The experiment procedure to acquire CT projections is as follows:

1. Preparations: (a) Install camera with lens, (b) scintillator, (c) set needed field of view (FOV), (d) focus the system using a test pattern, (e) Use a water level to adjust the rotation plane to the turn table. (f) Measure the pixel size. It is needed to obtain the correct dimensions and attenuation coefficients.
2. Place the sample as close to the detector as possible.
3. Make sure that the entire sample fits in the width of the detector for all rotational directions.
   Typical test procedure: turn the sample in the intended arc 0 to 180 or 360 in steps of 45 or 90 degrees. Here, you can use short exposure times as we are only interested in visually observe the position.
4. Provide an open beam area outside the sample. This area shall never be shadowed by the sample for all projections. This area can be used for normalization of beam intensity if the beam is fluctuating.
5. Choose if you want a half turn scan or a full turn scan. The full turn has the advantage that the impact of the beam divergence is reduced. If you chose the full turn scan, you should also select an odd number of projections.
6. Select number of projections. The Nyquist theorem requires 1.5 times the sample width in pixels, but acceptable results can still be obtained 0.5 times the sample width. The cost of taking less projections is that you may start seeing aliasing artefacts in regions far away from the center of rotation.
7. Move the sample out from the FOV. The best way to remove the sample is to use the linear stages in your sample environment.
8. Choose exposure time, and avoid over-saturation of any pixel. The signal-to-noise ratio depends on the dose, i.e. longer exposures will increase the signal-to-noise ratio. Acquire between 5 and 10 flat field images (open beam). The number of projections also plays a role here - more is better.
9. Close the neutron shutters, and wait for few minutes to let scintillator after-glow fade, and acquire 5-10 dark current images.
10. Bring the sample back into the FOV and start the CT acquisition.

The acquired data needs to be reconstructed using a CT reconstruction software. If you don't have any, we recommend that you use the open source tool MuhRec. The workflow for reconstruction may vary between different software. But in general, the following steps are required:

1. Normalization of the projection data (possibly sinogram generation).
2. Find center of rotation. There may be guides for this in the tools.
3. Find tilt angle of the acquisition axis. This is important as even a fraction of a degree may have great impact of the reconstructed quality. As a rule of thumb: tilt angle should less than 0.3 deg. This is a parameter that should be minimized during acquisition.
4. Artifact reduction. The typical artifacts are rings and lines, removing them will have a great impact in the image quality. Cleaning is easier if you have high SNR images.
5. Start reconstruction of the entire volume.