First fiber boundle went broken by the crew when routing it to the LGSW rack, so a second one was installed:

On 11/20/2014 2:47 PM, Jim Wiese wrote:
Gustavo,

The box containing the spare fiber with the break in the end
connector is back in the box it came in and placed in the ARGOS conex
container in the high bay.

In conjunction with installing the air line for the Luci Air Knife,
I took pictures of the current install of the AGROS fiber bundle in
the RFBG Cable Wrap. The connectors at either end have NOT been
connected per your request. I have however left the strain protectors
used for pulling this set of fibers attached to the ends, and the ends
safely tied to the inner drum and cable tray to prevent any undesired
flopping around.

I also have additional pictures I took with my camera of both the
install as well as a few stages showing how the delaminated fiber
connector was taped and prepped for safely storing it in it's current
condition. If whomever first examines this connector carefully unwraps
the protective layers it can be safely opened and examined and still
preserve a final layer where the barrels and pins and fibers are
attached to a padded backing strip. I want to reiterate that the fiber
itself is NOT BROKEN only a separation between the pin plug end, where
it transitions to the larger diameter portion that continued into the
cable outer wrapping.

It might be possible for the vendor to look at this and re do the
epoxy (it looks like) connection portion they used to create the joint
in the assembly.


I will prepare and send additional pictures tomorrow when I am at
Tucson. I am packing up to leave the mountain as we speak.


I wish this had not occurred, especially on my watch. I do hope
everyone knows how careful we have been in handling the fiber bundle
as you have also observed.


With regards,

Jim Wiese

Also the second fiber boundle went broken on 25 Nov when Marco B. and Gilles tried to route it inside the LGSW rack:

Michael J. Lefebvre via arcetri.astro.it 
25 Nov

to Gustavo, Jim, Jesper, Sebastian, Julian, Marco, Gilles, Wolfgang, Dave 
Dear ARGOS Team and other interested parties:

Today an "attempt" was made to complete the installation of the ARGOS tip-tilt fiber. This installation was attempted by Marco, Gilles and myself. First, we carefully unwrapped Jim Wiese's protective layers covering the fiber ends and successfully installed the 4 fibers onto the AGW WFS patch panel. Next, the fiber bundle was routed into the LGS WFS rack and again, Jim's protective layers covering the fiber ends were carefully removed. Lastly, the 4 fibers were attached to the corresponding APD ports in the LGS WFS rack. During installation of the final fiber, the ferrule broke (as before). I have attached a photograph of the failure below.We believe this failure is intrinsic to the manufacturing method utilized for final assembly of the fiber connectors. The occurrence of two bond line failures during normal (actually gentle) handling of the fibers indicates a fundamental flaw in the manufacturer's assembly method. 

There are several factors that may have contributed to these failures including:
 inappropriate epoxy selection - must have a sufficiently large lap shear strength for this application
inappropriate temperature range - chamber is currently at -3C
contamination of the bond line, parts not properly cleaned prior to assembly.
if a two-part epoxy was used, improper mixing or constituent ratio
other?

There is a slight chance that we can connect the final fiber and provide some strain relief of the armored jacket. If successful, we will be able to test the tip-tilt fiber as planned. In the long term, this fiber should (must) be removed and sent back to the manufacturer for repair. I recommend that the manufacturer perform a root cause failure analysis and provide a detailed (and reviewed) failure mitigation plan before attempting to reinstall these fibers. I assert that all the terminations will eventually suffer this failure if not properly mitigated. Additionally, the location of the patch panel on the AGW is such that it adds some additional strain of the fibers during installation. Moving the patch panel approximately 5 cm clockwise along the AGW circumference will reduce the strain on the fibers.

Your comments and advice are now required before we attempt to attach the broken fiber. By requesting this action you accept the risk that the actual fiber may be broken in the process. This will be a "best effort" attempt to salvage the fiber but in no means constitute the final integration.

I sincerely apologize for any inconvenience this may cause. The "ball is now in the manufacturer's court"; who will be the POC to the vendor? How soon can we return the broken fibers? How quickly can the vendor commence a root cause failure analysis? This needs to be addressed ASAP. Please provide your timely inputs to these questions.

Thank you!

Michael L. is actually in contact with Jesper S. to solve this issue: see email exchange from 26 Nov 14. The first boundle will be shipped back to Europe by Micheal to be refurbished.

• ARGOS Tip-Tilt Root Cause 1.pptx: ARGOS Tip-Tilt Root Cause 1.pptx