diyAudio Forums Archive > Top > The Moving Image > DIY Projectors N00B needs help - Click HERE for Original Thread lone_wolf_84 Hello, Where can I go to find information about what triplets and doublets are, what condensors are, what "focal length" i need, etc etc. I have been looking and cant find a definitive source for this information. Thanks Guy Grotke There are lots of posts here about projectors people have built, and there are other forums on DIY LCD projectors. Read them! Some even have online stores or group buys with pretty cheap (and tested) components. Some have lots of public domain design examples, some sell plans for specific designs. www.diybuildergroup.com has a "light engine tool" free excel download you can use to see what lenses, panels, and projection distances give you a particular size image. Once you figure out what focal length lens to use, you can look for that as a singlet (single lens), a doublet (assembly of two lenses, selected to correct chromatic aberration), a triplet (assembly of three lenses, selected to correct more aberrations), or a multi-element lens. Try to find one at least 65 mm in diameter, and don't buy a curved-CRT projector lens to use with a flat LCD. Some people claim to get adequate results with a $6 singlet from surplus shed. Others spend hundreds. I suggest you start simple and then improve it. A condensor is usually a Double Convex (DCX) lens about an inch from a light source, used to capture a wide arc of that light and bend it all into the projection path. Since the lamp is closer than the condensor's focal length, it works in magnify mode (as opposed to projection mode). Here is a pretty good optics information website you can peruse: http://hyperphysics.phy-astr.gsu.ed...lenscon.html#c1 lone_wolf_84 ok, thanks. So a condensor lense, is that what a fresnel is? And what i ment by asking what a triplet was i ment what kind of lenses it used, same thing for the doublet. Thanks ace3000_1 Heya buddy, thats actually incorrect, they are normally PCX, they come in pairs or singles, (singles for what we need). Here is a correct PCX condenser setup run in pairs. Lone_wolf, have you seen the search page? use the forums search function to find the information you want, this forum covers prety much everything you need ;). Trev:) Guy Grotke These DIY designs actually have two different, but cooperative optical systems: 1) The light engine, that sends light through the LCD panel in such a way that most of that light strikes the objective lens. 2) The LCD panel, objective lens, and screen. You can project an image with just #2, but it will be very dim. This is the same as the 100" TV internet scam. In #1 most designs use a big, short fl converging lens (ie. a 100 mm diameter 200 mm fl DCX or PCX) very close to the lamp to capture 33% of the output (or closer to 66% with a spherical reflector behind the lamp) and converge it to a lower fresnel the size of the LCD, a lower fresnel (ie. about 220 mm fl) to turn those diverging rays into a parallel beam that sends light through each pixel of the LCD, and an upper fresnel (ie. about 330 mm fl) to converge the resulting "modulated" rays toward the objective lens. Other common designs put both fresnels together below the LCD, so the rays are already converging before they go through the pixels. In one sense, all three light engine lenses are acting as a condensor. But I was referring just to the first lens in the light engine. Cheap fresnels are made from PVC, good ones are made from acrylic, and "way more expensive than we need" ones are made from glass. In #2, an objective lens is used to focus the image of the LCD on the screen. You can do this with a simple DCX, PCX, or positive meniscus lens that has the right focal length. There are also more complex multi-element lenses that try to correct various aberrations, but those aberrations may not even be noticable in some projectors. (That's why I suggest starting cheap and simple, then adding improvements as needed.) Two aberrations to look for are chromatic aberration (red and blue fringes near the edges of the image), and spherical aberration (bulging edges). You can correct chromatic ab by upgrading to an achromat: Generally a crown glass DCX + flint glass PCV with a focal length ratio of 38:59 (optimal because of the different refractive indexes of the materials), and of course they have to have a combined focal length that is usable in your situation. For example, I am using a 4" diameter 200 mm fl DCX and a -308 mm fl PCV to get 570 mm combined fl, which is as close as I could get with cheap lenses from surplus shed. ($7.50 total cost) These correct the "prism effect" because the crown glass DCX bends the colors differently in one direction, and then the flint glass bends them differently in the other, so they recombine on the screen. I did this because I needed a combined fl I could not buy on the cheap. But you might be able to use an OHP objective lens, or one of the triplets you can buy from various DIY projector websites. I have seen 320 mm fl triplets for $29. lone_wolf_84 hi, thanks for the replys, yes i have searched the forums, alot of forums, but either I dont know what to search for specificly or something, cause ive learned more in the three post's on this thread then I have the last 3 days of reading threads with mixed jargon in it. While i got your nice peoples attention (Which i really appreciate btw) i have an old bell and howell 3870a overhead projector, how many parts could i salvage fromt this? And i would be extatic if some could tell me or show me a link telling me what types of lens's are in a triplet. Thanks Guy Grotke You can use it intact, just as it is. Assuming it still works, of course. If you get an OHP projection panel (ie. ebay $75-100) and lay it on the top, you are in business! Most of these have both VGA computer input and composite video input for watching TV & DVDs. (Look for at least 640 by 480 native resolution with 16 million colors. 1 million colors is ok but not as nice.) Many people are perfectly happy with this setup. (Typicall, people who like watching their big screen TV better than fiddling with a DIY projector. .:devilr: ) This also assumes your OHP has the light source below the transparency surface. Some of the real cheap newer OHPs have the light source up by the lens assembly, and use a reflector below the transparency surface. Those don't work with LCDs. The polarization twists done by the LCD result in no light reflected back through the pixels. Another option is to dissassemble it, to build more of a box DIY projector. You can use the fan, the fresnels, the lens assembly, maybe the focussing mechanism, the power cord & switch :D . You could start your development with the OHP lamp, but they don't last very long. I would replace that with a Metal Halide bulb, ballast, spherical reflector,and condensor lens. MH bulbs give you lots more light, less heat, and run 10000-20000 hours. Of course, they cost $20-40. The next improvement is to go with a better LCD. 15" LCD monitors are down to the $200 range now, and have great parameters. You should get a model that somebody else has already tried, so you know it is easy to strip away the backlight and then get all the PC boards out of the way, etc. The OHP fresnels might not be quite big enough to completely light a 15" LCD (ie. 9" by 12"). You might need to buy some larger fresnels. If you used a 14" LCD monitor, then they should be big enough. (But most 14" panels have 50 msec response time, so they show some ghosts and trails when stuff moves fast.) In general, start with what you have and then upgrade as needed. If you want more light, go to MH. If you want less image distortion, go a triplet. If you want higher resolution, higher constrast ratio, and better response time, go to a stripped LCD monitor panel. ace3000_1 quote: But I was referring just to the first lens in the light engine. Guy, are you trying to tell me somthing? lol:D I was only talking about the primary condenser aswell and you will never find a company who will sell you a DCX lens as a primary condenser because none are made. Sure maybe you will find one as a secondary for instance in a slide projector (very rare), but never as a primary. Primary condensers are either PCX or Asheric, the index of refraction on a DCX is too steep to be used as a primary condenser and it will therfore give you distorted light and a hotspot. The diameter of a condenser also matters, the correct size for a 250w mh bulb with an arc of 25mm is 65-80mm, flooding a condenser with light will distort the image, not having enough light in a condenser will give you a dull image so this calculation needs to be taken very seriously in a design and calculated correctly for optimum results. The power factor of a condenser also plays a role, ideally in our setups, a short focal condenser with a low power has far less chromatic distortion then a short focal with high power. Once these parameters have all been calculated out correctly only then a condenser will make the world of difference. I gather you got your information from the DIY Builder Group, the only reason they use the DCX over there is because thats the only lens that JCB could find to work in his system, and its scientificly incorrect. Here is a Aspheric condenser, infact these are the best, they work perfect in a singular formation with a spherical reflector, on a large panel. These are actually called condensers where as the PCX is an alternative. http://www.edmundoptics.com/onlinec...?productid=2454 And here is your PCX. http://www.edmundoptics.com/onlinec...?productid=1747 And your DCX http://www.edmundoptics.com/onlinec...?productid=1748 Notice how the DCX is designed for imaging while the other 2 are for light collecting?. The DCX is best used in a multi element lens system, ie: a triplet, not for collecting/condensing light. Trev:) Guy Grotke Here is a simple discussion of various lens properties: http://www.olympusmicro.com/primer/...ensesintro.html Here is a nice paper on Cooke Triplets: