The Saga Of The aLIGO Fiber Puller At LHO (up ~9/25/2024)

Editor's Note:  In an attempt at documenting recent SUS Fiber Puller activities and share overall status of work of the Advanced LIGO Fiber Puller Lab located in the LSB building here at LHO, have had requests to post our notes/summaries (which in the last few years had been relegated to notes in logbooks, email chains, hearsay, etc.) in the LHO Logbook and tagging the ALOG section as "X1" with a Primary Task of "SUS".

Summary Of The Last 2-Years

(Alphabetical list of people who have helped/been involved:  C.Gray, J.Kissel R.McCarthy J.Oberling, T.Sadecki, R.Savage, TJ Shaffer, R.Short, B.Weaver)

Upgrade To 200W Laser (Fall 2022)

Until recently, the last batch of Pulled Fibers were pulled in Fall 2022.  At this time, the Fiber Puller started having issues related to the laser (f100 CO2 laser by Synrad; 100W and 10,600um)--this laser was well over a dozen years old and at the end of its operational life.  Higher power lasers had already been purchased for the Fiber Puller about 5-years ago in preparation for an upgrade and a possible need to pull thicker diameter silica fibers for heavier suspensions.  Because of this, it was decided to upgrade the Fiber Puller with one of these 200W lasers (f201 CO2 laser by Synrad; 200W and 10,600um).  The 200W laser is bigger than the previous laser, and because of this, the back of this laser hangs off the edge of the optics table, but Rick made a nice breadboard which offers a platform to support the back of this laser (see image1).  Cabinets and furniture were removed along the wall to allow extra space to for walking around this area of the lab.

After RyanS and I installed the new laser & re-did the chiller plumbing, we moved on to aligning the laser into the Fiber Puller. Since alignment work was next, we took this opportunity to swap in new upper & lower conical mirrors in the Fiber Puller (Travis ordered these a while ago.). 

Alignment Work Begins (2023-2024)

Then, with the help of Karl Toland's thesis (Chapter 2), we started Fiber Puller alignment.  This included Mechanical checks of the Fiber Puller which looked mostly OK; it was noticed there is a slight tilt forward of the tower.  We decided to not change this.  The periscope which points the beam into the Fiber Puller was reconfigured to allow its translation stages (namely the horizontal one) to have better adjustability for pointing into the Fiber Puller--previously the horizontal translation stage only moved forward and back which did not allow for horizontal translations of the beam into the Fiber Puller (this only required us to rotate the translation stage 90deg).

After this it was all optical alignment work, and this proved to be non-trivial.  Additionally, one of the gold mirrors was burned.  This mirror had a glass substrate, so it was replaced with a gold mirror with a copper substrate (also took this opportunity to replace several dirty input gold mirrors).

Then there was several months of no progress with alignment.  Eventually Travis & Jason checked out the status of alignment.  And in Aug 2024, Jason was able to give us a decent looking beam on our stock and after almost 2-years, fibers were finally pulled again!

Return To Fiber Pulling!  ...For A Little While.

Between Aug 2-20th, six fibers were pulled (4 passes & 2 fails).  At this point, while preparing to pull another fiber, one of the 1/8" ID tubing to the shutter burst on Aug 20, 2024.

NOTE:  Here I will paste in the email summary I sent to the group on Sept 26th.

---

Recovery From The Plumbing Failure On Aug 20, 2024

[start of email]

Summary:  

After the 1/8" chilling hose burst (8/20), weeks were spent with recovery and buying new plumbing hardware.  There is now better hosing for the shutter + beam bender (mirror mount) and most of the fittings have been replaced with push-to-connects (excepted for barbed fitting of the beam bender (see image 6)).  We have been running the chiller w/ new plumbing since 9/17/2024.

For the plumbing fix above, this was done in-situ without removing the shutter or mirror mount with the hopes of preserving the alignment, but first looks at alignment by myself and Jason at separate times did see an alignment change on our fiber stock target.  On Sept 25th, we were both able to work on the Fiber Puller, and Jason was able to get us a decent alignment back, we pulled our first fiber since the 8/20 chiller plumbing link---and this fiber also passed on 9/26!

