At the New York State Capitol, non-destructive evaluation techniques were used to deterine existing structural conditions and to locate hidden structural elements. Most construction documents had been lost in an early 20th-century fire; data collected through NDE served to re-create the structural drawings. Photo: Marie Ennis, P.E.

THE NEW YORK STATE CAPITOL
Albany, New York

   The New York State Capitol, constructed between 1867 and 1899, is an opulent masterpiece of the late-nineteenth century, and one of the last American monumental load-bearing structures. Begun by Thomas Fuller in 1867, construction continued until 1876 when appropriations ran out. The next year, construction resumed under the direction of a design advisory board composed of Henry Hobson Richardson, Frederick Law Olmstead, Leopold Eidlitz, and Isaac Perry, the first New York State Architect. Extraordinary care was taken in every aspect of its design, and more than $25 million was spent on construction. In 1979, the structure was designated a National Historic Landmark by the U.S. Department of the Interior.

Problem

   The 400 by 300 foot granite structure is 5 stories tall, with a full basement and attic. Although basement foundation walls up to 16-feet thick support the 200,000 ton building load, the Capitol has suffered significant structural problems since construction. The building load is spread unequally over the foundation, and the site contains layers of local clay mixed with sand and water, which is susceptible to liquefaction. When these conditions were discovered during the initial excavation, the idea of using wood pilings was abandoned and a 3-foot thick unreinforced concrete slab was placed over the excavation to support stepped stone footings for walls and piers.

    By 1990, the Commission for the Restoration of the Capitol determined that a comprehensive structural documentation of the building was essential to its long-term preservation. Original structural drawings did not exist, and most construction records were lost in a 1911 fire.

    This new study would determine at what locations the subsurface clay was being overloaded; help explain and document the sources of previous failures in the building; identify those areas where code-required alterations, such as new fire stairs, would be best located; and enable the Commission to develop appropriate responses to ongoing requests for additional mezzanines and storage areas.

Evaluation of the Capitol
Using NDE Techniques

   Robert Silman Associates, P.C., was retained in 1991 to produce a structural analysis of the northwest quadrant of the Capitol, and a feasibility study that would explore the potential of the building to absorb additional live loads. The project would involve preparation of structural drawings documenting the size, location and type of structural components. An unusual aspect of the work was a research component to correlate results of NDE techniques with conventional destructive probe methods. If NDE techniques could be successfully applied, damage to the building and disruption to its occupants could be minimized in analysis of the other parts of the Capitol.

   The U.S. Army Construction Engineering Research Laboratory assisted in the research by providing staff and equipment through its Construction Productivity Advancement Research Program. By employing NDE in the same areas that would subsequently be opened with destructive probes, the research team intended to show the practical applications of NDE techniques, and demonstrate the reduced physical damage to historic buildings. NDE would also mean substantial cost savings for the work.

    The NDE techniques employed included radar, impact echo, ultrasonic pulse velocity and spectral analysis of surface waves, electromagnetic detection, infrared thermography, and fiber optics. Using these techniques, the work was intended to supplement archival research and a visual survey to locate and assess the conditions of the structural components (beams, girders, columns, and bearing walls). Finally, traditional destructive procedures would take place, as necessary, to confirm findings and to provide a comparative basis for determining how and when NDE techniques could be used in lieu of destructive procedures, and what degree of accuracy was achievable for the techniques tested.

Solution (Test Results)

   NDE at the New York State Capitol was undertaken through a cooperative agreement with the U.S. Army Construction Engineering Research Laboratory, under the supervision of Robert Silman Associates. The Laboratory relied on outside consultants to operate and interpret equipment. Approximately ten days of site work were required to survey one quadrant of the building from the basement through the roof.

The floor structure is scanned using impulse radar. An operator pushes the transducer across the floor at a constant speed. Photo: Marie Ennis, P.E.

    The NDE techniques utilized at the New York State Capitol were experimental in nature. Because the U.S. Army Construction Engineering Research Laboratory participated in the project as a means to develop the procedures for its own purposes, six techniques were used and evaluated.

   Impulse Radar. For the New York State Capitol project, the radar tests successfully located embedded beams, girders, columns, and metal ties/anchors.

   The readings enabled the investigators to profile brick arches spanning roughly 5 feet between beams, and to locate H-shaped steel columns in masonry walls showing the orientation of the flanges.

This is the raw output from a radar scan. Beams, brick arches, and concrete slab were noted. Photo: U.S. Army Construction Engineering Research Lab.

    The tests readily measured the thickness of masonry walls that were well bonded. They were also able to determine the depth-of-cover over beams and girders and brick arches, and the location of voids, pipes and flues in walls up to 2 feet 8 inches thick.

   Results were less conclusive for the thick stone footings, where the high water table, large quantities of metal near the surface, and the existence of conductive surfaces (false floors with air spaces below) impacted the ability of the equipment to locate the footings. More than any other technique, the radar penetrated the masonry effectively, and allowed large portions of the Capitol quadrangle to be scanned continuously and quickly.

