Sensor Corrosion Analysis and Fix [Merged]

[BrianS]

Advertisement (gone after registration)

On 10/13/2020 at 7:23 PM, jaapv said:

Note that the Monochrom M9 uses Linear DNG whereas the M9 uses Raw DNG. The Linear DNG format can hold more non-image data. I do not know why Leica made this choice.

 

 

 

 

I ended up writing my own DNG processing code, including converting the color files from the M8 and M9 to monochrome and writing them in the "LinearRaw" format.  This requires demosaicing the image, eliminating DNG Tags for the color filter array and associated fields, changing the value of Tag 6 from 32803 to 34892.

Both file formats can use the same tags for storing information. They are both in the same DNG spec, just different values for Tag 6 "photometric interpretation" and then filling in the other tags as they make sense.

PHOTOMETRIC INTERPRETATION
Page 20 of the DNG 1.4.0 spec.
https://www.adobe.com/content/dam/acom/en/products/photoshop/pdfs/dng_spec_1.4.0.0.pdf

The DNG Spec uses Tag 6 to set the DNG format.
The Specification uses two terms for the Raw IFD,
A value of 32803(dec),  '8023'x (hex),  specifies CFA (Color Filter Array)
A value of 34892(dec),  '884C'x (hex), specifies LinearRaw.

CFA is used for the M9, LinearRaw is used for the M Monochrom.
The CFA format has Tag settings to demosaic the image, typically using a Bayer filter pattern.
More patterns are available.

LinearRaw does not have the information required to demosaic the image, which is fine for thr M Monochrom.
You would have to demosaic the image in the camera to store as LinearRaw.

Code listing and examples of the conversion here-

https://cameraderie.org/threads/experiments-in-leica-m8-and-m9-conversions.38737/page-2

 

I've seen some sites use the terms "Linear" DNG and "Raw" DNG- but the actual DNG specification uses "LinearRaw" and "CFA".

 

Edited October 17, 2020 by BrianS

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[adan]

On 10/13/2020 at 5:23 PM, jaapv said:

Note that the Monochrom M9 uses Linear DNG whereas the M9 uses Raw DNG. The Linear DNG format can hold more non-image data. I do not know why Leica made this choice.

Linear DNG is specifically for use with "unusual sensor configurations" - which would include ones with no Bayer pattern (i.e. Monochrom, also Foveon).

 

[dummilux]

On 9/16/2020 at 7:13 AM, DandA said:

Can someone clarify something (in simply terms) please.  What was the original thickness or the original non coated (corrosion prone) cover glass used in the M9's.  0.8mm?  Secondly, once Leica replaced the corroded cover glass on a M9 with a coated no corroding one, what was the thickness of this newly replaced coverglass?  I assume it was 0.8mm too, like the original?  If so, isn't it possible for these third parties now having the ability to replace M9 cover glass, to spec a 0.8mm one instead of a 0.6mm cover glass, so focus stack remains the same and no refocusing or changes to the focusing of the M9 takes place in any measurable way, even using ultra wide angle lenses?  Thanks!

 

Dave (D&A)

I emailed Kolari to ask if they could replace the glass with one that was the stock .80. This was their response:

 

"Hello Gerardo,

 

Thank you for reaching out. Unfortunately, we are only able to use the BG60 at this time.

 

We recalibrate the sensor position so that you are still able to achieve accurate focus after the conversion service.

 

Best regards,

 

William Martinez

Kolari Vision"

 

 

So, they do reposition the sensor.

 

I'm sending my M9p to Kolari this week.

 

 

[BrianS]

Using thinner cover glass means the offset microlens array will not be in their optimal position, and that there could be some slight error in the corrections applied for lenses.

The thinner cover glass does have advantages, as some have noted with the M8- which uses 0.5mm cover glass.

Edited October 21, 2020 by BrianS

[liu2k3L]

It may be hard for people here to believe that I just knew my M9 CCD has the corrosion issue 3 days ago. I came back to buy another battery. When the salesperson told me, I was totally shocked. I didn't buy it abroad. My local retailer just failed to inform me of the recall, they blamed it to the change of distributor years ago. And the only option available now is trade in program.

