Open Access
Add to BibSonomyAbstract
We introduce a novel telescope consisting of a pinhole-like camera with rotatable MEMS micromirrors substituting for pinholes. The design is ideal for observations of transient luminous phenomena or fast-moving objects, such as upper atmospheric lightning and bright gamma ray bursts. The advantage of the MEMS “obscura telescope” over conventional cameras is that it is capable both of searching for events over a wide field of view, and fast zooming to allow detailed investigation of the structure of events. It is also able to track the triggering object to investigate its space–time development, and to center the interesting portion of the image on the photodetector array. We present the proposed system and the test results for the MEMS obscura telescope which has a field of view of 11.3°, sixteen times zoom-in and tracking within 1 ms.
©2008 Optical Society of America
1. Introduction
In the fields of astrophysics and atmospheric science there are many interesting phenomena that occur rarely, randomly and for very short time periods. These include Gamma Ray Bursts (GRB) and Transient Luminous Events (TLEs). GRBs have been detected with an average rate of once a day over the full sky and their typical lifetimes are from milliseconds to a few minutes [1]. One of three instruments of the SWIFT gamma-ray observatory is the UV/optical telescope [2]. It captures early UV and optical photons from afterglow of GRBs. The observation of the early photons is limited by slewing the satellite to track the GRB after trigger which takes several tens of seconds. TLEs have been reported in the form of Blue jets, Sprites, Elves and Gigantic jets occurring between the stratosphere and ionosphere with rapid development times (1~100 ms) and large dimensions. The size of TLE is of the order of magnitude 10–100 km [3–5]. The observation of the TLEs has been made using CCD cameras with time frame of tens of milliseconds. Observing extremely rapid phenomena such as these from space requires an instrument that possesses both a wide Field of View (FOV) to find the event, and is able to rapidly follow up the triggered event to allow more detailed study.
An idea of next generation telescope—based on leading edge technology in optical Micro- Electro-Mechanical Systems (MEMS), has been proposed for space-based observations of transient events and fast-moving objects [6,7]. The MEMS telescope has the unique feature of both a wide FOV, as well as the capacity for fast zoom-in and tracking. When used as a reflector, micromirror arrays are able to change the viewing angle, which allows the tracking of fast-moving objects. Multiple micromirror reflectors may be arranged with different focal lengths to allow for a variety of applications such as zoom-in. MEMS technologies have been widely used for many optical applications. The most commonly used optical MEMS device is the micromirror, which has a typical size of hundreds of micrometers. The micromirror is fabricated in a frame of standard silicon technology. It has a rapid tilting speed (typically several microseconds), and high reliability when compared to large conventional mirrors rotated via electromagnetic motors [8–14]. It has low power consumption and provides accurate and robust operation for optical applications.
In this paper we present a novel type of telescope, called an obscura telescope, in which the pinhole is replaced by a small micromirror array. This is the first step toward the application of MEMS micromirrors in large tracking mirrors for space experiments. Pinhole cameras can be used for observations of bright objects, and have the advantage that they are free of aberrations. The image is recorded using an orthogonal light sensor grid. The FOV of a camera with one pinhole is limited by the size of the photo sensor [15]. The proposed telescope consists of two cameras: one is a wide FOV camera and the other is a zoom-in camera with a micromirror array. Fast rotatable micromirrors allow important functions such as fast zoom-in and tracking of a given object. The micromirror is able to direct light quickly to photo sensors and allows the instrument to track a moving light source—which is impossible for pinhole cameras. A Multi-Anode Photo-Multiplier Tube (MAPMT) is used as the image sensor because of its high temporal resolution. In the following sections we present the principle of the proposed telescope, along with the test results.
