Sensor Corrosion Analysis and Fix [Merged]

[sandymc]

Advertisement (gone after registration)

4 hours ago, jaapv said:

Does that apply to "standard" monochroms as well, Sandy? 

Not as regards sharpness. The Leica Monochrom, for example, produces a so-called "linear raw" file that LR or C1 will correctly understand to be a monochrome image. The issue is that for a modified sensor, the raw processor sees a M9 file, and assumes it is Bayer, which it's not. AccuRaw Monochrome has an override for such situations. A dedicated monochrome processor may still have advantages in tonality.

[Einst_Stein]

Can you comment specifically if it works on converted Sony R1 or Leica Dlux 4?

[dllewellyn]

For a color camera converted to monochrome, when shooting in RAW, I recommend using Monochrome2DNG or AccurawMonochrome and then your normal image processing.

[james.liam]

I posted this below but it was swallowed into the Mods' event horizon:

Kolarivision announce its fix of M9 sensors that doesn't involve replacement but rather subsitution of the filter stack. Interestingly, it is 0.6mm instead of Leica's standard 0.8.

All for US$999.

[dllewellyn]

But then you need the change the shims *exactly* to calibrate focus.  To do it right, you would want to measure the flange-back distance - the distance the lens mount and the sensor.  If fixes were so simple, Leica would have already done that. With a really wide angle lens, the coverglass/ICF stack thickness does distort light towards the edges at wide aperture - but we are talking about lenses around 10mm.  If you are shooting 25mm and above, the stack thickness will have little effect.

[james.liam]

1 minute ago, dllewellyn said:

But then you need the change the shims *exactly* to calibrate focus.  To do it right, you would want to measure the flange-back distance - the distance the lens mount and the sensor.  If fixes were so simple, Leica would have already done that. With a really wide angle lens, the coverglass/ICF stack thickness does distort light towards the edges at wide aperture - but we are talking about lenses around 10mm.  If you are shooting 25mm and above, the stack thickness will have little effect.

The only wider-than-10 is one of the Laowa throwaways. So it doesn't matter and may in fact improve the resolution on many lenses. At least those few souls with sensors that haven't rotted yet, or were too lazy to check have an option.

[andybarton]

Advertisement (gone after registration)

1 hour ago, james.liam said:

I posted this below but it was swallowed into the Mods' event horizon:

Kolarivision announce its fix of M9 sensors that doesn't involve replacement but rather subsitution of the filter stack. Interestingly, it is 0.6mm instead of Leica's standard 0.8.

All for US$999.

That’s been mentioned several times in this thread. 

[jaapv]

1 hour ago, dllewellyn said:

But then you need the change the shims *exactly* to calibrate focus.  To do it right, you would want to measure the flange-back distance - the distance the lens mount and the sensor.  If fixes were so simple, Leica would have already done that. With a really wide angle lens, the coverglass/ICF stack thickness does distort light towards the edges at wide aperture - but we are talking about lenses around 10mm.  If you are shooting 25mm and above, the stack thickness will have little effect.

Leica uses laser-interference to measure sensor adjustment. 

[dllewellyn]

That sounds correct.  The Leica M9 is not an AF off the sensor mirrorless camera.  With a mirrorless type camera, the camera is setting focus off the sensor and you can change the ICF/AA stack somewhat (depends on firmware).  The M9 focus is calibrated at the factory so that the lens focus matches the sensor focus.  You are best off not second guessing the factory.  Theoretically, yes, you can put in a thinner ICF, but getting the sensor precisely re-positioned is not so simple. 

[Einst_Stein]

That 0.2mm light path means the sensor location is off by about 0.1mm too close to the lens, assuming the glass refractive index is 1.6 and the air is 1.0.  This implies when you focus a relatively close object it will be projected to the image plain supposed to be for infinite (and infinite will always be out of focus).  This does not count the light dispersion effect due to the glass thickness. 

The light dispersion effect can not be fixed by the shim. 

The focus error will make the following object to be projected like the inifinite.

1: focal length 16mm:  Objects beyond ~2.5m can not be focused.

2: focal length 25mm: Objects  beyond ~6.2m.

3: focal length 28mm: object beyond ~8m

4: focal length 35mm: ~13m

5: focal length 50mm: ~25m

6: focal length 90mm: ~80m

I guess this issue is almost irrelevant given the M RF accuracy.

 

[analog-digital]

What are you guys discussing?

