While removing my LCD from my box for a clean I accidently scratched the reflective area (dull side) leaving a slight noticable hairline mark on the panel. The mark can only be seen on screen when there is of course a bright scene. This may not bother some viewers but after spending the effort building my pj this little flaw that illuminates on screen irritates the hell out of me. So, I did some searching on how to remedy my woes.

Is it true that the "dull" side of the LCD screen is actually just an antiglare film and nothing else? If, in fact, it is just a film then could this be taken off thereby allowing more light from our bulb to illuminate through the panel, hence, giving us more lumens for our projection?

It may seem to simple of a question to ask and I'm sure it may have popped up in the past but the idea sounds so logical.

Sounds logical to me.

Since we are not viewing the lcd screen head on, we don't need the antiglare anyhow. The only issue I see is if the anti-glare coating is painted directly on the polarizer or filter screen.

There is a thread somewhere that links to an LCD scratch remover polish. Perhaps this may be useful in your situation. I cannot vouch for the effectiveness of this product though.

Spent the free time at the office looking for either scratch removers, polishers even chemicals to rid the antiglare from alcohol to ammonia and cd cleaners. I did find a company that refurbishes the polarized film by removing the original http://www.moniserv.com/. I think that if we are producing only 200 to 300 lumens out of our diy pj's and mind you these are just homebuilt units that some say are comparable to commercial units as far as depth but not brilliance then could we possibly push the envelope in this area? It's amazing how much we DIY'ers have accomplished thus far from LL's concept to Haas Man's and Puni5her's design and now with Mississippi Man's help on MMud and I believe I read somewhere here that one of us was experimenting with a 1500w bulb How about the heart of it all, the lcd? I think it's worth checking into granted that means sacrificing a spare panel laying around. I think this might add a great deal to our efforts in using our pj's with the lights on.

Well. I not quite sure about this, but I think the LCD's applied to commercial projectors don't have polarizers built in. This is very interesting none the less. Anybody gonna try it? It seems that if we can get more light through it without the polarzing film which blocks almost 90% of the light, we can possibly use much less powerful bulbs and greater color temperatures and get a very possibly crazy bright image. Someone try this, I have a spare panel laying around, but its not messed up, its brand new, kinda don't want to kill it. Anybody gotta a scratched up one they can use???

it sounds interesting, ut could someone tell us how does it work the lcd without the polarizing layers?

Scratched lcd's can generally be found on ebay for fairly cheap.

I don't think the polarizer can be removed without ruining the panel though.

Check out the sight where it says 'Get scratch or stain on your LCD screen?' this is what it says under more details:

Repairing LCD Monitor, LCD TV and Other LCD Panel related products are quiet similar. They usually has three basic technical levels of repairing:
1. Anything that can be done by traditional repair, swapping the modules, replacing the backlights...Etc.
2. Polarize Film Replacement (scratches or stains on Screen).
3. TAB/TCP Replacement (Horizontal or Vertical Lines on screen).

then further down it says:

Polarizing Film Replacement Capabilities
Polarizing Removal Process ( Scratches on the screen)
Scratched or damaged polarizers first need to be removed from the existing glass substrate, and the glass needs to be cleaned and prepared for the reapplication of the new polarizing film. Both the removal and subsequent reapplication of polarizing film need to be conducted under stringent antistatic conditions. It is a well-known fact that during the removal process of polarizing film material significant triboelectric charges are generated. These charges are caused when friction is the primary means of generating static electricity. Whenever two surfaces in close contact are separated, one surface loses electrons and becomes positively charged while the other surface gains the same electrons and becomes negatively charged. After separation, however, each surface retains its positive or negative charge, unless the surface is conductive and or path to ground is provided.

prices are as follows, now, I wonder since we aren't interested in replacing the film would it be cheaper to just have it removed and cleaned?:

Polarize Film Replacement 15.1" or Smaller LCD Monitor $95 Parts and Labor are included*
17" to 17.4" LCD Monitor $135 Parts and Labor are included*
17.5" to 18.1" LCD Monitor $169 Parts and Labor are included*
19" to 23" LCD Monitor $219 Parts and Labor are included*
24" and Up $call

I'm going to write them and ask a few more questions about this. This is seriously worth looking into. I mean come on, how much more would you give if you could use your pj with your windows drawn open. With the resolutions of just our 15" panels we can handsdown slam the door on the commercials!

