k-hawinkler Posted August 31, 2012 Author Share #41 Posted August 31, 2012 Advertisement (gone after registration) K-H, Well, interesting further pics and test you posted. Sounds like you were up all night waiting for take the weather in order to take the images. Welcome to the world of dedicated (is that spelled "c-r-a-z-y") observational astronomy Well, it's not that different from supercomputing in the old days, I mean staying up all night to get the CPU cycles for your work. Actually, my daughter and I were on a very pleasant midnight walk when we noticed that the area around the moon would be clear by the time we finished our loop. So, I took advantage of that. Yes, I agree that the filter is not part of the ghost reflection mix (after seeing these results). And, certainly that flat surface at the input is a prime candidate for part of the issue. But, there still must be a second surface toward the rear of the lens which would serve as the first reflection. That front, flat element alone can not do what you see. Another surface is involved. And, it must be some distance away from the first in order for you to see the ghost image separation you see. I did look at the lens element layout again and I see that the last lens element (not the flat filter) does have a relative flat surface. I wonder if the first reflection could be from that? You should be able to inspect that surface for cleanliness and should even be able to clean it. I wonder if that would make a difference. Of course, I agree, there has to be a second (flat or nearly flat) surface to be involved in order for the observed effect to happen. You point out one candidate. Let me ask, could another candidate also be the flat digital sensor in the NEX-7? When I read Puts descriptions of these lenses, he mentions that that the AR coatings were intentionally light because of the high curvature on some lens elements and the need to reduce any AR coating influence on aberations. Maybe that has something to do with all this. One thing that does puzzle me on the entire design is why there are these flat elements at all. Maybe someone can shed light of this design feature. Is it just for protection of the front element? An optional front filter would have served this purpose. I can see including the rear filter if one intends to always have the filter option available. But, I don't understand the motivation for the front flat element. RM Post #4 of this thread, lars_bergquist, quote: "The Apo-Telyt has a permanently mounted plane protective filter in front of the actual first element, which is said to be quite soft." Quote Link to post Share on other sites More sharing options...
Advertisement Posted August 31, 2012 Posted August 31, 2012 Hi k-hawinkler, Take a look here LEICA SERIE 5.5 NDx1 13026 GERMANY. I'm sure you'll find what you were looking for!
k-hawinkler Posted August 31, 2012 Author Share #42 Posted August 31, 2012 Hello Everybody, Perhaps the elements in the front & back inside of these flat protectors are Fluorite or some other more or less equally soft material that require protective elements. Remember: The 180 Summicron has the same types of very flat, thinly coated both front & back protective filters. Perhaps in both lenses the decision was made to sacrifice a small portion of the available image quality in order to guarantee the longevity of the lens elements. Most likely these would be reasonably thin single or possibly 2 layer coated (not counting transparent spacing, etc layers). Probably not multi-coated (green). K-H: What color/s are the surfaces that can be seen on each side? That is: What you can see from the front as well as both sides of the back. Best Regards, Michael Hi Michael, Thanks. Well, I claim no particular expertise when looking at glass. So, with that proviso, the glass or dust, when shining a white LED light through it, appears to me to have a reddish almost purple tint. When I bought the lens 8 months ago it came with a one year Leica warranty. So, provided I can dig out that piece of paper, I could have the lens inspected and if necessary CLA'ed by Leica. Otherwise Don Goldberg comes to mind. Is there anybody else out there with that lens who would be willing to help figure out exactly what's going on? Quote Link to post Share on other sites More sharing options...
