Cal Bench Alignment
Table of Contents
Verify Octagon Alignment and Filter Wedge Alignment
The first order of business is to verify that the various action light sources are all coming out at the same place. The best way to do this is to choose one arm to use (simulcal is recommended, see below) and block the other arms with index card at a location where the beam is large. On the chosen arm, place an index card right in front of a fiber, so that the image of the light source is near focus. With the brighter light sources (e.g. flat field lamp, aka “Broadband Fiber”) the spot should be visible by eye with the room lights off (make sure the ND filters are at a minimum OD setting).
Use the cal bench FVC to image the nearly focussed spot. Iterate through the octagon sources and verify that they all produce a spot at the same location.
Choose one light source and then iterate through the various filter positions on each wheel (this is why simulcal is a good choice for the arm on which to make this measurement). Verify that the spot does not wander with different filter settings.
Adjusting the Octagon Alignment
If one or more of the octagon source need alignment, the individual stages can be adjusted.
Here is a link to the stage itself: https://siskiyou.com/ixf2-0ta/?searchid=38075&search_query=66760000E
And some advice from Stave Gibson:
They feature two flexures at right angles to each other (one 'bend' is visible in your picture). The adjustment bolts for both pitch and yaw are clustered together. Each axis has a pair of bolts centered on it's gap; looks like the larger bolts do the adjusting, and the smaller bolts do the locking, as shown here. Don't touch the middle two bolts as I think they are set at the factory.
However, I wouldn't touch these unless as a last resort. This echoes what Sam Halverson mentioned earlier (to start at the fiber end). Adjusting here will likely open a can of worms that you can't get closed again (I say that from experience
Science and Sky Fiber Alignment
Next we need to align the science and sky fibers such that the light from the Octagon sources lands on the fiber tip. Use cards to block the other two arms (at locations where the beam is wide and not near focus). You can image the reflection of the light off the fiber tip and verify your understanding by putting a card in the beam near the fiber tip (near focus). The image of the spot on the card will be out of focus, but the position/shape won’t be influenced by the reflectivity of what is underneath while the image of the reflection off the fiber tip will be more detailed. Both are useful.
Compare that position to an image of the back illuminated fiber (using the built in LEDs controlled via kpfpower controlled outlets). Turn off the octagon light sources, send the octagon to home, and put at least 2 layers of card in front of the octagon opening for this step. There is alight leak from the octagon, even when it is at the home position, so blocking that light using cards reduced confusion in the FVC images.
Sky fiber alignment example.
The video above is a screen capture of images from the above process. This shows a verification of multiple octagon sources and comparing those positions with the sky fiber.
The above video is similar, but looking at the sci fiber.
Adjusting the Science or Sky Fiber Stages
If alignment is needed, use the 3 axis stages holding the science and sky fibers to bring them in to alignment. The stages appear to be Thor Labs part MBT616.
Simulcal Fiber Alignment
The simulcal fiber in the most difficult because there is no built in back illumination. Fundamentally, use the same process as the science and sky arms (block the other arms): image the spot from the octagon and compare that the to the fiber location as determined by back illumination.
In practice, I had to bring the system in to rough alignment by eye. I used the broadband fiber in the octagon (with minimum ND) and I was able to visually see the illuminated spot on the simulcal fiber. It was useful to have a card to insert in to the beam to verify. I was able to walk that on to the fiber tip by eye using the fold mirror adjustments and the simulcal fiber stage adjustments. I then used the FVC to verify alignment as per the video below, but I made no further adjustments using the FVC images.
Back Illuminating the SimulcalFiber
You can back illuminate the simulcal fiber at the scrambler. It is difficult because there is limited access. I was only able to get a very faint image to show up in 10-15 second exposures.
To back illuminate, remove the side panel of the scrambler (undo 2-3 screws and slide the panel down) and insert a card in front of the fiber. Then illuminate that card with as much light as possible.