The cold, stark landscape of Antarctica is home to penguins, ice, high winds, and two antennae owned by the National Science Foundation (NSF), the independent federal agency that supports and funds all fields of fundamental science and engineering in the United States. The severe conditions of the coldest place on earth might deter those faint of heart. But this is a genuine account of man and instrument going where no others dare to tread.
The National Science Foundation acquired a decommissioned 7.2-meter diameter VertexRSI Ku-band antenna from the National Aeronautics and Space Administration (NASA). The NSF also owns and operates an 11-m diameter DAI C-band antenna. Both antennae are located on Black Island, Antarctica, and housed in separate geodesic frame radomes. The NSF wanted to recommission the 7.2-m antenna and operate it at its originally designed Ku-band frequency. This would provide a backup system and allow the 11-m C-band antenna to go offline for an upgrade to also operate at Ku-band.
The 11 m antenna was the primary satellite communications (SATCOM) link for McMurdo Station, which is situated on Hut Point Peninsula on Ross Island, Antarctica, the farthest south solid ground that is accessible by ship on McMurdo Sound—approximately 3,500 km due south of New Zealand. This terminal serves as the logistics hub for half of the Antarctic continent. It is operated by Raytheon Polar Services, a subsidiary of Raytheon, developer of aerospace and homeland security technologies. Black Island is located roughly 40 km across the sound from McMurdo Station, and access to this icy wilderness usually involves a helicopter or an ice traverse.
VertexRSI, a division of General Dynamics based in Kilgore, Texas, which offers earth station and base station communications products and installation services, was commissioned by the NSF to perform a thorough investigation of the antennae. VertexRSI assembled a team of specialists that would conduct an evaluation of the major mechanical, structural, and electrical components on both antennae. The field team would measure and assess the geometric condition of the antenna structure so upgrade recommendations could be made.
Veteran metrologists are keenly aware of conditions affecting measurement instrumentation, such as vibration, wind, and varying or extreme temperatures. The Black Island project would bear its own special brand of challenges, and photogrammetry was the tool of choice for this particular job. Photogrammetry is a portable measurement technology that uses photographs as the fundamental medium for metrology. The system’s ability to operate in formidable environments is undoubtedly one of its key strengths. Precision measurement is achieved by installing a set of fixed control points on the object being measured. The position (exterior orientation) of the camera is calculated using resection. This feature facilitates the operator’s ability to work from ladders or other higher perches without the need for tripods or other instrument stabilizing devices.
Triangulation is the underlying principle used by photogrammetry. Triangulation is also the way our human eyes work together to gauge distance; this is also known as depth perception. When a photograph is taken, a real-world 3-D environment is captured and compressed onto a 2-D plane. Within the 2-D image, it is possible to make very precise measurements of the length or width of an object to a degree of varying accuracy. What is not possible is the ability to measure depth in a single image. By capturing multiple images of the object from convergent angles, lines of sight to common points or targets on the object can be mathematically intersected. Thus triangulation is used to resolve 3-D coordinates from 2-D measurements.
VertexRSI utilized a digital photogrammetric system called V-STARS, which is manufactured by Geodetic Systems Inc. of Melbourne, Florida. The system uses a high-resolution intelligent digital camera, known as the INCA3. To overcome problems of greatly reduced battery performance experienced at low temperatures, the camera was coupled to an online kit consisting of a continuous DC voltage supply and Ethernet link to the computer via a cable. This configuration allowed for direct transfer and subsequent processing of the images as they were being acquired.
This single camera photogrammetric configuration uniquely lends itself to measurement of large antennae and related equipment, especially in harsh or unstable environments. The INCA3 camera dependably operates within a wide temperature range for extended periods. V-STARS also has the unique ability to self-calibrate to the environmental conditions in which it operates. And finally, the ultra-portable, compact size of the system makes it ideal for fieldwork. The weight of the camera is less than 4 lb and its carrying case can be checked as carry-on luggage onboard a commercial aircraft. The total weight of the measurement equipment including camera, online kit, scalebars, targets, and shipping cases was less than 50 lb.
