Who's up for more experimentation?

I just bought this BEF (Brightness Enhancement Film) off of ebay. For those of you who can remember Mark writting about this way back but the the cost for one of these is astronomical like 10 sheets of 17"x17" for $380 from 3M. It is believe this recycling film will increase brightness by another 60% and with 2 films up to 120%. I bought one just to try.

what is it? is it some kind of polar replacement or something placed over the polar itself?

I bought 2 for testing.

Are you going to be using it in a standard build with a point source light?

Tgreenwood

Not exactly sure, I had thought it would be a replacement high quality polar back then but from this sellers description I'm assuming he just added it to his lcd's with good results. For the price I'm willing to find out and hopefully if this becomes beneficial pass the info on to everybody (maybe a group buy?). Afterall, it seems to be what we do best around here in LL



Rock on JJ, always can count on you to venture into testing new innovations and/or ideas



By the looks of it tgreen, anything goes. Maybe even your Fluorescent concept might benefit from this. I'm more or less broadcasting this for those willing to experiment. I'm thinking at this price it's a small risk worth looking into.

Per the description from the 3M brochure, it isn't a polarizer:

You know what, I'll jump in and experiment with this also. 2 sheets on the way .

It is just another film layer that goes in between the diffuser and the Lcd. It has me wondering if there is some sort of requirement for it to be fairly close to the light source and lcd. Because it is kind of described as being like a fresnel with grooves.

Oh well we'll soon see

Absolutely right JJ, it's appears to be prismatic and from the description when orientating the grooves can help with the spread of light vertically or horizontally across the LCD. This might help greatly for vigneting (dark corners) which would appear brighter overall. BTW, it also mentions compressing the light which is kinda like what a precondenser does.

Further thinking about this since it is placed between a diffuser and LCD it might be possible to eliminate the rear fresnel and place these films over a frosted glass plate like polished opal then place the LCD on top of that and then the front fresnel after. This will shorten the optics by removing the rear focal distance from lamp to rear fresnel....hmmm...speculations, speculations

Wow, so I just received my BEF in the mail today. Talk about fast shipping and handling.

Me too...but they could have put it in between some sturdier cardboard. I have an edge that is bent a bit

Provide your findings when you can Garf

Well mine came in.

Has anyone else noticed that it is the exeact same material that is already placed behind the diffuser in all the Lcds that we use.

It is the same film that they place between the Lcd and the backlight in all the lcds that I have stripped.

I have 4 backlights with the diffusion and enhancement layers and you can't tell the difference between them and the 3m.

Which leads me to the wonderfully scientific conclusion that it is the same stuff

i added some between my lcds and backlights and it just made the Lcd darker. Maybe better contrast but definitely darker. I don't think it will help to enhance brightness if there is already 2 sheets of the enhancing film already there.

But I will try it on my Diy today just to see, but i don't think it will do anything.

I notice this too JJ.

I used the 7" build that's destine for my son. It didn't help and it only got darker when adding 2 placed 90 degrees behind the lcd I'm leaning that it doesn't work in our application unless we are approaching the use of this film incorrectly such as with a point source, perhaps the backlight side approach which would probably still be inaffective for more brightness. I wish Mark were here to assist, maybe he'd have some ideas how this could or might work.

Yes I noticed about an 80% decrease in lumens.

Maybe it would be more practical for the LED application?

Here's my take on it.

I haven't tried mine in my projector yet, but I did try holding it up to the light in my room to test. When held up directly in front of my face in between me and the lamp (perpendicular to my face and perpendicular to the lamp), the BEF would refract the image of the lamp that my eyes would perceive out of my eye's field of view. I then tried holding the BEF up directly in front of my face but almost directly under the lamp (perpendicular to my face but parallel to the lamp), the image of the lamp that my eye's would perceive was now in my field of view.

This picture somewhat illustrates what I'm saying. The film is almost directly under the lamp being held parallel to it and perpendicular to the camera:

Photo Sharing


This leads me to believe that using this film in the direct light path of our projectors will be useless. My findings show that this is obviously meant for refraction from a light source that is near parallel to the film (aka the back light of an LCD). Maybe this could be useful is some kind of folded design, but defiantly not in my design.

JJ, yes the BEF is basically the same stuff used in most LCDs.

