Online Lens/Optics Simulator | Hacker News Reader

Online Lens/Optics Simulator

opticexplorer.sharedigm.com

154 pointsherodotus60 comments

So I need to find a way to draw a light circle (1.5-2m in diameter) on the floor. Not filled circle, but the circular line filled with light. Ideally by the small "gadget" that can be placed on the floor.

I first explored lasers solution, but seems like all options are just too big, complicated, expensive and regulated(only 5mW lasers can be used around people).

Then started looking into options of using cheap light torches with some custom mirrors. And that's where I discovered that software for simulating light, mirrors and lenses starts from 5000$ per license. Some tools posted here (including OP link) are interesting, but still I have no idea if it's even possible to do what I'm trying to do.

One close reference I found in this fantastic video on designing custom mirrors: https://www.youtube.com/watch?v=FzemhuI1knQ

Author also discovered the crazy prices of optics software and went with Blender + LuxRender.

If anyone have suggestions on how to find the solution for what I need, I'd appreciate that.

I'd bet there's a professional lighting product that'll do that. Something like the lights they use to project arrows on the floor at IKEA. I don't know enough about that domain to know what they're called, sorry.

There are, but they don't have constraints of being small, mobile and staying on the floor.

Here is one remotely similar stuff, but it's too bulky and more complex (I need static ring basically and gadget that can fit the backpack) https://www.youtube.com/watch?v=xXl3F5AoeQU

What you want is a stage/DJ light which takes gobos, and the right gobo. For example: https://www.fullcompass.com/prod/212208-rosco-81115-glass-go...

The light wouldn't be on the ground but above it, projecting down.

Edit: I should add, the gobo would need to be glass, as metal ones would have to interrupt the ring to support it. If the one linked above doesn't work for you, this place looks like it could custom make one to your spec and has a number of supported lamps under the "Gobo Size" drop-down. Explain what you need and they look like they can help you figure it out.

You may also be able find a good deal on lighting in one of the usual marketplaces, but just confirm the fit with the gobo maker so you know what you're looking for.

Thanks, I should have stressed that mounting bulky equipment is not an option. The challenge for this is that device should be relatively small (something you can easily carry in a backback) and work from batteries. The place where it'll be used is not a place you can control over (i.e. ask for electricity plug or mount something on the ceiling or the wall etc).

The two general ways of doing this would be [A] projection and [B] scanning with a laser.

Projector optics combined with a ring-mask (a "gobo") are a known thing in stage lighting and/or theatere. Getting them that tiny will be a challenge however.

If you are truly invested I'd suggest looking into how very small "pocket" projectors do this, maybe you could even salvage the optics out of a broken one.

Thanks! Can you provide an example of what you mean by pocket projectors? Author of the video I posted in comment actually built the table top projector with this curved mirrors. The closest thing I found at the moment.

The other comment mentions a gobo which is the industrial solution to this. But I am confused: you want to project a ring? There are all sorts of arrangements to achieve this but really you want a bright led, a collimated lens, and a mask. Or just use a gobo, or a light projector showing an image of a ring.

Right, ring is the right word :) (well, in the more perfect setup, I'd love to have more complex but based on the ring shapes, but ring would be a good a start).

LED+collimated lens and a mask sounds OK, but two questions: a) probably also need mirror, right? the device should be small and placed on the floor itself (you can't leave it or mount). b) mask is for, well, masking out the light from the center of the circle? Part of the challenge here is to make the ring as bright as possible (could be used in pretty bright environments), so ideally no light coming out from the optic system should be wasted (masked). Again, not sure it's even possible.

From my understanding, the 5mw limit doesn't necessarily apply when a laser is scanned and has appropriate interlocks. For example the valve lighthouse (VR base station) is classified as class 1, but contains class 3 laser sources.

I think the scanning laser path is more likely to be what you want. Projection-based solutions will probably end up being a lot larger (and more power hungry), as I think you'd need a reasonable amount of quality optics to make anything resembling a sharp line at 1m distance from anything described as a 'small' gadget.

Thanks for the comment. Scanning laser path is something like LaserCube is offering, right? I talked to them and they warned that it's a class 4 laser and shouldn't be used anywhere accesible to people for safety.

So yes, basically what I'm looking at the moment is a way to verify the intuition about if that's even possible. Or what minimum size/power is needed to do that ring. And it's either a lot of DIY that are beyound my skills and time at the moment, or some good simulation software. I'm playing with Blender + LuxRender, but it's still too slow on my M1 and it's basically bruteforcing ideas instead of proper design.

> LaserCube

Hmm, that looks like a 2D galvo scanner. I was thinking more like a laser pointer and a mirror attached to a motor.

