Mitutoyo/Nikon 2 - 5x infinity objectives for 24x36?

[Blame]

lauriek got it right. Yes the ruler has to be transparent to let the light in.

It is not so much a way of finding the magnification as the field width.

The point is that you get a rough value of the field width of the objective. Testing with a tube lens is the next stage. If the tube lens vignetes more than expected with the eyeball test you know that it is the tube lens and not the objective that is the limiting factor.

As to the maths.....

Looking through my mitutoyo 5x I can see 13mm of my ruler with a faily sharp cutoff. To cover the diagonal of a FF sensor I need 43mm so I need to multiply that 13mm by 43/13=3.3. A tube lens of 200mm focal length gives 5x magnification, so to get 3.3x I need (3.3/5)x200mm =132mm or 135mm to nearest findable tube lens length.

Or as a fraction: Minimum usable tube tube focal length =

(required diameter)(rated tube length)
________________________________

(field width, subject side)(rated magnification)


Thus if a tube lens of 135mm focal length gives full coverage but with quite serious vignetting in the exteme corners or a 150mm one works with insignificant vignetting I can conclude that it is not seriously contributing to the vignetting. I can also conclude that If I tried pushing my luck with a 120mm tube lens I could hope to get the sides but I can forget about the corners.

[dmillard]

lauriek got it right. Yes the ruler has to be transparent to let the light in.

It is not so much a way of finding the magnification as the field width.

The point is that you get a rough value of the field width of the objective. Testing with a tube lens is the next stage. If the tube lens vignetes more than expected with the eyeball test you know that it is the tube lens and not the objective that is the limiting factor.

Thanks for the useful idea, although I had a little difficulty understanding the concept initially. I think a picture would have been worth a thousand words here.

David

[ChrisR]

What bothers me is that this is "testing" the lens in the way it's not going to be used.
The parallel bundle of rays coming out, yellow, hits the tube lens, blue, which does whatever it does with them.
Looking along something like the red line, if that's what's being described, may not be very relevant? Unless the tube lens is placed further from the sensor than its focal length so not focused at infinity, perhaps.

[seta666]

I think a picture would have been worth a thousand words here.

David

Me too, I still do not see it, a picture please!!

That was fast, as I was writting you sent the picture, Thank you!!

[ChrisR]

Well as I still have my pencil out, what I think Blame means is, with paper tube green, ruler light blue:

[ChrisLilley]

What bothers me is that this is "testing" the lens in the way it's not going to be used.

I agree. I had misunderstood earlier, I thought the ruler was being photographed attached to tube lens and camera. Which is why I said measuring the long edge was easier than the diagonal.
Now I understand what is being proposed, it seems incorrect.

The parallel bundle of rays coming out, yellow, hits the tube lens, blue, which does whatever it does with them.

Wait, your diagram is wrong. It assumes that the subject is infinitely small and exactly centered; that will give the bundle of rays you show in yellow.



There are no off-axis rays being shown here, from within the subject imaging circle.

[ChrisR]

Right, thanks, got it. Penny's dropped!

As on this, which I posted ages ago!

[Blame]

But the whole point is not testing as used. The point is isolating one component and testing its limits. That way you can determine whether the limiting factor to field width is objective or tube lens.

The output from an infinite objective is as if seen from infinity. Assuming you are not short sighted that is exactly what your eye sees best. The output from an eyepiece is (give or take diopter adjustment) also converging at infinity. Simply put, an infinite objective makes a microscope all on it's own.

As the field of view of an eyeball is huge the limiting factor is the field of view of the objective. This is a case of what you see is what you get. - or at least will get if tube lens or camera sensor doesn't crop the image.

I don't know how to create pictures to explain, so my apologies. However you have worked out what I meant, although you need to stick your eye right on top of the objective like it was a microscope eyepiece. Finding out that it works is the act of moments. Just use it like a jewelers loop. Seeing is believing. It is not an exact test, but it works well enough.

[rjlittlefield]

I don't know how to create pictures to explain, so my apologies. However you have worked out what I meant, although you need to stick your eye right on top of the objective like it was a microscope eyepiece. Finding out that it works is the act of moments. Just use it like a jewelers loop.

Perhaps this illustrates the concept? This is with a Nikon CFI60 10X, as seen by a cell phone crammed against the rear of the objective.



The high-quality field on this objective is known to be about 5 mm (see HERE), versus the 6.5 mm seen in this view.

--Rik

[Blame]

rjlittlefield

Thanks. That does work to explain. A mobile phone camera makes a great eyeball.

Your example shows much the same as I have seen with mitutoyo lenses. The vignetting is similarly sharp and 6.5mm is exactly what I get with my mitutoyo 10x.

This trick is only really good for testing vignetting limits not quality limits, so 5mm quality diameter is quite valid - indeed just looking at your picture gives a visible quality diameter of about that.

[Chris S.]

OK, finally got around to testing Mitutoyo objectives on full frame. They do vignette: http://www.photomacrography.net/forum/v ... 3897#83897

[ChrisR]

Has anyone got any f=200, infinite objective lower than 4x, to cover 24 x 36?

