This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

DLP9000: What is the off-state fill factor so I can calculate T_array?

Part Number: DLP9000
Other Parts Discussed in Thread: DLP2000


I'm trying to determine that my chosen operating conditions are not too harsh for my DMD. In my application, most micromirrors spend a majority of the time in the off-state under intense light power. 

The following document details how to calculate T_array for a non-uniformly illuminated partial DMD array:

Clearly, one of the pieces of information I need for this calculation is the off-state fill factor, FF_off-state-mirror. However, this is not specified in the datasheets for the DLP2000 or DLP9000, both of which have 7.6 um-pitch micromirrors. The document above gives FF_off-state-mirror = 75.3% for the 10.8 um-pitch micromirror DMDs, but I believe it should be slightly lower for DMDs with a smaller micromirror pitch. 

I'd very much appreciate if you could share the needed information for all DMD types. I also understand this number might depend on illumination angle, but a nominal value (within 1%) would suffice.

Thank you very much!


  • Hello Sarah,

    These numbers are not published for all devices.  The device in the publication is an NIR device and typically has a smaller numerical aperture.

    For visible light the effective fill factor is very dependent on f-number (numerical aperture) illumination angle and wavelength.

    If you already have a kit set up the way you would like optically, the absorption with you system can be measured but putting all of the mirrors off position and illuminating with one or two watts and measuring the resulting temperature rise per watt.

    I wish it was a little more straight forward.

    How much power are you contemplating exposing the DMD to?


  • Sarah

    For 7.56 um pixel you can use the off state absorption factor of 0.39 for your calculation.

    This  assumes the array is fully illuminated with an illumination distribution of 90% on the active array and 10% overfill on the array border. It is
    strongly recommended to minimize illumination overfill as much as possible which will, in turn, maximize the active array illumination, increase overall system efficiency, and reduce DMD thermal heating.

    what area of the DMD active array do you intend to illuminate ? 

    Note R SILICON-TO-CERAMIC (C/W) increases as the illumination area is reduced and is specific to each DMD/package combination

  • Hi Carey,

    Thanks for your reply. I am curious, what f/# and illumination angle are you assuming with this estimate?

    I am intending to minimize overfill as you suggested. I have also already obtained the correct value of R_silicon-to-ceramic from TI for my intended illumination area.

  • Hi Fizix,

    Thanks for your reply. I understand this number is highly dependent on illumination angle and f/#. I imagine I can extrapolate this number using the on-state fill factor and some rough geometry/trigonometry.

    I am a bit confused as to how I could measure the off-state fill factor using temperature rise per watt as you suggested. Would you mind providing a bit more detail? I have the means to measure T_ceramic at TP1, and can calculate T_array using the information TI has given me for my specific application.

    Thank you!

  • Sarah,

    I was not suggesting that you can measure the fill factor, but rather the heat load per watt.

    First measure the temperature with all the mirrors in the on position and wait until it reaches equilibrium.  Then turn all the mirrors off and measure the rise in temperature when it reaches equilibrium again.  If you do this with small enough illumination power, you can determine the degrees per watt of added load when off compared to when on.

    Once you have that you should be able to estimate the all off load at full power and hence the heat solution needed to achieve the array temperature to maintain.  Does this make sense?