 |
 |
Members | Affiliated Professionals | Consumers | Press |
|
|
        |
|
|
|
||||||||||||||||||||||||||||||||||||||||||||||
|
|
|
|||||||||||||||||||||||||||||||||||||||||||||
 |
|||||||||||||||||||||||||||||||||||||||||||||||
|
Technical Resources Unitized or Stick? Exploring which type of curtainwall system is right for a project By C. Michael Minkoff Is a unitized or stick curtainwall better for this project? For decades, with advances in the development of glazed curtainwall systems, many owners, architects, consultants, general contractors, glazing subcontractors, and glazing system manufacturers have addressed this same question. There has been considerable disagreement as to the answer and a variety of reasons exists for these differences in opinion. The purpose of this article is to propose objective criteria that may be used in determining which approach is best for a given project. This criteria is in the form of several questions that should be asked concerning each specific project. The answers should help with the basic question: Unitized or stick? Before we get to the questions, here is a brief description of each of the systems under discussion. A "stick" curtainwall system (Fig. 1) is one in which the primary structural framing components are erected individually in the field like an erector set, typically with vertical mullions attached first to the floor slabs and the horizontals attached to the vertical mullions. Then, the vision glass and spandrel materials (typically glass, aluminum panel, or stone) are field installed into the assembled gridwork. All glazing gaskets and joinery/perimeter sealants are field installed. Then, if the finished interior trim is not integral with the primary structural framing components, this trim is installed. One of the earliest examples of this type of curtainwall is the Westin Peachtree Plaza Hotel in Atlanta, GA. A "unitized" curtainwall system (Fig. 2) is one in which the framing members are pre-assembled in a shop facility into units which are typically one lite wide and one story tall. Then, the vision glass and spandrel materials are installed into the units in the shop. All glazing gaskets and joinery sealants are shop installed; however, perimeter sealants would still be field installed. These assembled and pre-glazed units are then shipped to the jobsite for field erection. Vertical and horizontal mating joints can either be dry-sealed with gaskets or wet-sealed with field applied sealants. Mating joints can either be a male/female configuration as used in the Georgia Power Headquarters Building in Atlanta, GA, or a female/female configuration as used in the Los Angeles Convention Center Expansion in Los Angeles, CA. These descriptions are admittedly an oversimplification and don't include any number of other approaches such as "pilaster/stick" (Fig. 3) as used in the 181 West Madison Office Building in Chicago, IL, or other combinations or variations. They also don't address features that may be borrowed from one approach and added to another. For example, many stick systems have added a split horizontal or stack joint (Fig. 4) typically utilized in unitized systems to accommodate excessive live-load deflections, and many unitized systems utilize field applied exterior trim where this facilitates shipment or erection. The combinations and permutations are almost endless. However, for this discussion I will limit comments to more traditional stick and unitized designs. Is there a pre-existing bias towards one approach or the other? Before addressing any other design considerations, we must first ask if there is a pre-existing bias towards a stick or unitized approach by the owner, architect, consultant, subcontractor, or manufacturer. For example, the owner may say that he would prefer one system but that he can't afford it because it is always too expensive. Or the architect may have had a bad experience on a project and may attribute all or most of the problems to the basic system design. Similarly, a consultant may have a preference for one approach over the other because of performance or quality control issues. A subcontractor may have existing relationships with certain manufacturers that may not have experience in one or the other approach. A manufacturer may have a "one size fits all" mentality and insist that its approach will work for any application. However, if an owner assumes that one approach is too expensive, then this can only be confirmed by actually estimating costs for both approaches. The architect's bad experience may not even be related to the type of system utilized and careful analysis of the problem project may uncover other factors which contributed to (if not solely caused) the problems. For a consultant to assume that one approach will always outperform another or that the quality of one approach will always exceed the other is an overgeneralization and may ignore many other contributing factors. A subcontractor's existing relationships may actually limit its flexibility in addressing each project on its own merits and choosing the best-suited system for a given project. And a manufacturer with a one-size-fits-all mentality will soon find that many of the problems and failures in our industry have resulted from misapplication of systems. A square peg still doesn't fit well in the round hole. Sometimes, these pre-existing biases can be eliminated through education, testing, or more thorough analysis of a proposed approach. Sometimes, after all the investigations, the biases may be valid for that project. In these cases, a different system or manufacturer may be appropriate. However, we must carefully analyze our presuppositions if we are to understand why these biases exist and whether or not they are valid. Are there performance criteria which would promote or dictate one system over the other? Performance criteria such as seismic loading, windloading, air and water infiltration, and live load deflection are major considerations when choosing a system approach. The lateral movement imposed by earthquakes, or seismic loading, can cause unusual racking or component distortion in the curtainwall system. Unitized systems handle these movements better when the glass is silicone glazed, and either system, when properly designed, can usually be made to work well for conventionally