Green Bank, WV--Sometime in the coming months, astronomer Mark McKinnon will point the newly operational Green Bank Telescope at a starry sky. He'll be searching for distant celestial objects that up until now have eluded the reach of more conventional stargazing equipment.
"One of the first things we'll do is look in places where we expect to find pulsars or neutron stars formed in supernova explosions," says Mark McKinnon, an astronomer who has waited nearly a decade for this opportunity. He is assistant director of the Green Bank Telescope (GBT) project at the National Science Foundation's National Radio Astronomy Observatory in West Virginia.
Some of the kinds of objects that McKinnon hopes to obtain images of are normally obscured by interstellar dust in visible light. But they also emit radio-wavelength radiation, which is observable by radio telescopes like the GBT. As their name implies, these telescopes differ from conventional optical telescopes in that they use radio waves instead of light waves to create images of the sky.
Whole new heights. Ten years in the making, the sheer size of the GBT will take radio telescope technology, which has been used to detect cosmic radio emissions since the late 1930s, to whole new heights. With a main reflector surface measuring 100m in diameter, the GBT will be the largest fully steerable radio telescope in the world.
And when it comes to telescopes, big is clearly better. "Since the peak gain (or sensitivity) of a reflector antenna is proportional to the ratio of the aperture diameter to the operating wavelength squared, the larger the reflector's diameter, the better the sensitivity," explains Roger Norrod, an electrical engineer who has been with the project since the onset. Put another way, says Norrod, the GBT can achieve gains that are one hundred times greater than a typical earth-station antenna with a 10m diameter.
Dennis Egan, the mechanical engineer at the Green Bank site, fondly refers to the GBT as a "giant light bucket" that will help astronomers like McKinnon obtain better images from distant objects and weak sources. But in order to achieve such a spectacular level of performance, he says, engineers had to come up with an ingenious way to maximize the usable area of the reflector surface. They also worked hard to achieve exacting tolerances in several critical areas of the telescope's design.
"A drawback of conventional telescopes is that a portion of the collecting area is blocked by the feed support structure, which means about 10% of the incoming radiation is thrown away," explains Egan. In order to increase the reflector surface area and eliminate problems with beam reflection and diffraction, the design of the GBT departs radically from conventional telescopes.
With a high degree of accuracy, six electric actuators allow translation of the GBT's subreflector, which directs the signal from the main dish into the receiver, in any direction. Exact position is critical in order to keep the subreflector aligned with the telescope's primary reflector and receiver when the feed arm and supporter structure deflect. When the receiver is off-center, the signal power is diminished.
The Green Bank Telescope's unusual asymmetrical design posed extreme challenges for engineers seeking to meet position requirement of only fractions of a millimeter. In particular, they had to compensate for the effects of wind, weather, and gravity on the telescope's 290,000-lb feed arm, which may deflect as much as 8 inches of water under worst-case conditions.
For one thing, every joint on the telescope's supporting structure is welded, rather than bolted, in order to maximize stiffness and rigidity. Also, because of the offset, "clear aperture" design, the surface of the GBT's primary reflector is elliptical rather than symmetrical. Making up the surface of the dish are 2,004 reflector panels, the position of each controlled by an individual electro-mechanical actuator. To achieve the required level of performance, the reflector surface tolerances can only be a small fraction of the operating wavelength," explains Norrod. "Those actuators allow us to position the GBT panels with an accuracy of just 1/30th of a 3-mm wavelength."
Engineers have also developed laser-ranging instruments that allow for accurate, real-time measurements of the reflector surface and position of critical parts of the antenna with respect to fixed-ground references.
But perhaps the most striking aspect of the design is the 240-ft-long vertical feed arm extending from one side of the dish. Weighing approximately 290,000 lbs, it supports the receivers and a second reflector. This subreflector measures approximately 8m in diameter and refocuses the received energy to the receiver feedhorns 11m below the primary focal point.
Norrod notes that dual reflectors, which the GBT will use at wavelengths shorter than 25 cm, are superior to a single reflector in terms of both sensitivity and field-of-view. But in order to fully exploit the capabilities of dual-reflector optics, the subreflector must be in near-perfect alignment with the primary reflector at all times.
Pinpoint accuracy. "To underscore the critical need for stringent tolerances, consider that the antenna main beam shifts about 3 arc-sec when the subreflector is moved 1 mm," says Norrod. "As a case in point, the GBT beamwidth at 6 mm wavelength is 14 arc-sec. At that wavelength, an error in sub-reflector positioning of only 5 mm would shift the main beam by more than a full beamwidth, making it possible to completely miss the object to be observed. In general, main beam pointing errors of 1/10th beamwidth are detrimental to most astronomical observations."
