What exactly is this Pro Lens from Lumenlab? I've seen posts on their public forum stating that it's been long missed for about a year now, but nothing about it's spec's or prospective uses.
😀, yes I did not xplaing very well last time.
I mean, the diycomunity is really working with any sources, I have seen a lego based proyector 😀 but what I was triyng to say is that a part source would be better than "look if you can find it at ebay..." mmm, I mean there has been done group buys to a non selling companyes... so the diycomunity has found a "valid" source again...
I know there are real bargains at ebay, I have bought some parts there... but I see something strange we to design the ideal fresnell lense (and make it be done for us) based on a "hard finding" 450m wide viewing projection lense. This same procces lense does cost a lot if is bought at a photographic specialiced shop (500$-1000$?)
And concerning about lumenlab prolens, I just gave my opinion to the very friendly brainchild, Just like I have been done here (with the ideal fresnell lense for instance) but sometimes people does not agree with me ( I found this ideal fresnell not "ideal" in my opinion), well that time it looks like he did agree with me about the prolens development.
I would love if we could find a "infinite" 450mm focal source so the fresnells that are being done exclusively for us could be exploited that way.
Just wonder, how many of us has a 450mm focal wide view triplet? maybe looking for a procces lense in ebay is a hard work, now looking for a 450mm procces lense is a really hard work.
mmm, cant wait to the 550mm fresnell 😀 any group buy in the way?? 😀😀
I mean, the diycomunity is really working with any sources, I have seen a lego based proyector 😀 but what I was triyng to say is that a part source would be better than "look if you can find it at ebay..." mmm, I mean there has been done group buys to a non selling companyes... so the diycomunity has found a "valid" source again...
I know there are real bargains at ebay, I have bought some parts there... but I see something strange we to design the ideal fresnell lense (and make it be done for us) based on a "hard finding" 450m wide viewing projection lense. This same procces lense does cost a lot if is bought at a photographic specialiced shop (500$-1000$?)
And concerning about lumenlab prolens, I just gave my opinion to the very friendly brainchild, Just like I have been done here (with the ideal fresnell lense for instance) but sometimes people does not agree with me ( I found this ideal fresnell not "ideal" in my opinion), well that time it looks like he did agree with me about the prolens development.
I would love if we could find a "infinite" 450mm focal source so the fresnells that are being done exclusively for us could be exploited that way.
Just wonder, how many of us has a 450mm focal wide view triplet? maybe looking for a procces lense in ebay is a hard work, now looking for a 450mm procces lense is a really hard work.
mmm, cant wait to the 550mm fresnell 😀 any group buy in the way?? 😀😀
Actually a lot of people have 450mm or so projection lenses, 18" (457.2mm) is a very common opaque projector lens which has been mass produced for that purpose. Between people with those and people with 450mm lenses from diyprojectorcompany or lumens lab there is plenty of market. And there is certainly no shortage of 450mm or so lenses to be found.
yes, i have one as well 😀. This is what i consider a valid part source for instance (let's say infinite available, at rasonable price...)
And this is the lens I thought the fresnell was designed xclusivelly for, thing that has been rejected by one of the fresnell designers...
This 450mm triplet, or 18" focal triplet or 135 triplet or whatever you would like to call it (since I believe is very similar specs all of them) is not a wide viewing lens at all. And I say wide to those that are capable of 45 degrres FOV angle at least (well, a lens that will capture the full lcd source at least). The one at DPC is 24 degrees FOV angle for instance. (This needs to be tested yet but I believe in maths, I mean if the lens was designed to work in X conditions, It won't work on Y conditions as well as it would do at X).
Now I reask my last question, How many of you has a 450mm focal projection lense with stated field angle capable of capturing a 17" lcd (or a 15" if you prefer) at a rasonable throw? (Assume the throw needed for a 100" diag image if you want as it was done for the ideal fresnell design initial axioms). Is there any different from that well known 135 triplet?
And this is the lens I thought the fresnell was designed xclusivelly for, thing that has been rejected by one of the fresnell designers...
