(post 3 of 3.)
8. Irising at the lamp arc
If you partially obstruct the light path by sticking your finger in the way, you'll see a very out-of-focus image of the obstruction in the screen. The closer to the lamp arc image, the more out of focus. I've read a post where someone recommended sticking a piece of electrical tape on the front of the triplet to reduce a hot-spot in the center of the screen. This sounded ridiculous! You'd think it would obstruct 50% of the image. Well, it does... but in an out-of-focus way. The image of this tape of this will be so out of focus, it will smooth out almost completely, effectively reducing the brightness of the center of the screen. (instead of blackening the center, like you might expect).
Many triplets have their center lens smaller, effectively creating an iris. Engineering the projector so the lamp arc image is formed at this iris, will have a beneficial effect on crispness, with hardly any tradeoff in brightness as long as the lamp arc image cross-section fits into the iris. Guy's lens sounds like it's even smaller, so I guess he is trading brightness for crispness.
Another quote from Guy:
9. Commercial projectors
So why do commercial projectors use such small expensive bulbs, and why can we get away with such large cheap ones? Assuming both types of projector use similar quality and size lenses?
I believe the answer is that our huge LCD is a very optically forgiving subject, much more so than the tiny LCDs in commercial projectors. The lamps in those projectors are limited by the size of the LCD. Our lamps are limited by the precision of the triplet.
10. Summary
A decent lens will correct for a decent amount of arc size. The engineering challenge is to pick a lamp and a triplet which are balanced with each other in terms of price-performance.
11. Conclusion
There are some good diagrams in the following thread:
http://www.diyaudio.com/forums/showthread.php?s=&threadid=47096
I have to admit I didn't start to grok this stuff until I set up a test bench (using a single Staples page magnifier, a plain incandescent bulb, and a cheap copy lens triplet). I guess I had to muss with it hands-on. If you're still reading, you must be more patient than me!
Anyways, to help answer the original thread question, you could benefit from the bigger bulb IF your triplet was of sufficient quality.
Also, big thanks to everyone on this forum who has explained this same thing, more precisely, and repeatedly, to all us beginners.
8. Irising at the lamp arc
If you partially obstruct the light path by sticking your finger in the way, you'll see a very out-of-focus image of the obstruction in the screen. The closer to the lamp arc image, the more out of focus. I've read a post where someone recommended sticking a piece of electrical tape on the front of the triplet to reduce a hot-spot in the center of the screen. This sounded ridiculous! You'd think it would obstruct 50% of the image. Well, it does... but in an out-of-focus way. The image of this tape of this will be so out of focus, it will smooth out almost completely, effectively reducing the brightness of the center of the screen. (instead of blackening the center, like you might expect).
Many triplets have their center lens smaller, effectively creating an iris. Engineering the projector so the lamp arc image is formed at this iris, will have a beneficial effect on crispness, with hardly any tradeoff in brightness as long as the lamp arc image cross-section fits into the iris. Guy's lens sounds like it's even smaller, so I guess he is trading brightness for crispness.
Another quote from Guy:
9. Commercial projectors
So why do commercial projectors use such small expensive bulbs, and why can we get away with such large cheap ones? Assuming both types of projector use similar quality and size lenses?
I believe the answer is that our huge LCD is a very optically forgiving subject, much more so than the tiny LCDs in commercial projectors. The lamps in those projectors are limited by the size of the LCD. Our lamps are limited by the precision of the triplet.
10. Summary
A decent lens will correct for a decent amount of arc size. The engineering challenge is to pick a lamp and a triplet which are balanced with each other in terms of price-performance.
11. Conclusion
There are some good diagrams in the following thread:
http://www.diyaudio.com/forums/showthread.php?s=&threadid=47096
I have to admit I didn't start to grok this stuff until I set up a test bench (using a single Staples page magnifier, a plain incandescent bulb, and a cheap copy lens triplet). I guess I had to muss with it hands-on. If you're still reading, you must be more patient than me!
Anyways, to help answer the original thread question, you could benefit from the bigger bulb IF your triplet was of sufficient quality.
Also, big thanks to everyone on this forum who has explained this same thing, more precisely, and repeatedly, to all us beginners.
This is really a great discussion.
Your post is very good Charlie. I liked reading and thinking about it. I do have a slightly different idea of the way the objective lens works so maybe Hezz (he's designed lenses?) can clear it up for us. If I understand you the lens takes the smeared out image (cone dia.) from a big arc and focuses it to a point on the screen.
My take is the lens corrects for the distance tavelled to different areas of the screen for all wavelengths. Imagine the distance from the middle of the LCD to the middle of the screen through the lens. relatively short. From the top of the LCD to the top of the screen is quite a bit further. 1 minte of arc from the LCD would be wider (accurately turned into 10min.) lets say and deflected out the proper amount for the bigger screen. The colours would be off because of the different defelctions of the different wavelengths so the lens does this too.
If the lens corrects for the cone made from a large source passing through a point why do they go through the trouble of making the expensive small arc lamps? Just make a lens to fix it. No they cant.
