Tuesday, October 25, 2011

Jim Klein's Optical Design Guide

October 25, 2011
Glendale, California


My name is Jim Klein. I have a B.A. in Astronomy-Physics from UCLA (1968) and an M.S. in Physics from CSULA (1973). I worked as an optical engineer, optical designer and optical analyst from 1977 to 2007 at Hughes Aircraft Co., Aerojet Electro Systems in AZUSA, CA, TRW in Redondo Beach, CA and finally Raytheon SAS in El Segundo, CA. Then I took the rest of my life off to write this guide and do other fun non-optical things.

I'm writing this blog to degrease myself (wash my soul) of my obsessive interest in how new optical systems are designed, how existing optical systems work and how software to perform these calculations are written. I hope what I write will be helpful and/or interesting. If I get lucky, some of what I write might even be entertaining. If anyone wants to send comments, send them to ecalculations@earthlink.net.

I will be using my FREE optical design and analysis program, KDP-2 (Klein's Design Program). It is available for free (that's why it's called FREE) to anyone who can connect to my website www.ecalculations.com . If you live in a repressive regime or live behind a firewall that disallows access to my website, overthrow your dictator or change jobs. I sell nothing here as I made a good (relatively speaking) living as an optical designer and analyst (at the cost of my immortal soul and 30 years of time) and now I would like to pass along some, if not all, of what I have learned to the next generation of optical designers, optical analysts and optical design program software developers.

As I said before, I am an optical designer. I spent 30 years designing and analyzing optical systems. I don't design the mechanical housings which hold the optical elements of optical systems. I don't design the anti-reflection coatings used to maximize the amount of light passing through optical systems and I don't build optical hardware. I will refer you to some elements of these topics but I am not an expert in these "Optical Engineering" areas. I will refer you to texts from which you can learn as much as you might want to learn about these topics but I will concentrate of the design and analysis of the lenses, mirrors and other optical elements which refract, reflect or diffract light.

That's it for tonight,

Jim Klein

Saturday, August 4, 2012

Get a copy of "Principles of Optics" by Jenkins and White. Read chapter 4 as many times as is necessary until you fully understand the concept of "pupil imagery". Failing to reach a complete understanding of pupils and their imagery, become a department manager or a division vice president or some other type of manager.

If you understand pupil imagery, remember that most everyone else is a complete idiot and does not understand it, so plan on spending at least 25% or your career explaining why pupil imagery is important to system engineers and project managers who think everything important can be done with Power Point.

If you are self employed, designate yourself "employee of the month", every month. Staples sells pre-printed award forms and you can have a neighbor sign their name as your boss if you are too embarrassed to do it yourself.

If you work for a big company, randomly and without getting caught, change the signs on the MEN's rooms to "EXECUTIVES and TECHNICAL STAFF". Change the sign's on the WOMEN's rooms to "CLERICAL and SUPPORT STAFF". You might want to do this only once per career and on the night before your final retirement party. I never quite had the courage or stupidity to take my own advice and do this.

That's it for tonight.

Jim Klein

Monday, August 4, 2012

What is "your CUSTOMER"?

Recognizing who or what the "CUSTOMER"is, is indeed one of the most important stages in the process of creating any useful optical system. The customer could be:

1. Yourself (if you are simply playing at creating some new fun or useful optical concept which popped into your head whilst jogging, fishing, dreaming or smoking that last bit of medical cannabis).

2. Someone besides yourself.

a. If you are self-employed this might be a person who calls you up with an idea for anything from a new binocular design to a faster that light, laser powered, warp drive for the next generation star ship.
b. If you are not self employed, your customer might be the person who arranges a meeting with you that runs through lunch and who assigns you to the next seemingly impossible design task intended to save your favorite political system from being over-run by the mindless hordes of barbarians who live in a neighboring country. He might just be the guy responsible for keeping you employed with a regular pay check. You want to make your customer happy within limits.