It's history, Leica no longer repairs the corroded sensor / IR cut glass. Anyone who has such a corroded M9 reacted too late. Bad luck. Leica also offers an exchange program as an emergency exit.

Either you accept that or buy an M9 / M9-P with a NEW sensor. Which is cheaper than a repair.

[dllewellyn]

59 minutes ago, Einst_Stein said:

That 0.2mm light path means the sensor location is off by about 0.1mm too close to the lens, assuming the glass refractive index is 1.6 and the air is 1.0.  This implies when you focus a relatively close object it will be projected to the image plain supposed to be for infinite (and infinite will always be out of focus).  This does not count the light dispersion effect due to the glass thickness. 

The light dispersion effect can not be fixed by the shim. 

The focus error will make the following object to be projected like the inifinite.

1: focal length 16mm:  Objects beyond ~2.5m can not be focused.

2: focal length 25mm: Objects  beyond ~6.2m.

3: focal length 28mm: object beyond ~8m

4: focal length 35mm: ~13m

5: focal length 50mm: ~25m

6: focal length 90mm: ~80m

I guess this issue is almost irrelevant given the M RF accuracy.

 

 

[dllewellyn]

Yes, correct.  Depends on the lens, but basically the camera becomes nearsighted if the glass is too thin without changing the sensor position.

Edited September 13, 2020 by dllewellyn

[dllewellyn]

Well, there are 100,000+ M9 cameras out there.  Maybe some owners don't want to buy a new camera.  It isn't that Leica doesn't want to fix the sensors; the sensors aren't available anymore.  So you can use your M9 as is, sell it, junk it, trade in or fix.  For that matter, you can fix and turn it into a monochrome as well.

[Einst_Stein]

19 minutes ago, dllewellyn said:

Yes, correct.  Depends on the lens, but basically the camera becomes nearsighted if the glass is too thin without changing the sensor position.

Qualitatively, yes, you are right, but quantitatively, it's ignorable. 

[Einst_Stein]

24 minutes ago, dllewellyn said:

Well, there are 100,000+ M9 cameras out there.  Maybe some owners don't want to buy a new camera.  It isn't that Leica doesn't want to fix the sensors; the sensors aren't available anymore.  So you can use your M9 as is, sell it, junk it, trade in or fix.  For that matter, you can fix and turn it into a monochrome as well.

I agree the best choice in the past was to trade-up, I paid $800 to trade it for M240. But it's too late now. The cost is much higher than before. 

The best for now is to sell M9 to someone who really wants to convert it to monochrome. I think the monochrome has a very narrow interests group.

[rramesh]

It would be good if we can put up RAW files comparing (a) M9 second generation with (b) M9 first generation with replaced cover glass.

Note that rather than compare the raw thickness of glass (original vs replaced), we need to factor in the properties of the glass (original vs replaced). I believe the difference will be a negligible compromise for most Leica lenses (>21mm), to make it a reasonable option for those not willing to pay a high price for an upgrade to a M10.

If the performance is not up to scratch, wouldn't Kolari or other 3rd parties be able to use cover glass of the same thickness i.e. 0.8?

 

Edited September 14, 2020 by rramesh

[dllewellyn]

Let's also see documented evidence of the fixed cameras. I can show pictures of the damaged and fixed camera and give RAW example comparisons.  If someone is in the NY/NJ area, they are welcome to come and test the cameras.  I will be happy to provide lunch.

BYOL (Bring Your Own Lens).

Edited September 14, 2020 by dllewellyn

[dllewellyn]

As well, if they want to actually see the spectrometer measurements, look at the glass under a microscope, play with cameras side by side, I am totally open to it.  In fact I would enjoy it!  Just real users though.  Kolari and LifePixel can figure it out on their own.

[Advertisement]

Hallo,

Look here: Leica M

[compman10]

5 hours ago, dllewellyn said:

As well, if they want to actually see the spectrometer measurements, look at the glass under a microscope, play with cameras side by side, I am totally open to it.  In fact I would enjoy it!  Just real users though.  Kolari and LifePixel can figure it out on their own.

I would be curious to see a before and after comparison of ColorChecker chart.  It looks like the BG38 you replaced the S8612 with peaks similarly at 500nm, but has a higher transmittance at shorter and longer wavelengths.  I would also be curious to know if the meter is still accurate with more light coming in.  Finally, how did the leather re-attachment go?  Does it look as good as the original?