heya buddy, ok the lamps arc needs to go at the reflectors focal, im sure you already know this. To help you find the focal of the reflector or atleast the right spot the lamps arc tube in the reflector should be reflecting back at the same size, So in other words you should only see the bulbs arc tube and not the reflection. I hope that made sence, but thats the easy way to do it.
With the condenser place it 10mm from the bulb. Move it slowly closer towards the bulb to get your image brightness even and thats about it. The shorter the focal is on a condenser the closer it has to go towards the bulb, also the longer the rear frensel has to be. All we do with a condenser is just guide the light to the lcd, thats basically it. Depending on what type of specs your condenser has will depend on its outcome. Normally a condenser will make the world of difference and i dont recomend not using one.
Trev🙂
With the condenser place it 10mm from the bulb. Move it slowly closer towards the bulb to get your image brightness even and thats about it. The shorter the focal is on a condenser the closer it has to go towards the bulb, also the longer the rear frensel has to be. All we do with a condenser is just guide the light to the lcd, thats basically it. Depending on what type of specs your condenser has will depend on its outcome. Normally a condenser will make the world of difference and i dont recomend not using one.
Trev🙂
focal, focal point, center of curvature
There is a little confusion about which point in a spherical reflector we are talking about, because all of these terms get interchanged.
When you represent a spherical reflector in two dimensions (ie. on paper), you can draw it with a compass as part of a circle. The center point, where you put the pointy leg of the compass, is the "center of curvature". Then there is another point, halfway between the center of curvature and the reflector surface, that is called the "focal point".
When you look at a spherical reflector in a catalog, they list the diameter and the focal point. (Some also list the center of curvature.) If you are using the reflector as a telescope mirror, then you care about the focal point. This is where parallel light from a very distant object (ie. a star) will focus. But that focus goes way off if you try to use much more than 1/20th of a sphere. That's why telescope mirrors are 4-10" in diameter but have focal lengths of 1-2 meters. (Parabolic reflectors work better as telescope mirrors, but sphericals are much cheaper to make.)
If you are using a spherical reflector in a projection system, then you care about the center of curvature. When light passes through this point and hits the reflector, it will be reflected right back to this point. So if you put the lamp arc there, all the light that hits the reflector will come back through the lamp arc. So it will look like it is about twice as bright.
Of course, our lamp arcs are not point sources, so you can't put the ends of the arc at the center of curvature. But light from each end is reflected back to the other end of the arc. So it still reinforces the arc brightness.
All of that light diverges in a cone shape to strike the lower fresnel, where it is refracted to be parallel. ( The light is actually emitted in all directions by the bulb, but we only care about the portion of that light that actually hits the fresnel.) If you include a condensor lens, then it can be placed very close to the bulb, so it captures more of the light than would hit the fresnel without it. For a small panel like a 7" Lilliput using a 200 mm lower fresnel, a good condensor setup should more than double the light sent to the fresnel.
There is a little confusion about which point in a spherical reflector we are talking about, because all of these terms get interchanged.
When you represent a spherical reflector in two dimensions (ie. on paper), you can draw it with a compass as part of a circle. The center point, where you put the pointy leg of the compass, is the "center of curvature". Then there is another point, halfway between the center of curvature and the reflector surface, that is called the "focal point".
When you look at a spherical reflector in a catalog, they list the diameter and the focal point. (Some also list the center of curvature.) If you are using the reflector as a telescope mirror, then you care about the focal point. This is where parallel light from a very distant object (ie. a star) will focus. But that focus goes way off if you try to use much more than 1/20th of a sphere. That's why telescope mirrors are 4-10" in diameter but have focal lengths of 1-2 meters. (Parabolic reflectors work better as telescope mirrors, but sphericals are much cheaper to make.)
If you are using a spherical reflector in a projection system, then you care about the center of curvature. When light passes through this point and hits the reflector, it will be reflected right back to this point. So if you put the lamp arc there, all the light that hits the reflector will come back through the lamp arc. So it will look like it is about twice as bright.
Of course, our lamp arcs are not point sources, so you can't put the ends of the arc at the center of curvature. But light from each end is reflected back to the other end of the arc. So it still reinforces the arc brightness.
All of that light diverges in a cone shape to strike the lower fresnel, where it is refracted to be parallel. ( The light is actually emitted in all directions by the bulb, but we only care about the portion of that light that actually hits the fresnel.) If you include a condensor lens, then it can be placed very close to the bulb, so it captures more of the light than would hit the fresnel without it. For a small panel like a 7" Lilliput using a 200 mm lower fresnel, a good condensor setup should more than double the light sent to the fresnel.
reflector materials
There is a lot of data availible on the maximum reflectivity of various materials. This assumes a perfect spectral thick surface, not a matte finish.
stainless steel: 65% (soup ladle)
rhodium: 84%
aluminum: 88%
gold: 96.7%
silver: 98%
All but the stainless steel value are right out of optics catalogs, so these are as good as they get. These catalogs also have some info about the ability of the different materials to reflect light at different wavelengths (dichroic behavior):
Aluminum reflects 88% of visible light, but 90% of IR, which means it is what we DON'T want!