2. Principle and construction of MEMS obscura telescope
A schematic diagram of the obscura telescope with MEMS micromirror array is shown in Fig. 1. The telescope is composed of a trigger camera and a zoom-in camera. The mirror in the trigger camera (called the Trigger Mirror, TM) is positioned at a distance L 1 from the focal plane detector and is used for locating the object within the wide FOV. The mirror in the zoom-in camera (called the Zoom Mirror, ZM), is installed at distance L 2≫L 1, and is used to detect the object image with higher lateral resolution, which provides a zoom-in effect. The focal length of the TM (L 1) is 90 mm, making an FOV of 11.3°. The FOV in the ZM regime is 2.9° with a focal length (L 2) of 360 mm.
When the telescope uses the TM for observations from a satellite at a height of 800 km above the Earth’s surface, it views an area in the atmosphere of 160×160 km2, and one pixel collects light from an area of 20×20 km2. When using the ZM, the telescope views an area 16 times smaller than it does with the TM, meaning that an object can be observed in greater detail via the ZM. Figure 2 shows the simulated image of a light source at the focal plane detector (8×8 square pixels). The light source has a circular Gaussian intensity distribution with a diameter of 20 km. The simulated images are shown for both the TM and ZM. It should be noted that the ZM is tilted to align the light source with the center of the focal plane. As shown in the simulation results, the telescope can observe the light source detected by the TM at a higher lateral resolution when using the ZM.
Fig. 1. Schematic diagram of the MEMS obscura telescope. The Trigger Mirror (TM) is positioned close to the focal surface to give the telescope a wide Field of View (FOV) (area 1) for event finding. The Zoom Mirror (ZM) is installed with a longer focal length to give the telescope a higher lateral resolution (area 2).
Fig. 2. Simulated telescope imaging: (a) the input light source distribution at the ground, (b) the image at the focal plane, 800 km above the ground as viewed using the TM, and (c) the zoom-in image as viewed using the ZM.
In the trigger mode, the TM reflects light from sources within the observation area to the photodetector. The signals reaching the photodetector are quickly analyzed in order to determine the direction of the source within the FOV (area 1 in Fig. 1). The data are examined using integration times much shorter than the expected duration of the event. In the wide field mode a typical event will result in between one and a few photodetector pixels being above the threshold. Once the direction is determined, the ZM is rotated in correspondence to the TM information so that the image (shown in area 2 in Fig. 1) is located at the center of the photodetector. In the zoom mode, the lateral distribution in the photodetector can be used to locate the brightest point of the event at the center of a MAPMT.
A FPGA (Field Programmable Gate Array) circuit receives the digitalized photodetector signal and the data are then analyzed to make a trigger decision. In parallel with this, the position of the brightest point of the event is stored to determine the tilt angle for the ZM. When the trigger condition is satisfied, the tilting angle information is used to determine the bias values to be applied to the micromirrors. The micromirror driving circuit then applies the necessary voltage to change the orientation angle. In this way, zoom-in and tracking are achieved by the ZM while the event is selected from a wide FOV using the TM. In situations where still higher resolution is required, it would be possible to add a third mirror (not shown in Fig. 1), with an even longer focal length. It is important that the ZM should rotate its viewing angle rapidly enough to observe the event immediately after the trigger, so as not to miss a significant part of the event under investigation.
Figure 3 shows the structural design drawing and a photograph of a fabricated MEMS obscura telescope. In the telescope, a plane mirror with a size of 3 mm×3 mm is used for the TM and a MEMS micromirror array is installed for the ZM. Details of the design and fabrication of the MEMS micromirror array to meet the requirements of the telescope are discussed in the next section. A MAPMT with UV filters is used as the focal plane detector since it meets the response time and spectral bandwidth requirements for the instrument. The square MAPMT (a Hamamatsu H7546A) is subdivided into 8×8 square pixels, each 2 mm×2 mm. The size of the telescope window is determined by the required FOV. Inside the telescope, the walls are baffled to minimize the background due to the scattering of light within the telescope.
