this is very relative, you have comertial unts for homecinema from 700 ANSI lumens on the screen to HI-end 6000 ANSI lumens on the screen.

any comparison would be just personal.

Same for the Diyprojectors, We can talk about the LL one (becuse it is standar.. we should use same elements on a LL...
You will have 200 to 250 lumens on an average LL I think.

disgusted?

Mark,

I like your lens placement order. The only better placement would be the last polarizer on the triplet, but this may have some issues such as focus and the light rays not being parallel to the polarizer.

If it worked, however, it would probably reduce cost and maybe reduce stray light leaving the triplet.

As far as commercial units, they generally provide lumen stats that don't equate to reality. Objective measurments often show the actual lumens to be much less than stated and the lumens on econo mode to be much, much less than stated.

What is important to us is that a real increase of 50-100% in brightness would basically eliminate any real brightness issues on our end. If designed properly, our projectors are already plenty bright, IMO. Increasing the brightness will only allow us more flexibility in watching them with ambient lighting.

Edit: By the way, I read up on how lcd panels work. Your explanation, Mark, was dead on. My misunderstanding was in equating the liquid crystal to a polarizer. It isn't. It does actually "twist" the light, instead of polarizes it.

The reason 2 polarizers are needed is because, without the first polarizer, the liquid crystal would simply be twisitng randomly polarized light. The result would be randomly polarized light coming out of the liquid crystal. A single polarizer would simply polarize the randomly polarized light into a single polarization, regardless of the liquid crystal "twisting".

Yes!

This is why I was trying to say that your eye has no way of telling which orientation the light is in. You need that second polarizer to "Analyze" the angle and become darker or lighter due to the wave energy properties I described earlier. As you say, without the first polarizer, the wave energy is random anyway, so a twist by the crystals does nothing useful. And finally, even if crystals were able to work as polarizer (squishing any axis to one axis) the crystals preserve most of the energy in both axis when they twist, so they could not possibly work as an analyzer (they don't get darker and lighter as the angle they twist changes).

Polarizers "squish" all light vibrating on any axis to one axis. Energy is lost in the process (We utilize this property as described earlier to create darker and lighter pixels).

Liquid crystals "twist" all light on any axis, to another axis. Two light waves vibrating into the crystals on different axis will not exit the crystals sharing one axis. There plane of vibration will simply be rotated the same number of degrees. Very little energy is lost. I guess you could say: Liquid Crystals are a variable wave plate.

Until you've put the whole picture together. Looking at what each layer does, and what would happen if that layer weren't there (as you have done) this concept can be a little tricky. I'm sure you can see why everything must be where it is. It's a beautifully simple system where each element is placed based on fundamentals.

I hope we can figure why the polarization is getting funky the further from the panel, as Elken describes. Then their are just a couple reasons why placing the polarizer on the triplet could be a problem as I see it.

First off, the triplet rotates with focus as SIMUL8TOR points out. So you would either need to adjust the thing constantly or mount it separately, or use a sliding focus.

Second, polarizers (like most optics) don't work so hot at angles. It would be ideal to have the light collimated as it passes through the polarizer. If it is ever found that these projectors have a higher contrast ratio then the original displays, it would be for this reason. Backlights do not put out collimated light.

The thing is, these sheet polarizers are much less expensive then I thought. In fact, their cheap. So using the sheet method is pretty economical. Especially since you have to buy a minimum of 2 feet, anyway.

The reason why placing the polarizer at the lens was considered an great proposition early in this thread was the thought that camera polarizers could be used instead. It seems these sheet polarizers are more effective anyway. This still needs to be investigated, though.

Not sure why you feel less stray light would leave the triplet. Axis of vibration and direction of travel are not related. A lens is only concerned with the direction of travel.

Mark.

You guys are killing me Ok, so...

1) The first polarizer polarizes the light into a single polarization either vertical or horizontal path, then...

2) The polarized light passes thru the LCD where the crystals twist's the light to certain colors, then...

3) After leaving the LCD the light is then poloarized into a straight path once more either vertically or horizontally to be projected.

.....is this correct?........

If the light is being twisted after it leaves the LCD until hitting the next polarizer to be polarized vertically or horizontally than this would explain why Elkin found it much more brighter to 'butt up' the polarized film onto the LCD. Right?

