18 replies to this topic

[coinboy1]

Hello,

 

This morning I was playing around with the trial version of Optalix. It seems pretty easy to use (easier to me than OSLO). I have been interested in the lens design of Pentax lenses and found many patents while searching the U.S. Patent office.

 

US Patent # 4,545,654 is one that appears to be from the Pentax A* 400mm F/2.8 based on the website:

 

http://pentaxstudy.b...A428_Design.htm

 

The website describes ray trace results from Example 5 of the patent. 

 

 

This example is for a 100mm FL sample. From the website http://douglasviewfi...-designers.html states that a lot of the patents are scaled examples of the actual lens. 

 

So I figured I would scale all the radius and distances of Example 5 and plugged it into Optilax. It worked kind of... 

 

 

I multiplied everything by a factor of 4 and ran it into Optilax with this optical layout which element 2 looks really messed up.

 

 

And the spots dont really match the design study from the first link of this post.

 

The point of this post was to pose 2 questions. 

 

1) Am I on the right track as far as scaling from the patent, what do I need to do to match the results of the design study?

2) Is it legal/ethical to reverse engineer designs from patents for educational use? I am purely interested in finding spot diagrams for lenses from patents for educational use and not for commercial endeavors.

 

Thanks!

 

-Tony

Attached Thumbnails

• 

Edited by coinboy1, 01 August 2020 - 07:17 PM.

[gr5org]

That patent was filed in 1989:

http://patentimages....s/US4904070.pdf

 

The longest a USA patent can last is 20 years (same with japan but this is a USA patent) form time of *filing*.  So this patent ran out more than 11 years ago.

 

THE ENTIRE PURPOSE of the patent system is supposedly to create a knowledge base for all mankind to use.  In return, companies/individuals get a limited exclusive right to that patent which they can then license out that right if they want.  Notice I say "supposedly" as some people have other ideas but that was/is the stated intent for creating the patent system.

 

So yes, you can use it for whatever you want now that the patent is expired.  You can publish it yourself.  You can sell copies of it.  You can make the same lens based on the patent.  You can do almost anything with the information in that patent except claim it as your own idea.

[gr5org]

Correction.  The patent I linked to above is a different patent that mentions the japanese patent which therefore must be even older.  So it is also expired.

[coinboy1]

Thanks for that info! Thats great news and what I wanted to hear!

 

I am really interested in finding the Patent for my recently acquired Pentax 67 M* 400mm EDIF lens which is a medium format lens that once cost over $6,000 new. This was designed around 1986 and I can not find the patent for the life of me. I really want to ray trace it and get a modern spot diagram generated. Here is the lens design (also found in the original Pentax M* 400 EDIF manual). 

 

 

Attached Thumbnails

• 

[TOMDEY]

Patents comprise Teaching, Drawings, and Claims.

 

The claims spell-out what is excluded from others doing, making, or using... unless permission is granted by the assignee. That also discludes ~making one for personal use~, although the likelihood of that being enforced would be rare.

 

The Teaching (body of text part) and referenced/included drawings should allow, in theory, for one ~skilled in the art~ to successfully duplicate what the inventor(s) have discovered/developed. That would, however, require permission.

 

Composing patents is in itself an art. The assignee (inventor himself, or more often, his employer) --- need to teach enough to legitimize the claims, but are prudently reluctant to gratuitously offer additional (most often critical!) ~tricks of the trade~ to actually pull off execution. An illegal copy-cat is likely to experience significant to severe/debilitating problems, without negotiating permission, and further guidance. This explains why companies often buy license and also hire the inventor (or his sponsor) to aid in development. That's the dream of most hobby inventors... but rarely realized, in that amateur context.

 

1) scaling - yes, indeed. From what I've seen, most patent-exemplified lenses are scaled. One thing to look for is wavefront. If you discover that that it comes in either absurdly magnificent, or absurdly terrible... that can indicate scaling. I worked on a lens that I scaled down by 10x, because I was using an off-campus computing engine to run the optimizations (only a few such computers in the world). And then demanded that the computer shoot for 1/1000 wave RMS performance. That succeeded, and we built it. The actual aperture was indeed bigger than ten inches. It worked fine.

 

And the patent example is almost never the specific item they intend to actually build... but shows/discloses the claimed features.

 

2) you are free to ~play around~ with the disclosed teaching to your heart's content! That's both legal and ethical, and is exactly what the assignee provides, in return for temporary license to exclude others from actually executing on the claims.    Tom

 

PS: Vanity Patents... near always a dead end. If one wants to try that... at all... get a good local attorney to handle all the legal stuff, and coach on format, value, etc. Avoid the TV Infomercial ~Patent Mills~ They will tell you that your idea is the most brilliant thing since sliced bread and the mouse trap... and then happily slap you with consulting fees till the cows come home... which they never will...    Tom

[ButterFly]

Patent terms ran for 17 years from the date of grant back then.  One needs to consult the file wrapper to deteremine whether and when it expired.  Consult an attorney if you have any plans to make, use, sell, offer for sale, or import the patented invention.

