Field curvature and relevance of IPD setting (1 Viewer)

This has been on my ToDo list for a while and this is the draft that my thoughts have resulted in.

At times we encounter binoculars that behave like IPD setting is critical. The reasons may vary, but I would like to point out how the effect of inaccurate IPD setting affects a binocular with curvature of field.

The coloured "dartboards" show the wavefront of the beam pencil through the binocular, and simply show that the focus increasingly differs with the transverse distance from the center of the beam.
The thin black vertical lines mark the user's IPD.

Fig. 1 a shows the ideal situation with IPD correctly set and the binoculars centered in front of the eyes.
Fig. 1 b shows a horisontal displacement of the instrument but with correct IPD setting maintained. The eyes look through corresponding dioptrical parts of the beams.

Fig. 2 a shows an IPD setting that is too wide and one eye is centered.
The eyes look through dioptrically non-corresponding parts of the beams.
If the diopter is changed rather than the IPD setting, a temporary improvement may seem to occur but it will soon be obvious that the real problem is not adressed.

Fig. 2 b shows an IPD setting that is still too wide but the horisontal position change allows for dioptrical equilibrium - as long as the observer strictly looks straight forward.

Fig. 3 a is basically the same as 2 a but the IPD setting is too narrow and only one eye is centered behind the ocular.

Finally, fig. 3 b is the same IPD setting but with the horisontal position adjusted to find dioptrical equilibrium.

........|:S|

Considering the importance of a correct diopter setting and a correct IPD setting, adding the occasional need to adjust the IPD for nearer distances, which in turn is a slightly constrained way to handle convergence and parallax issues compared to using actual optical axis convergence by mechanical means, then the common need for different diopter settings for different distances, everything suggests that reasonably flat fields should be preferred.

//L

It should also be pointed out that while incorrect IPD setting can appear to deliver dioptrical equilibrium without any change of diopter setting, oblique gaze directions will unmercifully reveal that there is something wrong.
In theory, flat field binoculars should appear less sensitive with eye placement and IPD setting. Binoculars with field curvature should require a more careful IPD setting and then work pretty well.

Binoculars with a flat field should be easy to see incorrect diopter settings with because the focus discrepancy is more uniformly distributed over the FOV, and binoculars with a curved FOV and incorrect diopter setting might be tricky to find the ideal settings with.

Another thing came to my mind when looking at the sketch, namely that an incorrect IPD setting with binoculars with field curvature will produce different dioptrical discrepancies with different gaze directions. In particular changes of vertical gaze direction will differ significantly from horisontal ditto.
This will create the impression of a FOV with optical inconsistencies.

//L

Higher performance eyepieces will usually give more uniform focus and flatter field,
and even a bit more depth of field. Sometimes it's a trade-off, though. The 7x25
8-deg Bushnell Customs, for example. The curvature makes me dizzy when scanning,
but the sharpness is very good across the fov. A Plossl with an extra field lens
will usually give you flatness and a very wide field, and also be very forgiving with IPD
settings. 7x35 Customs, or most old Binolux, will be very forgiving, but they
have more elements in them.
The IPD sensitivity, I suspect, runs along with the 'blackout' or 'squint' issue.
The curvature could be a hint that the design limits are close to being exceeded
for a given number of elements. A little further, and you are into spherical
abberation on at least one element, and the image can't even form properly.

To boil the theory down, I think that a wide and very flat field is the product of
a higher element count design that also tames the other related demons.

looksharp65 said:

It should also be pointed out that while incorrect IPD setting can appear to deliver dioptrical equilibrium without any change of diopter setting, oblique gaze directions will unmercifully reveal that there is something wrong.
In theory, flat field binoculars should appear less sensitive with eye placement and IPD setting. Binoculars with field curvature should require a more careful IPD setting and then work pretty well.

Binoculars with a flat field should be easy to see incorrect diopter settings with because the focus discrepancy is more uniformly distributed over the FOV, and binoculars with a curved FOV and incorrect diopter setting might be tricky to find the ideal settings with.

Another thing came to my mind when looking at the sketch, namely that an incorrect IPD setting with binoculars with field curvature will produce different dioptrical discrepancies with different gaze directions. In particular changes of vertical gaze direction will differ significantly from horisontal ditto.
This will create the impression of a FOV with optical inconsistencies.

//L

Click to expand...

Hi Lars,

My take on the visual effects off-axis misalignment is that it is extremely complicated, and usually avoided except for those who inhabit visual science laboratories ("mad scientists").