http://www.willbell.com/tm/The%20Co...ar%20Lenses.pdf Triplets are generally made with a negative lens (ie. a DCV) between two positive lenses (ie. DCXes). They are designed to correct chromatic aberration at more than just two wavelengths, and also to do some spherical aberration correction. Designing them is so complex, that everybody uses computer simulations now. There are too many variables to wing it: focal lengths, spacing beyween lenses, lens materials, etc. Just buy one! If you want to make your own, get a degree in optics. You can assemble your own achromatic doublet: This corrects the chromatic aberration for red and blue, and makes it easier to get a longer focal length for a long-throw projector. But even a simple DCX can be used for low magnification applications. Most DIY projector applications should qualify as low magnification. Guy Grotke Thanks for the info, but I'm not sure I understand it all: >the index of refraction on a DCX is too steep to be used as a primary condenser What does the shape of a lens (DCX versus PCX, etc.) have to do with the index of refraction of the material it is made from? (ie. crown glass=1.52, flint glass=1.65) Did you mean the reflectance at the edges of the DCX lens gets too high? >a short focal condenser with a low power has far less chromatic distortion then a short focal with high power What do you mean by power? The lumens put out by the lamp? Why would chromatic aberration matter in a condensor lens? We are not trying to focus the lamp's individual bits accurately. Or do they end up contributing to the total chromatic ab in the final image because you get blue rays travelling in a different direction than red rays? Wouldn't that very small difference in angle be much smaller than the effect of the non-point-source lamp arc? I also noticed the edmunds 75 mm diameter condensor lens has a fl of 50 mm. So this is a very much stronger lens than the 200 mm fl lens I mentioned. Wouldn't this add more distortion? I suspect that either could be made to work just fine in a DIY projector application. After all, they are being used in magnification mode, rather than projection mode. It should just be a matter of adjusting the distance between the lens and the lamp arc, until the light fills the lower fresnel. It would be interesting to see if anybody could tell the difference, or measure it with a light meter. On the other hand, that edmunds lens costs $47.90, and you can get a 100 mm diameter 200 mm fl DCX from surplus shed for $3.50. I think the difference between "good enough" and perfect might not be worth the price. Has anybody actually tried both in the same system? lone_wolf_84 That is EXACTLY the info i been hunting for for the past few days, Good work ppl i think my mind finally has something to grasp onto. thanks alot!!!!!:D :D :D :D :D ace3000_1 Guy, not trying to be nasty here:), but may i ask, you have read this forum? All of the answers are in here. quote: On the other hand, that edmunds lens costs $47.90, and you can get a 100 mm diameter 200 mm fl DCX from surplus shed for $3.50. I think the difference between "good enough" and perfect might not be worth the price. Yess but have you thought about glass types? that lens from surplus shed isnt temperd, now that will go boom, and if it takes out your bulb, what will then be the final price? quote: Why would chromatic aberration matter in a condensor lens? We are not trying to focus the lamp's individual bits accurately. Or do they end up contributing to the total chromatic ab in the final image because you get blue rays travelling in a different direction than red rays? Wouldn't that very small difference in angle be much smaller than the effect of the non-point-source lamp arc? My advice in this area is to build a projector and see the difference yourself, any abberations coming right back from the source will show in the image. The blue and red rays dont travel in different directions, they travel at different lengths. Lamps also dont have bits, the rays must be acurate as possible to get our ever so perfect image. What would you rather watch, a washed out bright image or a shazor sharp bright contrast corrected image? Condensers arent to magnify the light, they are to condense the light into a beam to the frensel. Frensels have very high chromatic distorion as it is, we dont want to add any more. Chromatic distortions from condensers is undoubtly visible when using the wrong lens. quote: What do you mean by power? The lumens put out by the lamp? Magnifacation. quote: What does the shape of a lens (DCX versus PCX, etc.) have to do with the index of refraction of the material it is made from? (ie. crown glass=1.52, flint glass=1.65) Did you mean the reflectance at the edges of the DCX lens gets too high? That was a typo error, my bad, i should have said the angle of incidence is high, therfore we get reflection. The PCX and the Ashperic have a flat side, think about it, the DCX is curved. Have a read in here, it will also explain to you the various other distortions ect acociated with lenses of different types. http://scienceworld.wolfram.com/phy...berrations.html Trev:) lone_wolf_84 yes i have read this forumn and other forumns, but when u got to search through 6 pages per thread about ppl randomly going on about stuff that isnt relative to the thread, u get a bit ansy and start skimming, or skip pages all together, not much consentrated information if you ask me. But anyway, again thanks. Also while ur here, what do uv filters consist of, is there anyway to fashion up ur own without to many consecuences? ace3000_1 quote: yes i have read this forumn and other forumns, but when u got to search through 6 pages per thread about ppl randomly going on about stuff that isnt relative to the thread, u get a bit ansy and start skimming, or skip pages all together, not much consentrated information if you ask me. But anyway, again thanks. Heay buddy, my previous post was actually aimed at Guy, not yourself so no probs there:) . Regards to searching, try the "diyprojector part 2" thread, there is a tonne of info in there and that should be more then enough to get you on your way ;) . There is also the "newbies thread" which also contains alot of information to get you started. Be sure to check out the "diy compatible monitors list" in the LCD forum so you know which panels are best to use and so you dont go buying one that is unsuitible. As for UV filters, some say normal glass works, it does but it wont filter out a great deal, the best solution is to buy a proper UV filter in the long run just to make sure your coverd. On the other hand, you can buy metal halide bulbs which have a inbuilt UV filter and that will save you having to worry about buying one after, thats what i normally do ;). In a triplet lens we have a negative DCV and 2 cemented achromats. Some vary but this is the most common arangement. Trev:) Guy Grotke I have read thousands of posts in this and other DIY projector forums. (Not just diybuilder.com) I have also read lots of other optics sites, gotten out my college physics textbook, etc. I agree that a condensor lens should be made of a material that can tolerate some heat stress, so tempered glass would be a good idea. I also agree that a PCX shape would be better than a DCX, in terms of losing less light to reflection near the edges. But I still don't understand how a condensor's chromatic aberration could possibly matter, or how a 50 mm fl condensor could generate LESS chromatic ab than a 200 mm fl lens. The chromatic ab in a lens works just like a prism. It makes the various colors of a white ray bend in different directions. So a single white dot near the edge of the LCD panel. lit from a point source, would look like a white dot with blue on one side and red on the other at the screen. But this is the chromatic aberration introduced by the objective lens. The chromatic aberration introduced by the condensor would make the blue versus red components of a point source bend in slightly different directions as they head toward the fresnel & LCD. But since the lamp has a hugh arc size compared to the difference in the position of the blue and red fringes of a particular ray at the fresnel, there will be blue light traveling in exactly the same direction (from another part of the lamp arc) at the "red spot", and vice versa. I think the net result will be that it all averages out and looks like white light. The large arc length of the MH bulbs we use will send rays through each LCD pixel in many directions, so each pixel's projected light is a cone diverging toward the objective lens. A perfect objective lens would focus all of the rays in the pixel's cone, back to a single point on the screen. But no lens is perfect! Any spherical aberration in the lens will send the pixel's rays to slightly different screen positions, making a blurry image. The important point here is that the circle of confusion made at the screen by the condensor's chromatic aberration will be hundreds of times smaller than the circle of confusion introduced by the objective lens spherical aberration. So I think the chromatic aberration introduced by the condensor will be 100's of times less important than the size of the lamp arc. But hey, that's just my opinion. Maybe I'm wrong about this. I'll try some experiments to see. Also, when I used the term "magnification mode", I meant that the lamp is closer to the condensor than its focal length. So the resulting beam out the other side of the condensor does not converge anywhere. Instead it diverges to fill the fresnel with an unfocussed uniform beam. At least, that's what we hope it does! ace3000_1 Heya Guy, yeah your right in a few areas but some things ill clarify to try and help you in this and so its not so confusing for others too. quote: I agree that a condensor lens should be made of a material that can tolerate some heat stress, so tempered glass would be a good idea. I also agree that a PCX shape would be better than a DCX, in terms of losing less light to reflection near the edges. Ok, im glad we got that