BEF is a sheet made up of micro prisms. My understanding of its operation in a typical LCD monitor is that it redirects more of the diffusive backlight forward. I tried some experiments using BEF to combine two light sources. Although it does combine them, the angles of the micro prisms are not ideal and result in poor efficiency. What were needed were 60° prisms, BEF uses some other angle.
If you have a laser or some other collimated light source, point it into the BEF from one side and it will refract the light. Now if you point the light from the other side it with refract the light in the opposite direction. So rays from two different directions can be combined as one. Using two BEF crossed will combine four.
Click to view attachment

DJ

Would this be effective for making a 2 lightsource engine? Each lightsource is the same as before, lamp, collimating fresnel, etc. Each fresnel is placed facing towards one plane of the prisms on the BEF.

I realize though that it probably wouldn't work because in real life, nothing's perfect! It would be more likely to work if the prisms were 60° like dazzz said, but apparently they aren't.

Something makes me think that micro prisms like these have been discussed in the past. Wasn't someone trying to figure out how to route acrylic or something??

Me

Yep. . Where are you Mark?

Okay, the film is BEF. Brightness Enhancement Film. Not to be confused with DBEF. D for Dual. BEF is a prismatic filter. DBEF is a reflective polarizer. In other words: they are named horribly.

Of course, what you have is the prismatic filter. It is just an acrylic sheet with grooves cut in one side. It is only meant to be used with light entering the smooth side and exiting the grooved side. It will not perfom its intended function the other way around.

When you shine light straight into the BEF at Normal Incidence, all of that light will bounce straight back. As you increase the incidence more and more light gets through. This effect is only witnessed along one incident plain. That is why two films are placed at 90 degrees in the optimal application.

So it is easy to see why when placing this stuff right before your monitor in the supposedly collimated area of the projector, very little light will get through. Since it is at Normal incidence with the films.

The intent of BEF is to collimated the light entering the smooth side to within a 35° viewing cone on exiting the grooved side. This works just like a high gain projection screen, in that it forces the wasted light that would have sprayed out the sides into a viewing cone in which you are more likely to actually find a viewer.

It does this by reflecting back the light entering the smooth side that will not be refracted to within that viewing cone. That light returns back to a diffused surface that sends it off in a random direction to try again. The process repeats until the light finds an angle of incidence with the BEF such that it will refract to within that cone. As I understand it, if there is a second sheet, the light will need to pass the test with both sheets simultaneously. The thing is, though, that it isn't all or nothing. Zero light gets through if it will refract outside the cone, but gradually more as the refraction approaches a 0° exit. The rest bounces back.

What I can't remember is what the optimal entry angles are with two crossed sheets, for a 0° exit. I think it was 4 lamps with one lamp at each corner. DAZZZ ? Or someone could use a flashlight and figure it out.

The point here is that these sheets do not work as a brightness enhancer unless you provide a diffusive light box. It isn't that it magically collimates all light, it is just one element of a system that gives that effect.

The other film from 3M, DBEF, is a reflective polarizer. It works by reflecting back the light that is not polarized to the same plain as the first polarizer of the panel. Normally, light that is not on the correct plain gets absorbed and converted to heat by that polarizer, so to pre-polarize pulls heat away from the panel, but more importantly allows roughly 50% more light that hits the panel to get through.

But again, the DBEF works by reflecting back the light that is not right. It then supposed to impact a diffused reflector, which depolarizes the light and sends it off in a random direction for another try. So while we would get a lot of light reflecting back, we have no way to return it with current designs just as with BEF.

Forum member tgreenwood made a cool discovery, though. She found a device that suggests that a corner cube retroreflector rotates polarization exactly 90 degrees. This is awesome because this is a much more efficient result than that of a diffused reflector. And it is cool because I figure a corner cube retroreflector would make a pretty adequate replacement for a spherical reflector and they aren't too difficult to build. So it is something to try for sure. This only applies to the use of DBEF, not BEF. If only bead type retroreflectors had the same polarization effect, we could just line the inside of our light engines with sheets of the stuff and have the same effect as a spherical. That would be cool.

A couple years ago forum member mikyd1954 built a full sized light box, painted the inside white, placed DBEF and BEF crossed over the opening, and tried to get a projection. And nothing projected. Figuring out exactly what went wrong got confusing. But we concluded that since light entering the LCD at greater than around 5 degrees will miss the triplet, the output of 35 degrees is just too much loss to make up for in recycling gains.

But there were a lot of questions unanswered. One that has bothered me is the question of wether a flat white light box will allow more light from the bulb to eventually exit the box than our current spherical reflectors and black paint. The idea is that there is a lot of light missing the reflector, and the front of the box straight from the bulb, that if it had the chance to bounce around for a bit would eventually make it out the front. People have built them, we have turned the things on, but we have never got any measurements to make a conclusion.

