Baader Amici Prism Diagonal

[markse68]

not really an eyepiece question but made of glass and images go through it so...

Looking at the Baader Amici prism specs and wondering what they mean by "glass path"? Is that the effective physical length from mounting flange to mounting flange? Does it take into account the shift of the focal plane caused by the glass?

https://www.baader-planetarium.com/en/baader-t-2--90°-baader-roof-prism-with-bbhs-r-coating-(t-2-part-02).html

They say that the optical path is 47.5mm Do they mean the physical length inside the glass which would mean the focal plane would be shifted out 26.5mm (74-47.5)? Or is the physical glass length 30.45mm and the focal plane is shifted out 17.05mm (47.5-30.45) assuming ref idx of BaK-4 is 1.56?

They also say the clear aperture is 31mm but in the images it looks like the edges of the prism clip the aperture mask so would vignette? Unless that aperture mask is bigger than 31mm- anyone got one they could measure?

Thanks,

Mark

[globular]

My understanding is that glass path of 50 means there is 50mm/1.56=32.1mm of glass - with the light travelling 50mm due to refraction. 

And optical length of 47.5mm means there is 47.5mm-32.1mm=15.4mm of additional light travel from the glass surface(s) to the connection flange surface(s).  (Probably 7.7mm each side).

Hence the movement in focal plane of using the prism (compared with straight through) is 50mm+15.4mm = 65.4mm.

I don't have this particular prism to measure it for you - but I do have a baader prism with reported glass path of 50 and reported optical length of 100 and I can confirm my measured optical path in use (~117mm) agrees with this same formulaic approach, i.e. 50 + (100-50/1.56) =117.9mm

(I'm also happy that the reported clear aperture of my prism (47.5mm) is pretty accurate.  But again can't comment on the one you're interested in.)

Hope that helps.

[markse68]

Thanks Glob, but I don't think it can be as I modelled a roof prism with those dimensions and with a physical glass "length" of 32.1mm I don't think they could claim a clear aperture of 31mm like they do 🤔

Here is same prism with 47.5mm physical glass length- it fits better the 31mm aperture they claim but still clips a bit- looks more like their image though. 

I've messaged them to get some clarification

Mark

[globular]

Puzzling.  Please post back when you get an answer.

[markse68]

Hi Glob, just got this reply but it doesn't make sense to me so I think I must be misunderstanding something quite fundamental!

""The glass path is the length which the prism (or any other optical element) adds or removes from the backfocus of your telescope. So, if you remove a prism with 50mm glass path from the telescope, the focuser has to be moved 50mm farther out."

I've questioned them some more....

Mark

[globular]

Their answer is consistent with my response above and does seems to work with their quoted figures for my non amici version - but I agree the figures don't add up for the amici.

Surely for an amici prism the glass path has to be approximately twice the clear aperture - as the light starting on the far left has to travel to the far right (= clear aperture distance) as well as through the prism (another clear aperture distance (ish)).  So if the 31mm clear aperture is right the glass path should be around 60.... or if the glass path is 50 then the clear aperture must surely be less than 31.

The figures they quote for the 2" version of the amici are; glass path 80, clear aperture 44..... this seems just about ok.

I suspect one or more of the quoted figures for the T-2 Amici Prism diagonal is wrong.  My guess is the clear aperture more like 27.

[markse68]

The only other thing i can think they might have done is move the entrance aperture off the central axis of the prism  further toward the base (in my model) This would move the exit aperture further toward the roof apex which might be ok given it is smaller, allowing fir a smaller prism but that would make it directional. 

Mark

[markse68]

It gets more confusing…

According to Edmund Scientific https://www.edmundoptics.com/knowledge-center/application-notes/optics/prism-tunnel-diagrams/, the geometric light path through a roof prism is 1.7071A.

They don’t mark A on the tunnel diagram though 🤔 But in the diagram for a regular right angle prism, A is the length of the short side. 

Surely it must be the same for the roof prism? If you take the ray that lies on the plane through the roof apex it is geometrically identical to a RA prism?

Trying to understand this as i need to calculate size/suitability for my new finder project

Mark

[globular]

I think A in the roof prism is...

And if you equate this prism to a "normal" one then it's A' would be...

