John D. Monnier 2003Jun22 Experiment to optimize fiber focus on star ------------------------------------------- Conclusions: a) Telescope focus is our major impediment to faint star science. b) When switching between North/South FIX, one might refocus fibers c) This memo can serve as a data point of what the fiber explorer image "should" look like under optimized conditions. [WE SHOULD DO DOING BETTER!] ---------------------- Experiment: In order to maximize flux I tried to optimize fiber focus ON THE SKY. This is best since this allows each fiber to be focused at the best place for each (aberrated) telescope. The numbers below can be used as fiducial numbers when checking image quality using fiber explorer. Remember -- when fitting a Gaussian to the fiber image, make sure to have sampled the background by moving the fiber far from the peak; otherwise, the FWHM you get for the fit will not be accurate. The focus micrometer on each fiber mount was used below. I merely read off the least significant digit. I'm not sure what units these are -- would be interesting to know. Notice how the image gets asymmetric out of focus -- astigmatism. Also Note that these changes on the micrometer correspond to HUGE changes in image quality when looking at the fiber focus itself (backward illuminated) through an alignment telescope. Further proof that the telescopes focus is most significant factor in limiting coupling (not the focus behind the small Off-axis parabola). Date: UT2003Jun15 Configuration: FIX A N35 C10 S15 Readout Mode 1l7r Star: HD 126248 Micrometer Peak Flux FHWMs (Major/Minor) Reading ADU (ADU) Fiber FIX (Tel A): 0 15 530/470 -5 18 502/435 -10 (40) 20 473/417 -15 (35) 22 453/382 -20 (30) 20 460/380 -25 (25) 15 522/423 BEST: -17.5 22 429/376 ------------------------------------------- FIBER SD1 (Tel B): -15 (35) 15 1150/611 -10 (40) 19 680/451 0 23 644/500 10 24 625/550 15 24 586/550 25 21 700/674 BEST: 12.5 25 583/525 --------------------------------------------- FIBER SD2 (Tel C): -10 (40) 13 608/505 2.5 20 687/470 10 21 680/465 20 22 610/556 30 19 596/450 [This fiber in particular showed large focus SHIFTS (x,y) during this exercise. While the other two fibers showed optimal coupling when the astigmatism was minimized (i.e., the image was mostly circular), this was not true on this fiber. Coma?] BEST: 15 22 640/490 ** NOTE ADDED TO PROOF ** ON UT 2003Jun21, I changed Configuration from A FIX to B FIX. I found that the coupling in Fiber SD1 (NOW imaging TEL B) was very low, and it was not due to vignetting (star tracker counts OK). I found that the coupling almost tripled by shifting the focus from 12.5 (down) to -10(40) [The image quality of the fiber also noticably improved from FHWM > 1000 to ~500]. This suggests the TELESCOPE is out of focus, not the FIBER. I did not attempt to optimize this or the other fiber. CLEARLY we are not operating in a particularly efficient way regarding these coupling issues, and we should make telescope focus improvement a priority if we want to push sensitivity. ** NOTE NOTE ADDED TO PROOF ** When I switched back to config A FIX, I thought I would have to re-focus the fiber above. However, I found that the fiber did not need re-focusing. Not sure how to explain. ------------------------------------------------- John D. Monnier, Assistant Professor of Astronomy University of Michigan 941 Dennison Building 500 Church Street Ann Arbor, MI 48109-1090 monnier@umich.edu 734-763-5822 (FAX 734-763-6317)