Zemax Camera Lens Design

[Carlito Roby]

Theoptimization variables are just the distance from the opbject to the lens, and the EFL of the paraxial lens. In the merit function I target the marginal ray to a height of zero on the image surface, and the chief ray to a height of -7*.8 = -5.6. Hitting the optimize button gives:

Thank you sir it is very helpful, but I have one doubt sir why did you selected object size 14 mm as i can see it is showing 7mm a semi diameter of the object? And I want to keep object at a distance 14mm as working distance.

Thanks.

And more thing why we have used single paraxial lens sir? why not two? yeah you did mention about to replace paraxial lenses with real lenses. So EFL of collective should be 7.4 mm ryt? and both will be opposite to aperture stops?

Please check this files sir. I have design using two paraxial lenses and kept the object height 7 mm on both X and Y as a square of 7x7mm gives semi aperture diameter of 3.5 mm. As you suggested optimized for -3.5 . 0.8 = -2.8 and achieved?

Doubt ?

Zemax is a software program used for designing and simulating optical systems. It is widely used in the field of optics and photonics for designing and analyzing the performance of lenses, cameras, telescopes, microscopes, and other optical systems. With the software, the behavior of light interacting with various optical components can be modelled, and optical designs can be optimized for desired performance.

In 2011, Evergreen Pacific Partners merged Zemax Development Corp with Radiant Imaging to form Radiant Zemax. Under Evergreen, the Zemax software was re-architected under the .NET framework to provide an updated user interface and user experience. The rearchitected version was renamed OpticStudio and was launched in 2014. One of the key new features of OpticStudio was the inclusion of a comprehensive application programming interface (ZOS-API) to support software customization and automation from a variety of .NET languages (including MATLAB and Python).

Ansys Zemax OpticStudio is part of the Ansys Optics product collection, which also includes products from Ansys Lumerical and Ansys Speos. The Ansys Optics solution provides the capability of simulating the behavior and propagation of light through optical and optically enabled products from the nano- to the macro- level, allowing for integrated modeling and co-simulation that enables the accurate and robust design of such products. Ansys Optics allows optical products to be visualized before they exist.

Ansys Zemax OpticStudio is widely used in industries such as aerospace, defense, consumer electronics, medical devices, and more, where precise optical performance is critical. The software has evolved over the years with updates and improvements to keep up with the demands of the optical design community.

When I introduce this piece as 'a Layman's Guide' I refer, of course, to myself as the layman: for all I know, you have a PhD in astro-optical design and regularly grind your own elements. But for the rest of us, struggling with ever-increasing pixel counts and the need to understand our existing lenses whilst acquiring new ones wisely, field curvature and the impact it has on focus can be 'the last frontier' of learning and one of the most difficult to conquer: lens manufacturers and reviewers provide MTF data and information on distortion and chromatic aberration, sometimes even DOF tables but I have never seen data on field curvature. And yet it can have a significant effect on what is and what is not in focus in every frame you capture, with certain lenses. Understanding it can help you both visualise the field shape of an individual lens and adapt your focus technique accordingly.

A sign that we have graduated beyond Photography 101 is that we begin to develop a proper sense of depth of field*. We know that it is deeper with wider lenses, smaller apertures and greater subject distance and we know, with some experience, that we can 'place it' where we want it. We also pick up some nuggets about circles of confusion and viewing distances and so on. But as pixel counts rise, I know I am not the only person who has noticed 'shape of field' effects more often, and struggled to pin down both the causes and how best to manage the effects.

I am always jealous, when debating the potential merits of a lens online with other folk, to hear the suggestion that I should 'rent one and see.' The market for lens rentals is comparatively underdeveloped in the UK and I long for access to a company such as LensRentals, an Aladdin's Cave into which I can peer but not enter. And peer I do, especially at the excellent blog of Roger Cicala, the only man I have ever seen wearing a lens case as a hat.

Roger sits at the perfect intersection, for me, of theory and practice. He loves lenses, loves finding out what they are capable of, and has a profound technical understanding of both what to measure and how to measure it. I am very grateful and honoured that he has agreed to contribute to this article. So: I am going to write an explanation of my understanding of field curvature (won through sweat and a bit of layman's study) with a mixture of simple theory and practical examples and suggestions - and then Roger is going to tell us all where I am wrong and what other considerations we might take into account.