In any case, don't trust any new customer or their new and great idea until you have considered the following possible problems:

1. Is the customer insane? Look into their eyes. Do they have a crazed or dazed look about them? Are they wearing a uniform you are un-familiar with or last saw in "Plan 9 from Outer Space"? Do they have their heads wrapped in aluminum foil to keep the CIA from reading their thoughts? Do you get a creepy feeling that they are simply full of shit? Do they speak in complete sentences?

2. Is the customer even remotely able to understand the laws of physics which may or may not need be violated in order to achieve the final goals of the job? Violation of basic physical principles is not possible, even if a 4-star general gets up and tells everyone that money is no object, that he has a blank check from Congress and that failure means the certain fall of all of western civilization and the deflowering of the last remaining virgin in the realm.

3. Is the proposed use of the device you are about to design immoral? If you succeed in designing a fully functioning battery powered, hand held, 300 watt CO2 laser, will it be used to vaporize tumors and cure cancer or will it be mounted on a HUMVEE and vaporize people who don't believe in the same god your "fearless leader" believes in?

4. If you decide to not work for the "customer", will you starve to death, risk being put to death or worse, spend the rest of your life wondering if you did the right thing?

Final note about Customers

My first job at Hughes Aircraft Co. in 1977 was to make sure that the 16 experiment windows used on the Pioneer Venus decent probes did not blow out or otherwise fail as the probes descended into the atmosphere of Venus. The work was a success. One of the probes sent back data for over an hour after it had landed on the planet's surface, far exceeding operational expectations. This was the high point in my 30 year career. We had good managers for this project. Good science was done.

The low point was trying to undue the damage done by a "center manager" after he sold the idea to a an air force general that by using optical phase conjugation, he could bounce high energy laser beams off trash can lids and still maintain optical beam quality and destroy incoming USSR missiles thus saving western civilization from the evil red hordes. The center manager was later described by a friend of mine as a "surgically altered Ferengi".

That's all for tonight,

Jim Klein

Sunday, August 12, 2012

LESSON 3, Lens Units, Wavelength Units and the Object Distance.

Every optical design program I am familiar with allow the linear units to be used to be specified by the user. Being from the USA, I use "inches" as the default unit system but the user can specify mm, cm, or meters and the units set will be remembered when the optical system is saved or the program is stopped. Having units of "meters" available makes the modeling of very large space based optical systems easier, that is why I included "meters" as one of the four linear unit systems.

KDP2 and most other programs use "micron units" to specify the wavelengths used. KDP2 allows up to 10 wavelengths to be stored in any optical system at any one time.

Most commercial optical design programs (CODE-V, KDP2 etc.) have a way to specify the distance from the object being viewed (or the "source") to the optical system. In KDP2 and some other programs, this distance is input as the THICKNESS of surface number 0 (zero). Some other programs (like ZEMAX) allow surface number zero's thickness to be left 0.0 and the first surface to have a non-zero thickness will be used to represent the "object distance".

In my program, KDP2, if surface number 0 has a thickness of 0.0 it will be reset automatically to a value of 1.0D+20 lens units. That us usually a large enough distance to represent "infinity" for practical purposes. This methodology isn't superior or inferior to other ways to specify the object distance, it is just the way I learned to do it before I started KDP2 in 1987.

For optical systems such as microscope objectives, any small value larger than 0.0 may be used to represent "close" objects.

Lesson 4 will deal with the topic of "Optical Axis".

Before leaving LESSON 3, I would like to suggest that anyone interested in becoming an optical designer or optical analyst begin to collect as many optical text books as they possibly can. My first text was "Principles of Optics" by Jenkins and White which I used in 1966 as an Astronomy major at UCLA. It is still available as a paperback though there are thousands of old hardback copies available at minimal cost. Even copies dating from the 1950s are useful as long as you don't expect to see the word LASER in the index :-)

There is a copy of the Military Handbook of Optics (also called Mil Handbook 141) in PDF format available from my website at www.ecalculations.com. It is free to download and makes an excellent first text for those with limited funds.

