The etalon effect in the ITMs is presently not controlled: this induces an uncontrolled Finesse asymmetry and, in turn, a higher coupling of common noise at the differential port.

In the past, the mirrors temperature was varied by heating up the whole tower (#27259). However, at the moment the heating belts are no longer installed, and we tried to increase the mirror temperature by using the Ring Heaters as actuators. The backdraw of this method consists of the thermal lensing induced in the substrate by the heating, but on the other hand it's much more efficient than heating the whole tower.

Yesterday we made a first test to better study the effect of RH on the cavity power. 

The first step of the experiment was devoted to the search for a good working point: in order to do this, we wanted to scan a complete etalon fringe on one arm in order to find a point which could minimize the contrast defect. To do that, we left the interferometer in LN1, so to be able to see on B1p the effect of the contrast defect. We chose to actuate on the WI RH, in such a way we could be able to compensate for the thermal lens in the substrate with the WI Central Heating (which is currently used in "alignment configuration" on the North TCS bench). 

Following the hints given by the simulations (VIR-0631A-19), at 10.10 UTC WI RH was switched ON, the applied power was 0.98 W, which corresponds to a power absorbed by the mirror of 0.5 W; after 1.5h also the WI CH was switched ON with 220mW of power. Since B1p was getting worse as well as the sidebands, after half an hour the CH power was further increased to 360mW. However, after a while the ITF unlocked.

At this point (12:30 UTC) we halved the power on the RH (applied 0.4W corresponding to 0.24W absorbed by the mirror) and went to LN3, and reduced the power on the WI CH to 260mW. In this way we could monitor more parameters, such as the mirror lines behavior, the optical gain, and various noises in the sensitivity curve. Since we were close to the end of the shift, at  12:52 UTC we switched off the RH, and after a while also the CH.

In figure 1 we show the whole period of the measurement. What is evident is that the sidebands are strongly affected, as they go down almost immediately. B1p becomes brighter, which can be related also to a worsening of the alignment error signals/setpoints related, in turn, to the sidebands decrease. 

After the RH was switched off, there seem to be a point where the Optical Gain reaches the maximum (at around 14.00 UTC), which also corresponds to the maximum of the frequency excursion of the mirror lines. 

In figure 2, the West arm cavity power normalized to the PRC power is plotted against the frequency of one of the mirror lines calbrated in K (no offset added, that's why it appears to be negative). The temperature variation as seen by the line is about 0.03K, while. However, the mirror didn't reach yet the thermalization, and this could explain why the temperature variation is so small. On the other hand, the power variation in the cavity is of the order of 1%, which seem to suggest that some other effect is ongoing.

In figure 3, the time series of the normalized power for both cavities is shown. A variation is visible only on the West side, while the power in the North cavity stays the same (as expected). 

More investigations and analysis are ongoing.