Subject: RE: low vibration building Date: Fri, 6 Sep 2002 19:30:28 -0400 X-MS-Has-Attach: yes X-MS-TNEF-Correlator: Thread-Topic: low vibration building Thread-Index: AcJOJF2kZ/j6lw8BSd+6JSTrlWuu6QH19QOQ From: "Soueid, Ahmad" To: Cc: , "Hal Amick (E-mail)" X-OriginalArrivalTime: 06 Sep 2002 23:30:29.0953 (UTC) FILETIME=[63388310:01C255FD] Thank you for your question Prof. Packard, I am sorry it took me so long to answer your question since I was in the UK the earlier part of this week. I discussed your concerns with our vibration consultant Hal Amick and we offer you the following explanation: (Hal's detailed response plus attachments is also included below my reply) The short answer to your question is that it is possible to impose a very stringent vibration environment on the design of a building and have it meet that criterion at the time it is turned over to the owner (I will qualify this answer below). For a "reasonably" quiet site, such as a quiet part of a campus, this can be done. It is not easy to expect an advanced technology building to perform as well as an ultra-quiet site. If the site was ultra-quiet to start with, it is exceedingly difficult to create an environment that is equivalent or quieter than the pre-construction ambient. Unless that site is deep underground or in rock, the "cultural" vibrations (people, infrastructure, parking lots, mechanical systems and other new site influences that were not present prior to the construction) make this virtually impossible. Taking additional measures such as providing large concrete masses supported on air springs with either active or passive controls will provide with additional protection from potential noise sources. Hal's memo attached elaborates on this topic and provides you with criterion levels you are asking for. If I may elaborate, the questions you are asking are similarly applicable not only to vibration issues but also to other criteria such as temperature and humidity control, EMI, cleanliness, etc. INTEGRATION As you probably know, a building supporting an ultra quiet research and development program has an inherent mix of conflicting requirements. Instruments requiring low vibration environments frequently require a high level of accuracy on temperature and humidity controls not to mention particulate or biological contamination. Similarly, instruments often require a very stable "clean" power supply while being vulnerable to EMI. As you can see, the solutions for each of these requirements may have a negative effect on the criteria established by the other. No one could provide a 100% solution for all these criteria in the same space. A good solution will strike a good balance and provides a higher percentage solution for the criteria that matters the most. This issue is pretty important and we have a major session that addresses these "integration" topics in our upcoming "Buildings for Advanced Technology Workshop" at NIST October 15-17. Clayton Teague (NIST), James Murday (NNI and NRL), Hal Amick (CG&A) and I organized this workshop to provide a forum to answer some of these questions. The goal of the workshop is to provide a forum to share prior experiences and design solutions with others that are currently tackling issues we have lessons learned on. We tried to assemble a wide variety of experience in many aspects of buildings. These include laboratory end users, scientists, architects, engineers, builders, equipment manufacturers and organization specializing in such environmental control fields. If you cannot attend the workshop, the presentations for all the sessions will be posted on the www.nanobuildings.com web site soon after the workshop is completed. GUARANTEE Speaking as an architect, we've been brainwashed not to ever use the word guarantee on any aspects of our business. I would watch the fine print on the people that offer you a broad guarantee, as there are many ways to weasel out of meeting a stringent criterion. We've seen this done by others in the past by performing measurements in an environment that is not equivalent to the real anticipated operating environment of the building. Another scenario we've seen is when the A/E/Contractor team rolls over and plays dead saying that they just can't do it. This is the worst situation because by the time this phenomenon happens, the building design would have been executed in a fashion that would prevent you from receiving the good solution you could have had to start with. We prefer to engage our technical A/E staff with the technical staff from the owner's side to review and evaluate the many design solutions available for a particular problem and obtain a buy-in for a solution. While using tools such as finite element analysis, computer simulation and modeling