hmmm, what defines a meniscus lens apart from a normal one?
I'm thinking of using a 2.5inch lcd and a metal halide bulb and a fresnel lens to collimate the light, but i'm unsure what kind of lens setup to use on the projection side. I'd like to get a picture about 150" or slightly less diagonal on my screen...
I'm thinking of using a 2.5inch lcd and a metal halide bulb and a fresnel lens to collimate the light, but i'm unsure what kind of lens setup to use on the projection side. I'd like to get a picture about 150" or slightly less diagonal on my screen...
lenses
Xblocker-
You are right about not needing a lens that is the size of your panel. However, I have to disagree about needing a doublet (or triplet) with a focal length that is minimally the size of the panel being used. If you have a fresnel before your panel, then the light coming through the panel is converging and you can use whatever focal length doublet or triplet you want - as long as it is large enough (in diameter) to catch the converging cone of light. The practical reason that the focal lengths of typical doublets on OHP's are at least a foot, is to give the proper magnification to get a big image. Using the following equation
(focal length x screen width) / panel width = throw distance in inches
which is the same as
M = F2/F1, where F2 is throw distance, and F1 is focal length of the lense, and M is the magnification
Using these equations you can determine the focal length you need to get the size on the screen you want in a certain throw distance. This is how the
OHP's work. Now it is not by accident that the focal lengths turn out to be greater than the panel width, take for example:
throw distance = 10 ft = 120 inches
6 ft screen width = 72 inches
10 inch panel width
so now plug into our equation
FL= (TD*panel width)/screen width = 120*10/72 = 16.7 inches
So long focal lengths come directly out of this equation, but you could use shorter focal lengths,
you would just get a bigger magnification.
Anyway, this is how I understand things, but if you know something I don't, I would love to hear it! So, let me know what I am missing.
BTW I totally agree with you about needing a double or triplet, without them you will get chromatic and spherical aberrations, plus even higher order effects. That being said they could turn out to be small effects, which might be why Undream's picture looks fine. I have a buddy that knows how to use Z-max (a ray tracing program), and I have been trying to get him to model this, he said he would, but he is kind of busy building real astronomical instruments. Anyway hopefully I will get some kind of model going here sometime.
Thanks
J
Xblocker-
You are right about not needing a lens that is the size of your panel. However, I have to disagree about needing a doublet (or triplet) with a focal length that is minimally the size of the panel being used. If you have a fresnel before your panel, then the light coming through the panel is converging and you can use whatever focal length doublet or triplet you want - as long as it is large enough (in diameter) to catch the converging cone of light. The practical reason that the focal lengths of typical doublets on OHP's are at least a foot, is to give the proper magnification to get a big image. Using the following equation
(focal length x screen width) / panel width = throw distance in inches
which is the same as
M = F2/F1, where F2 is throw distance, and F1 is focal length of the lense, and M is the magnification
Using these equations you can determine the focal length you need to get the size on the screen you want in a certain throw distance. This is how the
OHP's work. Now it is not by accident that the focal lengths turn out to be greater than the panel width, take for example:
throw distance = 10 ft = 120 inches
6 ft screen width = 72 inches
10 inch panel width
so now plug into our equation
FL= (TD*panel width)/screen width = 120*10/72 = 16.7 inches
So long focal lengths come directly out of this equation, but you could use shorter focal lengths,
you would just get a bigger magnification.
Anyway, this is how I understand things, but if you know something I don't, I would love to hear it! So, let me know what I am missing.
BTW I totally agree with you about needing a double or triplet, without them you will get chromatic and spherical aberrations, plus even higher order effects. That being said they could turn out to be small effects, which might be why Undream's picture looks fine. I have a buddy that knows how to use Z-max (a ray tracing program), and I have been trying to get him to model this, he said he would, but he is kind of busy building real astronomical instruments. Anyway hopefully I will get some kind of model going here sometime.
Thanks
J
jco9w,
watching the interdependence between fl, object size and imagesize, you're right about the formulas. But now let's assume we have a projection lens with a fl=2" to project a 10" panel. You would have a field angle of 2*cos(2/5)=133degrees! Now tell me, which projection objective is capable of doing that? This must be a very special wide angle objective. Not to talk about a single lens! Another critical point is lens thickness and aperture. The thicker a lens, at a given FL and aperture, the more disturbing elements concerning imaging come into game. Excellent obectives have an F-ratio of 1.0, which means aperture diameter = focal lenght, this also is only possible with corrected lenses.
