Sigma 600mm f/8 Mirror with issues

cooltouch · Joined: 15 Jan 2009 Posts: 9097 Location: Houston, Texas

All the posts in the recent thread about 500mm mirrors, with some especially nice mirror photos members have shared, have got me to thinking about a Sigma 600mm f/8 mirror I own -- this one is in EOS mount and is not a very sharp lens. Honestly, after I bought it a couple years ago, I shot maybe a couple dozen pix with it, evaluated them in growing disappointment and then just put the lens away.

A few days ago, after looking at all the great 500mm mirror pics in this current thread, I decided to drag my Sigma back out and give it another try. I shot about a dozen photos, and this time I turned a more critical eye to what was going on. I discovered a couple of interesting things. First, the lens has a pronounced amount of flare -- or soft image doubling if you want to call it that. And second, I noticed that the actual distance was WAY off from what the focusing ring was indicating. It was showing a subject that was a good 300 meters away to be a bit less than 30 meters according to the focusing ring. Funny thing was, because it was set the way it was, this meant infinity was reached by about 50 meters -- so there was a lot of room beyond infinity that was being wasted. So I assumed that close focus must suffer. It does, but nowhere nearly as much as I suspected it would. Closest focus according to its focusing collar is just a shade under 2 meters -- call it 6 feet. But I was able to focus on a flower about 7 feet away and there was still a bit of focusing room left. Kind of interesting, that.

So anyway, because of the way the images looked and because of the focusing collar being off by so much, I began to suspect that this lens had been opened up before. I gave it a very close examination of its exterior, but I could find absolutely no marks that a lens spanner often leaves. Not even touch up paint to hide spanner marks. All the screws looked to be in unscrewed condition. So finally I peeled away the rubber grip from the focusing collar! Aha! The Sigma uses a metalized sort of tape to hold down things much the same way that Tamron does. There was a piece of metalized tape that ran the full circumference of the collar and it has obviously been pried loose. I suspect that this tape holds two pieces of the collar together and that this is the way infinity focus is set. I have a Tamron 24-70 AF lens, where Tamron did the same thing with it as a means of adjusting infinity focus. So this tape is nothing I'm not familiar with.

If it does indeed control infinity focus, then I'm gonna try and reset it. As for that pronounced flare -- I believe that the flare is largely responsible for this lens's loss in sharpness. So if I can find the reason for it, I can probably vastly improve this lens's performance. I've closely examined the front and rear elements and both mirrors (I was able to examine the rear facing mirror by the reflection it cast on the primary mirror), and I inspected and cleaned the 30.5mm drop in filter, just for good measure. Everything looks fine, as near as I can tell, at least.

Then I tried something. I held the mirror up so it was reflecting a subject behind me and concentrated on the exit circle. Then I pulled out the filter and reinserted it. It's not that easy to tell, but I began to think that just maybe the filter might be the culprit. It's a Sigma brand filter and it's obviously the one that came with the lens.

So, I took the lens out again, but this time without the drop-in filter, and shot several of the same subjects. Result? Well, it's kinda interesting. Without the filter, sharpness is improved by a slight amount, but the image becomes almost a full stop brighter and contrast suffers somewhat. Unfortunately, that flare that slightly doubles the image is still there.

Following are a few comparison shots. I didn't add any sharpness to any of the pics, but I did adjust the brightness and contrast some what on the pics that were shot without the drop-in filter, such that they were roughly equivalent to the shots taken with the filter in place. I made these adjustments using Photoshop's RAW image converter so actually these are not post processed images since the adjustments were done to the raw files (and can easily be undone). I've reduced the images to "web size," i.e., 1500x1000 pixels, so not much difference can be seen.

These subjects have become something of a sharpness standard for me. Both are about 300 meters away and both have sufficient detail where I can make some judgements with regard to sharpness and contrast.