glazed applications. Windloading is transmitted from the glass or infill material to the horizontal and vertical mullions. In stick systems, the vertical mullions can span two floors (twin-span) and provide greater structural efficiency than single-span mullions typically used in unitized systems. However, if the mullion expansion joint is near stool height (approximately 20 to 30 inches above the floor line) then this is not necessarily the case. This would allow for "zero movement" or continuous-span design, which is even more efficient than twin-span design, and can be utilized with either a stick or unitized approach. However, even with continuous-span design, the stick approach is more stable because it eliminates a domino effect if the curtainwall is struck or impacted due to the fact that each mullion in a unitized system typically has a single anchor point. Although this may sound improbable, it's not impossible. In my 25 years with Bruce Wall Systems, we have on more than one occasion replaced mullions that were destroyed due to impact from automobiles or other disasters. Air and water infiltration is always one of the owner's primary concerns. As a result, it is common for an architect or consultant to specify two lines of defense for water infiltration and, though less common, two lines of defense for air infiltration. Two lines of defense for water infiltration can be adequately handled with internal guttering with either system approach. However, two lines of defense for air infiltration typically requires the addition of sealed backpans in the spandrel area which is more easily accommodated in unitized systems. Another performance consideration is vertical interstory differential movement due to live loading, column shortening, and long-term creep. Often, these movements can exceed 1/2 inch, particularly with steel or cantilevered structures. Unitized systems or stick systems that have been modified to include a split horizontal or stack joint (Fig. 4) handle these larger movements best. This is particularly true when one considers that these movements must be added to thermal expansion and contraction requirements and fabrication tolerances as well as field erection tolerances to calculate the total movement that each expansion joint must accommodate. Are there aesthetic criteria which would promote or dictate one system over the other? As previously discussed, the use of structural silicone glazing may promote a unitized approach, particularly in high seismic areas. I recommend that four-side structural sealant applications always be done in a controlled shop environment because of the greater risk of disadhesion in field glazed applications, not to mention the potential liability involved. (An example of this is the Georgia Power Headquarters Building previously mentioned, one of the first four-side silicone projects in the United States) However, two-side silicone applications can successfully be accomplished with either system approach as evidenced in the 505 Waterford Park project in Plymouth, MN, a two-side silicone stick curtainwall system. If the architect's design includes continuous reveals at the vertical to horizontal intersections, then this can easily be accommodated with a unitized system, as in the Los Angeles Convention Center Expansion, but only with great difficulty in a stick system. If the design includes large vertical mullions with significant projections outside the plane of glass, such as on the Barnett Plaza Office Building in Jacksonville, FL, then this is more easily accomplished with a stick system. If the architect's design is very complicated without much repetition and particularly if there are a multitude of vertical and horizontal plane changes, then a unitized approach would most likely be more difficult and costly. If there are a large number of penetrations through the curtainwall such as steel or concrete structures extending to exterior columns or canopies, as in the Mobile Government Plaza in Mobile, AL, then this is more easily accomplished with a stick system which can be installed around the penetrations. Another design consideration is the geometry of the building. As mentioned previously, a multitude of corners, setbacks, and balconies make a unitized approach less practical. On the Cleveland Public Library in Cleveland, OH, the plan was elliptical and the mullions were staggered from floor to floor such that the mullions from one floor to the next didn't line up. Because the glass was segmented along the ellipse, this meant that the adjacent glass lites on either side of a common horizontal were in two different planes. This was accomplished using a concealed flexible silicone gasket in a stick curtainwall system and would have been virtually impossible in a unitized system short of using curved glass and stretch-formed horizontal framing. Are there site logistic criteria which would promote or dictate one system over the other? Site logistic issues would include site accessibility, availability of crane or hoist, and perimeter accessibility and availability of storage space on each floor. Accessibility both to the site and on the perimeter of the floors may preclude being able to use a unitized approach or make it less practical due to unit size limitations. If storage on the floors is limited and it is not practical or possible to hoist units from the truck directly onto the building, then a unitized approach may not be viable. The sequencing of other trades such as precast or masonry may also promote one approach over the other depending on the design of the building. The physical location of a project can have a major impact on the selection of a system type. Certain cities have a reputation for having more productive field labor than others. It is no surprise that in cities with less productive or more expensive field labor the majority of projects are unitized. Some manufacturers have developed expensive and elaborate anchorage systems to minimize field labor time by allowing units to be rough set quickly and adjusted later. Field labor savings and speed of erection are two of the biggest advantages of most unitized systems. The type of structure being used and the method of anchorage to the building that this type of structure will allow can also be factors. Most unitized systems use top of slab