DMD interference

Xiongfeng Zhu

We are testing with DMD 1100 using a checkerboard image displayed on the chip. The light source is 355nm laser. However by doing this, from the CCD imaging we saw an reflection image on the screen with white blocks in where it's supposed to be lit, but also with smaller light spots in where it's supposed to be no light at all. So is this caused by interference? If so, what's causing this (by that glass window?) and how can we remove that?

Thanks

Xiongfeng

over 13 years ago

0 Fizix over 13 years ago

TI__Guru 54725 points

Hello Xiongfeng,

The reflected image is caused by the Anti-Reflective (AR) coating on the DMD window. The DMD window is optimized for the visible range (400 nm - 700 nm). The reflectivity falls off rather dramatically in the UV. Althought the AR coating has been improved it is very similar to the coating now empoyed. Please see the following link for curves on Transmittance in the UV for the Visibile AR coating. --> http://www.ti.com/general/docs/lit/getliterature.tsp?literatureNumber=dlpa031b&fileType=pdf

From that documentation you can see that the Transmittance has falled to just below 75% at 355 (see Fig 2 on page 2).

Fizix

0 Fizix over 13 years ago in reply to Fizix

TI__Guru 54725 points

Hello Xiongfeng,

I thought I would also mention that if you are seeing many spots in a periodic pattern (array) then you are seeing the diffraction of the coherent laser with the periodic spacing of the mirrors.

Fizix

0 Xiongfeng Zhu over 13 years ago in reply to Fizix

Prodigy 75 points

Thanks for your reply!

As for the 'big' diffraction, we are aware of that, and we've already picked up the first order diffraction pattern to look at. The problem now is inside the first order diffraction the projected imaging is not the same as the pattern we want. Below are two sets of our tests.

The first one is the checkerboard pic we displayed on DMD chip, the second one is the projected image we saw through CCD camera. You can see there are unwanted small spots between normal blocks. That's the problem we are having.

This set also shows that projected image and display pic don't exactly match.

So will interference explain this phenomenon?

Thanks!

Xiongfeng

0 Fizix over 13 years ago in reply to Xiongfeng Zhu

TI__Guru 54725 points

Hello Xiongfeng,

I have several questions:

Have you looked at the DMD surface itself to make sure that the pattern is actually as expected on the DMD? What I am asking is if the first picture (at the top) an actual picture of the DMD or the image that you sent to the DMD?
Are your optical elements made for this wavelength? I have tried to use lenses made for visible light with UV and usually they will not pull sharp focus. You may need to get some lenses made for this wavelength of light.

Fizix

0 Xiongfeng Zhu over 13 years ago in reply to Fizix

Prodigy 75 points

Yes. The image I saw on the DMD is the same as what we want.

And the optics should be fine.

0 Fizix over 13 years ago in reply to Xiongfeng Zhu

TI__Guru 54725 points

Hello Xiongfeng,

I have asked our optical team to take a look at this and see if they have any thoughts. I will post again when they respond.

Fizix.

0 Fizix over 13 years ago in reply to Fizix

TI__Guru 54725 points

Hello Xiongfeng,

The response from our optical team is that it is very difficult to tell from the pictures you have posted. It appears to very poorly focused.

I would recommend that you contact one of our Design Houses if you are planning to develop with DLP Technology. They will have options available to you depending on your needs. Please see the following link to find a Design House that will meet your needs (location and capablities).

Click --> http://focus.ti.com/analog/docs/memsdlpdsgn.tsp?sectionId=622&tabId=2448

Fizix