Dark horizontal streaks or stripes in the spectrum are caused by dust deposits on the slit plate of the spectrograph. A partial attenuation of the spectrum results from dust spots (“Donuts”) on the camera sensor.
Let's start with the slit plate of the spectrograph. A dust grain deposited on the slit becomes apparent across the spectrum as a horizontal, dark line perpendicular to the spectral lines. The more frequently the DADOS is attached to the telescope, the more dust can penetrate. And dust can also penetrate through the 1 1/4 " nosepiece of the slit viewer when changing the guide camera.
Cleaning is critical and not recommended because wiping on the slit plate would cause serious harm. Gently blow away dust from the slit plate with DRY compressed air, being careful not to let any liquid propellant from the compressed air get onto the slit plate. Never blow on the slit plate, and close the DADOS after use carefully with the dust covers.
Dust on the slit does not dramatically affect the relative spectral comparability within the spectrum because it affects all colors almost equally, thus contributing to the pseudo-continuum and getting eliminated by normalization. If the star is covered by a dust grain, the total intensity is slightly reduced. Usually a star appears enlarged on the slit due to seeing, and the intensity is later summed up in y-direction to cover the whole light perpendicular to the slit.
Astrophotographers and spectroscopists fight with dust, especially with dust on the camera sensor, which is visible as dark spots or rings (donuts). The correct full calibration requires the application of a flatfield image on the stellar spectrum. However, a special calibration unit consisting of calibration lamp and flatfield lamp is not yet available in our range of accessories. If you already do astrophotography and own a flatfield unit, you may use it for calibrating spectra, too.
When you begin taking stellar spectra, however, you should start without any flatfields. Please take some time to get to know your spectrograph. Taking a good flatfield requires some experience in astrophotography. For example, it is not possible to obtain a uniformly illuminated flatfield over the visible spectral range (400nm-700nm), even by applying filters. Most flatlamps emit much less light in the blue spectral range than in the red. A less exposed flatfield reduces the signal-to-noise ratio in the division result and can thus adversely affect the quality of the spectrum. In a narrow spectral range (for example Hα), however, you will succeed with a well-exposed flat.