[Ed Jones]

  Some years back I would reverse engineer cell phone camera lenses by first sawing/grinding it in half, photographing it, then making a clear transparency overlay of the layout.  I'd place this on my monitor and with Zemax, manually chose curves, spacings and apertures to match the overlay.  This was a bit harder to do because they use some pretty wild aspheres on their plastic lenses. We could usually identify the plastic.   When I got it close I let Zemax optimize it and then see how close it fit the layout.  I had a lot of fun doing this and got good results too. 

  All glass lenses would be easier, you might try this on your Pentax lens layout letting it pick glass types (if your trial version will let you).  

[MKV]

This morning I was playing around with the trial version of Optilax.

Tony can you provide a download link? I'm curious how it compares to OSLO in ease and scope. I wasn't able to find any link for optical software under Optilax. Thanks.

Edited by MKV, 01 August 2020 - 05:59 PM.

[coinboy1]

Here you go Mladen http://www.optenso.c...d/download.html

 

I downloaded the LT version (30 day free trial, no limit on # of surfaces like OSLO)

 

I like it a bit better than OLSO but I couldnt access the glass database in Optalix. I used MODAS to get all my glass info and typed it into Optalix. 

 

Also download the "Handbook of Optical Systems. Optical Design Files" it has A LOT of telescope designs in there! 

Edited by coinboy1, 01 August 2020 - 06:24 PM.

[coinboy1]

Patents comprise Teaching, Drawings, and Claims.

 

The claims spell-out what is excluded from others doing, making, or using... unless permission is granted by the assignee. That also discludes ~making one for personal use~, although the likelihood of that being enforced would be rare.

 

The Teaching (body of text part) and referenced/included drawings should allow, in theory, for one ~skilled in the art~ to successfully duplicate what the inventor(s) have discovered/developed. That would, however, require permission.

 

Composing patents is in itself an art. The assignee (inventor himself, or more often, his employer) --- need to teach enough to legitimize the claims, but are prudently reluctant to gratuitously offer additional (most often critical!) ~tricks of the trade~ to actually pull off execution. An illegal copy-cat is likely to experience significant to severe/debilitating problems, without negotiating permission, and further guidance. This explains why companies often buy license and also hire the inventor (or his sponsor) to aid in development. That's the dream of most hobby inventors... but rarely realized, in that amateur context.

 

1) scaling - yes, indeed. From what I've seen, most patent-exemplified lenses are scaled. One thing to look for is wavefront. If you discover that that it comes in either absurdly magnificent, or absurdly terrible... that can indicate scaling. I worked on a lens that I scaled down by 10x, because I was using an off-campus computing engine to run the optimizations (only a few such computers in the world). And then demanded that the computer shoot for 1/1000 wave RMS performance. That succeeded, and we built it. The actual aperture was indeed bigger than ten inches. It worked fine.

 

And the patent example is almost never the specific item they intend to actually build... but shows/discloses the claimed features.

 

2) you are free to ~play around~ with the disclosed teaching to your heart's content! That's both legal and ethical, and is exactly what the assignee provides, in return for temporary license to exclude others from actually executing on the claims.    Tom

 

PS: Vanity Patents... near always a dead end. If one wants to try that... at all... get a good local attorney to handle all the legal stuff, and coach on format, value, etc. Avoid the TV Infomercial ~Patent Mills~ They will tell you that your idea is the most brilliant thing since sliced bread and the mouse trap... and then happily slap you with consulting fees till the cows come home... which they never will...    Tom

Thank you Tom,

 

That was a very detailed and informative post! Well written and I gained a lot of insight!

[coinboy1]

  Some years back I would reverse engineer cell phone camera lenses by first sawing/grinding it in half, photographing it, then making a clear transparency overlay of the layout.  I'd place this on my monitor and with Zemax, manually chose curves, spacings and apertures to match the overlay.  This was a bit harder to do because they use some pretty wild aspheres on their plastic lenses. We could usually identify the plastic.   When I got it close I let Zemax optimize it and then see how close it fit the layout.  I had a lot of fun doing this and got good results too. 

  All glass lenses would be easier, you might try this on your Pentax lens layout letting it pick glass types (if your trial version will let you).  

I remember when you reverse engineered my Takahashi BRC-250 a couple of years ago! I was super impressed that you were able to do that from an image and some basic parameters! Want to give it a go for my Pentax 67 M* 400mm ED IF lens?   These are the only two glass layouts I could find for them (The two red elements I believe are the ED elements), I haven't found the patent yet. I wish I had Zemax...

