Hi all, I've been meaning to ask this question for some time.

It seems as though savy lumenlab members are backing up "split fresnel" design, which I do not quite understand why.

I understand a condenser fresnel lens between the light source and the LCD panel is needed, however IMHO the collector fresnel between LCD panel and projection lens will inevitably degrade the projected picture quality. Necessity for this collector fresnel seems to be justified here as "light concentrator" for the projection lens, which I would rephrase as --> A common projection lens simply can not cover 10+ inch image circle, so collector fresnel has to first "shrink" the size(diameter) of the LCD panel so that the projection lens will be able to cover the diameter of the whole panel (In other words, a fast(low f-stop) lens with 10+ inch image circle, that is, a large format camera lens such as Rodenstock Sironar is too expensive)

Collector fresnel does save money but the projected image suffers A LOT (I believe elusive problems discussed here such as "Glowies" root from the collector fresnel itself)

So, why not an OHP lens assembly approach such as using 3M OHP Head design? (A head design that a front surface mirror is placed above a triplet lens in 45 degree angle) <-- This design eliminates need for the collector fresnel lens and IMHO will produce much better projected image than Split design, though I admit I have experimented with neither of these designs.

What am I missing here..?

This question is asked a lot.


http://www.lumenlab.com/forums/index.php?s...st&p=183734

Google search: site:lumenlab.com +split vs +unsplit

Well.. I happened to know a little about Optics.. (I got a LF Sinar camera).. and back to back dual fresnel setup behind LCD panel (Unsplit) will only make the projection worse since LCD panel is rather a "thick" board and the light passing holes in that thick board at non-perpendicular angle will produce all kinds of optical problems.

Bedrock of my point is --> why not totally do away with "dual fresnel" setup (Split or Unsplit) and go fo 3M OHP head assembly setup with single condenser fresnel behind the panel?

(There are simple and efficient ways to correct keystoning with OHP head approach)

Are these 3M OHP setups different from other OHPs which have dual fresnels sandwiched together as one unit?

Hmm.. irrelevant.
I said single condenser fresnel behind the LCD panel for DIY projector design (We are not designing OHPs, are we?)

OHPs place a back to back dual(Unsplit) condenser fresnels below the stage since the space between the head assembly and the stage is "open" (Light needs to be forcefully directed to the projection lens) However "enclosed" DIY projector is another story. Enclosing the space between the head assembly and the LCD panel with bellows or a cone(or a pyramid)-like enclosure will do.

A possible drawback for this setup is loss of light since the inner surface color of the bellows or pyramid needs to be non-glossy black for optimal picture quality. (Light will be absorbed to the bellow (or to the pyramid): Prolly upto 1.5x amount of light maybe required).

So you're saying that the Light Flow section of the The Lumenlab PRO DIY Projector Guide V.3 is not the best way?

If you have a better way, please feel free to build it and show us the results.

"Best" is a relative term. It seems to me Lumenlab design is best for low power consumption but not for the picture quality. I do plan to build one this summer, though no proof is needed to see that both Split and Unsplit design will only degarde projection quality rather significantly.

Sorry I asked.

np

I always enjoy being told I'm irrelevant in a sarcastic response.

Although both projection optics and camera optics abide by the same optic rules there are difference. A camera operates more like an infinite/ finite conjugate system where as a DIY projector operates as a finite /finite conjugate system. Put another way, a camera lens collects diffuse light. A DIY projector collects light from a point light source (theoretically) and directs it through the objective lens.
The majority of the projector lenses used here are in fact OHP lenses no different than the 3M OHP lens you have mentioned.
Both fresnels are needed to control the light through the triplet and onto the screen, split or un-split. Using just one will be inefficient. If you were to simply remove the front fresnel the majority of the light would simply not pass through the triplet. One option, if using a single fresnel, would be to increase the distance to the light source so that light would pass. But fresnels aren’t designed for this.
I do agree that it isn’t desirable to have a fresnel in between the LCD and the triplet. But sometimes it’s the only option with what is available. The Samsung 10.6” LCD is one such panel where diverging light causes very bad contrast issues , the only known fix is to use a fresnel in front of it so that collimated light passes.
A fresnel in front of the LCD won’t shrink the apparent size of the LCD, it will in fact increase the size. The triplet lens has to have a FOV large enough for the LCD in the first place.
If we had access to a very large triplet lens we could conceivably create a projector like you have mentioned.

