Making Lenses- VERY accurately. Pics.

[toastydeath]

The Keck secondary mirrors were done on the LODTM, and I believe the Keck primaries were done on DTM3.

You can indeed directly turn glass and zerodur to an IR optical finish, but it must be done with ductile regime machining. The figure must be in ballpark because the process cannot remove much material, and the tool wear is high.

Ion beam milling is a little like a coating process. A large ion gun sprays a charged stream at the workpiece through a rotating/moving mask. The ions that make it through the mask react with and erode the material. It's now a very common follow-up process to diamond turning, because many optics applications require a surface finish under 40 angstroms, which is the practical limit for commercially-available diamond turning machines. Custom lathes exist that can go down further, the current best being 10 angstroms. Ion beam milling also reduces diffraction caused by the tool marks, and will transmit more light for the same surface roughness.

[Tr2sa]

Another interesting way to get near 100nm peak to walley:

http://www.opticsexcellence.org/SJ_TeamSite/RS_mrf.html

The most interesting part is that the "tool" shape and hardness is controlled in realtime by magnetic field ("tool" is actually a lump of abrasive and magnetic material mix formed by field). It also allows continious exchange of the material in the lump.

[sc brian]

Nice work ImanCarrot - I do diamond machining as well. In fact, I have a need for adjustable height tool holders - similar to those now supplied on Precitech and Moore machines - to retrofit my older MSG-325's. Do you know anyone who sells just this item? I am tempted to just make my own, but I thought I would investigate what is for sale first.

[ImanCarrot]

MSG 325! good lord! that was the first machine I learned diamond machining on. You loaded the executive on 1" tape with punch holes in it. The keypad on the Allen Bradley controller wasn't even QWERTY, it was ABCD- engineers were'nt supposed to know how to type then

I've got an adjustable tool holder for that machine- it doesn't say who made it, but I'd try Taylor Hobson or Sterling Ultra Precision.

Incidently, if you're looking for them fuses that the machine seems to eat with a voracious appetite, I found them at Kempston Controls here in the UK.

[sc brian]

Thanks, ImanCarrot - I will try Sterling.

You will be pleased to know that I converted my MSG325's to have modern CNC controls, increased resolution interferometers, and linear motors on the slides. I just like to tinker with these things

[PEU]

Hi Iman, Im experimenting with curved surfaces on acrylic now that I have a working CNC lathe, I have some questions for you, sorry if they are too general

1) how do you calculate an aspheric lens? I want to use them for LED flashlights (im a custom maker www.neoca.com.ar)

2) Once you machine one curved side, and you want to finish the other side, how do you hold the lens centered if you have almost no material to hold it?

Thanks !


Pablo

[ImanCarrot]

The aspheric surface is defined by a standard equation detailing things like Radius of Curvature, Conic Constant and all asperic terms (A4,A6,A8,A10 etc).

My machine takes this data and outputs the X and Y co-ordinates for the tool. I've written my own software to do the same since I needed A1,A2 and A3 terms which most aspheric generators will not do.

I use a subchuck to hold the lenses. Either that or I use a collet type arrangement and a vacuum chuck. You have to be careful not to bend the lens though.

[PEU]

Do you know of some learning/tutorial for making aspheric lenses on the web? Obviously I dont need the accuracy of your work

Thanks!

[tauntdesigns]

Nice work,

Back in the day, I worked for a company that mfg lens making machines. I used to run the 4-axis cnc in production and later in QC on the DCC CMM checking lens and polishing pucks.

The company's name was Coburn Optical industries, one of their machines looked similar to yours by the name of 'Lensmaker'. They have since been bought by Gerber Scientific.

This thread brought back some good memories,
Jack

[ImanCarrot]

Turorials: Taylor Hobson or Sterling would probably do tutorials, and charge you the earth for it lol, but if you're starting from scratch it'll be a long uphill struggle. Patience is the key, and perseverance, and sometimes bloody mindedness (as in "this machine ain't gonna get the better of me!"). Oh, Imperial College (London) does general optical courses- again costing a lot and aimed at accedemic type people, not us mere engineers.

There's no general consensus on the aspheric profile formula, some designers get the signs all mixed up and use Conic Co-Efficient instead of Conic Constant, so I generate a sag table showing the Z (you might call it Y) value for any given X value in incremental steps. This forms part of the contract so they can't say "Oh! we ment the aspheric co-efficients to be the opposite sign!". And I get it in hard copy, black and white and signed by the customer.

I firmly believe that optical designers are only marginaly more bearable than a hangover They take six months tweaking the design, use all the project's budget and timescale then when they finaly do place the order they want it yesterday, in quantities of one off's when you've quoted for 100 off's and change the design half way through your machining run, or the material, or frequently both, then they remember that the material specified doesn't transmit at the wavelength they specified on the coating.

Fun and games... I wouldn't wish it on my worst enemy- think twice before making embarking on a lens manufacturing carreer hehe.

It is nice though when you see something on the telly that's dead complicated and think "hey! my lenses are in that!". I'd say it's not rocket science, but actually... it is.

I seen a programme last night about the Mars probe (Phoenix?) and thought that the guy who made the lenses for that must have been dead proud.


Taunt: you found me out! my machine is actualy made to make contact lenses- I have to convince, lie to and threaten it every hour of every day to get it to make what I want, it's always trying to second guess what I want it to do and I'm always trying to second guess what it's trying to guess

[Dr Pete]

Nice to see all the other "optic" people here
Nice work there ImanCarrot!

Way back 1980 one of the owners of the optical co I worked for told me
"You won't make any money in optics but you will always have a job"
Now I think back on that and wonder how nice it was back in the old days.
What is CNC.....LOL !!!

Keep up the good work.
P.S. Love the pics also, keep em coming!