PDLC film or glass for optical shutters or shutter glasses

[kekkked]

A PDLC film or glass also known as "smart glass" is a transparent film or glass with liquid PDLC crystal inside which, when electrical field is generated, can act as an optical shutter. Unlike the traditional LCD shutter it

1) doesn't block light but scatters it
2) it makes no use of polarizers so the light transmittance efficiency is about 85% over about 45% for LCD.

It is also very cheap and can be cut into any shape or size and as far as I know can be thermoformed to make curved goggles.

Why isn't it used for 3d stereoscopic shutter glasses you might ask? Probably because it
1) needs 60V AC to operate in normal operation mode (electric field on/off) over the 3.3-5V DC for LCDs
2) because ordinary PWM AC voltage fading speed of the material isn't fast enough
3) but most probably because 3d TVs died around the time several companies were developing PDLC shutter glasses and when it comes to 3D cinemas those use a different technology without shutter glasses so there hasn't been a market need to develop them.

For a shutter glasses usage PDLC needs to switch from on state to off state or off stage to on stage in 0.008 seconds, or 8 ms, but with ordinary electrical field switching the max speed it achieves appears to be 40 ms, which is darn close.

Luckily there's a different method to switch the state of the PDLC crystal discussed in this paper:
https://www.researchgate.net/publication/3254347_Reduction_of_the_switching_time_of_polymer-dispersed_liquid_crystal_using_field_oriented_addressing

By the method described there PDLC switching times can reach the level needed for 3d shutter glasses and unlike 50% light loss from LCD shutter glasses they will preserve 82% of the unpolarized light. And it might not even need 60V AC and work at lower and/or DC current.

...but the wording in the above paper is a bit vague. If anyone understands what "two finger electrodes" means and how they work in this context and if they could work with a transparent conductor such as ITO film instead of an aluminum reflector used for PDLC screens I can give this idea a try.

Using a 3 electrode pixel electrode (Fig. 1) it is possible to generate a horizontal and a vertical electric field. One transmissive indium tin oxide (ITO) electrode is on the cover glass and two Al finger electrodes (electrode A and B) are on the back plane of the display. The ITO electrode has a fixed potential.
When electrode A and B have the same potential different from the potential of the ITO electrode there will be a vertical electric field across the cell gap of the display. When the difference in potential between electrode A and the ITO electrode and between electrode B and the ITO electrode are the same in magnitude but with opposite sign, the electric field in the cell gap between electrode A and B will have a predominant horizontal component. In this way it is possible to switch between a horizontal and a vertical electric field in the cell gap between the finger electrodes.

I already have some spare PDLC films and a lab PSU with oscilloscope, just let me know if you are interested and let me know what experiments to perform.

PS. I'm interested in implementing PDLC shutter glasses in my VR system over the existing LCD shutter glasses to improve brightness and have no interest in making profit on the glasses but you might find some real business potential in helping get a working low power but high performance PDLC shutter glasses prototype made and use the gathered info in your own product. For example shutter glasses are still used by NVidia and home cinema DLP projectors. I don't mind.

 

[berkeman]

2) it makes no use of polarizers so the light transmittance efficiency is about 85% over about 45% for LCD.

That's pretty impressive. Do you know if any manufacturers are looking at using it in flat displays? I don't think the higher drive voltage requirements are much of an issue -- plasma displays and EL backlights have needed higher drive voltages for years.

It looks like the paper you linked is an IEEE proof copy? Do you know if anybody is going to publish something in the SID Journal?

http://onlinelibrary.wiley.com/journal/10.1002/(ISSN)1938-3657

 

[kekkked]

No I haven't heard of any company working on PDLC screens, may be because the subpixel-based LED screens are cheap and look good enough rather than color-sequential PDLC displays.
I don't remember which company was working on PDLC shutter glasses but it never came out probably because 3D TVs slowly disappeared and the stereoscopic 3d projector and monitor market wasn't big enough for them.
Sorry, not aware of any other publication on high refresh speed PDLCs.

 

[kekkked]

If there is a more suitable place to ask for help on this please let me know here or via PM.
Its just about switching the angle and orientation of electric field so should be simple for most people.

 

[Tom.G]

If anyone understands what "two finger electrodes" means and how they work in this context and if they could work with a transparent conductor such as ITO film instead of an aluminum reflector used for PDLC screens I can give this idea a try.

Two finger electrodes
Take two ordinary combs (hair combs) and lay them on a table with their teeth facing each other. Label one of the combs "A" and the other one "B". Now push the combs together so the teeth of one comb go between the teeth of the other one. That is the "two finger" configuration referred to in lower part of fig. 1 of the article. I called them teeth above for ease of explanation. They are normally called fingers. Of course when you build such a thing you don't want the fingers (teeth) to touch each other because the "A" fingers need to have a different signal on them than the "B"fingers.