I've attached a fairly comprehensive article by Bradley and Thibos that you (and others) might like reading. All I can say is, get 'em lined up and keep 'em that way.

Ed

PS. I might add that your notion about field curvature might be valid, but not easily separated, and possibly masked, by the effects of other aberrations, — in particular, CA and astigmatism. Field curvature control involves a trade-off with other aberrations.

Thanks Ed, that is a really interesting study and I'll try to read it carefully as soon as I have the time. However, I have one initial objection. The optical and the perceived quality of a virtual image formed by an afocal system will decrease if the instrument's optical axis is not aligned with the eye's optical axis.
But some part of the image will still hit the fovea and that is where the sharp vision is formed if there is any focused optical image.
I'll have to read more, but my initial impression is that the authors confused the image degradation as mentioned above with the eye's poor optical and neuroophtalmic performance in the periphery, that is outside of the macula lutea.
But I'm probably wrong there, not having read and comprehensed it yet.

In any case, I think my sketch and the posts are valid provided that the displacement is moderate so that other aberrations won't mask these effects.
In fact, moderately large oblique gaze directions through binoculars are not so unusual.

My idea was not to introduce complicated optical calculations, only to illuminate how field curvature interferes with IPD and diopter settings and more so with gaze directions that are not exactly parallell with the optical axes (axises?)
This is more of a draft, and it is more difficult to estimate how pronounced this effect could be in reality and whether or not it has any practical relevance for ordinary users.
But I do believe that the added faults of incorrect IPD setting, minor collimation issues, field curvature and diopter setting may account for some of the difficulties we can experience even with binoculars of fairly high quality.

//L

"My idea was not to introduce complicated optical calculations, only to illuminate how field curvature interferes with IPD . . ."

The focal plane is in a specific location that is not required to change as we move our heads or change IPD settings. The more field curvature, the more critical the placement.

Cheers,

Bill

WJC said:

The more field curvature, the more critical the placement.

Click to expand...

Agreed, and that is a logical consequence of what I wrote.
But it is also good to discuss and understand the reasons, namely that an incorrect IPD setting will affect the balance between the eyes, and more so when not looking straight forward, and even more when looking up or down.

So I'm thinking about how the binoculars would appear if there is a minor issue with the alignment of the optical axes, not really miscollimation but rather some lack of horisontal parallelism.

If the IPD is adjusted to deal with that, what happens with the right-left balance if there is curvature of field?

How does that relate to lack of full contentment and frequent diopter changes with otherwise fine binoculars?

I'm not sure that all brands and samples are made and QC-ed sufficiently to avoid this.

//L

"So I'm thinking about how the binoculars would appear if there is a minor issue with the alignment of the optical axes, not really miscollimation but rather some lack of horisontal parallelism."

There is a chance of mis-collimation of the elements in either or both telescopes. Then, there is mis-collimation between the optical axes of both telescopes without regard to the mechanical axis (hinge) leading to "conditional alignment" (SPIE 8491-14, 2012). THen there is mis-collimation WITH regard to the mechanical axis (true collimation). Then there is the placement of our eyes, in which we see the effects of axis deviation. And finally, few of us a perfectly collimated; my oldest son is only correctable to 7 degrees.

That's mostly for those who don't have enough to worry about. In the real world, most of that is fairly inconsequential.

I did some informal experiments and concluded that unwanted color fringing is actually quite sensitive to IPD setting. I got uptight about it, and put spacers on my binoculars so when squeezed down to the limit they were exactly at my parallel eye setting, which you can well believe I went to great lengths to determine. But that is a bad thing, because a person with narrower eyes than mine, which are rather wide at 67 mm, can't enjoy the view through my binocular.

I got over it. In fact, I got over any binocular adjustment based on objective measurement of eyes or eyesight. Things are always more complicated than any of my limited and simplistic concepts. If it doesn't look right, just adjust it a little. Sure there's a thousand things going on, but it all works out somehow. Must be a brain thing!

Now when I hand somebody my binocular I admonish them to squeeze it together until they can see well.

Ron

That's a very thoughtful analysis, looksharp65. I guess that's why I usually try to set my IPDs under low light conditions. It's easier to center the bin's exit pupils with my own pupils when my own pupils are closer in size to the bin's exit pupils. Under bright light conditions, when my own pupils can often be substantially smaller than the bin's exit pupils, a whole range of IPDs can appear to be "good enough".