The telescope tube contains a 30 cm primary and 7.2 cm secondary mirror which are both made from Zerodur. The optical train has a primary f-ratio of f/2.0 increasing to f/12.72 after the secondary. The primary mirror is mounted on a strong back for stability and the secondary mirror is mounted onto spider arms. To maintain focus the mirrors are separated by thermally stable INVAR metering rods. The boresight is near the center of each CCD.
For light rejection there are internal and external baffles. The external baffles are forward of the secondary mirrors and help prevent scattered light from reaching the detectors. The internal baffle lines the inner walls of the telescope tube between the primary and secondary mirrors. Secondary/primary baffles also surround the secondary mirror and the hole at the center of the primary. Behind the primary mirror is the Beam Steering Mirror which directs light to one of the two detectors.
Before the light enters the detector it passes through a filter housed in a filter wheel. Each filter wheel contains the following elements: a blocked position for detector safety, UV-grism, uvw2-filter, v-filter, uvm2-filter, optical-grism, uvw1-filter, u-filter, magnifier, b-filter, and white-light-filter. The characteristics of the UVOT lenticular filters can be found in Table 8. The lenticular filter responses (Figure 9) are also provided. The grisms supply a low spectral resolution. The magnifier offers a 4-times increase in the image scale which increases the f-ratio to f/54 in the blue and provides diffraction-limited images. It does not operate in the UV because of transmission limitations in this part of the spectrum. Because the focal plane is curved, the filters are weakly figured and the surface of the detector window is concave. The UVOT PSF has an approximate FWHM of 2.5 arcsec, and there are small variations in the width of the PSF between the UVOT filters. The PSF has also been found to vary with temperature, so it can not be assumed to be constant with time. At present the time-dependent variation in the PSF is not well characterized, but it approximately corresponds to a 15% variation in FWHM. Significant changes in the PSF beyond this should be reported to the instrument team immediately. The details of the UVOT PSF are currently being investigated and will be reported in an upcoming publication.