DLP3010: Frustrated TIR

Hello Marc,

Welcome to DLP forum and thank you for your interest in DLP technology.

I will follow-up with our optical expert and get back to you on this.

regards,

Vivek

Hello Marc,

Your observation is correct . The frustrated TIR effect is due to evanescent wave in the thin air gap.  The RTIR does not pose the same kind of problem, for two reasons.

  1. There is still an evanescent wave at the critical angle in the RTIR prism, but there is no air gap to another element at that surface, so there is no “leakage” of the evanescent wave back into the optical path.

  2.  The air gap in the RTIR prism is in the illumination path, and both surfaces of the gap are in transmission, not TIR. So, there is no evanescent wave issue at any thickness of air gap in the RTIR design. The thickness can be relatively large, even up to 1mm or so, and therefore the two prisms do not have to be cemented together in close alignment. This eliminates a costly assembly process relative to the TIR prism.

The minimum air thickness in a TIR prism is really just a manufacturing issue; some manufacturers can supply prisms with an air gap down to 5 micron reliably.

I could not find reference on  how to “calculate” the evanescent effect, but I am sure it is well described in a textbook somewhere.

The point is that it will be there, unavoidably, because the air gap in a TIR prism has to be very thin in order to prevent astigmatic distortion in the projected image. The smaller the pixel, the smaller the gap has to be to avoid significant MTF degradation due to astigmatism from the air gap. That is one advantage of the RTIR prism in that there is no air gap in the projection path, unlike the TIR prism. Not only that, but the air gap in a RTIR configuration does not cause any evanescent wave issues. Any image distortion issues due to the air gap in a RTIR configuration are in the illumination path, not projection path, so the effects of the distortion are orders of magnitude less evident or critical in the illumination path.

When done well, there is little difference in the contrast achievable with either prism type. In both cases, extra attention must be paid to getting the OFF-state light out of the prism without secondary reflections or scattering off prism surfaces or edges that can get back into the projection pupil. Prism edges are notorious scattering surfaces that will essentially become new lambertian light sources when illuminated by OFF-state light. There are more challenges in managing the OFF-state light and the many possible secondary reflections it may have in the prism as well as in the optical engine mechanics for a RTIR prism vs. a TIR prism, but it can be done.

regards,

Vivek

Hello Vivek,

Thank you for your quick reply.

To limit the stray light would it not be appropriate to use absorbent paints on the non-useful faces of the prism which can be in addition to grind surfaces to increase the absorption area.Do you have references of such paintings?

I have another question regarding the DLP3010 : on page 18 of the datasheet in paragraph 6.11, I don't understand what the tilt angle error of +/- 1.4 ° corresponds to?

Finally, can you tell me where I can find the 3D files (STEP or IGES) of the DLP3010? The exact shape and position of the opening (detail F and G in the package option addendum of the technical sheet) could be useful, as well as its size to add my optical illumination and projection system with sufficient space between the components

Regards,

Marc

Hello Marc,

Please find the response to your questions:

You wrote -

To limit the stray light would it not be appropriate to use absorbent paints on the non-useful faces of the prism which can be in addition to grind surfaces to increase the absorption area.Do you have references of such paintings?

TI response -

  Yes,  Problems are generally caused by scatter from prism/lens edges and mechanical surfaces. Please do following:

Blacken and carefully baffle lens barrel to minimize scattering from walls

Blacken all lens edges

I do not have reference for a type of paint to use.

You wrote -

  I have another question regarding the DLP3010 : on page 18 of the datasheet in paragraph 6.11, I don't understand what the tilt angle error of +/- 1.4 ° corresponds to?

TI Response - 

 The Specification for the tilt angle of TRP mirrors used in the DMD is 17 degree +-1.4 degree. It is driven by the manufacturing process control. This is allowed variation over all DLP3010 dmds. Typically, tilt angle variation with in a device is in much tighter range and also for a batch/lot. 

Step file - I will get back to you on this.

regards,

Vivek

Hello Marc,

STEP files for DLP3010 are attached.

regards,

Vivek2512014-3-720p-s245-DMD.zip

Hi Vivek,

thank you for your answer for the tilt angle tolerance and that's what I understood when reading the detasheet.

Then, if we make a lighting with an inclination of 34 °, in the case of the position in the passing state, the angle of the axis of the cone can vary by +/- 2.4 ° from the direction perpendicular to the DMD which can introduce vignetting phenomenon. As this angle can be different from one pixel to another as indicated by the note (4) in the paragraph this effectively implies an optical design which can take account of this phenomenon especially for an instrument with high optical quality.

Regards,

Marc

Hello Marc,

The mirror tilt angle variation with in a DMD is only few tenths of  a degree. For example if the mean mirror tilt angle in a device is say 17.3 degree then variation of individual mirror tilt angle will be < 0.2-0.3 degree. However, the mean tilt angle form one DMD to another DMD from a different lot could vary +/- 1.4 degree.

Since the variation in a device is in such narrow range, you will not see artifacts like vignetting.

For optical engine design,  the recommended  practice is to design the optics for 17 degree tilt (F/1.7). Marginally increase the illumination angle to 35 degree. The DMD manufacturing process for TRP based DMD is typically centered around north of 17 degree. Following suggested design practice will enable higher contrast ratio.

Hope this addressed your concern. Please let me know if you any further question.

regards,

Vivek