Your local cineplex will pay thousands for a small arc xenon because a lens cant do this.
A lens will adjust to equal lengths of rays and equal relative angles.
ps this is left to right: 😉
Your post is very good Charlie. I liked reading and thinking about it. I do have a slightly different idea of the way the objective lens works so maybe Hezz (he's designed lenses?) can clear it up for us. If I understand you the lens takes the smeared out image (cone dia.) from a big arc and focuses it to a point on the screen.
My take is the lens corrects for the distance tavelled to different areas of the screen for all wavelengths. Imagine the distance from the middle of the LCD to the middle of the screen through the lens. relatively short. From the top of the LCD to the top of the screen is quite a bit further. 1 minte of arc from the LCD would be wider (accurately turned into 10min.) lets say and deflected out the proper amount for the bigger screen. The colours would be off because of the different defelctions of the different wavelengths so the lens does this too.
If the lens corrects for the cone made from a large source passing through a point why do they go through the trouble of making the expensive small arc lamps? Just make a lens to fix it. No they cant.
Your local cineplex will pay thousands for a small arc xenon because a lens cant do this.A lens will adjust to equal lengths of rays and equal relative angles.
ps this is left to right: 😉
Attachments
Your part 3 really has me scratching my head. Thx, I was hoping to get some sleep tonight. lol
I still think mine is correct but im wavering with your point
8-10. 😕
I'll reduce it to the fundamentals of the logic
If the lcd creating the image was a piece of sphere (bent in 2 axis)the proper distance from the lamp at the center, and the screen was a piece of a bigger sphere also with the lamp at the center you would need no lens.
In addition if the light was not point- like (bigger like a basketball) a lens in between the lcd and the screen no matter what the shape would not be able to fix it.
I think my theory stands.
I still think mine is correct but im wavering with your point
8-10. 😕
I'll reduce it to the fundamentals of the logic
If the lcd creating the image was a piece of sphere (bent in 2 axis)the proper distance from the lamp at the center, and the screen was a piece of a bigger sphere also with the lamp at the center you would need no lens.
In addition if the light was not point- like (bigger like a basketball) a lens in between the lcd and the screen no matter what the shape would not be able to fix it.
I think my theory stands.
excellant discussion
Wow, charlie10: You really put some time into those. Very nice complete discussion of the issues.
But on point 3, my particular LCD does diffuse the light a bit. If I remove the LCD and the projection lens, then the fresnels give me a perfect image of the lamp arc on a piece of paper where the projection lens goes. If I put the LCD back in and drive it with a pure white image, I don't see the arc image anymore. Instead, I see a much larger blob of light.
And for Me2!: I think they make small arc lamps so they can make a very compact system. All of the optics and LCD are at the same scale. My 15" LCD projector is about 4' by 2' by 2'. I have seen commercial LCD projectors smaller than a car battery! A film projector has to use a small arc lamp because the film itself is so small. If you tried to use a large arc, then the light would be going through the film at a lot of angles that would not make it through the lens. Cinema projectors operate right on the edge of melting the film. You don't want extra unusable light heating up the film more.
Also, lots of people have made perfectly good video projectors using CRT projection lenses. These lenses go right up against the CRT or LCD so they can collect all of the diffuse light coming from a CRT surface or in our case, from a large diffuse light source behind the LCD.
As for masking unwanted light at the projection lens, this is very simple: The light coming from an extreme end of the arc will end up at the extreme end of the arc image in the projection lens. If the arc image is 100 mm wide and the lens only lets an 80 mm wide beam through to the screen, then you just masked the unwanted 10% of the arc at both ends.
I agree with charlie10: If you have a very small arc length, then you don't need a very good lens. If you have a very good lens (that can correct all the aberrations and distortions and flat-image plane issues) then you can use a large arc lamp. If you have a crappy lens and a large arc, then you will get a bad image.
Wow, charlie10: You really put some time into those. Very nice complete discussion of the issues.
But on point 3, my particular LCD does diffuse the light a bit. If I remove the LCD and the projection lens, then the fresnels give me a perfect image of the lamp arc on a piece of paper where the projection lens goes. If I put the LCD back in and drive it with a pure white image, I don't see the arc image anymore. Instead, I see a much larger blob of light.
And for Me2!: I think they make small arc lamps so they can make a very compact system. All of the optics and LCD are at the same scale. My 15" LCD projector is about 4' by 2' by 2'. I have seen commercial LCD projectors smaller than a car battery! A film projector has to use a small arc lamp because the film itself is so small. If you tried to use a large arc, then the light would be going through the film at a lot of angles that would not make it through the lens. Cinema projectors operate right on the edge of melting the film. You don't want extra unusable light heating up the film more.
Also, lots of people have made perfectly good video projectors using CRT projection lenses. These lenses go right up against the CRT or LCD so they can collect all of the diffuse light coming from a CRT surface or in our case, from a large diffuse light source behind the LCD.
