Tarmon SP 35-80/2.8-3.8 with strong field curvature

[pkmx]

I got the Tamron SP 35-80/2.8-3.8 and Tamron's Adaptall-to-FD adapter yesterday, and upon initial testing I found that my copy have very strong field curvature.

Tamron SP 35-80/2.8-3.8 @ 80mm f/3.8:


Canon FDn 35-70/4 @ 70mm f/4:


Tamron SP 35-80/2.8-3.8 @ 80mm f/8:


The centers are all sharp. However, as you can see, the corners are very very soft. The Canon FDn 35-70/4 at f/4 easily beats it even at f/8.

I adapt both lenses to Sony a6000 via either FD-NEX lens turbo or glass-less no-name adapter, the issue can be seen either way.

I know this lens is supposed to be very good, so I wonder if this is an issue with my copy or the adapter? Perhaps stacking two adapters (adapter-to-FD and FD-NEX) introduces way too many flange errors to the system?

[calvin83]

At 80mm, is the lens focus to infinity when you turn the focus ring to the infinity position? If not, what is the value of the distance scale?

[pkmx]

[Gerald]

First of all you need to find out if the problem is field curvature or decentering.

1. Focus the left edge, then the right edge, and record the corresponding focus distances.
2. Distances equal -> field curvature
3. Distances different -> decentering

Note: Decentering is a lens defect, but the same symptoms are produced by parallelism error of the adapter or camera's flange.

[calvin83]

There is at least two possibility: either your adapter is not in the correct thickness or the lens is faulty. First, check if there is de-centering. If not, check if the lens can focus to infinity( at 35mm and 80mm) when the distance scale is set to the infinity position on a Canon FD mount camera. If you don't have a FD mount camera, hold or fix the lens in front of your camera.

You need to make sure the lens can focus to infinity( at 35mm and 80mm) when the distance scale is set to the infinity position(or very close to the the infinity position). Otherwise, it may be hard to tell where the problem comes from. Don't use a lens turbo for testing as it will induce field curve to the lens.

[kansalliskalaCafe]

Tested my copy with Samsung NX10 (a6000 is a crop camera, right?)

80mm, the distance is about the same I suppose. Well, I think there is some difference, but not that drastic as yours.

look at the word "humala" (=hop) in each.

Yes, I might have used some less shiny objects for testing but didn't have any 3 similar at hand.

3.8


100% crop @ 3.8


8


100% crop @ 8

[pkmx]

[calvin83]

I think your lens maybe faulty with the data your provided and the samples by kansalliskala.

[marcusBMG]

I agree there is something amiss with your lens pkmx. This is one of my 01A's - probably the least pristine one actually, loose mechanics. Quick test pics with my NX20, 20MPx APSC, , I wasn't particularly diligent about lining everything up I'm afraid so slight edge softness might be slight drift out of focus because not square on.

I would say the results are typical 01A, loss of contrast wide open is main thing. Click on resized pic for full size (3MB - I increased the jpg compression from the 7MB OtC files).

80mm Wide open.


F8

[cooltouch]

I purchased my copy of the SP 35-80 last year, largely because of the great reviews the lens has gotten, but honestly I haven't used it much. So I decided to add a data point or two to this topic.

The 35-80 has an FD Adaptall-2 mount installed. I mounted it to my NEX 7 using a Fotasy FD-NEX adapter, which provides a tight, secure mount. I took photos at 35mm and 80mm, wide open and at f/8. Here are my results:

Wide open at 35mm:
[/url]

F/8 at 35mm:


Wide open at 80mm:


F/8 at 80mm:


As you can see, there is little to no fall-off in sharpness toward the edges of this camera's APS-C frame. Given that it is APS-C, I wouldn't expect there to be much. I also find it interesting that there isn't much difference between wide open and f/8 with these images.

So, I suspect that pkmx's 35-80 has decentered optics, and not field curvature. If there was curvature, it would be visible at the edge of his frames, and I see no evidence of curvature in his photos.

