The structure mounted on the hub of the secondary mirror is supporting a retroreflector which was used for all the tests made over the last week to recalibrate the AdSec: it was for the Thin Shell #3 (TS3) the first time it would be checked and characterized as a mirror for an active secondary.
The tests and calibration were made with a 4D laser interferometer. The series of three pictures below shows the evolution from the initial pupil (on the left) to what was achieved with 100 modes (on the right).
It became cleat that the mouse hole was likely going to be a problem. The area around it requires too high forces to be flattened. A close-up on the central hole and its mouse hole is seen below.
A first pass at 600 modes while masking a central area slightly larger than the central obscuration of the telescope produced a mirror flattened as shown below.
The feature at the top-left external edge is due to a sequence of three clustered not working (therefore disabled) actuators. The previously known circular defect (groove) close to the external edge is also evident. The groove is about 5mm thick (FWHM) and about 100nm deep (WF).
More tests were made and a lot of data were taken to prepare the simulation and preparatory work needed to use TS3 for AO corrections. A refined flattening scheme was devised which ended up with a flattened mirror shown below. It has a wave front error of 29nm rms, matching what was found for TS4 (the shell TS3 is replacing) and TS5 (the shell on the left side of the telescope).
Ad-Sec-Dx can now be used as a non-adaptive secondary mirror for seeing limited operation. The retroreflector holder has been dismounted and the mirror will be used for the coming nights of engineering where the telescope and the instruments will be brought back to life and checked out after nearly two months of shutdown.
Meanwhile, our colleagues in Arcetri will work on the AO software to slave the actuators which will be behind the masked area. The next, and final, step will be the calibration of the unit in closed loop for AO corrections. Work on the telescope sould start on Sep 13.
It is anticipated that TS3, which is only a spare shell, will not work as well as TS4. However, it should still allow excellent AO observations on the right side of the telescope while we wait for ts4 to be refurbished (not an easy thing to do...).
Another excellent team work with participants from Arcetri (on Mt Graham or remotely from Italy) and LBTO!