defines the maximum acceleration of the earthquake timehistory. It does not depend on the damping properties of the equip-ment. Thus, no matter what the equipment damping, the ZPA isalways the same, and is equal to the maximum acceleration in theearthquake time history.

All other spectral accelerations are possible only if the equipmenthas a dominant resonant frequency that aligns with the frequencyon the response spectrum curve (RSC). Thus, this curve tells theengineer that it is possible for a piece of equipment to experiencethe spectral acceleration defined in the curve, if the equipment has

a dominant resonance frequency matching the frequency onthe RSC.

It is important to understand that the damping properties of asystem are a direct indication of the systems ability or inabilityto dissipate the earthquake energy.

To further explain the effect of damping properties on the seismicresponse of systems, assume that two enclosures were similarlydesigned and built, but with one exception: One enclosure is awelded structure, while the other is a bolted structure. Aside fromthis difference, the enclosures are identical in design, mounting,and weight. Should both be subjected to an earthquake motion, thestructural elements in the bolted cabinet will move relative to eachother, producing friction and noise. Ultimately, these effects withinthe bolted enclosure result in increased dissipation of the energyproduced by the seismic event as compared to the weldedenclosure. The bolted enclosure will dampen the energy muchquicker than the welded version, resulting in reduced time forthe seismic response to build up.

For this reason, the RSC is usually computed and plotted for differentdamping propertiestypically 1%, 2%, 3%, 5%, 7%, and so forth. Itmust be recognized that all resultant plots on the RSC are producedas a result of the same earthquake time history input motion(see Figure 9). It should be apparent that the result and responsewill be higher for systems with lower damping properties, and lowerfor systems with higher damping. A very useful rule is: The higherthe damping coefficient of the equipment, the lower its responsecurve; the lower the damping coefficient of the equipment, thehigher its response curve.

Figure 9. Response Spectrum Curve

Figure 10. Triaxial Shake Table

Figure 11. Response Spectrum Curve, 5% Damping Curves

10

1.0

0.10.1

1.0 10 100

2% Damping, 2.4g Peak Acceleration

Acceleration(g)

Frequency Hz

7% Damping, 1.25g Peak Acceleration

Zero Period Acceleration (ZPA), Equal to 0.6g

5% Damping, 1.4gPeak Acceleration

Side-to-SideBase Motion

VerticalBase Motion

Front-to-BackBase Motion

4 x 36"

SwitchgearAssembly

90"

86"

10

1.0

0.10.1

Frequency Hz

1.0 10 100

Front-to-Back Seismic Base Input

Side-to-Side Seismic Base Input

Vertical Seismic Base Input

Accele

ration(g)


9/24




To further demonstrate the use of the RCS, consider this example:Multiple switchgear sections are mounted to a triaxial shake table(Figure 10). The section of equipment is then subjected to the baseseismic response spectrum shown in Figure 11.

The switchgear assembly has a dominant natural frequency of 6 Hzin the side-to-side direction, 10 Hz in the front-to-back direction, and55 Hz in the vertical direction. The assembly is a bolted structurewith a 5% damping coefficient.

If the switchgear assembly is subjected to test levels that couldproduce the RRS, shown in Figure 11, one can quickly determinethat the enclosure will amplify the base motion and could experiencea 1.5g acceleration in the side-to-side, a 2.0g in the front-to-back, anda 0.9g in the vertical direction. The resultant forces and moments onthe structural elements and internal components can now becomputed to confirm that the enclosure and components containedtherein will maintain their structural integrity.

To design the enclosure foundation, apply these accelerations at thecabinets center of gravity (C.G.), multiply them by the total mass ofthe equipment, and increase them with factors as appropriateto account for design margin. Thus, cross-coupling effects and close-mode contributions are taken into account. Next, determine theresultant moments, forces, and shear on the mounting bolts or welds.

Only the acceleration associated with the dominant naturalfrequencies of the enclosure need to be selected from the spectrumcurves. For example, the front-to-back direction RRS has a peakspectral acceleration (3.2g) in the frequency range 1.5 to 3 Hz.This acceleration has little or no effect on the enclosure, becausethe front-to-back frequencies of 10 Hz do not coincide with thisfrequency range (1.5 to 3 Hz).

A beneficial engineering practice is to design equipment withnatural frequencies that do not align with the frequencies foundin the earthquake time history.

Most earthquakes tend to include low frequencies (1 to 3 Hz).Eaton understood this phenomenon and designed equipment withresonance frequencies above those levels. All Eaton equipment isdesigned with frequencies above 3.2 Hz, which serves to minimizethe amplification.

This is further discussed in the next section (Part III) where the devel-opment of seismic requirements for electrical equipment is considered.

Part III

Seismic requirements

Consistent with Eatons commitment to produce equipment thatexceeds present and future code requirements, essentially allengineered-to-order assemblies and standard assembly productshave been designed, manufactured, and tested to meet rigorous

seismic requirements.