PLUMBING (see image 5):

The backstory is years ago a water-cooled shutter & mirror mount were installed (1/8" ID plumbing using a mix of push-to-connect & barbed fittings) for the Fiber Puller.  The water-cooled shutter & mirror mount were added in series to the laser (which has 1/2"OD tubing) via an aluminum manifold.  During the 100W to 200W laser swap by RyanS and I, we discovered everything downstream of the laser (1/8" ID stuff) had been clogged (most likely for years) from the manifold downstream to the shutter + mirror mount.  This led to us replacing all lines, cleaning the manifold, and starting to use OptiShield with the chiller water with hopes of preventing any corrosion clogs.  This worked for a while during the upgrade and even through six NEW fibers pulled in August, but after the 6^th fiber there was a huge burst in the tubing at the shutter.  

This led to repair work, with a few issues/notes observed:  

1. Non-Water Tubing FAIL:  The line which failed (I reordered what was originally installed) was very bendy and easy to work with.  BUT over several purchase orders, it was noticed that this line was rated for "air" use.  This led to being more mindful of types of tubing.  Firstly making sure we had (1) water-rated tubing and also (2) tubing with decent max-pressure ratings.  Perhaps overkill, but ultimately went with tubing with max pressure rating of 120psi.  This new tubing was definitely much stiffer than the previous tubing.  
2. Metric & Inch:   Previously--between the manifold, shutter and beam-bender we had a mix of different fitting connections.  Additionally, because we were looking for: water-rated, high max-pressure rating, consistent fittings, etc, it limited options.  Ultimately, went with metric tubing + push-to-connect fittings (see image 4).  Able to make this change for everything, except for the mirror mount (it remained with its barbed fittings, but it has 3 barbs and it feels like a good connection).  Additionally, our pressures are pretty low (under 50psi)---not measured for the new 6mm lines, but the chiller's pressure dial reads ~25psi for the system.
3. 90deg Elbow Fittings Under Shutter (see images 2 & 3):  When the group checked out this plumbing work a few weeks ago, TJ suggested elbow fittings under the shutter (mainly because this is where the previous tubing burst and the lines do take a bit of bend under the shutter here).  New push-to-connect elbows were added.
4. Pressure/Flow values & Adding Sensors:  Had not really paid attention to what the pressure & flow of the system are through all this work, but after the failure it was brought up by Travis and Jason. We have a Flow sensor on the supply line (downriver from the laser) and a pressure gauge (also on supply side at the chiller).  Just for a rough idea of numbers, here are values observed:
   1. Flow sensor:  this ran at 1.0gpm with just the laser, but after the new 6mmOD system was added it's increased to 2.0gpm
   2. Pressure gauge:  was running around 30psi at some points during the rework (more consistently at 26psi after the new 6mmOD system added)
   3. TO DO list:  More Sensors!  Would be nice to add more flow/pressure sensors at more points in the plumbing system--especially for the 6mmOD lines.  With focus on getting operation back, I did not get a chance to look into ordering this, but we should do this at some point.

ALIGNMENT Once Again:

For all of the plumbing work above, I opted to do it "in situ" by not "touching" the shutter or beam bender (I did not remove them from the beam path and only connected/disconnected their tubing)---this was all in the hopes of not changing the alignment---we had a nice alignment which produced decent fibers in August! 

Although I attempted to be careful with all of the plumbing work above, first checks by both myself and Jason later showed we were off a little when we looked at the fiber stock target with the red beam.  

[On Sept 25, 2024], Jason and I were both able to work together in the fiber lab.  Here are some of the notes I took while Jason did his alignment magic!

• Coalignment of shutter's red beam with the main CO2 beam:  It looked decent via a check of the co-alignment-beam-check path (via flipper mirror) that Jason and RyanS added earlier this summer.  We made an ever-so-slightly small change via the shutter red-beam adjustment screws.

Alignment Attempt #1:

Red beam alignment

Jason went through alignment touch-up via the periscope mirrors.  He was quickly able to get a decent beam on the fiber stock target.  At this point, he checked to see how this beam looked with the alignment irises which have been in place all the while (between the periscope and Fiber Puller).  After aligning beam to the irises, he once again had a decent beam on the fiber stock.  In both cases, there was a "hot spot" noted on the "back" side of the fiber stock.  

