Just playing. To tell you the truth I'm considering doing this to my pj. First the mothers, then strip, then if I'm still not satisfied - total removal. Mind you this route also cost less, approxiamately less than 10 US. But whos also to say that I may have been able to 'rip' a panel the first time but then crack the next.

sim

Paint stripper! Now that certainly fits into the extreme category

DJ

BTW, very important...don't get this on your bare skin. Had some on mine and it began to irritate.
sim

Dam, I can't believe I said this and here I am using Stripper

Click to view attachment

maybe this is how the "parallel trasmitance" is higher than the sinlge trasmitance (3M specs...).

Rox,

The way I see it is that an ideal polarizer blocks 50% of the light...say all light components on the y-axis, and passes 50% of the light...say all light components on the y-axis. This would produce a 50% transmittance. Since these are non-ideal they will reflect and/or absorb some on-axis light. This decreases their transmision to 38%(much the same way a tranparent glass reduces tranmittance from 100% ideal to 90% or so). If you put 2 ideal polarizers together, one that blocks x-axis light and one that blocks y-axis light, then you should ideally block all light. That's 100% blocked and 0% passed.

Most of these polarizers show a cross tranmittance of about 0.05%. What this says is that about 0.025% of unpolarized light leaks through each polarizer. Pretty good for an actual system. The drawback, of course, is that to get here they probably had to block a little extra polarized light. Hence your typical transmittance of a single sheet of 38%.

This is why the cross transmittance spec is important to me and why I am suspicious of the 3M polarizers that do not show a cross tranmittance spec. They show 43% light transmittance in one polarizer, but will not state a cross transmittance spec. I cannot believe that this is an honest oversight, since it is a simple experiment. The cross transmittance spec is integral to an lcd. It basically determines the black level and contrast. The 43% transmittance of one polarizer may be due to approx 38% polarized light passing and 5% unpolarized light leaking through. Kinda similar to the weak polarizers in camera lenses.

If this is true then the 3M polarizers would not increase the image quality, but decrease it. It would definitely brighten the overall image, but at the sacrifice of contrast and black level. Basically similar to shining a light on the screen.

Simult8R,

What did you use to "sand" your antiglare coating before applying the stripper/acetone. Is steel wool sufficient, sand paper, etc. What will be too much. I assume very light pressure, just surface scratches? I don't want to accidentally sand through the a/g and scratch the polarizer.

By the way, great work with the stripper discovery.

Mark & elken,

My CMV 529 seems to have a reflective polarized coating on the backlight side. I assume this is to increase transmittance as per Marks suggestion of using a reflective polarizer on the backlight side. Unfortunately, the backlight side is now facing toward the triplet in the pj.

How difficult would it be to strip both polarizers and swap them. I guess this is for elken, mostly. Do the polarizers become kinked, and strectched out during the removal process? Are those kinks and stretches noticable in the image afterward?

The sanding technique is use about 80 grit, and do light circular motions,, it is simply to break surface so which ever solvent you use, will be able to act on layer,, but be careful not to goto deep, I personally used the first one I found,, i think it was about 40 grit..

steel wool wont work, you need to cut surface,, the A/g almost seems to have a armour plate coat to it..

clive..

yes the only downside to all this is to reverse your panel for my style of setup,,

I cut my front polarizer into four pieces and used one to sit behind triplet.. this works for me. also your need your graphics driver to flip image,, NVDIA has NVkeystone which includes this mirror feature,, but it is aggressive on memory..

the ultimate is to replace polarizer with full sized new one, and avoid all the hassle of reversal issues.

oh also dont remove that shiny back polarizer,, even I am not game to try that yet. until new polarizers are fully tested,

suppliers of thse seem to omit a lot of info as to the full spectrum of specs,,

I have now been watching my unit for several hours,, TV DVD,s etc

and I must say there is a very big difference,, in sharpness, colour tones, and the best of all I can leave a light on in the room,, in fact this improves viewing pleasure,,

gone is the feeling that the picture is dim,, even with dark scenes..

one of the DVD's being The Matrix, which is full of dark scenes,, is quite watchable,, even in ambient light,,,

Elken,

You stated earlier to NOT strip out the first polarizer. You said something about the alignment being critical. Could you elaborate on this. And would it be an issue when swapping polarizers.

