Tuesday, December 27, 2011

1/f and RTS Noise Reduction

As mentioned in Theses post, Oregon State University published Drake A. Miller's PhD Thesis "Random Dopants and Low-Frequency Noise Reduction in Deep-Submicron MOSFET Technology". The thesis is quite rich in experimental data os pixel source follower noise. The figure below shows more than order of magnitude variations in 1/f noise across the wafer:

Noise spectral power plots of 10 devices taken from
10 different locations across the wafer (see inset).

Any channel doping, such as Vth adjust, significantly increases 1/f and RTS noise:

Box plots of source follower noise power spectrum plots.
Red (Dark) boxes are doped devices.
Green (Light) boxes are undoped “native” transistors.

Few Vth adjust splits were measured:

It's not clear why S4 and S7 are not shown, but S1-S3 clearly show noise improvement:

The total read noise histogram clearly demonstrates the advantage of lightly doped source follower:

RTS Statistics shows the same:


  1. Question:
    The source-follower & selection transistors have different threshold voltages in this work. Normal the SF transistor and selection transistor are located very very close. How this Vth splitting is possible???

  2. Vth adjust implants are quite low energy. Normally it's not a problem to separate them between select and source follower. Worst case, one might need a higher grade mask for Vth adjust implants.

  3. Vladimir, what are the typical Vth values for NMOS & PMOS without Vth implant in a 0.18um CMOS process? Thanks in advance!

  4. Depends on other implants in pixel. Typically it's in the range of 0-0.5V.

  5. If you take a standard NMOS/PMOS, if the Vth implants are skipped, what will be the effective Vth please?

  6. There is no "standard" here. Every fab, every process has its own Vth. There are different types of pmos/nmos also: core, I/O, natives, high-voltage, etc.

  7. I would like to know the response of such pixel at low light levels. Any non-linearity ar low level may mislead any noise measurement since it is affected by the compression in case it is present. In other words, the noise results should be given using the conversion gain atlow level or just calculated by the dark current.

    I would appreciate such data, the linearity at low levels, since many sensors in the market show this effect.

    Eve more, for one supply pixels and native SF it is always present somehow.


All comments are moderated to avoid spam.