Extension tubes on a Nikon 10X

All Ex · Post by **All Ex** » Sat Jan 23, 2016 6:05 am

Tomorrow with that set up :

Nikon 10X, 200mm tube lens, 68mm extension tubes.

I`ve managed to shoot that:

(appr. 2mm)

Is it possible to achieve 18X (or so) magnification that way?
(There is some vignetting and it is a dark photo, but if that indeed applies I think it worth's the effort)!

ChrisR · Post by **ChrisR** » Sat Jan 23, 2016 7:37 am

The objective is designed to have the "tube" lens focussed at infinity.
You are pushing it away from that, so you would expect quality to be impaired.

You already have enough pixels to get all the detail a 10x NA 0.25 objective can collect.

So, you may find the result is better if you simply enlarge the middle of an image taken at 10x.

Lou Jost · Post by **Lou Jost** » Sat Jan 23, 2016 7:47 am

You already have enough pixels to get all the detail a 10x NA 0.25 objective can collect.

ChrisR, are you sure about that? My results with a Panasonic bridge camera vs a Nikon APS-C camera showed that one could get much more detail out of a 10x lens by increasing the pixel density on the sensor.

I'm not arguing in favor of ext tubes, just saying the aerial image has far more detail than that sensor can record.

rjlittlefield · Post by **rjlittlefield** » Sat Jan 23, 2016 9:36 am

Lou, if I recall correctly your results applied mainly to the case where a 10X objective had been pushed down to lower magnification, and more generally to the case where the objective's entire usable field was mapped optimally to cover the sensor.

In this case, All Ex is using a 200 mm rear lens, so the objective would be running at 10X and effective f/20 if the rear lens were set on infinity and there were no extension tubes. In that situation, his Nikon D800, full frame with 36 megapixels, does indeed have enough pixels to capture all or almost all of what's in that optical image.

From there, it appears he is turning the focus ring on the 200 mm and adding extension tubes so as to get significantly more magnification, 17X in the example shown here.

I strongly recommend against trying to do that. The extended optics will be running at effective f/34, plus a significant amount of added spherical aberration that will kill the sharpness of fine detail. Yes, the image will be 18X, but it will be empty magnification.

All Ex would do better to learn how to produce a sharp image using the optics the way they are designed to be used. After he can do that routinely, then it might be time to play with other configurations.

All Ex, if you want more magnification from the same objective, it would be far better leave the rear lens focused on infinity and add a teleconverter instead of extension tubes. That will at least allow the objective to work as designed.

--Rik

ChrisR · Post by **ChrisR** » Sat Jan 23, 2016 10:33 am

ChrisR, are you sure about that?

Well, I put "may" on purpose;)

I did squint at the image to see it's a D800 and a 200mm lens. It's a while since I tried pushing things closer (you get better results from a higher magnification objective). I'm sure the ojective will "go off" noticeably when abused that far, but there may be a little to be gained if using say one more pixel per detail on the sensor, gives more than the abuse removes.

There are several tests around, some more rigorous than others, suggesting that poor or diffraction liimited lenses can give better results on a higher pixel-count sensor than seems reasonable, though. Only some of them are trying to promote a Canon 5DS!
Theoretically speaking, I doubt my understanding of the interaction of overlapping Airy discs on MTF curves, too much to say much.

As well as a teleconverter, a longer "tube" lens gives the pro without the con - before introducing its own cons of course. I believe Alex doesn't have either.

Lou Jost · Post by **Lou Jost** » Sat Jan 23, 2016 10:44 am

Rik, I did both kinds of tests. Here
http://www.photomacrography.net/forum/v ... highlight=
I showed that a Sony HX400V bridge camera , when it is used at 200mm with a Mitu 10x, reveals considerably more detail on the subject than I can get out of a Nikon D90 APS sensor. This means there is a lot more information in the aerial image than an APS-C sensor can record. I think your tests of aerial images showed the same thing.

So I think All Ex's desire to expand the image might make sense, though his chosen method is not recommended. A better solution would be to use a 400mm lens instead of the 200mm lens, as ChrisR points oout, and if that is not an option, a teleconverter might be a good choice, as you say. I've gotten nice results with Nikon 2x teleconverters behind good prime lenses with the Mitu 10x.

