Click here if you are having trouble logging into the forums
Privacy Policy |
Please read our Terms
of Service | Signup and
Troubleshooting FAQ | Problems? PM a Red or a Green Gu.... uh, User
GlennLeDrew
sage
Reged: 06/18/08
Posts: 467
Loc: Ottawa, Ontario, Canada
|
|
A too-small prism? Generally, no. It's almost always the use of a too-"fast" (short f/ratio) objective that's the culprit.
Let's consider only Porro prisms for now. In the great majority of binos, the front and rear prism apertures are the same. So if you use larger prisms so as to get larger openings, the optical path length through the prism cluster scales up by the same proportion. Therefore you gain nothing, because the larger opening is now moved up closer to the objective where the converging "cone" of light is wider.
I've written simple BASIC software to investigate this, and my conclusion is that for "standard" Porro prisms having same-sized front/rear apertures, the objective f/ratio should be no faster than about f/4 if the full objective aperture is to be utilized. But a *great many* binoculars have objectives faster than this (f/3.5 to f/3.8), and therefore suffer from undersized exit pupils, i.e., smaller effective apertures.
For example, an older 100mm binocular I examined had f/3.5 objectives, and therefore worked at an effective aperture of only 78mm (!!!) because the steep light cone could not be fully accommodated by the prisms. If the objectives were instead f/4, all would have been well.
All too often I hear that the "squaring-off" diamond-shaped pattern which causes exit pupil graying in the outer parts is evidence of too-small prisms. No. That's evidence of, again, a too-fast objective for the prisms' glass type. A prism must have an index of refraction high enough so that its critical angle of internal reflection is steep enough to accommodate the steep "cone" of light being focused by the objective. That's why BaK-4 prisms are more commonly used nowadays; they are of higher density, and hence higher index of refraction, than the ubiquitous BK-7.
The "diamond" pattern is the result of the prism cluster's four mutually perpendicular reflective surfaces, each of which fails to fully reflect light from its respective outer portion of the objective aperture. So the gray portion of the exit pupil represents that portion of the objective's aperture from which the light arrives at the appropriate prism's reflecting surface too steeply.
If one made a binocular with longer focal ratio objectives, good ol' BK-7 would serve just fine.
Whew! Hope this makes sense...
-------------------- Home-made 11X50 right angle bino, 8.1 deg. FOV
Modified 26X100 bino, 3.5 deg. FOV
Mediocre minds discuss people. Good minds discuss events. Great minds discuss ideas.
|
EdZ
Professor EdZ
   
Reged: 02/15/02
Posts: 12341
Loc: Cumberland, R I , USA42N71.4W
|
|
Glenn,
along with undersized prisms comes the ability of the manufacturer to build a smaller unit, so in fact it is not a myth it is a reality that binoculars are designed with undersized prisms. Obviously, some binoculars have been tested and proven to have properly sized prims. Is it any wonder that most of these units often have considerably more bulk then their undersized brethern?
Almost all binoculars I have tested that have too small prisms are too small by approx 10%. Increasing prism aperture by 10% (most are approx 20mm), would lengthen a prism by 4mm, probably lengthening prism pair light path by a total of about 8-10mm. If housing size were kept constant, that would require approx a change of 6mm in focal length, or 6mm change in lenghtening of binocular housing, and would require keeping the front prism aperture at the same position with respect to the objective. If housing size were allowed to vary, it would require no change to the focal length and the binocular housing could be allowed to get shorter.
Every minor adjustment in prism aperture that requires an adjustment in focal lenght requires another small iterative correction, or any change in prism size requires another change in housing size, but either way, obviously Nikon, Fujinon and a few others have got it all figured out.
Now I won't doubt that you can claim the error is not in the prism it is in the objective focal length. Since it is aperture that we use as a major criteria to buy, I choose to state that for the given aperture of the binocular, the rest of the system is underdesigned. Since you too, with a statement that it is not the prism, it is the focal lenght of the objective, would also require a redesign of the entire binocular housing, you really can't get away from the fact that the housing and the placement of its parts is imperfect.
In your example, where you stated that all would be well if the objective were f/4 rather than f/3.5, for a 70mm binocular to accomodate that with no change in prisms, you would be required to lengthen the binocular housing by 35mm, certainly no small mechanical redesign. You simply choose to redesign around the current prism, and I would choose to redesign around the current objective and change the housing by a tiny amount.
edz
-------------------- Teach a kid something today. The feeling you'll get is one of life's greatest rewards.
member#21
|
GlennLeDrew
sage
Reged: 06/18/08
Posts: 467
Loc: Ottawa, Ontario, Canada
|
|
Edz: you wrote, "Since it is aperture that we use as a major criteria to buy, I choose to state that for the given aperture of the binocular, the rest of the system is underdesigned."
Then how is it that the very same prisms are often used in binos from 35mm up to 100mm? Again, if the f/ratios of their objectives were to be just that little bit greater (f/4), all would be well. That's what I was stressing.
The issue of prism size is important in one area only: that it allow a reasonably illuminated edge-of-field for the chosen ocular. A big field stop diameter, as for wide angle designs, requires a commensurately large prism. But the prism aperture can actually be a bit smaller than the field stop diameter and still deliver an image that has no detectable vignetting.
And the larger the space between the prism's rear opening and the field stop, the greater can be the difference between prism and field stop apertures... within the limits imposed by the objective focal ratio, of course!
-------------------- Home-made 11X50 right angle bino, 8.1 deg. FOV
Modified 26X100 bino, 3.5 deg. FOV
Mediocre minds discuss people. Good minds discuss events. Great minds discuss ideas.
|
EdZ
Professor EdZ
   