I hate to dissapoint you guys,, but went thru this excersise a year ago,,


the polarisers are an essential part of the way the TFT junctions work,, you would open the flood gate,, however interestingly,, be nice to seperate the antiglare layer that is bonded to the polariser tho

By the way, I'm not a real techy kind of person more or less a 'back yard DIY'r' although some here are quite knowledgeable from what I've read especially if it involves how the polarized film affects our contrast and the like. If any of you want to inquire yourselves with this company, by all means help us out. Otherwise, I might be shooting in the dark with them. Heres their contact info and you can also find it at their site, see the beginning of this thread.


Mailing Address
28971 Hopkins Street
Unit 6
Hayward, CA 94545

RMA Department
Monday - Friday
8:30am to 5:00pm PST
1-888-785-7288
510-785-7288
RMA@moniserv.com

Customer Service
Monday - Friday
8:30am to 5:00pm P.S.T.
1-888-785-7288
510-785-7288
info@moniserv.com

Technical Support
Monday - Friday
9:00am to 5:00pm P.S.T. 510-785-7288
techsupport@moniserv.com

elkin: seems you know something about this, perhaps you can help us in this area and inquire if they could remove the existing film with the antiglare and replace it with the polarizer alone. Further, I've found while searching through the net that polarized film can be bought and manually added ourselves.

what do you mean, all you want to remove is the anti-glare layer?

i don´t know what differencies are in anti-glare/anti-reflection but this will not improve the lcd trasmitancy at all.

the multilayer anticoating technollogy does not make more light to go throw the glass. The reflected light is still there, but what they do is unfase the light oscilation by 180 degrees so reflected light is canceled with incident light, aparently no light is reflected (really, it is minimized). as i said, i think it does not make more light to go throw the glass, and the lcd trasmitancy will remain the same in my opinion.

(edit: speeling)

About a year ago Arkay and I looked into this idea, thinking that we could get more light through the panel, I experimented with a dead laptop panel, I found that I could peel the layer off very easily, but with care, to reveal how it would work,, the difference was quite signifigant in amount of light passing thru..

At the same time I researched how the TFT panels are made,, well it is very complex in the physics,, I cant remember exactly how it all works, but each tft junction is charged by a voltage, which changes its ,,,, hmmm it generates a photon of light, the pixel array is masked with red green blue filters,,, hmmm damm memory is lost here,, but the polarisers, are set to control leakage or flooding of the TFT gates, the angle of each TFT gate is important to correct operation,,, I give up thought writing would remaind me,,, I will dig up the info again,,, how ever,,, I am still wondering if the layer that is two part,, the polariser plus the antiglare were peeled off,,, then placing just a polariser sheet only would improve transmission,, this is a new test I am going to do using a laptop that is going to be lead to the alter by me,,,
PS I have peeled the anti glare off a dead LCD panel, but because of the bonding process, the Polariser was scarred, too much marking,,, but a test in PJ increased light by at least 20%

The polarizer is defintiely needed to make the lcd work. However, the idea of peeling off the original polarizer with the antiglare and putting on a mew one without does have some merit. The antiglare coating would decrease brightness, but by how much? If the increase in brightness is insignificant, then the risk of ruining your panel may just not be worth it.

However, there is only one way to find out. Someone with a bunch of broken panels should try it....Now, who was it that posted they burned through 3 panels?

It does not matter where in the optics path a polarizer is located as long as it comes before the screen. There just must be a polarizer before the liquid crystal layer, and one after.

The reason why the polarizer must come before the projection screen is because diffusing screens randomize the wave angles out of polarization.

What this means is, if one could get both polarizers (and the anti-glare layer) removed, then you could go down to your neighborhood camera shop and buy a polarizer the same size as your lens, and another the same size as the effective cone of your point light source anywhere along the line. Install one at your lens, and the other within the path of your light source, and (in theory) and you'd be laughing. Here is why:

1. Photographic polarizers are extremely high quality. They are made of high quality photographic glass using the best processes available. They should be higher quality then the plastic ones used on LCD's. I say this cautiously as photographic polarizers are a big industry, as are polarizing films, so it is hard to say for certain which will be better.

2. You would no longer be diffusing and absorbing light at the anti glare coating as it would now be gone. Though, my instinct says that a 20% improvement in light transmittance as mentioned above is a bit high of an estimate.

