Advanced_Laser_Alignment.md

title	layout	description
Advanced Laser Alignment	page	Aligning the light sheet from scratch

This page will walk you through aligning the light sheet from scratch. It is recommended that the detection axis be installed before putting the optics into place.

Preparation

This step will prepare you for a complete system realignment. This may take up to several hours, so be sure you have enough time to devote to the task.

1. If there are any refractive optics in the illumination path, remove them now. The illumination axis should only have:
   1. A 2-mirror beam walk in front of the laser,
   2. A corner mirror,
   3. and an optical slit (presently widened to maximum).
2. Complete a very-rough laser alignment: The beam emitted by the laser should strike the approximate center of each mirror in the path, ending at the center of the illumination objective.

Initial Alignment

During this step, we will align the beam with the optical axis and get its waist focused on the detection axis.

1. Walk the laser directly down the first segment of the optical axis.
   1. Insert two beam alignment disks at the beginning and end of the long, 300mm rail. Make sure there is as much distance between the two of them as possible.
      • Any alignment disk with a center hole will do; the parts list calls for frosted glass disks (DG05-1500-H1-MD), but you could also use LMR1AP on 1" mounts with the appropriate stilt, or even two irises adjusted very narrow.
   2. Use the first mirror to position the beam in the center of the first alignment disk. (Depending on how close the first disk is to the second mirror, adjusting the second mirror will have considerably less effect -- but not none!)
   3. Use the second mirror to position the beam in the center of the second alignment disk.
   4. Optimize both mirrors simultaneously to ensure the beam is running parallel at 50mm off the breadboard down the center of the illumination axis.
2. Align the second segment of the optical axis.
   1. Move your alignment disks to the beginning and end of the smaller, second-segment rail. One disk should be directly against the sample chamber, while the other should be very close to the corner mirror.
   2. Adjust the corner mirror to run the beam directly between the centers of both disks.
      1. If the corner mirror is on a kinematic mount: You may need to turn all three knobs a considerable amount to position the point of beam reflection onto both segments of the illumination path. Then adjust the horizontal and vertical knobs to orient the beam correctly. This may bear repetition, since adjusting these knobs does not rotate the mirror through the point at which the beam reflects!
      2. If the corner mirror is on a gimbal mount: You may need to loosen the rail carrier holding the mount and adjust it along the 45-degree rail so the beam bounces off the center of the mirror. Then adjust the horizontal and vertical knobs to orient the beam correctly. Although the horizontal position on the mirror should remain centered, it is possible that the vertical position will shift slightly; as a result, you may need to repeat this process.
3. View and adjust the beam.
   1. Make the beam visible to the camera.
      1. Fill the sample chamber with water, then activate the software and open the live view.
      2. Raise the laser power to approx. 15 mW, and the exposure to around 100 ms.
      3. If you cannot identify the beam in the live view, try adjusting the corner mirror's vertical component -- the motion may help your eyes pick out the borders of the beam.
      4. On the main Micro-Manager window, find a good contrast setting and make sure Autostretch option is unchecked.
   2. Add a 'reticle' to the live view.
      1. Using ImageJ/Fiji's Edit->Selection->Specify..., enter half of the image's pixel width and height in the X and Y coordinates and check Centered.
      2. Enter a reasonable width/height for the box to form a nicely-sized rectangle (try 500 x 100).
      3. If desired, use Image->Overlay-> Add Selection... to convert the ROI to an overlay. (However, the handles may be helpful to show half-points on the box -- and by extension, the entire image.)
   3. Use the corner mirror's horizontal adjustment to focus the beam in the center of the image.
      • Turning this knob will appear to move the beam waist (narrowest point) left and right.
      • Realigning this mirror now may lead to better results, but if the adjustment is not too significant, it is optional.
   4. Use the corner mirror's vertical adjustment to center the beam along the image's height.
      • Turning this knob will appear to move the entire beam up and down.
      • Again, consider realigning the second segment of the optical axis.

• At this point you should have a conic shape, with diffraction limited waist in the center of the view.
• If the beam seems to need an angle adjustment, it may indicate that the first segment of the axis is not properly aligned.

First Beam Expander and Intermediate Adjustment

During this step, we'll install the first refractive optics, then compensate for their changes to the beam path.

1. Install the beam expander.
   1. Realign the first segment of the optical axis -- the refractive optics will have changed the beam's endpoint slightly. Refer to the above instructions for details.
2. View and adjust the beam again.
   1. This time, the beam waist's horizontal location is decided by the distance between the two lenses.
      1. Adjust this distance until the beam waist is centered on the image.
      2. Then turn the corner mirror's horizontal knob until the waist is as sharp and bright as possible.
      • The distance between lenses affects how collimated the outgoing beam is. A perfectly collimated beam will be focused to the illumination objective's focal point, while a converging or diverging beam will focus before or after this point, respectively.
   2. The vertical adjustment is the same as above: use the corner mirror's vertical knob to center the beam waist vertically on the image.