The MEMS obscura telescope described above will be carried into orbit by a Russian microsatellite, Tatiana-II, scheduled to be launched in February 2009. The primary aim of the mission is to observe TLEs over a mission duration of at least one year. Recently, observations of TLEs have raised several questions concerning global electrical phenomena in the atmosphere, and the influence of these phenomena on atmospheric properties. The study of flashes at near UV wavelengths (300–400 nm) is important, as the radiation in this wavelength range is directly related to the ionization of the atmosphere by charged particles.
Fig. 3. The structural design drawing (a) and photograph (b) of a fabricated MEMS obscura telescope.
3. Design and fabrication of the MEMS micromirror for the obscura telescope
A mirror rotatable in two orthogonal directions is required for the ZM, because it has to align the event image with the center of the focal plane. The micromirror array employed in the proposed telescope is composed of 8×8 small square reflector pixels. The micromirror is designed to be driven by two-axis electrostatic vertical comb actuators that allow continuous changes to the viewing angle of the mirror plate biaxially. All the micromirrors in the array are actuated with the same tilting angle using the control voltage. Single-crystalline silicon has been selected as the structural material for the micromirror because of properties such as a negligible residual stress, high yield strength, high temperature resistance, and a flat surface.
A micromirror consists of three parts: an actuator with top and bottom comb electrodes, a mirror plate, and a glass substrate with electrical lines. A schematic view of the mirror actuator is shown in Fig. 4(a) and images of a fabricated micromirror taken using a Scanning Electron Microscope (SEM) are shown in Figs. 4(b) and 4(c). The vertical comb structures are fabricated on a Silicon-On-Insulator (SOI) wafer to construct the actuation part of the mirror. The bottom silicon layer of the SOI wafer is patterned to form the comb electrodes to which the actuation voltage is applied, and the top silicon layer is used as the ground electrode. The moving part of the actuator has a gimbal-like frame.
A DC bias applied to the comb electrodes attached to the frame provides the electrical torque to tilt the frame, while the DC bias at the comb electrodes located inside the frame generates the torque to tilt the inner plate. Two orthogonal pairs of springs allow the mirror plate to be tilted independently in two orthogonal directions. By combining rotations along the two axes, the tilt angle of the micromirror can be controlled in any direction. This is done in response to the data from the camera control circuit, which determines the direction of the event of interest. The mirror plate (with a size of 340×340 µm2) is formed at the inner plate of the actuator using a wafer bonding process and is coated with an Al layer to make a highly reflective surface. The actuator part and the glass substrate with electrical lines are bonded together to make electrical contact between the comb electrodes and addressing lines. The actuator and addressing lines can be hidden behind the mirror plate, resulting in a high fill-in factor of 84%.
Fig. 4. The MEMS micromirror for two-axis rotation: (a) a schematic view of the mirror actuator, (b) SEM image of the comb actuator with the mirror plate removed and (c) micromirror array.
4. Telescope performance
For the proposed telescope the most important parameters of the MEMS micromirrors are: (1) the angular tilting range available for the ZM, and (2) the angular velocity of the mirror actuators. In order to measure the motion of the mirror plate with an applied bias, a spot optical measurement system was used. The measurement system was composed of a fiber coupled helium–neon (He–Ne) laser source, a Position Sensing Detector (PSD), an XYZ stage and other optical instruments. The laser beam was collimated using a lens with a 20 mm focal length and was aligned to the center of a single cell of the micromirror array. The laser source was reflected by the mirror plate so that it was incident upon the PSD. The angular displacement of the mirror plate was directly measured using the output signal of the PSD. While the input voltage was applied to the electrode of a mirror cell, all the other electrodes were grounded. The static response of a single cell was characterized by applying the voltage to each electrode.