That's all for tonight,

Jim Klein

Monday, August 13, 2012

LESSON 4, The Optical Axis of an optical system.

Lesson 4 was originally going to deal with the topic of "Fields of View". After thinking about that for a few minutes, it became obvious that before we talk about "Fields of View", we need to define a reference axis for the optical system. I will restrict this discussion to centered optical systems only and deal with the more complex definition of the "Optical Axis" of non-centered optical systems later in this guide.

The most common centered optical system I am familiar with is the 35 mm camera. In this system all of the individual lenses comprising the camera lens are circular in profile and are mounted in a cylindrical tube (the lens housing or lens barrel). If we now draw an imaginary straight line which starts at center of the object plane, passes undeviated through the center of the camera lens and terminates at the center of the film plane, that imaginary line will be defined as the "Optical Axis" of the system.

In the next lesson, we will use a simple schematic diagram of this 35 mm camera system to show the optical axis and define the concept of the "Reference Field of View" from which other Field of View positions may be defined.

People who will help me write the rest of this guide:

1. My Wife Sally who will proof read the text from time to time. She is not a technically trained person so if she can understand what I write, you will be able to do the same.

2. My alter-ego, Dr. Image Nius. He holds a B.S. an M.S. and a Ph. D in theoretical trans-warp Physics from the Max Platt Institute in some town in a European country that even Borat won't visit.

3. Anyone who wants to contribute both technical and non-technical input to this guide. All of these inputs should be emailed to me at my personal email account at: jameseklein@earthlink.net .

4. My 3 year old female Terrier "Belle" who keeps me from spending too much time writing this guide by bringing me squeaky toys to play with and dead rodents and other strange objects. She acts as my court jester and she is my friend. She reminds me of how little I really know about the universe.

Parting Note to Lesson 4.

Last evening I was watching the first episode in the second season of the AMC program "Hell on Wheels". Every few minutes they would remind viewers that "this is not available on Dish Network". I think AMC and Dish Network may have some kind of business dispute.

This Guide is, on the contrary, available to everyone. People that like me, people that don't like me and anyone in between. At one time I would have had a list of people who were not allowed to use this guide. Then I realized that life is too short to be that way.

Watch "Hell of Wheels" if you don't have Dish Network service. It is a good western if the language doesn't offend you too much.

That's all for tonight,

Jim Klein

Wednesday, August 22, 2012

LESSON 5 The Layout of a centered optical system.

The above drawing is my first using the drawing tool I have in PAGES on my iMac. As I get better with this tool or add actual drawings of real optical systems done with KDP-2, the art should get better.

As mentioned in Lesson 4, this is the graphical representation of a simple centered optical system comprising a 35 mm camera with a 50 mm focal length lens. As you can see, the Optical Axis starts at the center of the Object or Source Plane, goes through the center of the 50 mm EFL lens assembly and terminates at the center of the Image or Film Plane. Yes, I have a digital camera but I still keep a film camera purely for nostalgia purposes.

I'm rather happy with this drawing as I put it together without reading the instructions for the PAGES program.

I will end this Lesson with some practical advice. I hope it contains a little wisdom.

When presented with any engineering task by any customer, make absolutely certain that you and the customer are in complete alignment with respect to the technical and pro forma aspects of that task. Get the specifications in writing, even if the customer is your immediate supervisor. If a customer wants you to violate a fundamental law of Physics, always be professional when you explain that fact and make no references to the concept only working in a Science Fiction movie. Never hurt a customer's feelings. Remember, most of your customers have feelings or families or both and report to some kind of a boss who they too must make happy.

NEVER accept any task from any customer unless you are 100% certain you can complete the task for the $ amount and time agreed upon. If for any reason, whether it be technical or personal, you have doubts about your ability to perform up to or in excess of your customer's expectations, turn the job down. If you think you will be laid off if you say, "I don't think I am qualified to do this task", say it anyway. It is better to be open and above board with yourself and your customer than to deceive yourself and leave your customer dissatisfied and angry.