Edit: nevermind, Mark beat me to it

From the information provided on howstuffworks LCD page it appeasr that the molecule which does the twisting works specifically on light in a specific polarisation.

So if we break it down section by section of the molecule, like the rungs of a twisted ladder (Typical visual model of a DNA molecule works) each rung of the ladder will work on light that is close enough to it's polar orientation and make it equal to it's polarisation. Therefore in it's "full off" state, assuming a 90 degree twist, slightly more than 50% of the light would be MADE polar, as it hits a rung of the "ladder" which is close enough to make the light polar. So any light near the 0 to 90 degree range would be caught by the rungs of the ladder, and made to be polar. Light which is in the 90 to 180 degree range would be ignored.

The Howstuffworks article seems to suggest that the liquid crystal material is aligned along microscopic grooves in the glass substrate, which leads me to the above conclusion.

It also then stands to reason that on the "full on" state, where the ladders get untwisted the light would pass through virtually unchanged.

What makes me curious is this: Since it is obvioulsy possible to twist the polarisation of light, would it be possible to make a light polariser that takes in light randomly polarised, and pass it through at high efficiency, but making it polar? This would mean that the first polariser would pass through more light, instead of filtering it out, which would in turn result in a much brighter projection.

While it is easy to twist light, it is not easy to "squish" light. It seems the only possible more efficient polarizer at this time are the reflective polarizers from 3M. These at least give you the opportunity to recycle the non polarizable energy into random orientation and give it a second chance at passing the filter. The filters commonly used simply absorb this energy and convert it to heat. While reflective polarizers don't pass more light, like the polarizer you describe, they at least don't consume light.

The only way a wave plate could act as a polarizer is if it could somehow dynamically change its phase for each wave that passes. I think that may be about a billion times more difficult than building that see-thru car that James Bond has.

The howstuffworks.com explanation isn't very advanced. Like all of their articles, it is a nice primer to further research. It is my understanding that without the grooves, the crystals do not rest in a helical state. The ends of the crystals fall into these grooves, and it keeps them locked into the twisted helical state (until a charge overcomes the forces here).

The idea that the light needs to be on the exact same axis as these grooves to be twisted seems unlikely. Even though it makes for a very easy thing to visualize (it looks like a playground slide) I don't feel this is the case. The twisting simply varies the amount of birefringence the crystal structure exhibits. All I can say is that based on my own limited experiences, and a ton of thinking, this is how I feel it must work. I'm sure you have found, as I, that the amount of conflicting information on this function in general is disgusting. For example, Some people draw the crystals oriented end to end on charge, others draw them uncoiled (HSW).

A panel disassembly will straighten out most of this. Elken?

Mark.

. Put differently: .

(I don't know what I'd do here without smilies ).

Seriously, I'm puzzed as to why the polarizer needed to be butted up. My only theory right now is because his polarizer may still be covered in glue residue. This will be diffracting the light all over the place before it hits the polarization surface. If he presses this up against, then the lousy surface will be squished flat again, and no difraction will take place. Seems unlikely that he would miss this, though.

Another thing: the glass substrate is a couple millimeters thick. That is, it is permanently distancing the polarizer from where the last light altering component can be found. The only reason the glass substrate is there is for structure. For things to suddenly be perfect at that point, and not any further just seems strange.

Mark.

Reflections. Light bouncing everywhere.

I think that the collimated light will help in this respect. Maybe the waveplates have something to do with this?

I have to agree that it should not matter where the polarizers are. It could also be simply refracted light hitting elkin's eyeballs.

I would be surprised if, once the wave plates are removed and collimated light is used, that we would still have this problem.


Actually, I am using the box in box design, so the triplet is stationary within the box (does not rotate). The inside box simply slides in and out to focus.

If you check some of these posts you will find that during dark scenes there is some light leakage around the edges. This is especially true for 15" lcd panels. For 17" lcd panels the reported contrast is generally higher. Since they use the same technology I think it is due simply because of the larger size of the 17" panels.They probably measure contrast dead center of the panel. Any light leakage around the edges would naturally be dimmer in the center for a larger panel.

Extrapolated to the situation with the polarizer covering the triplet, there is no way that light can reflect off the edges and into the image. The triplet would be essentially covered with the polarizer, meaning that all light going through the triplet would have to pass through the polarizer or be completly blocked.

This is, of course, just speculation on my part. But I would be very interested in the result.