As for masking unwanted light at the projection lens, this is very simple: The light coming from an extreme end of the arc will end up at the extreme end of the arc image in the projection lens. If the arc image is 100 mm wide and the lens only lets an 80 mm wide beam through to the screen, then you just masked the unwanted 10% of the arc at both ends.
I agree with charlie10: If you have a very small arc length, then you don't need a very good lens. If you have a very good lens (that can correct all the aberrations and distortions and flat-image plane issues) then you can use a large arc lamp. If you have a crappy lens and a large arc, then you will get a bad image.
Charlie10 you have successfully explained the way I think optics behave, well done.
Another take on why small commercial projectors use small arcs is the ratio of arc size to LCD size.
15”lcd/1” arc = 15:1.
2” lcd/ 0.2” arc = 10:1.
It’s all a matter of scale. I would assume that the aberrations would also be affected by this ratio as well.
DJ
Another take on why small commercial projectors use small arcs is the ratio of arc size to LCD size.
15”lcd/1” arc = 15:1.
2” lcd/ 0.2” arc = 10:1.
It’s all a matter of scale. I would assume that the aberrations would also be affected by this ratio as well.
DJ
Ok I got it. I was confusing the concepts of distortion and focus. When the light came on I realised I had struggled over this same problem years ago and forgotten all about it. Thanks for clearing that up of me all, especially charlie with your fine post.
Thanks for the responses!
I wish I could edit that first post to note that section 3 is <ahem> malarkey <ahem>, but now it's too late! I should've known not to ramble about something I have no direct experience of (since I have yet to strip an LCD actually
).
Anyways, really glad the rest of it made sense! 😀 cool.
I wish I could edit that first post to note that section 3 is <ahem> malarkey <ahem>, but now it's too late! I should've known not to ramble about something I have no direct experience of (since I have yet to strip an LCD actually
). Anyways, really glad the rest of it made sense! 😀 cool.
anyway i would lke to listen how would you use that par30 reflector and the 150 W lamp. How big is the reflector?
Yes... blackout I'm sure you didn't expect to get such a complicated set of answers to your simple question!
I read that the "30" in "par30" refers to size, in terms of 1/8ths of an inch. So a Par30 bulb should be 30/8 = 3.75" in diameter. I believe this would not be big enough to light the LCD, all 11000 lumens would hit the middle of the screen and none to light the corners.
Is there such a thing as a Par30 elliptical reflector? Par60?
I read that the "30" in "par30" refers to size, in terms of 1/8ths of an inch. So a Par30 bulb should be 30/8 = 3.75" in diameter. I believe this would not be big enough to light the LCD, all 11000 lumens would hit the middle of the screen and none to light the corners.
Is there such a thing as a Par30 elliptical reflector? Par60?
charlie10 said:
parX is not the size, X means the angle of the rays output.
A par60 is twice the par30. Thinking a par reflector would give you paralell light is wrong (it would be par0) but the number is very important.
elliptical reflector lamps
OSRAM makes some lovely 250 Watt and 400 Watt MH lamps in their own dichroic elliptical reflectors. Look at their HTI series.
Only problems are that they run for less than 1000 hours, and are pretty expensive. Originally, they made them for medical instruments with fiberoptic lightguides.
If somebody made an aluminum elliptical reflector that would fit a $65 US 10000+ hour lamp, then that would be very competitive with designs that use spherical reflectors and pre-condensor lenses.
OSRAM makes some lovely 250 Watt and 400 Watt MH lamps in their own dichroic elliptical reflectors. Look at their HTI series.
Only problems are that they run for less than 1000 hours, and are pretty expensive. Originally, they made them for medical instruments with fiberoptic lightguides.
If somebody made an aluminum elliptical reflector that would fit a $65 US 10000+ hour lamp, then that would be very competitive with designs that use spherical reflectors and pre-condensor lenses.
I googled for "Par30 eighths" and found the following helpful bulb reference.
http://www.servicelighting.com/catalog_bulb_finder_bbt_incandescent.cfm
According to that reference, the "30" refers to diameter not angle. Might not be the only measurement system, I don't have the experience to know...
After finding out about the bulb lumen distribution, I'm very curious again about ellipsoidal reflectors.
Tried to look up stage fixtures. Searching ebay for "leko" turns up some options in the $100-200 range, that's for the whole fixture. Seems like a reflector should be cheaper though.
http://www.servicelighting.com/catalog_bulb_finder_bbt_incandescent.cfm
According to that reference, the "30" refers to diameter not angle. Might not be the only measurement system, I don't have the experience to know...
After finding out about the bulb lumen distribution, I'm very curious again about ellipsoidal reflectors.
Tried to look up stage fixtures. Searching ebay for "leko" turns up some options in the $100-200 range, that's for the whole fixture. Seems like a reflector should be cheaper though.
on the link you posted, there are some pictures with more than one par number (par 30, 38, 60). It could be that they use it to mean the diameter of the reflector aperture but i have never seen it before. I can buy a 50mm refelctor dicroic alogens from par12, par24, par36, par42, par 60... all of them looking very similar but the only difference is the light angle at the output.
- Status
- Not open for further replies.