sorted:) quote: But I still don't understand how a condensor's chromatic aberration could possibly matter, or how a 50 mm fl condensor could generate LESS chromatic ab than a 200 mm fl lens. Well i didnt say it wouldnt, this depends alot on the frensels focal used aswell, having the frensels focal matched with the condenser is just as important as the condenser its self, having these matched you will get very minimal distortions.And yep your right, the shorter the focal length is the more distortions we will get on a single lens, but the less of an angle we have to spread the light out ie: the smaller the lcd, the more we can magnify things within a certain limit. Also there are a couple of different lens platforms, there is the thick lens design and the thin. By far the thick lens design carries far more distortions then the thin and personally try and stay away from the thick as the distortions are clearly visible. Aspherics can have the higher power because they have been corrected, thats why they have the ring around the bottom, this cuts back or limits somewhat the distortion levels at a short focal range, and that is why they are the "ideal" condenser. quote: The chromatic ab in a lens works just like a prism. It makes the various colors of a white ray bend in different directions. So a single white dot near the edge of the LCD panel. lit from a point source, would look like a white dot with blue on one side and red on the other at the screen. But this is the chromatic aberration introduced by the objective lens. Well yeah but the colours are at different wave lengths and they all focus at a different focal point with the lenses characteristics. I dont get what your saying about red being on one side of the lcd and blue on the other because they both will be on the same side, depending on how far apart or how bad the abberations is will depend on how far they are apart, in most cases this is just a blur to the human eye and its somewhat somthing thats found visible to the human eye on the edge of the light being focused from the lens. The chromatic abberations from the objective we will get into soon, right now it doesnt play a role. Put it this way, if the light is blured going to the lcd or distorted what happends? The pixel is flooded with distorted light, the pixel becomes distored as the light is what iluminates it, it would be the same as having as diffuse light sorce going through that pixel, we dont want this, we want clear light guided through that pixel rather then to flood it with light, and thats why duffused light is no good for this aplication nor distortion because it gives us a washed out blury image. quote: The chromatic aberration introduced by the condensor would make the blue versus red components of a point source bend in slightly different directions as they head toward the fresnel & LCD. But since the lamp has a hugh arc size compared to the difference in the position of the blue and red fringes of a particular ray at the fresnel, there will be blue light traveling in exactly the same direction (from another part of the lamp arc) at the "red spot", and vice versa. I think the net result will be that it all averages out and looks like white light. I agree on that, our arcs are definatley far too big for what we need and with a 25mm arc you wont get perfect results, but you can get close with the right optics. I think you will find our eyes see it as white as this is micro stuff we are talking about here, but its distorted light not clear. The center of the image is always clear because of this and because of the projection lens but this has probally just as much as a role to play as the projection lens. quote: The large arc length of the MH bulbs we use will send rays through each LCD pixel in many directions, so each pixel's projected light is a cone diverging toward the objective lens. Well i wouldnt say that because we are guiding the light through each pixel and the light comes from only one dirrection, the arc, thats if you dont go reflecting it from the sides. Each pixel forms part of the main cone, it never has it own, only that can happen if each pixel has its own indipendant lens. Some lcds do, the micro lens tech lcds do in pro projectors but they run different then our setups and i wont get into that right now. quote: A perfect objective lens would focus all of the rays in the pixel's cone, back to a single point on the screen. Nope cos this happends in thin air after the projection lens at its focal point. quote: But no lens is perfect! Any spherical aberration in the lens will send the pixel's rays to slightly different screen positions, making a blurry image. Your dead right there, enough for us to see a blur, with some lenses, ie singlets, its terrible! quote: The important point here is that the circle of confusion made at the screen by the condensor's chromatic aberration will be hundreds of times smaller than the circle of confusion introduced by the objective lens spherical aberration. So I think the chromatic aberration introduced by the condensor will be 100's of times less important than the size of the lamp arc. But hey, that's just my opinion. Maybe I'm wrong about this. I'll try some experiments to see. The thing i see you forgetting is the fact that when we project with the projection lens, what ever is behind it we magnify, so what ever problems we have inside of our light engine they actually become bigger and not smaller, we magnify the problem. So any chromatic distortions ect from our condenser actually gets magnified from our frensels, in most cases then we have the frensels distortion on top of that, the lcd is filled with distorted light on the sides and then better yet we have the distortion of the projection lens. So in a real world environment having a little bit here and there will become alot, and thats why i say try to use lenses fit for the job otherwise you will be left with a result thats somewhat not what you hopped for. When adding positive lenses together the distortion increases, we can only correct the distortion with a negative lens, how manny negative lenses do we have in our projector? none, well infact no we do, in the triplet but that by far doesnt have the capabilities or specs to correct the rest of the system, so the only way to fix this issue is to use the correct lenses right from the start. Reflectors play a very big role in this too, but im not going to get into that either as ive gone through that manny times and we would be here all night lol. I use 2 condensers in my light engine, the secondary isnt a PCX, nor is it a DCX, ill keep it a mystery, but it certainly corrects the distortion and gives me a perfectly even bright clear and sharp image ;). Trev:) lone_wolf_84 OH, u made me remember a question, How do i determine the specs of my fresnel from my OHP? Diagonally its 13 inches 1/8 inch thick and theres only one of them its a hard plastic/acrylic theres a number on it and its 202805 and i got 2 lense's and a mirror inside the focusing box at the top there. the 2 lense's are identicle, and there 4" wide 1/8" thick, and the totel depth of the lense is 1/2" (ie from the outer most center point, to the outer most edge of the rim.) ancorp Ok a guessing game. Is the secondary condencer a positive meniscus lens? is it a negative lens? is it a doublet/triplet lens? ace3000_1 quote: Originally posted by ancorp Ok a guessing game. Is the secondary condencer a positive meniscus lens? is it a negative lens? is it a doublet/triplet lens? LOL nope. Trev:) ancorp Well it has to be something decent... give me a clue, is it multi-element? is it even a lens? otherwise this game aint n fun:D Guy Grotke Lone_wolf_84: You can determine the most important lens characteristic using the sun, a yard stick (or meter stick for you in countries that use sensible measurements), and an ant: Hold the lens up so one side of it is facing the sun and lower it toward the ground until the ant is vaporized. Then measure the distance from the lens to the ant carcass. That is the focal length. That tells you how "strong" the lens is. Window glass has a focal length approaching infinity. A very strong magnifying glass may have a focal length of 200 mm. Then flop the lens over and measure the focal length on the other side. They may be the same, or they may be different. This only works for positive lenses, like a PCX, DCX, or positive meniscus. If there is a negative lens in your OHP lens assembly, then another method is used. If you can dissassemble your OHP lens assembly, then you can measure the focal lengths of any positive lenses. You can also measure the focal length of the whole thing. This is the critical number to know for calculating how the projection half of your system will work. Negative lenses (ie. PCV, DCV, negative meniscus) have a negative focal length. That means they have an imaginary focal point on the same side of the lens as the light source. To measure one of these, use the sun around noon, two yard sticks, and a patch of clean concrete: Have your assistant hold the lens a couple of feet above the ground, so one side faces the sun. (No ant required. :D ) Find the outline of light coming through the lens by moving it up and down a bit and watching the circle shrink or grow. Then put one yard stick on the one edge of the light circle and against the corresponding edge of the lens. Put the other yard stick on the opposite edges of the light circle and lens. Where the two yard sticks cross is the focal point. Have your assistant measure the distance between the lens and the intersection of the yard sticks. Or you can just combine a negative lens with a stronger positive lens, measure the combined focal length, and calculate back to the negative lens focal length: 1/fl(pos) + 1/fl(neg) = 1/fl(combined) Don't forget the negative lens focal length in that equation is a negative number! Your OHP fresnel is probably actually two of them solvent welded together. That's okay. Measure both sides focal lengths. As for UV protection, the biggest danger is that you will fry your