Leica True Sense Imaging, maker of CCD's for the M-E and M Monochrom, now ON Semiconductor


San Francisco
Makes wonder how much longer the sensors in the M-E and MM are going to be available. I can't imagine, based on how much they brought in last year, that much of their revenue is being generated from just those two models. I wouldn't be surprised if those sensors got cut from their product line, as part of a streamlining effort to go a long with the acquisition.

As an M-E owner I hope the CCD is still available, if I should ever need to have mine replaced.
I've seen companies bought out like this before, and the products continue to be manufactured. The 14MPixel sensor used in the Kodak SLR/N continued to be manufactured and available for years after the company changed hands, it was available even through DigiKey. CCD arrays continued to be used in the scientific market.


And guess who bought Cypress Imaging-



Cypress bought out Fillfactor, and engineers from Filfactor founded CMOSIS... Round and Round it goes...


The KAF-1603 is compatible with my 20 year old digital camera. It has not been used in a "Digital Camera" for a long time, but continued in scientific instruments. The KAF-18500 is the same family as the other 6.8um detectors in the available lineup, have been around for a long time. The KAF-18500 is probably higher volume than the others, but has equal performance. Probably what draws me to these cameras- I just prefer CCD's, been using them for a very long time.
Another article-


Personal Opinion- I think ON Semiconductor is a better home for the company. Platinum Equity is an investment firm, bought the Imaging Division from Kodak, showed it was viable. The Imaging Division is now part of a larger company that services the technical market. and they make microcontrollers... Gotta look at those datasheets.

Now we need to find out who is at the Top of ON and invite them to Leicaplace. Get em hooked on rangefinders.


Hall of Famer
Makes wonder how much longer the sensors in the M-E and MM are going to be available...
As an M-E owner I hope the CCD is still available, if I should ever need to have mine replaced.

Based on my past experiences, Leica will take care of you if that need should arise. I had a Digilux 2 with a bad sensor several years out of the already extended warranty period. I had neither thew invoice nor any warranty papers, and the sensor was in short supply. Leica gave me the choice of waiting up to 9 months for a sensor replacement or taking up their offer for a huge discount towards a new M8 (the current digital M at that time). The discount they offered was more than double what my D2 was worth at the time.
I just visited the Truesense site, after being acquired by ON semiconductor.



The KAF-18500, and many other detailed data sheets are now available for download. Kodak and Truesense did not make the datasheets for the KAF-18500 available before. I'm happy...


This is the first time the KAF-18500 has been offered for sale in the general list of products, rather than "exclusive" for Leica. Also note- the IR cover plate that is getting so much attention of late is made by Schott glass. Long story short- corrosion of the cover glass has been attributed to High humidity and repeated wet cleanings. This is not the first time that optical glass has shown problems, glass used in Canon lenses in the 50s and 60s was unusually susceptible to etching, and of course thoriated glass became yellowed from radiation. The 0.8mm cover glass used for the KAF-18500 is much more efficient in absorbing IR than the 0.5mm cover glass of the M8. The chemistry must have changed, different metal salts or concentrations of salts used. I'm guessing the new glass is more susceptible to liquids.

I do note- The KAF-10500 used in the M8 is not longer listed in the active line-up.
It is also strange that the MM sensor, supposedly the debayerized version of the KAF-18500, is also not listed...

Interesting that these past days, Leica has "pseudofficially", in the l-camera-forum, pointed the corrosion issues with the M9, MM and derivatives FF CCD and they have established a time frame limit in terms of cost for sensor swap (gone the "good will" changes)...
also they point that the M240 CMOS is not affected and that we should clean the sensor without touching it!!!...

As they textually say:

Based on this thread, we feel the need to clarifying a couple of things about the sensor marks issue you have been experiencing. The issue is linked with corrosion effects on the cover glass of the CCD sensor in Leica M9, M9-P, M Monochrom and M-E cameras. They manifest themselves as marks on images captured at smaller apertures (f/5.6-22). The new Leica M (Type 240) with the CMOS sensor is not affected by this problem. We are truly sorry for the inconvenience encountered and we have set up the following scheme for servicing the sensors of the products affected. Please be aware that a contact-free cleaning of the sensor is essential in preventing the issue.

• Customer care will perform sensor cleaning free of charge by prior arrangement.
• In the case of damage as a result of corrosion, the sensor will be replaced free of charge up to three years following the date of purchase. Leica Camera AG will cover the full costs of replacement, amounting to 1,800 euros plus applicable VAT. This does not apply to sensors damaged by scratching or breakage of the sensor glass.
• In the fourth and fifth year following the date of purchase, sensors damaged by the corrosion effects described will be replaced for a fixed charge of 600 euros plus applicable VAT. Leica Camera AG will cover the remaining costs of 1,200 euros.
• In the sixth and seventh year following the date of purchase, sensor replacement will be offered at a fixed charge of 1,200 euros plus applicable VAT. Leica Camera AG will cover the remaining costs of 600 euros.
• For the eighth and more years following the date of purchase, sensor replacement will be offered at a fixed charge of 1,500 euros plus applicable VAT. Leica Camera AG will cover the remaining costs of 300 euros.
• The prices stated apply for direct shipment of the camera to Leica Customer Care in Wetzlar or the Customer Care department of a national distributor. Additional costs may arise when the camera is sent to Leica through a dealer.
• Mandatory warranty conditions will apply after customers have taken advantage of the extended goodwill arrangement.
• As longer waiting times may otherwise occur, the camera should only be sent to Customer Care after prior arrangement.

