Telescope Reviews: Resolution and phase-correction

What's the difference between a porro prism and a roof prism? Mostly how they're used, IMHO.

They're both essentially 90° plano mirrors. This enables either prism to flip the image left to right or up and down depending on the orientation of the prism.

A porro prism is a roof prism with the roof edge ground off. A roof prism is a porro prism where the reflecting surfaces (planes) meet in an edge (line).

One could easily construct a first class porro style binocular using roof prisms that are not phase corrected.

The huge difference between porro and roof prisms is that the light cone stradles both reflective surfaces and the edge in a roof prism and is reflected sequentially in a porro prism.

Another odd thing about roof prisms is that if the reflecting surfaces are oriented like a porro prism, the reflection is not offset and is directed right back in the opposite direction from where it came from. To get around this, the incident light has to hit the reflecting surfaces at an oblique angle. Not a big problem, just interesting.

This straddling imposes stringent demands on the manufacturing of roof prisms in order to make the reflection visually acceptable. Forget about phase shifts for the time being. The faces have to be FLAT, they have to be ORTHOGONAL, and the edge has to be a STRAIGHT KNIFE-EDGE. Wierd things happen when roof prisms are not well made. These things don't happen in the sequential reflections of a porro prism.

Since one half of the light hits reflecting surface #1 then #2 and the other half hits reflecting surface #2 then #1, these surfaces and their intersection have to be nearly perfect for a roof mirror to pass itself off as a plano mirror with a reversed image. Can this be done, of course and is done every time a roof prism is made. The exacting tolerences do make them harder to make and therefore more expensive, which is common knowlage.

What about phase? I may be wrong, but I think that if a roof mirror or prism is made theorectically perfect, it would not need phase correction coatings. It may even be possible to make prisms to this tolerance, but not be economical to do so. The phase error originates in the splitting and recombining of an image that happens in roof mirrors. Small errors in the path length will produce phase errors when the image is recombined on it's way out of the prism.

How can these small errors be eliminated or minimized. One way is to always make perfect prisms. Another is to "post-process" the prism to reduce these errors. This can be done by adding or removing material in microscopic amounts. Phase coating implies adding material. The coating process is more than capable of laying down incredibly thin, accurate layers of material to a substrate. If the material has an index of refraction close to, or equal to, the substrate it will effectively enlarge the substrate. This process could be used to fine tune the geometry of a prism to minimze this phase error without the labor intensive effort to make "perfect" prisms.

I mentioned that the problem could be approached by adding or removing material from the prism surface. I'm not aware of making phase corrected prisms through the removal of material (other than good old polishing), but it should be possible. The lasers used in refractive surgery can only remove corneal tissue to correct the optics of an eye. These lasers also work on plastic and likely other materials as well. Are they accurate enough to sculpt roof prisms, maybe, maybe not, but they weren't designed to. Perhaps laser ablation will have a role in making optical surfaces when the right person puts his mind to it.

The above is offered as food for thought only. I'm not an optical engineer and my views on this whole subject may be way off base. Feel free to correct any errors so I can learn also.

Peter

--------------------
Peter

Telescopes 25 - 318 mm
Binoculars 15 - 88 mm