eyes working with a MH bulb. I found a paper on this that said most of the bulb envelopes filter out very little of the UV. Don't look at or work near a lit MH bulb. Or else wear lexan safety glasses: if you search the web for the absorbtion spectra of lexan and acrylic, you will see they absorb about 90% of the UV light. If your fresnels are acrylic, you already have pretty good UV filters! You can also buy Lexan-XL at home improvement centers with even more UV absorbtion. A piece big enough to use as a panel or fresnel support costs only a few dollars. There are also some better UV filter sheets you can buy here and there for <$10. it would be interesting to measure to see if they add much more filtering to a system with a piece of lexan and an acrylic fresnel between the bulb and the LCD. lone_wolf_84 You where one of them disturbed little boys wernt you? :D :D ANYwho even though in the middle there you lost me, after the whole any frying thing.. ""WOOSH"" i guess i will have to read it a few times, and give a couple attempts at controlling the ant population around my place.. I will let you know how it works out, I have a huge junk of lexan that I was using as a windshield this past race season, and its not to badly scratched up, so i will defiently give it a try. Guy Grotke If you can find an area that is still perfect, then that would be ok. But if this is going in your light path close to the LCD, you will see ANY imperfections. Another reason to spend a couple of bucks and get a new piece: Lexan-XL has better UV filtering. I was just kidding about the ant, in case any PETA members are reading. No animals have been harmed in the production of this forum thread! :angel: lone_wolf_84 LMAO so what did you use then? I was also wondering, this mirror thats in there inbetween my two lenses... is it important? lone_wolf_84 :apathic: it took 21 inches or 635 mm for the hypothetical ant to vaporize... same on teh back side, and i got the exact same from the other lense.. I couldnt get it to work with the fresnel though... lone_wolf_84 btw 635 mm is bad right? ace3000_1 quote: it took 21 inches or 635 mm for the hypothetical ant to vaporize... It takes under 2 seconds with a blow torch :hot: :att'n: Guys im sure we are aware of the saftey in here, please dont go and start fires with lenses, this is not only dangerous but foolish. The various posts above was only intended for a joke. :att'n: The safe way to measure the focal length is doing it inside on a table with a room light on. We simply focus the light bulb to the table and then measure the distance from the lens to the table to get the estimated focal length of the lens. Trev:) ace3000_1 quote: btw 635 mm is bad right? Actually thats good if your going to be using a large panel, it will give you more throw. Trev:) lone_wolf_84 lol I did the indoor light thing. as of an hour ago i got a 17" 400:1 lcd with 15 ms(R), 10ms(F). With the lenses i got and the fresnel i got should I have a good start? i wouldnt mind to get a fairly decent imagine to keep my spirites high during the rest of the tweaking. SOOooOo after lots of reading and what not can i expect a good image? Guy Grotke >focus the light bulb to the table and then measure the distance from the lens to the table to get the estimated focal length of the lens I don't know how big your room is (and how powerful your light bulb is), but under normal conditions this will give a VERY bad estimate! Focal length measurement needs a parallel beam, like sunlight. I just measured a lens with a light bulb about 10 feet from the lens, and got 680 mm. Then I took it outside and used the sun: 480 mm This is consistant with the lens equation: f = focal length u = the distance of the object from the lens v = distance of the image from the lens (may be negative) 1/u + 1/v = 1/f When you use the sun as your object, 1/u is near 0 so it simplifies to: 1/v = 1/f and then to: v = f If they don't have a sun where you live, then you can use a light bulb, but you need to plug the distance between the bulb and the lens into the lens equation. lone_wolf_84 As I look outside and see nothing but darkness and small drips of moisture i doubt i ill be able to test my focal length today.. Mr.Weatherman says we are going to get some sun today, but he as been wrong all week. So i will just sit here and cross my fingers. the biggest room in my apt here is about 12 feet X 12 feet. The OHP uses a EYB 360 watt 2000 lumans bulb, Oh.. i found this on a website regardin gmy OHP: Lens Speed & Focal Length: 14" Doublet and ill kill more hypothetical ants when the sun comes out. ace3000_1 quote: I don't know how big your room is (and how powerful your light bulb is), but under normal conditions this will give a VERY bad estimate! Guy it works for me spot on everytime. Lets face it, the sun is dangerous and its not a practise that should be used. Ive got over 100 lenses here and all work fine with the light bulb trick to their spec. If i go and get a 200mm fl lens out now and do the light trick, the distance from the table to the lens when the light bulb is focused on the table is just about dead on 200mm, give or take 1mm. Its exactly the same measurment with the sun. The only problems may arise is if we have a 1000mm focal length lens because we run out of room and the room light becomes too close. Have you tried this in reality and compared the 2 or are just relying on theroy? Frensels also need to be split to have their focals measured acuratley, you need to measure each focal length on each lens to get any form of an acurate result. Trev:) ace3000_1 Ill add if your going to measure a frensels make sure its turned around the right way first, this also goes for PCX lenses or any lens that is plano. Trev:) lone_wolf_84 well um, i only have one fresnel.. and if there are two, then there sealed tgether under the acrylic/plastic layers.. I did get that number off the corner though... still dont know what it means though, i think its porbably just a part number. ace3000_1 quote: well um, i only have one fresnel.. and if there are two, then there sealed tgether under the acrylic/plastic layers.. I did get that number off the corner though... still dont know what it means though, i think its porbably just a part number. Yeah its 2 cemented together around the edges, my advice would be to go to your OHP's manufactures website and look up that part number, it should state the specs, if not hit them with an email. You will then have your acurate results for each frensel, the top frensel ( normally with the sticker on it) will have a different focal length from the rear, hardly any OHP's have frensels that have the same focal length on both sides. Trev:) Guy Grotke >Have you tried this in reality and compared the 2 or are just relying on theroy? Let me quote my post: "I just measured a lens with a light bulb about 10 feet from the lens, and got 680 mm. Then I took it outside and used the sun: 480 mm" When I said "I just measured a lens", I mean that I really did before I made the post. I wouldn't post it unless I was pretty sure it was true. But the lens equation is the fundamental description of converging lens behavior. The very definition of focal length is the distance from the lens to the focal point when a parallel beam strikes the lens. If you want to verify this, pick out one of the longer focal length lenses from your pile, measure it with a light bulb (and measure the distance from the bulb to the lens), then take it outside to measure it with the sun. If the sun is unavailable, then you can just measure it with the bulb far and the bulb near (ie. 10 feet versus 6 feet). Your measurements will be radically different. Since you will also have the bulb to lens distance, you can plug it into the lens equation to get the real focal length. Then start remeasuring those 100+ lenses! So lone_wolf: You don't have to wait for the sun! :D lone_wolf_84 yip i luckly caught a break and just finished before it clouded back over and started raining, and same thing my best guess is that these are 600 mm, (im averaging inside, outside, then rounding) So now that i know the focal length, what calculations need to be done to get specs of the box i need to build? lone_wolf_84 If i didnt tap my credit card right out on my lcd purchase, id fork out the money for the lumalab forumn and design. but since i cant, i wont, and this place is better anyway :cool: :dodgy: lol so if anyone wants to give me some rough hel on an enclosure, id be extatic. Also does it matter how far away the fresnel is from the light source? and does it matter how far away the lcd is from the fresnel? and the mirror inbetweem the lenses in the ohp head, does it do anything special? lone_wolf_84 ONE more thing, ROUGHLY from one of the lenses to the mirror is 2" and same from the other lense to the same point on the mirror. So does that mean when i put these lenses in my box, that one lense should be 4" away from the other? ace3000_1 quote: When I said "I just measured a lens", I mean that I really did before I made the post. I wouldn't post it unless I was pretty sure it was true. Well from my 2 years hands on experience i didnt beleive it for a second, you also didnt state what lens you are trying to measure or how you had it placed/turned ect. quote: (ie. 10 feet versus 6 feet). Your measurements will be radically different. That kind of depends on the focal of the lens in the first place doesnt it? I measure mine with a diffused light sorce btw and i dont have any problem, if you turn your lenses around the wrong way you will. 0 focal barrier is atleast 2x the focal of the lens, having it within this barrier will give you inacurate results. The sun is also a point source not just a beam of paralelle light. Manufactures measure lenses with precision lasers. quote: Then start remeasuring those 100+ lenses! Why? they all have their correct focal measurment from their manufactures lol and that how i know the light trick works. Try using a energy saving light, i dont use a normal bulb. Trev:) ace3000_1 quote: So does that mean when