Today we have continued the investigations about the mode-cleaner transmission. 

First of all we have tried to carefully align the cavity. To do that we have blocked one of the beams of the MZ and turned the polarization of the incoming beam to have the "p" polarization (lower finesse). In this way the peaks are broader and more visible. In the final alignment, the power in the 01/10 modes was of the order of a few % (see figures 1 and 2). 

Note that, since in a 3-mirrors triangluar cavity the 01 and 10 are at different resonance frequency, we can align the vertical and horizonal directions and check immediately the effect on the peaks. 

We have also measured the visibility of the cavity using the reflected DC photodiode. Note that when the the two beams of the MZ are present, the photiode saturates and then we put an optical density OD1 in front of the photodiode. 

Measurements with the cavity locked

We have made three measurements. All the powers are in mW 

1) MZ blocked - "p" polarization  

trans = 6, in = 7.6, ref = 1.2, end 0.03 

--> transmitted power = 79%, reflected power = 16%, Losses =  5%

2) MZ blocked - "s" polarization

 trans = 4.5, in = 7.6, ref = 1.5, end 0.26

--> transmitted power = 59%, reflected power = 20%, Losses =  21%

3) MZ unblocked - "s" polarizaton 

trans = 11.7, in = 18, ref = 4.2, end 0.6

--> transmitted power = 65%, reflected power = 23%, Losses =  12%

Measurement with the ramp

In the three cases the cavity visibility (1-R_res/R_nores) ~ 20%, which is in rough agreement with the measurements with the cavity locked. 

Conclusions 

1) the measurement of the cavity visibility ~ 20% (and the reflected power at resonance) seems to exclude a problem of asymmetry in the in/out mirror reflectivities. 

2) a transmission of the order of 80% for "p" polarization and 60% for "s" polarization is possible. The 80% transmission for "p" polarization is compatible with the lower finesse of the "p" polarization. 

3) we still have some non understood losses. More investigations are needed.

Figure 1 and 2 shows the alignment conditions with "p" polarization (we see that the high-order modes are a few %). 

Figure 3 shows the cavity visibility