Radiographic imager using camera system

[DysonDepplbom]

The goal is to build an imager that is capable of high quality stills and videos of similar quality to commercially available imagers.   The finished project would be a general-purpose imager using either hard or soft rays. The system would interface with a windows pc allowing live viewing and remote capture.

Design Overview
A CCD is mounted above the direct beam (no gamma/alpha particles hitting sensor) and behind a phosphor screen. The CCD is pointed down at a mirror at roughly 45 degrees that reflects the light emitted from the phosphor screen. All of this is housed in a sealed container with a front made from a thin panel of aluminum.

This should be about as simple of a project as it gets (I hope). The components are only a light-proof container, phosphor screen, optical mirror, and camera.

My interest lies in the CCD/CMOS style imagers like the SAIC RTR4 due to it's simplicity. Here is the patent information regarding their imager. This seems to be the most popular non-medical imager used in mostly NDT applications.

Examples of  Imaging Systems
Google Patent (SAIC RTR-4)
https://patents.google.com/patent/US5608774

Also here are a few other examples with color photos of similar imagers.
https://www.neutronoptics.com/cameras.html


X-Ray Detector Types (Hamamatsu)
https://www.hamamatsu.com/resources/pdf/ssd/e09_handbook_xray_detectors.pdf



Considerations...
The flat front of the imaging device (where the beam makes first contact) has a panel of aluminum and/or carbon fiber.  Aluminum removes the undesirable lower output rays.

Industrial Monochromatic cameras are used in imaging.  High frame rate and resolution is desired. Often linescan cameras are used.  I don’t have any experience in these types of cameras, or the setup required.

The CCD is protected with shielding to prevent deterioration of image.

The phosphor plates come in different colors indicating their composition. The two most used types of screens/plates are “green” and “blue”. The size of the phosphor screen will likely be (10” x 8”) or (17” x 14”).  It might also be reasonable just to build two imagers and have the larger one semi-collapsible. 

The phosphor screen can also affect the end quality of the image. The size of the phosphor particles is correlated with the resolution. I’m unsure what type of screen (blue vs. green) is best for this application. (I assume green). I also don’t know if a “special” finer phosphor particle screen is used specifically in imagers.

The cameras tested so far (video only) include a Sony 4k DSLR (poor image even with adjusted settings), Hero5 (worst image, barely visible) and an old iPhone 6 (best image so far). No filters were removed from these cameras.

The resolution and quality of the image is of first importance and I feel a more suitable monochromatic camera is needed.  I don’t have much experience with these types of cameras and could use some recommendations in how to set it up and what type/model camera to use.

I found one source that indicates blue phosphor screens have a higher resolution (smaller particles) but I cannot find any other information supporting this. I figure green is still probably better due to the lumen output and “easier” wavelength for the CCD to detect.

I’m unsure if a trigger mechanism is needed. Most testing has been done with a 0.1-1.5 sec exposure giving the camera time to adjust. Lower exposure times are desirable even though the source is operated remotely in a lab setting. 

Some of the sources operate in fast nano seconds pulses. I suspect a trigger mechanism is needed for these type of systems. The decay time of these phosphors appear to be about 250 nanoseconds.

Hamamatsu states that CCD cameras in these systems can be considered “consumable” in that the sensor can deteriorate over time.  I figure that is only under heavy use and would not apply here.


The mirror being tested so far is a flexible adhesive mirror attached to single pane glass. I ordered some better suited “first surface” mirrors that will be tested when they arrive.

I considered using something more advanced like a TFT/ Amorphous silicone type system.  Everything seems doable and I would get to work with a new technology. However, the software that reconstructs the image doesn’t seem like it would be available or easy to find. 


Questions and concerns...
Anyone have suggestions in terms of what camera to use? and what type of lens?

General thoughts relating to the imager design and components?

Recommendations on how to improve design or image quality?

Is there an ideal phosphor screen for use in imagers?

If and how to setup the trigger system (if its needed)?


I’m happy to hear any thoughts on the design of the imager and components. Specifically what type of camera would be best suited for this application and the setup process for machine/industrial cameras. There are so many options for cameras across all price ranges.  I assume monochromatic, large sensor, high resolution and maybe even"cooled" might be desirable.