Chapter 4. C1 Telescope

In his pioneering work on the design principles of the coronagraph, Lyot identified five sources of stray light (diffracted and scattered light) in a simple objective lens telescope:

  • the diffraction pattern from the aperture of the objective lens
  • a spurious solar image produced by multiple interreflections in the objective lens
  • macroscopic inhomogeneities in the glass surface inhomogeneities of the lens (pits, scratches)
  • body-scattering within the glass

Lyot designed the coronagraph to allow viewing of the faint solar corona from the ground outside of times of total solar eclipse. In the usual lens version of the coronagraph, a solar image is formed by the objective lens. Photospheric light from the disk is blocked by a blackened occulting disk. A field lens re- images the objective lens and its diffraction pattern, but a "Lyot stop" prevents the diffraction ring from reaching the focal plane, and a "Lyot spot" blocks the spurious image caused by multiple interreflections in the objective lens. With this design, Lyot found that he could eliminate stray light caused by aperture diffraction and by multiple intrreflections in the objective lens, which are the major contributors to telescope stray light. The other three stray light sources mentioned above remain, but can be reduced by selecting very clear glass, and polishing it well. Lyot constructed coronagraphs with a residual stray light level of 5 x 10E-6 Bsun, where Bsun is the disk average solar brightness. The scattered skylight level, even on a very high mountain top, is seldom less than 10 x 10E-6 Bsun, and Lyot's coronagraphs were able to observe the corona from the ground out to radial distances of around 1.3 Rsun. The cover of this Handbook illustrates an image taken during the 1981 July 31 solar eclipse with a normal telescope; of course, the coronagraph obtains images outside of times of total solar eclipse.

Coronal observations from space are limited only by the stray light generated in the instrument. So far, all space-borne coronagraphs have been "externally occulted" instruments that use a principle first described by J.W. Evans (see Chapter 5 and Chapter 6 on C2 and C3). The inner corona, however, has not been observed with high spatial resolution because of the increasing distance in front of the objective lens at which the external occulter must be placed, because of vignetting at the inner edge of the field, as the occulted area becomes smaller in angular radius. With the advent of superpolished mirrors and extremely smooth coatings, a new modification of the internally occulted Lyot coronagraph is possible, here called the "Lyot mirror coronagraph."

4.1 C1 Optics

In order to image the corona with high spatial resolution very close to the limb, an internally occulted system is necessary. C1 implements this requirement with a mirror telescope design. An off-axis, superpolished, parabolic mirror forms an image of the Sun on a convex mirror. All photospheric light enters through a hole in the convex mirror, which serves as the internal occulter and dumps the light to the outside of the instrument. A Lyot stop blocks the diffracted light arising from the entrance aperture. Since there are no multiple interreflections in a mirror Lyot coronagraph, the Lyot spot can be omitted. In addition, a Fabry- Perot interferometer acts as a narrow bandpass, tunable filter to isolate individual coronal emission lines, further reducing the effect of the stray light background. The filter itself is located behind the Lyot stop, reducing the light that would otherwise be scattered from its surfaces.

Figure 4-1 (low resolution) or Figure 4-1 (high resolution) shows a conceptual optical diagram for C1. Light enters a 4.7 cm diameter entrance aperture A0. The objective mirror M1 is an off-axis paraboloid of 75 cm focal length, located 127 cm behind A0. The solar image is formed at the prime focus, on a convex annular mirror M2. This mirror provides a field stop, and also performs the function of the Lyot internal occulter by removing the unwanted photospheric image from the system. It does this by means of a smooth-edged central hole through which the disk image falls, to be subsequently ejected from the system by a diagonal rejection mirror. The field mirror M2 determines both the inner and outer radius of the field-of-view. The field mirror relays the light of the coronal image onto a second off-axis paraboloid collimating mirror M3. M3 is identical to M1, and is in principle a second segment cut from the same superpolished telescope mirror as M1. It is positioned to have the same axis and focal point. This fully symmetric arrangement accomplishes complete cancellation of coma in the system. The field mirror M2, together with M1 and M3, forms an image of the A0 entrance aperture in a plane 70 cm in front of M3. Here the 4.0 cm diameter Lyot stop A1 is placed to intercept the bright ring of light formed by diffraction of the sunlight incident on A0. The divergence of the image beam passing through A1 is a 0.8 half- angle, corresponding to the 3.0 Rsun field limit. This small divergence makes possible satisfactory performance of the Fabry- Perot filter.

