EEVblog Electronics Community Forum
Electronics => Projects, Designs, and Technical Stuff => Topic started by: compet17 on December 24, 2014, 10:09:28 am
-
Recently I got a scrapped c-arm x-ray unit into my hands. Among a lot of very interesting parts there was a custom made (Siemens) CCD camera that made the TV picture from the x-ray image intensifier. It has a monochrome CCD chip for normal visible light (Texas Instruments PAL) which is cooled by a Peltier element under the chip. There is also a temperature sensor attached.
BIG QUESTION: What is the red LED on the side of the CCD chip for? (No, there is no hole in the chip)
1 - complete camera
2 - lens removed
(http://s16.postimg.org/x2rxa8f5d/DSC00036.jpg) (http://postimg.org/image/x2rxa8f5d/)
(http://s12.postimg.org/numqfks3t/DSC00037.jpg) (http://postimg.org/image/numqfks3t/)
My best guesses are:
a - for calibration with red light.... but the angle of the LED is very unusual for that
b - they use the LED as a light sensor (yes LEDs produce a voltage when light shines on them)... but again the wierd angle.
c - to see dust on the sensor and subtract it by image processing software (but everything is dust proof)
-
c - to see dust on the sensor and subtract it by image processing software (but everything is dust proof)
It's only dust proof once fully assembled.
Detecting dust on the sensor does sound like a plausible reason. Perhaps its part of the factory test to confirm it was assembled with no dust present?
-
Can't see that it could be anything other than test/calibration - position is odd, maybe it's designed to be visible though the lens?
For x-ray use the sensor is probably very sensitive so maybe positioned to get a very small proportion of the light to the sensor - running LEDs at very low current gives very inconsistent brightness
-
PSI: Yeah possible... but I assume these things are assembled in a cleanroom. It must be something else...
Mike: There was an additional objective/lens on that camera. Maybe they had to illuminate the CCD to adjust that lens... but anyway, letting the LED shine into the edge of the chip makes not much sense.
-
PSI: Yeah possible... but I assume these things are assembled in a cleanroom. It must be something else...
Mike: There was an additional objective/lens on that camera. Maybe they had to illuminate the CCD to adjust that lens... but anyway, letting the LED shine into the edge of the chip makes not much sense.
if you look at the flex, it may have been the only sensible orientation. There would be enough internal reflection to get some light to the sensor.
-
Some high-end video cameras had "optical bias" light sources that would change the dark operating point of the sensors to improve performance.
It does seem like an odd placement, but perhaps it is just intended to "flood" the interior space. Whether during a calibration step,or during operation.
-
(http://s4.postimg.org/auq6gt649/DSC00041.jpg) (http://postimg.org/image/auq6gt649/)
(http://s7.postimg.org/7ysy2cilz/DSC00043.jpg) (http://postimg.org/image/7ysy2cilz/)
(http://s23.postimg.org/5ltp30zg7/DSC00044.jpg) (http://postimg.org/image/5ltp30zg7/)
I made some more pics... (1) Here you can see the exact location of the LED, maybe originally it was more centered at the chip edge.
(2) Other side with some part numbers. (3) Extensive (and expensive) mechanical adjusting with spring loaded stoppers and three angled screws from top and 4 setscrews from the side.
-
Upon closer inspection, it appears to be aimed at the substrate, exactly the same as the cameras I've seen. Is there a hole in the sensor package where the light can reach the substrate of the chip?
-
..oh and can we have pics of everything else please ;D
-
Upon closer inspection, it appears to be aimed at the substrate, exactly the same as the cameras I've seen. Is there a hole in the sensor package where the light can reach the substrate of the chip?
That purple ceramic is quite transmissive of IR and red light IIRC, the chips with those packages are pretty transparent, and you can see through the package in bright light to see the internal traces. It probably is an optical bias used to keep the sensor noise down.
-
Anti-dew? The tiny amount of radiant heat from the LED would be enough to heat up the CCD just enough to stop any tiny amounts of water from condensing on the CCD window. Seems an odd way of doing it though. In normal use the power consumed by the sensor itself would be more than sufficient, but perhaps in standby mode it needs some help? Just a thought anyway.
That would have made a really interesting astronomical CCD project!
-
SeanB: That would make sense
Mike: More Pics as demanded :-)
(http://s12.postimg.org/q4ca5y4eh/DSC00046.jpg) (http://postimg.org/image/q4ca5y4eh/)
(http://s7.postimg.org/5ez5dqqt3/DSC00047.jpg) (http://postimg.org/image/5ez5dqqt3/)
(http://s8.postimg.org/lwmtrohep/DSC00049.jpg) (http://postimg.org/image/lwmtrohep/)
(1) This is the c-arm unit where the camera comes from
(2) The optical end of the "x-ray image intensifier" tube (for details you may google that term)
(3) Dumpster dive success... 8 HP thin clients t5540... I assume all working :-+
-
Zad: For that purpose it would be simpler to reverse the current through the cooling peltier and make it heat up. I doubt that this tiny LED would have an impact on dew. A resistor would make mor sense for heating.
But... who knows.... Siemens maybe? ^-^
-
(http://s7.postimg.org/7ysy2cilz/DSC00043.jpg) (http://postimg.org/image/7ysy2cilz/)
Is that a green LED hiding under the CCD package?
