direct CCD x-ray with Digital Rebel

A friend has managed to take an x-ray using a Digital Rebel 350D:

http://yfrog.com/0urebelxrayj

The lens of the Rebel was removed and the CCD covered with a piece of thin
cardboard (to prevent extraneous light). The IC/ capacitor were then placed
directly on the CCD. Exposure time of both camera and 50 Kv x-ray source
was 1.5 sec. No intensifying screens were used anywhere within the system.
Poisson noise was significant and was reduced with carefully applied noise
filters.

On Oct 19, 2:09*pm, "Xrayman" wrote:
A friend has managed to take an x-ray using a Digital Rebel 350D:

That guys has balls.... at least for now.

On Mon, 19 Oct 2009 08:09:37 -0400, "Xrayman" wrote:

A friend has managed to take an x-ray using a Digital Rebel 350D:

http://yfrog.com/0urebelxrayj

The lens of the Rebel was removed and the CCD covered with a piece of thin
cardboard (to prevent extraneous light). The IC/ capacitor were then placed
directly on the CCD. Exposure time of both camera and 50 Kv x-ray source
was 1.5 sec. No intensifying screens were used anywhere within the system.
Poisson noise was significant and was reduced with carefully applied noise
filters.

Silicon has virtually no sensitivity to photons in this energy range, so
I'd guess that the x-rays are being converted to light by some
scintillation process in a passivation layer, microlens array, or cover
slip.

The 350D uses a CMOS sensor, not a CCD (so it is incorrect to call this
a "direct CCD x-ray"), which means the image may also be the result of
the x-rays interacting with the individual pixel amplifiers, and not the
sensor array at all.

In any case, this is probably an extremely inefficient imager, with a QE
of a few percent at most, and maybe well under 1%. I'd guess the source
was pretty high intensity? If you tried to use this technique on
something biological, you'd cook the specimen! Why not use a
scintillation plate with this camera, so the x-ray intensity can be
reduced to a safer level?
_________________________________________________

Chris L Peterson
Cloudbait Observatory
http://www.cloudbait.com

Chris L Peterson wrote:
On Mon, 19 Oct 2009 08:09:37 -0400, "Xrayman" wrote:

A friend has managed to take an x-ray using a Digital Rebel 350D:

http://yfrog.com/0urebelxrayj

The lens of the Rebel was removed and the CCD covered with a piece of thin
cardboard (to prevent extraneous light). The IC/ capacitor were then placed
directly on the CCD. Exposure time of both camera and 50 Kv x-ray source
was 1.5 sec. No intensifying screens were used anywhere within the system.
Poisson noise was significant and was reduced with carefully applied noise
filters.

Silicon has virtually no sensitivity to photons in this energy range, so
I'd guess that the x-rays are being converted to light by some
scintillation process in a passivation layer, microlens array, or cover
slip.

I'd be more inclined to think it would be fluorescers like kaolin as
filler in the cardboard converting Xrays into visible light.

The 350D uses a CMOS sensor, not a CCD (so it is incorrect to call this
a "direct CCD x-ray"), which means the image may also be the result of
the x-rays interacting with the individual pixel amplifiers, and not the
sensor array at all.

In any case, this is probably an extremely inefficient imager, with a QE
of a few percent at most, and maybe well under 1%. I'd guess the source
was pretty high intensity? If you tried to use this technique on
something biological, you'd cook the specimen! Why not use a
scintillation plate with this camera, so the x-ray intensity can be
reduced to a safer level?

Sounds like good advice.

Regards,
Martin Brown

On Mon, 19 Oct 2009 16:15:01 +0100, Martin Brown
wrote:

I'd be more inclined to think it would be fluorescers like kaolin as
filler in the cardboard converting Xrays into visible light.

Agreed. When I used "scintillation" I was referring to any luminescent
process; certainly something in the cardboard would be a good candidate.
_________________________________________________

Chris L Peterson
Cloudbait Observatory
http://www.cloudbait.com

Guys,

To help address your comments:

Chris, yes, CMOS is the right term, sorry about that.


The source was 50 Kv, 1 mA with the object/ camera placed about 6" away. If
there is some sort of scintillation occurring, it is not from the cardboard
as the same results are achieved either with or without it (the cardboard
was used only so the room would not have to be totally darkened each time).

Note that it was difficult to "convince" the Rebel to actuate without a lens
attached. A lens had to be disassembed and just the bottom circuit board
used.... wish there was a better way.


"Chris L Peterson" wrote in message
...
On Mon, 19 Oct 2009 16:15:01 +0100, Martin Brown
wrote:

I'd be more inclined to think it would be fluorescers like kaolin as
filler in the cardboard converting Xrays into visible light.

Agreed. When I used "scintillation" I was referring to any luminescent
process; certainly something in the cardboard would be a good candidate.
_________________________________________________

Chris L Peterson
Cloudbait Observatory
http://www.cloudbait.com