3. Your LCD would be very easy to cool, as the now missing anti-glare coating and polarizers would not be absorbing IR and converting to heat, both glass surfaces would be exposed, and there would be less material for the core to conduct it's heat through.

4. Photographic polarizers are relatively cheap as compared to sheet form (around $30 canadian).

Some things to consider:

1. You do not want circular polarizers. Just the normal type.

2. Some photographic polarizers will appear purple instead of black when paired. They may all do this (I am not sure why).

Would be cool if someone would give this a go.

andysharifi, the chip you saw may have been using the above technique, or may also not have been an LCD panel at all. DLP chips do not need polarizers at all as they use tiny moveable mirrors to either reflect or slightly offset light at each pixel. This is why they have such high brightness.

Another thing that someone may want to try is using a CD scratch remover kit (not a machine). And use it to buff the ripples out of the anti glare coating. This should improve transmittance a bit and would remove any scratches at the same time. It would take a long time, though, and I am not sure if it would be safe applying that much pressure to the panel. It probably wouldn't be safe at all. It is also a wet sand technique, so would be hard to avoid water damage (not sure how waterproof these things are). I think you may be able to use toothpaste and an electric buffer as an alternative.

Mark.

Mark,

Where can we get the photographic polarizers that you are talking about?

Edit: I just got what you said. The polarizer need not be as large as the lcd screen. You could simply buy a polarizer the size of the triplet and put it behind the triplet. .... Brilliant!!

That's the beauty of it. No fancy store to find on the web, just your local camera shop (even Wal Mart may have them). Polarizers are very common knowledge amongst photographers. Polarization is how good photographers remove glare and darken skies in their shots. It also removes the glare of UV haze that was invisible when you took the shot, but shows up in the print.

Again, make sure they are not circular. Ask for help at the desk because Circular does not mean round. It has to do with the coating, not the shape of the actual lens. Auto-focus cameras need circular polarizers.

This whole concept does not work unless you are removing the polarizer that the photographic polarizer will replace. Wether you choose to do both sides or not is up to you. How you get them removed is the unknown right now. That company looks promising, but I dunno.

I just did an experiment with one of my old photographic polarizers. Unfortunately, as predicted, I could not get anything blacker than a deep (but very dark) purple when in 90 degree orientation with an LCD display showing a white screen. Maybe this is just not a very good polarizer. I think it may also be because the anti-glare coating is diffusing the polarization a bit. Hope I can figure this out.

Mark.

Oh, and you wouldn't need to use photographic polarizers at all. Someone should give the basic idea a try (remove the anti-glare front one and replace with another sheet).

You could still get the cooling benefits mentioned here by placing the new sheet polarizer up against the fresnel instead of the panel.

I only introduced the idea of photographic polarizers to point out that there would be a LOT of (theoretical) options and benefits beyond just light transmittance if you could get the polarizers removed as well.

Mark.

Thanks man. I thought you'd like that

Excellent! Cooler LCD, more lumency 20% +-, I think this something definately worth looking into. You figure how much more can our LL pj's go if we take it a little further and every little millimeter counts .

I shot an email to them (moniserv.com) in the sense they could explain to me how removing the antiglare may be beneficial in the area of home theatre projection. I wasn't sure how I was about to approach them on this but I gave it shot and hopefully they will respond. As I mentioned earlier if any of you is knowleadgable in this area and would care to write to them then please do. Just get back to us with your findings.

So, Mark, once again, it is a polarizer before the collimating fresnel lens and then a polarizer anywhere up to the point cone before the triplet, similar to the drawing below?

Tiny mirrors? Looking at a halogen light from my deprived overhead projector I noticed how the reflector's inner spherical shape was riddled with tiny square like mirrors. Could this be how the ohp gets its projected corners evenly lit? See pics below.

i don´t know if the polarizers can be placed so far... imean, maybe the polarizer before the triplet will not do a good job. (contrast drop maybe?). Also, the polarizer before the rear fresenll assumes the fresnell lens preserves the polarization, a thing that i am not so sure about.

Finally; why do you think the anti glare remove improves the light trasmitancy? ( i suggest you to check some multi-layer AR-coatings tecnics. They reduce the reflected light but no improvement on the trasmited. (the reflected light is 180 degrees canceled with the incident light, thats how it works basically, so thinking that the reflected light is no more reflected BUT DIRECTED THROW LCD is wrong).