I followed this rule for 30 years and it worked every time with success. Then I got scared I would be laid off  before I was ready to retire as there was very little work available and I took on a task for which I was not qualified because of personal medical issues I was not in touch with and the job ended badly. Know when you are over your head. Know your limitations. Don't even sign up for a job that your abilities can't cash. If you get old and you don't feel well, don't make the mistake of thinking you are 35 when you are 65. It never ends well.

That's all for tonight,

Jim Klein

Thursday, September 6, 2012


OK, here is where it gets a little more complicated and where I introduce some terms and their definitions.

All real objects or sources in optical design have some real physical lateral extent. For example, a telephone pole, in the object plane and with its bottom located on the optical axis introduced in Lesson 6 will have its top located at the point on the object surface labeled "Object Center". When a ray starting at this "Object Center" passed through the camera lens, crosses the optical axis at the location of the lens's "Adjustable Iris" and then proceeds through the rest of the camera lens, terminating on the "Image or Film Plane", that ray and only that ray is always called the "CHIEF RAY".

Simple enough definition. Whenever the "CHIEF RAY" is referred to, it is this ray.

Now I will stop using capitals and quotes. We now have an object surface, and image surface, and optical axis, an Adjustable Iris and a chief ray. How that ray it traced from beginning to end needs not be worried about as the optical design program automatically determines how to point this ray so it travels along the chief ray path.

Thursday, September 6, 2012


Back in Lesson 1, reference was made to Chapter 4 of Jenkins and White. Similar chapters can be found in other basic optics text books. I found them all hard to follow when I first read them. They seemed like "magic". Most people who have a passing acquaintance with optics, either in Physics or Engineering, are literally driven to painful headaches trying to follow the convoluted descriptions. I will make it simple.

The VIRTUAL IMAGE is an image you can seen with your eyes. It can be a wall in front of you, your image in a mirror whilst brushing your teeth or your significant other, trying to get you away from the computer so they can have your attention.

The human eye, in the absence of any near or far sightedness is designed to focus light from distant of near objects on the retina of your eye. Muscles in the eye bend your lens to bring near objects into sharp focus. The ideal relaxed eye focuses on distant objects (say 100 ft or more distant from the eye). This means that the human eye can only focus parallel or slightly diverging cones of light to sharp focus on the retina. The eye can not focus converging cones of light on the retina. It just can't do it. There was never any evolutionary reason why we needed to do this, so nature being conservative, didn't design that ability into the eye. Cones of light which approach the eye in a converging cone represent REAL IMAGES. If you hold a magnifying glass near your eye and look through it toward a distant scene, you can't see that scene "in focus", but if you place a piece of frosted glass behind the magnifier and adjust the distance from the lens to the frosted glass, you can look at the image when it appears on the glass. This is how camera and camera obscures work.

Review: VIRTUAL IMAGES can be seen with the naked eye. REAL IMAGES can't be seen without the use of some projector screen type of surface.

Now, for the next part of this lesson you will need a camera lens. It would be best to use an interchangeable lens out of a manual 35 mm camera or a view camera or a lens taken out of a junk camera from the 50s to 80s. You can get one for essentially nothing. The lens has a front surface that faced the object and a rear surface that faced the film. It should have an adjustable iris diaphragm which can be opened and closed by moving the F/stop dial on the lens. Take this lens in your dominant hand (or not) and look into it with the front surface pointing toward your eye. As you adjust the iris diaphragm setting, you will be able to see the lens "opening up wider" or "closing down to a smaller diameter".

You are not seeing the iris diaphragm, you are seeing the VIRTUAL IMAGE of the iris diaphragm which is being formed by the glass lenses which lie between your eye and the actual iris diaphragm hardware. What you are looking at is called the ENTRANCE PUPIL of the camera lens.

