More info & 3D Models on www.thepulsar.b... This video introduces optical design and optical aberrations. We also assemble a custom 5x microscopy objective that has diffraction-limited performances from Thorlabs stock elements.
I teach this at a university and have designed optical system and lens combinations. Everything said here is very accurate. Thanks for such a great explanation.
Thanks for the great video! I was searching to understand photography lenses a bit better (elements and aberrations) and this was very very helpful, clearly presented and well explained, thank you!
Spherical aberration comes from the difference between a parabola, which is the correct shape, and a spherical lens which is much easier to manufacture. Spheres are so much easier to make that we just accept the aberration and compensate for it instead of making the right shape to begin with. Right?
That's true for mirror where you get one (and exactly one) stigmatic point but for refractive lenses the aberrations come from the non-linearity of the sine law. Small angles therefore lead to smaller deviations to the law and less aberrations.
Thanks for summarizing all the major types of aberrations! This helps me know what to look out for as I design my own lens. Do you have any resources on wide angle camera lenses you know of?
Not specific to wide angle lenses unfortunately; kingslake is probably the best book I can recommend on camera lens design that is more than just copies of prescription data from patents
Thks for your feedback :) I have many pending videos about optics put unfortunately so little time to make them! I have some days off coming soon, I'll do my best to make a video at that moment :)
ty for the video, it was nice and educational. Just one point, you have beautiful accent, which is fine to ear, but some words were hard to understand, where the accent went too rich, to to clear hard words in future.
If round glass bends light why can’t you just use a flat square piece of glass at the front to start with? The sensor is also flat and square? Would that help make lenses smaller?
Do you give private guidance? Paid of course. I need to select a lens for 3D reconstruction in an artificial vision system using "Depth from Focus", and your video helped me a lot. I learned more with this video than reading manuals from Edmund Optics, Keyence, Dr. Schenk and self-called specialists. It's excellent video to understand optical design.
No paid guidance; only free help :D within the (limited) time I can allocate every week. You can use the contact form on the OpenRAMAN (dot org) website to reach to me. Let's check together if I can help you with your problem/project.
The best book I can recommend for beginners is "optics" by Eugene Hecht. It's the classical textbook that covers (almost) all fields of optics and that is used as teaching material as university. For optical setup themselves, resources are scarces but you can start with those of my website (free)
Excellent initiative and good luck :) don't forget to check on the website (thepulsar) and particularly the DevOptical posts as they're rich in information
Thank you for this. I am a daguerreotype photographer and I have an optical problem. I have a camera built in 1840. The lens system is very primitive. There are four Plano-convex elements arranged in a tube in two cells. All the lenses have wooden spacers. I believe they have been removed and replaced incorrectly over the years as the camera exhibits severe spherical aberration. I am trying to sort this out. Would you be willing to help. I can be reached through my centurydarkroom website contact page.
I don't understand the difference between oblique astigmatism and coma. They seem to both be caused by the same pheonomenon (off-axis rays impinging on a lens), so can one have one without the other?
They are indeed caused by the same root cause (off axis rays) and their nature is only revealed when you perform a power series expansion of the raytracing formula. Think of this as a formula y=ax²+bx+c; for a given x you have only 1 y value but it's a mix of three bases (x², x, 1) which have different effects and can be studied separately
Each of these bases also have different properties, e.g. x² is a symmetric function while x is asymmetric. Also, on a range of different x values, you can only compensate a x² by another x² and a x by another x function (i.e. you cannot compensate x² by a x function -- at least not for a range of x values)
You can do the same observations on aberration: some are symmetric around the pupil and other are asymmetric. Astigmatism can only be compensated by another source of Astigmatism, coma by coma, spherical by spherical, petzval by petzval etc.
Thank you, all of this is very helpful. What you are describing are the different Zernike polynomials, right? But so, if the cause of coma and astigmatism is indeed the same, when I send off-axis rays on a lens, will I always have both aberrations at the same time? If I look on a tranverse plane, can I always see some coma, and if I look along the optical axis, can I always see some oblique astigmatism?
Some ray positions can cancel some of the aberrations and not others so you can have a surface with astigmatism but no coma. Zernike polynomials are (to my knowledge) essentially used in ophthalmology but more theoretical analysis of aberrations relies on slightly different bases (power series). For instance, zernike doesn't have the concept of petzval curvature or distortion
Before the pandemic Thorlabs had some nice thick nitrile gloves. At the office we buy from VWR. Also, you can impregnate a fabric with alcohol and rub your hands with it to remove any greases left on the gloves. Do not rub directly your hands with alcohol as it is both ineffective and dangerous (I spilled i-PrOH in my eyes twice doing that... even though I had glasses). Avoid touching optical surfaces as much as possible, even when wearing gloves; especially for delicate items like mirrors and gratings
Thank you very much for this fantastic video. I would like to ask one question. By looking at your website and using the parameters, I calculated the magnification of this optics system is about 2. so1: 6mm, f1: -20mm, f2: 35 mm, d: 15 mm Something wrong?
I re-did the maths quickly and 50+50 you get around 25.5; with the -20 & 15mm it goes to 50-55. Probably with the thicknesses of the lenses themselves the distance increases a bit which reduces the focal length to ~35-40 mm
@@thepulsar Thanks a lot for your kind and helpful reply. Yes, as you said, due to the distance between the lenses becoming longer (19-20 mm) so the effective f can be gotten as 35mm. I really appreciate your responses. When you come to Japan again, please don't hesitate to contact me😊 I will guide you
It's a combination of Adobe Illustrator and After Effects :) to get accurate rays I used the raytracing equations (website) and exported in vector format .svg using custom C++ software.
Standard for the industry are OpticStudio and CODE V but it's a rip-off and OpticStudio recently went from 5k$ for a perpetual license to 2.5k$/yr/user! To start, I would recommend the EDU version of OSLO. It is limited to 10 surfaces but it is already enough to learn the basics :)
Bonjour Vidéo et démonstration pratique très ludique ! J'ai pour projet de réaliser des loupes binoculaire grossissantes pour du travail manuel de precision. Je réalise pour cela des simulations sur le logiciel qioptic. Avez vous des sites de fournisseur de lentilles et de prisme ( à des prix resonables) à me conseiller ? Merci à vous!
Thorlabs et Edmund Optics en tête. En Chine il y'a FOCTEK avec qui je travaille régulièrement mais attention aux délais et aux frais d'importation. Ne prennez que des produits sur étagères car les lentilles custom sont impayables à moins d'acheter des volumes importants.
I've never used it :-/ at the office we use Zemax but its it's pretty expensive. There is also OSLO where you can download a demo version limited to 10 surfaces which is better than nothing.
Ideally you would optimize the radii and the thickness by gradient descent but with stock elements you have no other choice than substituting lens and check how the various combinations perform. Of course you would always apply the highlighted design rules such as power spreading over two lenses etc
@@thepulsar What do you think about next? You optimize and after that select only stock lenses with "near" characteristics to those coming from this optimization. After this, re-optimize again but maintaining fixed the parameters of the selected lenses but other lenses with variable parameters and also distances between elements... ¿Could this help?
@@guillermobaldwinolguin1697 this is the recommended procedure when I attended my opt design classes but the results are (most of the time) not that great. On the other hand, some combinations of lenses seems to be "magic" and give really good results. 3-4 years ago I started making an open source program to test the combinations but it's currently on pause. I would like to resume it once I have progressed enough on my OpenRAMAN.org project
It's special nitrile gloves with skin touch; once you try them it's impossible to work without after :) if you pull too hard on them they just tear appart and they prevent small cuts while giving you a good grip on slippery parts