Following these pics, I've shown some 100% crops. Note in the crops the "doubling flare" I was mentioning above. Any clues you may be able to provide as to what is specifically causing this flare, I'd be much interested in hearing about. I think that at least one element is out of alignment, although I don't see offhand how I'll be able to realign it, unless a previous owner put things back together wrong, such that a correct reassembly might allow for the elements to fall into proper placement.

A street sign about 300 meters from my front door. You can't see the flare coming off any of the detail of the sign in these images although you can with the 100% crops below.

Without the drop-in filter:

A fire hydrant right next to the street sign.

Without the drop-in filter:

100% Crops with filter in place. Camera was a 24.3mp NEX 7:

With filter

Without filter

With filter

Without filter

As you can see this soft doubling of the image is above and slightly to the right. The second image (without filter) of the fire hydrant shows it most sharply. I think this is caused by a lens misalignment, but I don't know which one it could be. I kinda don't even want to open it up; there's a good professional camera service center local to me that can probably repair it for an economical amount.

Anyway, I have a growing confidence that, once this lens is repaired, it may exhibit the same level of sharpness as my old Sigma 600 did. That would be truly outstanding.

Lloydy · Expire: 2022-01-01

The LZOS 'Matsukov' 500 / 8 that I bought the other day came with a M42 adapter already on it, so I did my first test shots with it and soon realized that something was wrong, very wrong. Infinity was about 10 metres away. I put my usual M42 adapter on and the lens is perfect. I measured the two adapters and the bad one was 0.78mm shorter, and that was enough to render the lens unusable. And probably why it was so cheap Wink

I think mirror lenses, and probably all long telephoto lenses, show up errors such as this in a much more noticeable manner with the narrow DoF.

cooltouch · Joined: 15 Jan 2009 Posts: 9097 Location: Houston, Texas

Your comment about the adapter is a good one, but I suspect I can discount it in my case. When I ordered the EOS->NEX adapter for my NEX 7, I also ordered a Canon FD, Nikon F, M42, LTM, and 16mm C-mount adapter, all from the same seller. They all work great so I'm disinclined to think the adapter may be the culprit with respect to infinity focus. But it's easy enough to confirm with one of my other EOS mount lenses, which I will do. Oh wait, now that I think about it, I shot some with it using my EOS digital, and it worked fine. Still back then I wasn't paying attention to the position of the focusing collar.

Honestly, I'm not worried about the focusing collar being off, I'm more concerned about the flare.

visualopsins · Expire: 2025-04-11

Actually no, adapter thickness errors are more affective with wide angles. 0.78mm is enough to see through 500mm lens, but 35mm lens probably would not achieve infinity.

More focus throw is easier to focus.

Of course you know filter degrades image.

Check the rear elements. One is probably flipped. You may also find some haze on one element. The focus should be adjusted finally.

e6filmuser · Joined: 12 Nov 2010 Posts: 560 Location: Reading UK

visualopsins · Expire: 2025-04-11

cooltouch · Joined: 15 Jan 2009 Posts: 9097 Location: Houston, Texas

John Shriver · Joined: 24 Dec 2009 Posts: 192

I tried testing my R. Topcor 300/2.8 lens through a 1926 glass window. One that is quite ripply, but I didn't think about that. Got awful results, took the lens apart to test if the rear group was in backwards, it wasn't. (That mushroomed into a lot of work.)

Then went to test it again, looked at the window, and realized how stupid I'd been. Worked GREAT with the window open.

As for the mirror lens, wondering if the silvering of the mirrors is failing.

calvin83 · Joined: 12 Apr 2009 Posts: 7553 Location: Hong Kong

In my experience, there is no need for rear filter on any of the mirror lenses I tried when used on my NEX/A7.

jamaeolus · Expire: 2015-08-20

I got a bargain AF minolta 500 reflex cause the filter tray was missing. (Just over a 100 USD) Taped over the hole and it works great on my a7ii with LA-EA4 adapter. So I can add weight to your claim Calvin.