This 450mm triplet, or 18" focal triplet or 135 triplet or whatever you would like to call it (since I believe is very similar specs all of them) is not a wide viewing lens at all. And I say wide to those that are capable of 45 degrres FOV angle at least (well, a lens that will capture the full lcd source at least). The one at DPC is 24 degrees FOV angle for instance. (This needs to be tested yet but I believe in maths, I mean if the lens was designed to work in X conditions, It won't work on Y conditions as well as it would do at X).
Now I reask my last question, How many of you has a 450mm focal projection lense with stated field angle capable of capturing a 17" lcd (or a 15" if you prefer) at a rasonable throw? (Assume the throw needed for a 100" diag image if you want as it was done for the ideal fresnell design initial axioms). Is there any different from that well known 135 triplet?
Sorry if I came off a bit harsh, it is late and all.. etc. Anyways as far as wide FOV lenses go is there any definite way to measure on an old lens? (As all mine says is the focal and manufacturer) But I'd assume that most lenses from opaque projectors have a pretty decent FOV because they are designed to capture as much light as possible reflected off of a generally non-transparent object, thus larger FOV brighter image etc.) I just roughly calculated everything with a 17" and came up with an angle of 35 degrees (although I know that is off because I forgot to add in the length of half of the lens and used an example with a 4:3 17" LCD which probably wouldn't be used by most). I'll try it again in the morning with better numbers and see how it comes out. Definitely drop me a line if anyone knows how to precisely/roughly or whatever measure the FOV of a lens though as Rox has gotten me interested now 😀
mmm I think we can´t meassure the Fov "preciselly". All we can meassure are FOV-s such us "this is my max acceptable personal fov on my lens". I mean, Let's say it is not the FOV but the focal what we want to meassure on our lense; I would meassure the distance from the focused point to somewhere between the central part of the lens housing and I could say "it is 440mm focal"
then you came and measure on yours, make the focusing point from the sun something biger than i did and meassure 470mm distance... Then we have relative meassuremts...
Now the FOV meassuremt is somehow the area where the focusing performances are under a critical value, so out of there the spotsize is too big. We have not tools for meassuring it.
Just it was meassured by the manufacturer, why don't we ask them? (Is what i did to get to 24 degrees fov on the awi lens).
Do you think the lens from a opaque has a verygood FOV because it was designed to collect as much as light as posible... mmm I wouldn´t say that, I know it was designed to collect as much as light as posible, that's why it is 125mm effective clear aperture diameter, but it has nothing to do with FOV angle.
Anyway, what people is looking for is a lens where the focusperformances are somehow acceptable, like minoten's projections on his 15.5" and 135 triplet. I understand it being enough for some of you, but since some people is going for hi res panels, we need optics that can handle that resolution (means keep the pixels without distortion on the projection), If not, a 800*600 resolution 17" panel would look as good as 1280*1024 17" panel with those bad focus projection lenses...
then you came and measure on yours, make the focusing point from the sun something biger than i did and meassure 470mm distance... Then we have relative meassuremts...
Now the FOV meassuremt is somehow the area where the focusing performances are under a critical value, so out of there the spotsize is too big. We have not tools for meassuring it.
Just it was meassured by the manufacturer, why don't we ask them? (Is what i did to get to 24 degrees fov on the awi lens).
Do you think the lens from a opaque has a verygood FOV because it was designed to collect as much as light as posible... mmm I wouldn´t say that, I know it was designed to collect as much as light as posible, that's why it is 125mm effective clear aperture diameter, but it has nothing to do with FOV angle.
Anyway, what people is looking for is a lens where the focusperformances are somehow acceptable, like minoten's projections on his 15.5" and 135 triplet. I understand it being enough for some of you, but since some people is going for hi res panels, we need optics that can handle that resolution (means keep the pixels without distortion on the projection), If not, a 800*600 resolution 17" panel would look as good as 1280*1024 17" panel with those bad focus projection lenses...
an image is better than 1000 words;
look at the images, they are form oslo spot size simulation form a demotriplet design (came with oslo). The triplet is originally designed to have 40 degrees FOV angle (20+20 half angles).