Figure 5 shows the tilting angles of the micromirror with respect to the applied voltage. The maximum measured tilting angles were 2.2° and 6.2° for the two rotational axes, for voltages of 45 V and 75 V, respectively. The tilting angle of the mirror is smaller than the simulated value (7°), because the designed dimensions of the actuator may change during the fabrication process. Another possible reason for this is that only a portion of the input voltage is applied to the comb electrodes because of poor contact between the silicon comb electrodes and the metal addressing lines in the fabrication process. The measured contact resistance for five of the connections is about 1 MΩ, which indicates that the metal and the silicon may not be fully fused. The dynamic response of the micromirror was measured. The torsional resonance of the structure was detected at frequencies of 2 kHz and 1 kHz for two rotational axes, respectively, which shows that the micromirror is capable of making observations of fast-moving objects such as TLEs when used as a component of the proposed telescope.
The performance of the fabricated telescope has been tested using UV LEDs as a light source, which have a peak intensity at a wavelength of 400 nm with FWHM (Full Width at Half Maximum) of 20 nm. The distance between the telescope and the light source was 6 m and the detector images from the TM and ZM were recorded, along with the control of the micromirror tilting angle for different positions of light source. The diameter of the UV LED light source is equivalent to a 20 km source observed from an altitude of 800 km.
Figure 6 represents the measured light intensity at each pixel as a function of time. The UV data from 64 pixels were obtained with an initial sampling speed of 10 µs to enable the transient response of the micromirror during the first 1 ms to be studied. The sampling speed was then reduced to 1 ms for the next 100 ms. As shown in Fig. 6, the micromirror can track the light source and center the image on the MAPMT within the first 1 ms. The pulse detected in some pixels in advance of the main signal is associated with an oscillation of the micromirror due to its finite settling time (hundreds of microseconds).
Figures 7, 8 and 9 shows the test results of the MEMS obscura telescope. The responses of 64 pixels of MAPMT from the TM and ZM are shown. The z-axis denotes the digitized light intensity detected by each pixel. Figure 7 represents the recorded image from the TM at the start of the event and Fig. 8 shows the images after the ZM points at the light source and zooms into the center of the source. Figure 9 shows the tracking of the light source by the ZM, when the location of the light source is changed. When the position of the light source changes, the control circuit adjusts the bias voltage for the micromirror to align the tilting angle to center the source. These test results show that, qualitatively, the main features of the proposed telescope—wide field of view, fast identification, zoom-in on trigger, and tracking of a given object—are successfully demonstrated.
In principle, the features of the proposed MEMS obscura telescope might be accomplished by using sixteen independent conventional pinhole cameras each associated with its own photo detector. Such a large multiple pinhole camera is, however, practically not possible in terms of space, power consumption, photo detector and electronics channels, and complexity in optical alignment of multiple cameras. While slewing of a single pinhole camera itself is potentially feasible solution, not only it would increase the packaging space and payload complexity, but also the slew rate is far below the tracking speed of the proposed telescope.
Fig. 6. The measured UV data from 64 pixels. Each box shows the light intensity of a pixel as a function of time. (a) Data taken from the TM. (b) Data taken from the ZM which moves the zoom-in image to the center of the detector.
Fig. 9. Detector images of TM (left) and ZM (right) which “track” the light source moved from the point shown in Fig. 8.
5. Conclusion
A new obscura telescope, exploiting recently developed MEMS micromirror technology is proposed. A two-axis vertical comb-driven micromirror array, which is suitable for observation of fast-evolving phenomena, has been designed and fabricated for the proposed telescope. The telescope is composed of a plane mirror and a rotatable MEMS micromirror array located at different focal lengths. We have demonstrated the performance of the micromirror and the important functions of the telescope: a wide field of view surveillance, zoom-in on the object of interest, and tracking of fast-moving objects with high accuracy. The proposed telescope can be readily applied to the space-based observations of fast transient phenomena such as GRB and TLEs.
Acknowledgments
This work was supported by Creative Research Initiatives (RCMST) of MOST/KOSEF.