I suggest you do a little more reading.

DJ

Yes we are. At least the same principals.


Unfortunately it won’t do. It will be extremely inefficient.


Light will be absorbed/reflected by the walls. 1.5 times the light is an extreme under estimation.

DJ

Sorry I missed this post.

I have a camera as well.
What will it be worse than?



Maybe I’m missing something here. Can you draw a simple diagram to explain it a bit better.



Lens shifting comes straight to mind. But it still relies on the FOV of the triplet.

DJ

DAZZZLA, come on... Quite frankly you do not understand what "image circle" is (Image circle is measured at film side, not object side. Otherwise most lenses would have infinitely large image circle). In OHP head setup, it is the "object side" the triplet faces the stage (LCD) whereas in lumenlab design (and for all film/digital cameras except lens-reversed macro-shooting setup) it is the "film side" that faces the LCD (You fail to grasp the fundamental difference in optics between OHP head design and lumenlab design. Learn what the front-surface mirror is for in 3M OHP head design) Any projection triplet's image circle is very very small and for that matter, most lenses' image circle is very very small hence the very purpose of collector fresnel in lumenlab design IS to compensate this small image circle. (BTW, it is Xtremely difficult to make a lens with 2.5+ inch image circle at low f-stop. Why do you think medium and Large Format process/enlarging lenses are so expansive?)
I wouldn't even touch other points you made that are simply not correct.

***
I do agree though, more than 1.5x light may be required.

And Daveak, you need a metaphoric condenser fresnel lens placed in your mind to straighten up inputs coming into you.

I would appreciate a productive discussion...

With or without a camera?

your choice

Um... I don't know what to say about your last post...

Let’s back up a bit. Would you like to elaborate a little more on this statement?

EDIT:
And maybe some info on the better way to do it.

I'll be very frank with you. You claim to have a Large Format camera, but do not even understand the concept of "image circle" and made a claim such as projection lenses have enough image circle to cover 10 inch LCD panel. I would rather not waste my time on you. Sorry.

(Again, I would love egoless, productive discussion with experts...)

Actually let’s back up allot.
I’ve tried to list all of your points as separate as I could.


A DIY projector lens (OHP lens) is designed to resolve an image (the screen) of an object (LCD) that is greater than its diameter. For example the standard 320-330 mmFL triplet can create an image from an LCD of up to 17” diagonal. It doesn’t require a fresnel to do this or any other optics. Shine a very bright light onto the surface of the LCD and the triplet will resolve it as an image on screen, the light would of coarse be enclose so as not to saturate the image. This is not the most efficient use of light, most of this diffuse light would simply not go through the triplet.
When a fresnel is placed in between the LCD and triplet it will create a virtual image of the LCD that is larger, not smaller. It is then the virtual LCD image that is used as the object by the triplet to create an image on screen. The distance that the fresnel is in front of the LCD will determine how much magnification will be introduced. It is better to keep this to a minimum to avoid seeing the aberrations the fresnel introduces.
My main point is that the triplet is quite capable of resolving an LCD that is larger than itself.



The front surface mirror does nothing but change the principal axis 90°. You can remove the mirror and turn the OHP on its side and get the same image on screen.



It needs to be “forcefully directed” for a DIY projector as well.



My understanding of an image circle (search on Google as I’ve never heard that term used before) is to do with covering the film with a circular shaped light overlapping enough to cover the rectangular film. Maybe you are confusing this term with FOV of a lens.



There is no difference between an OHP and a DIY projector in terms of lens orientation. I think it is you that fail to grasp the theory.



As I said before, to change the principal axis.

I’m terribly sorry that you feel as though I’m egotistical.
If you don’t want to waste your time on me could you at least explain to everyone else your ideas for an improved design?

DJ

I already did and please.. You do not know what you are talking about.

Sorry you don’t like my answers, your choice.

Usually I’d just ban but I’m in a happy mood at the moment, thanks.

Pity a moderator at lumenlab has to look up google for concept of "image circle" and still claim to own a large format camera... All large format camera users know what image circle is, I say you are a liar. And a projection lens with image circle of 17'? Please...

Um, where did I say I own a large format camera?

Up there

A link would be nice. or perhaps a post number.
DJ

look for it yourself.

Sorry, couldn't find it.

Take your time
(And also on finding a projection lens with 17 inch image circle...) zee..

I have to say this was the least expected thing from lumenlab forum, not to mention from a moderator of lumenlab...

