Does light that hits the rear fresnel from outside the cone (straight lines drawn from an estimated circle around the point of light to the outer edges of the fresnel) only cause pollution or does it still add to the brightness of the image? The same might be said for light hitting the triplet.
If this light is only polluting then why not put a solid barrier with a circular hole which only allows the cone light to reach the rear fresnel and the same just before the triplet.
Click to view attachment
Full Version: Masking Light inside the pj
This has been discussed and I've always wanted to try something like you're suggesting but never have and at the speed I move likely never will. Any off-axis light you keep out of the triplet is a good thing so far as I can discern. I think I've been told you'll probably only realize an incremental projection quality increase and using your "masking" technique should theoretically improve contrast - IIRC. Why not give it a try? 
There are some notable lumen geniuses here who've presented theoretical stuff about what light is good and what light is bad, etc. This is an easy thing to try but your results may be difficult to document and quantify. Good luck!
There are some notable lumen geniuses here who've presented theoretical stuff about what light is good and what light is bad, etc. This is an easy thing to try but your results may be difficult to document and quantify. Good luck!
I've been very lazy and mostly using a cheap pre-evo China built svga pj. It's convenient but at larger sizes the screen door effect is quite pronounced so I miss the better quality image of my 400 watt 15" (1024x768) or 17" (1280x1024) diy.
I've put off rebuilding for about a year but I have to due to space and placement issues. I have 2 complete sets of everything plus spare lamps (1 pro and 1 standard) and could even build another standard if I bought another ballast. While I'm redesigning I thought I'd look at ways to improve the image.
I'm convinced that a front flat black mask along with a black painted interior in front would be very beneficial by stopping most of the stray useless light from hitting the triplet. I know I would have to make the mask sheet slide back and forth to find the sweet spot and it might have to be rectangular to match the shape of the lcd. My thought is that it's not really a light cone leaving the field fresnel but an ever decreasing rectangular shape of the lcd. An expansion of this idea would mask to the shape of the input image ie... 4:3 or 16:9 etc... but thats another story and would require a more elaborate and flexible masking design since some movies letterbox more than others. I'll have to give that some thought too but it would be in addition to the existing rectangular basic mask.
I'm not so sure about the rear of the projector.
The light area is more of a question to me. Does any of the bouncing reflected light contribute positively to the brightness of the image. In other words would any of this light end up collimated even though it's not directly from the point of light source.
My feeling is yes it does contribute but the only proof I have is some old tests I did with cfl lights. Now I know cfl and fresnels aren't a good match but please hear me out these were tests. I had one light at the fl distance and centre of the fresnel and a second one a few inches away. As I recall when I switched off the second bulb I got about 10% to 20% less light. These are just guesstimates since I don't own a light meter. My experiment was also flawed because it was a temp box with light pollution escaping from the edges of the fresnel and lcd holders so what appeared like a brighter image may have been light pollution entering the triplet.
My point is that masking out light before the rear fresnel might reduce what could be useful light but I'm not sure and can't really prove it. I could design a mask for that also when I rebuild but hate to waste time and money on something if it has no benefit.
I've put off rebuilding for about a year but I have to due to space and placement issues. I have 2 complete sets of everything plus spare lamps (1 pro and 1 standard) and could even build another standard if I bought another ballast. While I'm redesigning I thought I'd look at ways to improve the image.
I'm convinced that a front flat black mask along with a black painted interior in front would be very beneficial by stopping most of the stray useless light from hitting the triplet. I know I would have to make the mask sheet slide back and forth to find the sweet spot and it might have to be rectangular to match the shape of the lcd. My thought is that it's not really a light cone leaving the field fresnel but an ever decreasing rectangular shape of the lcd. An expansion of this idea would mask to the shape of the input image ie... 4:3 or 16:9 etc... but thats another story and would require a more elaborate and flexible masking design since some movies letterbox more than others. I'll have to give that some thought too but it would be in addition to the existing rectangular basic mask.
I'm not so sure about the rear of the projector.
The light area is more of a question to me. Does any of the bouncing reflected light contribute positively to the brightness of the image. In other words would any of this light end up collimated even though it's not directly from the point of light source.
My feeling is yes it does contribute but the only proof I have is some old tests I did with cfl lights. Now I know cfl and fresnels aren't a good match but please hear me out these were tests. I had one light at the fl distance and centre of the fresnel and a second one a few inches away. As I recall when I switched off the second bulb I got about 10% to 20% less light. These are just guesstimates since I don't own a light meter. My experiment was also flawed because it was a temp box with light pollution escaping from the edges of the fresnel and lcd holders so what appeared like a brighter image may have been light pollution entering the triplet.
My point is that masking out light before the rear fresnel might reduce what could be useful light but I'm not sure and can't really prove it. I could design a mask for that also when I rebuild but hate to waste time and money on something if it has no benefit.
Blacking out the "front" of the projector is for sure much more important - in general - than doing the "back".
As for the shape of the mask(s) they'll best be rectangular - like the LCD.
I'd like someday to whip up something for mine which I can in real time slip in and out of the projection path to see if I can really "see" a difference. (Ha! Fat chance!
)
As for the shape of the mask(s) they'll best be rectangular - like the LCD.
I'd like someday to whip up something for mine which I can in real time slip in and out of the projection path to see if I can really "see" a difference. (Ha! Fat chance!
)
This is a "lo-fi" version of our main content. To view the full version with more information, formatting and images, please click here.