July 21, 2011
In a perfect world, the steel used to construct marine accommodations for the shipping and offshore oil and gas industries would be completely straight. In reality, this is never the case. Yet it is critical for fabrication tolerances to be within specifications in order to maintain structural integrity. Fortunately, technology is making it easier to identify and correct any deviations before they become too costly.
When QCI Marine Offshore, Pascagoula, Miss., was charged with building a five-story accommodation module and helipad in a Gulf Coast shipyard, the company wanted to ensure high quality at every stage. The vessel is designed to serve as living quarters for oil workers and features amenities such as kitchen and dining facilities, locker rooms, a hospital and a library. To obtain as-built surveys of the structure, QCI subcontracted Dale Stockstill & Associates (DS&A), a leading provider of surveying, mapping and laser scanning services on the Gulf Coast.
Determining whether the vessel was constructed according to specified 3D design tolerances was challenging due to the flexural forces of the steel, the malleability of the steel under welding and the difficulty of access. Also, crews were installing walls, air-conditioning ducts, piping and cable trays while the as-built surveying was to occur, and the sheer size of the structure made surveying difficult. DS&A was confident that traditional surveying combined with laser scanning would provide the required data, but QCI was uncertain. Although the company was comfortable with 2D drawings and models, point clouds were unfamiliar territory.
The team agreed that the project would move forward in two phases. Close collaboration and good communication between DS&A and QCI would be essential to ensure the success of the project.
After exchanging several e-mails with the client, DS&A’s survey crew headed out for the first phase of the project: scanning the recently completed helipad. The crew chief of DS&A’s scanning crew carefully chose and tied in his scan positions and then performed the necessary scans using DS&A’s GLS-1000 high-definition laser scanner from Topcon Positioning Systems. After the scans were registered, DS&A’s technicians provided 2D control drawings to QCI.
The firm also introduced the client to color mapping. At DS&A’s request, QCI provided a 3D model of the helipad. This model was inserted into DS&A’s PolyWorks software along with the registered point cloud. A best fit between the cloud and model was obtained, and the resulting deviations were depicted in a color map, which provided a visualization of design deviations. QCI was satisfied with the survey and deliverables and gave the go-ahead for the accommodation module survey.
Scanning a helipad was one thing. A precise 3D survey of a five-story steel box under 24-hour construction in a busy shipyard was quite another. The difficulty of the survey was further compounded by extremely high-tolerance specifications for some of the required items that ranged from 3 to 15 mm. Several meetings were held in which DS&A’s survey crew chief and technicians along with QCI Marine’s engineers and managers went over survey procedures, work schedules, and--most importantly--realistic, achievable survey precision.
After several ideas were proposed and rejected, the survey crew chief suggested using the laser plummet on his Topcon GPT-3100W total station to establish four control points that would start on the top deck and extend downward through all of the decks to the pavement beneath the structure. Everyone agreed that this approach would provide a good solution.
With the help of QCI Marine employees, the four points were positioned to avoid hitting critical components on the way down. Once finalized, the crew chief set up either the total station or the scanner on the top deck, and a hole was drilled in the deck, allowing the laser plummet beam to pass through to the deck below. This procedure was repeated, with a crew member setting a PK nail at each position, until the beam hit the pavement under the structure. A closed traverse loop was run around these four points and was later adjusted and balanced out using standard survey procedures.
The crew then used its Topcon DL-502 digital level to establish precise vertical control for each control point on all decks. Starting at a previously established benchmark, the crew alternated between standard and inverted level rod shots. The standard rod was used to obtain the height of instrument (HI), and the inverted rod was used on the overhead deck. After obtaining the precise elevation of the overhead deck, the survey crew chief would add 6 mm (the thickness of the deck), move to the next level and repeat the procedure.
This procedure was repeated for all four control points until the crew reached the top deck. Setting up on each deck level required the crew to communicate by radio. The crew member on the ground below guided the instrument operator as he positioned the total station, making sure that the laser plummet was precisely over the PK nail on the ground. That way, regardless of which deck the crew was working on, the control point was the same except for the HI, which was obtained by adding the deck heights to the normal HI measurement.
Once the control was established, precise measurements were taken on specific beams, frames and bulkheads using DS&A’s total station, the GLS-1000 laser scanner and even a certified tape measure. The client needed to determine whether the structure was built to specifications. Were the total outside dimensions within tolerance? Was the structure square? Did it have any “twist” to it on the way up? The best way to answer most of these questions was through a high-definition laser scan.
Starting at the closed control loop under the structure, the crew set “flypoints” for scan positions out in the shipyard. The structure was scanned on all four sides and on the top, and the point clouds were registered in DS&A’s ScanMaster software. To determine the global control measurements, the point cloud for the structure had to be reduced to a box that could be precisely measured on each side for length, width and squareness. A ScanMaster routine provided a “best-fit plane” along each side of the cloud. These planes were intersected to establish the corners of the box (see the point cloud images on page 33).
Besides the dimensional control measurements described above, the point cloud contained much more information that was extracted and used in the deliverables. Deck heights were confirmed and checked from welds and plates visible on the scans. The ends of beams located from tape placed on the bulkhead prior to the scan were checked against those located inside using the total station and a tape measure.
The last deliverable provided to the client was color maps of the bulkhead walls, which showed the amount of “hogging and sagging” in the bulkheads. Once again the client provided DS&A with its 3D model of the structure, and DS&A broke the model into the individual components that were to be color mapped. The model of a given section and the cloud were placed together in PolyWorks, where a color scale was chosen and comparisons were made. Along with PDFs of the color mapping, the client was also provided with the cloud and model data on an FTP site to view in a PolyWorks viewer.
The map that DS&A presented to QCI revealed a color-coded picture of the structure represented by different colors for each specified measurement range. According to QCI, corrections were made to the structure where structural plating was indicated by the model to have “deformations” between stiffeners in excess of the project specification. The color coding allowed QCI to define the deviations to a structural engineering firm to analyze the as-built structure in a global structural analysis in order to certify the integrity of the accommodation module.
QCI, a general contractor that does not typically perform the dimensional control but rather subcontracts the work charged in a lump sum, was not able to determine exactly how much laser scanning saved on this project. But everyone involved agreed that it would have been much more difficult--and perhaps even impossible--to provide the results using conventional methods alone.
Faster ScanningAfter this scan of the five-story accommodation module, an even faster scan technology was developed. In March 2010, Topcon Positioning Systems unveiled the GLS-1500, which speeds up point cloud collection at a rate of 30,000 points per second and a range of 150 meters. According to the manufacturer, the unit also has an “all-in-one” design that reduces the amount of equipment needed in the field.
The unit has a built-in 2.0 megapixel digital camera, so when it is connected to a PC and used with Topcon’s ScanMaster software, a live video feed of the job site can be streamed to aid in scan setup and data acquisition. Also, it has an onboard data collector with a keypad and LCD display that allows use as a stand-alone laser scanner. Data collected can be stored onboard on an SD memory card or logged into a PC. A built-in wireless LAN connection allows control of the scanner on a PC from the inside of a vehicle.
Topcon Precise Scan Technology is designed to allow high-accuracy measurements over a wide range of distances. Lens array optics technology maintains distance accuracy from 1–150 meters and additional ranging past 330 meters is available. The unit’s Class 1 laser classification allows scanning near airports, busy traffic areas and populated areas, as well as low power consumption and fewer battery changes.