Hi Xblocker,
I think the theory behind jco9w's design is that unlike the spherical arrangement that loses much light, the elliptical reflector captures MOST of the light from the bulb, and focusses it onto a single point - the new effective point source of light.
A condenser lens (?) is being used to diverge the light from this point source into a much wider cone of light than would be natural from the reflector alone - allowing the fresnel + LCD operate as expected.
The issues here involve the natural angle of divergence of the light coming from the reflector; the angle of divergence resulting from the condenser; the focal length of the fresnels; and the focal length of the objective.
jco9w,
I would suggest that if you are using OHP optics, then you should mimic the original OHP dimensions as much as possible: i.e. you want the divergence of the light coming from your reflector/condenser combination to be as close as possible to the original OHP as possible such that with the LCD panel fully and evenly illuminated, the rays hitting the fresnel can be traced back to an effective point source of light exactly where the fresnel's focal point would be.
If you are using a convex condenser, then you should check to see if you are merely reducing the divergence from the reflector, or converging it so much that you are actually re-focussing it through a new point. A concave lens might be better for this task if the light from your reflector is diverging too slowly. As mentioned the rays should be diverging at an angle equivalent to a point source placed at the fresnel's focal point.
Bill.
P.S. I'd draw a diagram but I can't draw...
I think the theory behind jco9w's design is that unlike the spherical arrangement that loses much light, the elliptical reflector captures MOST of the light from the bulb, and focusses it onto a single point - the new effective point source of light.
A condenser lens (?) is being used to diverge the light from this point source into a much wider cone of light than would be natural from the reflector alone - allowing the fresnel + LCD operate as expected.
The issues here involve the natural angle of divergence of the light coming from the reflector; the angle of divergence resulting from the condenser; the focal length of the fresnels; and the focal length of the objective.
jco9w,
I would suggest that if you are using OHP optics, then you should mimic the original OHP dimensions as much as possible: i.e. you want the divergence of the light coming from your reflector/condenser combination to be as close as possible to the original OHP as possible such that with the LCD panel fully and evenly illuminated, the rays hitting the fresnel can be traced back to an effective point source of light exactly where the fresnel's focal point would be.
If you are using a convex condenser, then you should check to see if you are merely reducing the divergence from the reflector, or converging it so much that you are actually re-focussing it through a new point. A concave lens might be better for this task if the light from your reflector is diverging too slowly. As mentioned the rays should be diverging at an angle equivalent to a point source placed at the fresnel's focal point.
Bill.
P.S. I'd draw a diagram but I can't draw...
woneill,
i'm not sure if the setup you described make things better!
The new point light source, which is generated by the elliptical reflector isn't a real point light source, it's an image of the real source. Reflected lightrays pass this point and diverge again under the the same angle as they come. If i understand right, a concav lens should amplify this divergence and the light should go the usual way to fresnel -- LCD--...?
But what's with the direct light from the bulb? It's wasted and will add more scattering light to the whole arrangement! I think the lighting concept of OHPs which is made for large imaging areas is not easy to beat, and you have the advantage of using direct and reflected light together with a small spherical reflector, doubling bulb's filament. Efficiency of elliptical reflectors is no question, but IMHO only if they are used as primary condensors!
xblocker
i'm not sure if the setup you described make things better!
The new point light source, which is generated by the elliptical reflector isn't a real point light source, it's an image of the real source. Reflected lightrays pass this point and diverge again under the the same angle as they come. If i understand right, a concav lens should amplify this divergence and the light should go the usual way to fresnel -- LCD--...?
But what's with the direct light from the bulb? It's wasted and will add more scattering light to the whole arrangement! I think the lighting concept of OHPs which is made for large imaging areas is not easy to beat, and you have the advantage of using direct and reflected light together with a small spherical reflector, doubling bulb's filament. Efficiency of elliptical reflectors is no question, but IMHO only if they are used as primary condensors!
xblocker
Gunawan W
I tried with the 2 fresnel lens an inch apart.