 

[markse68]

Aren’t those truncations arbitrary though- I thought they were just trimming unnecessary parts of the prism to save space/weight given you would usually be using a circular entrance aperture?  Edmund were very quick answering my questions and you are correct Glob but it seems a complicated way to say it’s the same path length as the RA prism if you got rid of the truncations and called A the length of the short side of the triangular prism 🤔

They then said “ The light gets reflected from the roof of the Amici roof prism twice and hence the optical path length is not equal to that of the length of the equilateral side of the prism. You can use the optical path length equation mentioned above. “

which confuses me again . Away from the roof apex plane the light path to the roof gets shorter compensating for the folding double internal reflection 

Mark

Edited April 27, 2022 by markse68

[globular]

I equated the two to try and make understanding it easier, not because it is the "right" way of thinking about it. 
The length A is really the clear aperture... slightly larger typically to give a crisper margin.  If you trimmed arbitrarily then, yes, you are changing A but you are not changing the clear aperture (unless you trim too much).  The 1.7071 only holds if your A is "optimally" trimmed.  If you trim differently then you might get 1.6, say. But 1.6A(arbitrary)=1.7071A(optimal)=A'

29 minutes ago, markse68 said:

Away from the roof apex plane the light path gets shorter compensating for the folding double internal reflection

Absolutely.  If it didn't then you would get some rather distorted images!
The reason you get an extra 0.7071A is not because of the roof, but because of the extra glass needed to achieve a roof over the whole clear aperture.

Maybe this way of thinking about it is easier....

The light path through the red bit of glass is A and the light though the two bits of blue glass is 0.7071A,  giving 1.7071A in total.

 

[markse68]

Yes, thanks Glob, I understand it now and it confirms that I was right with prior calculations based on a triangular profile minus the truncations, I think.

Mark

[globular]

I think my attempt to help probably hindered, so sorry about that.  At least it was in tune with Edmund not in opposition.

Good luck with your finder project... look forward to seeing it.

[Louis D]

So, to achieve the same clear aperture as a simple right angle prism diagonal, the Amici prism diagonal version would need a 71% larger enclosure by diameter to contain the larger prism?

[globular]

I think the diameter would be 35% larger.

Edited April 27, 2022 by globular

[globular]

Although actually the housing is not scaled up like that, rather the shape of it is changed.

As you see below, the RA Prism has a square face holding the round clear aperture; whereas the roof prism housing is a rectangle of the same width but 71% longer.

[Louis D]

1 hour ago, globular said:

Although actually the housing is not scaled up like that, rather the shape of it is changed.

As you see below, the RA Prism has a square face holding the round clear aperture; whereas the roof prism housing is a rectangle of the same width but 71% longer.

It makes sense you can't truncate from distance A' to D while keeping a triangular profile to the entire enclosure.

 If you did, it would look more like a GSO dielectric mirror diagonal along the hypotenuse:

[Don Pensack]

In a normal prism, the light path length is equal to A because the light only travels inside a path equal to A from outside to outside.

Due to refraction, it is typically shorter than if the prism was replaced by a mirror.  The light path is therefore A - a certain %.

In a reflector diagonal, the light path is A. [or the prism is A and the mirror is A+ some %--same result, different point of view].

In the Amici, the light path inside is longer because of 2 internal reflections.  It is what Edmund says because the prism is longer from outside to outside and also because of the internal second reflection.

1.707 seems about right.

 

Now, the light path through a prism is not the light path distance through the diagonal prism.

That is measured from the front looking face of the diagonal body that bumps up against the focuser to the opening where the eyepiece inserts.

And that can be reduced with clever machining and a shorter eyepiece tube.

In a 2" mirror diagonal, that is from 100 to 115mm on average, and about the same for a prism.

A standard 90° prism will focus  bit farther out from a mirror diagonal, however, due to the refraction of the prism.

Prisms can sometimes be used to solve a problem of not having enough in travel to achieve focus.

[markse68]

On 29/04/2022 at 00:22, Don Pensack said:

In the Amici, the light path inside is longer because of 2 internal reflections.  It is what Edmund says because the prism is longer from outside to outside and also because of the internal second reflection.

1.707 seems about right.