Also, I am curious if the focus would be at the triplet or still on the panel. While the image if viewable on the polarizer, it is actually formed on the lcd. I don't know.

I really don't know what everyones talking about.

It sounds to me that while the antiglare can be removed, it also has another function that is required for optical purposes to make the lcd work properly, and removing it prevents the lcd from working properly. So, we are trying to find a material to do the work of the antilare filter, without the antiglare effect, it sounds to me you intend to use a polarizer. But what exactly is a polarizer?

You guys probably know all this but when I stripped my lcd, there were several whitish sheets, and a thick plastic material that seemed to scatter light horizontally. I assumed this was meant to scatter the light to increase the viewing angle.

Anyways, we remove all that, and just keep the lcd itself. In the end light is moving through the lcd, collimated, we only need the triplet to "see the lcd".

I'm sure you guys are all aware of all this but my point is just to keep in mind, that what is generally required for an lcd to work properly, is not necessarily all thats required for it to work in our projectors, simply because the optics are differant.

Anyways, I'm getting impatient with this whole thing. I'm not sure what the current status is on this project, I haven't been able to keep up, so to me it just looks like we're going in circles. I'm sure this is not the case but I'm ready for some results.

Supposedly clear tape is enough to remove some of the antiglare effect, so I was thinking maybe I could find some large sheet of clear tape like material and carefully cover the antiglare with it. But I'm worried, first, could peeling it off pull the lcd apart or are lcds more durable then that? Do you think the tape would leave an unremovable sticky substance? Do you think whatever chemicles in the tape's sticky substance could "sink in" to the lcd and cause problems? I'm not sure what material an lcd is made of, but even a washed plastic cup can retain the smell and flavor of whatever was in it last.

You know, I thought he was trying to say that polarization in general does not work. This makes a lot more sense. It was just the way he describes a complete and drastic change in image quality, as if the distance is destroying the fabric of space and time . I see now he may have been simply describing a decrease in image quality. Not that it just flat out doesn't work. I hope I'm not the only one who read it that way. I agree collimated light will help here, but only if the three surfaces are kept perfectly in line. That said, it's hard to believe reflections effect the image that much. The fresnel and rear polarizer have always been reflective and that is not a huge issue. Further, I've seen panel substrates with an anti-reflective optical coating tuned roughly to RGB.

Are you suggesting that there is light additionally reflecting off the inside edge (not front and back) of the polarizer, or are you saying that if I work out mathematically the number of interfering reflected photons hitting the triplet, that there will be less with a larger area of reflection (17" vs 15)? I've never worked this out, or heard of anything to this respect. This is really hard to word. Hope it makes sense. If either of these situations exist, then I see what you are saying. It would be worth trying.You sound like me .

I wrote earlier that the wave plates may be circularly polarizing the light a bit. This could be a major factor. A quarter wave plate combined with a linear polarizer will circularly polarize light.FWIW, The full quote was:Mikau: please don't try any of this just yet, unless you want to risk for the sake of experiment. The glue absorbing in is unlikely, but getting stuck to the panel, is certainly possible. If you used care, I see no reason why removing the tape would damage anything. I broke down the optical caveats of using tape earlier. It won't likely be the optically ideal solution.

Mark.

sorry guys,, i have been very busy,,,, customers,,


ok had a quick scan thru recent posts,,, please listen to marks points,, and dont try to vary them,,, they are valid,,,

now about spacing of elements

the further away the polariser,, the more light pollution you get,, hence the rapid detoriation of contrast and sharpness,,

also if you leave the orig (shiny side polariser) on,, angle of rotation replacement front polariser is of no worry,,, it is simple an aligment issue,, and the reason comercial units have higher contrast is due to usage of a MONO lcds,, coupled with rgb filters,, ergo ummmm

think about a black only printer,, gives you pure black colour,,, but a RGB printer gives a composite black,, but not as black as (black only printer)

our blacks are made up thru the sub pixels thru filter masks,, but also the polariser quality also adds to the final contrast too,,

as a test I ripped apart a deceased commercial PJ 3 lcd type,, used the polariser from it,,, to test laptop test rig,,, and yes it worked fine,,, mind you it was mounted on a piece of thin glass, in a frame which had rotation adjusters,, ( I will post a pic of these assemblies later,,

its a great pity we dont have avi ability here, cause a short movie tells all,


Now as far as (tape) method,,, this is possibly the more attractive method for DIY, and those who dont the money or access to facilties to remove and apply new polarisers,,

