1、2008 ASHRAE 51ABSTRACT The newer building codes contain more stringent require-ments for both the attachment and the continued operation ofequipment used in critical seismic applications. This require-ment states basically that if the facility is critical, the equip-ment required to keep if function
2、al is also critical and it mustbe qualified to ensure that it will continue to perform after theevent. For the limited applications in the past, Shake table test-ing was preferred, however Shake table testing on large piecesof equipment can be extremely expensive, time consuming andoften provides re
3、sults that cannot be extended to other appli-cations for the same piece of equipment. In addition, the limitednumber of facilities that can do this kind of work often rendersthis approach impractical. This paper discusses the changesmade in the building codes leading up to the current require-ments
4、and offers some guidance as to various options that canbe pursued by equipment manufacturers or others seeking toqualify equipment for critical seismic applications. INTRODUCTION Over the years, seismic components used in the HVACindustry have been tested in a number of fashions for a numberof diffe
5、rent purposes. Most of the analysis and testing has beengeared toward the restraint of equipment and more particu-larly, in evaluating the anchorage hardware used to performthis function. Components that are evaluated are purpose builtrestraint components, housed isolators, cable and strutsystems as
6、 well as the actual anchor bolts, lag screws, capscrews or welds used to attach the restrained system to thestructure. Static testing has been the most common method of testfor these components. When performing these tests, a knownstatic load is applied to the device in any of several “worstcase” di
7、rections. This load is increased until the tested compo-nent has been determined to fail using accepted industry crite-ria. The “failure“ load is noted and a rating is set for the devicethat is some fixed fraction of the measured failure load. Testinghas historically been done on a small sample size
8、 and to ensureconfidence in the rating, the ratio between the test result loadand the rating is relatively large (typically 2:1).Equipment is treated as a “black box”. The assumptionused to develop the test procedures was that the equipment andstructure were considerably more rigid than the attachme
9、ntdevices/hardware and no consideration was given to continuedoperability. The goal was simply to hold equipment in placefollowing an earthquake. The “black box” concept allows therestraint devices to be selected as stand alone components andany interdependence that would result from a system approa
10、chcan be ignored.Concrete anchors are also tested and rated as stand aloneentities. For these tests, a reasonable sample size is identifiedand the final rating of the anchor is based on a review of theperformance of the test sample group as a whole. Past versions of the building codes have required
11、an anal-ysis of equipment installations using static analysis proce-dures. Because the actual load is dynamic in nature, a factor isapplied to the computed static load that varies with both thetypes of equipment and their mounting arrangement. Thisfactor “amplifies” the design load condition to refl
12、ect thedynamic factors (mounting arrangement and equipment dura-bility). This amplified load is compared to the static load ratingon the restraint component as well as to the ratings of themounting hardware to determine their adequacy.The Impact of Seismic Testing on the HVAC IndustryPaul MeiselMemb
13、er ASHRAEPaul Meisel is Vice President of Engineering at Kinetics Noise Control, Dublin, OH.NY-08-0092008, American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc. (www.ashrae.org). Published in ASHRAE Transactions, Volume 114, Part 1. For personal use only. Additional reprodu
14、ction, distribution, or transmission in either print or digital form is not permitted without ASHRAEs prior written permission.52 ASHRAE TransactionsWith time, the amplification factors have been modifiedto the point that the inter-relationship between these factorsand the capacities assigned to com
15、ponents by the static testprocedures produce a consistently reliable restraint inter-face. Except in a few extremely critical applications, the factthat equipment might or might not continue to function at theconclusion of an event has not until recently been deemedcritical.STATIC TESTING AND CERTIF
16、ICATIONStatic tests are performed on restraint components andseismically qualified isolators by applying a series of loads tothe device at the surface or bolted connection that forms theattachment between the restraint and the equipment. Theconnection point between the restraint and the structure is
17、 heldrigidly in place typically against a flat surface or “platen”.During the course of testing, various restraint samples areoriented in different directions to ensure that all “worst case”conditions are evaluated.Loads are applied at increasingly higher levels andrelaxed between cycles until eithe