International Building Code (IBC) 2006

On December 9, 1994, the International Code Council (ICC) wasestablished as a nonprofit organization dedicated to developing a

single set of comprehensive and coordinated construction codes.The ICC foundersthe Building Officials and Code Administrators(BOCA), the International Conference of Building Officials (ICBO),and the Southern Building Code Congress International (SBCCI)created the ICC in response to technical disparities among the threerecognized model codes in use at the time. The ICC offers a single,complete set of construction codes without regional limitationstheInternational Building Code (IBC).

Since the establishment of the ICC and the issuance of the 2000IBC (Rev-0), there have been two revisions: the first was publishedin 2003; the second in 2006. There were no substantial changes inthe code that affected the validity of the 2003 IBC Eaton seismiccertifications issued prior to the revisions. This paper addresses therequirements of the 2006 IBC, hereafter referred to as the IBC.

According to Chapter 16 of the IBC, Structure Design, the seismrequirements of electrical equipment in buildings may be computwith two pieces of information: 1) a determination of the maximuground motion at the site; 2) an evaluation of the equipment mouing and attachment inside the building or structure. This data canthen be evaluated to develop the appropriate seismic test requirements. The ground motion, the in-structure seismic requirementsof the equipment, and the seismic response spectrum requireme

are discussed below.

A. Ground motion

According to the IBC, the first and most important step in the proceis to determine the maximum considered earthquake spectral respoacceleration at short periods of 0.2 seconds (S

s) and at a period of 1

second (S1). These values are determined from a set of 24 spectral

acceleration maps contained in the International Building Code andinclude the numerous contour lines indicating the severity of theearthquake requirements at a particular location. Great care has beetaken in selecting the maximum values for the contour lines.

For example, the maps indicate low to moderate seismic require-ments for most of the continental United States of America (USAwith exceptions being the West Coast (State of California) and theMidwest (New Madrid area). The seismic levels in the New Madr

area are approximately 30% higher than the maximum levels of thWest Coast.

The maps also suggest that the high seismic requirements in botregions, West Coast and Midwest, quickly decrease away from thhigh magnitude fault areas. These high requirements are limited trelatively local area along the fault lines. Just a few miles away frothis area, only a small percentage of the maximum requirementsare indicated.

To provide a realistic estimate of the seismic requirements for thecontinental USA, attention will initially be focused on the WestCoast, where the values noted exceed the rest of the continentalUSA, with the exception of the New Madrid area. The New Madrarea seismic requirements will be addressed separately to prevenimposing unreasonable requirements on the rest of the USA.

The worst-case conditions are formulated by selecting the mappedMaximum Considered Earthquake Spectral Response Acceleration short periods of 0.2 seconds (S

s), equal to 285% gravity, and at a 1

second period (S1), equal to 124% gravity. These accelerations will

be used to calculate the Adjusted Maximum Considered EarthquakSpectral Response Accelerations. This combination of S

sand S

1

is identified using the contour maps in Figures 12 and 13. Thesenumbers are the maximum values for the entire country, except fothe New Madrid area. These particular sites are on the border ofCalifornia and Mexico (S

1) and in Northern California (S

s). Figures 1

and 13 are developed for Site Class B, at 5% of critical damping.

To determine the maximum considered earthquake ground motionfor most site classes (A through D), the code introduces site coefficients. When these are applied against the location-specific siteclass, this produces the adjusted maximum considered earthquakspectral response acceleration. The site coefficients are defined aF

aat 0.2 seconds short period and F

Vat 1.0 second period. From t

tables in the IBC, the highest adjusting factor for SS

( 1.25) is equto 1.0 and 1.5 for S

1(> 0.5). It is important to note that the CBC

mandates the use of site class D for California.

Therefore, the adjusted maximum considered earthquake spectraresponse for 0.2 second short period (S

MS) and 1.0 second period

(SM1

), adjusted for site class effects, is determined from thefollowing equations:

SMS

= Fa

SS

= 1.0 x 2.85g = 2.85g

SM1

= FV

S1

= 1.5 x 1.24g = 1.86g

ASCE 7-05 (American Society of Civil Engineers) provides a plotshowing the final shape of the design response spectra of theground (Figure 14). ASCE 7-05 is referenced throughout the IBCas the source for numerous structural design criteria.


10/2410





Figure 12. Maximum Considered Earthquake Ground Motion for Region 1 of 0.2 sec. Spectral Response Acceleration(5% of Critical Damping), Site Class B, S

S


11/24




Figure 13. Maximum Considered Earthquake Ground Motion for Region 1 of 1.0 sec. Spectral Response Acceleration(5% of Critical Damping), Site Class B, S

1


12/2412





Figure 14. Specific Response Spectrum CurveGround

The design spectral acceleration curve can now be computed. Thepeak spectral acceleration (SDS

) and the spectral acceleration at 1.0second (S

D1) may now be computed from the following formulas in

the code:

SDS

= 2/3 x SMS

= 2/3 x 2.85g = 1.90g

SD1

= 2/3 x SM1

= 2/3 x 1.86g = 1.24g

SDS

, the peak spectral acceleration, extends between the values ofT

Sand T

0. T

Sand T

0are defined in the codes as follows:

TS

= SD1

/SDS

= 1.24/1.90 = 0.653 seconds (1.53 Hz)

T0

= 0.2 SD1

/SDS

= 0.2 x 1.24/1.90 = 0.131 seconds (7.63 Hz)

According to the IBC and ASCE 7-05, the spectral acceleration (Sa)

at periods less than 0.131 seconds may be computed by using thefollowing formula:

Sa = SDS (0.6 T/T0 + 0.4)where T is the period where S

ais being calculated.