We cannot only discuss magnification ratio without looking at quality. There have to be always compromises between math calculations and real physical-optical components.
Also, the bigger (steeper?) the transmitting angle of light through LCD the poorer the results, but thats another topic.
In Photography all big format cameras have longer FLs than mini formats...Why?
well, all in all i cannot see, why this shouldn't be the case for DIY-projectors. Make yourself a test with 10" panel and a slide projector objective and tell me what you can see!!
(grr, what a torture in english!)
xblocker
watching the interdependence between fl, object size and imagesize, you're right about the formulas. But now let's assume we have a projection lens with a fl=2" to project a 10" panel. You would have a field angle of 2*cos(2/5)=133degrees! Now tell me, which projection objective is capable of doing that? This must be a very special wide angle objective. Not to talk about a single lens! Another critical point is lens thickness and aperture. The thicker a lens, at a given FL and aperture, the more disturbing elements concerning imaging come into game. Excellent obectives have an F-ratio of 1.0, which means aperture diameter = focal lenght, this also is only possible with corrected lenses.
We cannot only discuss magnification ratio without looking at quality. There have to be always compromises between math calculations and real physical-optical components.
Also, the bigger (steeper?) the transmitting angle of light through LCD the poorer the results, but thats another topic.
In Photography all big format cameras have longer FLs than mini formats...Why?
well, all in all i cannot see, why this shouldn't be the case for DIY-projectors. Make yourself a test with 10" panel and a slide projector objective and tell me what you can see!!
(grr, what a torture in english!)
xblocker
good point
xblocker-
You make good points. My goal in the previous email was not to say that practically you weren't correct (in fact I showed that you do need a long
focal length), but to make sure people understand the physics. Agreed, you need a giant diameter lens if it has a short focal length, and they are hard to come by and can cause other problems, however, you could (in theory) still make it work.
Point being, a rule of thumb is a rule of thumb and physics is physics. Your rule of thumb works well with the projectors we are talking about, but
I didn't want people to come away that it is the law.
Anyway there is no need to argue, I agree with you about the focal lengths, I just thought that I would put the formulas out there so people got the whole story.
J
xblocker-
You make good points. My goal in the previous email was not to say that practically you weren't correct (in fact I showed that you do need a long
focal length), but to make sure people understand the physics. Agreed, you need a giant diameter lens if it has a short focal length, and they are hard to come by and can cause other problems, however, you could (in theory) still make it work.
Point being, a rule of thumb is a rule of thumb and physics is physics. Your rule of thumb works well with the projectors we are talking about, but
I didn't want people to come away that it is the law.
Anyway there is no need to argue, I agree with you about the focal lengths, I just thought that I would put the formulas out there so people got the whole story.
J
jco9w,
i agree, i only made some points, because i often see a big lack of information about optical systems here.
Duo,
with a 2" LCD you're nearly in the dimension of the slide projectors. I would try any objective of an old slide projector or modify such a projector. Look at the small LCD projector thread. There you can find a lot of hints. You can also try 2 mensicus lenses, each with about 7" FL to get a total of 3,5".It should look like that: ( ) . Eye glasses have meniscus shape. I experimented a lot with used or missworked eye glasses, which i got for nothing from an optometrist.
Good luck!
xblocker
i agree, i only made some points, because i often see a big lack of information about optical systems here.
Duo,
with a 2" LCD you're nearly in the dimension of the slide projectors. I would try any objective of an old slide projector or modify such a projector. Look at the small LCD projector thread. There you can find a lot of hints. You can also try 2 mensicus lenses, each with about 7" FL to get a total of 3,5".It should look like that: ( ) . Eye glasses have meniscus shape. I experimented a lot with used or missworked eye glasses, which i got for nothing from an optometrist.
Good luck!
xblocker
Duo,
Eye glasses with focal lenght of 3? What measure? And, are you talking bout FL or dioptrics? Dioptrics are defined as 1Dpt=1/f (f in meter,Germany).
I don't understand why opticians shouldn't give away defective lenses. I've already built simple telescopes and projection objectives out of it. So what!?