The spot size is the lens performances sumary along the lens fov (0 degrees- left to 1-full half field side)
Left is a simulation where the field angle is the stated one (40 degrees) check the spot size keeps minimal. The corrections on the lenses are made so the spot size is keep under the max 0.02 value for the green line and 0.6 value for the blue line.
now right is the same triplet but I throwed 50 degrees FOV to it to see what happens. The performances up to 40 degrees are just like the previows image, but farther than 40 degrees, the spot size falls drastically. The green line takes 4 times the max value on the 40degrees test, and the blue line takes double value than the 40 degrees test.
So this lens will show worse performances on the 40 to 50 degrees range. Now consider the 24 degrees stated FOV on the 135 triplet and make it work on twice the stated FOV (somewhere 46 degrees for the 17") DO you see what a dissaster?
look at the images, they are form oslo spot size simulation form a demotriplet design (came with oslo). The triplet is originally designed to have 40 degrees FOV angle (20+20 half angles).
The spot size is the lens performances sumary along the lens fov (0 degrees- left to 1-full half field side)
Left is a simulation where the field angle is the stated one (40 degrees) check the spot size keeps minimal. The corrections on the lenses are made so the spot size is keep under the max 0.02 value for the green line and 0.6 value for the blue line.
now right is the same triplet but I throwed 50 degrees FOV to it to see what happens. The performances up to 40 degrees are just like the previows image, but farther than 40 degrees, the spot size falls drastically. The green line takes 4 times the max value on the 40degrees test, and the blue line takes double value than the 40 degrees test.
So this lens will show worse performances on the 40 to 50 degrees range. Now consider the 24 degrees stated FOV on the 135 triplet and make it work on twice the stated FOV (somewhere 46 degrees for the 17") DO you see what a dissaster?
Attachments
look at the bottom of this page. 3dlens large fresnel
Thanks to JCB for posting this news elsewhere.
Thanks to JCB for posting this news elsewhere.
nice example, Rox
It's good to see some actual spot size diagrams, so people can understand what we are talking about. The example lens you used is certainly optimized for 40 degrees FOV. It looks like the designer didn't care how it performed at 50 degrees. There are many possible similar designs that would not work quite as well at 40 degrees, but would work almost as well at 50 degrees. It just depends on the design goals. It would be silly to compromise the performance of your lens in the FOV region it needs, just to make it work at an FOV angle it does not need!
Every lens design I have looked at with OSLO has similar curves. The spot size is very low near the center, rises to about half-field, and then falls near the full-field point. (Or it may be the other way around: High at both extremes and low at mid-field.) That is because the designer chose the lenses and spacing to make that happen. He (or she) tried to minimize the average error, while keeping the maximum error below some value.
Another thing to keep in mind: You use the example of the spot size going to 0.08 mm as "bad", but a 0.293 mm LCD pixel with 6X magnification on the screen is 1.76 mm across. That is 22 times bigger than your "bad" spot!
One thing you are absolutely right about: Most opaque projector and overhead projector lenses are designed for a diagonal field of about 14.8", (ie. a 10.5" by 10.5" stage). So they will probably work very well with a 15" LCD, and not so well with a 17" LCD. One limiting factor is the lens length/diameter ratio. These lenses are usually longer than they are wide. That limits the passage of light from wide angles. My process lens is much shorter, because the designer did not want to block the light. I notice that a lot of the 80 mm wide projection triplets are also pretty short.
This is why I suggest a good custom LCD projection lens should be a wider version of a process lens. But when you consider the pixel size, it does not have to be a very good process lens.
It's good to see some actual spot size diagrams, so people can understand what we are talking about. The example lens you used is certainly optimized for 40 degrees FOV. It looks like the designer didn't care how it performed at 50 degrees. There are many possible similar designs that would not work quite as well at 40 degrees, but would work almost as well at 50 degrees. It just depends on the design goals. It would be silly to compromise the performance of your lens in the FOV region it needs, just to make it work at an FOV angle it does not need!
