Mark: thanks again for your further explanation, thought I had it the first time but it slipped away..... so the main point here is that
1) replacing the anti-glare polarizer will increase brightness , but we don't know by how much since because there is nothing that is totally transparent to light, some is always absorbed(glass,acrylic whatever)

2) removing the wave plates may increase brightness also for the same reason? assuming they are not needed by the panel for some even more esoteric reason....

3) stripping the panels polarizers and replacing them with polarizers that are placed say at the fresnels instead mean the 80% of the light that is currently absorbed by the lcd/polarizers is now still absorbed but by the polarizers alone which should aloow for the use of a much higher wattage bulb since the polarizers functioning is not affected by the heat (except in the same sense as fresnels being plastic) so we could then use say a 1000 watt bulb (or at least an 875 watt 100000 lumen bulb!) this is the real goal, immediate doubling of the light we can pump thru the system!


yes?

oh..wait.... if we can keep an lcd cool enought to function while it(and the bonded polarizers) absorbs 80% of the light(roughly?) then could we pump at least 4-5 times as much thru it(maybe theoretically as much as 8 times)? I realize at some point too much light will totally wash out the contrast ...of course with newest panels having 1500:1 contrast maybe using 3 times as much light would only degrade the contrast to the usual 500:1 we have now in the panels?.... imagine tripling your light output !!

hmmmm.... I can see that we'd probably have to have a multi segmented cooling system too ...... one for the lcd and then probably one for each polarizer and fresnel?
and another ? ..... would keystoning afffect the polarizer? my immediate thought was to place each polarizer with a fresnel(in a split config) but would the tilting(if kept to a single axis) affect the polarization ?

whatever, this is just so cool!

It is fasinating how much we have progressed with this idea. I'm just real eager to see the outcome of it all.

On the commercial projectors, they utilize a very small panel, correct? Do you suppose we could take a look at these to see how they are using polarization? Just curious.

elken,

I like your plan. Will you be adding an external polarizer to see if it would be sufficient for creating an image in those areas where all of the the films have been stripped. This would answer the question as to wether we need 2 polarizers, front and back, or just one.

I would be interested in the idea of putting a camera linear polarizer on the triplet to see if simply rotating the lens mounted polarizer would decrease and increase the overall brightness. This may be a substitute for the brightness control that is commonly disabled because we remove the backlight.

Heck, depending on your results, I may take that next step. It would be great just to get some results to see if the concept is even feasible. Just do what you can as soon as you can and we can evaluate other options afterwards.

Phutton: Yep. But we cannot expect cooling quite that good because the color filters are absorbing a lot of light as well. That can never be helped. The cooling advantage that you missed is that the heat generated by the color filters will have less layers to conduct through, before it can be removed.

If a reflective polarizer is used either either to replace the rear polarizer, or in addition to the current glued on rear polarizer, you can expect most of the cooling benefits no matter where you put the polarizer.

The real gold that has come out of this thread for me is the reflective polarizer idea. Imagine doing nothing to your projector at all. Then putting a film in the middle of your light path that appears to have very poor transmittance, and suddenly your projection is 32% brighter

That would be the coolest thing.

The amount of heat that the rear polarizer generates is constant. It's based on the output of the lamp. The amount of heat that the front polarizer generates is relative to how dark the scene is. The darker the scene, the hotter the front polarizer.

mikyd1954: If you mount the polarizers on the Fresnel's facing the panel, then they can share the same cooling system and still be separated.

If the new polarizers are aligned wrong you will have a gray level that is darker/lighter than your true black and white levels. The unit should still function. A half black, half gray image should be used while the shimmying takes place.

DeathRay64: As far as I know, STN's are still plagued with super slow response (150ms) and a lousy black and white point. I don't figure any of our displays are STN's. But that's not a for sure.

mikyd1954: It's my guess is it would be best that the polarizer not be mounted to a tilted fresnel. May need to build a separate frame.

Elken2004: Sweet.

Mark.

I will do this ASAP,,, just time is killer at moment for me,,, had a bunch of customer jobs that have been, some of the worst,, problems,, all in same time span,, ggrrrrrr

I do have some polariser sheets from astronomy days floating around here somewhere,,

at worst have several dead panels from laptops and LCD's that can be used too

also may do the antiglare peeled off polariser, half and half,, rough and ready but will indicate differences too

oh PS the polariser does come off quite easily,, too,,, but extreme care is required, due to glass panel,,

give me time I will also take some pics of process,, ie half peeled and tools used

mind you I used a xacto knife, to barely crack bond while working along,, most of effort is simply peeling back force,,,, the glass part is simply cleaned with acetone or acholol,,,,, after peeling the only glues left are where you pause,, hope you know what I mean,,,

I know exactly what you mean .

