This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

DLP5500: Equivalence between DLP5500 and 1076-6439B

Part Number: DLP5500

Dear all,

I'm currently prototyping a system based on the DLP5500, which for now comes from a Hg-HID lamp-based projector which I "hacked". The ref number of the DLP extracted from the projector is 1076-6439B. Analyzing the reflectance (with a spectrometer) of this DMD+protective window, I observe a severe drop of reflectance in blue area, which makes sense since the lamp produces a lot of blue light compared to the rest, and a projector is supposed to send an almost spectrally flat spectrum.

The problem is that this reflectance does not correspond to that of the Corning Eagle XG glass described in the DLP5500 datasheet. For some reason I need quite some flux in the blue area of the spectrum, and I'm therefore wondering if, apart for the protective window coating, I should expect any change on the electronic side between the genuine DLP5500 and the 1076-6439B? If not, then I could make a simple replacement of the 1076-6439B with the genuine DLP5500.

Thanks for your help,

Simon

  • Hello Simon,

    First, welcome to the DLP section of the TI-E2E community.

    Did you remove all of the projector optics or are you still using it's illumination path.  If so there is almost certainly a filter that starts cutting off UV light and some of the shorter blue wavelengths.

    If not, could you indicate what wavelength it starts significantly dropping off at?

    Fizix

  • Hello Fizix,

    Yes, I just did the measurement on the DLP itself, all other optical items in the path being removed.  In attachment you can find the reflectance I get for the DLP, nevermind about the absolute scale, only the shape matters here, and sorry for the bad quality ;-).

    As you can see, the maximum of transmission is around 600nm, and it drops fast as the wav gets to blue/UV region.  It also drops - although more slowly - towards the infrared region.

    S.

  • Simon,

    What is the f# of your illumination and pickup for the spectrometer?  If you are using a fiber system then you are likely seeing the limited f# diffraction efficiency.

    What angle is your illumination coming from and what angle is your pickup looking at the DMD from?  I would expect 24° illumination and 0° (normal) pickup.

    Unfortunately you would need to do a bidirectional reflectance distribution function (BRDF) measurement and integrate to get a better picutre if using small apertures.

    The other way is to us a particular illumination cone (say f3) and a pickup with a similar f# or smaller.  This will optically do the integration for you.

    Fizix 

  • Hello,

    There are actually two measurements that I do:

    *the direct one, i.e. halogen lamp->dlp->spectro, and I roughly respect the angles (24deg incoming, 0 deg outcoming). The halogen is placed quite far away (> 1 meter) so we can consider in this case that the light is relatively collimated, which can be translated as a large f/#.

    *the all-mounted one: Xe lamp -> projector optical block-> DLP-> integration volume -> spectrometer. There the ratio between the observed spectrum and the spectrum from the original Xe lamp has the same trend as the one obtained with the direct measurement. This mounting corresponds here apparently to a quite shorter f/#, perhaps between 2 and 3.

    The similarity between the two measurements seems to indicate that the f/# does not have a strong influence on the DLP+window reflectance, which makes sense from the info I could find concerning the Corning Eagle XG glass.

    I'm not sure yet, but I'll probably end up sacrificing one 1076-6439B to analyze the true transmission of the window, i.e., by removing it.

    Now coming back to my initial question, do we expect any difference from the connection/electronical point of view between the genuine DLP5500 as sold from ti website and the 1076-6439B? If not I'll anyway just go ahead buying the genuine DLP5500 as a replacement of the 1076-6439B.

    Best,

    Simon
  • Simon,

    The optical properties of both parts should be very similar.  The projector electronics will do some video type processing, but in your measurements should affect all wavelengths in the same way.

    What do you see if you replace the DMD with a first surface aluminum mirror?  In other words I assume you have characterized the measurement system separately.

    Your result is not expected.  Both the window transmittance and the DMD reflectivity are quite flat in the visible all the way down to 400 µm. 

    Fizix

  • Hi,

    By chance do you have a plot showing the custom part window transmission?

    I didn't try to replace the DMD by an aluminium piece in the projector layout (this brings some complications from the mechanical point of view which I had no time to address). But as said earlier, I measured each item in the optical path separately, including the DMD+window, with just an halogen lamp, the item to measure and the spectro (which is well calibrated, spectralon reflectance is perfectly flat). I'll most probably try to remove the DMD window and measure its transmittance directly, without any parasitic/diffraction effect which may come from the micro mirror array, and check if it fits the Corning eagle XG specs.

    Cheers,

    Simon

  • Simon,

    I do not have such a plot. I am not aware of any difference in the windows of custom parts.

    Fizix
  • Ok thanks, in case I find something interesting I'll post it here! Thanks for your kind help!