Yep. But I only suggest that to show the theoretical flexibility that is available once the polarizers are separated. It also accommodates the possibility that photographic polarizers are better than sheets. In practice, you could get the benefit of cooling simply by layering like so:

collector
sheet polarizer
air gap
panel
air gap
polarizer

The increased lumens comes from the theory that the removal of the anti glare coating will decrease light absorption and scattering. It really has nothing to do with the polarizers directly. This is why the concept of photographic polarizers comes in to play. It is possible, though I have no numbers on this, that the transmittance of photographic polarizers is better than sheet polarizers, giving higher brightness. There may also be a larger difference between the light and dark orientations with photographic polarizers, giving higher contrast. Or, one or both of these factors could be worse. I don't want to get your hopes up too much because this is all just theory, and it all hinges on someone actually getting the polarizers removed safely and building a test projector. The true adaptability of this relies on those polarizers being not TOO hard to remove/get removed, and that the results are really worth the extra effort.My guess on that one is that those mirrors are all either slightly convex or concave, so they spread the light out as if it came from a larger point. An issue that I have been thinking about is how a spherical reflector will force the light through the filament or arc tube. This will block some or all of the returning light. So the use of little convex or concave mirrors like this could shoot at least some of the light around the filament. Just a guess, though.

I think that you are probably more correct. Using a lot of little mirrors with varying diopters and sizes could be tailored to give more light where it is needed while still maintaining something close to a point source.

Whatever it is they have certainly gone to a lot of trouble to produce that shape, so it must do whatever it does pretty well.

Rox: Yep, I really don't know which way this could go. I'm just putting it on the table as a possibility.Light should remain polarized infinitely. However, in the photographic polarizer design light will be hitting the polarizers mostly at an angle. This is not optimal, but may prove to be a non issue. Should sheet polarizers be used, and placed within the parallel light source of the split fresnel lenses, then the distance should not matter at all.Normal lenses preserve polarization. I see no reason why a fresnel would not.The anti-glare coatings that you describe are not what we are dealing with here. The coatings you are referring to are optical coatings that work by throwing the reflected light 180 degrees out of phase with the incoming light. The incoming light then cancels out the reflection. This method has very little effect on light transmittance, and actually apparently increases light transmittance of optics. I assume that is because the lack of out-of-phase reflections prevents cancelation.

This only works for certain chosen wavelengths in the visible spectrum so is only useful for imaging systems that are sensitive to only those wavelengths. As a result, this process is not ideal for the human eye.

As I understand it, the coatings on LCD panels are mechanically "roughed" to throw reflected light into many directions. They still exist, but are simply spread out. It is this method that I feel may have a significant effect on light transmittance.

To clarify though, I personally think there would be very little improvement in light transmittance after removing the anti-glare portion. The real glory here will be in the potential that by removing the polarizers the panel will be much cooler, and we have the opportunity to use smaller potentially higher quality polarizers. That part could be a good find.

Whew. Hope that clears things up.

Mark.

ok so we could say that the reflected light was always there, so only it is more or less difused but always is the same amount of light that is being reflected (assuming there is no anti-reflection coatings, that in any case, wouldn´t either improve the light trasmitancy in terms of "trasmitancy=trasmited light/incident light").

so, why should we risk our tft?

if you are talking about some kind of layer to make the lcd wider viewing angles, then this would be a good idea to remove (if posible).

the TN+film technollogy is what it means, the "film" is an element to have wider vision angles on the tft.

edit: http://graphics.tomshardware.com/display/2...114/lcd-06.html

Guys, about this idea for removing the antiglare filter. I'm not sure how these filters work but aren't they designed to eat light that passes through in one direction, and allow light to pass through uneffected in the opposite direction?

The antiglare filter is on the outer surface of the lcd. The side you would look at on an unstripped the lcd. The backlight is behind it so the antiglare filter is not facing the light source.

Because the projection lens inverts the image we turn the lcd around and upside down. This leaves the antiglare filter facing the light source! If the antiglare filter is designed to block light on this side, then this could be knocking out a lot of the brightness. But if its designed to let light pass through in the other direction, uneffected, then maybe simply turning our lcd's around will improve the brightness, and we can use a video card or mirror to correct the reversed image.

Just a though. But I don't really know exactly how an antiglare filter functions so its just a guess.