Reged: 02/15/02
Posts: 12341
Loc: Cumberland, R I , USA42N71.4W
|
|
Quote:
Again, if the f/ratios of their objectives were to be just that little bit greater (f/4), all would be well.
You haven't gotten around the fact that in order for that to work, the binocular housing would need to be completely redesigned. That's a whopping extra 2 inches of length for a 100mm binoccular.
Actually it's not so much prism size as it is the issue of placement of prism aperture with relation to the objective. And either one of those requires a larger housing. However, the larger prism would result in the smallest increase in binocular length.
edz
-------------------- Teach a kid something today. The feeling you'll get is one of life's greatest rewards.
member#21
|
pcad
Pooh-Bah
   
Reged: 01/17/05
Posts: 1464
Loc: Connecticut
|
|
Hi Glenn,
I agree with you about how the focal ratio of the objectives are related to the glass used in the prisms. Makes one wonder why companies bother with BaK4 in binoculars with "slow" objectives like f/6.
Peter
-------------------- Peter
Telescopes 25 - 318 mm
Binoculars 15 - 88 mm
Microscope 50x - 1000x
|
EdZ
Professor EdZ
   
Reged: 02/15/02
Posts: 12341
Loc: Cumberland, R I , USA42N71.4W
|
|
Actually Peter, there aren't too many f/6 binoculars. The BT100 is f/6.2. Vixen makes an f/11. There's a few others. Of all the fixed power binoculars I've tested, none were more than f/5.1. Most were between f/3.8 and f/4.2. All the 80s and 100s were between f/4.4 and f/5.1.
For a 100mm binocular to be f/3.5 it would have to be about 12" or shorter. For it to be f/6 it would need to be about 21 inches long, (both not including shields).
edz
-------------------- Teach a kid something today. The feeling you'll get is one of life's greatest rewards.
member#21
|
GlennLeDrew
sage
Reged: 06/18/08
Posts: 467
Loc: Ottawa, Ontario, Canada
|
|
Edz, I've modified, designed and built binos for years, and am quite aware of the issues regarding physical dimensions based on optical components used. I just didn't think it was germain to this discussion about the optics, it being obvious that a designer would lay out the instrument accordingly before manufacturing components.
But as we all know, the manufacturer strives for compactness and light weight, and hence chooses the shortest f/ratio objectives he can get away with in order to address those more overriding concerns.
Again, he could stay with the same prism set, but choose a slightly longer objective (and suitable ocular for the desired magnification) and then design the body accordingly. The result would be an instrument which works at full aperture. But as it is he offers instead an instrument which might as well have a smaller diameter objective and hence slightly lighter weight. But then, of course, he could not boast of having that larger, but partly useless, objective.
-------------------- Home-made 11X50 right angle bino, 8.1 deg. FOV
Modified 26X100 bino, 3.5 deg. FOV
Mediocre minds discuss people. Good minds discuss events. Great minds discuss ideas.
|
ronharper
scholastic sledgehammer
Reged: 02/14/06
Posts: 964
|
|
Some of my favorite views are: 7x50 in bright daylight, when my pupils are 2 or 3mm 16x70 stopped down to 46mm for double stars (Hooray for wasted glass around the edge of the objective! I positively do boast about it!) 16x70 at full aperture for faint fuzzies (Hooray for compact, lightweight, large and overly fast optics! Stopping down easier than stretching out.)
By the way, Glenn, do you think BK7 is significantly more transparent and homogeneous than Bak4? And, is it really much cheaper? Ron
|
GlennLeDrew
sage
Reged: 06/18/08
Posts: 467
Loc: Ottawa, Ontario, Canada
|
|
Ron, The only thing I really know about BK-7 and Bak-4 is their differing indices of refraction. I'd venture to guess that the denser glass is softer and hence a bit faster to work. I'd also venture to guess that the denser glass is a bit less transparent, but if so probably not enough to get wrapped around the axle over. As to homogeneity, that most likely depends on the care taken in annealing.
The only real advantage of BaK-4 (in prisms) that I'm aware of, from an end user's point of view, is that its ability to accommodate more steeply converging light cones before light loss is incurred.
-------------------- Home-made 11X50 right angle bino, 8.1 deg. FOV
Modified 26X100 bino, 3.5 deg. FOV
Mediocre minds discuss people. Good minds discuss events. Great minds discuss ideas.
|
pcad
Pooh-Bah
   
