For the past year, we have been obtaining seeing measurements at the Mayall 4-m telescope with a camera mounted on the south port of the guider. Several times during a night, the telescope operators use this system to measure the "delivered image quality" (DIQ). In order to make comparisons to the nightly measurements at WIYN, we take these data through an "R" filter, and have demonstrated that the measured FWHM agrees with what one obtains with a 10s exposure with a direct CCD. Along with the seeing measurements, we automatically record the temperatures at various points around the dome.
During April 1996 to April 1997 we obtained 308 measurements on 180 nights; the median DIQ is 1.08", with no difference between the f/8 and f/15 secondaries. The results are shown in Figure 1. (Measurements at prime are somewhat worse due to degradation from the doublet corrector used with the CCD; as discussed elsewhere in this Newsletter, an improved corrector, along with an atmospheric dispersion compensator, has just replaced the old system.)
What causes our nights of poor seeing? Examination of the temperature data indicates that whenever the wrong things are warmer, the seeing is bad: physics happens! And, the converse is also true: that on nearly every night of poor seeing, the thermal environment is highly non-isothermal. As the three accompanying plots show, (a) whenever the primary mirror is warmer than the surrounding dome air, the delivered image quality degrades quickly---although a somewhat cooler mirror is OK (Figure 2). Presumably this is due to convection cells forming directly above the mirror. (b) If the dome air is warmer than the outside ambient air, then convection cells occur as warm air leaves the dome and cool air falls (Figure 3). This is even better demonstrated by the fact that (c) whenever the air temperature in the upper part of the dome is significantly cooler than in the lower part of the dome, the seeing is bad (Figure 4).
So, what can we do about this? We have taken the following steps: (1) If the primary is warmer than the previous night's low temperature, cold air is blown onto and around the primary during the day when the telescope is not being used and when the incoming air temperature is at least a couple degrees above the dew point. We are currently in the process of tweaking the system to improve the amount and duration of the cooling. (2) We have installed a powerful fan that runs during the day to mix the air in the dome. Since the floor is cooled, the effect is to take the cold air at the bottom of the dome and mix it throughout the structure. (3) We are designing large dome ventilation doors, the installation of which will start this summer and continue into the fall. The goal is to achieve 30-40 flushes of air per hour in modest (10mph) wind.
Phil Massey, Bruce Bohannan, Chuck Claver, George Jacoby, Richard Wolff