Gold reflects well from 1000 down to 700 nm, but then drops off gradually to level out at 35% reflectivity below 500 nm. (As a reference: Blue = 435-500, Green = 520-565, Red = 625-740) So gold only reflects blue light at 35%! Where gold shines (literally) is in the long IR wavelengths. This is why astronomers use gold mirrors for IR photography, but for the visible light we want in a projector, gold is a horrible reflector. The non-linearity is why a gold reflection looks so red.
Silver has an extremely flat 98% reflection spectrum from 1000 nm down to about 450 nm. By the shortest blue wavelengths, it is down to about 96%. Then it drops rapidly to under 10% by 320 nm. So it reflects all the visible light we want, and cuts out a lot of the UV we don't want.
Note that these are thick surface values. The thin coatings used in expensive dichroic reflectors are designed to cancel out light near a particular wavelength, so they can do tricks like reflecting 90% of visible but passing 90% of the IR. But since a concave cold mirror costs almost as much as a new economy-model projector, I don't think they will be found in many DIY projectors!
There is a lot of data availible on the maximum reflectivity of various materials. This assumes a perfect spectral thick surface, not a matte finish.
stainless steel: 65% (soup ladle)
rhodium: 84%
aluminum: 88%
gold: 96.7%
silver: 98%
All but the stainless steel value are right out of optics catalogs, so these are as good as they get. These catalogs also have some info about the ability of the different materials to reflect light at different wavelengths (dichroic behavior):
Aluminum reflects 88% of visible light, but 90% of IR, which means it is what we DON'T want!
Gold reflects well from 1000 down to 700 nm, but then drops off gradually to level out at 35% reflectivity below 500 nm. (As a reference: Blue = 435-500, Green = 520-565, Red = 625-740) So gold only reflects blue light at 35%! Where gold shines (literally) is in the long IR wavelengths. This is why astronomers use gold mirrors for IR photography, but for the visible light we want in a projector, gold is a horrible reflector. The non-linearity is why a gold reflection looks so red.
Silver has an extremely flat 98% reflection spectrum from 1000 nm down to about 450 nm. By the shortest blue wavelengths, it is down to about 96%. Then it drops rapidly to under 10% by 320 nm. So it reflects all the visible light we want, and cuts out a lot of the UV we don't want.
Note that these are thick surface values. The thin coatings used in expensive dichroic reflectors are designed to cancel out light near a particular wavelength, so they can do tricks like reflecting 90% of visible but passing 90% of the IR. But since a concave cold mirror costs almost as much as a new economy-model projector, I don't think they will be found in many DIY projectors!
Ok well there isnt any confusion here as the parabolic type spherical reflectors are totally different all together, the same goes with a gold mirror and gold dichoric vs silver dichoric and aluminimised surfaces ect. Telescope reflectors are no good for us, they have longer focals for starters as with their coatings are also different. The parabolics in projectors are the deep design, not the flat. The parabolics in telescopes are the flat sperical reflector type run at parabolic. Telescopes also have large mirrors to capture more light. The bigger they are, the more light you gather, and the more resolution is obtained, hence why deep space scopes have 30inch + sized mirrors.
I dont know where you are getting your prices from guy, but you can buy an eliptical dichoric coated reflector suitible for the CDM-T bulb for $69 at Edmund optics. Have a read around on thier site, they also explain coatings for various aplications.
My advice is to forget telescopes, they are a different ball game altogether, work on projectors only or you will get confused.
You talking about these: (see pic) Take note of the formula, its .5 of the radious of the sphere. Thats what you do to get parabolic rays out of a spherical reflector. Run the source at the full radious and you get all rays going back to its source. The calculation is also done on the R1 surface not the R2.
Trev🙂
Attachments
This is a pic of the spherical reflectors we use, (and the right ones), note the stronger curve for a shorter focal. To run any spherical reflector to have the light going back to its source, you go on the R1 spec only, otherwise you will have parabolic rays, (what you dont want with a condenser).
Trev🙂
Trev🙂
Attachments
edmonds reflectors
Those are great prices on the cold elliptical reflectors, and I think I will order one to play with. But where are the cold SPHERICAL reflectors? Those are the ones we need!