i put these lenses in my box, that one lense should be 4" away from the other? Yep;) lone_wolf_84 and that the first one of the 2 lense's should be 600 mm away from the lcd? lone_wolf_84 Is there any added bonus of splitting my fresnel? if it is, and i do, i put one behind and one infront of the lcd right? Guy Grotke lone_wolf: you might want to actually use the whole lens and mirror assembly on top of your DIY box. People pay some money to add a front surface mirror just before the projection lens, so the box can be smaller. You already have one! So anyway, here is my lab setup: I have a 6 cm wide halogen bulb pointing at a wall that is 1829 mm away. The bulb has a distinctive pattern on the front, so it is very easy to focus an image of it on the wall with a lens. I held a 5" diameter DCX lens between the bulb and the wall, and then moved the lens toward the wall until a sharp image of the bulb face appeared on the wall. Then I measured the distance between the lens and the wall. This distance was 420 mm. I tried doing the same with the other side of the lens, and the measurement was the same. The distance from the bulb to the lens was (1829 - 420) = 1409 mm. Using the lens equation: 1/f = 1/1829 + 1/420 f = 323.5 mm Then I took the lens out on my porch and focussed the image of the sun on a brick (no fire danger :D ). The distance between the lens and the brick was 325 mm. This is within 1% of the value I got from the lab experiment and well within the error of my measuring technique. So what is the focal length of this lens? 420 mm or 325 mm? I could repeat the lab experiment with a different bulb to wall distance and the lens to wall distance would change. But I could still plug those new numbers into the lens equation to calculate the true focal length: 325 mm If the distance between the lens and the wall changes when you move the bulb, than that can't be the focal length of the lens: The focal length does not change! Guy Grotke Sorry, typed it in wrong: 1/f = 1/1409 + 1/420 f = 323.5 mm ace3000_1 quote: So what is the focal length of this lens? 420 mm or 325 mm? I could repeat the lab experiment with a different bulb to wall distance and the lens to wall distance would change. But I could still plug those new numbers into the lens equation to calculate the true focal length: 325 mm All you need is a normal light thats mounted to the ceiling turned on, thats all you need to give your light , not walls or any fancy lights lol. Then all you have to do is focus the light bulb from the ceiling light onto a table. A 5inch DCX is kinda big and as i said in my previous post we have a limit on the size of the lens used, try a smaller one like a 100mm diameter pcx. I just tried a 80mm PCX with a focal of 124mm, i tried it on the floor and then tried it less then a foot from the bulb, there isnt any focal difference at all, the only difference was the light bulb that was focused changed size. Trev:) ace3000_1 quote: So what is the focal length of this lens? 420 mm or 325 mm? If that lens is 5inches round and if its not a very thick lens id doubt that its under 500mm. Trev:) Guy Grotke >id doubt that its under 500mm It is a 3 Diopter DCX from one of those desk magnifier/worklamps. But we don't really NEED to guess... Diopters = 1000 mm / focal length 3 = 1000 mm / focal length focal length = 1000 mm / 3 focal length = 333 mm 5" diameter is only 25 % wider than 100 mm diameter. No reason that the fundamental nature of a lens should be different because it is 25% wider than the one you think is "the right size". The lens equation works just this way for any positive lens of any size. As for your experiment, perhaps you live in a universe with different physical constants, so your lens equation works differently. I live in the one where Isaac Newton worked this out a long time ago. If you think I'm making up the lens equation, you can do a web search for "thin lens equation" and look at some of the hundreds of college physics course websites that use it. Lone_wolf: If you put the lens assembly at exactly the focal length from the LCD, your image would focus at an infinite distance. Use the lens equation to calculate how far away it has to be, for the size of your throw distance (projector to screen). Hint: It will be a bit more than the focal length. ace3000_1 quote: As for your experiment, perhaps you live in a universe with different physical constants, so your lens equation works differently. I live in the one where Isaac Newton worked this out a long time ago. If you think I'm making up the lens equation, you can do a web search for "thin lens equation" and look at some of the hundreds of college physics course websites that use it. What? i simply stated that if its a thin lens it will have a long focal, just that! Thin ie: 10mm thick for a 5inch diameter sized lens is thin. I know very well the thin lens equation, hell its on the site i posted a few posts back! Perhaps you need to try the way i do things for yourself first using the exact same gear to answer your own questions, a long halogen bulb wont exactly give you the best results for measuring anything, try the power saver bulbs like i use and you just may well see a difference. I dont have a problem with my lenses and out of the ones that i have measured in this room they have always measured out to their spec. I am simply telling you my findings, not trying to cause an argument. quote: I live in the one where Isaac Newton worked this out a long time ago. For the thin lens equation?, i see a Edmund Halley in there somewhere..........the famous astronomer. http://scienceworld.wolfram.com/phy...ensFormula.html Trev:) Guy Grotke Well Trev, I posted data from real experiments, and you suggested there was something "wrong" about my lens, my light bulb, and even the fact that I am sending the light horizontally against a wall instead of down from a ceiling fixture. :rolleyes: I tried changing from my halogen flood (which is roughly the shape of a 15 Watt incandescent bulb, not long and skinny), to a 60 watt incandescent soft white, and then to a spiral compact fluorescent 25 watt bulb replacement. All had exactly the same result: With the bulb 1829 mm from the wall, the 325 mm fl DCX lens focussed a perfect image on the wall from 420 mm. When you plug those distances into the lens equation (AKA the Gaussian Lens Formula), you calculate a focal length of 323.5 mm. Just to prove that the lens-to-wall distance is NOT the focal length, I ran another experiment against another wall: With the bulb 2851 mm from the wall, the same 325 mm fl DCX focussed a perfect image of the bulb on the wall from 375 mm. (NOTE: Not anywhere near the previous experiment's 420 mm lens-to-wall distance.) When you plug these numbers into the lens equation, you calculate a focal length of 325.7 mm. Sorry, but I just can't suspend my disbelief enough to try it vertical instead of horizontal. If light behaves that differently depending on the orientation to gravity, then we are being sucked into a black hole at the center of the earth! :D In summary, I ran several different experiments and they all showed that the distance from the lens to the image is NOT the focal length of the lens, unless the light source is at an infinite distance. Which agrees with the lens equation. On the other hand, you said: >I just tried a 80mm PCX with a focal of 124mm, i tried it on the floor and then tried it less then a foot from the bulb, there isnt any focal difference at all... Using the lens equation for bulb-to-lens distances of 2311 cm and 304.8 cm (7'7" and 1 foot, respectively), I calculate that you would measure distances from lens to image of 131 mm in the first case, and 209 mm in the second case. These are so different, that I have to conclude that you must have made a huge measurement error, or it was just a thought experiment. Since we seem to have results that do not agree, let's have another experimenter try it: Lone_wolf_84, why don't you try measuring the lens-to-image-on-the-wall distance of a simple magnifying glass, using a bulb that is 6 feet away from the wall and then again after moving the bulb 10 feet away from the wall. If this distance is the actual focal length of the lens, then it should not change. Then take the lens outside and measure the distance to another ant victim. (To hell with PETA, I do hate ants!)Post the three measurements you get. ace3000_1 quote: Sorry, but I just can't suspend my disbelief enough to try it vertical instead of horizontal. If light behaves that differently depending on the orientation to gravity, then we are being sucked into a black hole at the center of the earth! I never said that! lol i was simply stating that all you have to do is turn a room light on. quote: Since we seem to have results that do not agree, let's have another experimenter try it: Try a pcx, i garentee you the difference, thats what i measured. quote: Using the lens equation for bulb-to-lens distances of 2311 cm and 304.8 cm (7'7" and 1 foot, respectively), I calculate that you would measure distances from lens to image of 131 mm in the first case, and 209 mm in the second case. These are so different, that I have to conclude that you must have made a huge measurement error, or it was just a thought experiment. Well thats your calculation, mine is a real life measurement, my ruler doesnt lie and neither do i. I think you are getting somthing mixed up here, my focal didnt change, the size of the light bulb focused did. Ill try and take you a picture but ill doubt it will turn out as my camera is dark as hell even on a sunny day. Trev:) lone_wolf_84 hey um, not to break up ur discussion or anything ;) but i never did get an answer to my question regarding fresnel splitting and weather or not its better to have the bulb-fresnel- lcd, or bulb- one side of fresnel-lcd- other side of fresnel and I do believe this is my last question before i start construction sometime next week and i plan on taking LOTS of pictures. Guy Grotke We have gotten rather side-tracked. Regarding splitting fresnels: If you split them, you