You also can try 2 planconvex lenses. In every case it's better to have 2 thinner lenses than 1 thick!
xblocker
Eye glasses with focal lenght of 3? What measure? And, are you talking bout FL or dioptrics? Dioptrics are defined as 1Dpt=1/f (f in meter,Germany).
I don't understand why opticians shouldn't give away defective lenses. I've already built simple telescopes and projection objectives out of it. So what!?
You also can try 2 planconvex lenses. In every case it's better to have 2 thinner lenses than 1 thick!
xblocker
about "low e-glass"
Hay guys just a note about low E-glass this is not a great thing. If anyone has ever seen low e glass it looks like tinted glass. The loss in light transmittance is any ware from about 25% to 35% and there are more problems, the glass really has a "shelf life" due to oxidation problems. Not to mention that glass itself is pretty opaque is anyone has ever seen a case if glass… you cannot see thru it! So too much glass can be a problem! Just a thought.
Hay guys just a note about low E-glass this is not a great thing. If anyone has ever seen low e glass it looks like tinted glass. The loss in light transmittance is any ware from about 25% to 35% and there are more problems, the glass really has a "shelf life" due to oxidation problems. Not to mention that glass itself is pretty opaque is anyone has ever seen a case if glass… you cannot see thru it! So too much glass can be a problem! Just a thought.
Large Lenses...
Hi Guys,
Ok, you got me - yes you CAN have an image bigger than the lens and still project an image.
e.g. a "pin-hole camera" can focus the image of the sun onto a screen, and it is probably safe to say that the sun is generally bigger than most pin-holes...
However, if you want to catch MOST (if not ALL) of the light from the image produced by an LCD panel, and not just a tiny fraction of it, then you must either use a very big objective, or use some form of lens before/after the panel (at least as big as the panel) to focus the emerging light rays towards the smaller objective.
In effect, in almost all projectors, there is an optical stage (collection of lenses) that is as big as the image being projected. In the OHP, it is the fresnel under the glass panel, in other projectors it comes before/after the image.
The two issues many people seem to be objecting to are:
1) my statement inearlier postings that the objective should be bigger than the image. (This WAS too general - my bad - I should have talked about having an optical stage bigger than the image.)
2) the use of a fresnel after the panel rather than before it. (This seems to be a philosophical issue. I still maintain that the image degradation by such a lens placed after the panel is much less (when the fresnel is placed properly) than the degradation that occurs in trying to use an equivalent fresnel to focus light "through" an LCD panel from behind.)
Most of the TFT panels around today are not good optical filters - just hold them up to a lamp at the other end of the room and look at the lamp through the panel - in most cases you will see scattering and a diffuse image of the lamp due to MLA layers designed to collect as much light from a diffuse CCFL source as possible and funnel it through to the front with the widest viewing angle possible. This is not 100% compatible with the concept of forcing light through from behind at a very specific angle needed for it to reach the smaller objective in front of the panel.
Some LCD panels were specifically designed for OHP work, and they are much better in this respect. Others weren't...
As in all science, theory takes you half-way there, and practicalities take you the rest. I apologise, if in being too general I was giving bad advice.
Bill.
P.S. I still think the optimal arrangement is to have the light entering the LCD panel from behind, perpendicular to its surface, and then having the emerging light focussed towards a powerful, optically corrected objective by a low power, large diameter lens. (The lower the power, the less aberations etc.) And, as always fresnel lenses are poor compared to a nice optical glass lens.
Hi Guys,
Ok, you got me - yes you CAN have an image bigger than the lens and still project an image.
e.g. a "pin-hole camera" can focus the image of the sun onto a screen, and it is probably safe to say that the sun is generally bigger than most pin-holes...
However, if you want to catch MOST (if not ALL) of the light from the image produced by an LCD panel, and not just a tiny fraction of it, then you must either use a very big objective, or use some form of lens before/after the panel (at least as big as the panel) to focus the emerging light rays towards the smaller objective.
In effect, in almost all projectors, there is an optical stage (collection of lenses) that is as big as the image being projected. In the OHP, it is the fresnel under the glass panel, in other projectors it comes before/after the image.
The two issues many people seem to be objecting to are:
1) my statement inearlier postings that the objective should be bigger than the image. (This WAS too general - my bad - I should have talked about having an optical stage bigger than the image.)
