CurtPalme.com Home Theater Forum

cmjohnson · Joined: 03 Apr 2006 Posts: 5180 Location: Buried under G90s

I've decided to spend a little money and see if it's even possible to make a reasonably priced anamorphic lens that will work with HD10 series lenses.

This won't be super cheap but I should be able to build a prototype using wedge prism award glass (two pieces) to make an anamorphic lens, at a cost per prism of roughly 80 to 90 dollars each. So for, call it 250 dollars, I will have a single prototype to work with. If it yield satisfactory results I'll build two more and possibly be the first (and maybe last!) person to fully equip a CRT projector with anamorphic lenses.

Potential benefits: Run all rasters at 4:3 aspect ratio, giving 33 and 1/3rd percent more active phosphor area than 16:9 aspect ratio rasters, allowing for 33 percent more light output at a specific contrast setting, and 1/3 greater effective resolution capacity in the vertical direction, which can make the difference between fully merged scan lines at 1080p and scan lines that are visibly distinct. Tube life increases as you can back down the contrast to achieve your light output target if that is a prime consideration for you.

And of course your screen can become much larger. 1.33 times just by adding the lens.

Since, if you're using a CRT projector, you are probably not running a 4:3 aspect ratio screen and are more likely to be set up so that a 4:3 raster spills over the top and bottom edges of the screen, you're going to end up with an image that is both taller and wider.

My personal setup has an 8 foot wide screen and that would imply a screen that is 6 feet tall in 4:3, although actually my screen aspect ratio is 2:1 at this time.

Going to anamorphic lenses and running 4:3 aspect ratio on the CRTs and running in true anamorphic mode, my new screen dimensions would be 128 inches wide by 72 inches high. 10 feet 8 inches by 6 feet. Screen area increases from 36 square feet (assuming a true 16:9 aspect screen measuring 8 by 4.5 feet) to 64 square feet. That's a difference of 78 percent,
and that does bring up the brightness issue again. Going to a smaller screen may be needed to keep surface brightness up, which will require reducing throw distance as well. But that increases angularity corrections needed for the red and blue tubes.

I think it would be best to recalculate so that screen brightness remains constant. We get back 33 percent of the available light generating phosphor area so ideally the screen area should not increase by more than about 33 percent or we end up not just back where we started in terms of contrast settings, but actually have to drive the tubes harder to keep the light output level where it was before.

So, 48 square feet is the new target screen area with a 16:9 aspect ratio.

That works out to a new screen size, call it 112x63 inches, which is 49 square feet. Close enough, you can lose a foot
in the edge masking if you want to.

Or, a hair under 9 foot 4 wide by 5 foot 3 tall.

At this size you're not going to have to run a higher contrast level but the screen is very noticeably bigger.

Of course I will have to figure out the angles between the anamorphic lens elements to achieve a proper 1.333:1 stretch ratio.

Assuming I can get that figured out properly, and it works, then I'll likely have to invest in getting the lenses precision ground flat if needed, and having anti-reflection coatings applied to them
It's that part that could add significantly to the cost of equipping a CRT projector with a set of lenses.

For 600 dollars I can afford it. But if the price is more than twice that, it will become too expensive.

cmjohnson · Joined: 03 Apr 2006 Posts: 5180 Location: Buried under G90s

I neglected to mention that there may be some issue with a loss of brightness due to the fact that no lens is perfectly transparent. I have to expect some loss of available light output thru the lens system. How much, I don't know yet. Hopefully it won't be enough to cripple the final result.

Tim in Phoenix · Joined: 21 Oct 2006 Posts: 4409 Location: Phoenix

Hey CM

This has all been tried years ago, the prism effect killed it. The thirty-three percent gain in light output from using more phosphor is lost in the prism.

gjaky · Joined: 05 Jun 2010 Posts: 2802 Location: Budapest, Hungary

cmjohnson · Joined: 03 Apr 2006 Posts: 5180 Location: Buried under G90s

I'm aware of these potential problems but have decided to explore it for myself anyway. My curiosity demands it.

If it works well enough I can at least use the prototype on my JVC D-ILA machine.

Actually, you do gain the light output of the previously unused phosphor but in order to do that you have to adjust focus and beam shape such that there are no gaps between scan lines as Gjaky mentioned. That's easy enough to do.

