One of the three outputs (more than 4 mW) of the splitter is injected in another 50/50 splitter, the first output is sent to the EIB phase camera (planned at the beginning of AdV but never used and dismantled), the second one is sent to EQB1 for the SQZ PLL. We found that the former output delivers about 1 mW, while the latter about 0.2 mW, meaning that this splitter is not working properly. We will make further checks on this second splitter. Anyway, if more power is needed for the SQZ PLL, we can remove it.
We made a systematic investigation of the optical losses occuring in the phase Cameras AOM box.
First of all we measured the insertion losses of the AOM as a function of the RF driver amplitude. With reference to figure 1 we first measured the power injected by optical fiber into the AOM box. The port “P1” input power results 52 mW while the AOM output port “P2” power is
| RF Power [dBm] | AOM output power [mW] | Insertion losses [dB] |
| 0 | 13.6 | 5.8 |
| 1 | 16 | 5.1 |
| 2 | 18.9 | 4.4 |
| 3 | 21.3 | 3.9 |
| 4 | 24 | 3.3 |
| 5 | 26.3 | 3 |
| 6 | 27.3 | 2.8 |
Notes:
- The AOM power is measured directly from the AOM fiber output after disconnecting it from the mating slave.
- RF power means the input power applied to the commercial driver of the AOM
- The RF power at which AOM has so far operated is 4 dBm
Conclusion: Considering that the manufacturer declares a typical insertion loss of 3.5-4 dB (not specific to which RF field amplitude) our AOM seems to work within the specifications.
Subsequently, in order to measure the losses associated with the mating slave, we connected the AOM output port “P2” and an external 1 meter long optical fiber first to the mating slave “MS1” then to "MS2" and finally on an external mating slave "MSExt". The power measured at the 1 m. long fiber output is reported in the following table:
| Mating Slave | Fiber output power [mW] | Insertion losses [dB] |
| MS1 | 17 | 1.5 |
| MS2 | 18 | 1.25 |
| MSExt | 18 | 1.25 |
The measurements are made with an RF power of 4 dBm so the insertion losses are calculated assuming 24 mW at the AOM output (see first table)
The losses associated with the mating slaves appear not too far from the value specified by the manufacturer (<1dB.)
Finally we have measured the value of the outputs of the 3 three branches of the 1x3 fiber splitter after connecting the AOM output to and the 1x3 splitter input to “MS1”. The power is measured at the output of the mating of the AOM box which are labelled respectively R (right) C (center) S (left):
| Box Output Port | Output power [mW] |
| MS_L | 5 |
| MS_C | 4.3 |
| MS_R | 4.2 |
The total insertion loss is 10*Log(17/(5+4.3+4.2))=2 dB which is slightly higher than what expected (order of 1 dB).
In conclusion, the losses associated with the AOM box are only slightly larger (order one dB) than the expected value.
However we have noticed that the connection to the mating slaves is very delicate. Just shaking the connectors a little the transmited power change significatly. The values reported are the maximum ones .
At the end of the tests we leave the system in the same configuration it was in initially.
Today I measured the RF output of the phase camera AOM driver.
To do this in the time window between 10:10 and 11:10 UTC the phase camera signal has been turned off.
Measurements made in EE-room where the driver is located
-AOM driver input: -6 dBm
-AOM driver output +20 dBm ( before injecting the signal into the spectrum analyzer the signal was attenuated by 10 dB. The reported value takes this attenuation into account. Moreover a 5 meter long RG58 cable was used to bring the AOM output signal to the analyzer)
Measurements made in Athrium where the AOM is located
-AOM input: +19.8 dB (in this case the 5 meter long cable is not used)
Conclusion: in our standard configuration (4 dBm input) about 30 dBm are applied to the AOM.