But the whole point is not testing as used. The point is isolating one component and testing its limits.

Well in my case the point is to find out before I buy, rather than afterwards

The Nikon 4x NA0.1 does at least cover the frame, as does the Nikon LU 5x NA0.15. Obviously the higher NA might lead to higher quality, I haven't tested yet. At this stage it's a case of
"Never mind the quality, feel the width."
(This is with Nikkor 200 and 135mm primes)

Something odd happens in the The Nikon Finite 2/2.5x objectives, which have short WD and poor coverage compared to their higher mag brothers.
It would be interesting to know about a later (CFI60 or LU) Infinite Nikon

The Nikon Finite 4x NA0.2 Apo is fine in the middle but the quality falls off a cliff well short of 36mm diameter (let alone 43) so I wonder if the CFI60 4x NA0.2 Apo does the same.

Later Edit:
there's partial recent answer here,
http://www.photomacrography.net/forum/v ... 9698#79698
where Horstl's tests on the Mitutoyo 2x is tested with an MT4 tube lens. It nearly makes it, with about a 36mm coverage on the sensor.
The field coverage diameter (200mm tube lens) goes from about
43/5 = 8.6mm with the 5x objective and
36/2 = 18 mm with the 2x

Assuming
1) no more field of view is achievable from the 2x objective and
2) 135mm is as short a camera "tube" lens as can be used with the 5x,
is it "worth" having a 2x in the armoury, from the field of view perspective?
Widest field diameter from the 5x objective would be
43/5 *200/135mm = 12.7mm .
So yes it is!

The same calculation for a 4x objective gives 15.9mm though, which is running much closer (to 18mm).
A non-vignetting 120mm tube lens, would get there.

[ChrisLilley]

Something odd happens in the The Nikon Finite 2/2.5x objectives, which have short WD and poor coverage compared to their higher mag brothers. ...

Horstl's tests on the Mitutoyo 2x is tested with an MT4 tube lens. It nearly makes it, with about a 36mm coverage on the sensor.
The field coverage diameter (200mm tube lens) goes from about
43/5 = 8.6mm with the 5x objective and
36/2 = 18 mm with the 2x

Assuming
1) no more field of view is achievable from the 2x objective and
2) 135mm is as short a camera "tube" lens as can be used with the 5x,
is it "worth" having a 2x in the armoury, from the field of view perspective?

Chris, surely 2x is past the point where microscope objectives are superior to enlarger/duplicating lenses, and into the area where the latter is superior to the former?

Today I went through the exercise of calculating the effective aperture for the various lenses and objectives that I have, at the range of magnifications where I find them useful. The break-even point, I found, was around 3x.

By 4x even a low-end 4x/.1 objective like the Edmunds 'Nikon Finite Conjugate' is giving effective f/20 @4x while a reversed El-Nikkor 50/2.8 is giving effective f/28 when used at its optimal f/5.6 aperture at 4x. (This says nothing about the quality of the respective optical designs, of course, just about the effective apertures).

At 2x, the CFI Plan Achro UW 2x/.06 objective is giving effective f/16.7 with a WD of 7.5mm, while a reversed apo-rodagon-d 75/4.5 (optimised for 2x) is giving effective f/16.9 at its optimal nominal aperture of f/5.6. Admittedly the CFI Plan Apo 2x/.1 is giving effective f/10, but with a WD of 8.5mm, at high cost, and with what field coverage?

And bringing this back to your theme of high quality field coverage, the apo-rodagon-d lenses have published MTF which shows good quality out to 42.9mm. That is radius, not diameter. A lens optimised for 2x and with an 80mm image circle - isn't that what you should be using on an FX camera?

I don't have a note of the working distance, but from memory its better expressed in cm than mm....

[ChrisR]

No arguments there, Chris.

Just probing the boundary really. It's easy enough to contrive situations where overlap is useful, some of them actually happen!
If you're a buggist with a beast that's got a 3mm head on a 15mm body, it's useful not to have to change more than the lens to go from 10x to 2x.
If you're in the field and want to keep your auto camera features, it's not so easy with mag more than about 1:1 (unless you have a Canon MPE!)
(is there a ZOOM Raynox??)

I'm tight for space so I'd be interested in a finite objective I could use on a microscope-mounted camera and fill the frame. The Olympus and other rms lenses are handy for that.
I haven't made an "infinite" scope mounted arrangement I'm happy with yet.

There's something that bothers me about lenses which cover a huge field, like reversed enlarger lenses - are they likely to be great for the little bit in the middle that I want?

[rjlittlefield]

There's something that bothers me about lenses which cover a huge field, like reversed enlarger lenses

True, but countering that is the issue that low-power microscope objectives have to use complex optical designs so as to physically fit in limited space. Consider a standard DIN scope with 45 mm parfocal distance from the mount, forming an image 150 mm higher in the tube. That's a total of 195 mm to work with. At 2X, a simple "thin lens" would have to be located 65 mm from the subject, 20 mm up inside the tube! To keep from doing that, low power objectives have to be retrofocus designs, analogous to DSLR lenses that have short focal length but still have to clear the mirror.

--Rik