The upper part of fig. 1 shows the conventional construction of a PDLC. If you drive the "A" and "B" fingers with the same signal, the device will act just like the conventional construction; i.e. apply a voltage between the top conductor (the ITO) and the bottom conductor (the fingers), and the molecules will line up with the electric field to pass light, the "On" state. When you remove the voltage, the molecules relax to random orientations and block the light, "Off" state. The response-time problem arises because there is no external force (electric field) to drive this relaxation, it is determined by the Liquid Crystal mixture and is relatively slow.

The key insight in the paper is by using the "A" and "B" electrode configuration on the bottom side, you can apply a voltage between the "A" and "B" fingers to supply an electric field that is mostly perpendicular to the "On" state. The Liquid Crystals then line up with this applied field, rapidly rotating to the "Off" state and blocking the light transmission.

Finger Material
Nothing special about using a different material for the bottom conductors. The example in the paper was for a reflective display, therefore Aluminium was chosen for the bottom conductor, it makes a good mirror. You are after a transmissive display, so use something transparent like ITO.

Interesting project. Please keep us updated on progress.

Irrelevant footnote:
The layout of the fingers described above is usually called "interdigitated". The word comes from "inter-" meaning between, and "digit" meaning finger. When you clasp your hands together with the fingers of one hand going between the fingers of the other hand, the fingers are said to be "interdigitated".

 

[kekkked]

Excellent explanation, thank you!

I just had an A4 sized PDLC film with a regular power supply and switch shipped to me for testing.

The two finger electrode sounds simple enough, however it depends if we can both find a off-the-shelf material to use for it otherwise manufacturing cost will be too high unless for large corporations.
Also, it depends on whether we can get the electrodes translucent enough to not beat the whole point of going with PDLC over LCD (that is, more than 50% unpolarized light transmittance).

I thnk ITO films can be etched without complex equipment. Just to be 100% sure, this is the arrangment of the electrodes we are talking about?

An ITO film can be etched to create such a grid with gaps between them. question is for shutter glasses whether we can get the gaps small enough to not be noticeable by the naked eye. A two sided ITO film would solve this problem easily by etching each side accordingly.

Any suggestions for the width of each finger?

 

[Tom.G]

Just to be 100% sure, this is the arrangment of the electrodes we are talking about?

Very close. All the Blue lines should be connected together and all the Red line should be connected together.

A two sided ITO film would solve this problem easily by etching each side accordingly.

Two sided fingers would not be ideal, the electric field will be largely thru the substrate, whereas you need at least the fringing fields to be thru the PDLC.

This brings up the subject of finger spacing versus thickness of the PDLC. The larger the finger spacing the deeper the fringing field will extend into the PDLC. itself The tradeoff is a higher voltage between the fingers is needed. Sounds like a trial-and-error effort or much more research.

Any suggestions for the width of each finger?

Whatever is the smallest practical without manufacturing defects. Since the ITO is transparent and will be very close to your eye and out of focus, the actual width isn't likely to be critical.

I just held a black pocket comb against the front surface of my glasses and could see through it. Being opaque, it was obviously present and annoying. Spacing was 16 teeth per inch, (0.0625in C-to-C) with a gap of 0.020in. With careful setup, you should be able to do quite a bit better. Hombrew circuit boards with 0.050 center spacing of conductors are doable... and it sounds like you have done this before.

 

[kekkked]

I see, thank you.
Obviously for glasses (or visor) the working area won't be a perfect rectangle but I can make the density equally distributed via software by varying the width of the electrode lanes and their spacings slightly, accordingly.

I have a camera which can record at 480 fps at 224x160 resolution. That sounds very low resolution but for close up of a shutter should be more than enough.

I think there might be better conductive inks than ITO coated PET for just printing the ink from an ordinary injket or laserje printer-type machine from a high res bitmap image file. Or maybe there are options for ITO as well. But this is a question for another section.
I'll get back when I learn more on this.

 

[kekkked]

Any suggestions how to remove the original ITO pet film to be replaced with a custom etched one with least possible damage to the PDLC itself? PDLC layer is adhered to ITO PET films from both sides so there's a risk of breaking the PDLC film and leaving chunks of it on the unused PET film which is to be swapped with a custom one.

EDIT: Like this then?

 

[Tom.G]

Any suggestions how to remove the original ITO pet film to be replaced with a custom etched one

Yeah, don't remove it, just etch the new pattern into the existing ITO. (That is if the ITO is on the outer surface.) If the ITO is on the inner surface, find a solvent for whatever adhesive is used. Try the common ones first like corn oil or rubbing alcohol, then go on to Vinegar or Ammonia. You may have to find the outfit that made the assembly and ask what the base of the PDLC film is, what the adhesive is or a safe solvent to use. Or you can dissolve the layer of PET. For instance Lye (Sodium Hydroxide, drain cleaner) will attack PET, eventually, so will Acetone; but Acetone attacks many plastics. They are all available over-the-counter but be sure of good ventilation and wear safety gear when using the stronger solvents.

Like this then?

Yup.