and full-scale mock-ups, we (as a team) agree on a solution that will work. At the risk of sounding like a sales pitch, we could probably answer your questions better by sharing the individual project solutions we've encountered in similar buildings during the past ten years. You may know that our team designed the NIST Advanced Measurement Laboratory in Gaithersburg, MD and in Boulder Colorado (The Boulder project did not get built). We are also currently in the various stages of the design or construction of many other similar projects such as the Purdue University's Birck Ultra Performance Nanotechnology Center; Sandia's Center for Integrated Nanotechnology, Canada's National Institute for Nanotechnology at the University of Alberta; The UK National Physical Laboratory's new Advanced Metrology Laboratory and many other semi-conductor fabs, some supporting a 300mm line. If there is a particular concern you would like to be addressed, I would attempt to provide you with an answer from one of our team members. Thank you for the opportunity to share our thoughts with you. I hope that Hal and I answered your question, if you have any further questions, feel free to call or email either one of us. My contact information is below, however my cell phone number is the best way to find me (703) 447-1880. Ahmad Soueid | HDR Architecture, Inc. | 1101 King Street, Suite 400, Alexandria, VA 22314 | (703) 518-8552 office | (703) 518-8588 fax Hal Amick's response: **************************************************************** It is quite common to design vibration-sensitive R&D buildings with a specific vibration criterion in mind. The most common criteria use the format of the family of criteria used in the semiconductor industry, which are denoted VC-A through VC-E. The "official" version is given in IEST Recommended Practice RP-12, but there are presentations and discussion of the criteria in many other locations [e.g., Gordon (1999)]. Amick (1997) gives a brief history of the development of these criteria, as well as a discussion of the terminology and measurement technology associated with the criteria. The criteria are to be applied to vibrations measured in one-third octave bands of frequency, and presented in rms velocity amplitude. The general preference by the vibration control community is to use spectra, rather than time-domain representation of amplitude. The proportional bandwidth represents a compromise between "too much" data (i.e., an FFT spectrum) and "too little" data (a single time-history amplitude or a single weighted rms value). These issues are discussed in Amick (1997) and Amick & Bui (1991); the latter also examines the use of power spectral density (PSD) representation. The two most stringent VC criteria are VC-D and VC-E, with rms amplitudes of 250 and 125 microinches/sec, respectively, or approximately 6 and 3 micrometers/sec. These are routinely used in the semiconductor industry, where well over one hundred facilities have been designed and built to VC-D since that criterion began being used in the late 1980's. VC-E has been used since the mid 1990's, and because of the additional cost of meeting it, is less common. A few dozen designs have been carried out to this criterion. "Meeting" a criterion is generally taken by the vibration control community as being within 1 or two decibels of the design criterion. (Sometimes other interpretations are applied.) In the semiconductor industry, there is a general practice involving statistical combination of a spatially significant number of measurements. Successful implementation of one of these criteria requires coordination of design and construction efforts regarding structure, foundation, and mechanical systems. When the vibration consultant has been involved throughout the design and construction, the criterion is usually met. In some cases, where the design efforts were not followed through into later design and/or construction, the design criterion has been exceeded. Both VC-D and VC-E criteria require a "quiet" site. These can be in a campus setting, but care must be taken to avoid high-vibration settings, such as traffic, etc. A site vibration survey is generally carried out. They can be met using either a slab-on-grade floor or a suspended floor. The former is much less expensive to achieve, but many applications in the semiconductor industry require a "subfab" area, one or two levels below the cleanroom. Some of the structural approaches for different criteria are discussed in Amick et al (2002). A PDF copy is attached. It is often a problem to expect an advanced technology building to perform as well as an ultra-quiet site. Unless that site is deep underground or in rock, it will be adversely affected by what we call "cultural" vibrations, those