Why a pentaprism?

[roygoh]

OK, this may not be correct to the finest detail but I believe the basic concept is there. The dashed line traces the path of the tip of the green arrow within the prism to demonstrate the double reflection at the 2 roof-top surfaces.

Finally, I am clear about this. Thanks to all who have contribution to this discussion.:gbounce:

Both the diagrams are created by me using MS word drawing tools. Guess that means that I own the copyrights to these images. I hereby allow CS members to use them for non-profit purposes if they are of any use at all....

 

[tsdh]

Pentaprism in SLR viewfinder, is to reverse the image into its original position for eye-level viewing. As we know that the lens actually reverse the actual image upside-down and left-right, so the pentaprism do the reversal once more to get it back to its original.
The following diagram probably can help:

 

[tsdh]

Originally posted by ckhaos
However, why not have two mirrors? i.e., replace the pentaprism with another mirror that is fixed at such an angle that it reflects light from the lower reflex mirror into the viewfinder.

The image from lens is reversed upside-down and LEFT-RIGHT. Using the second mirror can reverse back the upside-down, but not the left-right.
Only pentaprism can do the job with minimum component, small and cheap.

 

[roygoh]

Originally posted by tsdh
Pentaprism in SLR viewfinder, is to reverse the image into its original position for eye-level viewing. As we know that the lens actually reverse the actual image upside-down and left-right, so the pentaprism do the reversal once more to get it back to its original.
The following diagram probably can help:

sigh...if you have shown this diagram earlier.....

Anyway, I don't fully agree with your diagram because I don't see why the tip of the black arrow should be reflected off the surface between the 2 roof top surfaces.

My believe is that the enitre image that the user sees should have gone through the double reflection on the roof top.

Point A1 should still be on one of the roof-top surfaces, reflected off the opposide roof top surface, then to point A2, then to the user.

The case is interesting for points that are exactly on the centre line of the focusing screen. Theoretically all rays of light will go through 2 reflections on the roof top. Points on the cerntre line will hit the parting line of the roof top surfaces and their path will not be defined. This seems not to be a problem as the line is theoretically infinitely thin so that it is not observable.

- Roy

 

[roygoh]

tsdh,

Mind sharing your source of the image?

- Roy

 

[tsdh]

Originally posted by roygoh
sigh...if you have shown this diagram earlier.....

Sorry roygoh, I just read this thread today because nothing to do at home.

Anyway, I don't fully agree with your diagram because I don't see why the tip of the black arrow should be reflected off the surface between the 2 roof top surfaces.
My believe is that the enitre image that the user sees should have gone through the double reflection on the roof top.
Point A1 should still be on one of the roof-top surfaces, reflected off the opposide roof top surface, then to point A2, then to the user.

The tip of black arrow need only two reflections to make it right before going to the viewer (A1-A2). But the tip of brown arrow need three reflections: two reflections (B1-B2) to reverse left-right, and the third reflection (B3) to direct the light to viewer.

The case is interesting for points that are exactly on the centre line of the focusing screen. Theoretically all rays of light will go through 2 reflections on the roof top. Points on the cerntre line will hit the parting line of the roof top surfaces and their path will not be defined. This seems not to be a problem as the line is theoretically infinitely thin so that it is not observable.

Yes, the line is infinitely thin, it is not observable.

 

[tsdh]

Originally posted by roygoh
tsdh,
Mind sharing your source of the image?
- Roy

I forgot where the image come from, it is inside my disk since I was student long ago. I always save diagrams, images, articles, etc. which I think may be useful someday. True enough, today after reading this thread, I searched my archive and found that diagram, then I added the wording to make it clearer to explain.

 

[roygoh]

Originally posted by tsdh

Sorry roygoh, I just read this thread today because nothing to do at home.

OK.... Im insulted already......

Originally posted by tsdh

The tip of black arrow need only two reflections to make it right before going to the viewer (A1-A2). But the tip of brown arrow need three reflections: two reflections (B1-B2) to reverse left-right, and the third reflection (B3) to direct the light to viewer.

You are missing the point, I am afraid.

In your diagram the black arrow is LR symmetric, so it does not really review the flaw. If you diagram is true then the image of the top of the black arror is LR inverted. ALso, you are implying that some part of the image is triple reflected while some part is only double reflected.

The whole discussion was about maintaining LR correctness. Well, the later part at least.

Since the prism will not know when it will get LR symmetric images to reflect, the entire image has to be double reflected on the roof top to maintain consistancy.

Originally posted by tsdh

Yes, the line is infinitely thin, it is not observable.

Well, no argument about that.

 

[roygoh]

By the way, the reflection at point A1 just isn't geometrically possible. Just look at the angle of incident of the ray from the reflex mirror hitting the surface containing A1.

 

[tsdh]

Originally posted by roygoh
Since the prism will not know when it will get LR symmetric images to reflect, the entire image has to be double reflected on the roof top to maintain consistancy.

I try to chart if the whole image was reflected on top of the pentaprism, but seems as something wrong, I don't know where..
please take a look at my chart (sorry about the imperfect lines..):

 

[roygoh]

The square you have added and traced seems correct to me, though there is one detail to note. The image on the roof top is the confusing one, because there is supposed to be a double reflection there.

So the dark blue and green sections go up to the nearer side of the roof top, and then get reflected over to the far side of the roof top, thus super-imposing with the incomming light blue and orange sections, then continue down to the surface where A2 and B3 is on.

The same then happens to the all the other sections. The rays come from the focusing screen, hit one side of the roof top (which ever side is depending on which side of the focusing screen they originalte from), get reflected to the other side, and then continue to the final reflection surface.

The problem with the original diagram is that the tip of the black arrow should land within the box you have added on the roof-top.

There is no way the path shown with the reflection at A1 is possible.

B1 should fall at the point where the orange meets the light blue on the far side rooft top, and B2 should fall at the point where dark blue meets green on the near side roof top.

If the Bacl arrow falls exactly on the centre line of the focusing screen, the point A1 should also lan right on the parting line of the roof top, and land at the point where the yeloow meets the red.

The surface where the original A1 point is on should not be involved in any reflection of the desired image at all.

Hey, isn't this New Year's Day in Singapore? Happy New Year!

- Roy

 

[tsdh]

You're right, the image actually do a double reflection on the top of pentaprism. So each point will be reflected three times by the pentaprism before reaching viewer's eyes.
I have draw a simplified diagram to show that.

Happy new year for you.... ok I have to go to airport now, ... see you one week later.