I suspect that the matrix array in LCD panel The black matrix is a vertical array,, and the polarizer is a 90 degree setup coincides with this,,

I have read some details on how they install them and they seem to do one relative to the panel,, then align the A/G one relative to the first installed one,,, cant fully confirm this point,, the info seems to be a bit elusive,, but if ones thinks about , it does make a kind of sense..

Ok elken, i used the acetone sand paper technique. I seem to have a slight haze left on my polarizer. What do you recomend i use to do a final clean up? Might be acetone damage, now that i think about it. I will go unsplit optics to test. After removing the aglare and putting the square in front of the triplet the image is ALOT better but still just a little hazy.


Thanks for everything so far

Mike try it backto front rotate it fully,, if not flip it and rotate,, there seems to be very definate one way only,, and the rotaion when you get reasonable image, is very fine at critical point

if you can take apic of polar,, and post it,, one with light behind it, and one showing the haze if you can..

just another point refering to reversing panel,, it can overcome by using PJ in vertical mode, and using front surface mirror,, as per my configuration,, be easier than using software to flip etc,,, because even NVIDIA, the mirror works on desktop,, but overlays fail on some ocassions,, as in MCE2005, when full screen reverts to orig panel defaults.... I have not worried too much due vert usage.. just thought I had better mention these facts

Is there an easy way to determine the orientation of the polarization on the polarizer? Otherwise we would have to purchase polarizers large enough to work in any orientation and trim them after application. I can see why some companies would use 45º polarizers as both could be the same and could work on either side.

It has been speculated by Mark (and that sounds right to me) that the polarizer system should work at any angle as long as the two polarizers are 90º offset.

If we are unable to determine the orientation, then we would have to continue to buy the larger pieces.

It would seem to me that complete stripping and repolarization might be somewhat easier in that there would be no question that you had a matched set that way... and if we bought presized sheets, there would be only one part # per panel size if we bought 45º sheets. (that does assume that the any angle theory is proven)

[edit] I should add that I assumed that the polarizers would work equally well regardless of the light direction. This may be a mistake.

We may be stuck with buying oversized polarizers... not such a bad thing.

Without removal:
Click to view attachment


With:
Click to view attachment



See how i still have some kind of hazy residue left?

windex rubbed it for about 10 minutes. put in front of triplet : BAM crisp bright image. Unsplit optics net fuzzy edges, but totally watchable until my replacement polarizer gets here.


ZING!

edit by fuzzy i mean glowy.

Two questions about the Stripper.

1. Does the stripper only remove the antiglare with no harm to the polarbear (just want make totally sure, because I got a Lowes gift card for sweetest day)?

That's right fellas SWEETEST DAY! you can thank me later if you forgot

2. If total removal of polarbear equals a 100% improvement what percentage improvement would you estimate the stripper method to be?

[quote=SIMUL8R,Oct 15 2005, 12:35 PM]
[quote=SIMUL8R,Sep 23 2005, 08:12 PM]
[/quote]
I have soaked a video card in Acetone overnight to remove heatsink adhesive, and the card still runs today (nice "giveaway" as previous owner thought he ruined a gf4 card) . Just make sure that tis is dry dry before you try to place it next to a 400 watt bulb. I think even acetone (fingernail polish) would work well for this project. I it makes it easier,just go to a manly Lowes and pick up Acetone. They have all kinds of solvents.
You want we could pay pal some money to fund this?

[/quote]

Hold up, acetone, isn't this a bit extreme?

[/quote]

Dam, I can't believe I said this and here I am using Stripper

[/quote]
Told ya it wouldnt hurt...Hehe. Yeah stripper is fun to use too. Hell I love strippers. If I could afford it I would use two or three a night.
And you were right it this being extreme SIMUL8R, even has its own forum now.

Now for a quick question to the brave pioneers who have replaced the polarizers I must ask,,,

Do you think you can now get-by with a 250 watt Metal Halide lamp?
The cooling problems would be a thing of the past if we could.

Thanks for doing this and I'm glad I could contribute in some way, even if it was only knowledge of solvents.

I would stick with the 400watt.

I would like to see the math. I was impressed that it boosted to 24%, I can't see a 7% improvement taking it any further than that.We don't know for sure what the extinction is with 550nm. It may well be Greater than 99.99%. The average is 99.98% but the 550nm may be much better (My math doesn't take everything into account).All of my math speaks in terms of 2 polarizers, not 3. Would it be possible for you to show the exact portion of my calculation that you feel is wrong?This post was meant to destroy any further questions on the matter .

SIMUL8TOR: that product seems to work perfect . Awesome work. You shouldn't see a difference between that and full removal.