Here’s one that will resolve a 17”LCD

Well this guy isn't going to win any friends soon.

You'll find this a very very useful forum if you just tone things down a little....

You will be lucky to get 5 inch image circle out of that lens. You DO need to understand the fact that this lens works in lumenlab design since the collector fresnel lens PRE-REDUCE the apparent diameter of the LCD panel. Please quit making ludicrous claims anymore, you are only making lumenlab look bad.

No YOU are making yourself look bad.

And now you're banned !!!!

Note to members reading this. Calling a moderator a liar is NOT a good move.

It's a shame he was so hostile. I was wanting to see what he had come up with for his build.

Oh well.

Props to Dazzzla for taking the high road and not banning wuxga himself (as he was "too close to the argument") But a word to the wise, we mods stick together 'round these parts...

Bottom line, keep it civil and you can stick around. Go around insulting people and you may get a tongue lashing or two but if you learn the error of your ways you can stick around. Trolling a mod, however, now that is just plain stupid...

Whoa. Late to the party.

I'd like to make a point about "more light needed."

The condenser fresnel is used to essentially cancel square law losses at the point where the light hits it, gathering up the light so that more isn't needed.

IF (And this is a big if) the light is diffuse enough so that it can/will still evenly strike the projection lens from the entire surface area of the lCD, you have to reintroduce square law losses again. Call the fresnel setup about 85% efficient at that point, this allows some losses for reflections, losses at the fresnel rings, and light that still doesn't strike the triplet. Using this determination, if you have 100 lumens leaving the LCD, you will transmit 85 lumens.

Now let's look at the alternative, as it was explained.

Your base efficiency (at the minimum distance that the triplet could possibly capture the whole LCD) is the ratio of area for the triplet vs the LCD size at that viewing angle. Assume a 15" standard lens build.The standard lens has an aperature of a bit under 1.5 square inches. We'll round up to be generous. The 15" LCD has an area of about 108 square inches to start. Say that we narrow it down to a 45 degree viewing angle (In practice it would have to be larger than this. I'm being generous here. A typical throw gives a LCD to lens distance of about 360mm (Default in the focal calc for a 120" projection) At 360mm, area increased at 45 degrees, the LCD viewing area (the area that the 100 lumens at the LCD would be scattered over) is increased by the same 360mm on all sides, or about 14.1" So instead of 12" X 9" we have to add 28.2" to each dimension. That gives us a collection area of 1495.4 square inches. Given our 1.5 square inch lens, this gives an efficiency of roughly 0.1%. In order to get the same 85 lumens on the screen, we'd need to have about 84,700 lumens at the LCD, so it's no matter of 1.5 times the light. It's 847 times the light.

You can't compare a projector to a camera in any but the most basic abstraction. It does not, and cannot work in the same manner. We make a camera work by increasing the sensitivity to light, and by exposing the film for a longer time. It works because the film stores the information. A projection screen does not.

Ah what a bummer, I really do like to see a pompous ass work his magic. I was hoping he/she/it wouldn't go too far and get banned but rather be formulaically dismantled...not like most pompous asses would ever admit that they were wrong.

A letter to Mr.Pompous Ass:
Since you have never created a projector before but you have a camera, please follow your instructions and create a projector and prove everyone wrong...and take pictures....with your camera.

Well if dazzzla is so unknowledgeable that he is a waste of time I would hate too know what the rest of us newbies are thought of. I would prefer you just take your 15 posts of glory and go hang with a more expert group. I hate trolls. And you have yet to supply any of us with working results or even a plan.

Wow..... is this forum flat earth society? I've been reading this forum quite a while and got occasionally amazed how so-called savy members line up behind some moderstors who simply do not have knowledge in Optics. Reminds me of a StarTrek NG episode where Data was scape-goated by ignorant aboriginals not knowing it is Data that will save their lives.

WUXGA's claim is correct, in fact there is nothing wrong in what he said. There are no-sub $700 lens that are fast enough to use as a projection lens that's got larger than 5" image circle. PERIOD.

A projection lens of 17" image circle at $200? OMG....

Any Given lens has its unique image circle characteristics and this image circle characteristics has nothing to do with any other factors of the lens such as folcal length etc. and as WUXGA has said, it is very difficult to manufacture lenses large image-circled, yet fast. Ask any lens experts(or LF camera users) about a low f/stop lens of 17" image circle at $200. They'll laugh at you.