Imagine, for now, that we live in a two-dimensional world where the scenes we photograph have depth and width only. This will help my diagrams make more sense and will also facilitate an initial understanding of the subject for the uninitiated. I will add height, the third dimension, later.

Schoolboy geometrists quickly spot, however, that the edge of the planar subject is further away from the sensor than is the centre. Thus, if a lens focusses at the same distance across the frame, then the 'shape' of the in-focus field will naturally be curved. Theoretically, that curve would be a section of a circle with the sensor at its centre (I am taking some short cuts here to keep it simple). And that means that the edges of our planar target would fall outside of this arc and therefore, if shot with very thin DOF, be out of focus.

Lens designers have several competing and often mutually exclusive aims: they want large maximum apertures, low distortion, minimal chromatic aberrations, controlled vignetting and low astigmatism to name a few. And correcting for natural field curvature (literally bending nature) is to some extent incompatible with some of their other aims. So in practice, all modern digital camera and lens systems shoot somewhere between 2) and 3) above.

The result of this is that almost no lenses you will ever use have a perfectly flat field, and almost none have a perfect 'section of a circle' field. Instead, they come in varying shapes such as 'croissant', 'wave' or 'runway' (my terms) and, these shapes can direct their acute angle towards or away from (rearwards field curvature) the sensor.

Now let us add in, with the help of some diagrams, the idea of depth of field to the idea of 'shape' of field. Note that these shapes are simplified and idealised in order to illustrate certain types of curvature. In the real world the shapes are more complex.

In the above diagram, the field shape is an arc which is not a perfect section of a circle. In other words, the lens design has partially corrected the field curvature by flattening it. And, because we have stopped the lens down a little, there is now some depth to the field of focus too. In fact, there just enough correction and just enough depth to get the entire planar subject in focus. But if we place the exact point of focus slightly differently, we move the position of the field of focus relative to the subject plane and this happens: the edges are out of focus. Bear with me here: the exact shape of the field of focus shifts as you change your point of focus and that makes matters more complex but for now, it might help to ignore that factor!

Field shapes can also be 'wavy' as noted above and as the following two diagrams show, this wave-shape (again a result of a compromise between competing design aims) can either not matter or matter quite a bit depending on exactly where you place focus:

The second diagram (Fig.5) is a simplified explanation of what might be happening for some cases of 'mid-field weakness' in a lens: the 'placement of the field of focus' has moved closer to the sensor relative to the first diagram because the lens has been focussed to a slightly shorter distance. The subject on centre is still in very good focus but parts of the mid-field sections of the image plane are slipping in and out of the field of focus. I have made focus-bracketed series of shots which seem to indicate that this is happening.

Field shapes vary widely: quite possibly both wave shaped and curved, for example. One frequent field shape (see my review of the Nikkor 28mm F1.8G for a very graphic illustration of this) is the 'runway':

If you ever have images where a middle-distance planar subject is in good focus edge to edge, the distant parts of the frame are sharp on centre but OOF at the edges, you have a runway on your hands. It looks like this: excuse the odd tint! The yellow area is an approximate focus mask, focus is at the distance of the easel. Clicking on it will load a full-sized D800E version of the same file but without the tint and mask so you can look in more detail at the exact field of focus, which actually extends to the very far horizon in places.

Firstly, let's add the third dimension: height. In the above diagrams we have looked at a 2D cross section of reality but of course, the shapes we have been looking at need to be visualised in three dimensions. This is too complex a topic to illustrate without interrupting the larger flow of this piece and in real world use it is often enough to stick with the 2D visualisation - but if you want to go further, think of it like this: the example above shows a roughly runway (or triangular) shaped field of focus in 2D. in 3D this is more like a cone. Of course some of that cone-shaped field of focus is bisected by the ground itself so what you see on the ground is a triangular shape because (schoolboy geometrists again!) a triangle is one possible cross-section of a cone.

3a8082e126