   Another very promising aspect of the research done at the Capitol was the application of the backward propagation imaging method to the acquired data in order to provide a computerized visual image of the hidden structure. Further developments of this system would allow architects and engineers to measure hidden features directly from a plotted image, making the technique more "user friendly."

Above: The impact echo equipment includes a hammer, receiver, and water soluble coupling gel and was used to locate a steel column concealed behind the granite facing. Photo: Dennis Sack. Below: Data is shown on the field computer screen. Photo: Marie Ennis, P.E.

   Impact Echo. Impact echo produced good data for reading wall thicknesses and the integrity of granite and sandstone columns, veneer walls, and brick walls less than two feet thick. While steel columns behind granite facing were located, the orientation of flanges was not discernible. The impact echo test provided excellent results in detecting stone cracking parallel to the surface. Its optimal use was when working with homogenous stone materials; with multiple layers of material (example: terrazzo over brick) it did not produce good results. Brick arches and beams were not detected, and the presence of energy absorbing materials like plaster proved an impediment.

   Ultrasonic Pulse Velocity. Ultrasonic pulse velocity proved to be an excellent technique for determining the thicknesses and relative soundness of solid granite and sandstone columns. In composite walls, the velocity and signal strength dropped across mortar joints. Low frequency pulse velocity signals can provide good results for thick brick walls where multiple mortar joints would otherwise block the transmission of higher frequency signals.

Ultrasonic pulse velocity is being used to obtain typical wave velocities for granite. Note transmitter on one side of the column and receiver on opposite. Photo: Andrea Lazarski.

  A disadvantage of the low frequency energy is that the long wavelengths result in an "averaging" effect, which decreases the resolution of the method and does not allow for the location of smaller scale damage, such as brick unit de-bonding. In general, the results correlated well with the impact echo results.

    The two techniques should be used in conjunction; ultrasonic pulse velocity provides characteristic wave velocity measurement for the specific test material and impact echo assesses voids and thickness with access to only one side of the surface.

   Spectral Analysis of Surface Waves. Spectral analysis of surface waves required access to only one side of the material being measured. It worked well on brick and solid stone, and on the composite slab on grade at the cellar floor, but was less successful when used on thinner walls and framed slabs. Due to the sensitivity of the equipment, the technique was not suitable for measuring thicknesses of floors and walls. Additionally, the presence of plaster and other energy-absorbing materials proved an impediment. This process was more time consuming than the others.

This electromagnetic detection instrument is designed to locate rebars in concrete for quality control inspections. The diameter of the bar and depth-of-cover is calculated and digitally displayed. Photo: Marie Ennis, P.E.

   Electromagnetic Detection. This imaging technique worked well at locating iron beams and girders supporting brick arches that had as much as 9 inches of cover, provided no wire mesh existed. It was also successful at locating steel columns behind up to 8 inches of granite, and locating iron anchors in stone walls. Where wire mesh exists, such as in flat concrete slabs, the results were inconsistent.

   Infrared Thermography. This imaging technique was useful in locating hidden pipes and flues within masonry walls. The images of deteriorated brick masonry (water and salt damage) and patched areas of exterior granite walls clearly showed as anomalies. A dense, undamaged portion of masonry will retain less moisture than a porous area. The moisture content will affect the surface temperature of the material, thus identifying it as being different from surrounding areas, even when the area appears to be the same as the undeteriorated area to the naked eye. Although, theoretically, one should be able to image voids behind the surface of a stone face, this was not achieved at the Capitol. The reason for this failure is probably related to the sensitivity of the equipment, the wavelength bands being recorded, and high ambient humidity levels during testing. This technique did not prove useful for imaging roof trusses through clay tile roofing.

   Fiber Optics. Fiber Optics. Since the Capitol is a solid masonry building with few voids, the utility of fiber optics in the project was limited. However, it did allow assessment of the interiors of wall chases and flues, including the condition of their mortar joints. It was also used to access the space above decorative hung ceilings; however, the focal length was not sufficient to provide overall views of beams and girders.

Project Summary

   Of the NDE techniques employed on this heavy masonry building, radar proved to be the most successful for imaging hidden structure and conditions. Next, in order of success, were impact echo, ultrasonic pulse velocity, spectral analysis of surface waves, and infrared thermography. Electromagnetic detection was very useful, but its scope is limited to buildings that contain some iron or steel and to locations where framing members are isolated from pipes, conduits, and other metal features. Fiber optics was found to be of minimal use in this type of building because of the limited areas where voids were present. One problem to consider is the need to have highly trained equipment operators present, as well as sophisticated computer programs that can translate the raw data into meaningful results. In some cases, despite inherent non-destructive testing costs, the number or conventional probes that can be eliminated will offset these expenses.

Princeton University (Whig Hall)
Case Study

 

Site Map

Home |  Search |  E-mail |  Technical Preservation Services |  NPS History & Culture