Lab testing and subsequent successful employment of the photogrammetry system on Black Island has shown the system’s ability to operate at -10ºC for extended periods. In the case of the Black Island project, the camera was operating for up to five hours per day. The results captured on Black Island at -10ºC showed no degradation when compared to results typically achieved at normal operating temperatures (0ºC to 40ºC).
The purpose of the photogrammetric survey was to determine what would be needed to upgrade the 11-m antenna for operation at Ku-band. If the reflector panels had a high enough surface accuracy, then it would be possible to merely change out the feed system. However, if the reflector panels were not accurate enough, then the feed and panels would need to be replaced.
The primary surface of the 11-m antenna consists of 36 contoured aluminum panels in two concentric rows. The inboard row is known as tier 1, and the outboard row is tier 2. Tier 1 panels are approximately 100 in. × 64 in. at the widest point. Tier 2 panels are approximately 100 in. × 50 in. at the widest point.
Because the 11-m antenna is the primary source of communications for McMurdo Station, it was critical that the outage time be minimized. For this reason, a subset of the 36 panels was selected for measurement. The panel selection was based on panels that would best represent the entire reflector surface.
Upon arrival, the VertexRSI team proceeded with visual inspections of the 11-m and 7.2-m antennae. The next day, field engineer Thad Ayers set up the photogrammetry equipment and computer in the radome for testing to ensure everything was working properly in the cold environment. The ambient temperature inside the radome was recorded at –10ºC. The equipment was left running for a period of two hours in the radome to allow it to acclimate to its environment.
The first scheduled antenna outage occurred at 5 p.m., and the team inspected the front surface of the panels in detail. Some ice had accumulated on the lower panels and had to be removed with the aid of a heat gun. The antenna was then run up to the zenith position, sometimes known as “birdbath position” or 90° elevation.
At zenith, the VertexRSI team placed retroreflective targets on panel 1, tier 2, and acquired test images to select the appropriate camera settings. Images were then obtained from several locations about the panel. This simple network design provided optimum target intersection angles. Photography was completed in less than four minutes.
Immediately after the data acquisition session concluded, Ayers processed the images using V-STARS software for verification of the data accuracy. In this case, the XYZ accuracy estimates were 0.0002 in., 0.0001 in., and 0.0001 in., respectively. This reported accuracy is consistent with other measurements performed on objects of similar shape and size in the normal temperature operating range, i.e., 0ºC to 40ºC.
Once the measurements were confirmed, the process was repeated on panel 2, tier 2. After panel 2 was photographed, the antenna was returned to its operational position.
On the following day, the second scheduled outage began at 5 p.m. Three more tier 2 panels and two tier 1 panels were photographed on the second night. In total, seven panels were measured. The antenna was then returned to its operational position. Due to the efficiency of the photogrammetric system, the antenna was returned to service an hour ahead of schedule.
In antenna manufacturing and inspection, surface accuracy is typically defined as how it conforms to a theoretical shape. The theoretical shape is defined as a mathematical model, the measured points are compared to the model, and the deviations or error vectors are displayed. The root mean squared (RMS) is calculated to condense the errors into a single value to describe the overall surface conformance. The average RMS of these panels is 0.0172 in. with an average peak-to-peak measurement of 0.1214 in. This is marginal at best for an antenna operating at Ku-band.
In the past, traditional measurement methods utilizing theodolites have played a primary role in antenna manufacturing, inspection, and alignment. But times have changed, and today’s ever-increasing levels of higher frequency communication antennae require higher surface accuracies over larger surface areas. These antennae are also located in more demanding and hostile locations. Manufacturers competing on the global market can no longer afford the accuracy limitations and time constrains of conventional inspection methods. The advent of lightweight digital cameras coupled to fast and reliable software has alerted many manufacturers to recognize the advantages of photogrammetry as their primary method for construction and diagnostic inspection.