After several years of development, High Definition Integration Ltd. or HDI is now coming out of stealth mode. Insight Media was invited to visit the facilities and get an update on their laser-based 3D technology. I was impressed with the level of innovation demonstrated in the rear-projection system.

Art Berman
Insight Media Consultant

For example, the light engine has been under development for the past three years and uses LCoS panels with exceptional flatness and speed of response, according to principals Ingemar Jansson and Ed Sandberg. The basic panel technology came from MicroDisplay Corporation, which HDI acquired in 2007. The FHD resolution panel includes an array of internal spacers that enable a high degree of flatness, which is required for laser illumination. The turn-on and the turn-off time of the LCoS are reported to be 0.5 milliseconds, which is fast for a nematic-based LCoS.

For the laser light source, HDI has developed a laser module that contains a red, green and blue laser, each having an output of 1 Watt. Thermoelectric cooling is also included in the package. The output of the red laser is at 635 nm. The blue laser is supplied by Nichia and has an output that is currently at 465 nm but will be 447 nm in the final product. The green laser, which has been developed at HDI, has an output at 532 nm and utilizes a diode pumped solid-state laser in which the IR output is frequency doubled. The outputs of the three lasers are coupled into a single 50-micron fiber having a suitable numerical aperture. The output of the laser module is about 1K lumens of white light. The overall energy efficiency is about 20%, Jansson and Sandberg said.

The laser modules are currently being assembled at the HDI facility-in a process described as "simple" but requiring "high precision." If a projector (or other) application requires a brighter light source, several fibers can be bundled together. For example, if it is hexagonally close-packed, a seven-fiber bundle still would be only a tiny 150 microns in diameter. The use of the fiber somewhat reduces speckle. I was told, however, that other "mechanical" means are used to reduce speckle in the rear-projection application.

It is interesting to note that the output of the lasers are polarized, but, after traveling through the fiber, the light is completely depolarized. Since the LCoS panels need polarized light, the light out of the fiber is split into two components with orthogonal linear polarizations. Each polarization is directed to one of two LCoS panels. This approach is highly efficient in that nominally 100% of the light emitted by each of the lasers is utilized. Each LCoS panel operates in the field-sequential mode producing red, green and blue images synchronously with modulation of the laser outputs. One LCoS produces a right-eye image and the other a left-eye image. The images are optically combined and leave the projector through a single projection lens.

In 2D operation, the same image is presented on each LCoS, producing a double-brightness image at full resolution. In the 3D mode, a full-brightness image is produced that is also at full resolution. Significantly, both eyes receive time-sequential images continuously. This would not be the case in a 3D system based on shutter glasses in which the eyes receive images alternately. The ability to rapidly modulate the lasers, which are always on, also enables other image-improving processes, such as blur reduction and dynamic contrast enhancement.

The prototype rear-projection system had a diagonal of 100 inches and an aspect ratio of 16:9. The "special" plastic screen had a gain of 1 and provided a wide viewing area. With a power consumption of 150 Watts, 300 nits were presented on the screen. Reportedly, 7 to 8 ft/Lamberts of circularly polarized light were presented to each eye. The contrast ratio was stated to be about 1000:1. The prototype projector was fabricated in part by cannibalizing several commercially available projectors for optical and electrical components.

I was provided ample opportunity to study both 2D and 3D images produced by the prototype system. There were a variety of deficiencies in the image, which I decided were reasonable to ignore. Spots, specks and scratches can be forgiven in any hand-built prototype system. In addition, I observed that the bright- and dark-state uniformity were not great, and neither was the contrast ratio. Once again, I found it reasonable to cut HDI some slack regarding these image attributes. I believe that a professionally designed enclosure, like the one that would be used in a commercial product, would successfully address these issues. I did not observe any pixelation in the image.

The red and blue colors were highly saturated. Interestingly, the green was not so saturated. I found this surprising in a laser-based projection system. When I made this observation, I was told that the saturation of the green depended on image content. Although that was somewhat cryptic, Jansson and Sandberg did not provide further explanation at the time.

The HDI business model also is currently under development with licensing being one option under consideration. We wish them all the best.