Michael Geschlecht Posted August 31, 2012 Share #43 Posted August 31, 2012 Hello K-H, I'm pretty sure the filters are to protect the soft elements. Why don't you just call Leica in Solms & ask them. These 2 lenses, 180/2 & 280/4, are 2 of the best of their types made by anybody. Leica would probably be willing to talk to you about them. I don't think there is anthing wrong w/ the lens or the filters. It probably wouldn't hurt to call them & ask & I'm sure all of us would like to know what they say. btw: Given those coatings are somewhat thin they might be somewhat fragile. Even if they are harder than Fluorite. Best Regards, Michael 2 Quote Link to post Share on other sites More sharing options...
jaapv Posted August 31, 2012 Share #44 Posted August 31, 2012 At the time i scratched the front element. Leica told me it was put there because Leica felt the "real" front element was too expensive to be exposed. I find that quite believable, as the repair even with this "cheap" element was horrifyingly expensive. 2 Quote Link to post Share on other sites More sharing options...
lars_bergquist Posted August 31, 2012 Share #45 Posted August 31, 2012 Lars, I am afraid we should not quote Mariana Heyerdahl as a Swedish sculptor. She is born in Oslo but grew up in Italy. More about her "Terracotta Woman" can be found: Terracotta Woman Marian Heyerdahl, Terracotta Warriors, Army for peace, peace army terracotta, Terracotta women /Eric - from Sweden I stand corrected. LB Quote Link to post Share on other sites More sharing options...
Robert_M Posted August 31, 2012 Share #46 Posted August 31, 2012 K-H, It occurred to me last night, after my last post, that the first reflection probably comes from the sensor plane surface. That would explain why the ghost image appears in focus and the same size. I see you tuned into that before I could post again... So, 1) A strong light at infinity enters the entire front aperture of the lens at that flat protective plate as a parallel beam. 2) When the light source is off-axis, the image comes onto the focal plane at an angle. 3) The front sensor surface is not perfectly transmissive and absorptive. In fact, I'd suspect that it is less efficient than the rest of the lens surfaces. 4) Some of the light is reflected back out at an angle symmetric to the incoming image angle. 5) The image was at focus at the sensor and is now diverging back through the optical chain. But, it is being perfectly focused back to a parallel beam at the front aperture. 6) That ghost image is now reflected back from the front plane surface in the same manner but at a much lower value. And, it is re-focused back to the image plane. Can you tell from the pixel count values the intensity ratio between the ghost image and the primary image? At least 30 dB down, I bet. I suspect that the color you see in the ghost image is due to the Bayer sensor mask. There will be a different reflection and absorption efficiency for each of the colors. Have you tried these type of images with only the Sony NEX-7 camera and this lens? I wonder if the results are slightly different with a different camera sensor. I doubt that there is really any way of eliminating it completely except by using the circular polarizing filter arrangement I mentioned before. And, cutting out so much of the light would then defeat the purpose of night photography. BTW, one should see this effect in any lens system and get some ghost reflections off the various optical element after reflection back from the sensor. But in those cases, the lens surfaces are curved and not in tune with the optical focusing. So the light is spread out and effectively just reduces the contrast of the entire image. This lens is unique with that front flat element. It provides the means for that in focus ghost image. RM 1 Quote Link to post Share on other sites More sharing options...
k-hawinkler Posted August 31, 2012 Author Share #47 Posted August 31, 2012 Advertisement (gone after registration) Hi RM, Many thanks and congratulations! I think you may have figured out what's going on. I'll respond more fully as soon as I have written down my thoughts to your last post. Quote Link to post Share on other sites More sharing options...