For example, the acceleration at 0.0417 seconds (24 Hz) is equal to:

Sa

= 1.90 (0.6 [0.0417/0.131] + 0.4) = 1.12g

The acceleration at 0.03 seconds (33 Hz) is equal to:

Sa

= 1.90 (0.6 (0.03/0.131) + 0.4) = 1.02g

At zero period (infinite frequency), T = 0, the acceleration (ZPA) isequal to:

Sa

= 1.90 (0.6 [0.0/0.131] + 0.4) = 0.76g (ZPA)

The acceleration to frequency relationship in the range of 1.0 Hz toT

Sis stated equal to:

Sa

= SD1

/T

where Sa is the acceleration at period T.At 1.0 Hz (T = 1.0) this equation yields the following acceleration:

Sa

= 1.24/1.0 = 1.24g

Testing has demonstrated that the lowest dominant naturalfrequency of Eatons electrical equipment is above 3.2 Hz. Thisindicates that testing at 1.24g at 1 Hz is not necessary. In addition,having the low end of the spectra higher than realistically requiredforces the shake table to move at extremely high displacements tomeet the spectral acceleration at the low frequencies.

0.131(7.63 Hz)

1.0(1.0 Hz)

0.653(1.53 Hz)

Period T (frequency)

(SDS

)1.90g

1.24g

0.76g

SD1

Spectral Response Acceleration Sg

It is common to over test by factors of two to three times if thelow end of the spectra accommodates this acceleration component.Through testing experience and data analysis, the seismic accel-eration at 1.0 Hz is taken equal to 0.7g, which will ensure that theseismic levels are achieved well below 3.2 Hz. This yields a morevigorous test over a wider range of seismic intensities.

In developing the seismic requirements above, it is important torecognize the following:

TS

and T0

are dependent on SMS

and SD1

. If SM1

is small relative to SMS

then T

Sand T

0will be smaller and the associated frequencies will shift

higher. The opposite is also true. This must be realized in develop-ing the complete RRS. Therefore, it is not adequate to stop the peakspectral acceleration at 7.35 Hz. There are other contour line combi-nations that will produce different values for T

Sand T

0. In accounting

for this variation in the spread between SMS

and SD1

and the resultingimpact on T

Sand T

0, it is almost impossible to consider all combina-

tions. A study of the maps, however, suggests that all variations withhigh magnitude of contour lines could very well be enveloped by afactor of 1.5. Therefore, T

0is recomputed as follows:

T0

= 0.2 SD1

/(SDS

x 1.5) = (0.2 x 1.24)/(1.90 x 1.5) = 0.09 seconds(11.0 Hz)

Based on past experience, most electrical equipment exhibitsnatural frequencies in the range of 5 to 10 Hz. Therefore, they aretested to the peak spectral accelerations required by the code. It isalso important to recognize that stopping the peak accelerationshorter than 11 Hz would produce questionable test results due tothe elimination of a portion of the spectra that may well contain thenatural frequency of the equipment.

Eaton has developed generic seismic requirements that enveloptwo criteria:

1. The highest possible spectral peak accelerations and ZPA

2. The maximum frequency range required for manydifferent sites

This approach results in a comprehensive and ultra conserva-tive methodology in certifying equipment to the IBC and oftenexceeds the approach utilized by other manufacturers.

Within the electrical industry, some manufacturers cease theseismic peak spectral acceleration at 7 or 8 Hz. This substantiallyreduces the amount of energy and frequency content included inthe input time history. There are many certifications issued by othermanufacturers that claim qualification to 3 or 4g spectral accelera-tion. This raises the question: What is the true acceleration thatwas measured at the natural frequencies of the equipment?

It is very likely that the equipment dominant frequencies were onlytested to a fraction of what is claimed. Therefore, the claimed curveshould be reduced to the actual spectral acceleration at the domi-nant natural frequencies of the equipment. Eaton accounts for thatby testing to peak spectral accelerations even beyond 11.0 Hz.

This completes the ground motion design response spectrum. Thespectral accelerations are equal to 0.76g at zero period (ZPA) andincreases linearly to a peak acceleration of 1.90g at 0.09 seconds

(or 11 Hz) and stays constant to 0.313 seconds (or 3.2 Hz), thengradually decreases to 0.7g at 1 second (or 1.0 Hz). This finalcurve is shown in Figure 15.


