DIY anamorphic lens

[JudeBarnes]

Tor, on a few of those pictures looks like back acryllic for the sides of your lens? Did you find some that had a flat (non glossy) non reflective surface facing the inside of the lens? I'm not convinced this element is actually that important (having black sides vs. just glass).

I filled my 2nd protoype last night and neither leaked the first try! We fired up a movie and put it into action, no focus problems and got the squeeze! A little keystoning on the right, but I think I just need to fiddle with the prisms a bit more.

Here's the big question, now, Tor, or anyone, do you know the angles required for a 2.35 squeeze (vs. 16:9), for the 2.35 content films, to use the full panel on the projector and squeeze it back down to it's 2.35 ratio?

Do you think it's possible to use the existing prisms and just move them at different angles to accomplish more/less of a squeeze. I didn't experiment much with this but I did notice I could get more of a squeeze by simply making the angles that the prisms sat more dramatic. I am figuring that trying it with my existing prisms I'll create some drastic barrel distortion, and need different angled prisms to accomplish this?

[tahustvedt]

I did not use black glass or acrylic glass for the sides. I used either an aluminium frame which is painted black on the inside (waterprisms) and ordinary glass which is painted black on the outside (oilprisms).

The idea with using square plastic pipes as frames for the prisms is a very good one. But you need to make sure you use a plastic that can be glued. I dont think PVC can be glued with normal Epoxy glues but I'm not sure. There's only one sure-fire way to find out.

You will get more barrel distortion with more compression. The 2.35:1 Panamorph also has more distortion than the 1.78:1 Panamorph.

I don't know the angles for 2,35:1-optimizd prisms, but I think it's possible to achieve more compression with the 1,78:1 prisms by tilting them. the geometry might become more uneven because of this, the bottom and top edges of the picture might be more compressed than the center. I don't know for sure but I can imagine that happening.


BTW.
I too am considering a NEC HT1000. It looks like a really good projector for HT.


Tor Arne

[JudeBarnes]

Hey Tor I'm pretty sure the focus problems you're having on the corners with your bent lexan is the same "problem" I saw with my lexan faces, in my all acryllic version 1st protoype... They weren't intended to be, but I think they end up bent, not perfectly flat as they are supposed to be, I think they had a slight twist and as a result one or both weren't at the appropriate angles for the whole width of the lens.

I went to glass and it worked well. I will expirement with shifting the angles of both prisms to see if I can get a 2.35 and will post my results.

Tor: Does the excel file that "DECinema" posted in the avsforum thread have the info we need, possibly? I downloaded it and will take a look. Don't want to waste too much time on it though if you can say one way or another defininitively.

Need to still build a frame for mine... I'll be making it out of wood, painting it black... with my father who has the woodshop. I don't have access to aluminum and metal cutting tools like Tor does.

[JudeBarnes]

BY THE WAY.... I'm still very curious what this PVC tubing is all about. Can someone at least describe to me how this tubing is used to house the prism? I just can't picture it in my head.. again, all I can see in my head is pvc sprinkler tubing.. even a large enough piece of that, I don't quite get how it could be utilized...
rwhitley? Tor? Anyone?

[tahustvedt]

I still haven't tried DEcinema's program for calculating the prisms.



Regarding the PVC-tube:

Imagine a square crossection pipe with a width of 10cm. If you cut it at the right angles you end up with a plastic frame for the glass.

I made a simple illustration.


Tor Arne

[JudeBarnes]

that's a great idea... Thanks for posting that illustration, Tor. Now, it's obvious and I should have been able to picture this... In effect that is very similar to some of my original ideas but even better!!

My idea was to make the prisms with two pieces, both acryllic. To keep the number of seams and joints and gluing to a minimum. One piece for the front and back faces, and one piece for the sides and bottom. These two pieces would of course need to be bent to shape, the clear acrylic piece needing one bend in the middle, and bent over something that was at the correct angle. (like a piece of wood or a pre-shaped wedge). The bottom and sides would be one piece of acrylic bent at two places where the sides meet the bottom.

Tor, why'd you make so many models?? Do all of them work?? Did you ever get the bent lexan to correctly compensate for the barrel distortion AND cause no focus problems?

[tahustvedt]

Not all my lenses worked well. Some were too small to fit the picture because of my short-throw projector.

I managed to get a pretty good correction but the bends weren't completely even all around so the geometry on a couple of edges were "wavy". I have a new technique for making perfect bends though which is the simplest way really. It's mentioned in the patent posted earlier in this thread.


Tor Arne

[tahustvedt]

I forgot to mention that I never managed to get rid of the focus-problem. I think it's just something we must accept if we want to correct the picture.


BTW. This thread should be made "sticky".


Tor Arne

[JudeBarnes]

Hey Tor, I wanted to let you know I didn't have any top or bottom "barrel distortion" with the model I just built (at least that I could see), I did have a keystone on the right side which I think I can correct once I actually mount the prisms. Anyway, if you can ahieve one without the barrel

Not sure if the size of my prisms made a difference there, both are 6" wide and they are 6" tall and 4" tall.

I remember having the panamorph actually setup in my HT room and the panamorph had a distinct bend on the top. On the "Arnemorph" I just built there was no bend at the top?

If I had to choose between barrel distortion and unfocused /wavy corners I'd probably choose the barrel distortion, after all, with black borders you can pretty much 'swallow' the bend into a nice straight line at the border anwyay.

[DArthur]

Jude,

I'm also constructing an anamorphic lens (thanks Tor!) and live in the Vancouver area. There is another low-reflective glass called Image Perfect glass made by Zuel, Inc. It has low-reflectivity without using external coatings. I've found a site on the web: http://www.framingsupply.com which will cut the glass to supplied dimensions. I'd like to try and find a local supplier here in the Portland area. Which glass companies have you contacted?

Thanks,

Darryl