There are lenses used in copymachines that's got 17" image circle but their f/stop is 11 (And these lenses go at least $150 used), which renders it useless for projection use. (This is why copier machines flood that bright a light).

I am quite surprised and disappointed some moderators here have not even heard of image circle concept. It is a basic knowledge in Lens Optics.

After all, it is the collector fresnel lens that makes cheapo projection lenses sold here work in Lumenlab design by pre-shrinking the apparent LCD image before it reaches the lens.

Yeah, we've heard it all before.
(Is this the same guy?)
Now, please, either prove your concept and document it, or mosey back to your LF Camera forum.
We're happy with the way things are done here - unless you can find a way to do it better and less expensively.

Why prove a fact? I'm going to advanced design forum. Bye

Camera lenses are one section of optics. A projector, telescope, microscope and lasers are all other sections of optics. Each section use certain terms more often than the others. Sometimes they are the same and have the same meaning. Sometimes they are section specific, different terms used in different sections of optics but have the same meaning. F# or F number is one term that stems from photography. In other optics sections F# is termed as numerical aperture (NA). Another section specific term is diopter used in optometry.
Now to this image circle term. I think you are referring to the same term used here as field of view (FOV), which is section specific to here at LL. A more accurate term I could use here is field angle.

Now to your assumption that the lenes used on DIY projectors can only resolve a 5” circle, you are mistaken. There are hundreds, if not thousands, of projectors using lenses that resolve a large object. The OHP lens used in an OHP is resolving the transparency which is larger than 5”.
Removing the front fresnel and using just one is not a good an efficient way of collecting light. I’m not sure why you are so keen to remove it. If it is behind the LCD it won’t degrade the image, the triplet can’t see it.

DJ

... amazing that some people like to spend their time typing such hateful comments... it takes away from what a forum is really about, communication and discussion. Not everyone is going to agree with each other but at least here we can make a hypothesis and then comment back and forth to see if it is really doable...

I too have been googling "lens circle" and just don't see how it applies to us. I have been trying to find a diagram that would visually explain this... anyone out there have one that they can share???

Heres a link to some image circle info that may or may not help.

Image circle info

Here are a couple of links I found regarding image circle.
http://photonotes.org/cgi-bin/entry.pl?id=Imagecircle
http://www.largeformatphotography.info/for....php/t-592.html
The second link discusses image circle and its conversion to angle of coverage. It relates to how well a particular lens can cover a rectangular film. In projection the film is our screen but we work on a similar principal at the rear of the lens back to the LCD. Basically it suggests, at least what I could gather, that image circle is how large the film (LCD) can be without blocking light and degrading the edges of the photo. And the angle of coverage is just the image circle at the FL of the lens stated in degrees instead of image circle (inches or mm).
So it seems that the image circle in photography is the same as our Field Of View / field angle.

DJ

And some conversions for a standard 320mmFL triplet used on a 15” LCD.

Image circle= 15” 381mm
Angle of coverage= 61° for a 12000” diagonal screen (close enough to compare to infinity)

BTW the focal calc will give you the triplet field angle.

EDIT:
These values are for projection purposes. I have no idea how they would relate to photography.

Here's a couple of interesting threads to read - apparently created by the same person (back in 2005).

Large Format Photography Forum: DIY Projector using a LF Process lens

diyAudio Forums: A solution for projectors with 17" LCDs!

Here's an interesting Google search which will show how often the two terms "image circle" and "DIY projector" appear on the same page: +"image circle" +"DIY projector"
Not a whole lot there.

We know what we do is not perfect - it's not a perfect world. But it works for us. And speaking for myself, it works well.
For this guy to come onto the LL Forum and say we're wrong and to call us liars... well, it's just wrong. The proof is in the puddin', and we've got plenty of documentation to show that our members are definitely enjoying their puddin'.

I imagine card board and duct tape would have constituted much of his build

But no worries, he registered a new account last night right after he was banned.

BTW, he was making no sense at all.



Now, I might just be making a fool of myself... but doesn't the condenser "condense" the light into a point and go between the LCD and the lens, not the LCD and the light? (maybe he meant collimator) And don't you pretty much "need" to have a condenser between the collimator and lens, to help reduce the image to a size the lens can handle?

I dunno, guess I'm the stupid one.

I would actually consider his points if there werent hundreds of documented, working projectors here that all use projection lenses that can be had for less than $200!