I am not an active user. As a first hand owner of M9, my shutter counts is only 18k. It is not reasonable for me to spend thousands of dollars to obtain a new M. For me now, I prefer to change my CCD. After contact Leica customer care, he replied me a standard answer .

If anyone here has any advise, I'll be appreciative .

thanks

[Raymondl]

3 hours ago, liu2k3L said:

It may be hard for people here to believe that I just knew my M9 CCD has the corrosion issue 3 days ago. I came back to buy another battery. When the salesperson told me, I was totally shocked. I didn't buy it abroad. My local retailer just failed to inform me of the recall, they blamed it to the change of distributor years ago. And the only option available now is trade in program.

I am not an active user. As a first hand owner of M9, my shutter counts is only 18k. It is not reasonable for me to spend thousands of dollars to obtain a new M. For me now, I prefer to change my CCD. After contact Leica customer care, he replied me a standard answer .

If anyone here has any advise, I'll be appreciative .

thanks

Unfortunately changing the CCD sensor is not a possible option, however you could buy another M9 with the latest sensor (in addition to many other options..)
 

In my situation I did a fair bit of analysis and decided to proceed with the “repair” option, I have written up my experiences and provided sample pictures on page 9. 

[Denys]

Advertisement (gone after registration)

6 hours ago, liu2k3L said:

...My local retailer just failed to inform me of the recall, they blamed it to the change of distributor years ago. And the only option available now is trade in program...

Hello and welcome to The Forum; it’s a pity that your introduction isn’t a happy one. There are a few options other than the official trade-in one..  ..I hope you find something that suits you. Just a small point, but an important one I feel; there was no ‘recall’ such as found with faulty cars, etc. Leica was made aware of the corrosion problem then offered a free replacement service for cameras affected.. ..then it offered a subsidised replacement service for cameras affected. Only when the availability of new sensors dried up was the replacement service stopped. It was always up to the owner of the camera to identify a corroded sensor and then instigate a replacement.

Good luck!

Edited October 22, 2020 by Denys

[liu2k3L]

to Raymondl :thank you for the useful information

to Denys: Maybe "recall"  was not a correct term to describe Leica's way of solving this. Actually Leica officially deny it as recall. Anyway,  they failed to inform me of flaw product (which is not prevalent among all CCDs) . In addition, they claimed that there were informations on their web page, forums, implied it as the customer's fault not to beware of it. That disappointed me. That may explain why I don't want to trade in and consider to sell some of my M lenses.

[BrianS]

Three companies are now offering service to replace the corroded glass of the sensor, all for less than what Leica was charging. 

[kristyansen]

Am 22.10.2020 um 03:42 schrieb liu2k3L:

It may be hard for people here to believe that I just knew my M9 CCD has the corrosion issue 3 days ago. I came back to buy another battery. When the salesperson told me, I was totally shocked. I didn't buy it abroad. My local retailer just failed to inform me of the recall, they blamed it to the change of distributor years ago. And the only option available now is trade in program.

I am not an active user. As a first hand owner of M9, my shutter counts is only 18k. It is not reasonable for me to spend thousands of dollars to obtain a new M. For me now, I prefer to change my CCD. After contact Leica customer care, he replied me a standard answer .

If anyone here has any advise, I'll be appreciative .

thanks

vor 2 Stunden schrieb liu2k3L:

to Raymondl :thank you for the useful information

to Denys: Maybe "recall"  was not a correct term to describe Leica's way of solving this. Actually Leica officially deny it as recall. Anyway,  they failed to inform me of flaw product (which is not prevalent among all CCDs) . In addition, they claimed that there were informations on their web page, forums, implied it as the customer's fault not to beware of it. That disappointed me. That may explain why I don't want to trade in and consider to sell some of my M lenses.

Hi, sorry to hear. If you read through the thread you will see you are not alone (me included).

There are three options if you want to continue owning an M9:

1. If the corrosion is weak you can still use your M9 for a while until it worsens (wide open shooting does not show the corrosion). This is my case for now, but I confess there is no joy anymore in using the camera like this - so I will go for another option.

