Warehouse reflector testing, Talk about massive! Options

[shivers20]

My father brought this reflector home today for me to play around with. I placed my halide bulb in the vertical position and the amount of light increased drastically. Probably 10x brighter than without the reflector or with the bowl reflector I have been using for my projector. I did not notice any hotspotting but then again I only shined it on the wall so I cant really be sure. I am going to try and get some screen shots up by the end of the weekend. The downside to it would be that the bulb has to be in the vertical postion which causes burn out quicker and the reflector is huge maybe 24 inches across.

 DCP_0525.JPG ( 22.16K ) Number of downloads: 46

 DCP_0528.JPG ( 34.5K ) Number of downloads: 45

 hhhh.JPG ( 13.29K ) Number of downloads: 49

[Shrivel]

I don't think that reflector will help a projector much - it looks as though it's designed to reflect the light OUT, rather than straight back to the source. But I could be completely off my rocker

[mikyd1954]

actually it looks vaguely ellipsoidal (which could be good)...but only vaguely

This post has been edited by mikyd1954: Dec 10 2005, 04:03 PM

[SIMUL8R]

Interesting, it might work well without the collimating fresnel and just use a diffusing plate to get even brightness. Something like what Mark and I beleive you mikyd are working on with that white light box, if I'm not mistaken.
sim

[arizonavideo]

If you look at the reflector surface there is a hundred or so facets this is important because each one will over lap the light on to the other and focus on a larger area than one point of light [the engineers know that the lamp has a 90mm arc or so] so you will get a softer and much more even light.
It is close to an ellipsoidal reflector so it has two focal points the fl may be 10 or 20. An ellipsoidal light system should have an 85% efficiently compared to 30% for a spherical one and that’s when the spherical one uses a condenser lens. Our current system may only be 20% to 25% efficient.
A couple of questions. What determents the size of the arc image at the triplet? We know the image will be fully diffused so we will not see the lamp like with a perfect dish. Second with this reflector their is a fair amount of non- culminated light. Will this harm the focus?
I am looking forward to the results.

This post has been edited by arizonavideo: Dec 12 2005, 09:20 AM

[shivers20]

Interesting, it might work well without the collimating fresnel and just use a diffusing plate to get even brightness. Something like what Mark and I beleive you mikyd are working on with that white light box, if I'm not mistaken.
sim

Yeah great idea Sim, I was also thinking the same thing. It looks promising. One of the reasons why we use the collimating fres is to collect and evenly distribute the light. I measure the diameter of the reflector (top part outer lip) and its 17inches across. The light bouncing off the reflector would be enough to cover a 17inch panel by the looks of it. All I have to do is avoid hotspotting which I think I can, it should work. I am going to build a test box later on in the day and continue to do some more testing. Has anyone ever seen this type of reflector? The insides are odd-shaped look almost prism like. Keep you guys updated.

[arizonavideo]

You can look at each facet as a mini lens or reflector system designed to add to the one next to it with an overlap to even out the light

[phutton]

The only thing that would concern me is that the light may not be collimated when it hits the lcd. But there is only one way to find out. Put it in there and see what you get.

[shivers20]

I don't think that reflector will help a projector much - it looks as though it's designed to reflect the light OUT, rather than straight back to the source. But I could be completely off my rocker

Yeah but dont you end up with the same outcome? Our goal is to get a lot more light through the fres. Why not just bypass the straight back to the source method? especially if I can get more light this way. We shall see.
I drew a picture to give you a better idea on how I plan on setting it up. NOw that I think about it more and more, Sim is probably right when he says that I wont need a collimating lens.
 ll.jpg ( 4.52K ) Number of downloads: 44

[shivers20]

You can look at each facet as a mini lens or reflector system designed to add to the one next to it with an overlap to even out the light

Interesting stuff Arizona. Thanks!

[Shrivel]

Yeah but dont you end up with the same outcome? Our goal is to get a lot more light through the fres. Why not just bypass the straight back to the source method? especially if I can get more light this way. We shall see.
I drew a picture to give you a better idea on how I plan on setting it up. NOw that I think about it more and more, Sim is probably right when he says that I wont need a collimating lens.
 ll.jpg ( 4.52K ) Number of downloads: 44

That's fine then. I was assuming you were using the fresnel.

[shivers20]

I sold all of my parts to raise money for the pro lens so I cant test it properly. I used a 17x17 piece of white paper just to get an idea of the evenness of the light. I place the sheet 220mm from the bulb and it looked evenly spread. I was going to build a rouch enclosure but the snow outside said I dont think so! Here are some more pics.

I found some old LL fresnels in my basement. I placed one lens in front of the light source and I got a huge image of of the arc. I also saw some of the reflector's facets on my white wall. Why do you think this would occur?

 IMG_0253.JPG ( 27.08K ) Number of downloads: 43

 IMG_0254.JPG ( 25.95K ) Number of downloads: 42


This post has been edited by shivers20: Dec 11 2005, 01:42 AM

[shivers20]

 IMG_0257.JPG ( 16.13K ) Number of downloads: 45

 IMG_0255.JPG ( 29.31K ) Number of downloads: 47

[shivers20]

 IMG_0259.JPG ( 19.48K ) Number of downloads: 47

[arizonavideo]

For the life of me I can’t figure out how to get the large cone of light from the reflector to focus into the triplet?

