Participants: Matteo B., Yuhang, Eleonora
In entry # 876 we reported a computation about the difference between the reflectivity of the input and output MC mirrors needed to explain the low transmission. The results in that is has to be of the order of 0.5%
In order to see if the low transmissivity of the green MC is due to such a difference, we have changed the input and output mirrors, hoping to find a combination with smaller difference.
The mirrors installed belong to a batch of 6 mirrors with nominal reflectivity 99.2%. (see pic in entry 850). Among them 2 have been used in the mode cleaner, other 2 have been used to extract pick off of the MC reflected and transmitted beam. One was still in the box and the last one I don't know.
Here the different combination we tired:
| in | out | tra pickoff | spare | transmissivity | |
| initial | 1 | 2 | 3 | 4 | 44% |
| 12/07 | 3 | 2 | 1 | 4 | 45% |
| 17/07 | 3 | 1 | 2 | 4 | 35% |
| 17/07 | 3 | 4 | 2 | 1 | 56% |
The best results have been found with the last configuration
INPUT POWER : 23 mW
TRASMITTED: 13 mW
REFLECTED: 7 mW
We added a lamba/ 2 to improve the polarization and we observed that we can get rid of the broad mode observed in the spectrum.
The alignment was good but can be further improved.
Note that between the second and third configuration I have reduce the modulation depth sent to the 15.2 MHz EOM of a factor 2 (selecting 1/2 scale in the windows 'channel control' of the DDS software). As expected it improves the transmissivity of few percents. The lock of the SHG doesn't seem affected by this change.