Every lens design I have looked at with OSLO has similar curves. The spot size is very low near the center, rises to about half-field, and then falls near the full-field point. (Or it may be the other way around: High at both extremes and low at mid-field.) That is because the designer chose the lenses and spacing to make that happen. He (or she) tried to minimize the average error, while keeping the maximum error below some value.
Another thing to keep in mind: You use the example of the spot size going to 0.08 mm as "bad", but a 0.293 mm LCD pixel with 6X magnification on the screen is 1.76 mm across. That is 22 times bigger than your "bad" spot!
One thing you are absolutely right about: Most opaque projector and overhead projector lenses are designed for a diagonal field of about 14.8", (ie. a 10.5" by 10.5" stage). So they will probably work very well with a 15" LCD, and not so well with a 17" LCD. One limiting factor is the lens length/diameter ratio. These lenses are usually longer than they are wide. That limits the passage of light from wide angles. My process lens is much shorter, because the designer did not want to block the light. I notice that a lot of the 80 mm wide projection triplets are also pretty short.
This is why I suggest a good custom LCD projection lens should be a wider version of a process lens. But when you consider the pixel size, it does not have to be a very good process lens.
Re: nice example, Rox
that example I used is not a valid one for your LCD pixel size observation, The example was done just to show what happens when we feed with a wider FOV than the "expected" one.
The triplet I used is the one that comes with OSLO, "demotriplet" is the name it has, and it is 50mm focal 40 degrees full field angle stated (or designed for). So the spot size Vs pixel size comparation is not what I was getting to in this example but the performances curve along the FOV.
Now, I agree that If we have a spotsize somewhere as big as the pixel, it would look pretty good on the projection.
Let´s see a fast calc; imagine we upscale this demotriplet lens to 450mm focal lens, then the spotsize will be 450/50=9 times biger as well; 0.08x9=0.72mm so it is more than twice the pixel size (this means that the pixels would be projected diffused, one on top of the other... bad focus on the edges...)
by the way let me tell you something I thing you are doing wrong; the spotsize simulation I did is a "inverse projection" simulation, just think on a photographic camera that takes far light and does focus on the film. Now this is not the case on our projectors, it is like the negative film is outputing the light and it is focused out on the wall, so the spotsize is not aplicable on the negative but on the wall.
Our spotsize needs to keep under the LCD size,i agre with you on that, but if we do the simulation on oslo at same conditions (changing the conjugates so the perfect light source is somewhere at a typical LCD to lens distance and the projection is at a far distance, then the spotsize will show MUCH larger size. But as you said if this value keeps under the lcd pixelsize (considering the magnificatin introduced to pixelsize) then we are safe as well.
Finally let me tell you that I would consider the subpixel size (0.09mm), not the pixel size (0.29mm) for our optical considerations, this makes the things harder for us.
Now that the 550 fresnells are availabe, who would be the first one showing us the 550/135 triplet/15"-17" images?
Guy Grotke said:
that example I used is not a valid one for your LCD pixel size observation, The example was done just to show what happens when we feed with a wider FOV than the "expected" one.
The triplet I used is the one that comes with OSLO, "demotriplet" is the name it has, and it is 50mm focal 40 degrees full field angle stated (or designed for). So the spot size Vs pixel size comparation is not what I was getting to in this example but the performances curve along the FOV.
Now, I agree that If we have a spotsize somewhere as big as the pixel, it would look pretty good on the projection.
Let´s see a fast calc; imagine we upscale this demotriplet lens to 450mm focal lens, then the spotsize will be 450/50=9 times biger as well; 0.08x9=0.72mm so it is more than twice the pixel size (this means that the pixels would be projected diffused, one on top of the other... bad focus on the edges...)
by the way let me tell you something I thing you are doing wrong; the spotsize simulation I did is a "inverse projection" simulation, just think on a photographic camera that takes far light and does focus on the film. Now this is not the case on our projectors, it is like the negative film is outputing the light and it is focused out on the wall, so the spotsize is not aplicable on the negative but on the wall.