18、r a brittle failure occurs orthe restraint device shows evidence of yielding. Typicallycompressive and tensile loads are applied in the vertical direc-tion, loads are applied along the horizontal East/West axis, theNorth/South axis and lastly, simultaneous loads combiningcombinations of the vertical
19、 and horizontal loads are used todetermine a peak capacity. Between load cycles, the plasticdeformation in the restraint device is measured and recordedand once a total deformation equivalent to 0.2% yield has beenachieved, the component is deemed to have reached its peakcapacity limit. This type of
20、 testing is easy to perform, offers consistentresults, is quantifiable and is not subject to the variables thatplague some of the dynamic testing techniques. Although wellsuited to structural design issues, it does not address thedynamic component interactions that are prevalent when eval-uating mor
21、e complex mechanical systems. Unlike many of the dynamic tests, because of its quanti-fiability and its treatment of equipment as “black boxes” thistype of testing allows a rating to be generated on a particularrestraint device without regard to the equipment type on whichit is mounted. This rating
22、can be used to “size” the restraintdevice based on code generated requirements and offers muchmore useful input than the Pass/Fail result of most dynamictest procedures.DYNAMIC TESTING AND COMPONENT CERTIFICATION1997 NEHRPThe 1997 NEHRP (National Earthquake Hazard Reduc-tion Program) Provisions, in
23、section 6.3.6 initiated the trendtoward ensuring the continued operation of critical equipmentat the conclusion of a design level seismic event. In this docu-ment, a Component Certification is defined as being a certif-icate provided by the Equipment Manufacturer signifying thatthe equipment provide
24、d “complies with the projected forcerequirements for the application”. Further, it indicates that theneed to satisfy this requirement can be so indicated in thecontract documents or at the request of the “Authority havingJurisdiction”.While this statement would tend to imply that only thestructural
25、integrity of the equipment is in question, thecommentary portion of the document indicates otherwise.Because these comments comprise only 2 short paragraphs, Iwill quote them here in their entirety. The first paragraph is from the 1997 NEHRP Provisions:Component Certification: It is intended that th
26、e certifi-cate only be requested for components with an impor-tance factor (Ip) greater than 1.00 and only if thecomponent has a doubtful or uncertain load path. Thiscertificate should not be requested to validate function-ality concerns.The second paragraph is from the 1997 NEHRPCommentary for the
27、same section:In the context of the Provisions, seismic adequacy of thecomponent is of concern only when the component isrequired to remain functional after an earthquake orcontains material that can pose a significant hazard ifreleased. Meeting the requirements of this section shallbe considered as
28、an acceptable demonstration of the seis-mic adequacy of a component.The general interpretation of this is that, while equipmentneed not remain running through a seismic event, it must beable to resume operation if restarted at the conclusion of theevent. In the case of hazardous materials, it must n
29、ot leak.2000 IBCThe 2000 IBC is patterned off of the 1997 NEHRP docu-ment and is slightly more specific. It indicates in section1621.3.5 that a “Manufacturers certificate of compliance withthe force requirements of Section 1621” be submitted to thebuilding official. This makes the requirement mandat
30、oryrather than discretionary. Again, there is no clear definition ofthe scope of the certification in the code language, but the orig-inal intent was for it to follow the same logic as is presentedin the 97 NEHRP document.More guidance is provided in section 1707.7.1 of the 2000IBC code. Here there
31、is a clear definition of acceptable meth-ods that can be employed to substantiate this certificate. Test-ing is specified to address both the component and thecomponent mounting system. Optional test methods includeshake table, three-dimensional shock tests, analytical methodsusing forces and dynami
32、c characteristics derived from theseismic section of the code or by “more rigorous analysis”.2003 IBCThe 2003 edition of the IBC calls for seismic design to bein accordance with ASCE 7-02. ASCE 7-02 also includesComponent Certification and is the first document that clearlyASHRAE Transactions 53iden
33、tifies in the code language that it is linked to the continuedoperability of equipment. As a result, this is the first occasionwhere this provision has been actively enforced and has had tobe addressed in the building community. For reference,Section 9.6.3.6 of ASCE 7-02 states that equipment with a
