LOL, i think mines fairly right, from memory the specs on the diylabs triplet has somthing like a 70 or 80deg FOV so i must be close. I dont do it till it like starts to distort or where the light cuts out, you see a sweet spot that all of a sudden drops away very rapidly and thats where i took my measurment, they wer combined angles aswell not per side.
Its a shame your not in madrid as i could show you what i mean and how i do it. Ive got some big condensers here that you can clearly see the FOV with your eyes, that same FOV i see with my eyes i soon see on the screen 🙂, and see what the lens has as in limits in the real world environment. Prety much the shorter the focal and the larger the lens is the wider the FOV (not counting projection lenses with varible FOV's 😀 ). One thing i think you may be mixed up on is the clear arperature and the FOV in your measuring, to this day i havnt found a lens with a full side to side FOV and you wont either, its typical of them having a ratio of 3 arperature and 2 FOV so 2/3 of the lens is FOV. Though not aways somtimes its higher other times its less, it just depends on the design and type of the lens.
In the manufactures test labs im sure alot of them use lasers for the fov (did read that somwhere though ages ago), they use paralelle light for the focal finding. There is a formula for the FOV on a single lens, stuffed if i know what it is on a multi ellement lens, kinda gets complicated then.
Trev🙂
field angle
Rox,
You DID determine the maximum field angle of the lens with your laser test. You know that light can get through the lens at a certain angle. It just may not give you very good performance at that angle. But "good performance" is a relative term.
LCD projection lens don't have to be all that good! The maximum spot sizes used for triplet design are good enough for photography. Film grain sizes are very small compared to the size of an LCD pixel. If a projection lens can get most of a pixel's light within the 2 mm pixel space on the screen, then it is good enough.
If you can get all the light through a corner pixel into the lens, so the most extreme ray is less than the maximum field angle, then all of that light will get through the lens. Here is a drawing of the first lens surface of the triplet. The red oval is the light that comes through one corner pixel. (It has the oval shape because it comes from all the areas of the lamp arc.) Since the part of the oval closest to the edge of the lens has a lower angle from the central axis than the maximum angle, all of this light will get through the lens.
Rox,
You DID determine the maximum field angle of the lens with your laser test. You know that light can get through the lens at a certain angle. It just may not give you very good performance at that angle. But "good performance" is a relative term.
LCD projection lens don't have to be all that good! The maximum spot sizes used for triplet design are good enough for photography. Film grain sizes are very small compared to the size of an LCD pixel. If a projection lens can get most of a pixel's light within the 2 mm pixel space on the screen, then it is good enough.
If you can get all the light through a corner pixel into the lens, so the most extreme ray is less than the maximum field angle, then all of that light will get through the lens. Here is a drawing of the first lens surface of the triplet. The red oval is the light that comes through one corner pixel. (It has the oval shape because it comes from all the areas of the lamp arc.) Since the part of the oval closest to the edge of the lens has a lower angle from the central axis than the maximum angle, all of this light will get through the lens.
Attachments
ace my good friend, 😀 if it was 70 degrees field angle, we could use 19" lcd with 32cm triplet 😀 and you know that 15" is the largest we can use with it.
where did you see it is somewhere 70 degrees the field angle for this 80 triplet? i would say it is far from reality. (i stimate 50 degrees in the best case for this lens)
Why don´t we trust on manufacturers specs? mean who thought it is not 450 mm the focal of the 135 triplet? why shouldent the filed of view be 24?
where did you see it is somewhere 70 degrees the field angle for this 80 triplet? i would say it is far from reality. (i stimate 50 degrees in the best case for this lens)
Why don´t we trust on manufacturers specs? mean who thought it is not 450 mm the focal of the 135 triplet? why shouldent the filed of view be 24?
field angle
Here is what happens if the fresnels are not working right. The blue oval is all of the light going through a corner pixel. Half of it is falling outside the triplet, so it will not get to the screen.
There are other factors that all contribute to making the corners dim:
1) The condensor fresnel geometry: A 220 mm fl fresnel for a 15" LCD makes the corners 1.8 times dimmer than the center, just by the inverse square law of light intensity.
2) In a non-split design, the LCD may not pass as much light at the corners because the viewing angle is higher.
Many people have built great projectors using the 450 mm triplet and 15" LCDs, so I think your 24 degree FOV figure is too low.
Here is what happens if the fresnels are not working right. The blue oval is all of the light going through a corner pixel. Half of it is falling outside the triplet, so it will not get to the screen.
