Hello all,

I am wondering if anyone here has attempted to build a laser backlight. I know it is possible to create a white laser beam by combining 3 beams from a R, G and B source. So I am wondering how difficult it would be to take it 1 step further and try one of the following methods:

1 - Force beam divergence in both X & Y axis to form a trapezoidal beam.
2 - Force beam divergence is either X or Y, then scan the opposite axis using a servo tilted mirror.
3 - Scan horizontally and vertically like the electron gun in a CRT, also using servos.

I believe that the 2nd method would be the easiest to implement, but I am hoping that the 1st method is possible since it is the simplest as there would be no moving parts. I am thinking that the servo's would have to move at an extremely high rate, although they wouldn't be turning very far.

Just wondering what you guys think. I have provided a dumbed down drawing of the first method.

1 is the RGB Lasers, 2 & 3 are mirrors, 4 is the Divergence optics package, 5 is the LCD

interesting , but whats the benifit

The best light source for a projector is a point light source. A laser beam would be ideal, but I really don't think it's doable in a DIY setting.

Laser = Hi lux, low lumen.

Even the lasers that they use for the big laser shows at concerts aren't putting out a lot of LUMENS. They're very bright, high lux lights, but even if you have 1000 lux over an area the size of a pinhead, that's very small in terms of lumen output.

A 30 watt laser is a cutting device, Not for steel, but can be used for most plastics, but still won't produce more than a handful of actual lumens. There may be 20,000+ lux in there, but because the area is so small, the total number of lumens is very small. You'd need a laser normally used for cutting steel to do this. Dollar store lasers aren't going to produce enough light. It would be more than bright enough over the area that it covers, but you'd need so many to cover the area...

A good idea. Had me thinking for a while on implementation, until the whole total output thing got in the way...

That's where the scanning comes into play. A CRT monitor only lights up one pixel at a time. But your eye retains the image for a small time interval. A CRT simply scans the entire screen within that time interval, so that the last pixel on the screen is drawn before the first pixel leaves your eye. By itself, a CRT's electron gun is only bright enough to light up a single pixel on the screen. But through the magic of scanning, we create the illusion of a fully illuminated screen with this weak light source. If we apply this concept to lasers, there's a lot of potential benifits:


Advantages of a scanning laser LCD projector:

1: Extremely low energy requirements -- imagine a projector with lower energy requirements than an iPod.

2: Perfectly even light -- Ordinary light suffers exponential falloff with distance. That's what causes the vignetting effect in our projectors -- The distance between the bulb and the corners of the screen is farther than the distance between the bulb and the center. Though the difference in distance is small, the light falloff is exponential, and therefore noticeable. On the other hand, a laser's coherent light doesn't suffer from exponential falloff, thus vignetting would be eliminated.

3: No lenses -- A scanning laser could project an image with exactly zero lenses. Put the laser behind the LCD, and rock and roll!

4: No focusing -- A laser-projected image is always in focus at any distance. (You can project at some pretty insane distances, too!)

5: No ballast

6: No bulb

7: No heat

8: No light leakage

9: Compact size -- In theory you don't need any lenses, but if you set the scanning scanning FOV to 160˚ or so, you could use an appropriately-speced collimator, and end up with a relatively flat, compact, projector.

10: Cheap!!!


Can we really build this thing?

Believe it or not, this technology already exists --Here's and example: (link).

I've been looking into creating a DIY scanning laser. I've seen various tutorials for building an inexpensive white laser. They typically use red and green laser pointers, and scavenge a blue laser from an HD-DVD player. So that part of the problem is solved. Now, we just need to come up with a scanning mechanism. So far, I've managed to make a line projector with a small motor and mirrors. It's just 4 perpendicular mirrors glued to a ball, and stuck to the shaft of an electric motor. But, it solves the problem of horizontal scanning. If we spin this motor fast enough, we could then combine it with a second rotating mirror such that the second mirror takes each complete horizontal line generated by the first mirror, and scans it vertically. The scanning radius would be determined by the number of mirrors used in each motor. In my 4-mirror example, the scanning radius would be 180˚. This corresponds to an extremely short throw distance, but it's perfectly fine for the sake of experimentation.

Anyway, that's the plan.

HI
Nice Idea. In CRT there are coils which provide deflection to the beam. Now we should think how to deflect the laser beam electronicaly ( its obvious that canot deflect them mechanicaly, due to the hight deflection speed requirements)
Keep discovering

If you have Red Green and Blue lasers why even bother with a LCD.
Just shine it directly on a surface. No need to focus either.
If think you can take a standard VGA connection and get your RGB signal from that.
Deflecting the laser over two axis with the right timing would be the hard part.
I think it is not beyond DIY.

Edit:
I found this: http://en.wikipedia.org/wiki/Acousto-optic_modulator