Now do the same exercise except look through the rear surface of the lens and adjust the iris diaphragm setting again. You can again see the lens opening and closing. You are not seeing the iris diaphragm, you are seeing the VIRTUAL IMAGE of the iris diaphragm which is being formed by the glass lenses which lie between your eye and the actual iris diaphragm hardware. What you are now looking at is called the EXIT PUPIL of the camera lens.

The definitions of these PUPILS are also elucidated at length in the KDP-2 reference manual.

Don't worry how to calculate the positions and sizes of these pupils, optical design software does these calculations for you based on rays which are automatically traced through the optical system under study.

That friends and neighbors, in as small a nutshell as I can put it, are the definitions of an ENTRANCE PUPIL, and EXIT PUPIL and how VIRTUAL and REAL IMAGES relate to them.

As an ending note, the adjustable iris diaphragm is also called the APERTURE STOP of the camera lens. It is the aperture which acts as a "limiting" aperture or opening through which light may pass.

I hope this helped. For those of you who understand these topics, this was the simplistic way I condensed my understanding of pupils so I could explain their role in the operation of an optical system to a non-optically trained person.

I wish someone would send me as simple a description of how servos and stepper motors actually work because my understanding of them is limited and engineering reviews that talked about servos and stepper motors always put me to sleep. I never understood the terms and the words and it was like trying to under the meaning of ancient scripture in the original ancient language.

That's all for today.

Jim Klein


  1. As a self-employed consultant, and before that in 20 years building instruments and new photonic devices at IBM Research, I've run into a lot of customers who don't know what's possible, or don't know what they need.

    But that's part of what consultants are there for. In cases like that, I usually write contracts with Phase 1 being a photon budget or feasibility calculation to find out what's possible, and in Phase 2, the customer's technical people and I write a specification. The customer signs off on the spec if they want to proceed with the project. That has prevented a lot of misunderstandings, I expect.

    Another grubby issue for consultants is making sure one gets paid. I try very hard to avoid contracts where large payments are due on completion, because once the system is working, my leverage is much reduced.

    I've had a very few customers who have had serious payment issues in the past, e.g. going over 60 days on an invoice. With those folks, I work only on retainer, if at all, because I don't like pushing ropes. (They seem to be about 5%--I'm very fortunate in that almost all of my customers are straight shooters and generally Very Good People.)

    Since you're retired, and of a salty disposition, I hope for some very trenchant posts in future. Good on you for starting up again.


    Phil HObbs

  2. Me again.

    Re virtual and real images: you can too see a real image. Those little clamshell gizmos that make a penny seem to float in mid-air are creating a real image, and (surprise) it looks just like the penny in the bottom of the clamshell.

    I'd boil the distinction down to this: You can't put a white card at a virtual image, and you can put one at a real image.


    Phil Hobbs

  3. What kind of custom optics are mostly used to see REAL IMAGES normally?

  4. Jim, I have long enjoyed your contributions to the now-destroyed Usenet group sci.optics. Are you going to continue with this blog? I realize that it must weigh on your time; those who are still (gainfully?) employed seem to believe that those who are retired have an unlimited amount of free time on our hands - which is patently untrue. If anything, I am more busy now that ever before, except that I can now decide when to get out of bed (not at 4:30 am any more!).

    Anyway, at least I now have time to try to learn more about optics & geology, so I hope you can find time to contribute more ideas to this blog of yours.

    Finally, a couple of points (you do read these, right?): I looked for Principles of Optics by J&W, found only Fundamentals of Optics by them; I suppose that is the proper working title?
    Second: one (& only one) major virus software was freaked out about my downloaded copy of KDP-2 4th edition. I allowed it to be deleted and grabbed a copy of the ver 5. Have you heard of anyone having problems with infected versions - I'm sure I had gotten it directly from your site.

    Anyway, Cheers!
    Joe Gandalf

  5. Thanks for the software. You are great! I hope that I learn as much optics as possible.