There are other factors that all contribute to making the corners dim:
1) The condensor fresnel geometry: A 220 mm fl fresnel for a 15" LCD makes the corners 1.8 times dimmer than the center, just by the inverse square law of light intensity.
2) In a non-split design, the LCD may not pass as much light at the corners because the viewing angle is higher.
Many people have built great projectors using the 450 mm triplet and 15" LCDs, so I think your 24 degree FOV figure is too low.
Attachments
well, i am all the time using the same example;
My good friend diyeitor, the famous 135 triplet srtipper, has better image with the same 330 field fresnell when he modified the triplet. We don´t know the actual focal lengh, in fact it is very easy to measure the projection size and the trhow and we can determine the focal, but the only important thing is that the larger focal triplet (lets say 50cm focal) works very well with 330 field fresnell, you only need to check his last results.
then, please start thinking about the field of view angle istead of the field mismaching fresnell.
You know what?, i think the 450mm focal is not true. i think the manufacturers lie on this as well (ironic).
My good friend diyeitor, the famous 135 triplet srtipper, has better image with the same 330 field fresnell when he modified the triplet. We don´t know the actual focal lengh, in fact it is very easy to measure the projection size and the trhow and we can determine the focal, but the only important thing is that the larger focal triplet (lets say 50cm focal) works very well with 330 field fresnell, you only need to check his last results.
then, please start thinking about the field of view angle istead of the field mismaching fresnell.
You know what?, i think the 450mm focal is not true. i think the manufacturers lie on this as well (ironic).
The highest lcd size on the triplet i have and tested works on a 17inch lcd. If you wer to get a longer focal one which are readily avail you can run a 19inch through em no probs. The ohp lenses are designed to have a wide feild of veiw, they are made that way for the stage they have to illuminate, also they would make the FOV wider on a ohp lens to elliminate distortions on the edges for when projecting text.
With my findings of the rear lens being about 80deg FOV and the top being about 40deg FOV, that doesnt neserserally mean that the overal FOV of the lens is 80deg, maybe they combine it ?
You got any answers guy?
The stated FOV for a typical ohp projection lens like mine is about 65-75deg. Ill try to find you a page that states this.
Imo its interesting now that we may have found a differing FOV triplet lens, now we can go and make stuff 🙂.
Trev🙂
answers?
>You got any answers guy?
Nope, sorry! If I had one of those 450 mm fl 135 mm diameter triplets then I would try some FOV tests with a thin laser beam and some projection tests with diffuse backlighting on a 17" transparency.
But I don't have one of them.
I guess the folks that have them will just have to experiment a bit to see what they can actually do with different size LCDs and fresnel systems. (I know that most of the people happy with them are using a lumenlab long fl field fresnel.)
I do have a 600 mm fl Rodenstock process lens that has a terrific maximum field angle. I had to block bits of light leaking around the frame holding my 15" LCD, because this lens projected them on the wall (off the normal screen). My frame is about 1 3/4" wider than the LCD, so I think this lens would work for a 19" LCD.
>You got any answers guy?
Nope, sorry! If I had one of those 450 mm fl 135 mm diameter triplets then I would try some FOV tests with a thin laser beam and some projection tests with diffuse backlighting on a 17" transparency.
But I don't have one of them.
I guess the folks that have them will just have to experiment a bit to see what they can actually do with different size LCDs and fresnel systems. (I know that most of the people happy with them are using a lumenlab long fl field fresnel.)
I do have a 600 mm fl Rodenstock process lens that has a terrific maximum field angle. I had to block bits of light leaking around the frame holding my 15" LCD, because this lens projected them on the wall (off the normal screen). My frame is about 1 3/4" wider than the LCD, so I think this lens would work for a 19" LCD.
Inkog said:
What kind of stuff?!?
LOL, ive got a few tricks up my sleeve, including a new light engine that recycles lost light, ill try that in a few days 🙂.
I basically made that comment as the more we know or discover things, it opens up our horizons in constructing our own custimizable lens system with available on the shelf parts. Making a projection lens isnt that hard, and the more we learn about them, the closer we come to customising/building our own.
Trev🙂
ace3000_1 said:
Yes please.
And to you that would like to meassure the field angle, if we don´t know the criteria, how could we measure the angle?
I mean I could say; This is the point where I CONSIDER THE MAXIMAL ABERRATIONS FOR MY SETUP, SO HERE IS MY MAXIMAL FIELD ANGLE. But this is not the criteria manufacturers use.