I did another test on my Benq 17" I poured glycerine oil over panel,, and then applied a sheet of glass to this mess,, 1/8 inch glass,, then used packaging tape to seal all edges,,, to antiglare surface that was still exposed,, crossed my heart no leakage occurred, then put sled back in PJ,,, my gawd stunning was result,,,,

refer back to my crude half screen test with bubbles,, this definately is the way to go,, somehow Now bear in mind,,, yes we are adding two more layers to exsisting, senairo,, but the gain is quite high,,,,

also it could be as simple as a DULUX high gloss laquer spray, very carefully applied,,, will have to test this on another dead laptop panel to see effects,,

and yes Mark,, I agree we still need to follow thru with acquistion of new polariser sheets,,,,, but only do the antiglare side ONLY!! now peeling the layer is not that hard to do,,, just due care..

as far as setup,, being that Polariser sheet is in mylar form,, it would either have to be mounted against the panel, or better still to a thin sheet of glass so that is able to be fine adjusted in rotation,,,,,,,,,,,,,,

oh PS I forgot to state that it must be a very precise distance from LCD panel at all corners IE no tilt at all ( this would explan why you cant use it on the other side of any of the lens), because it affects linearity of polarisation almost the same as rotation alingment... gawd my spelling sucks, lately,, brain ahead of hands,, hehehe


EDIT,,, the antiglare side is fairly robust,,, all my tape stickons and gloop dumped and cleaned has little effect,, even cleaned it with alcohol,, but BE CAREFUL,, but any marks will certainly show,, will show as like dirty screen effect,, and another note after three months of running,,, my fresnels were dirty with very fine dust,, boy cleaning them also tidied up image,,,

Using a new fine bristle paint brush,, running with circle grooves,, and finally blowing out dust with compressed air works very well,,, doi not brush across grooves,, they are delicate sharp edges....

So basicly you went with the apricot preserves concept I mentioned. Only you put it on toast.

Any pics of the results?

Would you attribute this to reflections between the panel and the polarizer? Exactly how nasty is the deterioration? Even the commercial PJ polarizer didn't work out (no glue)?We would still have a compensation film that was not removed. I would always wonder what removing that would have done.You ultimately feel that a professional layer removal would be needed?Does this mean you have dunked the laptop panel into a projector?

Makau: 3M sells "Magic Tape". It's removable scotch tape that should not leave any residue I would think. It's in the school supplies isle .

Mark.

WELL IT'S ABOUT TIME ELKIN!!!

This only encourages me to have my LCD professionally remove the antiglare and replace with a better polarizer (lamenated). As much as I would like to proceed with this DIY style it just seems to be to much to calibrate and maintain. Not to mention the slim to greater possibility of crackin the panel if I tried peeling it myself.

Mack-A-oo? Well thats better then McQuack.

Its "Mikau" pronounced "McCow" or "McKway" pick one.

Anyways, who or what is 3m?

Are our LCD panels a 2-way item or are they really designed to have light going through them in only one direction? The reason I ask is that to project our images correctly, we have the normal 'out' side facing the lamp - the direct opposite light path to 'normal' LCD operation.

Has anyone simply turned their LCD around to see if there's any difference.?

Maybe we need to swap all film layers/polarisers front to back also to achieve the best result (this is of course assuming that the central element, the glass substrate, is bi-directional)

There are some screens coming on the market with special film layers designed to increase brightness, viewing angle etc. These panels are definitely a one-way light path design and would be unsuitable for out PJ's in standard form.

Sorry McDonald's .

It's 3M Innovations. You've never heard of them? Huge company: www.3m.com

They don't make the things you use, they make the things you use better.

Mark.

I mentioned that earlier and someone did a lumen output test with a luxmeter. No noticable differance in light output.

ok further testing done


hmmmmm,,,,,, interesting

using a frame mounted polariser from DOA comercial PJ

at five feet away rotating the polarisor,, yes the image is clear and ok

this one must at 45 degrees to frame because only got good contrst and colour balanced image at that angle..

so it may be possible to have one in front of triplet EDIT::: (between triplet and field fresnel),, ergo reducing the size and cost considerably

I need to set up an optics rig in pj to run this thru the fresnel,,, FRESNEL is the only possible negative to all this,,, unknown effects with ray angle change passing thru polariser,, as before mentioned,, the aligment of polariser is critical as far as,, parrallelism to lcd panel... but not impossible,, I think

EDIT again: keystoning would also effect parrallism too,, so dont jump gun yet folks,,,
DAZZZLA,, can you set up a program to deal with this???????? ray trace that was,,, time is against me here,,,

Elkin: I don't suppose you got any pics to ease our minds.