k-hawinkler Posted August 31, 2012 Author Share #48 Posted August 31, 2012 (edited) K-H, It occurred to me last night, after my last post, that the first reflection probably comes from the sensor plane surface. That would explain why the ghost image appears in focus and the same size. I see you tuned into that before I could post again... Yup, that sensor plane surface was my first hunch. Rational: My APO lens is from 1993 and presumably was developed when Leica was still deeply embedded in film photography, remembering the first digital camera I bought was Apple's QuickTake 100 camera (February 17 , 1994) History of the Digital Camera. Film doesn't reflect that strongly. How else to explain the fact that Leica missed such a strong effect when their lens is being used with a digital sensor? I am convinced, if Leica had tested this lens with a digital sensor/camera, they would have noticed something weird going on. According to Leica Camera AG - Culture - Leica Products, quote: "1998 also saw the launch of the first digital camera - the Leica Digilux." Leica came to digital photography rather late. So, 1) A strong light at infinity enters the entire front aperture of the lens at that flat protective plate as a parallel beam. 2) When the light source is off-axis, the image comes onto the focal plane at an angle. 3) The front sensor surface is not perfectly transmissive and absorptive. In fact, I'd suspect that it is less efficient than the rest of the lens surfaces. 4) Some of the light is reflected back out at an angle symmetric to the incoming image angle. 5) The image was at focus at the sensor and is now diverging back through the optical chain. But, it is being perfectly focused back to a parallel beam at the front aperture. 6) That ghost image is now reflected back from the front plane surface in the same manner but at a much lower value. And, it is re-focused back to the image plane. Yup, many thanks for the lucid physics explanation. I agree, very likely that's what's going on. Can you tell from the pixel count values the intensity ratio between the ghost image and the primary image? At least 30 dB down, I bet. Thanks for the suggestion. I will have to look at this carefully. However, with Dropbox I could make the raw and JPG files available to you, so you can check for yourself. Please, let me know if you want to pursue this. Thanks. I suspect that the color you see in the ghost image is due to the Bayer sensor mask. There will be a different reflection and absorption efficiency for each of the colors. Thanks for this key insight. The greenish color of the reflections from particularly strong light sources is so striking, also in the images of Black Mesa at night, earlier in this thread. Your explanation takes this important clue fully into account. Occam's razor ... Occam's razor - Wikipedia, the free encyclopedia Have you tried these type of images with only the Sony NEX-7 camera and this lens? I wonder if the results are slightly different with a different camera sensor. I doubt that there is really any way of eliminating it completely except by using the circular polarizing filter arrangement I mentioned before. And, cutting out so much of the light would then defeat the purpose of night photography. Yes, only with this combination of the moon - too short on time so far, but I have been thinking about it (however, I used the Nikon D800E + Leitz Telyt 280 mm f/4.8 in post #6 http://www.l-camera-forum.com/leica-forum/2160708-post6.html of the Black Mesa neighborhood). That's why I asked if anybody else out there who has this lens would like to participate in this exercise. Of course, if somebody with a film and digital back for a Leica R camera and the lens could photograph some extremely strong point light sources, that would be the ultimate test. I could use my NEX-5N as well. In order to use a Nikon camera, I would have to leitax the lens first. Not on aption I would consider at this time. I could also try my M9 with R to M adapter and focus bracketing via a tethered laptop. An M10 with LiveView would be a great tool for this problem as well BTW, one should see this effect in any lens system and get some ghost reflections off the various optical element after reflection back from the sensor. But in those cases, the lens surfaces are curved and not in tune with the optical focusing. So the light is spread out and effectively just reduces the contrast of the entire image. This lens is unique with that front flat element. It provides the means for that in focus ghost image. RM Yup, I have been thinking to use various other long Leica/Leitz lenses and check, that in the absence of the characteristic flat front plate, the described effect doesn't happen, certainly not as strongly. Aside from the APO-R lens, my other long Leitz lenses are of the V-type. Edited August 31, 2012 by k-hawinkler Quote Link to post Share on other sites More sharing options...
Robert_M Posted August 31, 2012 Share #49 Posted August 31, 2012 I guess one could simulate the same effect with another long lens and putting a flat filter in front. If the weather clears at night this weekend, I'll go try this. It was a full moon last night, so it is a good time for this experiment. RM 2 Quote Link to post Share on other sites More sharing options...
k-hawinkler Posted September 1, 2012 Author Share #50 Posted September 1, 2012 I guess one could simulate the same effect with another long lens and putting a flat filter in front. If the weather clears at night this weekend, I'll go try this. It was a full moon last night, so it is a good time for this experiment. RM Hi RM, Excellent idea, I'll try the same first with Sony NEX-7 + Leitz Telyt-V, version 3, 280 mm f4.8 that didn't give flares/ghosting in http://www.l-camera-forum.com/leica-forum/2162789-post10.html. So, in order to get best focus peaking in the NEX-7, I set under Menu, Creative Style: Contrast = -3 Saturation = 0 Sharpness = +3 That's the parameters I have used so far. When on the tripod, I use manual operation with ISO=100, exposure time=1/100 s, and f=11 to get good focus, off-axis. Then I only vary f from fully closed down to completely open while the moon slowly drifts through the viewfinder, shooting both raw and jpg. In post-processing I use the raw file and set Clarity=50 and Exposure to somewhere between +2 and +5 when processing the most overexposed image in CS6. The moon is already up, but there are some clouds around. Good luck with your experiments. Quote Link to post Share on other sites More sharing options...