The problem is that this "roughed" technique theoretically causes some of the light to be thrown off axis. It may still enter the layer just fine, but off axis light in our design means lost light, and lost precision. A large scale example of what I mean is a wavy surface of water, or a flat surface. Which optic is better? Which surface would you rather project through to get a descent image at the bottom of the lake? The other issue here is wether the anti-glare is etched directly onto the polarizer or is a separate sheet. If it is a separate sheet then the additional material to material transfer could lose even more light.Again, it is because of the other cooling and polarizer quality benefits that I feel this may be worthwhile for serious/perfectionist builders to consider. I personally weight very little of the value of doing this towards improved transmittance at the panel, but I do feel there WILL be an improvement there.Why is it that you feel this film would make a difference, and yet the anti-glare (bearing it is not an optical coating) would not?

I think you may have just solved why I cannot get the polarization in my test to work as well as it should. If this film is present on my monitor, then it will likely be randomizing some of the polarization, thus the light leaving the film is not as polarized as when it entered. This means any polarization strength test done from that point on will not be the same as if the film were removed.

You are asking all of the right questions. The big answer is that I just don't know if this will help at all, or be "worth risking" a panel. I'm just presenting it due to it's potential. Someone needs to research the quality of photographic polarizers vs sheet polarizers or the transmittance of LCD style (not optics coatings style) anti-glare filters. Some real numbers may help clear up the potential optical advantages, if any. And even if the numbers present no advantages, there may still be something to be learned by giving this a go.I'm pretty sure they just take light that would have reflected, and spread it out over a larger area. Transmittance would be the same effected in both directions. I don't think there is a technology that allows more light to go in one direction than another. But I'm not sure on this.

Mark.

I think this could be tested with our own pj's, I recall someone had linked a pj made by a German which showed some remarkable results, I'll see if I can find it. Although, he (German) did leave the antiglare on the lens but faced the other way he had some very good results and just by using the collumating lens from an OHP and no collector if I'm not mistaken. Hold one while I'll search.

Simple enough to check, actually.

If the anti-glare really is one-way material, then a folded projector will be brighter than a straight path projector using the same panel. I haven't seen signifigant evidence that this is the case.

I know that it scatters the light, which is easy enough to see. You cannot get a nice celar image of the light envelope through the fresnel lenses. (But they WILL produce a nice clear projection of the lamp without the LCD in the way.)

The most likely explanation being that the LCD scatters the light going through it. As stated, light that does not hit the projection lens is lost, as far as we're concerned.

I think joe chevvy has a vertical projector with a mirror. That probably has a reversed lcd.

I believe vroom did this. He said it made the image alot brighter. His was so bright people were accusing him of using a commerical pj. You theory may be correct.

Ok, here it is: http://translate.google.com/translate?u=ht...pic.php?t=10650

I read thru this link real quick and I notice the maker was using a bigger triplet. Go figure. I tried searching real quick if he mentioned reversing this panel but I realized just by looking at his there was a reflection which tells me that his panel was facing front towards the arc as ours is but without a collecting fresnel. Not bad results though using OHP parts and that HUGE triplet.

I get the feeling that one of us if not some might just try this on our own pj's and come back with some revelation of some kind. My pj is kinda patterned in Haas's style and might be kinda difficult but I'm going to give it a shot somehow, by using my Nvdia's reversing features. Guess we'll see who races back here with our results

btw, for our sakes, for those of you who might attempt this provide us with some pics for comparison.

I would if I had a ballast at the moment . But there are problems, If i were to switch them over, the electronics and cables will all be on the light side and they will get might toasty, how can I overcome that??

I suppose if this works to our advantage then we either have to inquire with the company I posted to repolarize our lcd's without the antiglare or redesign the LL box to accomodate the electronics and cables in this fashion which would be the cheaper way.

Mark, question, I don't suppose you would know exactly what sort of 'higher end polarizers' would you suggest if we were to have this 'company' refilm an lcd? And if they do not have the film we seek, do you know of what and where this 'higher end polarizer' could be found. Seeing you seem to be the most knowledgable so far in this thread about this, think you could come up with some approximate 'numbers' as you mentioned?

GAH! Having polarizers re-bonded to the panel would defeat most of the glory that is to be had here (efficient cooling, and polarizer options/upgrades). Just have them strip off the polarizers and send you the new polarizing sheets that they would have glued on (they likely only have one type, or get the best ones they have). Having the polarizers bonded would not be good at all. Chances are, they would have just glued another anti-glare polarizer on the front anyway.