A quick Pros and Cons sheet.


Polarizer Full Removal:

Pros

Cooler Panel,
- removes material from the panel improving conduction from core.
- moves polarizer generated heat from panel.
- if using reflective polarizers, will be near equivelant to moving polarizer heat from the panel, but without having to move the polarizer.

Polarizer Upgradeability.

Possible light recycler.

Cons

Reflections between the polarizer and the panel could diffuse and thus pass when they should not.

Risky procedure.

Possible alignment issue (but I deduce not).

Risky procedure.

Any dust/fingerprints between the polarizers will show as dark dots with glowing edges.

Analyzer Full Removal:

Pros

Cooler Panel,
- removes material from the panel improving conduction from core.
- moves polarizer generated heat from panel.
- if using reflective polarizers, will be near equivelant to moving polarizer heat from the panel, but without having to move the polarizer.

Polarizer Upgradeability.

Cons

Reflections between the analyzer and the panel could diffuse and thus pass when they should not.

Reflections between the analyzer and the panel will have an effect on image quality.

Risky procedure.

Any dust/fingerprints between the polarizers will show as dark dots with glowing edges.


Anti-Glare Full Removal when using standard panel orientation:

Pros

No chance of glowing particles/fingerprints.

Maintain factory accurate alighnment.

If your panel makes use of compensation film, zero chance of defeating it.

Slightly Cooler Panel,
- removes material from the panel improving conduction from core.

Less risky?

Reduced off axis reflections could mean less diffusion.

Cons

Cooling not as good.

Glue transmittance, but the glue also cuts down on reflections.

Zero upgreadability.


Anti-Glare Full Removal when using reversed panel orientation:

Same as standard orientation except:

Pros

Fewer possible reflections in projected image.

Mark.

The number I found on the sandpaper was 216, whether thats the grit I'm not sure. Treat this procedure delicately, the a-glare is pretty tuff but realize that the polar is just below it. Only score the film ever so lightly and in several rotations and this will make the stripper work rather quickly. Only wish I had done this earlier and sanded the entire area before proceeding.

For those of you 'nuts' enough like me to try this, let us know your experience for others to understand better.
sim

I think im going to convert my setup to unsplit, basically end up with exactly like elkens setup. I can say without hesitation my white level is very very improved. No more yellowing .

1. Yes, seems the polar is even tougher than the a-glare. The only way it starts show weakness is when it looses it's adhesiveness to the rear plastic which I believe holds it in place, but thats only if the chemical of your choice gets behind it or in between the layer. Thats where the weakness is, it's edges or underneathe.

2. Sorry, your gonna have to ask Elken or Mark on this. To be honest, before my ohp blew its bulb, I placed the test 'polished' piece on the panel as well as behind the triplet. For a moment there, I found the contrast incredibly brighter and healthier while behind the triplet. Now, I'm not one to offer scientific knowledge to all this but could it be that the cone leaving the lcd/collector to the triplet is such that the light is brighter and more intense as it goes thru the triplet where it's reverse for projection. That would account (my opinion) why the contrast was so dang bright. It's like taking magnifying glass under the sun then we see a tight bright dot, put your polar right in that dot and 'BANG' intense directed projection.

sim

hehe, I likes your world griff, let's say we both try mechuric acid and see what we come up with..hehehe

250 bulb?...lets tackle the polar first and see from there.
sim

I just finished repeating some of the experimentation. I had too many questions.

1. The anti-glare is about the thickness of a piece of paper. It is about the consistency of a piece of photographic film (but thinner). Very strong.

2. The anti-glare is only roughed on one side.

3. The entire assembly is layered as follows:

a. plastic protector,
b. layer of 45 degree polarization polymers (brushed on to the surface of plastic protector),
c. glue,
d. anti-glare.

Note: this means the polarization polymer is exposed once the anti-glare is removed. Thus it is exposed to any chemical or scrapping technique.

4. The polarizer polymer is an exceptionally thin layer. You cannot perceive it's thickness, it looks like watercolor paint.

5. The plastic protector is the thickest of all the layers.

6. 2cm x 2cm piece soaked in ammonia for 36 hours broke down the glue, but also partially broke down the polarization polymer. The good new is the anti-glare just pulled up easier than a piece of tape would, the bad news is the polarization polymer now is dissolved completely around the edges (where the ammonia seeped in) and has partially dissolved (gotten thinner, but consistent) everywhere else. It now lets in a small amount of blue light (higher energy, tighter wave).