Michael Geschlecht Posted September 1, 2012 Share #51 Posted September 1, 2012 (edited) Hello Everybody, Actually not that unique. While Leitz/Leica may only have a few lenses w/ flat exterior front or rear elements there is no shortage of lenses w/ internal flat elements, ie: 1969 & 1979 50mm Summicrons both have a # of internal flat surfaces. As to other makers who produce flat front & back elements: Makers of mirror lenses & Makers of diving cameras have produced front & rear elements w/ flat surfaces on a regular basis for a long time. Best Regards, Michael Edited September 1, 2012 by Michael Geschlecht 1 Quote Link to post Share on other sites More sharing options...
Michael Geschlecht Posted September 1, 2012 Share #52 Posted September 1, 2012 Hello K-H, What happens if you use the Lunar 16 rule: Exposure = 1/ISO @ F16 w/ a full Moon? Best Regards, Michael 1 Quote Link to post Share on other sites More sharing options...
k-hawinkler Posted September 1, 2012 Author Share #53 Posted September 1, 2012 Hello K-H, What happens if you use the Lunar 16 rule: Exposure = 1/ISO @ F16 w/ a full Moon? Best Regards, Michael You get something like the last two images in http://www.l-camera-forum.com/leica-forum/2165068-post36.html. Recall where I noticed the problem first? http://www.l-camera-forum.com/leica-forum/2160547-post3.html That was pretty much a close to correct exposure. I used the moon shots to make minute differences visible to illuminate the problem and said so in http://www.l-camera-forum.com/leica-forum/2163770-post14.html 1 Quote Link to post Share on other sites More sharing options...
k-hawinkler Posted September 1, 2012 Author Share #54 Posted September 1, 2012 (edited) Hi RM, Excellent idea, I'll try the same first with Sony NEX-7 + Leitz Telyt-V, version 3, 280 mm f4.8 that didn't give flares/ghosting in http://www.l-camera-forum.com/leica-forum/2162789-post10.html. So, in order to get best focus peaking in the NEX-7, I set under Menu, Creative Style: Contrast = -3 Saturation = 0 Sharpness = +3 That's the parameters I have used so far. When on the tripod, I use manual operation with ISO=100, exposure time=1/100 s, and f=11 to get good focus, off-axis. Then I only vary f from fully closed down to completely open while the moon slowly drifts through the viewfinder, shooting both raw and jpg. In post-processing I use the raw file and set Clarity=50 and Exposure to somewhere between +2 and +5 when processing the most overexposed image in CS6. The moon is already up, but there are some clouds around. Good luck with your experiments. So, here are my results for Sony NEX-7 + Leitz Telyt-V, version 3, 280 mm f4.8 for ISO=100, exposure time=1/100 s, and f=4.8, off-axis. In CS6 Clarity=50 and Exposure=+5 First case, no filter 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! Then, with filter LEITZ WETZLAR GERMANY SERIE 8 UVa, 14169 CANADA VIII My take: This seems to nail it. At a minimum, this makes a pretty convincing case for the mechanism described previously in this thread. Edited September 1, 2012 by k-hawinkler Quote Link to post Share on other sites Simply register for free here – We are always happy to welcome new members! Then, with filter LEITZ WETZLAR GERMANY SERIE 8 UVa, 14169 CANADA VIII My take: This seems to nail it. At a minimum, this makes a pretty convincing case for the mechanism described previously in this thread. ' data-webShareUrl='https://www.l-camera-forum.com/topic/186459-leica-serie-55-ndx1-13026-germany/?do=findComment&comment=2101699'>More sharing options...
