Switch to English Language Passer en langue française Omschakelen naar Nederlandse Taal Wechseln Sie zu deutschen Sprache Passa alla lingua italiana
Members: 69,750   Posts: 1,515,756   Online: 1071
      
Page 7 of 8 FirstFirst 12345678 LastLast
Results 61 to 70 of 73
  1. #61

    Join Date
    Aug 2004
    Location
    USA (Utah)
    Shooter
    4x5 Format
    Posts
    165
    Quote Originally Posted by Leigh B View Post
    Hi Charlie,

    Sorry to disagree, but glass certainly is machined, at least as far as the grinding process is concerned.

    The grinder is a machine. It can be controlled by computer, and the results can be analyzed by automated measurement.
    The results are used as input to the control system to fine-tune the position of the active tools (grinding surfaces).

    This results in a much higher achievable uniformity than was previously possible.

    - Leigh
    Yes but you are grinding a super-cooled liquid, so it isn't quite the same as "machining" a solid. Glass flows. And "tar pitch" isn't my idea of a precision machining fixture. Maybe it is for you. Just because something is a "machine" doesn't mean it produces "machined parts", or why don't we call the prints we make from enlargers "machined prints"?
    (Posted with all the seriousness that such a weightly academic topic deserves--or doesn't!)

    Charlie

  2. #62
    ic-racer's Avatar
    Join Date
    Feb 2007
    Location
    Midwest USA
    Shooter
    Multi Format
    Posts
    6,103
    Quote Originally Posted by voceumana View Post
    Glass flows.
    Leigh is correct. Glass does not flow when it is in the super-cooled state. Read more here: http://www.glassnotes.com/WindowPanes.html

    In May 1998, Zanotto wrote an article in the American Journal of Physics relating to the false notion that observations of thick glass in old windows translated to the fact that glass is a liquid. Zanotto sought to calculate the flow of glass and found that at 414 Celsius (777 °F) the glass would move a visible amount in 800 years, yet at room temperature he found that it would take glass 10,000 trillion times the age of the earth.

  3. #63
    Leigh B's Avatar
    Join Date
    Jan 2011
    Location
    Maryland, USA
    Shooter
    Multi Format
    Posts
    1,035
    Images
    1
    Quote Originally Posted by voceumana View Post
    Yes but you are grinding a super-cooled liquid, so it isn't quite the same as "machining" a solid.
    Grinding is the most precise and accurate of the standard machining methods.

    Every high-precision part in industry, regardless of the material, is set to final form and dimension using grinding techniques.

    - Leigh
    “Wise men talk because they have something to say; fools, because they have to say something.” - Plato

  4. #64

    Join Date
    Aug 2004
    Location
    USA (Utah)
    Shooter
    4x5 Format
    Posts
    165
    OK--I give up, none of my suggestions seem to explain why nearly all enlarging lenses have an element or elements that are not perfectly centered. Ctein reported this in an article for a Creative Camera Darkroom Guide (the article was on choosing an enlarging lens), and I trust his work.

    To me, the reason why there are off-center elements doesn't really matter, what matters is that it exists. I believe Ctein did enough work on the subject to count as fact for me. And his statement that this occurs explains to my satisfaction why specific lenses seem to produce better images than others of exactly the same design, same model, same manufactuer.

    Charlie

  5. #65
    Leigh B's Avatar
    Join Date
    Jan 2011
    Location
    Maryland, USA
    Shooter
    Multi Format
    Posts
    1,035
    Images
    1
    Quote Originally Posted by voceumana View Post
    OK--I give up, none of my suggestions seem to explain why nearly all enlarging lenses have an element or elements that are not perfectly centered. Ctein reported this in an article for a Creative Camera Darkroom Guide (the article was on choosing an enlarging lens), and I trust his work.
    Amazing how he finds so many problems that nobody else finds or reports.

    Why would anybody trust his work?

    All products in the world are manufactured to tolerances. This is a universal truth.
    No product exists that is absolutely nominal in all respects. It's not physically possible.
    So the question becomes: "How much deviation is acceptable?"

    Manufacturers put considerable resources into investigating and answering that question,
    and come up with production standards that reflect their desired level of quality.
    They're interested in establishing the reputation of a product line in the market.

    Ctein on the other hand is only interested in filling his wallet.

    This tactic has been used forever. In the news business it's called "yellow journalism".
    Find some nit, blow it way out of proportion, and create a perceived problem where none really exists.
    The author has nothing to lose, since the whole idea is to get folks to read his work, regardless of its veracity.

    - Leigh
    Last edited by Leigh B; 08-29-2012 at 12:46 PM. Click to view previous post history.
    “Wise men talk because they have something to say; fools, because they have to say something.” - Plato

  6. #66

    Join Date
    Jul 2011
    Shooter
    8x10 Format
    Posts
    2,432
    Ctein is a friend of mine. I sometimes disagree with him about this or that, but know him well enough to recognize he actually does his homework and is not simply bluffing like so many other article writers. Some of his information about lenses etc is simply out of date, but possibly still relevant if
    you're considering lenses of an appropriate vintage. Quality control varies less than it used to, at least among the major manufacturers in their upper-end products. Sometimes cheaper series of enlarger lenses have distinct quirks like focus shift at different f-stops. Quality control is NOT just
    a matter of machine tolerances, which are not themselves always perfect, but of human assembly
    of components too. But the odds of getting a bad modern lens from any of the big four is pretty low.

