Replace the window of DLP1700

Ying,

It is NOT recommended that you change the window on the DLP1700. The window is there not only to protect the silicon, but also to ensure the proper workings of the entire structure.

What is your application and reason for needing to replace the window?

I want to use terahertz as the light source, so I have to remove the window which will block terahertz.

If I cut off the window, will the gas or other staff in it (if exist) will goes away? Will this affect the function of the mirror? Thank you very much.

Hello Ying,

Yes, this will affect the function of the mirror if you remove the window. 

However, please be aware that at the wavelengths of teraHz light (0.1 - 1.0 mm) that the DMD will not be effective even if the window is removed.  These wavelengths are 10 - 100 times greater than the size of a mirror so that the DMD will not switch the beam direction when turned on and off. 

Fizix

Hi Fizix:

I'm thinking about  using an array of micromirrors as one large mirror for terahertz, but the first thing is to replace the window. Do you have any idea what will be changed or how bad the DLP will be if I remove the original window very carefully in the clean room and then attach another window to it?    

Thanks,

Ying

but i think it may be possible if the mirrors  work in blocks.

Right now i am also curious whether there is such DMD with terahertz windows available.

Hi Ying,

Unfortunately it is not that simple.  You are correct that doing so will adversly affect the DMD.

Hello Lina,

Welcome to the DLP section of the TI E2E community. 

To answer you question, unfortunately there are no DMD with TeraHz windows available.

I have given this some thought, and I do not think that the idea of a "super-pixel" (group of pixel that work as a unit) will solve the issue. 

I offer for thought the following thought experiment:

• Imagine that we consider each 2x2 group as a "super-pixel". 
• Now turn all "super-pixels" on. 
• What is the difference in state between this and turning all individual pixels on? 

I conclude that there is no difference, therefore the bulk of the energy will go to the 0th diffraction order.  Not to a sub-order  introduced by the larger period of the superpixel.

Fizix

what we care about is one channel of DMD, say, either positive 12 degree or minus 12 degree. And we can randomly control the direction of the each superpixel  just like the low resolution imaging system. Different super-pixels will be on and off randomly, so we can still get different information.

what i concern about is that even the mirrors in the same super-pixel flip to the same dirrection, but there are certain distance between them, i mean they are not in the same plane, whether they can switch the beam correctly.

Hello Fizix,

Is the interior of the DMD filled with a vacuum or some gas?  Then by removing and replacing the window (even in a clean room), will the presence of air about the mirrors eventually degrade the mirrors and cause the device to fail?  The reason I am asking is that I had an extra DMD that I was able to successfully remove the front window from in a clean environment.  It worked well for a period of time, but after some time it not longer functioned.  I took precautions by covering the DMD when not in use to prevent contamination from entering the mirror cavity, and it was only briefly exposed whenever I ran some experiments with the device.

Charlie

Charlie,

You are correct that removing the window of a DMD will adversely affect continued operation of the mirrors.  TI does not recommend removal of the window.

Fizix