k-hawinkler Posted September 1, 2012 Author Share #55 Posted September 1, 2012 (edited) Now that we have reason to believe that the failure mode of the LEICA APO-TELYT-R 1:4/280 in photographing night scenes with intense specular highlights entails both a digital sensor as well as the flat frontal protective plate of the lens itself, my question is what could be done to improve the performance of the lens in these extreme circumstances without modifying the lens proper or the sensor. That leaves the insert filter. Ideally one would need a filter that is completely transmissive on one side and completely absorptive on the other, or as close as possible to 100% efficiency. As I have absolutely no experience in optical properties of materials, I would like to pose the following questions: • Do technologies already exist that could be used to develop such a filter? - It not, could they be developed? What would that require? - If yes, what are those technologies? What would it take to develop a commercially available filter with the desired properties? Filters with these properties - if achievable at acceptable cost - certainly would give lens designers new options in the digital age. Edited September 1, 2012 by k-hawinkler Quote Link to post Share on other sites More sharing options...
Robert_M Posted September 1, 2012 Share #56 Posted September 1, 2012 K-H, Nice test. You have confirmed the explanation of what is going on. It was cloudy here last night, so I did not attempt the same. We are still in monsoon storms through the weekend. You ask for a cure to the problem. I believe that a Cir-P filter in the back S5.5 tray of that lens will cut out most of the problem. Of course, these filter do not seem to be inexpensive. I think the part number is 13338. With the Cir-P filter in the correct way around, the input strong light will be first converted to one polarization of linear and then come out the back side of the filter as circular polarization. Upon reflection off the sensor plane, the circular polarization is reversed. When it then hits the back side of the S5.5, it will not be transmitted. That light will either be reflected there or absorbed. So, the reflected, focused image of the Moon no longer gets to that front flat element of the lens. Any reflected light off the S5.5 will come back to the sensor out of focus. There may still be some "ghost", but it will not be in focus the same. If you do not already have that filter, it is pricey to try this experiment. I have seen one of those 13338 filters on offer (current) for 180 euro plus 30 euro shipping to US. This is not an item that I look for regularly so I don't know what the fair price is. A less expensive way to verify this scheme is with the 400 f/6.8 lens which you and I both have. Use a flat, UVa (or other) filter at the front of the 400mm lens to simulate the flat element of the 280 APO/4 lens. Then use a series 7 cir-P filter in the filter tray of the 400/6.8 lens to simulate the 13338 filter of the APO 280 lens. I'll see if I can give this test a try. Of course, the downside of this scheme is that you will loose about 2.5 stops of light for you night pictures. RM 1 Quote Link to post Share on other sites More sharing options...
Robert_M Posted September 2, 2012 Share #57 Posted September 2, 2012 K-H, Yet another suggestion to try for a "cure". And this one doesn't cost anything Try stopping down the aperture to f/8 or f/11. It may be that the reflection from the sensor plane is off axis enough (and at the wrong angles) that it will not make it back through a reduced aperture stop. This is sort of like a "Lyot stop" in reverse. I'm sure that a rigorous optical designer will correct my use of the term, but it is the same idea. RM 1 Quote Link to post Share on other sites More sharing options...
