s
normalized to mv = 20th mag observed at zenith
For a number of years, the Burrell Schmidt has offered CCD imaging at the Newtonian focus and traditional photographic imaging, with the plate holder located inside the telescope tube. The high demand for CCD imaging with this telescope, coupled with the declining photographic requests, resulted in a decision to offer only the CCD as a detector, both during KPNO and CWRU time (see NOAO Newsletter No. 41).
Consequently, the opportunity arose to remove the hardware inside the tube that accepts the plate holder. It was apparent that this hardware presented significant obstruction. This obstruction undoubtedly contributed to the fact that the Schmidt's CCD count rate was below that expected based on the aperture relative to our other telescopes and the S2KA CCD's performance relative to that of our other CCDs. It also seemed likely that the extra material in and near the beam contributed scattered/stray light.
The Schmidt was shut down during the March bright run to remove the photographic plate holder and blacken exposed metal surfaces. As a secondary goal, the lead screw in the CCD focus mechanism was replaced in the hopes that this would reduce the sky-position-dependent focus variations. The mirror was also realuminized. Because the collimation has always been done using the plateholder for reference, this also provided the opportunity to use an improved laser collimation procedure, devised and carried out by Chuck Claver.
The count rates displayed below show the improvement in throughput:
e s normalized to mv = 20th mag observed at zenith
late January 3 March improvement U 0.54 0.86 1.59 V 1.70 2.71 1.59
Thus, we have gained half a magnitude by removal of the plate-holder and associated hardware! It is true that in the intervening time, the primary was realuminized, and the corrector and dewar windows cleaned. However, the similarity of the gain at U and V suggest that the realuminization was a minor component. (Thanks to Nigel Sharp and Mike Corbin for obtaining the exposure of the spectrophotometric standard.)
The best images obtained were considerably undersampled (1.35 pixels FWHM = 2.7" ), as good as we have ever seen. This indicates that the new collimation procedure has worked well. Unfortunately, the good image quality reveals a focus gradient previously not apparent. The gradient is across a diagonal of the chip, consistent with a simple tilt. Thus we achieve 1.3 pixels in the center, but in two of the corners of the chip, the images degrade. The images are < 2.0 pixels FWHM over nearly the entire chip.
The new lead screw appears to work well, and there is no longer any change in the displayed focus values as one moves the telescope around the sky. However, there is still a noticeable shift in focus as a function of position in the sky (+250 focus units going from +70 declination [0 HA] to 0 declination [0 HA]). This indicates that some other component is likely responsible. We have documented the magnitude of the effect (and the sign!) and expect that observers can compensate for this now that the focus readout is reliable. The direct imaging manual has been updated to include all the above information. One final note: the Schmidt now has a second monitor for the data taking Sun, Loden. This enables one to easily juggle the multiple windows needed for data acquisition, display, and reduction.
Taft Armandroff, Phil Massey, Bill Schoening