-pages to create: Histogram; sharpness/unsharpness of voxels; artifacting and how to correct; voltage/power and relation to wavelength; Projection v. Slice; Ghosting and its correction
Once you have positioned your specimen, you can begin to adjust the parameters.
Before you begin adjusting parameters, turn on the histogram to monitor the pixel values. Right click on it to show a logarithmic scale.
Voltage (kV) is the energy of the photons being produced. Note, that the source will always produce photons with a range of energy (polychromatic) and the number you set only defines the maximum power of those photons. Try to set the voltage so that it fully penetrates the sample: when minimum signal count displayed in the histogram is greater than at least 170 and preferably over 200. Do not set the voltage above 90% of maximum (216kV for 240kV microfocus tube, 162kV for 180kV nanofocus tube). There are other parameters that increase penetration if voltage alone cannot.
Current (μA) is the number of photons being produced. Increasing current will increase penetration and increase the range of grey values. A wider range of grey values makes a good scan and should be optimized. However, using a current higher than 200μA can be hard on the system and can cause a blown filament. Power is a product of current and voltage. In most cases, power controls the focal spot size so that power = spot size. There is an exception to this when using the nanofocus 180kV tube with the spot setting of 1, 2, or 3. In these cases the spot size is fixed:
spot 1 = 2μm
spot 2 = 1.5μm
spot 3 = 1μm
Power is also restricted by the maximum power allowed by the tube. The Power should never go above 70% of this maximum.
max power for 180 tube = 15 W 70% of 15 W = 10.5 W
max power for 240 tube = 320 W 70% of 320 W = 224 W
Timing (ms) is equivalent to exposure time on a camera. Increasing timing will increase penetration, but will also increase the time of the scan. Increase timing if you need more penetration than what the power limits will allow at your current setting. Increasing the overall time of the scan can be hard on the equipment too. Aim for scans of less than five hours and run scans of more than 10 hours only with extreme caution.
The Average number will determine how many projections will be averaged together for per projection. This normalizes out the natural variation in the signal and reduces noise. Normally, this is set at 3, but can go up to 6 if a cleaner image is needed or if you are scanning a sample with a voxel resolution finer than 2 um. Increasing the Average will increase scan time.
Noise: To get a measure of noise, select a region of air and look at the range listed in the histogram.
The Skip number determines how many images will be thrown out per projection. This allows the specimen to settle if it wiggles during rotation. Skip default is 1. If Average = 3 and Skip = 1, then there will be four images taken for each projection, the first of which is discarded.
Sensitivity also increase penetration, as well as increasing noise. This is usually more helpful for very dense metal samples, not already noisy fossil samples or embryo samples. Increasing the number of averages can counteract the noise, but will increase scan time.
The Images number determines the number of projections to be taken for the full 360°. A sample that is narrow on the screen and therefore takes up less pixels across it’s widest part will need fewer projections to be reconstructed than a sample that takes up the full width of the screen (2048 pixels). The pixel width determines the number of images needed. The recommended setting is number of images = widest pixel width of sample. Therefore, number of images is usually ≤2000. If the number of images < pixel width, then small artifacts may occur on the outer diameter of the reconstruction. There will be no improvement to quality if number of images > 1.5 x greatest pixel width. Filters are small square sheets of copper or tin. Because the beam is polychromatic , filters block the lower energy range of photons, giving the beam a more consistent energy. This reduces beam hardening artifacts. Filters are recommended for very dense samples. Filters also prevent pixels that detect air from being oversaturated when the signal is pumped up to penetrate dense samples. When you add a filter, remember to enter this information with the Filter button . Do not bend or crease filters.
Ghosting: To prevent ghosting from occurring on the detector panel, please do not let the machine sit with the x-rays on when a sample is present longer than necessary. Ghosting will leave an image on the panel, which will fade in time, but is not healthy for detector longevity.
Oversaturation: If the background turns red, this indicates that a pixel is oversaturated. This may mean that you need to decrease the voltage/current/timing, add filters, or that there is a bad pixel. Quad Screen View and Snapshot: Quad Screen View is used to compare how one set of parameters looks (with averaging) side-by-side with a second set. Start by taking a Snapshot of the image you want to compare (on toolbar next to Live View). Double click on the image and select Quad Screen. You can then use the other three empty windows to take additional, comparative Snapshots
Quick Guide to Relationships Between Voltage and Current:
↑ sample density = requires ↑ penetration
↑ voltage = ↑ penetration ↑ current = ↑ contrast
↑ timing = ↑ penetration ↑ timing = ↑ scan time
(voltage) x (current) = power focal spot size ≤ voxel resolution
focal spot size in 240 tube = power
focal spot size for 180kV microfocus tube:
microfocus spot 0 = power
microfocus spot 1 = 2 um
microfocus spot 2** = 1.5 um
microfocus spot 3** = 1 um
**Reminder: you will need to increase the Average when using these settings
↑ focus spot size = ↑ unsharpness
(voltage) x (current) < 70% of max power
max power for 180 tube = 15 W 70% of 15 W = 10.5 W
max power for 240 tube = 320 W 70% of 320 W = 224 W voltage < 90% max power (216kV for microfocus, 162kV for nanofocus)
Go back to Step-by-Step.