I have read that sharp blades are not a good combo with that glass. As you probably know, all it takes is a light score line, and you can easily snap glass along the score. It provides a sharp point for the energy to transfer. In airplane skins, if an inspector ever notices a crack forming, they drill a hole at the end of the crack. This spreads the energy over a larger area.

I realize that you have done this before, but I also read that the safe way to pull the polarizer off is not by pulling perpendicular to the panel, but instead to let it fold over itself and pull parallel across the panel. Hope that makes sense. I somehow doubt these things are flexible enough to do that, though.

It would be perfect if you could take some pictures.

So long as you have those polarizers kicking round, do you think you could test if the fresnel preserves polarization or not? This question is driving me nuts.

I have been wondering if a bit of heat would help? Not too much, obviously.

If you find your astronomy polarizers, do you have any specs on their efficiency?

Mark.

a combo of back over works,, its really on the fly,,, the xacto is merely to tickle the seam,, the polariser is fairly flexible, even with antiglare coat,,, and yes the anti coat part is quite dark too,,,, as dark as the polariser itself,, single piece,,, it alone reduces light consideralbly as seen with 60 watt globe behind it,, once panel is free of polarisers,, they are very transparent,, and the rgb filters are very visible,,,, i will try some macro photos,, of all these things,, hopefully sometime today,, to show all what all has been discussed,, in practice,,


ok additionally i will take some pics with peeled parts held up to panel with 60 watt light source,, this i can do later this morning,,, allready lying around on my bench, just to show the transparency effects,,

all thats missing in this test,, are the plasma conduits energised heheheheh

geez I amm stupid, stupid stupid !!!!!!!

easier and quicker test with laptop,,,, strip panel section,,, leave laptop on bench with stripped display supported,, and just use 60 watt lamp behind,,, this alone will answer all questions,,, results will be quick and easy,,, using previously quarters setup,,,, dont need the the PJ which would take longer to setup,,,, hehehehhe

I'm really getting anxious here. It's like Christmas. I'm thinking a 250watt bulb or even a bank of CCFL tube will work after this is done. That is unless you want to watch movies outside.

patience,, just took bunch of pics,,,,,, let me reduce and post shortly

Mark,

I found a piece of polarization film.

Fresnel - no affect
Buhl OHP lens - no affect

I'm assuming the piece I have is horizontally polarized.

3 different laptop screens - horizontal
Numerous old B&W small LCDs - horizontal
One old B&W LCD - vertical
Every new B&W LCD - diagonal
Casio watch - seems to be circular

The B&W LCDs are on various type of electronic gear.

And the little trick of using cellophane as a 90 degree polarization
twister works too.

A little more info later.

no further words,, let pics say....

oh this one is the victim to sacrifice,,,

ditto

yitto



ok you will see that the part left is part polariser and part antiglare top piece,, the sides shown are as you would look at display normally "side"

on this disturbed panel the rear or backlight polariser is still attached,, I will look for panel stripped both sides did last year (july) and take more pics of closeup of pixel arrays and rgb masks

WOW! You are doing this right now??? I guess I didn't know exactly what you meant.

Those pictures are a bit hard to decipher. I just have a few questions:

1. You are able to pull 2 layers off of one side of the fresnel?

2. If the above is true, then you should be able to get 3 layers from the front side:

a. The anti-glare coating.

b. The polarizer.

c. The wave plate.

Are you sure that the one layer is the polarizer and the other is the anti glare? I would have expected that one layer would have been the polarizer and the built in anti-glare, and the other layer would be the wave plate.

3. If you can be sure of what each layer is, what are your initial thoughts on each individual layers effect on brightness? I'm having a hard time deciphering those photographs.

4. Will your experimentation ensure that we know the visual transmittance of the panel:

a. originally. with original wave plates polarizers and anti glare.

b. with just the anti glare removed.

c. with identical polarizers in place, but no anti glare or wave plates.

Seeing the transmittance of each component may not take everything into account.

Thanks a billion for your work here.

paladin: just to be clear, if you pass polarized light through the fresnel from the focal point, it all exits the fresnel polarized on the same axis? Or are you saying that polarization has no effect when passed through the filter? That is to say, when passing polarized light from the focal point of the fresnel through the fresnel, it is no longer polarized after it exits the polarizer.