We will continue watching this thread so you are welcome to react here.

^JJ with Leica-camera

For whom wants to go through all the post:

Kodak, Truesense, and now ON semiconductor typically combine the color and monochrome versions of the sensor in one datasheet. The sensors are the same, the dye added to the microlens layer is different. A KAF-18500 "Monochrome" is just a KAF-18500 "color" without the RGB Dye added to the microlens array. The "Bayer-Pattern Mosaic Filter" is made of RGB pixels in a specific pattern.

This post is also a good one, also from a Leica employee:


I wonder if the KAF-18500 can be converted for IR use, I suspect it can. It depends on how the cover glass was attached to the sensor. I would love an M-Monochrom for IR. "Usually" the IR absorbing glass is removable.
I found the datasheet for the cover glass used by the KAF-18500,


It is made by Schott. Apparently, it was also used in the Kodak SLR/n of 2004.


This filter is listed as "sealed" in the KAF-18500 datasheet. I'm going to venture a guess that some solvents will eat away at the sealing, allowing the underlying salts to react to moisture and liquid.

This glass is also sold for use in screw-in filters, here is a discussion on FLICKR.


SO- from everything that I've read about the problem on various forums, the problem is with the type of IR absorbing glass used in the KAF-18500. It is not the only sensor to use this Schott glass. I wonder how wide-spread, ie how many sensors really affected and under what situations. It sounds like Schott needs to issue cleaning instructions. I also wonder if the IR filter is annealed to the sensor, or if it can be replaced separately from the sensor. Usually this is the case.

Looks like it was cemented on the DCS SLR/n,


and you can see the problem is with the Schott glass filter, white-spots.

So- these IR filters have been in use for a long time, and by a lot of manufacurers. I've found articles mentioning it from the 90s. All of them are not suffering from white spots, some of them are. If they were all corroding from "just age", I'd expect to be reading about a general problem. My advice- avoid wet cleaners, use a rocket blower, send in for a pro cleaning at Leica when it's bad. Myself- I've used a wet cleaner one time on each of the M9 and M Monochrom when they were a few months old, and twice on the M8. I'm not too worried.
The M8 uses Kyocera B-7 glass. 0.5mm.

Stupid IR Absorbing glass. I told Kodak they should just leave it off the KAF-18500 when I called them in 2010. They remembered me from calling in 1993 and asking to leave off the IR filter on the DCS200. They listened then...
From Schott glass-


Two categories of IR absorbing glass, the first designed for severe conditions and the second designed for steep cutoff. Note that the first group is designed for 1000hrs of resistance in rough environments. The second group- you can see that S8612 provides steep cutoff for IR, but passes visible light efficiently. The "QE" for the M9 in Red is much higher than that of the M8.

Problems of corrosion in IR absorbing glass is due to the metal salts used in the chemistry of the glass, it is not unique to Schott. These groups of glass, including S8612, have been around for decades. If all S8612 glass corroded from age alone, there would be a lot of cameras with problems around. Problems will be more prevalent in areas of high humidity.
Having a look at the NIR graphs, certainly the M9 must have a much better QE with red light because at 600nm, the S8612 already has "only" 50% of transmittance and at the 650nm red light, just 10% ???...
From Scheneider Optics (B+W filters transmission curves, see bottom right red filters 090, 091):


I have not been able to locate the Kyocera B-7 Datasheet, but having a look at the Schott BK-7 (probably close) it has a lower refraction index and a higher abbe number... I really think that the M8 is a great machine showing a big amount of work to achieve the first digital M... and I am with you, get rid of those glass slices... ;-) :)
The long sheet for the KAF-10500 is off the website, I have it downloaded.


The "Short Sheet" shows a QE of 21% in Red.

I cannot find a datasheet for Kyocera B-7, "BUT" reading into the QE of the M8, the response looks like what you would get IF you used glass from "Schott Group 1". This is the type of glass that is more resilient to environmental damage, but does not have a sharp cutoff. I wonder of the Leica and Kodak engineers went for more durable glass for the M8. After the M8 was attacked for the IR contamination issue, they moved to high cutoff glass. All manufacturers are held to the same laws of chemistry, and IR absorption is well known.

A lot of cameras use S8612 glass. They typically have a low-pass filter in front of the sensor, the IR cover glass is not the first layer in the stack. It's not what you wet-clean. With more cameras getting rid of AA filters in the stack, some tough decisions will have to be made. Thick cuts of IR absorbing glass in themselves act as an AA filter, it has often been observed that the M8 produces a "crisper" image than does the M9.
"I've been there before for obsolete data sheets"....

I downloaded a number of data sheets going back 20 years. I strongly believe that the M8 cover glass is the type that is resilient to humidity, but is not as effective in absorbing IR.
Lifepixel does not work on Leica cameras. I suspect even if they could take the camera apart, getting the cemented IR cover glass off is the problem.

Latest threads

Top Bottom