| ??? 05/05/09 05:27 Read: times |
#165053 - I wouldn't be surprised if Chico has a working projector Responding to: ???'s previous message |
I love you.
But did it take you 50 posts to figure out that Chico may not be interested in video or static pictures or presenting a full page of text? I would think that 99.9% of the chances people have of seeing a laser projector is the XY-controlled kind seen at many discos. So the probability that Chico did get interested about the concept and did start this thread because he wanted to implement a XY-projector (not aiming at video or still images) should be quite high. Here is an experiment for you Richard. Take your laser pointer and aim it at a wall. Look at the nice strong dot. Now try to move the pointer in circles as fast as you can, slowly increasing the diameter of the circles. Do you still clearly see the traces on the wall? The traces will very quickly dim down depending on the diameter of the circles, caused by the much lower power levels/area. The bigger diameter you draw, the lower power densities, and the harder to see. As long as the diameter isn't too large, you will see the individual traces even if they are not repeated, and even if the time spend at every point is very short. What is required is enough power/mm of the trace. To do it a bit more controlled, you can take a Dremel or similar and mount a small piece of metalized plastic at an angle. Now rotate this tiny mirror while shining on it - you will get a perfect circle or ellipsis with good overlapping of the consecutive traces. Now watch the intensity of your circle as you either increase the distance between the Dremel and the wall, or you increase the angle of the mirror. The above experiment is very quick, and can be used to deduce how useful an existing laser source will be in a laser projector design. If the traces gets too dim to be useful, then there will be no use starting to work controlling any mirrors. Note that this test only helps for a XY-driven projector. To get an idea about a raster projector, you would have to activate your rotating mirror too and instead try to draw many concentric circles of varying diameter to simulate the additional path length in the raster display. What you can see is that the rotating speed (fps) of the Dremel will not affect the perceived intensity of the circle (unless you combine the Dremel with the rotating mirror). But the circumference of the circle will affect the intensity. The only way to compensate for the intensity loss is to switch to a stronger laser, or a laser with better color. The optimum would be a green laser, since a green laser has up to 80 times better visibility than the first-generation red laser pointers. Richard said:
So long as the output power of his laser is not known, the speed requirement isn't readily discernible. Speed isn't the matter here. Total path length for one frame is what will control the intensity of the display. And fps and repeatability, also called registration, is what will affect if our eyes will see a flickering display or not. But a 50fps or a 200fps display will project the same amount of energy in each part of the trace and will seem to have mostly identical intensity, even if the 200fps display will require the laser beem to scan at four times higher velocity, i.e. each individual frame will be contain one quarter of the energy. Richard said:
How many kW of laser power is required to produce a 33 ms uniformly persistent dot, as perceived by the human retina, when struck by the laser for 62 ns? ... at a distance of, say, 10 meters from the source, and the viewer ... on a matte-white painted wall? You do realize that a single 62ns pulse (not repeated continuously) would probably not be seen at all, since our brain would filter the information away as random noise. Unless the power is strong enough to produce significant ghosting in the eye, in which case you will be able to look in other directions and still see a dot in the middle of your view. And the next step up is that you have so strong laser that you will never see any light because you will permanently blind that part of the eye. But a very important factor when talking about visibility is the wavelength of the laser, since the sensitivity of our eyes is very much affected by wavelength. The next thing that is important is how glossy the surface is, since that will affect the percentage of light that will affect the size of the reflected lobe. The special case is a perfect mirror, where 100% of the light comes back. For a normal white wall, most of the laser energy will scatter in all directions, which is the reason why you can see the laser dot from multiple angles. But don't talk about kW-class laser output unless you are expecting to light a video display for an arena. My video projector would - wild guess - probably manage 10W of projected optical power when displaying white. A two meters wide image projected on a diffuse surface is quite dim when viewed from 3m distance in daylight, but with the room darkened, you will blink when seeing a sunrise in a movie. A laser projector would get away with less output power, since the laser light will give a higher contrast. Projecting on a white wall means that our "black" is no darker than that white wall is when the projector is off, so the contrast is very important. For a laser projector, we might possibly talk about 3W required output. Sounds tiny, but still quite problematic if you look at the price of LED-lasers or DPSS lasers. Richard said:
So long as the output power of his laser is not known, the speed requirement isn't readily discernible. That sounds like you mean sweep speed. Don't focus on the sweep speed. Focus on surface area covered, and the length of the path that the beam is swept. The sweep length is very high in a raster dipaly, since the beam has to be swept past all pixels even if the pixel should be off. A XY projector should - at worst - have 50% dark time - if drawing images but could get down to 0% dark time if used for drawing Lissajous curves, where there are no discontinuities the laser never has to be turned off. The issue with a XY projector is the rotation speed and precision of the mirrors, which affects how complex graphs that may be drawn. With long throw distances, you can trade precision for speed, since you would require a smaller modulation angle to get a specific size of the drawing region. Note however, that most XY projectors are used to draw simple patterns in which case almost any home-made solution will be good enough. Richard said:
When Chico lets us know what his inputs are, and what he's really trying to do, in some detail, perhaps we can stop all this speculation and address his problem. I imagine, however, that he's not reading much of this because it really doesn't address his interests/needs. Why do I think that Chico has left the thread? Because he could build the mechanics of the simplest form of XY projector in less than an hour. If settling for AC-driven modulation, it would then be ok to just take sound from the soundcard of a PC to start playing with that sub-hour design. AC-driven output would not allow absolute positioning, but Lissajous output works fine. Required material would be two powered PC speakers, some metalized plastic, a bit of glue and a PC to emit sound. Very red-neck, but working quite ok and buildable by almost anyone on this forum. |