The second question is whether with the BEF films in place, more light still exits than with just a spherical reflector. This would mean that more light can be collected off the sides of the box, not directly from the bulb than from the bulb with a spherical reflector.

Then we could presumably optimize by placing the bulb in the corner of the box. This would allow us to still collect light from the sides of the box, but also collect light directly from the bulb with the films (since the BEF reflects light straight from the bulb if head on).

After all that is done, we could add the DBEF and as long as the losses are less than the gain, could know that all of the light exiting the light box has a chance of passing through the panel. That would be saying a lot more than our current designs. The problem then is, after all those potential gains, can enough of that light impact the projection lens to see a gain. Again, mikyd1954's experiment had previously allowed us to jump to a negative conclusion here. But we just didn't take this scientific an approach. We were left to assume that the breaking point was the light missing the triplet.

Vikuiti BEF films product page: http://solutions.3m.com/wps/portal/3M/en_H...fo/Product/BEF/

Vikuiti DBEF films product page: http://solutions.3m.com/wps/portal/3M/en_H...o/Product/DBEF/

A couple neat things from the PDF are that the monitor 3M tests with has a tranmission of 7.9%. And they have Vikuiti™ Thick RBEF meant for high heat since it is pretty thick.

edit I should have mentioned that the light that reflects off the reflective polarizer is also polarized. That is why the corner cube might work so well. It is also just cool.

Mark

Geez Mark, you’re like a specter around here. Every time there’s a discussion in experimenting with light refraction and sorts only brings your presence out. It practically mimics a séance BTW, good to read from you again.


Yes, at the corners where 2 sheets crossed @ 90 degrees was the optimal but the flashlight was angled 45 degrees (best judgment) from the corners of the 2 sheets, almost perpendicular to the 2 films and pointing to it's center. The light was kind of diffused and not crisp compared to just the flashlight when seen reflected off the wall. *see pics below*


Great just spent $20 for a small sheet of BEF @ ebay to find out it was DBEF instead that might work better for us or in tgreenwood's case
BTW, I emailed the seller and he pointed me to another forum that were experimenting with this as well in brightening their vehicle's LCD's. I believe I read that 3M offers samples of their products.


Great ...back to square one I believe, once again.


This is why I believe my current polished ladles work well because I immerse half the arc within the sphere of the ladle and place a large precon enough over it to gather as much light and not loose much.


Well, I tried this. Using just one sheet of BEF and a flashlight and grooves pointing towards the wall. The light seemed diffused and tainted lacking the brilliance compared to the flashlight alone pointed at the wall. Quick lux readings showed half to lower numbers (numbers were fluctuating and hard to lock in) with the film compared to without. Whether your idea of light bouncing around getting recycled forwad is still a question. I'm thinking tgreenwood's fluorescent lamps use might be a better model for this kind of test...tgreen??


Well, lacking the DBEF we are still in the dark with this unless someone else has it or takes advantage of 3M's product samples which might not be large enough to test with just the same. It was by chance that I found someone selling separate sheets of BEF at just enough size on eBay without having to spend tremendous amount of $$ for a complete package just for tests purposes.

Good. That's what I thought.Did you happen to notice how much light was reflected back out the films, though? It should be pretty much the other half you are missing on the wall.True enough, but they both are fun to play with. And in the case of a diffused light engine, they are both necessary. Agreed, BEF is not applicable with the corner cube concept. Something to note is that most every monitor sold will have BEF inside. But DBEF is only found in higher end monitors. If you can find a single sheet reseller like with the BEF, that would be awesome.What I am getting at is that those experiments should be done again, but a bit different.I figure it would play out the same with either setup.I actually have a sheet. In a 17" size. DBEF is really cool stuff. The model that I have looks slightly embossed, and diffusive. But apparently you can get the stuff completely clear. mikyd1954 thought he had the clear stuff, but I can't recall him showing a photo to confirm (another thing we didn't do). It is really great stuff because it has nearly perfect transmission (47%). I believe the reason it isn't used in place of the polarizer is the lower contrast and it doesn't seem to perform well at wide viewing angles. I believe that last point is also one reason why they want it to come after the semi-collimation of the BEF in the stack.