7. Soaked another section 36 hours in Alcohol. Seems to do nothing.

8. Soaked another section in orange cleaner. Non toxic, easy on hands, did nothing . Didn't even break down the substrate to protector glue still left on the test piece.

I can't imagine based on this structure how one could use stripper, or scrapping methods and not harm the polarizer polymer etching. Could you test the crossed transmittance of those sections under bright light, compared to untouched pieces?

Mark.

Simul8r, using that stripping chemical, did you say you have to sand your panel first?

When you began removing was it difficult to first break through the outer surface?

That method does not require polarizer replacement does it? You just strip and go?

Seems in explaining the use of the 'Stripper' I didn't fully explained the procedures enough as I should of. So, I made the neccessary changes. Hope it helps.

sim

Oh and Mark, I knew you'd like that I was able to totally remove antiglare without ripping.

Is the sand paper required or just helpfull? That sounds scary.

So the edges will be ok if you just tape them?

Does this work by itself? You don't need any replacement polarizers?

(edit) oh yeah, and what are the metal brushes for?

All I can say is that I tested a piece of the a-glare/polar from my first rip. Making sure that I stayed away from the plastic protector and I was able to remove it. I pretty much turned my garage from a carpenter shop into a chem lab. Bought 50 bucks more of supplies from Lowes.

Tried M.E.K. seems to react to the antiglare but also the plastic. Still sitting in there hoping to get the polar free of both layers and place in between plexiglass.

Acetone - 2 days, signs of deterioration but the polar is quite brittle and hard to remove both layers without polar seperating into small pieces.

Goo Off - no reaction

Stripper - found it controllable and liked how it worked, but when it touched the plastic protector then there were issues. Thats when I realized this would be better if used while the antiglare/polar was still adhesed to the subtrate protecting the plastic and the with do care protect the edges as well.

sim

(edit) so check cross transmittance while the panel is on under bright light or off.....and what am I looking for?

Again, more corrections on the 'Stripper Method' in the area of the plastic scrapers. Please review.
sim

This is what I was surprised by. It seems there is no encased polarizer. Just the polarizer polymer "paint" on a rear plastic protector. As such, the front polarizer protector that I was expecting to find turned out to be the anti-glare. What I am getting at is we need to be more specific .

The other thing that surprised me was what people meant by "fraying". I thought people were refering to the plastic protector getting cracked at the edges, but now I see you probably meant that the circumference of the polarization polymer (paint) dissolved (and because it is made up of microscopic strings it takes on a frayed look) while the overall structure of everything remained intact.

So what was seperating into small peices? Was the paint flaking off? Or was the protector cracking?Compare the crossed transmittance with an untouched piece.

I am hoping that the ammonia is particularly reactive with the polarizer polymer. What we need is something that will allow the anti-glare to come up as easily as it did with ammonia (it was rediculous), but not be reactive with the polarization polymer. If that could be done, then we need not risk scraping the polymer surface. Has anyone tried Petroleum Benzine yet . It is said to break down the glues, while not affecting the actual polarization polymer or panel electronics at all.

Mark.

I really appreciate your attention. I might want to try this. Today. So I want to be clear on all the details so I don't screw up real bad.

Ok heres what I envision doing:

buying all the equipment you reccomended, except maybe sandpaper (is that really necessary)

taking my lcd out of the projector and laying it on its back, taping up the edges of my lcd very carefully. Applying the stripper to the middle lcd and letting it sink in a few minutes, then gently, scraping at the surface with a plastic scraper working my way out from the middle. When I get to the edges, inch my way closer and closer to them applying the stripper very lightly as I get close.

What I'm worried about is applying the stripper close to the edes. Even with tape it may seap through the aglare and work its way over the edge.

Mikau, it took me roughly 6 or more hours reapplying the stripper to react to the antiglare. The first try left me spots where the stripper worked (not sure why). So, I kept reapplying and scraping, reapplying and scraping until I decided to try an area w/antiglare and sanded. After applying the stripper and waited 10 to 15 minutes it came off quicker.

You might be able to get away with removing the antiglare at the edges if sanding was done (hence quicker). Again, I haven't done the edges on the LCD since it was still in it's case but this stripper is stronger than acetone (I beleive) and I have seen what acetone can do to the film if left there long. I suggest trying a corner first...sand then add stripper, if it comes off quick after scraping then you know your time limits from there. Taping the edges is a choice for precaution. Good luck, and let us know your experience from this should you do it.