2. Update your sensor glass through one of the two (non Leica) options from the US to change your sensor glass: KolariVision or MaxMax. Price USD 1000 Or USD 1500 (+ shipping and if you do not live in the US to pay customs on the repair service). I will go for this scenario.

3. Sell your M9 for a low price to somebody who wants to upgrade to a new model. They will save around USD 2000 and you may even share the savings. With what you gain you can purchase a used M9 with a Leica replaced sensor. It should cost you USD 2000+. 
 

Hope this helps to go from frustration to action.

[kristyansen]

vor 5 Minuten schrieb BrianS:

Three companies are now offering service to replace the corroded glass of the sensor, all for less than what Leica was charging. 

We know Kolari Vision, MaxMax. What other company offers this service? 

[Jadon Rosado]

Hello!

My name is Jadon Rosado, I’m the third source for M9 sensor repairs. I have a lot more info on this Thread. The glass I use is BG40 at 1mm, the difference is the closest you can get without disturbing the micro lenses. The method I’m using is the least invasive as it preserves most aspects of the sensor. The only major change is increased transmittance across the entire visual spectrum, this causes colors to be slightly more vivid and increases shadow detail. I’ll post some sample photos below. 
 

i was also informed by Brian that some users are interested in just installing a new s8612 filter, this will have a chance of corroding again but it got me thinking; has schott improved the AR coating on the the s8612? Maybe there have been improvements made to the glass that reduce the chances of corrosion (not eliminate it) I’m certainly open to trying it if I can find the glass at the proper .8mm thickness. The glass can’t be thicker or thinner as it would change the transmittance curve. 
 

You can inquire via email JadonRosadoBroberg@gmail.com

 

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

[dummilux]

20 hours ago, Jadon Rosado said:

Hello!

My name is Jadon Rosado, I’m the third source for M9 sensor repairs. I have a lot more info on this Thread. The glass I use is BG40 at 1mm, the difference is the closest you can get without disturbing the micro lenses. The method I’m using is the least invasive as it preserves most aspects of the sensor. The only major change is increased transmittance across the entire visual spectrum, this causes colors to be slightly more vivid and increases shadow detail. I’ll post some sample photos below. 
 

i was also informed by Brian that some users are interested in just installing a new s8612 filter, this will have a chance of corroding again but it got me thinking; has schott improved the AR coating on the the s8612? Maybe there have been improvements made to the glass that reduce the chances of corrosion (not eliminate it) I’m certainly open to trying it if I can find the glass at the proper .8mm thickness. The glass can’t be thicker or thinner as it would change the transmittance curve. 
 

You can inquire via email JadonRosadoBroberg@gmail.com

 

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

Hi Jadon, great work. I read your post on the other site. 
 

What is the advantage of your method over Kolari’s?

[Jadon Rosado]

2 hours ago, dummilux said:

Hi Jadon, great work. I read your post on the other site. 
 

What is the advantage of your method over Kolari’s?

Thank you!

the main advantages of my method is that it requires the least amount of calibration prior to installation, I only have to shim the sensor .1mm while kolari has to shim it .5. It may not seem like a lot but shimming the sensor that much will throw the micro lens array way out of tolerance making the the lens presets useless. The glass I’m using is the closest I was able to find to the s8612 that didn’t compromise performance. The BG40 glass has slightly higher transmittance across the entire visual spectrum making all colors slightly more vibrant while also bringing out the shadows better. In the test photo I metered for the highlights on the hood of my car just to give an example of the improved shadow detail. I’ll include another photo from a portrait session just to show off the colors a better, this was shot with a 90mm summicron v1. 
 

 

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

[dummilux]

7 minutes ago, Jadon Rosado said:

Thank you!

the main advantages of my method is that it requires the least amount of calibration prior to installation, I only have to shim the sensor .1mm while kolari has to shim it .5. It may not seem like a lot but shimming the sensor that much will throw the micro lens array way out of tolerance making the the lens presets useless. The glass I’m using is the closest I was able to find to the s8612 that didn’t compromise performance. The BG40 glass has slightly higher transmittance across the entire visual spectrum making all colors slightly more vibrant while also bringing out the shadows better. In the test photo I metered for the highlights on the hood of my car just to give an example of the improved shadow detail. I’ll include another photo from a portrait session just to show off the colors a better, this was shot with a 90mm summicron v1. 
 

