May be a couple of lenses and the broadband coatings.  You'll probably save a couple hundred bucks, and spend an extra $20k in designing, specifying, and procuring the lenses.  There will be an additional markup due to the fact that your lenses will be one-off, as opposed to production lenses that Mitituyo would make in large volumes.

TTFN

FAQ731-376: Eng-Tips.com Forum Policies
Chinese prisoner wins Nobel Peace Prize

"Couple of lenses" sounds very promising. Produce them possibly is not that complicated also if glasses are not so exotic. Coatings is not a problem for me at all (it is not broadband). Just design costs of ~$20K a bit terrible. Is there really that much work? For experienced Zemax  designer few lenses on single wavelength shouldn't very big deal I believe. Or I'm wrong?

I would really appreciate more input on this subject. Especially from someone who related to optical designs.      

Most Optical Design consulting firms will provide a free estimate of the fees to design the system you are talking about. Do search in a photonics directory and contact someone about a quote. IRStuff is probably close depending on how much of the design needs to supplied. When you buy a lens system from Mitutoyo you get more than just a ZEMAX prescription. There's the mechanical mount to be designed and manufactured in addition to manufacturing the lens elements. Designing for NUV can be expensive due to the materials required; the coating requirements will impact the cost as well but not as much as much as might be as expected but you won't know without a quote.

Harold
SW2011 SP1.0 OPW2011 SP1.0 Win 7 Ultimate
BOXX 8550 Xtreme Dual Xeon 5680 @4.2Ghz
nVidia Quadro 4000
www.lumenflow.com

Well, I guess I dont' see why you're having difficulty thinking that this won't be cheap.  Color correction is almost always the last step in a long process of balancing competing requirements.  

Just how many microscope objective designers do you think there are in the world?  And what makes you think that you can get to hire one to do your one-off?  Just because a ZEMAX or CodeV designer is "experienced" does not mean that they're experienced in microscope objectives, particular a 100x.  Moreover, unless that "experienced" ZEMAX designer knows everything there is to know about the rest of a Mitituyo microscope, he'll burn hours just trying to figure out what he even needs to design to mate to the rest of the microscope.  What your asking for is not dissimilar to asking for an "experienced" electronics engineer to design an 28-bit ALU that mates into Intel i7 quad-core processor.  Is a 28-bit ALU that much simpler than a 32-bit ALU?  Do you know enough about an i7 processor to know the timing and interfaces required to get an ALU to work in an Intel processor?  Even when I might have been considered a "experienced" designer, I wouldn't have know anything about an Intel 486 internal workings unless I was part of the design team.

I am seriously perturbed by your lack of understanding of AR coatings.  You need AR coatings to reduce reflections, regardless of whether it's a single wavelength or a spectrum.  Otherwise, you get ghost images up the wahzoo.  4% reflections from every surface is a non trivial problem for imaging.  Reducing the bandwidth simply cuts down on the complexity of the AR coat, but does not eliminate it.

There no optical designs that only use a single type of glass.  The variety of glass compositions simplify the optical design.  Unless you are an experienced lens grinder and polisher, it ain't going to happen in your garage.  Most production lens systems run batches of lens in a single run.  Running a single 1/2" lens by itself would be absurdly expensive in an actual optical shop.  You can buy a $50 moderate quality lens from Hamilton, that's made in batches of 100, or you pay $10,000 to build a one-off.

TTFN

FAQ731-376: Eng-Tips.com Forum Policies
Chinese prisoner wins Nobel Peace Prize