Reged: 01/17/05
Posts: 1464
Loc: Connecticut
|
|
Hi Ed,
My 22x70 is definately f/6. The Miyauchi Saturns also have slow f/#'s. I agree that nearly all handheld binos are faster for the reasons above. My question was why do some of these rather expensive, high quality astro binoculars continue to use BaK4 prisms when they aren't necessary. Marketing issues perhaps?
Peter
-------------------- Peter
Telescopes 25 - 318 mm
Binoculars 15 - 88 mm
Microscope 50x - 1000x
|
EdZ
Professor EdZ
   
Reged: 02/15/02
Posts: 12341
Loc: Cumberland, R I , USA42N71.4W
|
|
Peter, about how long are the 22x70s, from front of objective to back of eyepieces? Approx 15" makes them approx f/6. IIR, they seemed quite a bit longer than that, so may be closer to f/7
edz
-------------------- Teach a kid something today. The feeling you'll get is one of life's greatest rewards.
member#21
|
pcad
Pooh-Bah
   
Reged: 01/17/05
Posts: 1464
Loc: Connecticut
|
|
I'll check tonight when I get home. William Yang said they are identical to his new Megrez 72mm ED scope and it's listed as f/6.
Peter
-------------------- Peter
Telescopes 25 - 318 mm
Binoculars 15 - 88 mm
Microscope 50x - 1000x
Edited by pcad (06/19/08 10:49 AM)
|
EdZ
Professor EdZ
   
Reged: 02/15/02
Posts: 12341
Loc: Cumberland, R I , USA42N71.4W
|
|
Quote:
Edz,
I've modified, designed and built binos for years, and am quite aware of the issues regarding physical dimensions based on optical components used. I just didn't think it was germain to this discussion about the optics, it being obvious that a designer would lay out the instrument accordingly before manufacturing components.
But as we all know, the manufacturer strives for compactness and light weight, and hence chooses the shortest f/ratio objectives he can get away with in order to address those more overriding concerns.
Again, he could stay with the same prism set, but choose a slightly longer objective (and suitable ocular for the desired magnification) and then design the body accordingly. The result would be an instrument which works at full aperture. But as it is he offers instead an instrument which might as well have a smaller diameter objective and hence slightly lighter weight. But then, of course, he could not boast of having that larger, but partly useless, objective.
Considering that,
keeping prisms the same would require changing every other single component in the binocular, the objectives, the housings and the eyepieces, and
that the prisms are quite likely the most expesive pieces of glass in the binocular, and that the manufacturer may stand to save the most by using smaller prisms,
I would say that calling prisms too small a myth, is an incorrect assessment.
As I said before, certainly you could solve the same problem by your method, but I doubt very much changing all the components and making the binocular an inch (for 60-70mm) or two inches (for 100mm) longer to achieve that result would be the realistic approach to solving the problem.
Assuming the approach that it is undersized prisms that need to be replaced, in fact, for smaller binoculars near 50mm, it may be possible to fit the larger prisms without even changing the size of the housing and perhaps only changing the eyepiece focusing assembly by actually making it 5-6mm shorter.
edz
-------------------- Teach a kid something today. The feeling you'll get is one of life's greatest rewards.
member#21
|
SaberScorpX
Post Laureate
   