I will asap,,,, time is problem,, I just wanna get some solid ground with good reference point then step by step tests and results,, all I amdoing right now is quick reporting of my trail,,,, so no one goes of on a tangent,, ergo getting into trouble by doing irreversable damage,, by following a false lead,,,,

If I had not of decided to sacrifice a poor old HP laptop,, ( which was still fully functional and useable)we would all still be theorising as to what where and how...

we have two paths here

1/ remove antiglare/polariser, and buy new one, ergo higher light transparency
but frought with some unkowns still

2/ add a clear sheet of very thin glass optically glued to current antiglare surface thereby filling all the little pits ( frosted glass effect wetted becomes transparent)
this I have proven beyond a shadow of doubt and also much more DIY freindly,, suits 90% of us out here

if one wants ultimate refer point 1,,, then costs will be involved for sure

we must remember that most dont have one the skills or the want or desire to spend too much,, cant lose sight of that point,,

I am still amazed that those here have achieved good results with little skills and or knowledge

The art of DIY still lives,, against all odds,, I have been a DIYer for years across many areas,,, it sort of died off, several years ago,, used to be able to go to electronics stores and get allsorts of stuff that appealed to DIYers now limited type of supermarket junk,,,, heheh getiing off my highhorse now,,,

Understood Elken: Will hold back until your ready with a final and elaborate report. God's speed, my DIY brotha!!

Mark: The issue I have with putting another pane of glass on the panel is 'heat'. As you explained earlier and which sounds logical, stripping the panel of unwanted layers will allow it to cool efficiently thereby furthering the life expectancy of our LCD's. Adding another layer, in my guess will generate more heat and deteriorate it's function even further. So far, base on 'MAD Elken's' findings it would be a 'risk' if others decide to go this route, don't you agree?


no offense Elken, just debating....don't mind us, just work...hehehehe

. My brain just about exploded trying to figure out why it didn't. Thanks for double checking this .

Apparently the Fresnel's preserve polarization. Will be interesting to see how well, and whether the angles will be too steep for complete polarization. It is possible the fresnel material will act as a wave plate, so the polarizer may need to be rotated to a different angle then expected. I don't think you will run into this problem, though.

Apparently, different manufacturers will choose different polarizer orientations. Some will be 45 degree and some will be 90 degree. I derive that should the rear polarizer be removed as well, the angle will not matter so long as they remain 90 degrees to each other.

SIMUL8TOR: If I were Elken, there is no question that there would be one section of that panel with everything removed (both sides). I don't really see this as an experiment anymore. Too much has happened exactly as it should. To be blunt, the only thing that has really worried me was wether the layers could be physically removed. Not that there would be some catastrophic failure if you did.

Mark.

ok guys,,, I need help here????

can anyone explain what happened HERE??

is it possible this had anything to do with why !!!!

ok,, further crude testing,,,, confirmed

as I suspected,,, polariser must be between the field fresnel and the panel,,

especially after four glasses of cab merlot,,, i cant even pinpoint the model that is happening,,,

but the closest orientation for proper colour polarisation after the frensel is a further 45 degrees more angle than without the fresnel,, it the aforementioned item seems to act like another kind of polariser,,, wierd but true

not acting in logical config,,,, but there is logic bedded in there,, maybe merlot is stuck in da middle,,,

anyhow seriously,,,, apart form wierd angulation,,, there are wierd starbursts,,, as i suspected the redirection of rays to a convergent point are palying havoc with polarisation,,,, makes sense if you think real hard about it,, something that currently eludes me

so please no more theory about it,,, dazzzzla would be able to model it with his stupendious program,,,, i knew this would happen myself,,, but had to prove it to myself too,,,

Mark I think you would be the closest to understanding this reason,,, so whilst I snooze downunder here,,, run it thru your central core proccessor, and transalte for me ... hehehheheh

Can’t quite decipher what you’re asking here. What would you like me to model?

