2.1 Scientific goal and requirements
The scientific goal of the solar FUV-UV Spectrum measurement experiment is aimed to assess the energy from the sun deposited in the near-space environment. This requires precise measurement of the solar irradiance in a broad-band spectral range. Based on the study result of Haigh in 1999, the solar X-ray and EUV-FUV in wavelength less than 170 nm is completely absorbed in the upper atmosphere 50 km above the ground, and the atmospheric compositions of O, O2, O3 and N2 have a complex transmitting window in the wavelength between 170 nm to 210 nm that allows some of the solar FUV spectra in wavelength band to path through the upper atmosphere to arrive in altitude between 20 km to 50 km, which depends on the atmospheric parameters and the level of the solar activity. In the wavelength from 200 nm to 350 nm, the solar FUV goes through the near space as a function of the wavelength, from completely absorbed, partly absorbed to almost completely transmitting.
The SUVS provides precise spectral data in the wavelength 170–400 nm to fulfill the scientific requirements, with high sensitivity in the whole wavelength band, and a wide dynamic range of the flux magnitudes. The special characteristic parameters of the SUVS are listed in the Table 1.
Table 1
Performance of the SUVS onboard the “HongHu-6” high altitude balloon
| Wavelength Band | 170 ~ 400 nm |
| Spectral Resolution | 0.1 nm |
| Flux Dynamical Range | 2.32×10− 10 ~9.92×10− 7 Ergs/pixel/s |
| Field of View | ± 1° |
| Cadence | 1s |
2.2 Instrument
The SUVS was installed on the payload platform dragged by the balloon. The rope between the downside of the balloon and the payload platform was about 80 meters in length. In the flight site located in about 37º northern latitude, the Solar Zenith Angle (SZA) changes between about 100º and 38º from sunrise to noon in late September, as shown in Fig. 1. This allows the SUVS FOV to avoid blocking by the balloon body. In order to keep continually monitoring the solar FUV-UV spectra, the SUVS was mounted on a two-dimension Solar Pointing System (SPS) to automatically search and trace the sun. The SPS was guided by a solar sensor, which provided accurate position information of the sun in 20ms cadence. The SPS was locked while the balloon flied lower than 10 km or in the night time. After the balloon arrived in altitudes upper than 10 km and the sun rose up, the SPS was unlocked by telecommand and started to search the sun. Once the sun was caught in sight, it was kept in the center of the SUVS’s FOV, with the pointing accuracy better than 0.1º.
Since the solar disk is a surface light-source about 0.5º in FOV, and the payload platform of the balloon always kept in motion during flight experiment., the Roland optics provide an ideal solution for the SUVS to keep the spectra positions stable to achieve high spectral resolution performance. In a simple configuration of Roland optics, distance between the slit and the grating, and that between the grating and the focal plane, should be long enough to achieve high wavelength-resolution, and a large detector is needed to cover the spectra space from 170 nm to 400 nm, which introduce great challenges to develop an instrument in normal dimensions carried by a balloon platform. In order to make a compact instrument to adapt to the flight requirements, an improved Roland-optics was designed, with two planar mirrors to fold the incidence arm and a concave reflecting mirror to fold the diffracting arm and focus the diffracting spectra to a much smaller focal plane, as shown in Fig. 2.
The incidence slit is 60 µm in width and 10 mm along the slit. The concave Roland grating is 498.1 mm in radius, with line density of 2700 lp/mm, provided by Horiba company. The focal plane is in concave shape with the span-length of 170 mm, which is unable to be matched by a single plane detector. In order to cover all the spectra from 170 nm to 400 nm, 2 sets of the Roland optics and 6 silicon detectors were used. 3 one-dimension silicon detectors, 1 piece of back-thinned CCD sensor and 2 pieces of back-thinned CMOS sensors, work as a group for each set of the optics. The two groups of detectors are alternatively placed along the concave focal plane to avoid structural intervention, respectively covering the spectra of 170–210 nm, 210–250 nm, 250–290 nm, 290–330 nm, 330–370 nm, 370–405 nm, as shown in Fig. 3. The CCD sensors are more sensitive in shorter wavelength of 170–250 nm. Both the CCD and CMOS sensors, producted by Hamamatsu company, have quartz glass windows to allow FUV-UV irradiance in wavelength longer than 160 nm incidence. Each of the CCD sensors has 2048×1 effective pixels, 14µm × 500µm each pixel. And the back-thinned CMOS sensor has 2048×1 effective pixels and 14µm × 200µm each pixel.
In the total energy of solar electromagnetic irradiance, the FUV-UV spactra are small in ratio comparing to the white light and the infrared compositions. The zero order of the diffraction grating is trapped inside the SUVS, to supress the zero order energy. On the other hand, since the second order of diffracting spectra in 170-200nm are overlapped to the first order of 340-400nm, as shown in Fig. 5, we used a type of band-passing filter, FGS900-A provided by the Thorlabs company, to suppress the higher diffracting contamination for the CMOS-1B and CMOS-2B sensors. The transmission efficiency of band-pass wavelength filters used for CMOS-1B and CMOS-2B are shown in Fig. 6.