 

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

Thanks for your thorough reply and for offering this service. I’m under the understanding that Kolari is only shimming .2mm. They also confirmed that they would reposition the sensor. From their site:

“We were able to delaminate the corroded filter from the sensor, and replace it with BG60 glass. This glass will last indefinitely without oxidizing per Schott test reports, and also allows us to reduce the Leica sensor stack by 0.2mm...”

I guess in an ideal world, no shimming would be necessary. A S8612 with better coating would be optimal. This is the option I would go for if it was available. 
 

When will you start accepting cameras for repair? 

I was planning on sending mine to Kolari this week but I’m still going back and forth between options. Maybe I should just wait it out.

[kristyansen]

vor 40 Minuten schrieb Jadon Rosado:

Thank you!

the main advantages of my method is that it requires the least amount of calibration prior to installation, I only have to shim the sensor .1mm while kolari has to shim it .5. It may not seem like a lot but shimming the sensor that much will throw the micro lens array way out of tolerance making the the lens presets useless. The glass I’m using is the closest I was able to find to the s8612 that didn’t compromise performance. The BG40 glass has slightly higher transmittance across the entire visual spectrum making all colors slightly more vibrant while also bringing out the shadows better. In the test photo I metered for the highlights on the hood of my car just to give an example of the improved shadow detail. I’ll include another photo from a portrait session just to show off the colors a better, this was shot with a 90mm summicron v1. 
 

 

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

Dear Jadon, first of all, welcome to the site.

Second, delighted to see your technical, science skills as well as your perseverance resulting in this very attractive offer that Leica M9 owners have now at their disposal for the Leica M9 sensor glass repair!

[Jadon Rosado]

31 minutes ago, dummilux said:

Thanks for your thorough reply and for offering this service. I’m under the understanding that Kolari is only shimming .2mm. They also confirmed that they would reposition the sensor. From their site:

“We were able to delaminate the corroded filter from the sensor, and replace it with BG60 glass. This glass will last indefinitely without oxidizing per Schott test reports, and also allows us to reduce the Leica sensor stack by 0.2mm...”

I guess in an ideal world, no shimming would be necessary. A S8612 with better coating would be optimal. This is the option I would go for if it was available. 
 

When will you start accepting cameras for repair? 

I was planning on sending mine to Kolari this week but I’m still going back and forth between options. Maybe I should just wait it out.

Strange I swore I read somewhere that they were using a .5mm filter on their sensors... At .2mm that brings it very close to mine. My bad

They are still using the Bg60 glass which has a similar curve to the s8612 but much lower transmittance of roughly 92~93% the s8612 and BG-40 (what I’m using) is 98%. Using the BG60 will cause under exposing and the dynamic range would take a hit. I believe Leica used the 60 or something similar for their repair but it requires new firmware to make the camera compensate for the underexposure.

I’ve read some thing about schott improving their coatings a few years ago, it’s a long shot but maybe they improved the Coating on the s8612 but no ones ever bothered to try it out... or maybe applying a very thin Mylar wrap (similar to how the Sony pelical mirrors are) to the s8612 to completely seal it off, this would barely change the thickness but it would be a lot more prone to scratching. Just food for thought.

I’m working on getting my business set up, for the time being I am accepting a limited amount of repairs, I’ve yet to have a sensor repair come in but the Film Ms have been keeping me busy!

I should also mention my repair is only $700 so it’s definitely a better deal if you like the new rendering, I understand that some will prefer the original better but for the time being this is the most reliable solution I have to offer
 

Edited October 25, 2020 by Jadon Rosado

[kristyansen]

vor 2 Minuten schrieb Jadon Rosado:

Strange I swore I read somewhere that they were using a .5mm filter on their sensors... At .2mm that brings it very close to mine. My bad

They are still using the Bg60 glass which has a similar curve to the s8612 but much lower transmittance of roughly 92~93% the s8612 and BG-40 (what I’m using) is 98%. Using the BG60 will cause under exposing and the dynamic range would take a hit. I believe Leica used the 60 or something similar for their repair but it requires new firmware to make the camera compensate for the underexposure.