jaapv Posted September 2, 2012 Share #58 Posted September 2, 2012 The cheap way to improvise a 5.5 filter is to use a 39mm filter glass without the rim.It will rattle, but it will work. K-H, Nice test. You have confirmed the explanation of what is going on. It was cloudy here last night, so I did not attempt the same. We are still in monsoon storms through the weekend. You ask for a cure to the problem. I believe that a Cir-P filter in the back S5.5 tray of that lens will cut out most of the problem. Of course, these filter do not seem to be inexpensive. I think the part number is 13338. With the Cir-P filter in the correct way around, the input strong light will be first converted to one polarization of linear and then come out the back side of the filter as circular polarization. Upon reflection off the sensor plane, the circular polarization is reversed. When it then hits the back side of the S5.5, it will not be transmitted. That light will either be reflected there or absorbed. So, the reflected, focused image of the Moon no longer gets to that front flat element of the lens. Any reflected light off the S5.5 will come back to the sensor out of focus. There may still be some "ghost", but it will not be in focus the same. If you do not already have that filter, it is pricey to try this experiment. I have seen one of those 13338 filters on offer (current) for 180 euro plus 30 euro shipping to US. This is not an item that I look for regularly so I don't know what the fair price is. A less expensive way to verify this scheme is with the 400 f/6.8 lens which you and I both have. Use a flat, UVa (or other) filter at the front of the 400mm lens to simulate the flat element of the 280 APO/4 lens. Then use a series 7 cir-P filter in the filter tray of the 400/6.8 lens to simulate the 13338 filter of the APO 280 lens. I'll see if I can give this test a try. Of course, the downside of this scheme is that you will loose about 2.5 stops of light for you night pictures. RM 1 Quote Link to post Share on other sites More sharing options...
k-hawinkler Posted September 2, 2012 Author Share #59 Posted September 2, 2012 K-H, Yet another suggestion to try for a "cure". And this one doesn't cost anything Try stopping down the aperture to f/8 or f/11. It may be that the reflection from the sensor plane is off axis enough (and at the wrong angles) that it will not make it back through a reduced aperture stop. This is sort of like a "Lyot stop" in reverse. I'm sure that a rigorous optical designer will correct my use of the term, but it is the same idea. RM Hi RM, Many thanks for another excellent idea. You sure seem to have an unlimited supply! Much appreciated! I plan to go back to the location shown in post #3 http://www.l-camera-forum.com/leica-forum/2160547-post3.html and try it out. Actually when I shot that series of images, I meant to shoot with a higher f-number than I did, but I forgot to stop down after I had focused. ... Oh well, that gave us a nice problem to work on. So, I looked at my moon shots from the other night in which I had kept ISO and exposure time constant but varied aperture to eventually get an intense source of light by overexposing the moon deliberately. So, from analyzing those images it becomes clear when properly exposed and stopped down the APO-TELYT-R 1:4/280 enables excellent images of the moon. There should not be any doubt about that! But that's not what we are trying to do here. In a way we are trying to use the moon in order to set up a pretty well controlled laboratory environment to enable shots that give us some insight into the inner workings of the camera/lens system in question. So, in this spirit I did a few more moon shots by keeping exposure time and brightness (for lack of a better word) of the moon in the camera constant while varying aperture. So, these parameters are captured in the next image: 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! I then took six shots with these parameters and processed them in the usual way to find out if and when the problem goes away. Here are my results: f=4 f=5.6 f=8 f=11 f=16 f=22 So, indeed as you expected, the annoying flaring/ghosting is considerably diminished by f/8 and barely noticeable by f/11. Of course we don't know the apparent brightness of the lights in the initial photo in relationship to the moon, but I agree with you that it is worth a try to retake those images with the aperture stopped down. Thanks again for your help. 1 Quote Link to post Share on other sites Simply register for free here – We are always happy to welcome new members! I then took six shots with these parameters and processed them in the usual way to find out if and when the problem goes away. Here are my results: f=4 f=5.6 f=8 f=11 f=16 f=22 So, indeed as you expected, the annoying flaring/ghosting is considerably diminished by f/8 and barely noticeable by f/11. Of course we don't know the apparent brightness of the lights in the initial photo in relationship to the moon, but I agree with you that it is worth a try to retake those images with the aperture stopped down. Thanks again for your help. ' data-webShareUrl='https://www.l-camera-forum.com/topic/186459-leica-serie-55-ndx1-13026-germany/?do=findComment&comment=2102702'>More sharing options...
k-hawinkler Posted September 2, 2012 Author Share #60 Posted September 2, 2012 The cheap way to improvise a 5.5 filter is to use a 39mm filter glass without the rim.It will rattle, but it will work. Hi Jaap, Many thanks for the tip. Quote Link to post Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.