FPGA for DMD Controller

Salvatore,

I am reading your question as "Can I configure an FPGA to drive/control a DMD directly?" I am sorry to offer discouragement, but this is not easy, nor is it supported by the data sheet information which is available to customers. DMDs are sold only as chipsets, which include the DMD AND the DMD controller (and sometimes other required chips). The detailed methodology for driving the DMD is TI IP, and is not available to the user. This arrangement is primarily to assure the reliability and robustness of the DMD operation.

Best regards,

Pascal

I wish Ti would be a little more open source about this issue.  DMD have a tremendous promise for all kinds of medical imaging applications, scientific sensors, and displays, but by bundling everything together like they do with their development sets, I think it really inhibits innovation in this area.  My application would require  me to buy 500K in DMDs yet probably the circuitry and the DMDs themselves would be a few thousand real dollars with an FPGA and some used DMDs from old projectors (they run $3/each on some sites).  The basic mechanism to control DMD is not that big of a deal and can be done by an FPGA circuit, but I think their development circuitry is too bloated, nany-ish, and limited really.,,it is a lot of small frills but no information about how to actually control the signals to the DMD or push it too its max capacity (which is MHz not kHz).    By beihg a little more open minded, they could really encourage much more innovation that they may not see themselves, yet benefit from substantially.   Hiding and obfuscating stuff is not a good solution to the IP problem, and it does no one any good in the long run.  There are just too many examples of this in the modern world to justify it.  Just my opinion though.  I have a feeling though like the horseless carriage, we have 30 years of IP issues before society can truly benefit from the tremendous potential of these miraculous mini-machines. 

Hi Tim,

Thank you for taking time to share your thoughts.  We agree with your opinion that DLP Technology holds tremendous promise for medical, industrial and scientific applications.  We also recognize the frustration with the current situation.  You and others, reading this post, likely know all too well the difficulty in harvesting DMDs for these applications from commercially available projectors.  

Steps we have taken to ease some difficulty and frustration include:  reselling the most fitting DLP chip sets through our distribution partners (e.g. Arrow, Avnet, Digi-Key, etc.); creating development platforms with our Design House Partners (e.g. DLP LightCrafter, Discovery 4100, etc.); creating support documentation as well as supporting this E2E forum.  

However, we know there is still work to do.  We are listening to you and others on what we need to do next.  Please keep pushing us on what you need to be successful.  We want to make DLP Technology available for everyone.  The way we make it available is less about protecting IP and more about making the technology easier to use and maintaining the high reliability that you expect.

Thanks again,

Eric

Hi Tim,

I'm running into the same problems as you. Did you finally try to control the DMD with a FPGA? If so, how did it go?

Best Regards,

Daniel

Hi, good day

 i also have same concern,   however, from some of the document i see there are one FPGA for processor,  while controllor,  drivers are still asic.   is there one programable SOC solution to achieve dmd become pico projector?

expecting for the answers

many thanks.