[mikyd1954]

I sold all of my parts to raise money for the pro lens so I cant test it properly. I used a 17x17 piece of white paper just to get an idea of the evenness of the light. I place the sheet 220mm from the bulb and it looked evenly spread. I was going to build a rouch enclosure but the snow outside said I dont think so! Here are some more pics.

I found some old LL fresnels in my basement. I placed one lens in front of the light source and I got a huge image of of the arc. I also saw some of the reflector's facets on my white wall. Why do you think this would occur?

well, first off you will need the collmating fresnel..light needs to hit the lcd within a certain range of angles, basically just a little (7% or so) off perpendicular to the lcd panel, thats what the collimating is for, the only reflector that you don't need a collimating fresnel for is a parabolic one, because it produces collimated light..... you got that on the wall because thats probably what was in the focus of the fresnel.... same thing happens in a pj if you focus on the fresenls(or the reflector) instead of the lcd.... take a big piece of cardboard and nove it in front of the reflector(towards the reflectror then further away) until you get the smallest spot of light(assuming there is one)

[mikyd1954]

For the life of me I can’t figure out how to get the large cone of light from the reflector to focus into the triplet?

thats what the front(collector) fresnel is for....

[mikyd1954]

try using a transparency and the front fresnel to see if you can get a projection of the transparency... also try moving the bulb further in and out of the reflector to find the first focal point....

[shivers20]

I placed a fres behind a sheet of white paper to check for evenness, tell me what you think. I am going to try the fres and transparency test.
 IMG_0264.JPG ( 10.93K ) Number of downloads: 42

 IMG_0265.JPG ( 15.45K ) Number of downloads: 45

 IMG_0266.JPG ( 7.08K ) Number of downloads: 43

 IMG_0267.JPG ( 8.82K ) Number of downloads: 44

 IMG_0268.JPG ( 14.31K ) Number of downloads: 46


I used a cd jewel case and I projected an image of it on the wall. I seem to be having a problem with the reflector's facets showing up on the wall as well. I also see a bit of rainbow effects going on. It could be that the fresnel is too close or too far from the bulb, not sure. As for the repositioning of the bulb to find its first focal point I can not do, Unfortunately I need an etra hand which I dont have right now, maybe tomorrow.

This post has been edited by shivers20: Dec 12 2005, 06:55 PM

[mikyd1954]

I placed a fres behind a sheet of white paper to check for evenness, tell me what you think. I am going to try the fres and transparency test.

I used a cd jewel case and I projected an image of it on the wall. I seem to be having a problem with the reflector's facets showing up on the wall as well. I also see a bit of rainbow effects going on. It could be that the fresnel is too close or too far from the bulb, not sure. As for the repositioning of the bulb to find its first focal point I can not do, Unfortunately I need an etra hand which I dont have right now, maybe tomorrow.

tomorrow you'll have an extra hand? cool...I could use 3 hands too anyway I'm not sure exactly what those pictures are showing, could you describve each one?

[shivers20]

The pics are just to show how even the lighting is. I was worried that the reflected light would not cover the entire fres or lcd panel. I was also worried that the light would not diffuse evenly. By the looks of the pics (nice pic) especially the fourth one I believe shows it the best. I wish I held onto my triplet... doh! Anyway I am going to go ahead and build a test box.

[Mikau]

Interesting! Yeah there are some big reflectors like that at the parks near me only they're more roundish. I've always wanted to try one in a projector but their too high to climb.

You might try what Lucky did. Just shoot the thing across the room onto a wall, no fresnel or anything. To see how well the light is collimated.

If not, stretching a thin plastic bag with letters on it will work fine to measure light distribution. Who doesn't want gigantic "have a nice day!" letters, spread across their living room wall?

This post has been edited by Mikau: Dec 14 2005, 08:17 AM

[Lucky_Me]

The faucets are showing up because it is a lens that is focusing what it sees. If you use a triplet at the fresnels focal point this effect will disappear. If you have a triplet place it at the focal point of the fresnel. If you do not, try to compare your results with these results:

http://members.shaw.ca/dberladyn/projector/parabolic.html

All of teh distances and explainations are there for you to see what I did. If you can, adjust your digital camera to take pictures as dark as it can. reduce your ISO to 100 or lower and there should be another setting like EV, reduce that as low as possible. If the camera is dimmed if will show you how even the light pattern is. Again refer to the webpage I gave you above. It will show you how to conduct tests to compare your reflector to the parabolic.

P.S. Good Find! I actually looked for one of those but I didn't find one at the local used building supply.




#1051

This is what happens with the parabolic reflector, a fresnel and no triplet. You shouldn't have any problems if you have a triplet.

This post has been edited by Lucky_Me: Dec 14 2005, 09:13 AM

[Dzignr]

Shivers,
I have seen some reflectors like that around here. I am sure those facets are to help even out the lamp image when projected 30 feet or so down to the ground. They are used for large warehouses and things like that.

The optical engineer here at work that has been helping me with my reflector told me that if the light is uneven coming from my large reflector I can diffuse it with frosted thermal glass. I am running into some roadblocks here at work trying to buy one for myself...it may be easier just to go dig around in the junk closets and borrow one permanantly . I will let you know when I have more to talk about regarding this reflector. We may be able to reduce the wattage of the HID lamp and still get great output.

This post has been edited by Dzignr: Dec 20 2005, 08:44 PM