Mark.

NOW THATS WHAT I'M TALKING ABOUT, FINALLY! We are all behind you Elken. dang, I'm getting goose pimples now.

Mark,

I don't have a projector yet, waiting on parts.
I do have a 220/330 fresnel set.
I used a laptop as a source of polarized light.
The polarization film did the exact same thing when rotated
with and without the fresnel between the laptop screen and
pol film. I just tried it again at various different distances
including from the FP, all the same. Fresnel didn't change
the polarization.

Tomorrow I'll scrounge up another piece of pol film.

Thanks paladin. This was the expected result, as a normal lens preserves polarization as well. I wouldn't figure another polarizer would give different results.

Looks like a light recycler may be a little trickier to figure out.

Some sort of matte white reflector will need to be used (unless the reflector or arc tube will diffuse the polarization enough).

Thanks again ,
Mark.

wow,,, what added effects are there what,,,, heheheheh wait for pics

ok here tis first part tests

the last pic produced the most fascinating effect

rotating corner peice cut off,,, reversed the colours !!!!!!!!!!!!

need to allow complete boot to win 98,, because i suspect red would become green and green red etc etc etc !!!!! colour negavive,, thru polarisation,,,,

this changes all the discussions totally,, however,,,

now need to find polarisation sheets and do light level tests with my 250nM to 800nM,, light meter,,, used to use it to calibrate lasers,,,, extremely accurate... mind you it reads miro watts to watts in scale,,, but will be good for difference testing,, the rest can be exstrapolated,,, hmm spelling,, oh who cares heheheh

Any sub pixel that used to be aligned for no pass through the analyzer will now be aligned and vise versa (and everything in between accept middle gray will be inversed as well).

Awesome work.

Just another quick question:

1. Does the polarizer need to be flipped horizontally, and then rotated 90 degrees to achieve the reversal? Or are you simply flipping the filter horizontally (such that it is upside down, but the side that used to be closest to the top of the panel is still closest to the top).

I ask because if you are just flipping it over such that the left side becomes the right, and the top and bottom stay in the same orientation then we know this panel has a 45 degree polarizer. That could be a bit more tricky to replace.

Hope everything is looking promising. It is awesome to see the panel still works after all that ravaging

And I think that seeing that transparent non polarized area is worth a thousand words. Just counted the words in my posts .

Mark.

well you will be pleased to hear that,, it works


oh 90 degree rotation, gives negative,,, relative to its orignal postion


i cut a piece from a rear off another panel the shiny single layer one,, and put it in front of test panel,,,,

and guess what,,,, it is much briter,, but contrats is very very very good viewing perpindicular to panel,, ( all we need )

also the light is increased crudely by my count 50%

a strange "extra", the image is so sharp its unbeleivable,,,,,

and another point,,,,,,,,,,,

having the polariser off the panel about 15mmm gap,, the image forms on the polariser,, with ghost on LCD glass,,, ( virtual effect )

and yet another thing too

the backlight sheets also form an image too,, although very light,,,

hmmmm must try a fresnel too,,, no real value just extra info,,, BRB

yes a fresnel also forms an image too,, very light tho

this is over section that had antiglare+polariser removed,, this justs adds to our info,,,

must find my large sheets of polariser's


I will add pics later,, just had a customer ring up,,, system assist,,, but now have sold them complete system, thru all of this ,, hehehehhe


However, altho these tests give a very big positive to all the posts in this thread,, let me do a full test, as per few posts back,, and dunk this setup in my PJ,, need to get full screen up,, even if it is just an AVI file,,,, for this poor murdered HP laptop,, screamed when I grabbed it for sacrifice,,,,,

this test with 60 watt lamp,, as per pics before,,, is crude but effective,,, but need acid test, to see how 400 watts hits it,,,,

Well, where is the ICON for eating forum text?

Interesting work guys. You know after a private PM with Mark, I was thinking about those three LCD panel projectors. You know the ones that use a prisms to seperate the light source, run each beam through each coloured LCD panel, and then remix that seperated light back to one beam... those LCD panels you can see through... well maybe not exactly, but it seems to me that I have a memory of seeing those panels and each pixel in the panel if you know what I mean.