DJ

""Hic"


heheh sorry,,,.

ummm mm how do i put it,,,


ok,, collimated light tranfering thru LCD panel,, then polariser,, then field fresnel,,,then triplet... (normal state) as we have but currently polariser and panel are consided as one item...

now collimated light thru LCD panel,, then field fresnel,, then Polariser,, then triplet,,

I know of no way to model this,, almost a 3 dimensional model, to show what happens to light that is at various angles of twist from panel, then refracted, and then gets subjected to polaristion while coming to a convergent point,,, ( geez sorry this is hard to describe),,,, I already know what happens,, but how do you prove this before others say otherwise,,, no being narky or like,,, just dont want anyone expending valauble energy,,,

I've been following this thread and early signs indicate that you may certainly be onto something here. Congratulations to all involved... My question (and the reason I quoted above) is that, how much of a concern is heat in this application ? It appears to been a fairly major point in even attempting this procedure, however, never has this really been an issue, execept for a few people requiring some "fine tuning" of their cooling circuit, and even then, no reported problems concerning failure of the LCD due to heat problems. Certainly the major focus on this thread (IMHO) should be trying to get the most lumens thru the panel without sacrificing picture quality. If this involves placing another layer of glass on the LCD, then I believe this should be highly considered (especially if it is a inexpensive altertative to using expensive replacement polarizing filters)... after all, there are other ways of reducing heat energy (IR) at the LCD, such as IR glass used near the lamp.

Okay. So let's take it as a given that the polariser needs to be as close as possible to the LCD element, probably attatched to it.

What we need are some light transmission figures for the LCD.

I measured mine last night. I got 8% using a method that I believe to be reliable.

I took out all of the optics from my projector, no lenses, no lexan, nothing. I placed my luxmeter in the hole where the triplet would be, took a lux reading. Then, with the meter still in the same location, I placed the LCD panel next to the triplet opening, and took a lux reading. This should negate any ambient light effects. I believe this reading to be reasonably accurate.

Now, what we need is a reasonably accurate reading on the transmissiveness of a panel WITH polarizers, but without antiglare.

gee, you mean you don't want to put glycerin on your panel first? wheres your DIY spirit ;-)
but actually if Elken is close to it being a 30% increase for coating the front polarizer with some clear substance, I'd happily settle for that! well, for a while anyway....

Thanks for replying Gadget, your right, since signing up I have not read any topics on overheating the LCD mostly temp readings. I just wanted to let others be aware of what would evolve from all this should they persue this route. Thanks for supplying your input as well as the rest of you all in this effort. Greatly, greatly appreciated.

Don't forget Supra, there are other layers of film material that Mark pointed out including their adhesives. My understanding thus far is have the LCD stripped down to the substrate then add more efficient polarizers both sides would yield better lumens. As Elken also suggested, adding the scotch tape to the antiglare side boosted an additional 30%, makes you think what more we can achieve if we rid other layers. Thank Mark, in this regard.

With no voltage applied to the lcd and the collimated lightbulb on, do we get a white screen or a black screen. The reason I ask is because this should reduce speculation on wether the two polarizers are 90 degrees out of phase or are aligned in phase. Either method would work. It's just the preference of the lcd manufacturer as to which one they designed the controllers for.

If you get a white picture then the polarizers would be in phase. This is good and easier to simulate with 2 external polarizers. Simply cut a polarizer in half and apply it to each side of the lcd in the same up/down direction.

If the screen is dark then the polarizers are 90 degrees out of phase. This means that you have to cut the polarizer in half and then resize them at 90 degree angles before applying them.

Also note that if the front polatizer can handle the angles of the collimated light after the front fresnal then the twisting of the light by the fresnal may not be all that big a problem - If we can put the polarizer on the triplet. Simply cut the polarizer and attach it to a ring that fits the triplet. Then rotate the ring until you get the right angle. Leave it there and enjoy (assuming no artifacts created by the angled light or bad focus).

When the LCD is off, the screen is white.

Hi guys,

Elken: About the twisting at the fresnel: I gave a quick warning to this effect earlier. Looks like the wave plate function of the fresnel has been confirmed. The exact same thing that cellophane tape would be doing. It's no big deal, though. Just means the polarizer must be at an unexpected orientation as you have discovered.

