Ushiba, Akutsu

Summary

Searched for a ghost beam coming back from the viewport window but not found, so did not install an additional beam dump in the MCE chamber. Also tweaked the beam path to the MCE gige camera so that the IMC transmission spot came to the "nominal" point on the screen.

Ghost beam search

After installing the super expensive viewport window (see 28971), letting the IMC flashing with P-pol 1 mW input, we searched for a ghost beam, which should be originated from the MCe transmission and reflected at the window's AR coat. But we could not find none of them (should be at least two). Note that even with a sensor card, the main MCe transmission beam itself could not be seen! With a beam viewer, we could see this main beam when it was directly incident on the viewer. With a SONY's de-IR-filter camera (Miyakawa-san's super camera), we could see the main beam when it was directly incident on the camera, but hard to find the relevant ghost beam. Interestingly, with the latter camera, the image caught stray light generated with the camera's lens or something. So we gave up to put an additional beam dump this time. These result would not be inconsistent with theory; see the below, calculation note.

Re-alignment of the beam path to the gige camera

After yesterday's flange sealing work, the transmission beam spot shown in the gige camera at MCE was shifted to the right edge of the screen. So we tried to make it nominal. But it was not so simple.

First of all, we checked the beam spot on MCe was mostly centered. So, what was moved would be the beam axis after exiting IMC until getting to the gige camera. So we tried to re-align this with tweaking a steering mirror just in front of the gige camera, but for some reason, we failed; It seems the similar issues reported in 17092... but we were not fully sure. Today, we just tweaked the camera's location, and somehow got back the beam spot location in the screen.

Calculation note

According to the wiki, MCi and MCo transmission would be ~6000ppm, while that of MCe ~4.7ppm. So the ratio of the transmitting power off MCe to that off MCo would be 4.7/6000 = 8e-4. Assuming the IMC is at critical coupling, the MCo transmission should be ~1mW, so that off MCe should be only ~0.8 uW. Moreover, the AR coat reflectivity of the super expensive viewport window (category D in this wiki) would be ~0.15% (see the relevant JGW-D1402927), so the ghost beam should be ~0.8nW or so. No surprising if unseen!