Anyways, its looking promising here. I wonder about the quality of different polarizers and if this ended up working, if we could go down to 250W bulbs instead of 400W. Perhaps I'm getting optimistic, I'm watching your results.

yes those 3 lcds are clean,, they are mono and have colour filters on light source side to give RGB,, reason more transparent, is because they are mono lcds,, there is no RGB masks in substrate,,,, its the the mask in colour LCDs that darken them

So we can say that this thing worked successfully, or are there still shortcomings??

It works,,, but as I said before,, final test is to put in PJ for full test,, it is possible with out diffuser might end up with introduced hotspotting form our 400 watt light source,,,, mind you a correctly setup condensor fresnel, hopefully should deliver light evenly,, fingers crossed

I cant find my sheets of polarisers,, will acquire some today, from photoshop,,

also as an extra note,, the only way for this to work is to have polarisers hard up against LCD glass substrate,,,, does not need gluing, just against it

Mark, up against the LCD and not glued will still give us some cooling somewhat don't you think?

ok,,, tried to photo the effect of just polariser piece, impossibly difficult with lamp, which kills pic,, but look at lamp rim, in pics and see the difference in its sharpness,, in life the portion is very much brighter, and colour intact,, also look at black icon rh side lamp,, mmm



in the second pic,, the black rectangle icon,, centre high in lamp section,, to the left of black icon there is a wedge shape part extending downwards left,, that part is polariser only,, you can see clearly that more light is coming thru without loss of image colour and contrast,, lousy photo, but live it is pronounced,,, later later will srtip entire panel and use sheet polariser,,,, ps rear shiny polariser is still intact,,,

The remote polarization thing seemed to be a real longshot anyway...

Other than that, this seems like possibly a huge step up in brightness might be possible. A roughly estimated 50% increase in transmittance could give us roughly 100 extra lumens at the screen. That would far surpass anything we would be able to do with the light engine.

From your pics it looks like the antiglare layer is by far the most important layer to remove. It looks like it would be impossible to remove this layer only and not damage the films below.

If different polarizers don't improve things much, the best course of action may be to strip only the antiglare side and replace only one polarizer.

I wonder what your impression is of the impact on transmittance the birefringement film has? It appears to have a minimal impact so we might be best to leave the backlight side intact.

Eagerly awaiting your further update.

yes I am inclined to agree,,

also another accidental discovery


the pieces i used were taped with scotchtape,,, just that alone seems to have same effect,, so i oiled the panel,,, hmmmmm ergo defeating antiglare,, which is basically clear anyhow,,,, geez this gets more interesting

If my aquirement of polariser sheets today is successful,, I will strip my benq 17 SHIVER,,, after i conclude that dunked laptop works,,,

hmmmm,,, i might as a test put some glycerin on my panel in PJ and run some tests,, be easy to do,, cotton bud dipped will do,,,, just another experiment

If you have a piece of glass, why don't you sandwich that film between the glass and the panel? At least that way, you'll know the polarizer is tight and flat - something I think that would give the best possible outcome for determining quality tests.

I'm kind of glad I haven't chosen a bulb/ballast yet.

Edit: Oh, one more thing... wasn't Mark or someone talking about rotating the polarizer to control brightness, or was that you? Maybe no one will do this, but if this works, potentially we could have a brightness control by having the replacement polarizer on a rotating piece of glass mounted next to the LCD. I'm getting ahead of things but... it's an idea.

I believe Mark was referring to a third polarizer that would increase transmittance through the other two polarizers. I fear however that this (if I understood it correctly) would reduce contrast while it increased transmittance. It would somewhat override the polarization of the other two films making it impossible for the panel to make a true black.

Or, if it's possible, use a light dimmer. This might be a better way only if it's possible with the Eballast and/or regular ballasts.

observe of effect of scotchtape on lcd panel,, it allows good image of a brochure behind panel,,, theory has it if you see it better, should be same with light thruput,, not scattered,, also I have noticed it is very cone effected,, which should not be a prob with our setup's

Are you saying that the scotch tape act as a polarizer or a light diffuser? I don't see what you mean from the pic.

I am a little worried about what a ~7º tilt to the panel might look like in our projectors with the birefringment film removed. ...for those who want to get better focus when keystoning.

Elkin, just left your plog. You have a light fusion screen with MM's MUDD set up at your shop already? I was about to suggest how this test of yours added with MM's LF might blow the roof off the entire DIY community and right here, right now your just about ready to do THIS????....THE EXCITEMENT IS KILLING ME!!......I'm getting a beer.............

the scotch tape simply negates the diffuser surface,, just an interesting side effect,, where it is stuck panel is intact both sides,,,,

bifringement,, well the big dunk will show those issues up if it is a problem,,,

er dunk being dropped in PJ

arhhhhhh I getting confused,, doing too many things at same time helllllppppp!

