Why is there no 'Amici mirror' for erect images / mirrors vs prisms

[skysurfer]

I have an Amici prism for my small 8cm telescope and sometimes I use it with my Televie Genesis to avoid mirror images to compare it easily with star charts.
A 'normal' diagonal has only one reflecting surface, an Amici two and binoculars have four. In the latter two cases prisms are used which also add one extra glass-air transition with light loss. Although modern higher end multicoated prisms have light loss of no more than 1%, every % is one too much.
Better quality normal diagonals as the one in my Genesis are mirrors not prisms. And in SLRs the same: high end cameras such as the Nikon D4, D800 or Canon EOS 5 and 1 have a pentamirror instead of a pentaprism.
But there is no such thing as am 'Amici mirror', i.e. two 45 degree mirrors which make the image erect over a 45º or 90º viewing angle.
My questions:
- why is there no market for an Amici mirror as technically it is possible ?
- why do binoculars not use mirrors as 4 mirrors have less light loss than 4 prisms ?
- why do binoculars with 90º viewing angle not use Amici prisms (or mirrors) ? Less weight and light loss.

No I don't like prisms particularly not PRISM.

[Merlin66]

Comes down to the limited market...and bang per buck.
Easier to get a prism or two out of India/ China than a couple (or few) 1/10 wave coated front surface mirrors and support systems to align them....

[dannat]

Baader make a 2" amici prism diagonal, a good prism glass surface while expensive can rival a mirror, & the coating won't degrade.

Mirrors a flat & would be hard pin down -to collimate & adjust than a solid little glass prism, which has little screws pushing against it

The other point about an amici at high mags the glass ones can cause little degradation spikes on bright objects, like planets

[Steffen]

I think you've got it backwards. As far as reflectivity is concerned, nothing beats the 100% total reflection a prism can offer. Prisms also have less light scatter than metal coated mirror surfaces.

All higher end SLR cameras (incl the ones you've mentioned) have pentaprisms, only the cheap models use pentamirrors. With the multiple internal reflections necessary to make a pentaprism/pentamirror work the mirror type will lose more light and result in a dimmer view finder image.

Prism diagonals lose light through reflection on the entry and exit surfaces, but modern multi-coating can bring that down to as little as 0.1% per surface IIRC.

Cheers
Steffen.

[Wavytone]

Two-mirror solutions do exist, and have ben used many times before, e.g.. in periscopes and in big binocular telescopes where prisms would be impossibly large.



They don't - 2 mirrors are sufficient to produce an erect image and without left-right reversal.



Porro prisms suffice, and the cost of manufacture is very low. Historically, after WWII many of the plants that made pentaprisms for SLR film cameras could crank out Porro prisms for next to nothing and small roof prisms weren't much harder. Roof prisms are still used in the small binoculars and Zeiss, Leica also make larger but rather expensive pairs. Porro prisms in particular provide a very useful aspect for the average binoculars in the range 20-70mm aperture where, for mechanical reasons, it was desirable to make the spacing of the optical axes of the two objectives somewhat larger than the inter-pupillary distance between your eyes - and the binoculars also had to provide an easy way to adjust for variations between users with different IPD. The traditional binocular design with Porro prisms provides an elegant way to do this - and achieve a very compact length as well. In comparison roof-prism binoculars are quite a bit longer and heavier, and due to the fact that most peoples eyes are about 70mm apart this poses an upper aperture limit on any design that uses co-axial optics (such as roof prisms).

Amici prisms aren't a solution for binoculars.

As indicated below total internal reflection is a very useful thing, better than a mirror. In addition they are easier to keep clean, and better yet, the surfaces do not deteriorate with time the way first-surface mirrors do. If commercial binoculars were based on mirrors they would have a relatively short life in the hands of most owners before the mirrors are so degraded as to be useless, whereas the life of prism binoculars is limited only by the time for fungus to set in, or corrosion or mechanical misadventure (dropped) to ruin the assembly.

Large prisms however pose several issues:
- mass, which increases as the cube of the dimensions;
- passing the convergent beam through a long optical path in glass between plane surfaces introduces spherical aberration. The objective must be designed to compensate for this, or alternatively mirrors can be used (which contribute no spherical aberration).
- manufacturing cost, which increases as the 5th power of the dimension, due to the difficulties in maintaining accurate angles, while making large optically flat surfaces.

The other bit of history is that until the 1950's the vast majority of binoculars were sold to the military - in Japan, the US and europe. Small size and light weight were high priorities - whether it be the army footsoldier in the trenches who had to carry everything, aircraft (every pound of weight mattered) or warships (size and weight). The Porro prism design was always preferred because it was the most compact and lightest. The same applies even now if you are bushwalking or backpacking.

You have a wee bit to learn about optics, and prisms in particular, Grasshopper.

[The Mekon]

Well posted Wavytone!

[Don Pensack]

I'll add the problem with Amici prisms for astronomical use: the line down the middle. Though relatively invisible at low powers (binoculars), it is very visible at high powers, especially when a star drifts across the line.
The split and recombination of the light path can produce phase issues upon recombination as well.

As if that weren't enough, there is the issue of chromatic aberration with oblique angles of entry ("prism", remember) which confines the best use of a prism to long f/ratio telescopes, where the angle of incidence varies little.

As far as light transmission goes, I see little difference between a multi-coated prism and a mirror diagonal. However, at least in theory, a 1/10 wave mirror would result in a 1/7 wave error after passage, while a prism with 1/10 wave surfaces would result in a 1/3 wave error after passage. A prism would have to have more accurate surfaces to produce the same wavefront error after passage. Yet another reason why prisms are typically confined to low-power use.

The question of having the image not match star charts is one that typifies someone with a little less experience in the field [not that, from the list of scopes, this applies to the original poster]. I felt the same way for most of the first few months I observed (I had a reflector, where the image was upside down relative to my charts), but quickly adapted to the difference in orientation. Computer charts can be printed with any orientation, even the mirror-image orientation of refractors and catadioptrics, but pre-printed chart sets can be difficult to use if one has a variety of scopes with a variety of orientations.

Assuming that high-power use is planned, it is hard to beat a good mirror for use in a star diagonal, IMO.

[Don Pensack]

Info on "roof" prisms:
https://en.wikipedia.org/wiki/Roof_prism

Note: Olivon is bringing into the US a 2" roof prism 90 degree diagonal.
At this time, it remains to be seen what quality this prism will have, but a correct-image 2" 90 degree diagonal is not a common thing. I can see this being very popular with refractor users who also want to use their scopes for daytime use. Expected to arrive in the next couple months.

[Steffen]

Interesting, I'd like to see that. Wouldn't a 45˚ prism be more suitable for daytime use, though?

Cheers
Steffen.

[MortonH]

I think so. I have a 2" William Optics 45-degree diagonal for that purpose.

[Don Pensack]

Well, it depends on the height of the tripod.