Getting a good X-ray image is not a matter of finding the right settings by trial. It follows from three properties of the sample, worked through in order.
Everything starts with the sample itself. How dense and thick is it? What is the smallest feature that has to be visible? How large is it physically? Those three questions map directly onto the three things that determine whether an image will work.
Step 1: Power
Power is voltage multiplied by current, and sample density determines how much of it is required.
An engine block needs substantially more power than a printed circuit board, because there is far more material for the beam to get through. Insufficient power means the photons never reach the detector, and no adjustment elsewhere recovers an image that was never formed.
Voltage and current do different jobs within that total, and the distinction matters when tuning. Voltage governs photon energy and therefore penetration. Current governs how many photons are produced and therefore signal strength.
Step 2: Resolution
Resolution determines how much detail the image can actually show, and it comes from three things: the focal spot size of the source, the pixel size of the detector, and the magnification at which the sample is imaged.
The requirement is set entirely by what needs to be found. Millimeter-scale porosity in a casting demands very little resolution. Micro cracks in a BGA joint on a laptop board demand a great deal of it.
This is worth being honest about in both directions. Specifying more resolution than the application needs costs money and usually costs throughput. Specifying less than it needs produces images that look fine and cannot answer the question being asked.
Step 3: Magnification
Magnification is set by geometry: the distance from the source to the sample, and from the sample to the detector. Bringing the sample closer to the source magnifies it more.
Unlike the first two steps, this is a physical property of the machine. A larger system with more sample manipulation options offers a wider range of achievable magnification, because there are more positions the sample can occupy. A compact system with fixed geometry offers less.
It is also where sample size becomes a constraint. A large or awkward sample may simply not be able to get close enough to the source to reach the magnification the resolution requirement demands.
Why the three interact
Presented as a list, these look like independent checkboxes. In practice each one constrains the others, and that is the part worth understanding before choosing a system.
Power affects resolution. Increasing tube current requires spreading the electron beam over a larger focal spot to manage heat, and a larger focal spot blurs fine detail. So a dense sample requiring high power arrives at step two with less resolution available than a thin one.
Magnification affects resolution as well, since resolution depends on magnification directly, and there is an optimum rather than a maximum. Pushing magnification past that point makes images larger and less sharp.
And sample size constrains magnification, which constrains resolution, which may not meet what step two required.
That is why difficult samples are difficult. A large, dense part with small features to find pushes all three parameters in directions that fight each other.
Using this to choose a machine
Worked in order, these three steps produce a specification rather than a preference.
Sample density and thickness set the power the source must deliver. The smallest feature of interest sets the resolution required. Sample size and the magnification needed to reach that resolution set the geometry and manipulation range the machine must have.
Answer those, and the requirements for a system follow from the answers instead of from a brochure.