I did a LF last sunday been using it all week,,, yes it is great,,, but this was my first shot,,, I think I will do another test panel 4 x 4 foot,, with the new ideas,, and also I might have overdone spraying,,,, but let me tell you the colours are surely more vibrant,,,

SIMUL8TOR: I have to say I am surprised that the polarizer must be against the glass. We'll need to think about that one for a bit. If it is against the glass but not glued, there may actually be worse cooling. This is because the heat from the color filters will not be able to conduct through the polarizer too well. Of course, at least the heat form the polarizer will not conduct into the panel too well either, but I'm not sure what the end balance would be. I am hoping there is just some error here, as with collimated light, polarization should continue through space indefinitely. Something tells me this test light is just not collimated enough. My concern right now is that the wave plates are still on the glass, and all that has been removed are the anti-glare, and the polarizer. The wave plate could be throwing the light into circular polarity, and thus, unless you are within very tight tolerance to the wave plate, then the electric fields will not be consistently aligned across all colors, at all distances. I'm thinking using a circular polarizer instead would correct for this (or figure out how to actually remove the wave plates), allowing any distance to be used.

Deathray: A third polarizer will not increase transmittance or contrast. I was only trying to explain the weird conflicting effects of adding in a third polarizer, to help people understand the wave theory.

Elken: Fascinating stuff. I think what you have discovered with the cellophane tape is how some everyday items are birfringent. While it is not a polarizer, it is able to twist the light wave a bit, just as the liquid crystal, but not variable. I have to say I am impressed by how high the transmittance is on the raw panel with just the rear polarizer. Unfortunately, I'm still not deciphering a lot from the images, but keep them coming I think in the end it will all make sense. You don't happen to have a circular polarizer on hand do you? As mentioned above, I feel it may correct for the distance issue (so long as it isn't just because you are not using truly collimated light). Great to see you're having fun with this. It's awesome you would consider doing this to your primary panel SHUDDER

Lucky_Me: Welcome back

And adding a third polarizer would result in around a 10% decrease in brightness. Probably not worth it. The eBallast has no dimming capability as far as I know (current and voltage must be kept constant in gas lamps).

Mark.

yeah sorry for detail,,, been doing this whilst working etc..

the sticky tape was an accidental observation,, much the same effect when you wet frosted glass,, becomes transparent,,,, that why later I will fire up PJ and put some glycerine oil on panel,, thru cooling slot,, over a few inches,, and see what it does to image,, er apart from obvious distortions,,,, a wetting agent is all easy to clean off,,,, it would be stupily simple if this gave us an increase of collimated thru panel,,,

thinking about it the antigalre is a mottled surface,,, and it should eefect our collimated light and scatter it,,, exactly its prime purpose in life as a normal lcd panel..

Are you saying that the viewing angle has decreased? That would be in line with predictions. What specifically happens at wide viewing angles now. Darker? Lighter? Different colors than expected?

Mark.

ok there is a colour shift,, and lighter,, that is why polariser needs to be against panel,, further away effect increases,,,

in effect the polariser is also a lens,,,, i used a 330mm frensel and saw a light image,,,

but head on the blacks are very black,,,,

wait for me to do dunk test,, this will really show what happens,,, I now so many tests to do,,, not funny hehehehh

but if we gain even 50 lumens,, we are ahead,,

I also have a standard three mini array in old PJ,,, i will photo it and post it so you can all see what a mono lcd is with its filters,,

Ahh...I think I just peed myself.

I'm so happy

But head on there is no effect? That is all that matters. The bulb that you are using is not truly collimated light, so I would expect most of that problem will clear up in the projector. To be clear, are you finding if you hold the polarizer too far away, that even head on colors and brightness get pooched? Try standing accros the room with the polarizer in hand. That'll be pretty collimated.

Mark.

For those of us that have our panels keystoned there may still be a way to get a keystoned image. Put the second polarizer on the front fresnal, which is keystoned. I think this would be a really good experiment. If the front fresnal is 330mm from the triplet then the image would form exactly where it is focused on the triplet. May be the best of both split and unsplit fresnal configurations.

That is assuming that the double images don't transfer through the projector.

be careful guys,, let me test this all properly first,,,

well one thing established is that 400mm x 400mm sheet polariser is $250 aud
here is australia,,, but that was from one call only so far