Reged: 01/12/05
Posts: 4116
Loc: illinois, usa
|
|
Welcome Glenn-
Thanks for the refreshing perspective and expertise. Looking forward to your future contributions.
Cheers, Stephen Saber
|
GlennLeDrew
sage
Reged: 06/18/08
Posts: 467
Loc: Ottawa, Ontario, Canada
|
|
When the workload around here gets reasonable and I can devote an evening to it, I'll make some annotated screenies from my bino planning software (REALLY crude, in QBASIC). Perhaps I'll show setups with a particular objective f.l. and f/ratio coupled with a fixed field stop diameter, changing only the prism parameters.
If you wish to, you can still download from S&T this software to play with yourself. It was submitted along with the article I wrote about my home-made right angle binos for the July 2002 issue.
Get it here:
http://www.skyandtelescope.com/resources/software/3304921.html?c=y&page=3
-------------------- Home-made 11X50 right angle bino, 8.1 deg. FOV
Modified 26X100 bino, 3.5 deg. FOV
Mediocre minds discuss people. Good minds discuss events. Great minds discuss ideas.
|
Shortymx
super member
Reged: 10/31/07
Posts: 127
Loc: 20º 41' N 101º 24' W
|
|
Thanks, Glenn for your comments. It's refreshing to reaed something that dosen't just relay the claims of manufactures. It seems, today, that all you read is FLP-53, and BAK-4. I'm sure these are worthwile in many appliations, but not necessary in all. I'm afraid amature astronomy is getting caught up in the advertising gimmic enviorenment that so many consumer products are in.
-------------------- Shortymx
|
pcad
Pooh-Bah
   
Reged: 01/17/05
Posts: 1464
Loc: Connecticut
|
|
Ed,
Measured the length of the 22x70's to be 15 7/8" with the eye cups rolled down. The eye lens was recessed ~3/8" and the edge of the objective was recessed 7/8" with the dew shields retracted. That leaves ~ 14 5/8" from end to end. With the eye cups and dew shields extended the total length is 18 3/4". The focal length should be ~ 16.5". I'm guessing 20mm FL (0.8")eyepieces. Do you think the prisms path can account for the ~ 2.7" difference? Still seems pretty close to f/6.
Peter
-------------------- Peter
Telescopes 25 - 318 mm
Binoculars 15 - 88 mm
Microscope 50x - 1000x
|
sftonkin
sage
   
Reged: 02/25/04
Posts: 395
Loc: Kent, UK
|
|
I must remember to tell my doctor that it is a myth that I have too much girth for my height: it is (by the OP's logic) merely that I am too short for my girth.
-------------------- Stephen
Hindsight: The only truly diffraction-limited system
|
GlennLeDrew
sage
Reged: 06/18/08
Posts: 467
Loc: Ottawa, Ontario, Canada
|
|
As promised, I did up an illustration using my (crude) Binoplan software, with fiddling done in Photoshop.
A commercial 20X100 available 20 or more years ago was measured out. It had rather fast f/3.5 objectives which made it impossible for the prism front aperture to fully take in even the on-axis cone of light, much less any off-axis light. The result is an effective aperture of about 87mm.
The prisms were standard 1-inch jobs, with the front and rear apertures being the same. One might naively think that using larger prisms would help, and that using smaller prisms would be sheer folly. To show that there would be no real difference, either way, I checked to see how another common prism size would perform, namely those with smaller 20mm clear apertures.
In both analyses, I chose typical prism-to-field stop distances, they being equal to the prism's clear aperture. You may be surprised to learn that as laid out, the smaller prisms actually resulted in a slight *gain* in effective aperture! But they do cause somewhat worse vignetting, although not enough for even the most discriminating observer to discern.
In both bino configurations, getting the field stop as close as possible to the rear prism aperture will result in a gain in effective aperture. This is why many binos do indeed have their focal surfaces (hence field stops) in about as far as physically possible. But unfortunately, with objectives faster than about f/4 there will almost always be a reduction of aperture.
Here's the link to the image...
http://www.cloudynights.com/photopost/showphoto.php?photo=15120&password=&sort=7&thecat=500
EDIT:
I should add that the software does take into account the effects of refraction within the prisms. For an index of refraction of 1.5, the focus is moved back by an amount equal to 1/3 the prism's physical length. So for standard 1" Porros, the glass length is 110mm, and therefore the focal position is moved back by about 37mm.
The two layouts shown in the illustration are scaled a tiny bit differently, in that the objective-to-focus distances are drawn as exactly the same length. In reality, the lower layout using the smaller 20mm prisms should be a little bit shorter (and smaller overall) because the focus with them is moved back 29mm, 8mm less than with the larger prism set. But this is small enough to ignore here.
-------------------- Home-made 11X50 right angle bino, 8.1 deg. FOV
Modified 26X100 bino, 3.5 deg. FOV
Mediocre minds discuss people. Good minds discuss events. Great minds discuss ideas.
Edited by GlennLeDrew (06/22/08 02:32 AM)
|
pcad
Pooh-Bah
   
