DLPA052A November   2014  – August 2025 DLP9000 , DLP9000X , DLP9500 , DLPC900 , DLPC910

 

  1.   1
  2.   System Design Considerations Using TI DLP Technology down to 400 nm
  3.   Trademarks
  4. 1Introduction
  5. 2Thermal Considerations
  6. 3Duty Cycle Considerations
  7. 4Coherency Considerations
  8. 5Optical Considerations
  9. 6High De-magnification System Considerations
    1. 6.1 Incoherent Sources (Lamps and LEDs)
    2. 6.2 Coherent Sources (Lasers)
  10. 7Summary
  11. 8References
  12. 9Revision History

Duty Cycle Considerations

All applications benefit from operating the DMD near 50% landed on/off duty cycle, but this consideration becomes more important in the shorter wavelength arena and in higher temperature operations. The landed-on/landed-off duty cycle indicates the percentage of time that an individual micromirror is landed in the on state vs. the off state. The switching time between states is considered negligible and ignored when determining duty cycle.

Duty cycle for DMDs is expressed as the landed-on percentage/landed-off percentage. For example, a pixel that is on 75% of the time and off 25% of the time is denoted by 75/25. Note that the two numbers always sum to 100.

Operating at or near 50/50 promotes the longest DMD performance. There are two possible scenarios of operation under this consideration.

The first scenario is when the pixel histories are not known or tracked. Operating the DMD at 50/50 whenever the DMD is not actively being illuminated(1) drives the average back toward 50/50. The longer this is operated at 50/50 in quiescent periods, the closer the overall average is to 50/50.

Note:

Shutter (or turn off) illumination any time that DMD patterns are not needed or being used at the fabrication surface. Do not use the DMD as the primary illumination shutter, instead shutter at the source, such as a mechanical shutter.

The second scenario is when the history of each pixel is tracked. In this case, the pixels can be operated in the inverse duty cycle for an equal period of time when not actively being used for patterning and at 50/50 after that. For example, if a pixel is driven at 62/38 for four hours during operation, then driving at 38/62 for four hours during quiescent periods averages to 50/50(2).

Note:

The 50/50 average is an oversimplification since during operation when being illuminated heats the pixels, whereas operation when the illumination is off cools the pixels. For high fluence systems, experimentation needs to be done to determine how much longer to run the reciprocal duty cycle or how much to weight the reciprocal to restore the residual tilt to flat.