Moreover, design engineers had to compensate for the effects of wind, weather, and gravity on the telescope's 290,000-lbs feed arm, which can deflect up to as much as 8 inches. So in order to keep the subreflector aligned with the primary reflector, even under worst-case conditions, they designed a mechanism that translates and tilts the subreflector with five degrees of freedom using six electric actuators.
"Three points define a plane and by moving these three points you can rotate a plane," explains Egan. "But to translate the plane, you need a more complicated mechanism. The six actuators allow us to translate the subreflector in any desired direction, and to change its angle with respect to the receiver room."
The ActionJac(TM) electric actuators supplied by Nook Industries (Cleveland, OH) consist of a worm-gear speed reducer that transmits motion from one plane to another; a precision ball screw that converts rotary motion to linear motion; and support bearings, all enclosed within a housing. Each actuator has a capacity of ten tons, while individual stroke lengths vary between 21.75 and 41.25 inches.
|An Accurate Actuator -- The six electric actuators (Nook Industries) used to position the 4,500-lb subreflector consist of a worm-gear speed reducer that transmits motion from one plane to another, a precision ball screw that converts rotary motion to linear motion, and support bearings, all enclosed within a housing. To achieve the desired accuracy, design engineers matched the worm gear set for minimum backlash (play between the meshing teeth and mating gears); selected a ball screw with ground threads for high precision; and pre-loaded the nut to eliminate backlash. Further, a transducer mounted within the hollow shaft of the screw provides position feedback of a brushless dc motor.|
"The advantage of a power screw is that by using a motor brake, the actuator tends to stay in place. Plus, a relatively small brake holds a large load," explains Egan. "Other options that can transmit high forces, such as hydraulics, tend to drift and would require more feedback control to obtain the required accuracy."
Nook Industries, which manufactures linear motion equipment designed for precise positioning and lifting, began working with GBT contractors on the electric actuator requirements for the telescope in 1992. "A particular challenge was to provide a high level of performance under the harsh conditions to which these electric cylinders will be subjected," says Rick Christyson, chief engineer at Nook and head of this project.
In order to meet the stringent positioning requirements, Nook engineers who designed the electric cylinder surpassed the lead accuracy of conventional electric actuator technology by a factor of ten.
One way they accomplished this was by producing a ball screw with threads that are ground (as opposed to rolled) for higher precision. They also eliminated backlash (or play) in the screw, which typically ranges anywhere from about 0.002 to 0.013 inch depending on the specific geometry, by using a nut with an internal preload. Similarly, they matched the worm gear set to minimize the amount of play between the mating teeth of the worm gear and wheel.
Servo control was another path to accuracy, and it does not totally depend on the mechanical accuracy of the components. "In many cases, when engineers talk about accuracy, what they really want is repeatability," says Christyson. "In the case of positioning, it's so important to know not only what combination of mechanical actions will orient the device exactly as desired, but also that once that combination of actions is identified, it can always be repeated."
"In other words, we don't position the subreflector," explains Egan. "We tell it where we want it to go, and the software moves the individual actuators."
To achieve this high level of repeatability with the GBT's electric actuators, Nook engineers mounted a position feedback directly inside of the hollow ball screw. A magnetostrictive transducer provides an analog pulse-width measurement to a control card, which converts it to a digital length. Positioning resolution and repeatability is 0.00025 inch.
The end result? Sometime soon, when Mark McKinnon goes to look for a particular neutron star, he'll find it. And a whole lot more.
An accurate actuator
The six electric actuators (Nook Industries) used to position the 4,500-lbs subreflector consist of a worm-gear speed reducer that transmits motion from one plane to another, a precision ball screw that converts rotary motion to linear motion, and support bearings, all enclosed within a housing. To achieve the desired accuracy, design engineers matched the worm gear set forminimum backlash (play between the meshing teeth and mating gears); selected a ball screw with ground threads for high precision; and pre-loaded the nut to eliminate backlash.Further, a transducer mounted within the hollow shaft of the screw provides position feedback of a brushless dc motor.
Incredible in both scale and precision, the 100m-diameter Green Bank Telescope will allow astronomers to study a wider range of celestial objects. Engineers have built receivers for the telescope that work within a range of wavelengths from 100 cm to 6 mm (frequencies of 300 MHz to 50 GHz). Plans are underway to build receivers working to wavelengths as short as 3 mm (100 GHz). Conventional telescopes, on the other hand, require the receivers to be swapped out to study different frequency ranges.