Wednesday, December 31, 2008

Weekly Patent Review

Kodak filed an application US20080315272 on control of 4T pixel floating diffusion capacitance in such a way that it's low for small signals and large for large signals. The pixel schematics is shown below:


Transistor 201 opens only for large signals, increasing floating diffusion capacitance to accomodate the larger charge.

Omnivision's application US20080318358 talks about using indium doping for the photodiode pinning layer. Indium allows for a shallower implant layer with higher concentration than commonly used boron. An indium-boron combination is also applied for the patent.

10 comments:

  1. I wonder how you know to turn 201 on? What happens if it is off and the signal overflows?
    On the other hand, if 201 is half on and acts like a capacitor, I wonder how this patent will fare against all the prior art out there on cascaded integration and overflow caps for increased dynamic range?
    -EF

    ReplyDelete
  2. Hmm, maybe it is just for interscene dynamic range, not intrascene....
    -EF

    ReplyDelete
  3. According to the application, 201 is not half on, it's either on or off. It's probably for interscene dynamic range.

    If they connect RG2 vertically, rather than horizontally, one could think about some column circuit logic which turns 201 on in columns where signal is large. But I don't think this was the intention of the applicants.

    ReplyDelete
  4. Good for FSI? Looks like it takes up more space and would let in less light. I don't think EK is working on BSI because they have that white pixel thing with truesense that is supposed to let in more light. Putting more stuff in the front kind of defeats the purpose. This kind of thing would make more sense for somebody with a BSI solution like Sony or OVT.

    ReplyDelete
  5. I think indium doping intention was to make the pinning layer shallower, which is good for FSI and does not matter for BSI. Actually, it's a divisional application of US7432543 patent, which Howard filed in 2004 soon after joining Omnivision. I don't think he targeted BSI by that time.

    ReplyDelete
  6. Good point about OVT's patent. But that still leaves EK's. Looks like it adds more bulk to the circuit (as opposed to the last patent you posted which took some away). EK's patent might let the circuit respond better to a range of lighting conditions, but it would block some light too. I'm not sure its worth it for FSI.

    ReplyDelete
  7. As for the Kodak's application, usually floating diffusion capacitance should be balanced with photodiode capacitance. But smaller floating diffusion cap has an advantage of lower noise. So, Kodak controls the capacitance, providing high-sensitivity mode for low light and low-sensitivity higher DR mode for stronger light. I think it's mostly useful for larger pixels, where the extra transistor and wiring might be of less problem.

    ReplyDelete
  8. A comment completely different from the foregoing comments : the serial combination of two reset transistors shown in this patent is exactly the same as the one described by Bart Dierickx in his "Silencing Reset" idea ....
    The world is even smaller than I ever expected it was going to be. Happy 2009 to everyone.
    A.T.

    ReplyDelete
  9. Yes, I have not realized this at first, but on the pixel schematics level Caeleste's and Kodak's ideas are almost identical. Good point!

    ReplyDelete

All comments are moderated to avoid spam.