 

[kekkked]

ITO is on the inner surface of the PET films.
I could dissolve it, but I don't see how you can dissolve just one of the ITOs. And I'm not sure if I follow you, you mean to dissolve just the ITOs of the PET films, then overlay 2 extra ITO PET films on those pet films? Each PET only has light transmittance of 85 percent. 4 together adds up.

For dissolving the PET films themselves, i hope that will work. Will need pdlc and the polymer it is cured with remain as one piece after the pet laters are gone.

 

[Tom.G]

I'll try to clarify. There were three different approaches suggested.

1) If ITO on outer surface. It isn't so next approach.

2) If PET/IDO glued to PDLC try dissolving the glue to expose the ITO

..2-a) With ITO exposed etch the existing ITO to create the interdigitated fingers. Then reattach the PET to PDLC

3) If 2) doesn't remove the PET, then dissolve the PET; assuming that is possible without damaging the PDLC. Then create the fingers and attach new PET layer

 

[kekkked]

Maybe better to first try to dissolve the glue keeping the PET (and the ITO on it) attached to the PDLC as PET is a bit tricky to dissolve.

 

[kekkked]

By the way the lines are only connected on one side in the diagram, is that fine?

 

[Tom.G]

By the way the lines are only connected on one side in the diagram, is that fine?

Yes. Your current is tiny so there is no problem with voltage drop.

Another source would be a flexible printed circuit. At least for convential circuit boards, there are many suppliers online that supply free layout software where you do the design, upload it to them along with shipping and payment info, and you receive the boards several days later. I haven't looked, but there must be flexible circuits available the same way; you may have to search, investigate/negotiate about transparent/translucent conductors though.

 

[kekkked]

I'm not concerned about voltage drop but rather whether this layout will generate the electric fields with the orientation and angle needed.

 

[Tom.G]

I'm not concerned about voltage drop but rather whether this layout will generate the electric fields with the orientation and angle needed.

Well, it comes down to calculate it or try it. I understand there are some pretty good Field Solvers available but I've never used them. Perhaps others here can point out some useful ones.

I did a search here on PF for 'fringing fields' and found many hits. Here are a few:
https://www.physicsforums.com/threads/what-causes-fields-to-fringe.124150/
https://www.physicsforums.com/threads/calculation-behind-fringing-field-sensor.125786/

Here is a 3 page paper you may find useful, it also has a sketch of capacitor field lines on the third page.
http://physics.usask.ca/~bradley/pll_plate_edge.pdf

 

[kekkked]

To update on this I have cut 3 small peices from a bigger smart film and submerged each in 1) acetone, 2) paint strip solution, 3) 35% HCl Zn. Will check back after few hours see if any affected the bonds.

By the way is there any info whether there's some adhesive sticking the PET to the PDLC layer or is the polymer in the PDLC the binding agent?

 

[kekkked]

Neither acetone nor paint stripper detach the PDLC and ITO PET layers. Acetone and paint stripper also don't seem to have visible effect on the PET so far. Acetone does remove the glue from the protective layers on both outher sides of the smart film (PET and PDLC layers) so I'm guessing the PET and PDLC are bonded together with a different adhesive or more likely by the polymer in the PDLC itself. Thoughts?

It is possible to peel off the two PETs from each other and switch one of the ITO PETs with a custom etched ITO PET for testing but it will be a grainy pattern in the opaque state with transparent holes rather than uniformly opaque as some of the PDLC will remain on the other discarded PET layer. But it is enough for testing the switching methods described in the paper. My 1000 fps camera will ship soon.

So
1) Time to test the patterned electrodes described in the paper. Anyone ready to give advice upon checking my design? Will give back all the info and results I get from my tests.

2) Any ideas how to reapply or apply a PDLC film on an ITO PET? The original PDLC can be scrapped from the surface and reapplied like grease, but it's hard to apply at as uniformly and also it doesn't bond the ITO PETs together very good anymore.
Second option is getting PDLC in liquid form and applying on a clear ITO PET yourself. I have only managed to find one supplier of PDLC mixture from Taiwan but they only sell 500g bottles which cost approx. 3000 USD each. If anyone can find a different supplier let me know.

 

[Tom.G]

Before proceeding you might want to check the switching speed of the unmodified film and the number of cycles to end-of-life.
Faster switching could probably be reached with a higher voltage, but may reduce life.

This brochure states Off -> On time as 100 milliseconds. (Specifications on pg.5)
http://www.dmdisplay.com/data/DMDisplay_Catalog_2015.pdf

This site shows cycle life as 3 million On->Off cycles. (see Specifications, about 60% of the way down the page.)
https://www.smartfilmglass.com/

 

[kekkked]

Thank you very much for the link. However I'm a bit skeptical about the cycle life mentioned on that website, sounds more like product marketing than some result backed by experiments. But that said that is for around 60V AC while we need few DC volts.

If it were the case, a PDLC shutter switching at 120 Hz would die in just 6 hours, which is pretty bad.