13/24




Figure 15. Specific Response Spectrum CurveGround

This curve indicates the ratio of peak spectral acceleration tomaximum input acceleration (ZPA) is 1.90/0.76 and approximatelyequal to 2.5. This ratio is maintained throughout this document.

The code does not provide formulation for the seismic spectralrequirements inside buildings or above grade. Instead, the codeprovides formulation of the equivalent loads at the center of gravity(C.G.) of the equipment internal to structures or buildings. Thepurpose is to ensure the structural and mounting integrity of theequipment during and immediately after a seismic event. Theserequirements will be discussed to determine realistic seismicrequirements for equipment mounted anywhere from the ground

level to the roof of a particular building.

B. Seismic requirements of equipment installed internaloron topof structures (buildings)

The code provides a formula for computing the seismic require-ments of electrical and mechanical equipment on ground level of astructure or a building. This formula is designed for evaluating theattachment of the equipment to the foundation directly supporting it.

The seismic loads are defined in ASCE 7-05 Section 13.3 as:

Fp

= 0.4 ap

SDS

Wp

(1+2 Z/h) / (Rp/I

p)

15, 0.09(11 Hz) (T

0)

1.0(1.0 Hz)

0.313(3.2 Hz) (T

S)

(SDS

)1.90g

0.7g

0.76g

SD1

Spectral Response Acceleration Sg

Period T (frequency)

Where:

Fp: seismic design force imposed at the components C.G. and

distributed relative to component mass distribution

ap: component amplification factor that varies from 1 to 2.50

SDS

: spectral acceleration, short period, as determined in theprevious section

Wp: component operating weightR

p: component response modification factor that varies

from 1.5 to 6.0 (ASCE 7-05 Table 13.6-1)

Ip: component importance factor of either 1.0 or 1.5

Z: highest point of equipment in a structure relative tograde elevation

h: average roof height of structure relative to grade elevation

To produce the maximum required force, the following parameterwere chosen:

Z is taken equal to h (equipment on roof)

Ip

is taken as a maximum equal to 1.5

ap

is taken equal to 2.5 (maximum value allowed by the ASCE cod

Rp is taken equal to 2.5 (lowest value allowed by the ASCE codefor electrical distribution and control equipment).

This combination of ap

and Rp

provides the most conservativeseismic loading requirements.

SDS

has been computed in the previous section equal to 1.90.

The acceleration at the equipment C.G. when roof mounted is thecalculated as:

Acceleration = 0.4 x 2.5 x 1.90g (1+2) / (2.5/1.5) = 3.42g

For equipment on grade, the acceleration at the C.G. is thencalculated as:

Acceleration = 0.4 x 2.5 x 1.90g (1+0) / (2.5/1.5) = 1.14g

Based on this criterion, in order to establish the seismic acceptab

ity of equipment inside a structure or a building, one must imposean equivalent static load at the equipment C.G. and record theresults. This approach is very difficult and perhaps impossibleto apply.

The C.G. would first have to be located and then physically coupleto a forcing mechanism supported by some type of a fixture and areaction mass. This approach would provide incomplete data or anysis. Applying a static load will push the entire system as one unithe force direction without revealing sufficient data about the equment flexibility, the relative motion of internal components to thecabinet structure, or the dynamics and resonance of the equipme

A more realistic approach with enhanced test results is to exposethe equipment to floor motion, causing the inertia forces to occuin the opposite direction when the mass is suddenly accelerated.Bolting the base of a piece of equipment to a shake table, then

quickly accelerating it, results in exposing the equipment to inertialoads higher than the source input.


14/2414





As explained previously, many seismic test programs clearly indicatethat electrical equipment, which is supported at the base, tends tovibrate in the equipments natural dominant frequency, much like afree cantilever beam that is supported at the bottom and free at thetop. The seismic response at the middle of the equipments C.G.is at least 50% higher than the floor input of the equipmentsnatural frequency.

Therefore, the base forces associated with the static loads at theC.G. of the equipment could be computed as 3.42/1.5 = 2.28g. TheZPA associated with this spectral acceleration may be computedper the previous relationships established.

In the context of this discussion, Eatons seismic requirements tomeet the IBC (Reference 1) are:

For equipment on grade, the base seismic requirements areshown in Figure 15.

For equipment on the roof of a structure, the base inputacceleration at the equipment natural frequency mustdemonstrate the ability to withstand levels of 2.28g baseacceleration or 3.42gs at the equipment C.G.

C. New Madrid seismic requirements

According to the IBC, the New Madrid fault maximum considered

earthquake spectral response acceleration is Ss = 3.69g andS

1= 1.25g. The method to develop the required spectrum and

required forces at the C.G. is the same as described above. Basedon the exercise in the previous section, and by virtue of theequations being of the first order, the requirements can be directlydetermined by linearly increasing the complete levels and static forcerequirements by the ratio of 3.69/2.85 = 1.29. The resultant RSC isshown in Figure 16.