References and links
1. D. Band, J. Matteson, L. Ford, B. Schaefer, D. Palmer, B. Teegarden, T. Cline, M. Briggs, W. Paciesas, G. Pendleton, G. Fishman, C. Kouveliotou, C. Meegan, R. Wilson, and P. Lestrade, “BATSE observations of gamma-ray burst spectra. I - Spectral diversity,” Astrophys. J. 413, 281–292 (1993). [CrossRef]
2. P. W. A. Roming et al., “The swift ultra-violet/optical telescope,” Space Sci. Rev. 120, 95–142 (2005) [CrossRef]
3. V. P. Pasko, “Electric jets,” Nature 423, 927–929 (2003). [CrossRef] [PubMed]
4. Y. Yair, C. Price, Z. Levin, J. Joseph, P. Israelevitch, A. Devir, M. Moalem, B. Ziv, and M. Asfur, “Sprite observations from the space shuttle during the Mediterranean Israeli dust experiment (MEIDEX),” J. Atmos. Terr. Phys. 65, 635–642 (2003). [CrossRef]
5. H. T. Su, R. R. Hsu, A. B. Chen, Y. C. Wang, W. S. Hsiao, W. C. Lai, L. C. Lee, M. Sato, and H. Fukunishi, “Gigantic jets between a thundercloud and the ionosphere,” Nature 423, 974–976 (2003) [CrossRef] [PubMed]
6. I. H. Park, G. K. Garipov, and B. A. Khrenov, “Tracking mirror for measurements of extreme energy cosmic rays from space,” Proc. 28th ICRC Cosmic Ray Conf. (Tsukuba) HE1.5, 1109–1112 (2003).
7. I. H. Park, “MEMS based space telescope for extreme energy cosmic rays experiments,” Nucl. Phys. B -Proc. Suppl. 134, 196–201 (2004). [CrossRef]
8. P. F. Kessel, L. J. Hornbeck, R. E. Meier, and M. R. Douglass, “A mems-based projection display,” Proc. IEEE 86, 1687–1704 (1998). [CrossRef]
9. D. Hah, C. A. Choi, C. K. Kim, and C. H. Jun, “A self-aligned vertical comb-drive actuator on an SOI wafer for a 2D scanning micro mirror,” J. Micromech. Microeng. 14, 1148–1156 (2004). [CrossRef]
10. Y. Hishinuma and E. H. Yang, “Piezoelectric unimorph micro actuator arrays for single-crystal silicon continuous-membrane deformable mirror,” J. Microelectromech. Syst. 15, 370–379 (2006). [CrossRef]
11. I. W. Jung, U. Krishnamoorthy, and O. Solgaard, “High fill-factor two-axis gimbaled tip-tilt-piston micromirror array actuated by self-aligned vertical electrostatic comb drives,” J. Microelectromech. Syst. 15, 563–571 (2006). [CrossRef]
12. J. Kim, D. Christensen, and L. Lin, “Monolithic 2-D scanning mirror using self-aligned angular vertical comb drives,” IEEE Photon. Tech. Lett. 17, 2307–2309 (2005). [CrossRef]
13. P. J. Rodrigo, I. R. Perch-Nielsen, and J. Glückstad, “High-speed phase modulation using the RPC method with a digital micromirror-array device,” Opt. Express 14, 5588–5593 (2006). [CrossRef] [PubMed]
14. D. Dayton, J. Gonglewski, S. Restaino, J. Martin, J. Phillips, M. Hartman, S. Browne, P. Kervin, J. Snodgass, N. Heimann, M. Shilko, R. Pohle, B. Carrion, C. Smith, and D. Thiel, “Demonstration of new technology MEMS and liquid crystal adaptive optics on bright astronomical objects and satellites,” Opt. Express 10, 1508–1519 (2002). [PubMed]
15. G. K. Garipov and B. A. Khrenov, “Camera obscura for observation of EAS images in Cherenkov light,” J. Phys. G: Nucl. Part. Phys. 20, 1981–1988 (1994). [CrossRef]