I have looked at several optics manufacturers that have online catalogs. The prices for cold spherical reflectors were all about the same: Way too expensive! The $517.40 US price is quoted is from Rolyn Optical, for one with 65 mm diameter and 43 mm center of curvature.
The ironic thing here, is that you don't really need a cold elliptical reflector to protect the bulb: In a normal elliptical reflector, all the light (including IR) goes away from the bulb to the second focal point. At that second focal point, it would be very easy to use a flat hot mirror or a flat cold mirror to seperate the visible light from the IR. I can get either of those for $5 or less.
Those are great prices on the cold elliptical reflectors, and I think I will order one to play with. But where are the cold SPHERICAL reflectors? Those are the ones we need!
I have looked at several optics manufacturers that have online catalogs. The prices for cold spherical reflectors were all about the same: Way too expensive! The $517.40 US price is quoted is from Rolyn Optical, for one with 65 mm diameter and 43 mm center of curvature.
The ironic thing here, is that you don't really need a cold elliptical reflector to protect the bulb: In a normal elliptical reflector, all the light (including IR) goes away from the bulb to the second focal point. At that second focal point, it would be very easy to use a flat hot mirror or a flat cold mirror to seperate the visible light from the IR. I can get either of those for $5 or less.
Ya i agree im thinking in scooping up one for myself aswell, the prices there are reasonably high for diyers but atleast at Edmunds they sell top quality gear. They sell the dichoric spherical reflectors aswell you need to look in their cat under mirrors, 1/4 wave focusing from memory.
Trev🙂
Edmonds reflectors
>They sell the dichoric spherical reflectors as well
Not that I can find. I see aluminum, enhanced aluminum, gold, and uncoated mirrors. In other sections I see flat cold, hot, and extended hot mirrors. No cold spherical reflectors anyplace.
I suppose you could buy an uncoated spherical mirror, and then pay them to put a cold mirror coating on it. But I don't see any uncoated mirrors that have a good shape for a projector.
Just by some quick calculations, I think the small elliptical reflector could send light to a 200 mm fl fresnel for a 7" LCD, or a 220 mm fl fresnel for an 8" LCD, or 330 mm fl fresnel for a 12" LCD. The big one could send light to a 650 mm fl fresnel for a 17" LCD.
>They sell the dichoric spherical reflectors as well
Not that I can find. I see aluminum, enhanced aluminum, gold, and uncoated mirrors. In other sections I see flat cold, hot, and extended hot mirrors. No cold spherical reflectors anyplace.
I suppose you could buy an uncoated spherical mirror, and then pay them to put a cold mirror coating on it. But I don't see any uncoated mirrors that have a good shape for a projector.
Just by some quick calculations, I think the small elliptical reflector could send light to a 200 mm fl fresnel for a 7" LCD, or a 220 mm fl fresnel for an 8" LCD, or 330 mm fl fresnel for a 12" LCD. The big one could send light to a 650 mm fl fresnel for a 17" LCD.
Yeah i had a look earlier tonight too and i couldnt find any, maybe they stop doing them, i know they used to have them. Ive noticed that they have also cut back on the elliptical reflectors too, they used to carry a larger range. Ive got a few contacts in Germany that do dichoric sphericals tell me what you need and ill see what i can do, but no promisses.
Yeah same as mine. Im going to buy the small one and whack a CDM-T in it, trust me, it will be just as bright if not brighter then a 250w in spherical setup condensing the arc.
What i like about the eliptical is the fact that 90% of the light is going to be Cold light coming from the source, (going on the beam pattern of the CDM-T). Also the light will be at a much higher colour temp which is going to make things not only cheap and easy but also have the correct contrast and brightness ect. I dont think we will need a condenser on it going on the brightness from the pro parabolic reflector i have. It should be bright enough without one, less distortions and money anyway.
With the big one you should be able to run a HQI 250w in it, a small modd on the bulb might be in order though, and i dont realy like the extra long focal.
Trev🙂
For confirmation's sake, then, I want to put the arc at the center of curvature for the reflector and start at about 10mm out from the outside of the HQI bulb on my condensor then? and work my way closer from there?
Explain to me why I can't run the light parrallel from the reflector to the condensor? It will cause a hot spot in the center with dark edges, is that it? I'm trying to understand everything B4 I put everrything together because I dont have a lot of materials to play around with. So I need to get my design as close to accurate on paper or on a bench as I possibly can.
It would be fun to watch, at least...
Kind of like putting a firecracker in a model ship, just to see it blow up.
If the distance to the fresnel matched the focal length of the condensor lens, then sending the light parallel into the condensor would focus all of that light in one tiny spot at the center of the fresnel. You wouldn't have a just a "hot spot": You would have a melted spot! And then it would destroy your LCD...