The resulst were blury in the corners.
Almost like having a round TV.
Then I also tried setup #2 and found that when I go to focus the
image, I can see the fresnel lines on the screen.
By puting it behind the LCD caused me to project the LCD image first, before seeing the lines on the fresnel.
As for as wasted light, Im only using a 250watt 4K MH bulb and
getting a very bright, clear and no distorted image.
And the room lighting does not need to be pitch black.
I have the lights on in the back of the room and the picture still looks good.
I did this on a trial and error basis, went throught what you guys
went throught, and after alot of confusion got it to work
beyond belief.
I tried with the 2 fresnel lens an inch apart.
The resulst were blury in the corners.
Almost like having a round TV.
Then I also tried setup #2 and found that when I go to focus the
image, I can see the fresnel lines on the screen.
By puting it behind the LCD caused me to project the LCD image first, before seeing the lines on the fresnel.
As for as wasted light, Im only using a 250watt 4K MH bulb and
getting a very bright, clear and no distorted image.
And the room lighting does not need to be pitch black.
I have the lights on in the back of the room and the picture still looks good.
I did this on a trial and error basis, went throught what you guys
went throught, and after alot of confusion got it to work
beyond belief.
Hi Xblocker,
An image of the real source that acts as if it is a point source, and contains most of the energy from the original point source is actually hard to beat - it is a very efficient, directional, virtual bulb.
If the elliptical reflector is any good then all the light from the bulb will be projected through the point source image in the direction of the screen. If the divergence of the light from this image equates to the divergence expected by the fresnels then focussing the fresnels on this image is perfect - much more light than any other method.
Otherwise the rays from the projected image should be diverged to give the same angle as an OHP would do.
The other OHP optics could stay as normal except they take their input from this efficient virtual bulb instead of an inefficient normal one.
Also, in most elliptical arrangements, the bulb is fitted longitudinally so that most of the light is emitted along the transverse axis and focussed by the reflector into the point source image.
Compared to the spherical arrangement that loses most of the available light from the bulb, any light that does get directly through the front of the reflector will be negligible. (A small lens could be also placed in front of a transverse bulb to focus this light too - discussed in the other reflector thread.)
If you really wanted to be fussy, a diaphragm should be used at the focal point of the elliptical reflector to remove any off-axis light. The end result will still be much more efficient.
Bill.
An image of the real source that acts as if it is a point source, and contains most of the energy from the original point source is actually hard to beat - it is a very efficient, directional, virtual bulb.
If the elliptical reflector is any good then all the light from the bulb will be projected through the point source image in the direction of the screen. If the divergence of the light from this image equates to the divergence expected by the fresnels then focussing the fresnels on this image is perfect - much more light than any other method.
Otherwise the rays from the projected image should be diverged to give the same angle as an OHP would do.
The other OHP optics could stay as normal except they take their input from this efficient virtual bulb instead of an inefficient normal one.
Also, in most elliptical arrangements, the bulb is fitted longitudinally so that most of the light is emitted along the transverse axis and focussed by the reflector into the point source image.
Compared to the spherical arrangement that loses most of the available light from the bulb, any light that does get directly through the front of the reflector will be negligible. (A small lens could be also placed in front of a transverse bulb to focus this light too - discussed in the other reflector thread.)
If you really wanted to be fussy, a diaphragm should be used at the focal point of the elliptical reflector to remove any off-axis light. The end result will still be much more efficient.
Bill.
Hi Redevil,
Well done - all successes are good.
When you put the fresnel after the screen, did you try it both ways around? Fresnels are directional - usually (but not always) they will only work with the rough side facing the LCD.
The trick is to get the fresnel as close to the LCD as is possible without getting the Moire lines from the fresnel interacting with the LCD.
