[Halld-pid-upgrade] [Revised Logentry] PMT installation and cookies

[mpatsyuk at jlab.org]

Logentry Text:
--
Today we were working on the bubbles. The task was to improve the optical coupling on row 17 (second from the bottom) for two adjacent PMTs. The following picture shows the view on the problematic row in the 3 segmented mirror through bar box opening [figure:1]

The plan was:

1. Make sure all the screws are tighten (if some are not, tighten them, which may improve the coupling)
2. massage the module by loosening the screws on the bracket and wiggling the module
3. see what happens to the coupling
4. maybe we need to reinstall the bracket 17

What environment we had:

The optical box is already attached to the bar box, the photodetection plane is vertical, and we did not yet try detaching a bracket in vertical orientation. [figure:6]

 Also, it is hard to see the quality of coupling: it is dark inside the box, and many reflective surfaces make it difficult to identify the bubbles. 

We started with inserting a periscope (a small camera with a flashlight on a stick, the picture from the camera is sent to a smartphone via wi-fi)
[figure:7]

 through the port on the top of the optical box [figure:8]

 to be able to see the PMT coupling from inside. The camera can be located in the gap where the LED light sources are: between the flat mirror parallel to the radiator side and the three segmented mirror. The camera light was too weak to allow seeing the quality of the coupling, and we managed to add some light by shining a flashlight on the bottom of the window (from outside of the optical box) [figure:3]

Even with additional light it was very difficult to identify problematic coupling.

We tried wrapping the LED strip which was used for the bar box transportation (Greg's LED, which helped to view the interior of the bar boxes) around the camera stick. It helped a lot and the PMT faces became much more visible [figure:2]


With the camera light it is even lighter: [figure:4]

But nevertheless, the camera does not allow seeing the imperfections/make photos of them. We "scanned" the surface with the camera to be able to see the reflected light off different spots on the same PMT and off different PMTs to compare the known problematic spot to the surfaces, where we knew the coupling was good. And this method was not very convincing - the differences were hard to see and identify.

Eventually we decided that the best way to inspect the row 17 was to look with a human eye into the gap on the bottom of the window leaving the LED strip on inside the box. [figure:9]

 In such a way one could see the coupling in details in 3 segmented mirror, but still it was impossible to make clear photos of the view. The reflections were spoiling the picture. By moving the head left-right-top-bottom and looking on the inspected area with different angles, one could get an idea about how the imperfection look. A picture would not show it, since it is only a one view.

We tighten the screws on the brackets. Actually, the screws on the row 17 had some freedom to tighten, which helped to slightly improve coupling. The two adjacent PMTs still had a large area of poor coupling as shown on the first picture on this page. To massage the module we loosened the four screws around the right nut moving the right module to/from the window: [figure:5]

After that we wiggled around the cases, which holds the metal bar with the large nut on top. We did that and tighten the screws back. After the procedure the bubble got shifted completely to the edge PMT on the module (one of the two affected), but the total area of the bubble stayed approximately the same. The bubble occupies the upper ~75% of the PMT (see first pic on this page for orientation). We tried a couple more times, but it did not help much: the best we could get is the 75% of the edge PMT is poorly coupled, and the other two PMTs have good coupling.

We are not sure about the reasons for the poor coupling.

After about 1 week since we applied all 18 brackets, the greasing oil does not drip out, which was my concern. The following photos show that there are only about several drops of oil on the window leaking from the bottom modules:

[figure:10]

[figure:11]

Conclusions:

1. The inspection of the PMT coupling is easier to do while the PMTs are being installed - by looking into the window.
2. It is important to make sure all screws are tight.
3. The LED strip adds enough light inside the optical box to be able to inspect the coupling. Several angles/views are needed for a clear idea on the bubble shape. By moving the camera around/changing the angle of light hitting the inspected area one can get a better impression of the coupling quality.
4. We did not dare to detach the module and decided to leave "one controlled bubble", because it is in the less relevant area - further from the beam, where the tails of the rings are. It is only one of 90 PMTs, which assuming a non-uniform occupancy, impacts the photon yield on the level of 1 % or less. Also, having a known area of poor coupling one could directly observe the difference in the occupancy.
5. Before the installation of the PMTs to the second box it might be useful to test how robust is the optical coupling against massaging and wiggling of the module described here. Preferably, with the adjacent brackets attached to the test window as well.

---

This is a plain text email for clients that cannot display HTML.  The full logentry can be found online at https://logbooks.jlab.org/entry/3643199
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.jlab.org/pipermail/halld-pid-upgrade/attachments/20190118/0d8ffb0d/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: application-pdf.png
Type: image/png
Size: 346 bytes
Desc: not available
URL: <https://mailman.jlab.org/pipermail/halld-pid-upgrade/attachments/20190118/0d8ffb0d/attachment.png>