leica.gmbh 中文 · EN · DE

技术记录

卖的是「德国品质」,盖玻实际是什么

Owners bought an 18MP full-frame Kodak CCD (KAF-18500 family) marketed as the digital M. The failure that destroyed files at f/5.6–f/22 is not a dirty sensor. It is corrosion/oxidation of the cover glass / IR cut filter sitting on the CCD — white halos, not black dust specks.

Construction (as described in press and repair literature)

  • Sensor: Kodak full-frame CCD ~18MP, unique microlens offsets for M rangefinder ray angles.
  • Cover: Schott S8612 / BG-type IR absorbing glass. Thorsten Overgaard and others noted coating peel as early as 2014 on Monochrom bodies in for CLA.
  • LDP LLC (PetaPixel 2020): spectrometer on a stripped ICF found transmission to 300nm — “That filter doesn’t have coatings.” BG/UG glass oxidises with time, humidity, ozone.
  • Leica’s own early replacements sometimes used the same vulnerable family (owner CCD-ID lore: IDs below 15 still corrode; 15/16 treated as the later attempt). That is why people who already paid Wetzlar once still show up on forums in 2025.
  • After 2020, independent repair = replace the glass (Kolari BG39 etc.), not a new Kodak CCD, because Kodak CCD production for this part is gone — matching Masney’s “sensors are not available any longer.”

Quality questions the complainant put in 2019 and still lists

  1. Why sell 2014–2016 bodies without a prominent, durable public warning that the cover glass was known to fail?
  2. Why did the first replacement campaigns put back a still-vulnerable glass/sensor stack?
  3. Why is “German quality / a Leica is forever” still the marketing wrapper while the CCD generation has a cottage-industry glass-swap market?
  4. Why is the official 2014 notice URL a 403 rather than a still-readable product-safety page?

These are questions, not laboratory findings of this archive. The laboratory-style findings cited are third-party (LDP LLC, Kolari, forum CCD-ID checks). This site exhibits them so Leica can admit, deny or correct them in writing.