The collimated beam of coronal light from M3 passes through the Lyot stop, and is sent through a narrow bandpass, tunable Fabry- Perot (FP) cavity filter, where the coronal light is analyzed spectroscopically. The FP is an interference (comb type passband) filter, which passes many interference orders. It works in conjunction with a broader blocking filter, with a bandpass sufficiently narrow to eliminate all but one selected order passed by the FP, but broad enough to allow tuning of the FP over nearly a full free spectral range (the wavelength range between successive orders). A second, very broad bandpass filter (110 nm) passes many orders, and serves as a "white light" channel. During a series of exposures, the FP bandpass is stepped across a part of the free spectral range to cover a desired spectral line, with the scanning actuated by piezoelectric drivers.

The beam from the FP is sent to a telephoto lens. Filter and polarizer wheels, and the shutter, are located between the telephoto lens and the focal plane. The telephoto lens focuses the final coronal image onto a 1024x1024 pixel CCD camera. The 21 m square pixel size subtends an angle of 5.6 arc seconds in the coronal image, for a resolution of 11.2 arc seconds (2 pixels).

Laboratory measurements of the C1 system demonstrated adequate imaging of the solar disk for sharp occultation at 1.1 Rsun, imaging of A0 on A1 with sufficient quality for the required stray light reduction, and spatial resolution over the entire field to 3.0 Rsun of better than 3 arc sec (although the actual instrumental resolution is set by the CCD pixel resolution).

The M1 objective mirror is able to perform small angular movements (see Chapter 7). One use of this ability is in dynamic imaging, where a series of four images is obtained in which the M1 mirror is moved through an angle corresponding to half a CCD pixel in both dimensions. This is equivalent to oversampling the image, and can potentially improve the spatial resolution by a factor of 2.

With the door closed, the coronagraph will see the rear of a Sun-illuminated diffuser set in the door. This simulates a flat field source at the corona, with the same spectral character as the stray disk intensity, providing a known level of solar disk illumination at the CCD. In addition, internal calibration lamps (redundant) located near the front door can be used to illuminate the diffuser to provide an additional, calibrated incremental intensity. A photodiode (redundant) will monitor the absolute level of illumination from the lamp to detect any changes in the door surface characteristics or lamp degradation. Both illumination sources are used to calibrate the telescope optics and instrumentation.

4.2 The Fabry-Perot Interferometer

The scientific objectives of LASCO required a spectrometric instrument, capable of recording the inner E (emission line) and K (electron scattered) corona over a three solar diameter field-of- view, with high spatial, spectral, and temporal resolution. In the C1 design, a Fabry-Perot interferometer is used as a narrow passband, tunable filter. Sets of individual monochromatic images can be obtained by scanning over a range of wavelengths around each observed coronal emission line. The tunable passband also allows the faint coronal signal to be accurately separated from the instrumental stray light background using differential imaging techniques.

The low coronal signal levels, in combination with the high stray light level of an internally occulted coronagraph, result in background noise - limited signal detection over most of the field- of-view. The Fabry-Perot has a bandpass (FWHM) of approximately 0.07 nm, and a tunable range of approximately +/- 1 nm (+/- 500 km/s doppler velocity), both differing slightly for each of the chosen lines. Blocking filters, which are used as order sorters, have a free spectral range (interorder separation) of 3.5 nm. With the 0.07 nm bandpass, this can only be achieved if the finesse (the ratio of the free spectral range to the spectral bandpass) is larger than 50.