What I don't have is large sheets of the BEF. Flat white paint. A reliable lamp and ballast. And time.

edit Those are great photos for people to get the idea. What they don't show, though, is the light reflecting back that doesn't hit the wall. There will be other spots in the room behind you. In the case of the head on shot where the films are letting no light through, that light is actually reflected straight back into the flashlight. The film looks like a mirror to the flashlight in that case, and less and less so depending on other angles.

Mark

Yes, I would say about that much. Almost as if a reflection off a tainted mirror back towards my shirt.


Well, I already cut my sheet to small pieces of 7" diagonals. I have a can of flat white paint some where around in garage..I think. I have an extra 150w bulb but no ballast for it...and well, time...I think were in the same boat here.

Here’s the thread where I tested a sheet of BEF to combine two Light sources. Post #10 has a good pic of its affect on refracting two collimated beams into one. There was a reflection (not shown) but it didn’t reflect back to the source, it reflected as a normal mirror would, angle of incidence=angle of reflection. I can’t recall if the light was entering the prism side or the smooth side. I’ll see if I can find the sheet again.

DJ

I couldn’t find the original pieces of BEF but I did find some from a 17” Samsung. It also had a sheet of diffusive DBEF. One thing I noticed about the DBEF is that its crossed efficiency isn’t as good as a standard polarizer; no big deal just a note.
Collimated light entering the prism side perpendicular causes the beam to be split into two. Collimated light entering the smooth side at an angle refracts and exits perpendicular.
Click to view attachment
The dotted line is the reflected beam. There could be some reflection back to the source but I couldn’t see it because of the size of the lamp. I’ll try it with a laser.

DJ

Edit:
The laser showed no reflection back to the source.

Geez, I guess I must have missed this back then. And 2 sources to boot with lower readings compared to without the BEF.

Thanks DAZZ for contributing whatever you can.

Based on your description after testing the diffused DBEF it seems to be acting in a similar manner as the BEF. Is that a fair assumption?

I don’t like diffusive films so I decided to give the DBEF the sticky tape tests ... It works, sort of. Hmmm…
Why not try to strip the diffusive layer (A/G )…
It does separate but the polarizing layer is very delicate and partially separates from the bottom layer. I only tried a corner and pulled it up dry...
I wonder if a water soak would work .
The next problem is that the bottom layer with the polarizer attached has a diffusive surface as well. I’m not sure if it is a separate layer, four in total, or just three layers.

BTW I have no idea what I’m going to do with the DBEF, just blindly fiddling with things.
I was laughing to myself while I was stripping the DBEF, what a fitting thread to post this info in, back to the beginning again.

DJ

I'm sure you caught this, but that is what I meant by lower contrast. This is why it is always installed in conjunction with a standard dye polarizer.I never meant to suggest the light reflects back to the source. Only if head on. In case this is in response to something I wrote.Without a complete recycling engine, these things do nothing for brightness enhancement. At that point they are just another element in the light path absorbing and messing with the light. They need to be part of a complete recycling system.DBEF is a reflective polarizer, and BEF is a prismatic filter. They share very little in common.
I wouldn't get water anywhere near my LCD if I was you . That's just stupid. .

Mark

I got exited that I had a sheet of DBEF and just happened to mention it in that post. The tests and pics in post #3224 were with BEF.

The DBEF is a reflective polarizer, I checked it with a laser and it does separate S and P polarized light.

DJ

What about paint stripper?

Bwhahaha . Or I got one too: drive it over with a bus. Hahahah. .

Mark

PAINT STRIPPER??!! that's just plain retarded

So, where are we at with this....ism?

I tested the DBEF to find out if the angle of incidence affects its operation. It doesn’t seem to, at least from my crude handheld tests it doesn’t. So here’s an idea that may work:
Click to view attachment
After the light has been collimated by the rear fresnel, placing the DBEF at 45°, it will act as a mirror to the p-polarized light and reflect it to the LCD. The s-polarized light passes through the DBEF and then through a sheet of polycarbonate that rotates it 45°. The light is then reflected back through the polycarbonate again, rotated another 45° so it can pass through the DBEF and on to the LCD with the correct polarization. The polycarbonate is a very crude ¼ wave retarder it may not rotate all wave lengths then same so there may be some color shifts but it may work.

DJ

Edit:
After some more thought, I think there might be a problem with the light passing back through the DBEF after it has been rotated 90°

Yes, I see what you mean. The DBEF reflects p-polarized light. It passes s-polarized light, but after the s-polarized light passes through the lexan twice it becomes p-polarized and can't pass back through the DBEF because the DBEF reflects or absorbs p-polarized. Nuts.