Abbazz · Joined: 23 Jun 2007 Posts: 1098 Location: Jakarta

cooltouch · Joined: 15 Jan 2009 Posts: 9097 Location: Houston, Texas

visualopsins · Expire: 2025-04-11

Ha!

Try lens again under optimal lighting -- no front lighting! I'm thinking perhaps your choices show lens at it's worse in difficult lighting.

If you still get the doubling there is definitely something flipped in rear group imho.

cooltouch · Joined: 15 Jan 2009 Posts: 9097 Location: Houston, Texas

To me, the lighting was optimal. Mid-afternoon on a sunny day. The subjects were well lit. These are not the only photos I've taken with this lens. All photos display this ghosting of a double image. I think you're right, though. At leest I hope you are -- one of the elements is flipped, or possibly canted in its receptacle.

cooltouch · Joined: 15 Jan 2009 Posts: 9097 Location: Houston, Texas

I just came across this somewhat stale thread as I was doing a search here, and realized I never got back with the latest news on this lens. I dismantled the rear of it. I didn't remove the helical to get to the interior -- didn't see much point. But I did completely dismantle the rear of the lens, finally exposing the rear element. And the way it is designed, it is impossible to flip. The element is bonded to a rear metal standard that fits tightly into the pieces only one possible way, so it can't be flipped, and there is no room for movement.

This lens is clearly defective, exhibiting a sharp and clear double image. If I can get it corrected such that the double image is merged, it will probably be a good optic. I'm thinking about taking it to the local Professional Camera Repair place and see if there's anything they can do with it.

visualopsins · Expire: 2025-04-11

I meant to say "no backlighting" Smile

Mid-day light is very harsh imho.

woodrim · Joined: 14 Jan 2010 Posts: 4060 Location: Charleston

paulhofseth · Expire: 2018-06-28

I have not bothered to test with&without rear transparent filter, but in theory it should make a difference since the rays exiting can not be perfectly perpendicular to the glass (easily understood since rear lens elements are smaller than the film format they will cover). Hence both the filter thickness and the dispersion in the glass type used will make a difference and has been part of calculating the optics. Whether its absence is visible at internet size or first becomes noticeable with a magnifier at A3 size is another matter.

Minolta specified a particular filter for their mirror lenses. In the corresponding Minolta price lists, the filter was listed as having a different glass type to the ordinary 39mm filters (and cost very much more).

p.

Big R · Joined: 29 Nov 2010 Posts: 42

I had one of these Sigma 600 f8 mirrors, too and interestingly, my images looked about the same as yours. Removing the filter does make the images different but not really better...
There were some dark spots on the mirrors that i tried to clean, but that didn't really help. I also suspect that these weren't the ones which did harm the image but that there might be damage to a lens element or a misalignment.
At the end I sold the lens for a few bucks.

paulhofseth · Expire: 2018-06-28

Addendum to my comment above; Minolta and Leitz clearly specify a rear filter, but on his thoroughly researched and illustrated website, Marco Cavina describes a Nikon mirror lens as not designed with a filter in the light path.

p.

cooltouch · Joined: 15 Jan 2009 Posts: 9097 Location: Houston, Texas

Woodrim, I read about your success with your Rubinar, so I was bearing that in mind when I dismantled and "remantled" the Sigma. No screws were overtightened and the tripod collar, while removed and then reinstalled, gave no indication of binding when it was loose or tight, so I discounted it as a possible culprit. After tearing into the back of the lens, I've come to the conclusion that the problem does not lie with the rear element, unless it is defective, in which case there is nothing I can do about it.

I'm aware that, in general, lenses that have filters in the light path require them to be there because they are part of the overall optical design. So I didn't expect much in the way of positive results when I tested the lens without the rear filter in the light path.