The maximum seismic forces at the C.G. for equipment mountedat the top floor will be equal to 1.29 x 3.42 = 4.41g or 2.94g peakspectral acceleration.

Eatons seismic requirements for (equipment on or in proximity to)the New Madrid area is:

1. For equipment on grade, the base seismic requirements are

shown inFigure 16

.2. For equipment inside a structure or on top of the roof, the

base input acceleration at the equipment natural frequencymust exceed the levels of 2.94g base acceleration or 4.41g atthe equipment C.G.

Figure 16. Response Spectrum CurveGround(New Madrid Area)

Period T

2.46g

0.83g

1.0g

To Ts 1.0

California Building Code (CBC) 2007

The 2007 CBC, effective January 1, 2008, adopted the 2006 IBC asCBC-Title 24. The seismic requirements are essentially the sameas described in the IBC, with some minor modifications. Whenconsidering the maximum seismic requirements, the IBC and CBCare basically identical.

Again, as in the IBC, the RSC starts at 1.24g (Sa) at 1.0 Hz, and

increases to 1.90g (SDS) at 1.53 Hz (Ts). The peak spectral accelera-tions then cover a wide band of frequencies up to 7.63 Hz (T

o) then

linearly decrease to 0.76g at the ZPA.

Combined Seismic Requirements for Eatons Distributionand Control Equipment

To better compare all levels and determine the final envelopingseismic requirements, the IBC standards are used for California andNew Madrid areas, as plotted in Figure 17. All curves are plotted at5% damping. All curves are determined for equipment mounted ongrade or in the basement of the structure.

An envelopment of the seismic levels in the frequency range of 3.2Hz to 100 Hz are also shown. This level is taken as Eatons genericseismic test requirements for all certifications. These levels are alsoplotted in Figure 18. The levels are listed below:

Frequency Acceleration

1 0.719

3.2 2.28

11 2.28

33 1.02

100 1.02

Many standards require that seismic levels be increased by 10% toaccount for differences in commercial hardware. Applying this willbring the spectral peak acceleration to 2.51g and the ZPA to 1.12g.

Frequency Acceleration

1 0.7

3.2 2.51

11 2.51

33 1.12

100 1.12

The vertical levels are taken equal to 2/3 of the horizontalrequirements.

In addition, Eaton performs seismic tests on the equipment atapproximately 120% of the generic enveloping seismicrequirements (see Figure 18). This testing is designed toestablish margin in anticipation of future changes in the codes.For seismic certification of equipment located in the New Madridarea, Eaton proceeds as follows:

Complete testing to the generic levels in Figures 17 and 18.

Perform additional tests at approximately 20% higher seismic levelsthan shown in Figures 18.

During September 2008, Eaton performed experimental seismictesting on electrical equipment levels higher than the combinedrequirements. Some of the equipment test results are shown inFigures 19, 20, and 21. The levels are provided in the front-to-back,side-to-side, and vertical directions. As indicated, the actual testlevels recorded were much higher than current codes require.

19 See discussion under A. Ground motion onpage 9 for

acceleration at 1 Hz.


15/24




Figure 17. RRS Comparison

Figure 18. 100% vs. 120%

Figure 19. Test Response Spectrum Curve (Front to Back)

Acceleration

(g)

0.1

1

10

1 10 100

Frequency (Hz)

Eaton Seismic

IBC 2006 New MadridIBC 2006/CBC 2007

Acceleration

(g)

0.1

1

10

1 10 100

Frequency (Hz)

Eaton 100% Seismic Envelope

Eaton 120% Seismic Envelope

Acce

leration

(g)

0.1

1

10

1 10 100

Frequency (Hz)

FB TRS

FB RRS

Figure 20. Test Response Spectrum Curve (Side to Side)

Figure 21. Test Response Spectrum Curve (Vertical)

Acceleration

(g)

0.1

1

10

1 10 100

Frequency (Hz)

SS TRS

SS RRS

Acceleration

(g)

0.1

1

100

10

1 10 100

Frequency (Hz)

V TRS

V RRS


16/2416





Part IV

Test facility and test methodology

Test specimens

Since the inception of Eatons test program in 1985, more than100 specimens have undergone seismic testing. Since it was notfeasible to test every single configuration, it was necessary to

select a number of test specimens that adequately representthe total product portfolio. Each product line was reviewed andevaluated to determine the number and design configurations ofthe test specimens. Criteria were established for representationof all equipment in each product line:

1. The test unit structure shall be similar to the major structuralconfigurations being supplied in the product lines. If morethan one major structure is being offered, then theseconfigurations shall also be selected for testing.

2. The mounting configuration of the test units to the shaketable shall simulate the different mounting conditions for theproduct line. If several mounting configurations are used, thedifferent product variations are required to be included inthe testing program.

3. The major electrical components should be included inthe test specimens. The components shall be mounted atsimilar locations to their mounting locations in productionconfigurations. The components shall be mounted to thestructure using the same mounting hardware used in thetypical design.

4. The weight of the test units shall be similar to the typicalweight of the equipment being represented. Equal andhigher weights than the typical weight shall be acceptable.