For instance, Ace says that 32focal tirplet does good work with 17" (we should know the throw he uses, since is very important as well) then others say that it is 16" the maximun LCD we can use with it. The only value i would agree with is the one stated by the manufacturer specs, this is the functionality of the specs, so enginers do know if it will work or not before they try it.
If someone finds the exact criteria to determine the Field angle, then ok, you can meassure it using the same criteria as manufacturers but until then i will not trust on any subjetive meassurement of field angle.
Kool!
I tried to make 1 with a polarizing reflector and 1/4 wave retarder plate in front of a first surface mirror.
Idea was the light of the right polarity goes through to the LCD. Light with wrong polarity is reflected, turned 90deg, hits the mirror, turned 90deg and goes to the LCD with the right polarity!
Great!
Only problem is the 1/4 wave retarders only work at specific light frequencies so you need 1 for red, 1 for blue , 1 for green which is damned expensive. Also all these layers absorb a bit so you dont gain that much. I torpedo'd it.
I'd love to hear your solution.
Guys,
I hate to admit it but one thing you should be aware of and Rox is right about this, is that the laser test will not give you the correct FOV. It will only tell you the maximum angle that the lens system can pass light through it.
The problem with the laser method is that it is only one wavelength of light. Or at least a very narrow band. The FOV is the angle where at least three critical light wavelengths for daylight have been corrected for. The FOV is where the correction has been made to a designed standard where the spot sizes of these three wavelengths are sufficiently small and aligned to be considered acceptable performance for the intended application.
The lens will still image to a lesser degree outside of the FOV.
Now it would not surprise me based on the cost of this lens that it only has a 24 total angle FOV like Rox claims the manufacturer has said. The lens was probably cobbled together from existing stock lenses to make it cheap. This would explain why the edge blurriness is a problem and the only way to deal with it would be to reduce the corner to corner distance of the LCD by using a 4:3 aspect ratio or smaller LCD.
That being said we must as Guy said put things into perspective. THis is the best lens for a 17 inch LCD for a low cost that we have at the present time. And we in all likelihood will have to get by with it until someone can get a manufacturer to have a fully optimized large format triplet made. This will cost a lot more. The last quote that someone got out of India was about 500 - 600 USD per lens for at least 100 quantity.
Rox,
If you use a high quality 10 or 12 inch LCD with this lens you should have a lot less edge problems.
I would rather live with the edge problems until I can get a better lens either by making it by hand or something else.
Hezz
I hate to admit it but one thing you should be aware of and Rox is right about this, is that the laser test will not give you the correct FOV. It will only tell you the maximum angle that the lens system can pass light through it.
The problem with the laser method is that it is only one wavelength of light. Or at least a very narrow band. The FOV is the angle where at least three critical light wavelengths for daylight have been corrected for. The FOV is where the correction has been made to a designed standard where the spot sizes of these three wavelengths are sufficiently small and aligned to be considered acceptable performance for the intended application.
The lens will still image to a lesser degree outside of the FOV.
Now it would not surprise me based on the cost of this lens that it only has a 24 total angle FOV like Rox claims the manufacturer has said. The lens was probably cobbled together from existing stock lenses to make it cheap. This would explain why the edge blurriness is a problem and the only way to deal with it would be to reduce the corner to corner distance of the LCD by using a 4:3 aspect ratio or smaller LCD.
That being said we must as Guy said put things into perspective. THis is the best lens for a 17 inch LCD for a low cost that we have at the present time. And we in all likelihood will have to get by with it until someone can get a manufacturer to have a fully optimized large format triplet made. This will cost a lot more. The last quote that someone got out of India was about 500 - 600 USD per lens for at least 100 quantity.
Rox,
If you use a high quality 10 or 12 inch LCD with this lens you should have a lot less edge problems.
I would rather live with the edge problems until I can get a better lens either by making it by hand or something else.
Hezz
Hezz said:
THANKS!!!!! (I really thank that at least one of the menbers here agree in some litle point with me) thanks for this hezz.
I agree with you. I think tis lens is a good choice for our setups yet. But WE HAVE TO ASSUME SOME LIMITATIONS THAT MOSTLY CAME FROM THE TRIPLET ITSFELF, NOT FROM THE FIELD FRESNLELL AS WOULD BE THE UNFOCUSED CORNERS AND A MAXIMAL THROW SO THIS DISTORTIONS KEEP UNDER ACCEPTABLE VALUES).