[pkmx]

Thanks for the inputs, guys! That's really a bummer. I was really looking forward to this lens since it has a good reputation, and I had to look for a good while since the lens is quite rare here.

I found that the corner softness mostly vanished while shooting macro. This one is shot at 80mm wide-open at 1:2.5 magnification on APS-C crop:



Now only the extreme corners are soft, and the rest of the frame looks good, which is not too bad considering it was shot wide-open. So it seems like I can use it as a macro at least.

I also found that the rectangular hole of Adaptall-FD adapter doesn't line up with the sensor, illustrated here when mounting onto the Lens Turbo II:



I wonder if this could be the culprit of the bad corner performance, is it supposed to look like this?

[calvin83]

These may help.

http://forum.photozone.de/index.php?/topic/830-do-it-yourself-centering-sanity-check/
http://www.lensrentals.com/blog/2012/05/testing-for-a-decentered-lens-an-old-technique-gets-a-makeover .

[cooltouch]

[calvin83]

The alignment is not the cause of the softness as you are not testing it on your lens turbo.

[pkmx]

[marcusBMG]

I had a slightly similar problem with another lens. Couldn't see what was up, then I noticed a rattle that proved to be a loose element.

[cooltouch]

Actually, that's a pretty good point. I've owned three different Tamron lenses that had element groups come loose. One even completely fell out and was just laying inside the lens tube. Fotunately, in all three cases, it was a fairly easy matter to open the lenses up and tighten down the loose groups.

[calvin83]

If there is no rattle, it is possible that the rear element is flipped. It is hard to know until the problem is sorted out.

[pkmx]

There is no rattle, and the glasses are pretty clean as well. Maybe I could open it up and figure out what could be wrong with this lens.

Oh and, I also performed the decentering test linked above (http://forum.photozone.de/index.php?/topic/830-do-it-yourself-centering-sanity-check/) and couldn't produce the problem.

[marcusBMG]

This lens is one of the more complex ones. I am in the process of making a custom tool to open it up (cos I have a good one but with a nasty blotch of fungus on the third element in).

More technical problems like collimation, shimming, decentering would probably need an optical bench to diagnose, and an engineering shop or at least a supply of parts from defunct lenses to fix.

Opening the lens up might reveal an obvious and fixable fault, but the practical advice is just to regard it as a dodgy one and shop around for another one. These aren't so expensive, 50 quid is average ebay auction here (UK).

[cooltouch]

A few years ago, I picked up an earlier adaptall 35-80, like this one here:

http://translate.google.com/translate?hl=en&sl=ja&u=http://www.tamron.co.jp/data/old-lens/qz35m.htm

It also had some fungus on an inner element toward the front of the lens. So I decided to take it apart from the front. I used a lens spanner to loosen the front retaining ring. It was very tight. The spanner slipped off the ring and put a nice scratch in the front element. Whoops! But I finally got it off. It was holding on a very thin front element, and there was another ring behind it. That ring was on so tight, I was never able to bust it loose, so I never made it to the element that had the fungus on it.

Now, I'm sure that this early 35-80 had a different optical formula from the SP version. And hopefully, if you try to open it up, you won't run into the same problems I did. I'd be interested in knowing how your repairs work out.

[pkmx]

I opened the lens up and couldn't find anything wrong after looking around for a bit, so I guess that without proper tools this is going to be very hard to diagnose and repair.

Since there weren't none left on my local market, I decided to just order another one from KEH. Let's hope that this one will be good!

Thanks to everyone who helped in this thread.

[cooltouch]

Well, even if it isn't good -- or you just don't like it -- KEH has a good return policy. So you won't get stuck with it if you don't like it. They're good people. Periodically KEH gets a pile of my money in exchange for some cool gear. Robert's Camera, out of Indianapolis, Indiana (http://robertscamera.com/), is another good source for used gear at good prices.