can make the light travel through the LCD pixels perpendicular to the panel. If you don't split them (or if you split them but put them both between the light source and the LCD), then the light will be converging as it goes through the pixels. Fresnels are much lower quality than glass lenses, so you get the sharpest image if you don't put one between the LCD and the projection lens. On the other hand, you might have an LCD panel with very poor transmission of light at at angle (ie. viewing angle). I think this difference in viewing angle characteristics leads different experimenters to have differences of opinion about the necessity of splitting the fresnels. If I were you, I would try an optical bench-type setup to see if you get good results without splitting. In other words, use some cardboard, wood blocks, tape, string, etc. to position your light, un-split fresnel, LCD, and projection lens so it can project an image the desired distance to a screen (use the real projection distance you want to end up with). You will probably have to make a cardboard cover to limit light leaks. Then you can move things around a bit to get the best possible image. (I totally agree with Trev on this: you need a finely- detailed high contrast image, not a brightest possible but blurry image.) If you consistently see a lot of viewing-angle artifacts around the edge of the image, then try it with a split fresnel design. If that change gets rid of the artifacts, then you have your answer. The fresnels are there to send most of the light to the projection lens. You can verify that they are doing that by removing the LCD (or just driving it with a white image), and then moving a sheet of white paper along the path from the LCD or upper fresnel toward the projection lens. The light should converge toward the lens aperature. It if converges before you reach the lens and then begins to diverge, then you need a different focal length fresnel or projection lens. (The fresnel fl should be a bit longer than the projection length fresnel, if you do have the perpendicular light situation I described above. But you may not: It depends on the spacing between the lower fresnel and light, and the lower fresnel focal length.) Once you have a setup that you are happy with (hopefully ecstatic :D ), then measure the actual distances very precisely,and make a box for it all. If you have trouble getting your fresnels to work right, you can always do some light condensing experiments with cheap page magnifiers from an office supply store, before ordering a more expensive higher quality fresnel. A 2 Diopter magnifier fresnel is 1000/2 = 500 mm focal length, a 3 Dipoter is 1000/3 = 333 mm, etc. Another good trick: If your LCD is not stripped, mounted in a good protective frame with all the PCBs secured, and running, you can make a mockup frame with some clear plastic or transparency film in it. Put a test pattern on it with dry markers. Then use that for your first experiments. Guy Grotke Trev, I just figured out why you saw what you did in your experiment! Sorry I doubted your findings. What we had was just a communication problem. When you put the PCX lens a foot from the lamp, you saw a very large projected image on the floor or table top. This fits into the thin lens equation perfectly: 1/d(light-to lens) + 1/d(lens-to-image) = 1/focal length In any converging lens projection setup like this, the direction of the light is interchangeable. So long as the bulb and the image screen are more than 2 times the focal length apart, this precedure actually produces two different points of focus for the lens. One point has the short side nearer the screen, and the other has the short side near the bulb. But for a particular lens and particular bulb-to-screen distance, both of the short sides have the same length. So you can plug the short side and long side measurements into the lens equation and get exactly the same focal length value using either point of focus. I was talking about the point of focus nearer to the screen. When you put your 124 mm PCX very close to the bulb, then you had the short side between the lens and the bulb. Try again with a table lamp exactly 10 feet from the wall and then move the lens starting from the wall toward the lamp until you find the short focus point. Measure the distance between the lens and the wall. Then move the table lamp so it is exactly 6 feet from the wall and then move the lens starting from the wall toward the lamp until you find the short focus point. Measure the distance between the lens and the wall. The two measurements will be very different. And that's why you can't use a light bulb that is a short distance away to directly measure the focal length of a lens. If you have 30 foot high ceilings, then a ceiling lamp might be far enough away that the value of 1/d(light-to lens) will be insignifigant compared to the value of 1/d(lens-to-image) for a short focal length lens. But with 8 foot ceilings or longer lenses, you have to measure both distances and use the lens equation. Or just go outside and use the sun, which has 1/d(light-to lens) ~ 0. Then the other point of focus would be with the lens 124 mm from the sun! :hot: lone_wolf_84 Ok so i guess ill sit here and unpatiently wait for my lcd, then after that i will start all the testing and what not. FYI the discussion you two are having aint bad, ive learned alot from it :P so on that note, Countinue on with the debate :D Guy Grotke The whole thing about the lens equation stuff IS right to the point of your project: If you can duplicate the spacing of your OHP, but replace the OHP lamp with a MH bulb, and the transparencies with an LCD panel, then you will have a working projector. But if you want it to behave differently, like having a longer throw (ie. 100" image from the back of the room), then you can use the lens equations to figure out what the proper distance will be between the new projection lens and the LCD. Then you can use it again to figure out the combined focal length you need for your pair of fresnels to get the light into the projection lens. ace3000_1 quote: If you have 30 foot high ceilings, then a ceiling lamp might be far enough away that the value of 1/d(light-to lens) will be insignifigant compared to the value of 1/d(lens-to-image) for a short focal length lens. Mines are 12ft lol, as i explained before it wont make a difference so long as you are out of the lenses focal length you have to be past its focal length measurment for it to work, if you are within its focal length then i think its obvious you wont focus the bulb on a table. The only difference i saw was the size of the bulb when focused, the focal point didnt change. As for most of the calculations ill be honest with you, you dont need to use them, most of those are for designing lenses from scratch rather then using them on lenses that you've bought that already have the specs. All of my lenses have the specs for them and only at certain times i may do a calculation to make sure all is correct in a multi lens light engine array for abberations. The room light test is just a quick an simple way to measure the EFL, and yess its just a very estimated EFL, (the EFL on a bought lens is also very estimated i might add). Building a projector isnt also what you think it is, having everything at its perfect focal points doesnt always work out, most of the time they need to be just off, this applies especially to condensers. So realy at the end of the day the thing that realy matters most to a projector builder is his/others hands on experience rather then his maths. Ive got 2years of full time experience behind me, both practical and theroy, i can draw you up a projector in 5mins and it will be garenteed to work perfectly, i wouldnt garentee you it would work if i didnt have any background practical experience. We have our base points in theroy aswell as in practical. I did read in another thread that you have tried to build a projector and havnt succeeded, there is nothing wrong with that and for first time builders that is very common, but ill give you some friendly advice. Use this forum and others to help you, forget the calculations for now, use the parts that others have worked with and make a working model of your own first, then calculate your theroy differences, and at times you will soon see that the theory sometimes conflicts with a working design in reality. Why you may ask? ask yourself the very same question of where the calculations came from to begin with, they came from mostly cameras and telescopes from over 100 years ago. The theorys are old and resonably acurate, but in a modern real world environment somtimes they arent always acurate enough, and because of this new ones emerge every year. There was just an upgrade to a calculation widley used by world leading scientists that was just proven wrong, i cant remeber the name of it (was somthing in quantum mechanics), but anyone who has a PHD in such and such using that old calculation is now polictically wrong. I hope that made sence i just woke up:rolleyes: Trev:) Guy Grotke Trev, I think you may draw some incorrect conclusions from things you read in these posts: I have never written that I tried to build a projector and failed. I wriote something like "the number of sucessful projectors I have built so far = 0". But that's just because I am still waiting for jcb to send me an FFC extension I ordered. My projector is coming along just fine, thank you! I have very high confidence that it will work, because I do understand the math, I have run several experiments with projection lenses, fresnels, a transparency mock-up of the LCD panel, and a light source (effectively lots of prototypes), and the results of the experiments agree perfectly with the math. That has been my experience with all sorts of engineering endeavors. I have worked in medical research labs, written embedded software for air pollution monitors and the baseband chip used in most of the Japanese PHS cell phones in the world, designed and built a multi-camera automated body