distance of light from fresnel

[part2wanksta]

is there a certain distance that the light has to be from the fresnels. i am using the fresnels from a ohp. they are split.

[SupraGuy]

Shouldn't it be the focal distance of the fresnel?

As I understand it, the purpose of the first fresnel is to make the radial light coming from the light source into parallel light going through the LCD.

Therefore, the light source should be at the focal point of parallel light going the opposite direction.

Similarly, the projection lens should be very near the focal point of the second fresnel in order to capture as much as possible of the light that has come through the LCD panel and project it onto the screen.

What I'm not so sure of is this: If you have a parabolic reflector, do you need the first fresnel lens at all, since you should have a (mostly) parallel light source to begin with?

[part2wanksta]

how do i find out what the focal length is

[Guy Grotke]

If you have a parabolic reflector that has a slightly wider diameter then the LCD, and is being used correctly, then you don't need the condensor fresnel. You still need the field fresnel to converge that parallel light into the projector lens.

You can measure the focal length of individual (ie split) fresnels just like you measure it for any other lens: Take it out at midday, and face the rough side to the sun. Move the lens down toward the ground until you get the smallest possible image of the sun. Measure the distance from the lens to the image spot on the ground. Try not to set anything on fire!

Put the condensor fresnel at it's focal length from the lamp arc to start. Then once you see where the projector lens has to go to get your screen image focussed, you can adjust the lamp arc to fresnel distance a bit to get the image of the lamp arc focussed into your projection lens. You should not have to move the condesnor fresnel more than a centimeter or two. If that doesn't get it, then you may have to put the field fresnel on the other side of the LCD to get the arc image into the lens.

[Megaman]

I'm wondering. If a parabolic reflector is used, will not the light that comes directly from the light source (without beeing reflected) cause errors? That light will not come in parallel to the other light that comes through the lcd and the fresnel, and thus the light going through a specific lcd pixel will be placed in different parts of the final image.

I'v attached an image that hopefully will make you see what I mean.

Using a reflector in front of the light source so that only reflected light will come to the lcd would make a dark spot in the center of the image where the reflector prevents the light to pass. Placing the light source a little bit farhter away from the parabolig reflector would "fix" this, but then the light that comes to the lcd won't be parallell.

And also, doesn't this problem exist in most setups that uses reflectors?

[Guy Grotke]

Yes, using a parabolic reflector with an unshielded lamp will give you a lot of direct light that may cause a fuzzy image. (Most of that light will not even end up at the projection lens.) But you can add a tiny spherical reflector to send the useless light back through the lamp arc to the parabolic reflector. Lots of commercial projector lamps have such a tiny reflector built into the lamp. It would make a hole in the center of the light, except for the fact that the arcs of the lamps we use are so big that the light is never really all going in parallel.

Spherical and elliptical reflectors don't have this problem when they are used correctly. But still do have the problems associated with the large lamp arc. A really good projector lens will focus all the different rays passing through a pixel to a single point on the screen. A bad projector lens will send them to slightly different parts of the screen to give you a blurry image. You can get a good image with a small lamp arc or a good lens. If you have neither one of these, you will not be happy with the image.

[Megaman]

How can a lens compensate for the size of the lamp arc?

[Megaman]

How large are the arcs in the HQI lights? How large is an arc that is too large to be useful?

[ace3000_1]

How large is an arc that is too large to be useful?

Anything over 5mm. I wouldnt say useful, efficent is a better word.

Trev

[Guy Grotke]

>How can a lens compensate for the size of the lamp arc?

If the arc length was 0 mm, then you would not even need a projection lens: Every pixel on the LCD would be lit by a ray from a single direction. The condenser system would direct that ray to one point on the screen. In effect, this would be like using a pinhole lens. Or without a condenser system, the light would form a perfect image at any screen distance just by simple projection of the rays. With a condensor system and a projection lens, the ray from each pixel would strike only a single point on the lens so the lens would not have to focus anything. The screen image might be distorted by different refractive bending by different parts of the lens, but all of the light through an LCD pixel would end up at a single point on the screen. Even with a very bad lens, you would get a sharp image.

If the arc length is very small, (ie. 1mm), then the cone of rays passing through a single pixel will strike a very small area of the projection lens. The amount of aberration and distortion over that small area will be very low, even in a bad lens. So the rays will be focussed well on a very small area of the screen.

If the arc length is long, (ie. 24 mm), then the cone of rays passing through a single pixel will strike most of the lens. If it is a bad lens, then the result will be light scattered all around the desired pixel position on the screen because all of the lense's aberrations and distortion will affect it. If it is a good lens, then the rays will all be focussed to one point on the screen.