If you have a laptop or LCD monitor, then that serves as a good way to test this out. Literally put the fresnel as close to the LCD as possible without getting the moire effect, and place your eye at the focal point. You should be able to see the screen being magnified and undistorted. If your eye is too close to the panel, you will not see the corners. If it is too far then you will see massive distortion. Fresnels as well as being directional have a very narrow viewing angle - your eye (the simulated objective lens) needs to be placed within a very narrow range of positions. The idea is to have the fresnel and objective matched so that they can interact to focus a reasonable sized image at a reasonable distance... (OHP optics are already matched to work this way.)
The correct fresnel orientation should collect most of the light from the panel - including the corners. If it didn't do this then something fundamental was going wrong - either the fresnel was the wrong way around, or the objective lens was too close to the LCD/fresnel, and was literally only getting light from the center of the light cone.
Bill.
Well done - all successes are good.
When you put the fresnel after the screen, did you try it both ways around? Fresnels are directional - usually (but not always) they will only work with the rough side facing the LCD.
The trick is to get the fresnel as close to the LCD as is possible without getting the Moire lines from the fresnel interacting with the LCD.
If you have a laptop or LCD monitor, then that serves as a good way to test this out. Literally put the fresnel as close to the LCD as possible without getting the moire effect, and place your eye at the focal point. You should be able to see the screen being magnified and undistorted. If your eye is too close to the panel, you will not see the corners. If it is too far then you will see massive distortion. Fresnels as well as being directional have a very narrow viewing angle - your eye (the simulated objective lens) needs to be placed within a very narrow range of positions. The idea is to have the fresnel and objective matched so that they can interact to focus a reasonable sized image at a reasonable distance... (OHP optics are already matched to work this way.)
The correct fresnel orientation should collect most of the light from the panel - including the corners. If it didn't do this then something fundamental was going wrong - either the fresnel was the wrong way around, or the objective lens was too close to the LCD/fresnel, and was literally only getting light from the center of the light cone.
Bill.
woneill
You are correct. The fresnel lens must be in one direction only.
In my case I do have the lines facing the panel on the back side
and yes it is close as possible. I removed the glass plate and replaced it with the fresnel. It was a bitch to cut the fresnel but I
managed. My image is so good that you can see the actual TFT
thin film transistor. This is at 640X480. My avermedia box, wich I use to interface the panel has the option to give me 800x600.
But to get that it uses a scan doubler (horizontal scan x 2)
It looks like crap in that mode. The image is so clear that I can
see the jagged edges when motion is viewd. But I will be hooking
in my computer soon. Cant wait. This panel can handel 1024x768.
When HDTV comes out I'll be ready.
Later
You are correct. The fresnel lens must be in one direction only.
In my case I do have the lines facing the panel on the back side
and yes it is close as possible. I removed the glass plate and replaced it with the fresnel. It was a bitch to cut the fresnel but I
managed. My image is so good that you can see the actual TFT
thin film transistor. This is at 640X480. My avermedia box, wich I use to interface the panel has the option to give me 800x600.
But to get that it uses a scan doubler (horizontal scan x 2)
It looks like crap in that mode. The image is so clear that I can
see the jagged edges when motion is viewd. But I will be hooking
in my computer soon. Cant wait. This panel can handel 1024x768.
When HDTV comes out I'll be ready.
Later
Mitch
Right now Im using a white bed sheet. Couple of wrinkles.
Im doing the rear screen thing, and need to knock a wall down.
But I already have the screen material (Grey). I unrolled it, held it
in front of my projector and the image on the other side is good
in image and contrast (little less brighter). Make sure image is projected on dull side. The side you watch TV on is shinny. This prevents the light from going into the stuff behind.
The sreen needs to be streched over a square frame.
It is I think 52"H and 64"W
Later
http://www.rosco-ca.com/products/screens/roscoscreen.html
Right now Im using a white bed sheet. Couple of wrinkles.
Im doing the rear screen thing, and need to knock a wall down.
But I already have the screen material (Grey). I unrolled it, held it
in front of my projector and the image on the other side is good
in image and contrast (little less brighter). Make sure image is projected on dull side. The side you watch TV on is shinny. This prevents the light from going into the stuff behind.