34、nimportance factor of greater than 1.0 in seismic design cate-gories C, D, E and F shall not only be durable enough to remainin place through a seismic event, but also “must remain oper-able following the design earthquake”.Section 9.6.3.6 further indicates that qualification for thecertification ca
35、n be met with either shake table testing, three-dimensional shock tests or experience data and the Manufac-turer shall provide the necessary documentation. Analysis forsome reason is not identified as an option.2006 IBCThe most current code version is the 2006 IBC. It links toASCE 7-05 in a similar
36、fashion as the 2003 code links toASCE 7-02. The basic design criteria of continued equipmentoperability for critical equipment is clearly identified in thenew code for “Designated Seismic Systems”. “DesignatedSeismic Systems” are those services that must be maintainedin the structure following the c
37、onclusion of a seismic event.Testing and experience data have been retained as possiblequalification techniques, however analysis has been added asa third option and three-dimensional shock tests have beenomitted. Also specified is that the Certification shall be providedby the Equipment Supplier.TE
38、STING OPTIONS AND LIMITATIONSShake Table TestingIn order for a test to be valid, the test must as a minimum,accurately reflect the peak load conditions that would beencountered during the design seismic event. Because manu-facturers do not want to certify equipment over and overagain, it is typical
39、that the testing is done to the highest levelfor which the manufacturer anticipates selling the equipment.This allows the equipment to be used in all less stringentapplications. Because the equipment being tested is often large andexpensive, it is typical that only a single unit is evaluated.General
40、ly ignored during the testing process is that seismicprofiles for different events in different geographic areas varysignificantly and that these variations can affect internalcomponents and their reliability differently. Normally a seis-mic profile from a past recorded event (preferably from theare
41、a to which the equipment would be going), is scaled up ordown to meet the desired peak criteria. The end result fromtesting is a Pass/Fail verdict and unlike the older static analysisand testing procedures, it cannot be easily extrapolated forother applications. The equipment support system will als
42、o affect the loads towhich the equipment is subject. Codes generally call for abouta 2:1 increase in design forces for the same equipmentmounted on isolators as compared to the forces used if directlybolted to the structure.To make matters more complicated, there are a limitednumber of tri-axial tes
43、t facilities with the capability ofhandling large pieces of equipment (I believe that there areonly 3 in the US). Test time is expensive and the waiting listis long.The combination of the above factors makes it very diffi-cult for manufacturers to comply through testing and still meetcost and delive
44、ry targets. In an attempt to simplify part of the analysis/testingprocess, ASHRAE is working in conjunction with NSF(National Science Foundation) to clearly evaluate the shockfactors that result from isolating equipment. The goal is toallow a single dynamic test to be run on the equipment thatwould
45、be valid for both hard mounted and isolated equipmentconditions. These relative shock factors could be directlyapplied to the dynamic test data and used to determine limitingseismic load conditions for non-isolated and isolated cases.These factors can also be used in conjunction with the devel-oped
46、static ratings on restraint components to further ensuretheir applicability to different applications. The tests beingconducted at MCEER (Multi-Disciplinary Center for ExtremeEvent Research, part of the State University of New York inBuffalo) use special generic restraints that allow gaps (orclearan
47、ce), snubber element thickness and material durometer(hardness) to be adjusted. The restraints are then installed onvarious types of equipment and subjected to a range of differ-ent amplitude seismic inputs to establish appropriate impactfactors for a variety of different component types and systema
48、rrangements. Combination Test/AnalysisA second method employed to evaluate equipment dura-bility is a combination of component testing and a rigorousdynamic analysis of the structure. This offers significantbenefits when equipment manufacturers offer a range ofequipment made out of common modules or
49、 components.The first benefit is that these smaller components can betested on more readily available, less expensive shake tables.Second, the test results can be shared amongst all of the equip-ment that might use the same component. In some cases, thetest data for one component can even be extrapolated to othersimilar components. Moving beyond the tested mechanical components, thestructure of the equipment can be evaluated using a wide rangeof techniques ranging from finite element to more classicalprocedures. Where required, additional reinforcement can bespecified and added for part