> what I'm looking at the moment is a way to verify the intuition about if that's even possible

You just have to build it! It's definitely possible, and the only question I have is the power of laser needed to be visible in your environment for a given circle size, and a simulation will not tell you that.

You could maybe take a look at this video if you wanted more complex shapes.

https://youtu.be/wk67eGXtbIw

Thanks! I loved that video of Steve, and tried to find how can I actually design such lens like olympic rings myself, but ended nowhere atm.

Collimate a LED light, and bounce it off a conical mirror or through a conical lens, depending on if you want to have the LED pointing up or down which depends on the size constraints.

Many comments below seem to be from persons “interested” in optics, but who cannot design simple optical systems using first order optics, and who have never set up simple ray tracing with basic components or 2 x 2 matrices.

Just as in electronic design with SPICE, in sizing components and choosing materials using FEA, or in fluency with ONE computer language, advanced tools are useless if you haven’t put in the basic study in the field, and have never used these components in the lab.

No program can turn you into a practitioner by itself.

Could a complete amateur learn how to make a cheap DIY telescope lens in their own garage?

Yes!

While I haven’t done so myself, I looked into home-made telescopes as a way of getting much larger apertures on a budget.

There is a whole group of hobbyists making primary mirrors up 80cm diameter for an order of magnitude lower cost than any off-the-shelf telescope with the same dimensions.

Of course this takes hundreds of hours of work (fun!) and weeks of waiting while the glass is sent off to be mirror coated.

Definitely. People have made DIY scopes for at least a century. There's still a lot of old literature around. Among optical systems, a basic telescope is one of the least complex.

Check out Huygens Optics on YouTube. Lots of practical advise and tutorials.

Yes there are many books. It’s the motivation.

Could you recommend one?

Google “amateur telescope making.” There are a hundred books.

I'm not sure if it will necessarily be cheap, maybe cheaper than an off the shelf mirror of equivalent size. This is what John Dobson (of the Dobsonian) is famous for. However you still need to do stuff like coat the primary mirror once it's polished and you can't easily do that at home.

See https://youtu.be/snz7JJlSZvw

Silvering mirrors is coming back in fashion, either with the old recipe à la Jean Texereau / How to make a telescope, or with a spray silvering process. This week I silvered my last mirror with spray silvering, which was quite economical and convenient. I used 1.6g of silver nitrate for a 200mm mirror.

If by chance you read french, there is the reference for french amateur telescope builders here http://www.astrosurf.com/texereau/

yes: a simple scope is really just two lenses in the right positions with some wrapping around it. I sat around with lenses of known focal lengths, the wikipedia page for keplerian telescope, and just sort of 3d printed enclosures until everything was the right distance from each other.

I like visualizing things. Tools like geometers sketchpad were useful when I was learning trigonometry, even though the equations weren't very hard.

The basic optical equations aren't very hard, but equations aren't sufficient for developing an intuition (for me!). Sticking a bunch of lenses on an optical breadboard and shining light through them was very enlightening.

Playing around in SketchUp won't make you into a mechanical engineer. If there were an equivalent for optics it wouldn't make you into an optical engineer. But that doesn't mean it wouldn't be very useful for it's tiny market.

The few introductory guides I've found seem to dive straight into some very dense maths.

There's also a difference between having an intuitive understanding of a subject and a mathematical one. You can have the former and know how to use tools, eg, using a fourier transform is different to doing one manually.

actually I just learned SPICE in about a day and it completely revolutionized how I make circuits. Same for optical tools- a little time spent with a visual tool (like this one) can greatly enhance my ability to work creatively and make useful things in reasonable amounts of time.

I too learned a lot from Spice. But I was familiar with linear systems, circuit theory, op amps, and had a decent lab with components, a soldering iron, a scope, a signal generator, and a meter. For me, you have to build stuff to really understand. Imagine my surprise when I discovered that a simple 7805 regulator’s output voltage wasn’t 5 V with no load!

The other good, free simulator I've used is https://phydemo.app/ray-optics/ although I've struggled to get it to do anything real because it doesn't seem to allow you to set the focal length of lenses except by editing their sizes/shapes.

I like to crank the # of rays way, way up and then sort of move lenses around to visualize somethign I'm working on the optical table.

+1 for also looking for a tool like this (online or otherwise) before, and being disappointed with what I found previously. Two projects immediately spring to mind: firstly, we have a theodolite to align some lab equipment, which I'd like to add a live digital camera to, and secondly we're designing a material heating system based on halogen tubes, for which we need to optimise the shapes of reflectors around the bulbs, subject to some geometry constraints.

I'd not seen the sharedigm.com platform before either, possibly itself also interesting for other things.