The sreen needs to be streched over a square frame.
It is I think 52"H and 64"W
Later
http://www.rosco-ca.com/products/screens/roscoscreen.html
HI-
Since there seems to be some confusion about my setup, I will elaborate. I have a MH lamp enclosed in an elliptical reflector. The reflector makes the light from the lamp converge to a point. At that point I need to put a condenser lens. This lens makes the light diverge in a cone. The rate of divergence, is I believe the problem. This rate of divergence of the condenser must correspond to the fresnel I am using, so that I get the light condensing through the fresnel and LCD to a point where the focal length of my objective can focus the light.
Schematically
reflector fresnel LCD objective
( | || }
woneil, I agree that I should match an OHP as much as possible, however I don't have the original OHP to look at. I realize that I must match the condenser to the fresnel to the objective, and I guess that is my problem. I think that my fresnel/LCd has to be closer to my reflector, however, the condenser must make the cone of light expand fast enought to encompass my LCD or I will get uneven illumination. Therefore there is some miniumum length that my fresnel/LCd must be from the condenser. This minimum length dicatates the focal length of the fresnel, and hence the focal length of my objective. The problem I am having is that they don't match each other.
I can't do much testing until I get my new condenser from surplus shed, so I won't be able to do too much till then. However, I do want to say that I think a elliptical reflector will help, NO MATTER WHAT THE SIZE! I will image your bulb light to a small point, which you can use as a point source. It is very efficent as woniel says.
Anyway I will keep in touch, and try and get some pictures tomorrow.
J
Since there seems to be some confusion about my setup, I will elaborate. I have a MH lamp enclosed in an elliptical reflector. The reflector makes the light from the lamp converge to a point. At that point I need to put a condenser lens. This lens makes the light diverge in a cone. The rate of divergence, is I believe the problem. This rate of divergence of the condenser must correspond to the fresnel I am using, so that I get the light condensing through the fresnel and LCD to a point where the focal length of my objective can focus the light.
Schematically
reflector fresnel LCD objective
( | || }
woneil, I agree that I should match an OHP as much as possible, however I don't have the original OHP to look at. I realize that I must match the condenser to the fresnel to the objective, and I guess that is my problem. I think that my fresnel/LCd has to be closer to my reflector, however, the condenser must make the cone of light expand fast enought to encompass my LCD or I will get uneven illumination. Therefore there is some miniumum length that my fresnel/LCd must be from the condenser. This minimum length dicatates the focal length of the fresnel, and hence the focal length of my objective. The problem I am having is that they don't match each other.
I can't do much testing until I get my new condenser from surplus shed, so I won't be able to do too much till then. However, I do want to say that I think a elliptical reflector will help, NO MATTER WHAT THE SIZE! I will image your bulb light to a small point, which you can use as a point source. It is very efficent as woniel says.
Anyway I will keep in touch, and try and get some pictures tomorrow.
J
hey, I gotta question. On those LCD screens that you can buy that don`t have a covering over them, they have two wires sticking out. One is red and the other is yellow.
Does anybody know what those are for? Are they for splicing a RCA cable to it. Cause its says they are made for DVD, TV, Cars, ETC.
Thanks,
Mitch
Does anybody know what those are for? Are they for splicing a RCA cable to it. Cause its says they are made for DVD, TV, Cars, ETC.
Thanks,
Mitch
Here, but you can`t see he red and yellow wires. Its just two that are sticking out from the back
http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=1371718636
http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=1371718636
Mitch I hate to break it to you. But those look a lot like laptop panels and the fact that they are linking you to www.earthlcd.com is not a good sign. It looks to me like they are selling laptop lcds and making it look like they are specially designed for cars.
Hey everybody! It's been a while since I came here.