Tgreenwood

I figured you could just take the mirror out. That was the plan in the diffused light engine thread. Or use a corner cube reflector. I think the actual plan was to put a small high heat wave plate behind the bulb, and up against the reflector.

Yeah, the colors may get strange. I found a company that sold sheets of 1/4 wave compensation film, but I am not even sure where to dig that up. Ultimately, I think what we are looking for is a specular reflective circular polarizer mounted polarizer side towards the LCD. I don't think it exists. But my point is that a circular polarizer is just a linear polarizer bonded to a 1/4 wave plate.

Circularly and elliptically polarized light will pass through the LCD regardless of brightness at the screen, so we need to be careful when adding wave plates after polarizers.

edit What I mean is you go reflector, lamp, wave plate, reflective polarizer, LCD. Or maybe ideally: reflector, wave plate, reflective polarizer, LCD.

Mark

This seems like a good spot to re-post what I had in the Light Recycling engine thread:


Another version would be using an LCD that doesn't have a reflective polarizer already glued to it, but add a sheet of DBEF, that way you could build a straight-shooter or folded, as you please.

Another version again would be a diffusive light engine consisting of fluorescent bulbs in a lightbox shaped like a corner cube retroreflector (mylar or mirrors) using two sheets of BEF and one of DBEF.

Lots of possibilities.

Tgreenwood

PS: You can get inexpensive quarter-wave film from polarization.com here....Circular Polarizers and Waveplate Retarder Sheets .

As I see it, the diffusive engine doesn't play well with the corner cube or wave plate engine. What design did you have in mind?Sweet. polarization.com pulls through again. The site I was thinking of was much more obscure than that.

Mark

Could you explain why the diffusive engine wouldn't play well with the corner cube or wave plate engine? No matter the source of the light, the light wave will have its polarization rotated.

Tgreenwood

It really depends on what you are trying to accomplish. If you want both BEF recycling and DBEF recycling, you need to use a diffuse reflector. Corner cubes and specular reflectors won't randomize both direction and polarization as a diffuse reflector will.

What I am really specifically getting at is that BEF and specular reflectors do not play together for recycling light. You need to supply the BEF with a new incidence angle for each attempt.

If light recycling isn't what you are after, the BEF plays fine with specular reflectors. For instance, using BEF purely for its light redirecting properties from say, 4 seperate light sources. Is this all you are after with the BEF?

I think we should start refering to direction recycling and polarization recycling seperately since the mechanisms are so much different. .

Mark

I was thinking about the corner cube reflector and how it would apply to us. I am not sure that a corner cube reflector will reflect back all the light that can impact a mirror directly back to the source. Have a look at this:

http://farside.ph.utexas.edu/teaching/302l...es/node133.html

If light can hit 2 mirrors, then every path I see seems to work. If light can hit just one mirror head on, that seems to work as well. But what if light can only hit one mirror and is not impacting that mirror head on? It would just do what a normal mirror would do. Seems a lot of light straight from the bulb in our case would do that. Fortunately, I think this would have been lost light anyway, with a spherical reflector as it would have missed the Fresnel. But worth paying attention to.

And if the individual corner cubes are too large, it also seems to me that you could have the light return in a parallel direction, but may be too far apart.

A while back DAZZZLA demonstrated how difficult it may be to get light returning to the spherical reflector from the DBEF to return in exactly the same direction, or even in a direction that will impact the Fresnel, for that matter while still passing through the usable bulb area hoop™.

This could also be another benefit to corner cube reflectors, as -if small enough- they can send light straight back through as long as it hits the reflector. And since the reflector can be a sheet of tiny corner cubes, that shouldn't be a problem.

That said, light reflected back that does not pass though the hoop™ on the way to the reflector is rarely going to be useful regardless of the reflector type. I know I kinda went in a circle there, but my point is that it is really important that our reflective polarizer be specular enough and accurately placed enough to make that happen.

FWIW, here is me and DAZZZLA's hoop model: http://www.lumenlab.com/forums/index.php?s...812&hl=hoop It comes in very hand when working with these problems.

Mark

You're right. Using both the way I was thinking of won't work, because the specular retro-reflector will continue reflecting the same incidence light that was reflected by the BEF in the first place. Back and forth it will go, over and over, until losses make it zero.
What I'm after is to squeeze every last lumen out of the light engine that I can even remotely make usable.