Hi Don, thanks but I built a model and if the prism is sized correctly to give a real clear aperture of A’ in Glob’s diagram (ie the roof edges are tangential to the circular aperture) then A’ x 1.7071 is exactly equal to A. I don’t think it has anything to do with the double reflections.

This brings up an issue with the specs for the Baader T2 prism though. They describe the clear aperture as 31mm but that would give a geometric light path through the glass of 53mm. Which is bigger than any of their quoted dimensions. The image of the prism does seem to show the roof edges visible in the aperture of the housing so they could be clipping the quoted clear aperture. As Glob estimated above I think the real clear aperture is about 27mm. Baader are busy preparing for a big exhibition but hopefully I’ll get actual prism dimensions from them soon.

Mark
 

 

[markse68]

I’m now thinking that the Baader prism might be too large for my application. Does anyone have one of the WO erecting prisms- either the regular 1.25” or Redcat version would mind measuring the clear aperture? I’ve emailed WO several times but they don’t seem to be there at the moment, and the suppliers don’t know the specs.

TIA,

Mark

[markse68]

WO came back to me and say it is 20.2mm clear aperture which is very small! Too small for me. Baader too big, WO too small. The search continues… Really I need a 25mm prism I think (24mm clear aperture)

Mark

[markse68]

Another question has emerged. When i asked Baader about their T2 amici specs they told me bluntly that it would not work for my application, that i was wasting my time, didn’t give me the specs I was hoping to get and no longer respond! They said that it wouldn’t work because the focal ratio i want to use it with (f3.3) is too fast, that it would need phase coatings or suffer double images, and that they couldn’t condone it! I tried to point out that they seem to work just fine in RACI f4 finders and that from what i’ve read phase coatings in roof prism binoculars improve contrast (not prevent double images) (not only that but mirroring the roof has similar effect to phase coatings and their prism is mirrored 🤔) That’s when they stopped responding…

But they got me worried- any experts here could explain if and why it would work at f4 but not at f3.3?

Mark

 

[DaveL59]

Hi Mark, the phase coating does a slightly different job than simply mirroring the prism and in a bino of x10 or more does make a difference in the brightness and definition of the image you see. As to what Baader are implying I can't say but perhaps they believe that with un-phased reflections you'd get some possible ghosting when the image is re-merged at the eyepiece, perhaps also some loss of contrast?

[markse68]

2 minutes ago, DaveL59 said:

Hi Mark, the phase coating does a slightly different job than simply mirroring the prism and in a bino of x10 or more does make a difference in the brightness and definition of the image you see. As to what Baader are implying I can't say but perhaps they believe that with un-phased reflections you'd get some possible ghosting when the image is re-merged at the eyepiece, perhaps also some loss of contrast?

Hi Dave, I have read a few differing explanations of why phase coatings improve the image in roof prism bins but this one seems typical. Here they say this:

“One remedy is to silver the faces of the roof prism, as Buchroeder found experimentally in 1982. An even better way is to apply dielectric phase-correcting coatings. Zeiss Oberkochen started doing this in 1988, and other firms quickly followed suit.“

https://skyandtelescope.org/astronomy-resources/astronomy-questions-answers/why-do-the-best-roof-prism-binoculars-need-a-phase-correction-coating/

Still am left wondering why it’s deemed acceptable to use amici prisms (are they phase coated? unlikely- are they even silvered?) with f4 finders (Baader even sell one!) yet it won’t work at f3.3. At what focal ratio does it stop working?That may well be the case but i’d like to understand why.

Mark

 

[DaveL59]

6 minutes ago, markse68 said:

 

Still am left wondering why it’s deemed acceptable to use amici prisms (are they phase coated? unlikely- are they even silvered?) with f4 finders (Baader even sell one!) yet it won’t work at f3.3. At what focal ratio does it stop working?That may well be the case but i’d like to understand why.

Mark

 

Doesn't sound like they are using phase coatings and perhaps not even silvered? Most Binos are in the f5-f4 range I think so perhaps they've no experience of using them in faster optic assemblies, or maybe they've done this in R&D before and found it just doesn't work well enough? hard to say as they didn't seem to want to share much in that reply.

As to why the difference, I guess it's a bit like viewing a scene in a camera with/without a polarizing filter. 

Edited May 20, 2022 by DaveL59