Anybody selling a cheap panel? If I can't get one, then I'll just go back to my computer hardware phase... then come back every month to check up on things. 😀
I still have my nView Spectra with power adapter sitting here. I'm thinking it's a bad cord cuz the one I have now has a male/female adapter on it (for some reason, all the ones on eBay have the same thing. I figure they got the wrong cable too, so they're trying to ditch it. Coincedence?) Still going For $20 + shipping. Unless somebody's nice enough to find me a male VGA to female EGA cable.
Anybody selling a cheap panel? If I can't get one, then I'll just go back to my computer hardware phase... then come back every month to check up on things. 😀
I still have my nView Spectra with power adapter sitting here. I'm thinking it's a bad cord cuz the one I have now has a male/female adapter on it (for some reason, all the ones on eBay have the same thing. I figure they got the wrong cable too, so they're trying to ditch it. Coincedence?) Still going For $20 + shipping. Unless somebody's nice enough to find me a male VGA to female EGA cable.
EURIKA!!!!!!!!!
I think I have finaly figured out why some people can get a good image and some can't.
Everything must match...
The light source must sit at the focal point of the first fresnal and it must iluminate the entire lens.
next the panel
next another fresnal
and finaly the objective lens.
The hard part is, the second fresnal lens creates a larger virtual image of the lcd. The objective lens must have a focal length that is equal to or greater than the diagnal of the virtual lcd but is also large enough to catch the entire light cone from the second fresnal.
Now for #s
light => fresnal 12"
panel as close as possible to first fresnal
panel => second fresnal 1"
second fresnal=> objective 12"
assuming 12" fl on both fresnals and 13" fl on the objective lens.
I am going to buy the rest of the parts I need this weekend and see if it works.
Wish me luck
Joe
I think I have finaly figured out why some people can get a good image and some can't.
Everything must match...
The light source must sit at the focal point of the first fresnal and it must iluminate the entire lens.
next the panel
next another fresnal
and finaly the objective lens.
The hard part is, the second fresnal lens creates a larger virtual image of the lcd. The objective lens must have a focal length that is equal to or greater than the diagnal of the virtual lcd but is also large enough to catch the entire light cone from the second fresnal.
Now for #s
light => fresnal 12"
panel as close as possible to first fresnal
panel => second fresnal 1"
second fresnal=> objective 12"
assuming 12" fl on both fresnals and 13" fl on the objective lens.
I am going to buy the rest of the parts I need this weekend and see if it works.
Wish me luck
Joe
Undream, tech head,
I totally agree. You have to match ALL the optics up. Marlakar was partially lucky, and partially smart. Using all the stuff from a single OHP was smart, he was lucky to have all that stuff 🙂
There are 3 condensers available at Surpuls shed, two have similar f #'s, and one is much faster. I ordered the Melles Groit one, and will check that out.
I found this website which is partially useful for those of use trying the Marlkar method, but mixing and matching optics:
http://www.cairnweb.com/tech/tech_lamp1.html
This site is talking about a Xenon lamp, but the principles are the same. It explains why with no condenser you get a donut beam (which I get). It then at least partially explains how the focal lengths must match up.
J
I totally agree. You have to match ALL the optics up. Marlakar was partially lucky, and partially smart. Using all the stuff from a single OHP was smart, he was lucky to have all that stuff 🙂
There are 3 condensers available at Surpuls shed, two have similar f #'s, and one is much faster. I ordered the Melles Groit one, and will check that out.
I found this website which is partially useful for those of use trying the Marlkar method, but mixing and matching optics:
http://www.cairnweb.com/tech/tech_lamp1.html
This site is talking about a Xenon lamp, but the principles are the same. It explains why with no condenser you get a donut beam (which I get). It then at least partially explains how the focal lengths must match up.
J
Condenser/Fresnel combo
http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=1371429214
Look at this projector. One of you could rob the fresnel and condenser off of it.
This is the same projector that Marklar had originally (Dukane 680).
It does not work. I don't want it.
http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=1371429214
Look at this projector. One of you could rob the fresnel and condenser off of it.
This is the same projector that Marklar had originally (Dukane 680).
It does not work. I don't want it.