Cavity length and wedge defect (departure from parallelism) of the interferometer plates are controlled from a microprocessor that uses capacitance micrometers for position sensing. Three piezoelectric transducers determine the cavity length. This combination allows automatic adjustment of wavelength and finesse. The coronal spectrum is referenced to the solar disk Fraunhofer spectrum using three optical control channels, which are located on the perimeter of the interferometer plates.

The coatings of the interferometer plates limit the useful wavelength range to approximately 120 nm. The coating was selected to cover the range from Fe XIV at 530.3 nm to H-alpha at 656.2 nm. Two additional channels have been selected for the coronal emission lines of Fe X at 637.4 nm and Ca XV at 569.4 nm. The Na I D line channel at 589.0 nm allows absorption line spectroscopy in the corona for separation of the K coronal and stray light components of intensity. A channel from 530 nm to 640 nm, covering many orders of the interferometer, is used for white light imaging. As discussed previously, blocking filters are used to eliminate all but one selected order passed by the Fabry-Perot. There are four of these blocking filters, one for each of the spectral lines of Fe XIV, Ca XV, Na I D, and Fe X, in the filter wheel. The last filter wheel position contains the broad bandpass "white light" filter. The polarizer wheel contains three polarizers at 120 degrees, a blocking filter for H-alpha, and a clear glass position. Observations of the Fe XIV, Ca XV, Na I D, and Fe X lines are made using the appropriate blocking filter, and the clear glass polarizer wheel position or one of the three polarizers. Observations of H-alpha are made using the H-alpha blocking filter in the polarizer wheel, and the "white light" filter in the filter wheel. The white light observations are made using the "white light" filter in the filter wheel, and the clear glass polarizer wheel position or one of the three polarizers. Table 4-1 summarizes the available channels using the Fabry-Perot.

The Grotrian technique will be used to separate the K coronal and stray light components of intensity, using the Na I D Fraunhofer absorption line. Intensity measurements at the Na I D line center and at a nearby continuum wavelength are made with the door open and closed. With the door closed, the diffuser plate (discussed above) introduces a photospheric solar disk spectrum, with an instrumental stray light level which is ignorable. Its Na I D line profile is similar in its wavelength behavior to the instrumental stray photospheric light which is present with the door open. With the door open, a measurement of the coronal spectrum gives the K coronal and stray light components of intensity at the Na I D line center wavelength, and the K coronal and stray light component at the nearby wavelength. With the door closed, the ratio (Na I D line center) / (continuum) is found, and with the door open, the ratio (K corona+Na I D line center stray light) / (K corona+continuum wavelength stray light). Assuming that the first ratio also gives the ratio (Na I D line center stray light) / (continuum stray light) for the open door ratio, the K corona intensity can be separated from the stray light component.

The E corona intensity component of the coronal emission lines observed by C1 can be determined by normalizing two sets of measurements, with the door open and closed, at wavelengths outside the emission line, followed by subtraction of the closed door (diffuser) set from the open door (coronal) set at all passband central wavelengths where the emission line is present. Given a reasonably stable instrument, this method should provide coronal measurements whose precision is determined primarily by the background noise - limited observation statistics.

Table 4-2 lists the expected precision of two of the emission line measurements, calculated using estimated stray light levels which were taken from the original proposal. These have since been further reduced, and a significant improvement can be expected.

Table 4-3 lists potential diagnostic methods for use by C1. Coronal temperatures (assuming T(e) = T(ion)) can be deduced from line intensity ratios using the three forbidden lines, but Ca XV will be useful only in active regions. The three lines can also be used in density diagnostic line intensity ratios. Nonthermal velocities can be derived from line profiles, and coronal dynamics from doppler shifts. Magnetic field directions can be determined from the Hanle effect polarization in Fe XIV. Intensities of the Na I D line can be used to separate the K corona from the stray light component of intensity. Outflows in the corona may be measurable by the doppler dimming in H-alpha, but this remains to be demonstrated.

Return to Title Page or Go to Next Chapter