  7. #67
    Steve Smith's Avatar
    Join Date
    May 2006
    Location
    Ryde, Isle of Wight
    Shooter
    Medium Format
    Posts
    8,529
    Images
    122
    Quote Originally Posted by DREW WILEY View Post
    Quality control is NOT just a matter of machine tolerances, which are not themselves always perfect, but of human assembly of components too.
    And where the tolerance off of tooling is not accurate enough, other methods are used such as shims or matching an over sized part to a mating under sized part to cancel out the error.

    However, I would think that a lens element would have to be off centre by a lot more than would be possible in even a very poorly machined housing to make any noticeable difference.


    Steve.

  8. #68
    Leigh B's Avatar
    Join Date
    Jan 2011
    Location
    Maryland, USA
    Shooter
    Multi Format
    Posts
    1,035
    Images
    1
    Yes, there are many techniques that can be used in final assembly and in the preparatory
    stages to minimize the effects of errors in individual components.
    Reputable manufacturers (i.e. all of the major mfrs) take the time and expense to do that.

    To say that the elements in a multi-element lens are "out of alignment" is gratuitous nonsense.

    To take a moronically trivial case:
    It is not possible to make a single plano-convex lens in which the optical axis is exactly coaxial with the physical axis.

    Note that I said 'exactly'. It's easy to get the two axes quite close, but measurement techniques do not exist of sufficient
    accuracy to make that alignment exact.

    So saying that the elements in a multi-element lens are "off-center" is a tautology. It's true by definition, and completely irrelevant.
    This is a typical example of blowing up a nothing into a cause celebre.

    - Leigh
    “Wise men talk because they have something to say; fools, because they have to say something.” - Plato

  9. #69

    Join Date
    Jul 2011
    Shooter
    8x10 Format
    Posts
    2,432
    Just depends how fussy you need to be. As I've stated elsewhere, I cut my teeth on big precision
    Ciba prints, so became pretty damn nitpicky about enlarger lenses. I also made very precise 8x10
    dupes as part of the workflow, which needed a different category of enlarging lens. That being said,
    there were still plenty of relevant choices. And anyone making reasonable-sized black and white prints would be like a kid in a candy shop right now, there are so many bargains out there, unless you're after something distinctly rare, like an Apo El Nikkor - which is too heavy for many typical
    enlargers anyway, and sheer optical overkill.

  10. #70
    polyglot's Avatar
    Join Date
    Jun 2009
    Location
    South Australia
    Shooter
    Medium Format
    Posts
    3,053
    Images
    12
    Alignment is an issue, though a rare one and only really with certain designs. Mass-manufacturing techniques got a lot better in the 90s, concurrently with improvements in software for the optimisation of optics. The result we see is that complex designs (zooms mostly) that were not previously commercially feasible due to rejection rates due to poor alignment suddenly became a lot more workable. Some optical designs are more sensitive to alignment (both radial and axial placement of an element) than others and we can now build the more-sensitive but higher-performing designs reliably.

    For example, see all the 18-250mm crazy-zooms appearing for DSLRs with decent sharpness (though it seems manufacturers/consumers of those lenses care not for reducing distortion); they were not previously possible because of the assembly tolerances required. And they're hugely contentious on the digital forums because very poor examples slip through QA somehow and cause a lot of angst - you get people saying "mine is pixel-perfect sharp" and posting images while others are saying "but no this lens is shit" and posting images.

    If you think of a graph of lens performance (in whatever dimension: resolution, contrast, chromatic aberration (CA), coma, etc) vs alignment error, it's going to be a peak with a drop-off to either side. Some designs have broad flat peaks therefore are easy to manufacture, other designs have a performance/alignment function that looks more like a thumbtack: a big spike with sudden dropoff. If there weren't this performance/alignment variability, we wouldn't be seeing sample-to-sample performance variability on brand-new lenses.

    That said, enlarger lenses tend to be very much of the older robust designs that are easy to manufacture. While there is some sample variation, they still tend to be diffraction-limited and with little lateral CA; the APO versions have reduced CA. Enlarger lenses are not ultrazooms and they are not complex. While I'm sure you can find an occasional Really Bad one, it's not common and not nearly as bad as the sample variation seen on the really complex new designs.

Page 7 of 8 FirstFirst 12345678 LastLast


 

APUG PARTNERS EQUALLY FUNDING OUR COMMUNITY:



Contact Us  |  Support Us!  |  Advertise  |  Site Terms  |  Archive  —   Search  |  Mobile Device Access  |  RSS  |  Facebook  |  Linkedin