5. Other variations, such as the number of structures inproduction assemblies, and indoor and outdoor applications,will also be represented by the test specimens.

Figure 22. Test Specimen

Side-to-SideBase Input

Front-to-BackBase Input

TypicalTest Unit

VerticalBase Input

Shake Table/Base Motion

Accelerometers

ResponseAccelerometers

ShakeTable

Test facility and test table

Test specimens for current production products are tested ona truly independent triaxial shake table at the locations such asWyle Seismic Test Laboratory, located in Huntsville, Alabama.Wyle Laboratories is accredited by the American Association forLaboratory Accreditation (A2LA) in the field of vibration testing. TheWyle Laboratories, Huntsville Facility, Quality Management Systemis registered in compliance with the ISO-9001 International QualityStandard. All instrumentation, measuring, and test equipmentused in the performance of test programs is calibrated in accordancewith Wyle Laboratories Quality Assurance Program, which complieswith the requirements of ANSI/NCSL Z540-1, ISO 10012-1, andISO/IEC 17025. The table and control systems are capable ofexciting the test specimens in all three directions simultaneously,using statistically independent and phase incoherent seismic inputsignals. A sketch of a test unit mounted to the shake table is shownin Figure 22.

Test sequence

The seismic verification testing consisted of the following 10 stepsfor each specimen:

1. Receipt and inspection

2. Functional operation

3. Hi-pot electrical testing

4. Resonance search testing

5. Seismic test at 50% of the combined seismic requirements

6. Seismic test at 100% of the combined seismic requirements

7. Seismic test at higher than the 100% combined seismicrequirements (typically 120%130%)

8. Functional operation

9. Hi-pot electrical testing

10. Final inspection

Resonance search test

Resonance search (sine sweep) tests are performed on all testspecimens. The sine sweep tests are performed in the three prin-cipal axes of the test specimens: front-to-back, side-to-side, andvertical directions. The sine sweep tests are conducted at amplitudeof 0.2g. The sine sweep tests are performed from 1 to 50 Hz at asweep rate of 1 octave per minute.

Seismic test input

The seismic inputs are generated using random signals with afrequency and energy content up to 100 Hz. The test inputs areindependent in the three principal directions of the test specimens:front-to-back, side-to-side, and vertical directions. All seismic test

inputs are 30 seconds in duration (see Figure 23).

Data acquisition

The test inputs to the shake table are monitored using threeaccelerometers mounted on the shake table. The accelerometersare oriented in the shake table principal axes, which coincide withthe equipment front-to-back, side-to-side, and vertical directions.The seismic response of the test specimens are monitored usingseveral accelerometers mounted on the test specimen and orientedalong the three principal axes of each test specimen. The testinput and seismic response of the equipment is recorded on andanalyzed using a shock spectra analyzer. The test responsespectra are derived at 5% damping (see Figures 23 and 24).

Electrical connection and test specimen monitoring

As stated previously, the acceptability of the test specimen requiresthat all equipment maintain structural integrity and perform its

intended function before and after the seismic test.


17/24




Test assembly and mounting conditions

At the beginning of each test, the test units are mounted to the shaketable (Figure 25) using the specified seismic mounting conditions.

Figure 23. Test Input

Figure 24. Test Data Acquisition

3.0

G

0.0

2.00.00 30.00

Time (sec) x interval = 2.0000

MIN = 0.1921E+01 MAX = 0.1588E+01

Min/max x:B in size = 8

Figure 25. Typical Equipment Mounting and Installationof Accelerometers

Test procedure

All test specimens identified in Part VI (Figure 26 See note on

pp. 20 regarding list of products.) are subjected to the seismic tesrequirements specified in Figure 18. Testing is conducted inaccordance with IBC (ASCE 7-05), CBC, and ANSI C37-81 testrequirements. The test programs are documented in third-partylaboratory test reports.

Acceptance criteria

The seismic verification of the test specimens was based on thefollowing acceptance criteria:

1. The test specimens structure shall maintain structuralintegrity with no major structural failure that may impactthe electrical performance of the test specimens or impactadjacent equipment.

2. The test specimens shall perform their electrical functionimmediately after seismic testing.

3. The test specimens shall pass one minute dielectricwithstand testing per the associated industry standardsafter seismic testing.


18/2418





Part V

Shared responsibilities

The equipment manufacturer, specifier, Authority Having Jurisdiction(AHJ), and installer have a shared responsibility to ensure that theinstallation will meet the seismic requirements of the code. Theequipment manufacturer determines that the equipment will befunctional following a seismic event. The equipment specifier and

installer must ensure that the equipment is rigidly supportedand will not leave its foundation during a seismic event. The AHJshall confirm that the installation method conforms to themanufacturers guidelines.

Previously in this paper, the Eaton interpretation of the various codesand standards, as well as the levels of the test response spectraused in testing, was described. The test results ensure that Eatonequipment will perform the intended function after the seismicevent. However, the foundation and the anchorage must also meetthe codes and standards for the entire installation to be functionalafter a seismic event. Equipment poorly anchored or mounted to aflexible foundation will not meet the requirements.