The problem is the rainbows that you describe. This is likely due to stress fractures in the fresnel itself. These stress areas will introduce birefringence into the fresnel. This will cause light to split and scatter producing the rainbow effect. This is the way glass sheets are inspected. I worried about this early in the thread as well. My conclusion was that if there were rainbow effects, it's just not worth putting the polarizer on the front side of the fresnel. To be clear: any time you place a polarizer on each side of a transparent substance and pass light through it, the stress fractures/impurities appear as colorful streaks. There is no way around this but to have a perfect transparent substance. Although, I would think that the effect could be minimized by placing the polarizer right up against the front side of the fresnel/as close to the fresnel as possible.

Ray-tracing will not aid in the investigation of either of these issues. They are both concerned with the orientation of the light's vibration, not its direction of travel.

Mark.

It's not the primary concern for sure. Lumens and picture quality are much more important, I agree. However, the more heat your panel can take, the more options you have as far as light engines go.

You also won't need to pump as much air through your box, so quieter fans could be used. The worry is that by encapsulating the panel, heat will become an issue, where it wasn't before. It's also possible that an additional layer will act as a beneficial heat sink, though.

IR glass will cut down your lumens, while polarizer rearranging will not.

The side project in all of this remains the possibility of using 3M reflective polarizers. As described earlier, this could result in greatly improved light output, should an effective light recycler be designed. It would also immediately keep the panel a potential 52% cooler, by my rough figuring.

Mark.

oopss hole in what I said about modeling effects

did not mean for ray tracing as per say,,, i meant more a diagram to show what happens,,,

we normaly have antiglare side to lamp source,,,, but that raises another interesting question

If one looks at the internal structure of the lcd,, the last item towards the normal viewing situation, is the RGB mask, now I wonder what effect it has by firing light thru that mask before crystal array,, as opposed to light normally coming from backlighting thru crystal array, then mask, then viewer,, gawd more tests to do,,,

If one assumes that we remove the antiglare side only,, then what i stated last night is true,, but my thinking was wrong side..

I took some closeups other day,,, now need to take close up with panel reversed, and have a look at pixel quality,,

Is the briteness enhancement layer which is the first layer on the backlite side just simply a very "shiny" as it is, but they give it a fancy name for marketing reasons,, because all it is doing allowing full transmission etc.

also I will check the half cut polariser, to account for full brite screen on power off


ummm the gylcerin test was to see what it would do it worked very well,, except had to keep test short due to leakage,, used scottch tape to seal,, ,, twas fun with that goopy mess,, and yuk cleaning crappy too,, but ok,,, hehehe

but panel heat to was reduced,,, my temp gauges built in were lower by few degrees,,, so the reflectivity of glass "sticky tape" actually aided in cooler panel,, the antiglare is an absorber by diffusion effect,,,

Bring out the 3 year old in you!

So are we getting anywhere?

I see a lot of argueing about how close the polarizer can be to the lcd or whatever, and how it causes distortions and what the limits are, and all.

I don't see the point of this arguement. Won't it work just on the surface of the lcd? If it will, then heck put it there! Why are we arguing about this?

So is there a positive answer on the brightness differences due to LCD orientation? Would a folded, with mirror, design be brighter because of the antiglare filter ?

there simply is no argument,,

Mark has been researching the science, very well,, and I agree fully with most of what he has posted,,, just some things needed to be put to the test, for real,, because it is the unkown side effects, that have to find out about..

also,, if we could get away with a small polariser... means less cost..

there is absolutely no doubt that what we have discovered thus far is that removal of antiglare layers,, and use of straight, shiny surfaced polariser,, works and it also works very well away from panel,,, just not with a lens inbetween,,

and if it wasnt for someone having marked their panel,, and looking for repair of it,,, "Sim" I may not have looked back into this whole thing of removal, for quite a while, because I was working on screen tech,, even before MMan came on the scene,,, had just built a LF screen a week before,, but having had built this LF screen,, now back to panel,,, thanx to "sim" for triggering it again

So please Mikua,, dont go starting theoritical arguments here,, it only throws the research off,, no offense,, Having said that, if you have any ideas,, add them,,


Clive..