I have been well maneredly kicked from a popular forum right now (my rights have been suddenly dropped to a point where i can´t even post replys, this is a well maered way for me :d)
Thisis right now the only forum where the information can´t be delted if the owner doesn´t like it. Actually I understand that if someone sells an item, doesn´t like to read it isn´t the correct item for the setup on his own forum. But I don´t like people being cheated. I really started myinvestigation for me, but then thought this information is very usefull for the comunity.
just let me tell you that i would say 80 triplet would do similar work with 17" or even better than 135 triplet. Thats because the fov on the 80 triplet is much larger than the 135's. (but the bad side of the 80's is the aperture, this is obyously better on the 135)
Finally let me tell you that it was me the most disgusted person when found this len's field angle.
anyway my challenge is open yet for you that try a setup with 135.
http://www.diyaudio.com/forums/showthread.php?threadid=55209
I really thank you verymuch hezz. Have a good day.
now here comes the one millon question;
we have told to 3dlens our ideal fresnell lens. This lens was bassed on 135 triplet and a 100" image. A thing that it has not been checked/demostrated yet. Do we agree with this setup now? mean is the 550 fresnell our ideal fresnell lens?
this was a present from the 3dlens owner, and i start thinking we did things too fast. Please what do you think about?
we have told to 3dlens our ideal fresnell lens. This lens was bassed on 135 triplet and a 100" image. A thing that it has not been checked/demostrated yet. Do we agree with this setup now? mean is the 550 fresnell our ideal fresnell lens?
this was a present from the 3dlens owner, and i start thinking we did things too fast. Please what do you think about?
ROX,
Something you should know about the 80 mm lens. I had that lens and tested it and even though it has claimed wider FOV it does not have as good an image as the 135 mm with a 15 inch test image and given the same light illumination.
Meaning that it is designed to an even lower standard of resolution.
But in the end both lenses will work and what you should really be thinking about is where your projector will be placed and how big the image is. This will determine what focal length lens that you need.
For me the 327 mm FL will not work as I need 450 - 500 mm FL.
Hezz
Something you should know about the 80 mm lens. I had that lens and tested it and even though it has claimed wider FOV it does not have as good an image as the 135 mm with a 15 inch test image and given the same light illumination.
Meaning that it is designed to an even lower standard of resolution.
But in the end both lenses will work and what you should really be thinking about is where your projector will be placed and how big the image is. This will determine what focal length lens that you need.
For me the 327 mm FL will not work as I need 450 - 500 mm FL.
Hezz
Light recycling: slightly off-topic, but I haven't seen it discussed recently. I'm very happy to since I'm in the middle of tinkering with this. I've obtained a few 3M Vikuiti film samples (DRPF, DBEF), unfortunately all are "diffuse" since that's appropriate for wide-angle LCD backlights. Do they have one that's totally clear / non-diffuse? I guess I should ask my contact there, but wondering if anyone here had experience.
Also obtained a quarter-wave retarder sample. Yes there is a peak wavelength, my sample is 440nm but I think it would still do an ok job of the reds and blues? I'm having a hard time lining up all the fast axes though! I barely even know what a fast axis is. Is there a good thread to discuss this on? I'd love to hear more about your results me2.
c
Hezz said:
yes i agree that image size and throw is the most important spec to check in our designs. But we shouldn´t forget about the field angle like have been done untill now. this is the next most important thing for me.
Lets say i try to do a projector with 15" and 80mm slide projector lens. I work out the throw, image size.... and everything looks like it would work. The main problem if i do it would be the field angle from the triplet would be just projecting a 3" central image of my 15" lcd. Thats becaouse i didn´t take care of the field angle.
it is you the first one who told me there are limitations with the throw with this lens.
And it is you as well te one that told me you got the triplet at 52cm from the lcd. If we add 20mm gap and say triplet optical center is at 6cm fartcer then you would use 2+52+6=600mm focal field. Don´t you think the 550 ideal is precipitated?
And it is you as well te one that told me you got the triplet at 52cm from the lcd. If we add 20mm gap and say triplet optical center is at 6cm fartcer then you would use 2+52+6=600mm focal field. Don´t you think the 550 ideal is precipitated?
The plate I got was for a red scanning laser. It was designed for ~700nm. Yes, everything goes a mild pink.
Yours might work in blue. 440nm is in the purple. Not sure why they gave you a sample for that region. Unless you mean 440labda? That would be red. Usually these things are fairly specific to the wavelength. YOu might have the best source because their stuff is designed for LCDs and might have multi-layered coatings already in. Did it come in multi sheets? I'll have to check out thei stuff. Let me know how yours turns out 🙂
- Status
- Not open for further replies.