The first task could be done with a tool like this, but it is not a project for beginners. The second requires a different tool, a non-sequential ray tracer. See https://community.zemax.com/got-a-question-7/sequential-vs-n...

I used OpticSim.jl [1] in the past and it does the job. The documentation is far from perfect, but it's really capable. I managed to reproduce subtle distortions observed with Zemax in hyperspectral imagers. There is also a python ray tracer that I didn't have a chance to try, but it seems that the development has stopped [2].

[1] https://github.com/brianguenter/OpticSim.jl [2] https://github.com/quartiq/rayopt

This is awesome! I spent a lot of time looking for a tool just like this and didn't find it. I've designed some simple-ish optical systems and I've been really frustrated by having trouble visualizing how different elements of the system will affect the light (especially moving parts). Also frustrating to have to calculate positions manually, etc. All the tools I could find were high-end quantum-optical simulators that cost more than my salary.

It looks like it doesn't support mirrors or things like right angle prisms. Mostly targeting camera lenses, it seems. Not super useful to me at the moment. I'll be watching with interest.

Have you got any recommendations for open-source, or free, optics design software?

No, I never found anything satisfactory. I ended up using SOLIDWORKS with some 3D sketch lines and reflection constraints to approximate it.

I have done this myself, and I have a lot of experience in optics. I have also written a ray tracing program in Excel, which was able to reproduce aberration diagrams. By doing this first, I could use “professional” tools.

Solidworks is excellent for reflective optics, and for prismatic, mainly planar systems. You can simulate Snell’s law either with an equation, or a simple construction using line segments of unit length, and of length equal to the refractive index of the surrounding medium. If you have nothing else, but you understand optics, you might get by with Solidworks. And there are commercial optics addins that have been written using the SolidWorks API.

No. There just isn't. It's a shame there are so many wheels for artistic rendering ray tracing with amazing performance, but simply no one bothered to invert camera as sources and objects as detectors, add more geometry and a tiny bit more of physics, and make a opensource non-sequential ray tracing engine suitable for optics design.

Commercial tools like Zemax cost a leg (> $13k, and they were changing to subscriptions even before being bought by ANSYS) and have a rather steep learning curve. There are newcomers like COMSOL offering a ray tracing toolbox, haven't tried but they are offering generous free trials and the price seems to be more reasonable.

Edit: maybe the abadonware Beam4 mentioned by another post worth trying.

You can use some software for free with limitations eg OSLO

https://lambdares.com/oslo-edu-download

FRED also has a demo mode

https://photonengr.com/fred-software/demo-request/

That might be enough for simple designs.

Beam Four is also decent and allows you to use most common surface types including gratings.

I'm interested in Optica (for Mathematica) but I don't know anyone that's used it.

What sort of thing have you designed? I’m curious what the real world uses are for these sorts of tools.

It was a kind of laser scanning instrument, for my job. Measuring the intensity of reflected light. Don't want to get into it too much.

Stealthy answer

So I'm quite uninformed on the topic, but why can't modern path tracing raytracers do this? They don't help with calculating positions, though, but might they still be suitable for checking the results?

You might be interested in reading https://benedikt-bitterli.me/tantalum/

Got so excited when I saw this as I was recently trying to design a simple optics setup for focusing an infared light source into a collimated beam.

Unfortunately the components seem very limited and there's no option to create your own custom ones by inputting surface, refractive and reflective characteristics and geometry.

... I got excited when I saw this because I thought it might be an interactive tutorial about lenses and optics in Haskell.

Is there a consensus on the standard optics 101 textbook?

The one I've mostly heard about is Hecht.

The optics chapter(s) within a general physics textbook would usually be a good place to begin without diving too much. My univ used Tipler, but other Univ Physics books would likely just be equally good.

Then after that, as already answered, Hecht to go further while still being within a "101" level.

Almost all of the camera lenses have moving parts, for focusing and possibly zooming. Is there any way to reproduce that movement in this tool?

Doesn't seem like it, though Bill Claff's tool, the lens database of which is used here, can: https://www.photonstophotos.net/GeneralTopics/Lenses/Optical...

Of course. Change the spacing.

Interesting, does it have a solver?

Shame it isn't open source, something like this that could be iterated on would be huge.

Beam4 is available as fully open source optics software:

https://www.stellarsoftware.com/

Wow, the interface looks like it's from the 80's

Wait until you see EZNEC for antenna simulation... At least it has a GUI, the original NEC was punchcards!

Wait until you see Zeemax a $4000 a year Software (last i recall) which is one of the 2 or 3 main players in this field. The interface was absolutely atrocious when I had to use it 8 years ago or so.

Reply on news.ycombinator.com