• In this step and the next, to achieve the best possible alignment would theoretically require 'walking' the corner mirror whenever it would be adjusted, in particular if this mirror is on a kinematic mount: any adjustment moves the point-of-incidence of the beam, thus moving it off the optical axis.
  • In practice, these changes are often slight enough that this adjustment can be overlooked.

Second Beam Expander and Final Rough Pass

During this step, we'll put in the second two-lens system and again compensate for their influence on the beam.

1. Install the BFP lens system.
   1. Realign the second segment of the optical axis. Note that the position of the second lens may make it difficult to put in the far alignment disk; if so, be careful about adjusting the corner mirror, or you may lose the beam in the live view.
2. View and adjust the beam again.
   1. Restore the horizontal centering.
      • Again, the beam waist's horizontal location will be affected by the lens spacing. You may need to move both mirrors to keep the objective's back focal plane aligned with the second lens' forward focal plane.
   2. Once this is done, use the corner mirror's horizontal adjustment to optimize the beam waist.
   3. Restore the vertical centering (same as above).

• After this step, you can replace the cylindrical lens on the optical axis. Although it should be focused on the corner mirror, and the first lens of the second (BFP) system should have a focal point in the same place, this doesn't seem to be an absolute requirement -- the position of the cylindrical lens makes little apparent difference.
• Once the cylindrical lens is in place, the view through the live window will be mostly useless: you will be looking at a sheet of light, parallel to the camera's chip.
  • However, physical inspection of the sample chamber should show the sheet sharply narrowing near the center of the field of view, then expanding to as many as two or three centimeters on the far side of the chamber.

Light Sheet Refinement and Characterization

In this step, we'll observe the light sheet directly and finish adjusting it, then measure its width.

Important: At this point, make sure you have an appropriate combination of filters in the combined optical path. Directing laser light onto the camera's CCD is can permanently damage the camera.

• The infinity space should contain a filter blocking the laser's primary emission wavelength. A band-stop or long-pass filter at the appropriate wavelength will suffice.
• It may be necessary to remove any cleanup filters in the illumination path.
  • These filters remove all light except the laser's primary wavelength, which the later filter will block.
  • For viewing the light sheet directly, these secondary wavelengths are actually very useful: they are emitted at a lower power than the primary light, and may pass through the detection filters.

1. Turn off the laser.
   • This can be done by disabling its power supply, setting the laser power to 0, or 'closing' Micro-Manager's shutter (disable auto-shutter).
2. Set up the alignment sample.
   1. We assume herein that the alignment sample is a fragment of a light-microscope alignment slide, with a partial or complete grid, and an intact mirrored section spanning the camera's field of view.
   2. Insert the sample with the mirrored face facing the camera. Make it as parallel to the camera as possible.
   3. Using the software, move the stage to focus on the grid.
   4. Adjust the angle until the grid lines are equally out of focus across the camera's field of view.
      • It may help to focus on the grid in between angular adjustments: look for sections of grid lines growing unfocused as they progress to the left or right.
      • Optionally zero the rotational motor; this way, the following rotation is as simple as entering (-)45 in the stage controls window.
        • This function is not yet implemented in the OpenSPIM software, but may be performed using the Picard Twister control software. It is safe to do this while MM/OpenSPIM is running.
   5. Rotate the stage 45 degrees to reflect the laser light down the detection axis.
      • Be careful the rotation is in the correct direction. Watch the sample as you rotate it through 45 degrees and make sure the mirrored surface is now 45 degrees between the two objective lenses.
   6. Refocus on one of the grid lines. The grid to the left and right of center will be unfocused; this is expected as the grid's surface is no longer parallel to the focal plane.
3. Step the laser up in power slowly, until the light sheet is visible.
   • Try 0.1 mW steps, starting from 0.5 mW. OpenSPIM 1.0's 488nm Cube gives best results at around 0.91 mW. (The second decimal place may not display, but the Coherent Cube recognizes it correctly.)
4. Adjust the corner mirror's horizontal controls to center the sheet on the center of the image.
   • It may be useful to create a reticle as before, but with the height and width swapped to give a vertical rectangle.
5. Adjust the final lens in the sequence back and forth along its rail to focus the light sheet.
   • Keep an eye on the 'max' statistic in Micro-Manager. If you begin to saturate the camera, lower the laser power and/or reduce the exposure.
   • The first beam expander also contributes to the focus, and may require fine-tuning.
6. The light sheet diverges in either direction from the middle; the vertical slit must be narrowed to counteract this.
   1. Take a Z- or X- stack of the mirror surface and observe the beam's divergence at the edges of the field of view.
   2. Narrow the vertical slit until the light sheet is reasonably consistent across the camera's view.
      • As the slit narrows, the centered beam will widen and lose amplitude.

• You'll have to make a qualitative judgment about the consistency of the light-sheet vs. its width. A narrow slit will cause a wider light sheet, but it won't change as dramatically across the view of the camera.
• Keep in mind that if the light sheet is very wide at the edges of the view, this will illuminate out-of-focus parts of the sample.