associated with the presence of people and infrastructure. For example, there are usually parking lots for the users. Vibrations are generated by activities around the site (usually not under the control of the designer) and by systems within the building (initially under the control of the designer, but not after transfer of ownership). The issue of vibrations generated by systems within the building is part of a process we call maturation. This is illustrated in an attached figure labeled "Study of Fab Maturation", file "maturation.jpg". The red curve represents vibration at the "as-built" stage, when ownership transferred to the owner. The base building mechanical systems were operating. The other two curves were taken later, after the owner had itself installed a great deal of process support equipment, but without involving the vibration consultant. [This figure was developed to show the owner that it was degrading its own environment.] The vertical axes of these graphs are in velocity level, in decibels re 1 microinch/sec. The level is 20*log10(amplitude) if amplitude is in microinches/sec. In this scale, VC-D is 48 and VC-E is 42. A number of nanotechnology facilities are currently on the boards (CG&A is involved with eight at this time). Only one, Duffield Hall at Cornell, is under construction. No performance data are available yet. All involve some space at VC-D or VC-E. The NIST Advanced Measurement Laboratory (AML) is under construction in Gaithersburg, MD. When completed, it will be the most sophisticated large laboratory in the U.S. Its basic "on-grade" criterion is denoted NIST-A (discussed in Amick, et al (2002)), which is the same as VC-E at frequencies above 20 Hz and 1 microinch/sec below 20 Hz (more stringent). We are quite confident this criterion will be met at startup. If one wants to take a very good site and make it better, one must resort to special measures. NIST will employ pneumatically-isolated floors in its most critical spaces. A prototype was built and studied, and is described in Amick et al (1998). It is CRITICAL that one define the time between site selection and completion at which one measures the vibrations. A memorandum (to be the basis of a paper presented at IEST conference) shows the results of vibration surveys at four different times at a given location. [Memo re Measurement Times.pdf] REFERENCES All of the following documents, and more, are available at the "Papers" link at www.colingordon.com . [Amick et al. (2002) has not yet been posted] Gordon, C. G. (1999). "Generic Vibration Criteria for Vibration-Sensitive Equipment," Presented at International Society for Optical Engineering (SPIE) Conference on Current Developments in Vibration Control for Optomechanical Systems, Denver, CO (July 1999). Amick, Hal (1997). "On Generic Vibration Criteria for Advanced Technology Facilities: with a Tutorial on Vibration Data Representation," Journal of the Institute of Environmental Sciences, pp. 35-44, (Sept/Oct, 1997). Amick, Hal and Sean K. Bui (1991). "A Review of Several Methods for Processing Vibration Data" (with S. Bui), Proceedings of International Society for Optical Engineering (SPIE), Vol. 1619 (November, 1991) Amick, Hal, Bea Sennewald, Norman C. Pardue, Clayton Teague, and Brian Scace (1998). "Analytical/Experimental Study of Vibration of a Room-Sized Airspring-Supported Slab" (with), Noise Control Engineering Journal, v. 46, no. 2, pp. 39-47 (March/April, 1998). Amick, Hal, Michael Gendreau and Colin G. Gordon (2002). "Facility Vibration Issues for Nanotechnology Research," Presented at Symposium on Nano Device Technology 2002, May 2-3, 2002, National Chiao-Tung University, Hsinchu, Taiwan <> <> <> **************************************************************** -----Original Message----- From: packard@socrates.Berkeley.EDU [mailto:packard@socrates.Berkeley.EDU] Sent: Tuesday, August 27, 2002 7:56 PM To: Soueid, Ahmad Cc: tkalil@uclink.berkeley.edu Subject: low vibration building Dear Mr. Soueid, Since you are listed as a contact person for the planned meeting on nano-buildings planned at NIST on October 15-17, I wonder if you can answer the following question which is relevant to any nano-building design. Has any engineering/architecture firm been able to guarantee a vibration spectrum specification for a quiet R and D building? If yes please tell me what that spec. was (or is) if you know. The goal in any nano research building is to have vibration/sound levels no greater than the ambient levels on the ground with the building absent. I have not yet seen any building that even came close to that spec. Thank you in advance for your answer. Richard Packard Professor of Physics HA-MG-CG Taiwan Nano 2002.pdf Maturation.jpg Memo re Measurement Times.pdf