-
Would it be possible to upload the photos somewhere else? When i click on a photo I get taken to a site from the Cybercrime Coordination Unit Switzerland saying that the site had been banned because of distribution of child pornography.
Thanks.
-
Is that a green LED hiding under the CCD package?
Most likely epoxy coated thermistor for TEC feedback loop.
-
Is that a green LED hiding under the CCD package?
Most likely epoxy coated thermistor for TEC feedback loop.
I did wonder about that, but the device appears to have square section leads like an LED, and all the bead thermistors I have seen have had very light gauge round leads.
-
One more guess - similar to Richard's.
CCD backlight - to reduce the static noise floor of the sensor - by artificially washing out the image - then by necessity to crush the 'black levels' slightly.
The (type of) power source for the LED may tell you more.
-
It might be to produce a flat field in order to detect cold pixels.
-
It might be to produce a flat field in order to detect cold pixels.
There is nothing to ensure an even distribution of light over the array, so it wouldn't be a particularly flat field, but could detect dead pixels, and over time also measure any degredation by comparing previous reference images - sensors in x-ray systems can suffer damage over time due to the effects of the x-rays.
It could also be a faultfinding aid - as it has an image intensifier, having an internal light source would provide a quick test that the sensor is OK, to distinguish sensor faults from intensifier problems when all you can see is a dark screen.
Due to downtime, parts cost, dust issues etc., being able to give a quick answer to the question of whether a fault is in the camera or intensifier would be useful, certainly well worth the cost of a LED.
I wonder if there is also a LED that can fire some light at the input side of the intensifier, to distinguish intensifier faults from x-ray problems.
-
Would it be possible to upload the photos somewhere else? When i click on a photo I get taken to a site from the Cybercrime Coordination Unit Switzerland saying that the site had been banned because of distribution of child pornography.
Thanks.
I got several popups on my phone telling me I had spyware and needed to install some program immediately.
-
interesting seeing that led there, surely it cant be important for normal operation or it would be better integrated into the ccd and its housing
as mike suggests it must be just for fault detection
would be interesting to know for sure, would tracing the connections back reveal anything?
-
interesting seeing that led there, surely it cant be important for normal operation or it would be better integrated into the ccd and its housing
as mike suggests it must be just for fault detection
would be interesting to know for sure, would tracing the connections back reveal anything?
Probably not, except maybe to see if it's brightness is variable.
You'd need to fire it up & see when it lights to get a better idea of what it's for.
-
ooops.... seems this photo-hoster probably has some problems... I will check that. But I am from Switzerland too and I have no problems to see the pics.
No, this is not a green LED. It looks more like a temperature sensor that is glued to the CCD by some green gunk.
Mike: It is not possible to check the image intesifier with any kind of light source. The input surface is a dome of aluminum which is penetrated by x-rays. Underneath the aluminum dome (inside the vacuum tube) is a screen of cesuim-iodide that converts the x-rays to beams of electrons that are focused electrostatically to an output screen on a glass window. That picture is then picked up by the camera.
(see wikipedia or google "x-ray image intensifier")
-
I remember an odd trick where a severely saturated CCD can be cleared more quickly if flooded with red light. Makes no sense to me, but maybe a search will turn something up.
-
CCD flooding?
http://gxccd.com/art?id=418&lang=409 (http://gxccd.com/art?id=418&lang=409)
-
CCD flooding?
http://gxccd.com/art?id=418&lang=409 (http://gxccd.com/art?id=418&lang=409)
Most likely ;)
-
To erase a residual image... hmmm sounds not bad. Still it is an unusual position for the LED. But it would make sense. I think I accept that as a final answer. :-+
-
Today I have spoken with an x-ray technician. He says that this machine is capable of making so called "instant x-ray prints". That means with a reduced x-ray dose and a longer exposure time, more detailed still images can be made. Normally the machine is used to make TV video live images.
Due to the longer exposure (some 100 ms) it might be necessary to remove residual images with that red light.
-
copy/paste from
http://gxccd.com/art?id=418&lang=409 (http://gxccd.com/art?id=418&lang=409)
The whole CCD is flooded with light during Preflash, so its substrate is completely saturated. It is better to use near infra-red light (this is why “NIR” Preflash abbreviation is used), which penetrates to the substrate more easily. The detector is cleared after flooding with light to remove electrons from active detector areas (image pixels, horizontal register, output node, ...) so the regular exposure can be started. Because all portions of the CCD were completely saturated, CCD clear operation must be repeated several times, one clear is not enough.
If the above described procedure (saturation + repeated clear) is performed before every single exposure (dark, light, flat), initial conditions are always the same and calibration (dark frame subtraction) eliminates RBI. Ghost images are not visible, any residual charge left after previous exposure is overrun during light flooding to maximal value and removed during subsequent dark frame subtraction.
-
More red LEDs in video stuff:
This is from an old endoscope (medical) camera. You can see the tiny red LEDs at the neck of the video tube. This ia also an unusual position to get light to the photo-sensitive area but it seem that very little light is needed to have an effect.
(http://img3.fotos-hochladen.net/thumbnail/dsc000706yq3uikah7_thumb.jpg) (http://www.fotos-hochladen.net/view/dsc000706yq3uikah7.jpg)
(http://img3.fotos-hochladen.net/thumbnail/dsc00071r4w8o0hdav_thumb.jpg) (http://www.fotos-hochladen.net/view/dsc00071r4w8o0hdav.jpg)
There is a LED on each side of the tube.