2) the use of a fresnel after the panel rather than before it. (This seems to be a philosophical issue. I still maintain that the image degradation by such a lens placed after the panel is much less (when the fresnel is placed properly) than the degradation that occurs in trying to use an equivalent fresnel to focus light "through" an LCD panel from behind.)
Most of the TFT panels around today are not good optical filters - just hold them up to a lamp at the other end of the room and look at the lamp through the panel - in most cases you will see scattering and a diffuse image of the lamp due to MLA layers designed to collect as much light from a diffuse CCFL source as possible and funnel it through to the front with the widest viewing angle possible. This is not 100% compatible with the concept of forcing light through from behind at a very specific angle needed for it to reach the smaller objective in front of the panel.
Some LCD panels were specifically designed for OHP work, and they are much better in this respect. Others weren't...
As in all science, theory takes you half-way there, and practicalities take you the rest. I apologise, if in being too general I was giving bad advice.
Bill.
P.S. I still think the optimal arrangement is to have the light entering the LCD panel from behind, perpendicular to its surface, and then having the emerging light focussed towards a powerful, optically corrected objective by a low power, large diameter lens. (The lower the power, the less aberations etc.) And, as always fresnel lenses are poor compared to a nice optical glass lens.
Xblocker,
You mentioned how large format cameras generally have larger FL lenses than the smaller ones. The main reason for this is perspective.
In 35mm photography, a "normal" lens is usually quoted as 50mm. This lens gives roughly the same field of view (perspective angle) as the eye - producing a "normal" looking picture.
Wide-angle (35 mm FL and below) give a much wider field of view, and telephoto (70mm and up) give a much narrower perspective. (Never do a facial portrait with a wide angle lens - noses can become enormous...)
For the larger format cameras, the image is projected onto a larger area of film, and the dimensions of the cameras are such that to achieve a "normal" perspective, you need a longer focal length lens. The larger the format, the longer the lens...
Bill.
You mentioned how large format cameras generally have larger FL lenses than the smaller ones. The main reason for this is perspective.
In 35mm photography, a "normal" lens is usually quoted as 50mm. This lens gives roughly the same field of view (perspective angle) as the eye - producing a "normal" looking picture.
Wide-angle (35 mm FL and below) give a much wider field of view, and telephoto (70mm and up) give a much narrower perspective. (Never do a facial portrait with a wide angle lens - noses can become enormous...)
For the larger format cameras, the image is projected onto a larger area of film, and the dimensions of the cameras are such that to achieve a "normal" perspective, you need a longer focal length lens. The larger the format, the longer the lens...
Bill.
Using an LCD screen and a very, very bright light to project the image on a white object has some cons.
1) Needs a very bright light.
2) Bright light produces a lot of heat.
3) Fan needs to be used to cool off the light emitting device.
4) Needs a large lens
5) LCD screens cost to much.
6) Visible black lines on the projector when sitting 6 feet away from the source.
There are so many other cons when using a LCD and light emitting device. I think using lasers might be of some interest. Lasers act like electron tubes but lasers can be much finer in detail. Laser diodes, found in CD players and laser pointers, prices are coming down. There are red and green laser diodes but blue laser doides are coming soon. The next best way besides using laser diodes is LEDs. LEDs can handle a lot of voltage around 10 times their orignal forward voltage. Laser diodes can not handle this amount of voltage.
Lasers work on scheme of pulsing a high voltage to a light emitting device for a short period. LEDs can be great for projecting like a laser. For more information go to How Lasers Work
LEDs come all sorts of colors such as red, yellow, green, orange/amber, purple, and blue. The colors need to emit all the colors in the rainbow are red, green, and blue. Each color has its own brightness so adjustment is needed for accurate color.
I found a circuit in my books that uses a LED to be used as a laser. I will post it if anybody have some interest.
Below is a reference:
Iannini, Robert E. "Build Your Own Laser, Phaser, Ion Ray Gun & Other Working Space-Age Projects". TAB Books. 1983
1) Needs a very bright light.
2) Bright light produces a lot of heat.
3) Fan needs to be used to cool off the light emitting device.
4) Needs a large lens
5) LCD screens cost to much.
6) Visible black lines on the projector when sitting 6 feet away from the source.