The anchoring of electrical equipment as recommended by thestructural or civil engineer is critical. If the equipment is not attachedto the building structure in accordance with the minimum standards

recommended, the complete equipment installation might becometoo flexible and may overturn or shear the attachment devices andslide off its foundation. Such movement may damage either thebuilding structure or other components, including items connectedto the equipment.

Structural and civil engineers formulate methods of attachment thatare applicable to each specific condition based on past experience.They evaluate the equipment, methods, and techniques of attach-ment along with tested anchoring systems. The structural or civilengineers responsible for the structural design review the proposedmethod of attachment. Based on both established criteria and directcalculation, they verify its performance and the capability of thebuildings structural elements to accommodate the seismic forces.In many states, registered professional civil or structural engineersmust attest that the design is adequate for the seismic environmentand apply their seal.

In evaluating the equipment mounting, the structural or civil engineerperforms calculations based on data received from the equipmentmanufacturer specifying the size, weight, center of gravity, andmounting provisions of the equipment. The embedded concreteanchors, wood, or steel attachments must be adequate to resist thesite-specific seismic forces. For either attachment, bolts of theproper grade of steel must be correctly sized and tightened torecommended torque levels. If an embedded anchor is used, engi-neering data for the anchoring hardware will allow the engineer todetermine the size required. The mounting depth and the strengthof concrete to contain it will also be determined. The embeddedanchors must be correctly installed in accordance with the methodspecified by the anchor manufacturer.

The reliability of electrical connections within the system must alsobe considered. Electrical equipment is installed as part of a system.Busway or conduits connect individual components of the electrical

system to each other. Interface methods that will prevent damagefrom an earthquake must be specified. For example, bottom entryof conduits is recommended for transformers and switchgear.If top entry is specified, seismic fittings or a flexible interfacedesigned to accommodate the necessary enclosure motionare needed. Transformers are often close coupled to switchgearwith a flexible connector to minimize transfer of relative motion.Likewise, a flexible connector can be used to connect generatorsto the bus duct, and the addition of insulating boots improves theintegrity of such installations.

The availability of electrical power following a disaster is oftencritical. It is certain that earthquakes will occur in the future. It isthe responsibility of the engineer to design and specify reliableequipment and systems that will withstand them. The IBC and CBCestablish minimum requirements for equipment seismic design and

installation. As required by the IBC and CBC, Eaton has equipmentavailable that has been seismically certified. When specified, suchequipment increases the likelihood that the electrical system willfunction in the aftermath of an earthquake.

Part VI

Typical Eaton seismic equipment specifications

1.01 The manufacturer of the assembly shall be themanufacturer of the major components withinthe assembly.

1.02 For the equipment specified herein, the manufacturer shallbe ISO 9001 or 9002 certified.

1.03 The manufacturer of this equipment shall have producedsimilar electrical equipment for a minimum period of five(5) years. When requested by the engineer, an acceptablelist of installations with similar equipment shall be provideddemonstrating compliance with this requirement.

1.04 Provide seismic qualified equipment as follows:

ote:N To spec writer: To help understand the 2006 IBC/2007CBC seismic parameters for a specific location, the attached linkto the U.S. Geological Society will be extremely helpful:

http://earthquake.usgs.gov/research/hazmaps/design/

Download the file Java Ground Motion Parameter CalculatorVersion5.0.8 (4.6 MB) and save it to your hard drive, then run the executablefile (.exe) that was downloaded.

Enter the latitude and longitude of your project location.(To find exact latitude and longitude for a given address,go to http://geocoder.us/)

The IBC seismic criteria for that location will then be displayed.It is simply a matter of verifying that the criteria shown for your specific

building location is less than the equipment certification levels.

1. The equipment and major components shall be suitable forand certified by actual seismic testingto meet all applicableseismic requirements of the 2006 International Building Code(IBC) Site Classification [enter classification from above Web

site]. The site coefficients Fa = [enter value from above Website], and spectral response accelerations of SS

= [entervalue from above Web site]g, S

1= [enter value from above

Web site]g are used. The test response spectrum shall bebased upon a 5% damping factor, and a peak (S

DS) of at least

[enter value from above Web site] gs (312 Hz) applied atthe base of the equipment in the horizontal directions. Theforces in the vertical direction shall be at least 66% of thosein the horizontal direction. The tests shall cover a frequencyrange from 1 to 100 Hz. Guidelines for the installationconsistent with these requirements shall be provided by theequipment manufacturer and based upon testing of repre-sentative equipment. Equipment certification acceptancecriteria shall be based upon the ability for the equipmentto be returned to service immediately after a seismic eventwithin the above requirements without the need for repairs.