ChrisR · Post by **ChrisR** » Sat Jan 23, 2016 11:20 am

How much more, was that? At the time, in the link you give, you wrote:
"I conclude that the Sony makes useful tube lens + sensor for the Mitu! I think it may resolve a bit more detail than the Nikon, though not much more."

Bear in mind Alex's pixels are about 3/4 the dimension of the D90 you used, so there would be less to gain.

Perhaps you did another comparison to be able to say "considerably more detail"?

A while ago I tried a camera lens which zooms to 300mm on a Canon, 18MP APSc, so similar density to the D90. It did show more detail at 300mm.
It's here somewhere..

Edit - found it, over 4 years ago:
http://www.photomacrography.net/forum/v ... hp?p=92328
I'm not sure if I'd call that "a lot" more detail, though Rik did at the time.

Lou Jost · Post by **Lou Jost** » Sat Jan 23, 2016 11:31 am

Chris, I don't have resolution targets (other than my favorite white butterfly scales on my reference butterfly wing...) so I can't be very precise about it. I made that post when I was just getting started in this. My subjective impression now is that the small sensor shows significantly more fine detail than the APS-C sensor, though there is more noise too. And that particular Sony camera doesn't let me control the amount of noise reduction it uses to make jpgs, nor does it shoot RAW, so it is hard to compare.

I have used a Nikon 400mm ED-IF with the Mitu as well, and it does resolve more detail on the subject than I can get from a 200mm lens. But not as much more as I expected. I'll do a test and post it one of these days.

All Ex · Post by **All Ex** » Sat Jan 23, 2016 12:42 pm

In this case, All Ex is using a 200 mm rear lens, so the objective would be running at 10X and effective f/20 if the rear lens were set on infinity and there were no extension tubes. In that situation, his Nikon D800, full frame with 36 megapixels, does indeed have enough pixels to capture all or almost all of what's in that optical image.

From there, it appears he is turning the focus ring on the 200 mm and adding extension tubes so as to get significantly more magnification, 17X in the example shown here.

I strongly recommend against trying to do that. The extended optics will be running at effective f/34, plus a significant amount of added spherical aberration that will kill the sharpness of fine detail. Yes, the image will be 18X, but it will be empty magnification.

All Ex would do better to learn how to produce a sharp image using the optics the way they are designed to be used. After he can do that routinely, then it might be time to play with other configurations.

I suspected that, otherwise a set of bellows, lets say, would do all the dirty job and everything would have been all wright, I mean without having to mess with objectives and other thinks like that.
Your sayings Rik are highly appreciated, your explanation on the degradation of the quality in the image that setup produces is very helpful.
And to avoid confusing the things, the initial photo was taken before the final adjustments on the lens.
I prefer to use (in the future) a 20X objective than trying to do it`s work by other ways, in between I`m experimenting (as all do) with what`s in my hands.

I have used a Nikon 400mm ED-IF with the Mitu as well, and it does resolve more detail on the subject than I can get from a 200mm lens. But not as much more as I expected. I'll do a test and post it one of these days.

That is a new for me approach, I`ll try to follow and understand it.
Thank you all,

rjlittlefield · Post by **rjlittlefield** » Sat Jan 23, 2016 12:59 pm

ChrisR wrote:A while ago I tried a camera lens which zooms to 300mm on a Canon, 18MP APSc, so similar density to the D90. It did show more detail at 300mm.
It's here somewhere..

Edit - found it, over 4 years ago:
http://www.photomacrography.net/forum/v ... hp?p=92328
I'm not sure if I'd call that "a lot" more detail, though Rik did at the time.

Yeah, it does look like I was feeling generous that day. Looking again just now at the blue scales in the one with 300, versus same scales in the one at 166 resized to 200%, I can see a lot more detail in the 300 at the scale ends. But right now I'm thinking that was some matter of illumination, flare, or some stacking artifact killing the contrast, because other areas of the scales look pretty similar.

--Rik