There are so many other cons when using a LCD and light emitting device. I think using lasers might be of some interest. Lasers act like electron tubes but lasers can be much finer in detail. Laser diodes, found in CD players and laser pointers, prices are coming down. There are red and green laser diodes but blue laser doides are coming soon. The next best way besides using laser diodes is LEDs. LEDs can handle a lot of voltage around 10 times their orignal forward voltage. Laser diodes can not handle this amount of voltage.
Lasers work on scheme of pulsing a high voltage to a light emitting device for a short period. LEDs can be great for projecting like a laser. For more information go to How Lasers Work
LEDs come all sorts of colors such as red, yellow, green, orange/amber, purple, and blue. The colors need to emit all the colors in the rainbow are red, green, and blue. Each color has its own brightness so adjustment is needed for accurate color.
I found a circuit in my books that uses a LED to be used as a laser. I will post it if anybody have some interest.
Below is a reference:
Iannini, Robert E. "Build Your Own Laser, Phaser, Ion Ray Gun & Other Working Space-Age Projects". TAB Books. 1983
Electro said:
1: This one is true. Whaddya know, you are 16.667% correct.
2: They don't produce a lot of heat. in fact, I even accidentally bumped into my MH bulb that had been burning for 45 minutes and didn't get burned.
3: Actually, they don't need to be directly cooled. Marklar's setup has a single fan mounted in the back, and isn't even pushing air upon the bulb. Most of these bulbs are designed to be in closed fixtures like street lights and things like that.
4: completely and utterly incorrect. I can use my ~3" diameter overhead projector lens to project my 10.4" diagonal LCD panel. You are just plain wrong. Want an even smaller lens? no problem. you can find one that will work. I just decided to use my 5 7/8" lens because it seems to work slightly better, and I want to use parts in my projector that can all be purchased again for cheap.
5: $100 is too much? How much would you spend on LEDs or Lasers? More importantly, how many man hours would you put in trying to get something like that to work? its all relative, I don't think this is an issue.
6: Put it slightly out of focus. Problem solved. Personally, I'll be sitting 12' from the screen, so, I'll be able to sharply focus without problems.
woneill,
agreed, my point was someting different. I mentioned the relation between math formulas and physical reality. One the mathematical side, we could theoretically get shortest FL and biggest magnification, but not one the physical side, 'cause real lenses (the thicker they are) unfortunatly produce many undesirable side effects, which can be only eliminated with a bunch of correcting elements! That's why it IMHO would be better not to choice a FL below the diagonal of the panel.
xblocker
agreed, my point was someting different. I mentioned the relation between math formulas and physical reality. One the mathematical side, we could theoretically get shortest FL and biggest magnification, but not one the physical side, 'cause real lenses (the thicker they are) unfortunatly produce many undesirable side effects, which can be only eliminated with a bunch of correcting elements! That's why it IMHO would be better not to choice a FL below the diagonal of the panel.
xblocker
Latest from the frontlines...
you'd be surprised what you can do with a cordless drill, a 45 cent protractor, a hand saw, and a calculator. damn I need some decent tools :\
Lens mount/front assembly. turned out darn good.
from the front
here it is, not fully mounted, but, just slid on so you can see what its gonna look like. Not beautiful, but, it should work. Still have to mount 4 focusing bolts/wingnuts, cut some vents for the LCD panel, figure out how I'm going to get the LCD panel remote to work, put on one or two more braces, and then of course, mount/install the four feet that its going to be standing on. Sucker is heavy altogether. really heavy.
you'd be surprised what you can do with a cordless drill, a 45 cent protractor, a hand saw, and a calculator. damn I need some decent tools :\
An externally hosted image should be here but it was not working when we last tested it.
Lens mount/front assembly. turned out darn good.
An externally hosted image should be here but it was not working when we last tested it.
from the front
An externally hosted image should be here but it was not working when we last tested it.
here it is, not fully mounted, but, just slid on so you can see what its gonna look like. Not beautiful, but, it should work. Still have to mount 4 focusing bolts/wingnuts, cut some vents for the LCD panel, figure out how I'm going to get the LCD panel remote to work, put on one or two more braces, and then of course, mount/install the four feet that its going to be standing on. Sucker is heavy altogether. really heavy.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.