I gave you the polish method which shows the equivelent brightness of tape but realize that the antiglare is still there, I'm just offering another method I have done that removes it completely. This method is directed solely for those fearful of ripping the film off, otherwise experiment with all five (5) methods....Tape, Polish, Stripper, Total Removal/Acetone wash, Replace Polar. Your call how far you want to take it.

sim

Sorry SIMUL8TOR but,

Mikau: I wouldn't touch a thing. We have no confirmation that the polarization polymer is fully intact, and I can say pretty confidently that scraping that stuff is probably not the best.

I do not see this as production ready yet. Based on what I have seen this afternoon. We need crossed transmittance results, confirmations, and time, before this can be considered something you should do to your prized panel.

Mark.

The polar was so brittle that when I tried removing the plastic or antiglare off (acetone soak) the film some of the polar as ripping and seperating in a diagonal parallel pattern. My best description of the polar is like seaweed paper you get from grocery stores, very delicate.

Cross transmittance, help me here, are you referring to looking directly close to the monitor and looking for any abnormalities on the stripper side?
sim

The diagonal pattern was the polarization polymer. It is literally just strings of polymer microns thick strung across the polarizer protector at whatever angle the polarizer is to be set. Your polarizer is a 45 degree polarizer, hence you saw polymer strings at 45 degrees. They were so pronounced because:

a. The glue is just not encapsulating the polymer strings anymore, so they are clearly visible due to their diffusion.

b. And it is possible the polymer has been partially broken down, causing a wider separation between the polymers. This is what we need to figure out with the crossed transmittance test.

What you need is a section of treated polarizer, and 2 sections of untreated polarizer. Set a treated and untreated polarizer together such that as little light gets through as possible. Hold it up to a bright light. Observe how much light you figure is getting though, and what color (blue is bad).

Set two untreated peaces such that no light gets through. Hod up to a light and make the same observations. Any change?

If no change, try 2 treated polarizers together to be sure.

Mark.

Will do with what left over test piece I have left (few). Will have to do this sometime later (work calls).
sim

Well maybe I'll just go with the polish for now.

Note: I put a temporary edit on the 'Stripper Method' until it is verified there is no damage to the polar. After proven I'll remove or change the edit.
sim

Something I thought I should mention...

Me, rox, supraguy and some other people were discussing the distortion effects caused by fresnels. The fresnels are not color corrected, and thus bend differant colored light at differant angles. So we get some slight color seperation with field fresnels. As an experiment, I've tried placing the collimator and collector in front of the lcd. Extreme color distortions, you see three slightly offset, red green and blue images. I asked "why does this not happen when we place them behind the lcd?" rox and the others agreed that when we place the fresnels behind the lcd, the colors are separated but the lcd "scatters" the light and helps to recombine the colors. If we remove the antiglare, this scattering effect is greatly decreased and could introduce more severe color seperation even in an unsplit set up.

Clearly removing the antiglare makes a killer improvement but just some minor diffusion may be a good thing for this reason.

Mikau: Im beginning to wonder the same thing too. When I get my ohp's bulb replacement I want to try a side by side w/my pj. The ohp works w/out a collector fresnel but gives a nice image.
sim

YAY! I finally made a usefull comment!

Okay guys, I'm begining to feel like Mikau - completely ignored

I think the post may have been missed due to Simul8r's discovery of using paint stripper to take off the anti-glare...

Not that I am not thrilled that it worked, but that stuff is VERY corrosive. I am no expert, but in my web-surfings, I gather that the polarization sheet is basically Polyvinyl Alcohol (PVA) with a suspension of Iodine heat-impregnated into the matrix as a dichroic element. As such, anything that attacks PVA would presumably not be a good solvent to use. Now industrial gloves are also made of this material and I tracked down a PDF that outlines the chemical resistance of PVA. Perhaps a good starting point in searching for stripping solvents?

http://www2.umaine.edu/SEM/Documents/MiscT...stanceGuide.pdf

Here are some cropped jpgs of the pertinent info.

more:
Click to view attachment

Click to view attachment

Click to view attachment

I hope this info may be of use because it looks like anti-glare stripping is going to be the preferred "minimal-risk" procedure.

My roomies want the projecter off boxs and back on the ceiling, so im going to have to remove about a 7 inch wide band on my antiglare polarizer and tape it back on and rehang it.

Grrrr