However, what I found was an increase in contrast, if not sharpness, when the filter was removed. The double image did not go away, but in fact became more pronounced. Now, it makes sense that adding an element to the light path can reduce contrast, especially if it isn't perfectly clean, so I made sure the filter was spotless when I went to test the lens with the filter in place. Nonetheless, tests with the filter in place showed images that had a slightly reduced contrast compared to those when the filter wasn't in place. I could see no other optical issues caused by the lack of a filter in the light path.

woodrim · Joined: 14 Jan 2010 Posts: 4060 Location: Charleston

Michael: I'm guessing that you've put the lens aside again. It does get frustrating when time spent does not yield results.

I have just today received a Sigma 5.6/400. I have been reluctant to buy a Sigma mirror, mainly because I have two very good 500mm and don't need to risk getting a lesser lens. However, I saw an opportunity to try the 400mm for cheap. I have only taken some BS test pictures and have mixed results. I did find that with one shop particularly, the exposure created ghosting around bright spots, although that picture had a large dark background and I was in multi-point exposure mode. I need to be in center mode, if not spot when using the mirrors.

There is evidence of decent sharpness but also some image doubling, or ghosting, that is puzzling. I wonder if it is similar to some of what you are experiencing. In the following test example, the center where I focused seems sharp but the leaves on the left have the double image ghosting. Any thoughts?

ZoneV · Expire: 2011-12-02

I think the image quality without filter is not too bad for a mirror lens.
Photo filters always create additional straylight, so on a high contrast scene this could be a reason for straylight.

Removing filters behind a lens which are designed to be there changes flange back distance. On fast lenses + high resolution sensors there is influnece on the optical quality. But this is more for f/2.8 and faster. But: Our old manual consumer lenses are designed for film, so there is no ir-cut filter, anti-aliasing filter and sensor cover glass in the optical design. From this removing back filters from lenses could rather improve image quality (when flange back distance is adjusted).

The mentioned front filter influence is a common problem. But I think still not that well known. FIlters are not perfect parallel plane plates of glass, but rather have some structure, thickness variations, may partial act as a lens!
So close to the sensor this is no problem, but with for example 500mm lens, or a fast 200mm lens wideopen, these small deviations are enlarged much, and can affect image sharpness, or and bokeh.
Many had problems with front filters and the Canon Ef 100-400mm lens, and some even want to sell the lens because of bad quality - but it was the filter.

A university friend of mine has co-authored this text about filter quality. She is working at Schneider Kreuznach at the industry filter department:
https://www.vision-systems.com/articles/print/volume-16/issue-10/features/filters-factor-into-optical-imaging.html

Consumer lens manufacturers don´t told me their filter flattness quality when I asked a lot of them on the Photokina fair some years ago. I suppose it is not that good Smile

cooltouch · Joined: 15 Jan 2009 Posts: 9097 Location: Houston, Texas

Woodrim -- Wow! That is some really pronounced doubling on the left side of your image. The center looks decent, though, but still I don't consider that doubling to be normal. There is also some rather pronounced doubling on the right side as well, but not to the extent as on the left.

So, yes, I suspect your lens definitely has a collimation issue. Can this be repaired? I dunno. I'll let you know more when I let my repair people take a shot at it with my Sigma. And yes, you've guessed correctly. I have set the lens aside. I'm saving my ducats at the moment for more pressing needs -- like paying for parts to rebuild the top-end of my old BMW R90/6 motorcycle, which is in dire need of a top-end overhaul. I've been holding back on any major outlays until I've bought the parts I need for this project. Although, I couldn't entirely resist, as you might read in my most recent post in the 500mm lens thread.

ZoneV, thanks for that link on filters. It's quite technical, so I've bookmarked it where I'll be able to go back and read it when I have the time to expend the brainwork on understanding it more thoroughly.

When I finally do get that lens looked at by the repair shop here, I'll update this thread and let you know if there's any difference after they've had it apart.

woodrim · Joined: 14 Jan 2010 Posts: 4060 Location: Charleston

Michael: After that first test shoot, I pulled the filter, which needed cleaning. The next day the doubling was not apparent. If not the dirty filter, perhaps it wasn't seated properly. I'll know soon if the problem is still present.