volume measuring system on a NASA grant, designed and built a three-D mouse and drawing system for the PC using a pair of CCD cameras and LCD shutter goggles, etc. Now I write functional verification software for a company making 100 Mbit/second DSL chips, so we can have high confidence in the designs before spending $1000000 for the first silicon. (Writing software for chip companies pays a LOT better than any of the optical stuff.) In all of these, working designs are based on the known properties and relationships of materials and forces. And that means designs based on physics and math. And the the lens equation I have been suggesting IS a thin lens simplification of the general lens equation. It is also exactly the same form equation as the one used for calculating the combined focal length of two lenses, or the effective resistance of two parallel resistors! Funny how the same math describes how so many different thing work. Reading all of the posts in this and several other forums has been educational, in the sense of learning what works together well, from the items available now at reasonable prices. But I have been looking at LCDs with the idea of building a projector for about 20 years. Now their price/performance lines have intersected at the point that you can build a great projector for less than you would pay for a new TV, and I have the time available to put into it. The LCDs available in 1984 were pretty worthless for this purpose and ungodly expensive. At that time, my sister-in-law's boyfriend started writing a flat panel display industry report and newsletter. I read some of his early reports and got interested. (And no, I didn't pay the $250,000 report fee! :bigeyes: ) On the other hand, experience with all of the useful items that now fit within the budget of a DIY builder, is a great asset that I am short on! So, based on all your experience with projectors, have you ever seen a real 14" or larger LCD projector that used a 45 degree angle cold mirror? These have wonderful IR removal properties and there are several of these available at surplus prices, but those are all under 5 inches in diameter. That means the light beam has to fit in about a 3 inch diameter circle when it strikes the mirror. That seems way too small for any part of the light path before hitting the LCD, unless the condensor gets a lot more complex. So for all the suggestions to use a cold mirror, I don't think anybody is actually doing it in state-of-the-art DIY projectors. Maybe that is why these cold mirrors are so cheap! :D Have you seen one used? ace3000_1 quote: I have never written that I tried to build a projector and failed. I never said you did ;). quote: On the other hand, experience with all of the useful items that now fit within the budget of a DIY builder, is a great asset that I am short on! So, based on all your experience with projectors, have you ever seen a real 14" or larger LCD projector that used a 45 degree angle cold mirror? These have wonderful IR removal properties and there are several of these available at surplus prices, but those are all under 5 inches in diameter. That means the light beam has to fit in about a 3 inch diameter circle when it strikes the mirror. That seems way too small for any part of the light path before hitting the LCD, unless the condensor gets a lot more complex. So for all the suggestions to use a cold mirror, I don't think anybody is actually doing it in state-of-the-art DIY projectors. Maybe that is why these cold mirrors are so cheap! Ive seen a 15inch projector in person, a guy down the road from me has one that i helped him build but not one with a ir cold mirror in it.. Its very doable but as you say 3inches diameter is kinda on the limit, 4 inches should work though and surplus shed sell these quite cheap, about $5 from memory. Im all for the cold mirror and dichoric range of reflectors, in the new design im doing now i use a dichoric reflector and the image is very white without the red tinge often given from the low band ir. I personally wouldnt go back. The colour shift of the reflector i use has a shift of 300k in colour, most good quality cold mirrors shift the colour about 300k up from the bulbs normal colour, it just makes those colours just stand out! and gives us an even better contrast level. You would think with the light being whiter the contrast would be less, but for some reason atleast to my eyes its a whole lot better. Using the dichoric reflector will also save the life of your bulb if its spherical because the light going back to its source is cooler then what it would be if the ir was being reflected back. I say go for your design, it sounds interesting and its totally doable, i recomend using 2 condensers in that setup and all should be sweet. Trev:) lone_wolf_84 WhoA WHOA WHOA:yikes: How come noone informed me of this whole cold mirror, dichoric thing? seems like good info to know, so ace, mind giving me some words of wisdom about them?:yes: :spin: Guy Grotke Do a google search for "cold mirror" dichroic ,and you will find lots of general info. These would be wonderful for use in a projector: You bounce the light of it at 45 degrees (some are for 0 degrees), and >90% of the visible light we want gets reflected. >90% of the IR passes through! Almost magic! You can also look up the counterpart "heat mirror", that reflects IR but passes most of the visible light. These are easier to use (just because of the physical geometry of the light path), but not quite as good: They cut out a bit more visible red than a cold mirror, and I have read that they wear out after a while. I am not surprised to read that this gives you better contrast: The red range is 625 to 740 nm, which is a big difference in wavelength. If the filter cuts out some of the longer red, then you should get less chromatic aberration. In some work I did with CCD image sensors, I added IR cut filters and got MUCH sharper images. This is why most video cameras include IR cut filters. I plan to use a heat mirror in my light engine, but I would LOVE to find a dichroic cold mirror spherical reflector. That would send about half of the IR right out the BACK of the light engine before it even gets to the condensor. I was thinking I would have to buy a high temperature PCV lens and get somebody to put the coating on it. That can't be cheap! Have you really found one we can buy at a reasonable price, Trev? please, please, please... ace3000_1 quote: The red range is 625 to 740 nm, which is a big difference in wavelength. If the filter cuts out some of the longer red, then you should get less chromatic aberration. Yeah your right, i first noticed the pixels being very sharp and bright, it certainly makes a big difference. I suspect it does cut out some of the longer red because whats meant to be black is black and not a blackish red, there is still a slight red tinge wich is very faint but i think thats obviously because we are only filtering the reflected light and not the light from the front of the bulb. So if i wer to use a cold mirror aswell the red would disapear. The colours i noticed changed alot, red was a bright viabrant red, the green was the right shades of green and stood out much more where as before they wer abit dormant, and the blue was also clour corected and vibrant. Im using a 5100k colour bulb this way and with the reflected light being 300k up the reflected light becomes 5400k so realy its abit of a mix lol but half way there to perfect colouring. The backlight in the lcd is 5500k so its getting close to colour spec. After doing this i tried a 4200k colour bulb with a standard reflector and by all means i just cant turn back, the 4200k colour is just too low and makes things look dirty. To me making half of a decent image is having the right colours, it realy makes things come alive, things look real as they are meant to be rather then looking dirty and dull, a daylight scene looks like its day rather then it being over cast ect. Another thing that stood out changing to a dichoric reflector as i meanioned before was the contrast, colours blended well and stayed sharp but wearnt over powering where it wasnt meant to be, the blacks wer black and the only pixels that yealded light wer the ones that wer meant to, it realy makes a big difference. quote: In some work I did with CCD image sensors, I added IR cut filters and got MUCH sharper images. This is why most video cameras include IR cut filters. Yeah thats right, a good example is web cams. Actually You can make them see through clothes if you only have a ir filter in lol ( with the right wave length ofcourse ), but we wont get into that. quote: I plan to use a heat mirror in my light engine, but I would LOVE to find a dichroic cold mirror spherical reflector. That would send about half of the IR right out the BACK of the light engine before it even gets to the condensor. I was thinking I would have to buy a high temperature PCV lens and get somebody to put the coating on it. That can't be cheap! Have you really found one we can buy at a reasonable price, Trev? please, please, please... I can prety much get you anything you want buddy, you got any specs for me? like focal of the reflector and its diameter? I got a few normal sphericals kicking around you could use one of those if your using a cold mirror and not too worried about your bulb life. I can get you good condensers, the ones i use, they are temperd borosilicate optical glass with a strong dark green AR coating, (that makes a big diff within its self for a clear image). What size bulb are you planing on running and what type? Trev:) Guy Grotke I am trying out a Ushio S250DD metal halide bulb first. I like the specs on this bulb: 5200 degrees K color temperature (lone_wolf84: closer to sunlight than most MH bulbs) 90 Color Rendering Index (lw84: 90/100 is much better than most MH bulbs 65/100) And since it is an HPS replacement bulb, it is in a pretty small package. (So it can fit in a lot of existing HPS fixtures) The outer glass is a 46 mm diameter cylinder, and the arc length is 24 