If the light goes into a lens parallel, then it is focussed at the focal length on the other side. If the light comes from a point at the focal length, then it comes out parallel on the other side. In a condensor lens, we make the light come from a point closer than the focal distance, so it is "less diverging" but not focussed on the other side. When you put the bulb arc at the center of curvature of a spherical reflector, then the condensor "sees" all of the light, direct and reflected, as coming from the bulb arc.
Kind of like putting a firecracker in a model ship, just to see it blow up.
If the distance to the fresnel matched the focal length of the condensor lens, then sending the light parallel into the condensor would focus all of that light in one tiny spot at the center of the fresnel. You wouldn't have a just a "hot spot": You would have a melted spot! And then it would destroy your LCD...
If the light goes into a lens parallel, then it is focussed at the focal length on the other side. If the light comes from a point at the focal length, then it comes out parallel on the other side. In a condensor lens, we make the light come from a point closer than the focal distance, so it is "less diverging" but not focussed on the other side. When you put the bulb arc at the center of curvature of a spherical reflector, then the condensor "sees" all of the light, direct and reflected, as coming from the bulb arc.
Test Bench
Guy, and ACE,
Here's what I come up with so far, using the condensor & reflector I mentioned earlier. The thing is, in the bench, I'm using a 75 watt incandescant bulb in a desk lamp. I didn't line anything up with measurements but I just eye-balled everything in a line and on center. (I am pretty good at this, I used to be a draftsman.)
Here's the test bench
Guy, and ACE,
Here's what I come up with so far, using the condensor & reflector I mentioned earlier. The thing is, in the bench, I'm using a 75 watt incandescant bulb in a desk lamp. I didn't line anything up with measurements but I just eye-balled everything in a line and on center. (I am pretty good at this, I used to be a draftsman.)
Here's the test bench
Attachments
test bench
Eye-balling the alignment is okay for the testbench experiments.
I see a big circle of light around your projection lens. That tells me your fresnel(s) are not making a nice image of the lamp arc (or in this case, the light bulb) at your projection lens. All that light falling outside the projection lens is wasted.
Eye-balling the alignment is okay for the testbench experiments.
I see a big circle of light around your projection lens. That tells me your fresnel(s) are not making a nice image of the lamp arc (or in this case, the light bulb) at your projection lens. All that light falling outside the projection lens is wasted.
Alignment
Gents,
I would recommend this trick for alignment and finding your focus:
Dip a sharpened pencil in white paint and let it dry. Make an adjustable clamp to hold it parallel to the table (clothepin and base or whatever).
Get a laser pointer pen and duct tape it even with a long level. again make a height adjustable clamp.
Both bases need to roll side to side and adjust up and down.
Put the laser and level where your objective lens is and move it around square to the reflector. If the tip of your white pencil is always red from the laser dot all is good. Otherwise some part of the system is out of alignment or your focus pencil tip is not at the focus.
😀
Gents,
I would recommend this trick for alignment and finding your focus:
Dip a sharpened pencil in white paint and let it dry. Make an adjustable clamp to hold it parallel to the table (clothepin and base or whatever).
Get a laser pointer pen and duct tape it even with a long level. again make a height adjustable clamp.
Both bases need to roll side to side and adjust up and down.
Put the laser and level where your objective lens is and move it around square to the reflector. If the tip of your white pencil is always red from the laser dot all is good. Otherwise some part of the system is out of alignment or your focus pencil tip is not at the focus.
😀
Guy,
RE: The big circle of light,
The lamp in the test bench is a white frosted incandescant 75W light bulb. That makes the arc about 45-55mm I didn't measure, but that's what it looks like. I'll set up with the HQI on Wednesday.
RE: The big circle of light,
The lamp in the test bench is a white frosted incandescant 75W light bulb. That makes the arc about 45-55mm I didn't measure, but that's what it looks like. I'll set up with the HQI on Wednesday.
Re: Alignment
Whats wrong with marking out all of your centers on a plan? Seems more acurate to me 😉
Trev🙂
Me2! said:
Whats wrong with marking out all of your centers on a plan? Seems more acurate to me 😉
Trev🙂
awwww but Trev thats no fun mate LoL just too easy and accurate! (allthough its how I do it to) wheres your Aussie sense of adventure gone bud 😛 Laser pointers, markers, levels and clamps mate thats where it's at!(take me half a day to knock the contraption up though LoL Then id prolly take me eye out with the laser pointer) You could even make a little sticker for the front of your PJ when its finished "Laser Aligned Optics" hahaha
Its a good idea Me2! but Trev's right it is quicker, easier and accurate to simply draw it out on the plan mate, great thought/idea none the less very clever thinking mate.
- Status
- Not open for further replies.