The prism effect is something to be explored, and maybe it can be addressed by the usage of chromatically corrected cemented doublets, but that's likely to change the cost and not by a little. If I take it to that point, I'll be enlisting the services of an actual optical engineer for some professional help.

cmjohnson · Joined: 03 Apr 2006 Posts: 5180 Location: Buried under G90s

Actually, if the output from the CRTs is fairly monochromatic, the prism effect (color fringes) should not even be an issue,
as each assembly would only be passing a single narrow color range thru it. Color fringing is something that has to be carefully corrected for when the full color spectrum is passing thru a single lens or prism, but should not be an issue if you have a separate
anamorphic lens assembly for each tube.

I ordered the blanks a few days ago, I'll do a quick first test once they arrive. And then they go off to an optical lab at the local
college for evaluation for flatness and parallellism on all relevant axes. Depending on the quality of the prisms, I then get to decide what to do next. Cost will be a deciding factor as well.

Ideally I'd like to find that they're flat and parallel (on axis) to 1/4 wavelength and then get them multicoated to reduce reflections as much as possible, thus preserving system transmission efficiency and maximizing light output.

That will cost something. Not sure how much yet.

cmjohnson · Joined: 03 Apr 2006 Posts: 5180 Location: Buried under G90s

I got the prisms today. First impression is that they're pretty promising. But all I've done so far is just placed them on a board in front of my RS45 in anamorphic mode. Finding a suitable alignment for good geometry was easy.

Reflections are going to be a problem but I'm going to get the prisms multi-coated on the surfaces of interest,
and I'm flat grinding and black painting the rest of the surfaces so they won't be a reflection source.

It does look promising on first impression. It also looks like the lens assemblies will not be small.

Update: I've already flat ground the edges and blackened them, and have determined that the completed anamorphic lens assembly is going to be heavy enough, and bulky enough, that simply making it in a way so that it clamps to the outer barrel of the HD-10 lens body is probably not going to be satisfactory. Although I'm sure that this will be needed in order to assure proper lens alignment, I think another method of structurally supporting the weight of the assemly will be required. So the assemblies will probably require their own complete separate mounting system which has enough adjustability that each lens system fits properly on the projector lenses in addition to being well supported.

But that's stage 2. Stage 1 is just getting one assembly working optimally. Then I can look at mounting considerations for all three lens systems.

It does appear, so far, that I was right about chromatic aberration/color fringing. Since each assembly will only see one color of light, there doesn't appear to be any such issue to deal with.

Now, using this assembly on a single lens projector, that may be different. That will demand fully achromatic lenses, but on CRT, it doesn't appear to be a consideration at this point.

**AnalogRocks** · Posted: Thu Jul 06, 2017 7:09 pm Post subject:

I remember someone trying this back in ohhh 2004ish(?) They had problems with mechanical alignment in that; the beam divergence between the R,G and B didn't have enough mechanical and/or electronic adjustment to get an aligned picture. Sorry don't remember which projector they were testing with. That may have been on AVS I read that.

**kal** · Posted: Thu Jul 06, 2017 7:19 pm Post subject:

There have been multiple attempts/tries over at AVS since I first started reading there in the late 1990s. So research into those threads and why it wasn't cost feasible would be warranted.

Kal

cmjohnson · Joined: 03 Apr 2006 Posts: 5180 Location: Buried under G90s

I want to do the exploration for the personal enjoyment of it. That others have tried and not found the results viable for whatever reason is truly of no concern to me.

The journey itself is worth the effort.

It already appears that the cost of having the prisms coated with anti-reflection coatings will take them out of the realm of affordability.

No worries, I already have a Plan B. It will require me to make hollow prisms filled with coolant, but allows me to use pre-made, optically flat, multi-coated glass plates of adequate thickness for the construction. But this will require precision glass cutting services, which, fortunately, are locally available and rather inexpensively.

**kal** · Posted: Thu Jul 06, 2017 7:50 pm Post subject:

Nothing wrong with experimenting/learning.

Kal

**AnalogRocks** · Posted: Thu Jul 06, 2017 11:34 pm Post subject:

cmjohnson · Joined: 03 Apr 2006 Posts: 5180 Location: Buried under G90s

At this point I'm not sure if I will!

But like I said, the experiment itself is rewarding.

cmjohnson · Joined: 03 Apr 2006 Posts: 5180 Location: Buried under G90s

I acquired some glass windows that were made for a defense contractor, which were surplus after the project ended,
and they are fully multi-coated and large enough that they can be used to make prisms after they're cut to shape
and assembled into the appropriate configuration. They'll have to be liquid filled with basically the same thing we use
in our LC chamber assemblies. There weren't enough pieces to make three complete assemblies but there is enough
to make one set for proof of concept.