 

 

[MKV]

Here you go Mladen http://www.optenso.c...d/download.html

Thanks, Tony!

Mladen

[Ed Jones]

 

Want to give it a go for my Pentax 67 M* 400mm ED IF lens?

Do you know any dimensions like R1 to R18, the aperture of lens 1, BFL?  I need something to scale it by.

I really loved to reverse engineering, I worked in R&D and had to reverse engineer competitors lens designs for patent infringements.

[Mike I. Jones]

Tony - you'll like this site:

 

http://www.lens-designs.com

 

Optalix is good stuff, I especially like the graphics.

[RajG]

  Some years back I would reverse engineer cell phone camera lenses by first sawing/grinding it in half, photographing it, then making a clear transparency overlay of the layout.  I'd place this on my monitor and with Zemax, manually chose curves, spacings and apertures to match the overlay.  This was a bit harder to do because they use some pretty wild aspheres on their plastic lenses. We could usually identify the plastic.   When I got it close I let Zemax optimize it and then see how close it fit the layout.  I had a lot of fun doing this and got good results too. 

  All glass lenses would be easier, you might try this on your Pentax lens layout letting it pick glass types (if your trial version will let you).  

I'd like to see the design of the lens on the IPhone 11 Pro. Will Apple Care cover the replacement after you saw the lens in half? 

Edited by RajG, 02 August 2020 - 08:33 AM.

[coinboy1]

Do you know any dimensions like R1 to R18, the aperture of lens 1, BFL?  I need something to scale it by.

I really loved to reverse engineering, I worked in R&D and had to reverse engineer competitors lens designs for patent infringements.

Hello Ed,

 

I may end up taking apart my Pentax lens because I am having collimation issues; if I do I will measure one or more of the elements for you. I just got the lens from eBay and they offer a 30 day return so its either disassemble and try to fix the issue or return it. I started a thread here: https://www.cloudyni...if-camera-lens/

 

P.S. Very cool about your reverse engineering experience, that sounds like my dream job ;-)

[RajG]

I've had fun trying to replicate old patents, which I've found on google patents. Some work well, others don't -  I suspect some inventors show broken designs to throw people off the scent. Another problem with very old patents is that they use obsolete glasses and replacements aren't always obvious.

 

Here are a couple that I've successfully replicated in OSLO EDU:

 

Pentax (Kobayashi) Cooke triplet, works well at 100mm focal length f/8, but doesn't scale well.

 

https://patentimages...5/US3486805.pdf

 

Herzberger & Hoadley 100mm focal length f5:

 

https://patentimages.../US2487873A.pdf

 

The first one is very good, use fluorite and N-SK16 for the middle element. I think this can be scaled up to about 250-400mm fl. Some examples use a glass with Abbe number 100.5, not sure what that is.

 

This last one is Herzberger's superachromat, 100mm focal length f/2.8. Too many surfaces for OSLO EDU. If you are able to replicate this, let me know how it works.

 

https://patentimages...2/US3395962.pdf

Edited by RajG, 02 August 2020 - 10:51 AM.

[RajG]

Following up on my prior post, here's the Herzberger Hoadley #1 f/5 lens at the original 100mm fl & scaled up to 250mm fl. I adjusted the spacings slightly to make it work better. This is an awesome design. The 250 f/5 with a suitable field flattener would be a cool alternative to the popular WO Redcat 51.

 

100mm focal length version:

 

 

250mm focal length scaled version:

 

 

 

 Herzberger Hoadley 100 f5.len   831bytes   0 downloads

 

 Herzberger Hoadley 250 f5.len   653bytes   0 downloads

 

[Ed Jones]

 

I'd like to see the design of the lens on the IPhone 11 Pro. Will Apple Care cover the replacement after you saw the lens in half?

  Well I never knew what cell phone they were used in but I think it might have voided the warranty.    We were never successful at getting into that market, however..  I'm surprised I still had a design on my computer, I would have bet they were all on 3.5 floppies and my computer has no such drive.  I'll attach it.

  Yea back in the hay day of projection TV US Precision Lens owned many patents on projection lenses and had more than half the TV market.  Enforcing our patents kept our lenses in TVs that nearly were all made overseas.

  Another tricky thing to reverse engineer were cemented doublets in optical designs and they didn't want them separated or destroyed if at all possible.  I could measure the thickness, paraxial BFL, and confocal distances (2) on a Zygo.  I couldn't know the thicknesses, internal curve or glass type but could get a reflection off the internal curve for confocal distances.  I had to set up multiple configurations linked correctly and the merit function if set up right would let Zemax solve for these unknowns, and least the glass indices at 633.  Yea them were the fun good ole days!

Here's one cell phone design, I know nothing else about it:

 

Attached Thumbnails

•