I hope we are on the same page here

Mark.

lol, patience is a virtue Mark and thanks.

I'm working on a design that will have the electronics sit on the outside of the box. The LCD, FFC's and electronics will all be inline (not an L like usual) and will slide in and out of the top of the box like a big cassette. This is the best alternative I could come up with for people with panels that do not have long enough FFC's to clear the optics easily. In my case, however, I will be doing this to have a way to quickly pull out the LCD in case of heat or worse , and have a portable "image engine" for other designs. When I shut the light engine down, I could just slide the LCD out so fans would not need to continue cooling the thing as the unit cools.

The down side, I guess, is that the electronics would probably have to stick up vertically out of the box.

Maybe you could adapt something like this to your current box? The biggest design challenge I can see so far with the removable image engine idea is keeping the unit air tight so the cooling system will flow properly.

Mark.

If someone could just remove their fresnels so the lcd is standing alone in the box, we could do a simple test. Get a flash light and put your hand behind the lcd, shine the flashlight on your hand so it glows, then look through the lcd at your illuminated hand. Then do the same thing with your hand on the other side and see if your hand looks brighter one way then the other.

I was going to try this last night but I couldn't find a flashlight.

Better yet, I have a luxmeter.

Well, I tried and couldn't reverse my LCD in my Haas like box . However, since I had it opened I began tweaking somemore and decided to cut my reflector even closer for the Ushio. Plus brought the collector out a little further. MAN, did it get better! Will post some pics at my plog.

So, any luck gents..or ladies?

There are no ladies here! Only madmen.

Let us know how your luxmeter test works out!

It appears that with my panel (CMV CT-529A) it does not matter which direction light is passing through, it is about the same efficiency. I got absolutely negligible differences in measurement.

This measurement was done with a flashlight suspended in the pj case, 25mm from the LCD. My lux meter was held directly in the beam path, 25mm from the LCD on the opposite side, then the light and meter were switched. The measurement was in the neighborhood of 300 lux, this should give you an idea of the accuracy that I could measure. The readings fluctuated slightly, but they seemed to be only a few lux apart, which is most likely due to some inaccuaracy in the placement of the light and/or meter.

I did a preliminary test for light transmission, and my initial results are at about 11% -- but while thinking about my test, there may be something that I'm missing, I think I may have picked up some reflections with my meter, which would invalidate those results. More to come on this matter.

I also tried some testing on a extra panel I have as backup. I didn't have any equipment, but I held up both sides to a lightbulb, and couldn't see any difference between either side. If I couldn't see a difference at all with my own eyes, maybe its just not worth the trouble to switch it over for some people because of FFC issues and wire placement. Maybe its just the older panels that sucked with a certain placement, but i think the newer panels are so efficient in that respect that either side doesn't matter. i think we can try to focus on the anti-glare coating.

These are two data points that suggest that forward and reverse placement have no difference but other LCD vendors may be different.

I forgot to mention that i have a Westinghouse 15" lcd, which uses the same core as the CMV as supraguy. So there is one panel that doesn't benefit.

Considering what Mark had to say about seperating the polarized film and placing them not laminated to the lcd in the pj would mean creating a sled that now would have 6 different divided sections I would guess (alot of moving parts now), I am so tempted to just try 1 polarizer either front or rear for the hell of it. Sorry, madman inside getting impatient.

I just found this quote from http://www.varitronix.com/handling/

PETROLEUM BENZIN is recommended to remove adhesives used to attach front/rear polarizers and reflectors, while chemicals like acetone, toluene, ethanol and isopropyl alcohol will cause damage to the polarizer.



There is some other good information there as well. Looks like removing these things may be DIYeable

I googled it, and there are a lot of people succesfully removing these things from smaller displays themselves just by peeling. There are also a lot of people screwing their displays up by peeling. This chemical may make the job safe and easy.

I also just quickly looked into alternative polarizer options. A neat one is these "wire" polarizers. They do not absorb much light like film polarizers so they work with very hot lamps. They work by reflecting the light that was not able to be polarized back. Apparently that light can be "recycled" . The recycler idea is not a new one (even amongst DIY'ers), but it is to me. I'll have to look into it some more. I'm sure it is complicated, though.

Mark.