Not perfectly plane-parallel filter

[JPD]

I have a set of 29 mm filters by Voigtländer från the 1950's-60's and I noticed that the medium yellow glass is not perfectly plane-parallel. The view shifts a very tiny bit when I look through it. Would this degrade the picture quality?

 

[wiltw]

The point of the lens is to taken rays of light from all portions of the lens and focus them all at the same point. A filter which is not plane-parallel would seem to defocus some of the rays of light...suboptimal.

 

[Nodda Duma]

wiltw it seems like it would but an ideal filter in front of the lens -- tilted or no -- doesn't impact performance.

A filter has no optical power and -- as long as it is in object space -- does not affect the optical performance of the lens behind it. This is primarily due to fact that rays entering a thin glass plate -- tilted or no -- leave at the same angle, so the sum of any non-zero aberration coefficients (lateral color) cancel out to zero.

There are a couple caveats to that: 1) color filters reduce the spectral band passing through the lens and therefore reduce chromatic aberration. 2) surface errors cause a waveform deformation and reduce slightly the performance (the advice to buy a good quality filter is good, quality advice). 3) wedge error induces chromatic effect but practically speaking is negligible.

A tilted filter vs. a non-tilted filter in front of the lens will shift the image as a function of the index of refraction of the filter glass and the thickness, but that's it.

Just keep in mind the filter is part of your optical system, so all the advice about using good quality optics apply to it as well. Use good quality, AR-coated filters so you don't get a 4% back reflection overlaying onto your primary image and manufacturing errors (wedge and surface figure as described above) aren't a concern.

Filters inside the optics or between the lens and image plane are a different story and must be accounted for in the design when possible.

 

[AgX]

But a tilted filter is something different than a filter with not parallel planes.

 

[Nodda Duma]

Ahhh you're right, it's not. Darn my sporadic reading comprehension.

A filter with wedge will induce a slight chromatic effect. But I think even for cheap glass this chromatic effect - especially for a filter that cuts down the spectrum -- is negligible.

Yes it'll shift the image. A tilted filter will also. A filter with poor surface accuracy may. That in and of itself doesn't impact performance. But if the filter is so crappy that wedge induces a noticeable shift, then I'd toss it in the dirt and get a better quality one. While the wedge itself doesn't reduce the quality, such a cheap filter will have other issues (surface figure, striae, etc).

 

[JPD]

Thanks for the replies!

A filter with wedge will induce a slight chromatic effect. But I think even for cheap glass this chromatic effect - especially for a filter that cuts down the spectrum -- is negligible.

Yes it'll shift the image. A tilted filter will also. A filter with poor surface accuracy may. That in and of itself doesn't impact performance. But if the filter is so crappy that wedge induces a noticeable shift, then I'd toss it in the dirt and get a better quality one. While the wedge itself doesn't reduce the quality, such a cheap filter will have other issues (surface figure, striae, etc).

This is a original coated G2 Voigtländer filter of high quality in their excellent chromed push-on mount. Not sure how it escaped their quality check. I will try to find another one, and in the mean time use the lighter G1 filter. I have a 30 mm filter too that I guess I could adapt to fit, but then I'd need to bring a 30 mm hood as well as the 29 mm.

I remember reading in a Rollei book the tip that you can use the Rolleiparkeil parallax correction accessory for the Proxar close-up lenses to get straight lines when shooting tall buildings. The Rolleiparkeil is a glass wedge, with a much greater shift than this filter.

Just keep in mind the filter is part of your optical system, so all the advice about using good quality optics apply to it as well. Use good quality, AR-coated filters so you don't get a 4% back reflection overlaying onto your primary image and manufacturing errors (wedge and surface figure as described above) aren't a concern.

These filters are AR-coated. I would buy new ones if B+W or Heliopan still made filters with push-on mounts. They stopped making them a few years ago. :-/

 

[Nodda Duma]

Speaking of wedges, one of the techniques used in optical systems for precision real-time boresight alignment is the use of a Risley prism - a pair of wedges rotated with respect to each other to adjust beam steering of a laser or optical field of view.

 

[John Koehrer]

Take a look at the inner rim of the filter, the glass is usually held in by a retaining ring or a snap ring.
If it's a snap ring it should just press in.

 

[JPD]

Take a look at the inner rim of the filter, the glass is usually held in by a retaining ring or a snap ring.
If it's a snap ring it should just press in.

It looks like this:
http://www.pacificrimcamera.com/images/160854.jpg
https://www.fk-secondhand.com/shop/...laenderuv-filter_32mm_20150423_10_600x600.jpg

There's no retaining ring, and it looks like the metal is pressed firmly around the glass.