I’ve read some thing about schott improving their coatings a few years ago, it’s a long shot but maybe they improved the Coating on the s8612 but no ones ever bothered to try it out... or maybe applying a very thin Mylar wrap (similar to how the Sony pelical mirrors are) to the s8612 to completely seal it off, this would barely change the thickness but it would be a lot more prone to scratching. Just food for thought.

I’m working on getting my business set up, for the time being I am accepting a limited amount of repairs, I’ve yet to have a sensor repair come in but the Film Ms have been keeping me busy!

 

Hi Jadon, I would also really want to see a S8612 option, for me this is why I have the M9 (Leica used BG55 to replace the original sensors, but changed the sensor basically).

Without knowing if Schott made changes to the glass, therefore assuming that S8612 would sooner or later corrode after some years, it would be good to understand: 

1. Can S8612 still be sourced by you?

2. Are there better versions of S8612 less prone to corrode? 

3. If corrosion is still an issue, are there ways to add the glass on the sensor, so that can be easily re-serviced in a few years?

4. What would be a much more convenient price for the potentially needed re-services than the original service?

Thanks for any thoughts here. 

 

[Advertisement]

Hallo,

Look here: Leica M

[Jadon Rosado]

4 hours ago, kristyansen said:

Hi Jadon, I would also really want to see a S8612 option, for me this is why I have the M9 (Leica used BG55 to replace the original sensors, but changed the sensor basically).

Without knowing if Schott made changes to the glass, therefore assuming that S8612 would sooner or later corrode after some years, it would be good to understand: 

1. Can S8612 still be sourced by you?

2. Are there better versions of S8612 less prone to corrode? 

3. If corrosion is still an issue, are there ways to add the glass on the sensor, so that can be easily re-serviced in a few years?

4. What would be a much more convenient price for the potentially needed re-services than the original service?

Thanks for any thoughts here. 

 

Those are all very interesting questions, I’d need to do some research and if I can reliably source the .8mm s8612. all schott BG glass is phosphate based glass just like the S8612 so really any of these can corrode if the coating is compromised. (Leica BG55, kolari BG60, RDR BG40) 

it leads me to believe that at some point in time schott imporved their AR coatings, whether or not that carried over to the S8612 I don’t know but it would make a lot of sense for them to use a newer coating as to keep the glass up to date. 

a potential solution to the issue may be to use a thin sheet of clear Mylar to add another protective layer on the glass, this would definitely prevent the glass from corroding again but Mylar is very prone to scratching so sensor cleaning would have to be done by a pro. The application of the Mylar to the s8612 would definitely have to be done in a clean room for it to work, the smallest spec of dust would ruin it. 
 

the setup I currently have allows me to replace the glass with little issues, but worst case I can always remove the glass to remove any debris that snuck in. 
 

if I can source the proper S8612 I would do the initial installation for $700 as I would the BG40. Removing the original glass is the hardest part hence the high price. I use a less extreme adhesive to attach the new glass so it can be removed later for potential Full spectrum converting... if the s8612 began to corrode within a few years I would probably just charge the same as a M series CLA so $250.

I should clarify, this improved S8612 is merely a hypothesis. I’ve never considered the s8612 as an option so I’d just have to try it out and hope for the best. Testing will be tricky as I live in Oregon, it’s a very dry climate though we do get a lot of rain. The best way to test a replacement s8612 would be in a more humid environment as it would give us any adverse effects sooner. 
 

I rambled on for a bit but I hope this information is useful,

Jadon

[DandA]

Jason, welcome and thanks for all your posting and information.  You mentioned that Leica got around the underexposure issue by issuing new firmware for those cameras in which they replaced with the latest sensor.  If that's the case, if someone wanted you to replace the coverglass in a M9 that has the latest sensor to obtain your replacement BG40, would they then have to find a way to revert to the firmware revision that was just prior to the latest installed by Leica upon replacing the camera with the latest sensor?

 

Dave (D&A)