Reged: 01/17/05
Posts: 1464
Loc: Connecticut
|
|
Hi Glenn,
Nice illustration, thanks.
Several times you've mentioned "standard Porro prisms" as opposed to "asymetric Porro prisms". Is the use of asymetric prisms an attempt to correct the problem of "too small prisms"? I imagine that using progressively smaller prisms would reduce the overall prism path length and move the prisms further away from the objectives. Am I on the right track, or am I missing something?
Peter
-------------------- Peter
Telescopes 25 - 318 mm
Binoculars 15 - 88 mm
Microscope 50x - 1000x
|
KennyJ
   
Reged: 04/27/03
Posts: 10029
Loc: Lancashire UK
|
|
< Am I on the right track, or am I missing something? >
Peter , I'm not sure if YOU are missing anything , but I'm beginning to think that my eyes must be ! :-)
Having earlier today congratulated Glenn on his ability to explain his thoughts without diagrams or sketches , what do I find in his next post -- ?
-- but a link to illustrations ! :-)
Kenny
-------------------- Two eyes and a preference to use both
Zeiss 7 x 42 BGAT
Captain's Helmsman 7 x 50
Nikon 10 x 42 Superior E
Swift Audubon Kestrel 10 x 50
Helios 15 x 70 Observation
Strathspey 20 x 90
Televue 76 APO
Zeiss 85 Diascope
Helios 102 f5 refractor
Various eyepieces barlows tripods mounts etc.
Panasonic Lumix DMC - TZ5 digital camera
|
GlennLeDrew
sage
Reged: 06/18/08
Posts: 467
Loc: Ottawa, Ontario, Canada
|
|
Quote:
Hi Glenn,
Nice illustration, thanks.
Several times you've mentioned "standard Porro prisms" as opposed to "asymetric Porro prisms". Is the use of asymetric prisms an attempt to correct the problem of "too small prisms"? I imagine that using progressively smaller prisms would reduce the overall prism path length and move the prisms further away from the objectives. Am I on the right track, or am I missing something?
Peter
I use the term "standard" for those systems that have equal size apertures at the entrance and exit of the prism train. More precisely, I mean that both prisms are identical in size (some binoculars may actually have the rear aperture stopped down a bit in an attempt to provide a baffle to block non image-forming light scattered off the prism walls.
Asymmetric prisms are indeed used so as to be able to accommodate at least the on-axis portion of the steeply converging light bundle. All that's needed is to make the front aperture, and its immediately following 45 deg. reflecting surface, a bit larger. In the illustration you can imagine how, if the front aperture was made just about 10-11% larger (with a concomitant but sufficiently small increase in total prism system length), the full light cone could make it into the prism.
My Celestron 25s100s use just such an asymmetric prism arrangement, and in spite of the ~f/3.85 objectives, the instrument works at full aperture. If they didn't do this, I'd never have bought them.
-------------------- Home-made 11X50 right angle bino, 8.1 deg. FOV
Modified 26X100 bino, 3.5 deg. FOV
Mediocre minds discuss people. Good minds discuss events. Great minds discuss ideas.
|
EdZ
Professor EdZ
   
Reged: 02/15/02
Posts: 12341
Loc: Cumberland, R I , USA42N71.4W
|
|
Quote:
Ed,
Measured the length of the 22x70's to be 15 7/8" with the eye cups rolled down. The eye lens was recessed ~3/8" and the edge of the objective was recessed 7/8" with the dew shields retracted. That leaves ~ 14 5/8" from end to end. With the eye cups and dew shields extended the total length is 18 3/4". The focal length should be ~ 16.5". I'm guessing 20mm FL (0.8")eyepieces.
Do you think the prisms path can account for the ~ 2.7" difference? Still seems pretty close to f/6.
Peter
typical prism light path can range from about 80mm for very small prisms up to about 120-130mm for very large prisms. Focal length thru the prisms is light path divided by index of refraction, so the prisms can reduce the overall length by 80/1.56=~50mm to about 130/1.56=~80mm. F/6 works out well.
-------------------- Teach a kid something today. The feeling you'll get is one of life's greatest rewards.
member#21
|
|
5 registered and 10 anonymous users are browsing this forum.
Moderator: EdZ
Print Thread
|
Forum Permissions
You cannot start new topics
You cannot reply to topics
HTML is disabled
UBBCode is enabled
|
Thread views: 1029
|
|
|
|
|
|
|