mm. Not quite the smallest sized bulb available, but very nice color. It is rated to run 20000 lumens (at first, anyway :D ) with an average life of 15000 hours. If my projector needs a new bulb in only 7500 hours because of the reflector & hot mirror, that would be fine with me. I picked a 250 Watt bulb because I plan to use an optical-grade front surface spherical reflector (lw84: effectively doubles the light output), and the geometry of my media room demands a 95" image from 14 feet. I am hoping the low magnification (lw84: 95/15 = 6.3 X) and good throughput of the LCD give me enough lumens on the screen. I can also go to a higher gain screen material if I need more light, since we will hardly ever have more than two people watching. (lw84: So we will be viewing the screen from directly in front.) I am still thinking about the light path to the lower fresnel. I would like to get a cold mirror in there, but my 1:1 scale drawings seem to indicate that is not possible with a 220 mm fl lower fresnel and a 100 mm condensor. It looks like I could get a 10" diameter cold mirror in there if I use a 500 mm fl lower fresnel. Maybe a smaller condensor, or no condensor at all... I have a 75 mm diameter 65 mm fl spherical reflector on order from jcb, but his store still shows that as backordered. (He also NEVER returns emails!) I planned to work out the light engine geometry to fit that reflector, since it looked like the best one I could find without paying $100s. (lw84: surplus or commodity parts can be a LOT cheaper than trying to order 1 of something from an industrial optics supplier) But what I really would love to find is a reasonably-priced off-the-shelf drichroic spherical reflector. These would be so cool (pun intended) that I think somebody could make some money by getting a few hundred made and then selling them to the DIY projector community. Apparently lots of people are running 400 Watt bulbs with no reflector (or a bad reflector) and they are trying lots of things to keep excessive heat away from the LCD. They could pretty much just drop one of these into their current projector to get twice the light or half the heat. ace3000_1 Heya Guy, one thing ill tell you and im sure you knows this, Dichoric are the most reflective reflectors we can buy, its the best surface made. Darn why you get that bulb for? lol you should have gotten a HQI-TSD lamp, they have a built in uv filter, are 25mm in diameter, 165mm long, 20mm arc, 20000lm, 5100k, and with a 90cri, they are also 80-90lm per watt eficientcy. With this bulb we can capture more light and have a much tighter light engine thats more acurate and efficient. About the reflector from jcb, thats not too good, he runs them at half focal or there abouts so the rays arent acurate and the focal is just too long, (the reason why they are run at half focal), you will loose light with it. However given that if it has an enhanced aluminium surface you will see the difference, i have one here thats not the same as his but has the enhanced alloysurface (50mm dia, enhanced aluminium surface, 50mm focal, 1/12 wave length) and ill tell ya right now, becarful, because those things burn holes in lcds. Mine nearly did with a 150w mh CDM-T bulb and a heap of cooling on the lcd. :xeye: You can get a decent dichoric reflector for yours BRAND NEW for about $50 in that diameter, but if you had of gotten the bulb i sujested up top, you could of gotten a smaller reflector that will give you a brighter image and possibly reflect more light for about $30. The size of your cold mirror will depend on where you place it, the closer to the light the better for size, a 4inch should do it with those focals. Trev:) Guy Grotke Interesting... I'm not married to that Ushio bulb! An HQI-TS form would be much easier to fit to a spherical reflector. And a smaller arc length would be nice, too. (Just a bit harder to use a double-ended bulb with a big parabolic or eliptical reflector, maybe. But those are just ideas for experiments.) But all of the HQI-TS bulbs I found seemed to be the older 4200 K 65 CRI formulation. You said "5100k, and with a 90cri". Who makes that bulb? I would like to look at the spec sheet. I am always willing to try something better! And who sells dichroic spherical reflectors for $30-$50? ace3000_1 quote: An HQI-TS form would be much easier to fit to a spherical reflector. And a smaller arc length would be nice, too. (Just a bit harder to use a double-ended bulb with a big parabolic or eliptical reflector, maybe. Actually you can, im suprised no body has thought in how to do this yet, im holding it back for now (atleast until i try it first :) ), same with the reflector link as its been proven in the past my links and designs are what some others like to profit on. quote: You said "5100k, and with a 90cri". Who makes that bulb? I would like to look at the spec sheet. I am always willing to try something better! Osram, here is their spec sheet. Trev:) ace3000_1 Here is a picture of the other bulb i have, the CDM-T 150w, the specs for that is 4200k colour, cri 90, 12k hr life, 10mm arc, 19mm dia and a MOL of 100mm. Trev:) Guy Grotke Thanks, I have been poking around on the Osram site and never saw that bulb! Maybe it is newer than their website. The little ceramic bulbs look nice, too. But I have not seen one above 150W. What do you think about these bulbs: http://www.ushio.com/technicalspecs/Emarc.pdf They are DC short-arc MH bulbs designed for projection, among other applications. example: 270W 41 volts @6.7 Amps 5700 K 1.5 mm arc length 14000 lum but only 2000 hour life. <== nothing is perfect! They also have one of these in a dichroic eliptical reflector intended for driving an optical fiber, but from the geometry specs I think you could couple it to a 330 mm fl fresnel to light a 15" LCD panel. In this design, all the light would go through the reflector's 10 mm wide focal point before diverging to the fresnel, so it would be very easy to add a small cold mirror. I don't see anybody selling these on the web with a google search. I will call up their sales office on Monday to ask for prices. Then maybe I will stop by and pick up one or both. They are only 17 miles from my office. :D ace3000_1 They are a great system but then we are heading away from diy and the cost of those little things are as much as a normal projector bulb, as you can see they have the same/similar specs. If you go out and buy ones of those then you might aswell buy a projector wich kinda sux lol. Prety much how this all started in here is because the replacement prices of a bulb was the problem and people try to do things on the cheap. I like to compare cars to projectors, you can go out and aford that sports car but the thing is running cost and repair costs will be the part you cant, its very similar to bought projectors as where we can aford the actual unit but we cant aford the bulbs, you get my drift? ;) But either way the bulbs in that site you posted are great but just too darn exspensive. Trev:) Guy Grotke >im suprised no body has thought in how to do this yet Oh, I've thought about a lot of these sorts of things. If I just wanted a projector, then I would have bought an OHP panel on ebay and an OHP at a yard sale. But I'm more interested in learning about all the possibilities. An elliptical reflector that is bigger than the LCD would be very cool! It could converge the light through the LCD panel, and right into the projection lens. No condensor, no fresnels! I once made a 4 foot diameter parabolic reflector when I was a kid out of masonite, heavy paper, and aluminum foil. I might just try that again as a prototype, once I have a lamp and get my panel running again. ace3000_1 quote: Oh, I've thought about a lot of these sorts of things. If I just wanted a projector, then I would have bought an OHP panel on ebay and an OHP at a yard sale. But I'm more interested in learning about all the possibilities. Im the same mate, i try to get the best out of somthing lol or try to improve on an aspect in a design ect, isnt life boring just buying stuff and not learning? quote: An elliptical reflector that is bigger than the LCD would be very cool! It could converge the light through the LCD panel, and right into the projection lens. No condensor, no fresnels! Possibly, not saying that it wont work cos it will just how well is the question as the light wont be going through the lcd paralelle. quote: I once made a 4 foot diameter parabolic reflector when I was a kid out of masonite, heavy paper, and aluminum foil. I might just try that again as a prototype, once I have a lamp and get my panel running again. I do alot of sheet metal work and ive made a few:) , but its just one of those things where as its just as easy and cheap to buy one, acuracy is always questionable aswell, ive got few light engines that ive made in the past that basically recycles light, ive made others aswell that are basically just a big reflector that houses the actual normal light engine to capture the lost light. At the end of the day with the wide range of optics and projectors, we have unlimited possibilities, having this technology available to comsumers now days also helps. Trev:) Diy101 quote: Originally posted by ace3000_1 Actually you can, im suprised no body has thought in how to do this yet, im holding it back for now (atleast until i try it first :) ), same with the reflector link as its been proven in the past my links and designs are what some others like to profit on. Osram, here is their spec sheet. Trev:) Anyone knows if this bulb is available in the US? Or of any other places that can be ordered from ..online ? zzonbi That can't excuse measurements' tolerance either. The conjugates are unique. If you have one, you have exactly another one to form a pair for a given thin lens (their name suggests this too). Humorous debate, though ;)

Page generated in 0.28418707847595 seconds with 17 queries,
spending 0.00946426 doing MySQL queries and 0.27472281 doing PHP things.

Powered by: Search Engine Indexer and vBulletin
Copyright ©1999-2009 diyAudio.com