CO2 beam alignment

With the main CO2 beam on a new fiber stock, we needed to turn the power up to about 55-60% on the laser to be able to see the beam via the cameras in the Labview app.  With the cameras, I tweaked the (periscope mirrors) to get a better alignment (which allowed us to lower the power to about 50%).  Jason had a try at tweaking alignment via the cameras (I believe this was his first chance getting to do this—especially since the side camera was better-focused on our stock back in August).  

We were able to optimize a little more, but it still looked not great (see image 7).  I thought it might have looked like what we had when we were pulling fibers in August, but we could clearly see it wasn't great.  There was an angle on the fiber stock and the rear camera could clearly see the angle even more.  I mentioned we could probably pull a fiber, but we'd definitely have to increase the laser power.  We did not pull a fiber however.  We broke for lunch with hopes at more checks in the afternoon.

Alignment Attempt #2 With Mechanical Offset Check Between Top & Bottom Clamp Translation Stages:

Translation Stages:

Thinking there was still something amiss mechanically, Jason wanted to check the Top & Bottom Clamp Translation Stages.  We do this using a special tool to set the Upper translation stage and the Lower translation stage (the latter we have not touched for a while).

Jason went through several iterations of making small adjustments to both the upper and lower translation stages (see image 8).  On one of the final adjustments, he intentionally overshot the upper translation stages.  This was to hopefully help us later (it did!).

Tower:

NOTE:  At this point, Jason continued with mechanical checks and we looked at the Tower with a bubble level.  There is an observable tilt forward of the tower albeit small.  (this is mentioned above)  Karl's thesis mentions this should be fixed, but there are no tolerances given or go/no go values.  And it's not clear how to do this without a major change to the system.  There are alot of surface-to-surface mechanical contacts in the tower, so one would need to loosen lots of bolts and then possibly install shims and retorque down everything.  Anyway---a big job.  Since it is a small tilt, we are just noting it and decided to move on.  In the future, the Fiber Puller will need to be upgraded to a taller tower, this will definitely give us the opportunity to remove the tilt from the tower (note:  This could be a reason for [the known years-long issue of ] why we always have to make adjustments to the upper clamp translation stage almost every time we pull a fiber).

Red beam alignment

Once again, Jason very quickly went through an alignment onto our fiber stock.  One nice thing:  no "hot spot" on back side of stock!

CO2 beam alignment

A new stock was installed, and we heated it.  We could clearly see the beam with the cameras at a lower power of 45%!  (lower power than Attempt #1!   Our 1st good sign!)

And the 2nd good signs was the forward & back alignment of the stock was GOOD (in other words, there was not an offset in this direction!  Another sign the translation stage mechanical checks did us good!).  The stock did show an offset in the other direction ("left/right"), so the translation stage was adjusted accordingly.  

Additionally, the spot on the stock looked MUCH better on the cameras (3rd good sign! [see image 9]).  There were minimal angles seen with the "ring spot" on the stock!  Touched up the alignment a little more, but it was already better than how we looked in the morning.  We were ready to pull!

Vaporization

Before pulling, wanted to get an idea of what laser power we could go up to before silica vaporization begins.  Ryan can correct me here, but I believe Alan told us, we should do pulls at laser powers just lower than the laser power when silica vaporization occurs.  The laser power was slowly ramped up, and I started seeing vaporization somewhere around 75-80% (it was very bright on the stock).

Polish & Pull

Not sure of what is the ideal values to use here, but to just do it, went with:

• Polish power = 55.5%
• Pull power = 65.0%

Although late in the afternoon, we went through the whole process.  Fiber was profiled and then analyzed:

• PASSED!
• It had a bit of a low fundamental = 493.892 Hz

This is where we ended for that day.

Have continued to pull more fibers and get more data points with an eye on the fundamental (violin) mode frequencies.  One knob we have here are those Polish & Pull powers. 

NOTE:  A reminder that even with the 100W laser toward the end, our fibers have been populating the lower range of violin mode frequencies (see fiber table).  Something to think about.

Additional Note On Alignment & Power

With our nice alignment, it should also be noted we are still running at higher powers than what we are used to (which was at 85% or ~85W with the 100W).  With the current 200W laser, we were running at ~140W in Aug for pulls and yesterday we were at ~135W.  This would point to a the overall alignment in need of some improvement.

[end of email]