This is really now an interesting question,, as per my post before,, going back to the structure and the way the layers from backlite side thru,,,, to exit side antiglare,, I am begining to wonder if this would be so,, and also improve image quality too by going with the flow,, instead of backwards,,, (this has been the way, simply because of the FFC's and not bending them backwards against their natural curve, and also, having to flip image) and this way would aid with antiglare removal and replacement of single polariser,,,, bear in mind this is if we leave the exsisting backlite side shiny on the panel,,,

BTW,,, yesterday I had ARKAY up for a visit,, we had both his 15" benq 567 and my 17" benq, running side by side,, with a spanned image of 2048 x 1024..

gawd cinerama screen 16 foot wide,, very very bizarre,,

but subject was to test eballast + sd400 + 17" + std LL optics and
mag coil ballast + HQ double ended lamp + 15" + std LL optics

and ran tests with sticky tape,,,, well my results still stand,, the gain was proportional on both,,, and another note,,, we adjusted both PJ's to same levels,,
the only thing was my panel was superior on contrast,, as 15" was slightly briter but due to lower contrast range (benq 567)

our tests elimenated several things that people have been arguing over lately

I only brought these points in because it is all relavant to what we are doing here too,,, I dont want to pollute this thread other issues,,, so this is a once off

Clive

Well I didn't really mean arguement, more like lively discussion. I just don't get what it is we're still trying to figure out.

Ok so first we remove the antiglare, with EXTREME care. Then what?

Hope your a cheap date Elken cause I sent you a bouquet of economy grade polarizers

In hindsight I'm sure glad you took back the experiment you left and tried again. There was a moment there when I thought you and Mark would be butt'n heads rather than collaborat'n

This is exactly the same question I asked Mark earlier. 'Would allowing the panel to work with the crystals pointing the direction they were designed originally be more efficient.' Although, it has been done where the LCD was reversed in a pj, some say there was a better projection others say there was no change in lux. I would have to say give it a test without the antiglare. I have no clue about how the crystals move, obviously they react to electrical current and definately not mechanical (weeee little sockets n' springs) but I'd have to speculate that there might be another source of lumens we might not know unless testing this route is tried. Hell, you already murdered the laptop, I say 'drag' it around the city for all to see!!!

Did a quick search on how the lcd looks discected, check out this site. Just wish I knew how to copy/paste the picture rather than linking the site.

http://www.manufacturing.net/ctl/article/CA528274

Heheheheh I am never cheap,, I just do it cheaply...

hey new purpose for polariser's,,, when the spouse starts nagging,, usaully rightfully so hehehhe,, new version of "the cone of silence" "the cone of polarisation"

nah only goats butt heads,,

nope mark has proven to have done his research well,, even gone beyond mine, on this issue,,, but his cause is extremely worthy,,, because at the end of the day,, geez its nearly the end of the year too....

if we can gain a 20% amount without major hassels,, we are streets ahead..

the only part where I thought, we might butt heads was about the "tape senario"

but however these are two parrallel paths, one is the high road, the other the low path,, both will work,, just need the process to do it,

upto this point I have not dunked the laptop,,, but will do before the end of the weekend,, had several sidetracks along the way,,,

I now have the small polarisers out of a prof PJ,, these are better than retrieved from dead panels,,, I will setup rigs inside PJ to take pics of both pre and post Fresnel and between fresnel and triplet,, and also post triplet, all of these will intially be with the original "shiny, backlite side intact,,, then I will strip part of that aforementioned surface too,,, it would be good to get big sheet of polariser,, but so far cost here in aus is $250 for 400 x 400mm,, grrrrrr

another source for polarisers maybe from astronomy suppliers, so if you guys, can follow up, that,, edmund is one,, cant remember some others, been a long time, for me and optical glues..

one method i can think of is surface wetted with "whatever", then slide thin sheet of glass onto LCD glass,, this would overcome the trapped bubble issue,,
discovered this with the glycerin test,,, mind you my lcd is in metal frame,, but was also worried about it sticking too well,, because if you have two sheets of glass and simply put water on one and then slide the second on,,, nothing will move them,, even sliding after a while is very difficult,,, try it yourselves

it is possible to get optician to do the job,,, they do this sort of thing often,,

as another example,, doublets and some triplets are bonded together,, so the glues are out there,,, just need the method for us..

clive..

When you "Add Reply" in the area where you can change the font is a box "IMG" , click on it and put in the URL

quote sim

not a lumens gain,,,

i beleive it is the correct order of tranmission,,, the last active (passive really) should be the colour filter mask,,, just seems logical

maybe a closer look at pixels on screen,, would show, this to be true or not, possible no one, who has done this have checked that yet

now macro photo's need to be done,,, darn another thing on my checklist, heheh