type anchors which are set in pockets in the slab and grouted over or set on top of the slab if there are raised floors or interior finishes that will cover up the anchor. Most stick systems anchor to the face of the slab by means of embedded plates or inserts. Some thin slab construction will not accommodate anchor pockets, which may preclude the use of a unitized system or increase its cost. Erection tolerances are critical in the anchor design and can make one approach more practical or cost-effective than the other. Unitized systems typically require a particular sequence of construction. The units are typically erected sequentially laterally and floor by floor. It is with some difficulty that interruptions such as hoist bays or leave-out bays are accommodated with typical unitized construction. The exception to this is female/female type unitized construction which was utilized on the Los Angeles Convention Center Expansion. This allows the insertion of units into the curtainwall at virtually any location, providing similar flexibility to a stick system. Are there schedule considerations which would promote or dictate one system over the other? System development time is typically greater with a unitized system but the amount of time required for erection is usually less. The reason for this is that the fabricated materials and assembled units must be stockpiled before beginning the erection process. Due to the sequential nature of the erection process discussed previously, each of the typical units required to complete a given floor must all be ready at about the same time in order to prevent delays in the erection process. However, in order to be more cost-effective, identical parts would typically be fabricated together and be stockpiled for use by the assembly crews as needed. If one part is missing, then the unit can't be completed and the whole erection process can be affected. By contrast, stick systems are usually quicker to the site because the sequencing and stockpiling requirements are not as critical. If a part is missing, one simply moves on and then comes back when that part is available. However, these systems typically take more time in the field to complete the installation. If there are long lead-time items, such as curtainwall supported granite or limestone, a stick approach allows some work to be done on site while waiting for delivery of the stone. If properly designed, a unitized system may also be modified to allow for field installation of stone or long lead-time items. System selection becomes a major factor relative to schedule when there is a limited window of time for the field erection process. This can be particularly important in renovation construction in occupied buildings where the building can't be left open at the end of each day. Are there intangible issues which would promote or dictate one system over the other? The availability of manpower, the availability of manufacturing space, the experience level of the subcontractor or manufacturer, and other intangible issues can also affect the selection of a system type. There may be a shortage of qualified field personnel which may mean that a unitized system, because of its greater dependence on shop labor, would be the best approach, or vice versa. My company has experienced shortages of manufacturing or assembly space in some parts of the world which could promote a stick approach. There may be other intangible issues such as the comfort level of key contributors to the curtainwall with a particular system type. Trying to get a manufacturer, subcontractor, or erector to use a particular approach when a company has limited experience with that approach may be unwise. Are there cost effectiveness issues which would promote or dictate one system over the other? One of an owner's primary concerns is always the cost. However, when this issue is posed, it is really two questions. The first is: What will my initial cost be for this project? The second is: What will my long-term costs be for this project? For a speculative office building, initial cost is usually the major consideration whereas corporate or institutional clients want to factor in mechanical costs, long-term maintenance and repair costs, and variations in the usable life of a particular system. Which system is initially cheaper may be irrelevant if the heating and cooling costs are substantially higher or if maintenance costs are significantly greater or if major repair or replacement of the system may be required early in the life of the building. As I have noted, there are numerous factors to consider when determining the most cost-effective approach when considering either initial costs or long-term costs. One system approach is not always going to be the best value for a given project. Each project must be evaluated carefully and thoroughly on its own merits. It may be concluded that the most cost-effective approach is not stick or unitized but a different approach altogether such as a pilaster/stick system or combination of systems. The tools for selecting a system type discussed here are merely guidelines to help narrow the selection process. They are intended to be neither exhaustive nor rigid in their application. Some of these issues have cost ramifications only and can be overlooked if there are other compelling reasons to do so. Hopefully, it is clear that one system approach is not always right for every project. Each project must be evaluated based on the specific criteria for that project. Then a more informed decision can be made in the selection of a stick, unitized, or some other curtainwall system approach. C. Michael Minkoff is vice president of Bruce Wall Systems Corp., Tucker, GA, which he joined in 1974. He has a Bachelor of Architecture degree from Georgia Institute of Technology, and is a member of CSI and the American Arbitration Association. The information in this article is adapted from a presentation he made at the recent GANA Building Envelope Contractors Conference in Las Vegas, NV All projects pictured are Bruce Wall Systems projects. The exploded isometric diagrams have been provided courtesy of Harmon, Ltd. Source: Appeared in Glass Guide for Architects, Glass Magazine June 2000. |
|
|
| Copyright © 2007 NGA - National Glass Association. All Rights Reserved. |
Site Design by WebFirst |