-- OR --

2. The manufacturer shall certify the equipment based upona dynamic and/or static structural computer analysis of theentire assembly structure and its components, provided it isbased upon actual seismic testing from similar equipment.The analysis shall be based upon all applicable seismicrequirements of the 2006 International Building Code (IBC)Site Classification [enter classification from above Web site],site coefficient F

a= [enter classification from above Web

site], FV

= [enter classification from above Web site] andspectral response accelerations of S

S= [enter classification

from above Web site]g, S1

= [enter classification from aboveWeb site]g. The analysis shall be based upon a 5% damp-ing factor, and a peak (S

DS) of at least [enter classification


19/24




from above Web site]g, S1

(312 Hz), applied at the base ofthe equipment in the horizontal directions. The forces in thevertical direction shall be at least 66% of those in the hori-zontal direction. The analysis shall cover a frequency rangefrom 1 to 100 Hz. Guidelines for the installation consistentwith these requirements shall be provided by the equipmentmanufacturer, and should be based upon testing of repre-sentative equipment. Equipment certification acceptance cri-

teria shall be based upon the ability for the equipment to bereturned to service immediately after a seismic event withinthe above requirements without the need for repairs.

A. The following minimum mounting and installation guidelinesshall be met, unless specifically modified by the abovereferenced standards.

1. The contractor shall provide equipment anchorage details,coordinated with the equipment mounting provision,prepared and stamped by a licensed civil engineer in thestate. Mounting recommendations shall be provided by themanufacturer, and should be based upon the above criteriato verify the seismic design of the equipment.

a. The equipment manufacturer shall certify that theequipment can withstand, that is, function following theseismic event, including both vertical and lateral required

response spectra, as specified in above codes.

b. The equipment manufacturer shall document the require-ments necessary for proper seismic mounting of theequipment. Seismic qualification shall be consideredachieved when the capability of the equipment meetsor exceeds the specified response spectra.

Eatons equipment test levels and ICC-ES-AC156

In December 2006, the ICC-ES issued an Acceptance Criteriafor Seismic Qualification by Shake-Table Testing on NonstructuralComponents and Systems. The criteria was made effective January1, 2007. Eatons methodology for seismic certification of electricalequipment is consistent with the proposed criteria and meets thetesting requirements specified. Eaton, however, differs in one impor-tant aspect: Eaton has taken the ratio of the equipment response

modification factor (RP) to equipment importance factor (IP) equal to2.5/1.5. This ratio provides the minimum ratio required by the codesfor electrical distribution and control equipment, and also consid-ers that the acceleration is required to be measured at the centerof gravity of the equipment. The ICC-ES-AC156 employs a factor of1.0 to this ratio producing unnecessary and overtesting conditions.One additional difference needs to be mentionedEatons electricalequipment, with high natural frequencies (per ICC-ES-AC156), arealso tested and certified to the same seismic test input as flexibleequipment. ICC indicates that this equipment may be tested to 0.4of the seismic levels developed for flexible equipment. Eatons testprogram is more conservative by testing all equipment types to thehighest levels.

In addition, an important note should be made regarding the mount-ing configurations of the test units. Eatons equipment is mountedto the shake table in their most conservative and common mounting

configurations to establish the lower bound of the equipmentseismic capabilities. For example, Eaton seismic certification curvesfor motor control centers (MCCs) are based on a test unit mountedat the base as a free cantilever item, free at the top and supportedonly at the bottom. This test configuration encompasses allother mounting configurations because of its conservative nature.The test capability in Eatons certificates, therefore, coversall other applications.

When the MCC test units were tested supported at the bottomand with the top attached to a lateral wall, the seismic capacityof the test units were found to be much higher than their seismiccapacity when supported at the bottom only. The seismic capac-ity of equipment presented in some papers appears to be basedon testing of equipment with both top and bottom supports. It isimportant, therefore, to recognize that those published curves onlapply to equipment mounted using top and bottom supports. Oth

mounting arrangements without top lateral supports will need to bre-established based on new testing programs.

3RD PARTY TEST ENGINEER IN CHARGE

S E I S M I C Q U A L I F I E D

TEST CERTIFICATE OF SEISMIC WITHSTAND CAPABILITY

Eatons Cutler-Hammer equipment identified below was tested for seismic withstand capability and tested in accordance

with the combined requirements specified in the International Building Code, California Building Code and the Uniform

Building Code. As required by the codes, the equipment demonstrated its ability to function after the seismic tests.The seismic capability of the equipment exceeds the worst-case required levels, as illustrated in the figure below.

For interpretation of testing datarefer to Eaton

Publication SA12501SE

The frequency sweep

tests revealed that the

lowest equipment natural

frequency is:

0HWDO(QFORVHG/RZ9ROWDJH6ZLWFKJHDU0DJQXP'6

)URQW$FFHVVLEOHZLWK7\SH0'6&LUFXLW%UHDNHUVRU

&01HWZRUN3URWHFWRUV

1DWKDQ*OHQQ3(:HVWLQJKRXVH(OHFWULF&RPSDQ\//&

7(67('%

earthquake requirements and seismic capabilities for eaton’s electrical distribution and control...

Documents