Our spotsize needs to keep under the LCD size,i agre with you on that, but if we do the simulation on oslo at same conditions (changing the conjugates so the perfect light source is somewhere at a typical LCD to lens distance and the projection is at a far distance, then the spotsize will show MUCH larger size. But as you said if this value keeps under the lcd pixelsize (considering the magnificatin introduced to pixelsize) then we are safe as well.
Finally let me tell you that I would consider the subpixel size (0.09mm), not the pixel size (0.29mm) for our optical considerations, this makes the things harder for us.
Now that the 550 fresnells are availabe, who would be the first one showing us the 550/135 triplet/15"-17" images?
I forgot to mention;
The 10,5" square on the opaques, I wonder what is the throw for that FOV. In fact, the most important value is the field angle, (24 degrees as I have been told) so something is worng there...
And I believe it is no problem at all for working at something wider FOV than the "expected one by the lens", but I find 46 degrees while the expected is 24 degrees not correct. Well it all depends on what performances we want on our projectors.
The 10,5" square on the opaques, I wonder what is the throw for that FOV. In fact, the most important value is the field angle, (24 degrees as I have been told) so something is worng there...
And I believe it is no problem at all for working at something wider FOV than the "expected one by the lens", but I find 46 degrees while the expected is 24 degrees not correct. Well it all depends on what performances we want on our projectors.
prime said:
Awesome. I'll be putting in my order later tonight when I get home along with a 220mm fresnel.
@Rox and Guy- My projector doesn't have a case or anything because it is hidden behind a wall therefore it is easy to swap stuff in and out, so when I get the 550mm fresnel I'll post some images with a mock up 17" and 15" and then my actual 14" LCD. I have always found that no matter what the data says the best way to evaluate most things is just to try it. That way we'll at least have an example of how an opaque lens will perform with an opaque projector lens.
ok tjh, thanks for your colaboration. As much as posible of tests will show a more realistic result, then we can have an idea of how it works. Good luck
Actually, one quick question before I order: Does it make any difference if the two fresnels used have different groove pitches?
Because the 550mm is .2mm groove pitch where as the the 220mm focal 310mm x 310mm fresnel is .5mm groove pitch. (the only other one with .2mm groove pitch is only 10in wide, and not big enough for even my 14" LCD.)
Because the 550mm is .2mm groove pitch where as the the 220mm focal 310mm x 310mm fresnel is .5mm groove pitch. (the only other one with .2mm groove pitch is only 10in wide, and not big enough for even my 14" LCD.)
diferent groves pitch
no problem at all.
just wonder you are placing fresnells split or both between lamp and lcd?
no problem at all.
just wonder you are placing fresnells split or both between lamp and lcd?
Thanks for the quick response Rox!
As far as your question goes: I prefer to place them both together between the lamp and LCD. I used to do it the other way but I found it magnifies the image and it may have been just me but it seemed to be worse quality. Although that is an on-going debate in the DIY community.
As far as your question goes: I prefer to place them both together between the lamp and LCD. I used to do it the other way but I found it magnifies the image and it may have been just me but it seemed to be worse quality. Although that is an on-going debate in the DIY community.
If your fresnel is 550fl and your 135 triplet is 450fl then how far apart should the 2 be in a split-fresnel design?
reed if you read this thread from the begining, you will find all that info. The first post was from the 3dlens owner asking for the "ideal" fresnell, then there are some assumtions took... check the posts Guy grothke wrote... you will find your answer there.
WOW! that was fast. My 550mm fresnel arrived about 30 mins ago by standard air mail.
I thought I would have had more time to work on my box.
Thanks jimhuang, 3Dlens.
p.s. I don't have one of the 135mm diameter, 450mm focal lens
- Status
- Not open for further replies.