I do have a sheet of DBEF-D400 which has a diffuse surface. The diffuse surface is on both sides of the reflective polarizer. The only effective way to use it is in a projector that uses a TV Projection lens (LCD on the lens and the same size as the lens) so that the outgoing direction diffusion doesn't matter much. I'll also use a cube retro-reflector lightbox to recycle some of the reflected polarized light that has been diffused in direction by the diffusive polarizer. I could use BEF in it but it would be pretty pointless. Actually, it would be better to use a specular polarizer since I can't use BEF anyway, and it would reduce the amount of light that would be wasted on the inside of the lens.

Anybody want a sheet of diffuse reflective polarizer?

By the way, the DBEF-E (embossed) on reflection (from the smooth side) only spread my laser pointer spot from 0.05 inch diameter to 0.4 inch diameter at 10 feet. Not bad. About the same with the DBEF-II.

Tgreenwood

You are thinking two dimensionally here, but you need to think three dimensionally.
Click to view attachment
There are three mirrors, not two, and the point where all three meet would be on the centerline of the fresnels-LCD-triplet (if using just one large retroreflector). The light coming from outside of it is reflected at a compound angle from each mirror surface. Placement of the bulb inside of the retroreflector will be an interesting problem to solve. Light emitted from the back half of the bulb will be retroreflected with offset, I think. I'd love to see someone do a raytrace of it. Or just do it, for that matter.

Tgreenwood

Wow. What DBEF models and how many do you have?

I am wondering if you or DAZZZLA and I actually do have the diffusive models, or if we all just think we do and DBEF is naturally slightly diffusive. The surface of my DBEF has a slight embossed pattern. But I wonder if that is as good as it gets. I should get a picture.

What is a TV Projection Lens? Sounds cool.I drew one of the rays I am talking about on the picture. There would be a lot of these if the bulb was inside the reflector.

Click to view attachment

Mark

I remembered this thread and I didn't fully comprehend it at the time but after reading it all over again from start to finish it really put things in order like a good puzzle. Mark,..


I gather we are not discussing 'hoops' here but a diffuse light engine without a collimator? I'm trying to keep up with whats on the bench right now.

There seems to be a number of different DBEF sub types. D2-400, D260, D280, D400, D550, -E, –M and –P2. I think -P2 type is non-diffusive and only sold as an OEM product so it’s going to be hard to get if this is what we need. Guessing I’d say that I have D550 or lower.

DJ

Edit:
I split the corner of my sheet and the reflective polarizing part doesn’t appear to be diffusive by nature. It’s very reflective, almost looks like a mirror.

Here’s another PDF I found. The D400 and D550 types appear to have a thick reflective polarizing layer so the diffusive layers could be removed.

DJ

Perhaps this?

I made a jig to test the DBEF as a recycler as in the last diagram I posted. It’s just a corner cut from the sheet and suspended above a mirror. The Polycarbonate doesn’t do a good job as a ¼ wave retarder so I’ll just post the pics of it through a polarizer that was rotated 90°.Sorry for the poor quality, my camera didn't like the dirty polarizer.

Polarizer horizontal to the camera:
Click to view attachment



Polarizer vertical to the camera:
Click to view attachment

The effect is much more dramatic in real life.

Here’s a shot with some polycarbonate placed in front. Although it looks good in this pic, it’s actually very discolored. The reflected light from the mirror is only passing through the polycarb once so I think it may be retarding at some other angle.

That is really sneaky. Would probably have to go into the collimated area, and be super precisely built, but it just is really cool. The cool part is it doesn't rely on light going all the way back to the reflector through the fresnel and back. Just takes care of it right there. If light is not perfectly collimated there would be some losses as light sneaks behind the mirrors.

Someone should just spring for the polarization.com 1/4 film and build the easy one. The odds of the Lexan being a perfect 1/4 compensation is really unlikely, and perfection could help us draw an accurate conclusion. If your films are in fact "like a mirror" (mine isn't) then it should work. We will probably have to either use glass heat shields, or risk the Fresnel, and film on the bulb side of the shield.

Anyone have any idea what happens to unpolarized light through a 1/4 wave plate? I figure nothing but a bit of loss to the film, but that is based purely on deduction.

What do we think the odds are that the Fresnel is also a wave plate?

I am trying to figure out what it is about corner cube reflectors that rotates polarization 90 degrees. It must be the 45 degree angle of all the mirrors. But that ray that I drew would definately not be rotated. Is it possible a spherical reflector could do the same thing? If a ray came in at an angle while hitting the edge of a spherical reflector, such that it impacts another edge, maybe the same rotation would happen? If so, that could be a problem.

Mark