Usability of a 100x-lens in photomacrography?

A forum to ask questions, post setups, and generally discuss anything having to do with photomacrography and photomicroscopy.

Moderators: rjlittlefield, ChrisR, Chris S., Pau

keks
Posts: 44
Joined: Sun Mar 13, 2011 10:41 am
Location: Austria

Usability of a 100x-lens in photomacrography?

Post by keks »

Hello all, first a big thanks for the guys here who don't get tired answering the load of questions, that comes in here.

I am currently working on an considerable update of my all-directions-tabletop-setup that i have posted some years ago. It includes a rear focussing mechanism (fixed lens, camera moving for focussing), similar with ploum's setup, but with finite lenses and combined with a stackshot. The other things are a very flexible mounting/positioning system and a CRI 95 led lighting system. All done with realatively simple means. When that is finished I will again contribute something here:)

For now, being a non-optician, I have a question about a matter that has puzzled me for a while. Its about the resolution limit for photography with full frame sensors. Im not talking about the very small steps for stacking here.

When I calculate the effective aperture E for some Nikon M Plan lenses, using E=NA*(M+1) I get:

10x, NA 0.25: E=2.8
40x SLWD, NA 0.40: E=16
40x, ELWD, NA 0.50: E=21
100x, ELWD, NA 0.80: E=76

Im not understanding the optical laws behind this, so i hope this calculation represents my thoughts. For me the limit of usability without getting limited by diffraction is somewhere around 40x. A setup with this 100x-lens or any other lens of similar magnification no matter what NA appear severly limited by diffraction to me.
Am I correct with this consideration? If yes, whats the use of a 100x lens? Can they even be used for photography? Is there a different situation when watching through it with the naked eye?

Thanks for enlightening me!

Pau
Site Admin
Posts: 5518
Joined: Wed Jan 20, 2010 8:57 am
Location: Valencia, Spain

Post by Pau »

I'm not good with the numbers, someone will revise them, although they seem OK.:oops:

A 0.8NA objective will resolve more (smaller details) than a 0.5NA, but the effective aperture is smaller due to the much higher magnification, so on sensor the image will look less sharp despite possibly containing smaller details on subject. In fact you need less pixels to capture all the detail provided by the 100X. In both cases you're diffraction limited, but more severely with the 100X.
If you really don't need the modest increase in resolution you can live with a 50X or 40X with less frustration.

Resolution is not about sharp pictures but about smallest details resolved although we want both :D
Last edited by Pau on Thu Dec 29, 2016 1:25 am, edited 1 time in total.
Pau

Charles Krebs
Posts: 5865
Joined: Tue Aug 01, 2006 8:02 pm
Location: Issaquah, WA USA
Contact:

Post by Charles Krebs »

When I calculate the effective aperture E for some Nikon M Plan lenses, using E=NA*(M+1) I get:

10x, NA 0.25: E=2.8
40x SLWD, NA 0.40: E=16
40x, ELWD, NA 0.50: E=21
100x, ELWD, NA 0.80: E=76
The relationship you posted is not correct. It is for photographic lenses (symmetrical ones) that are marked in f-numbers. In that case the effective aperture is (f-number)*(m+1)

To find the effective aperture (in photographic terms) for a microscope objective you should use:
effective-aperture = m/(2*NA)

10x, NA 0.25: f-effective = 20
40x SLWD, NA 0.40: f-effective = 50
40x, ELWD, NA 0.50: f-effective = 40
100x, ELWD, NA 0.80: f-effective = 62.5

While it is informative to think in terms of the effective aperture, with microscope objectives it is very useful to look at the NA directly. The higher the NA, the higher the potential resolution. And it is a direct relationship. For example the 40X ELWD above has an NA of 0.50 while the 10X has an NA of 0.25. In terms of the ability to resolve fine subject detail, the 40X can resolve details 1/2 the size of the 10X. But it magnifies them 40X. So there will actually be smaller details that can be discerned using the 40X, but there will also be more overall image blur from diffraction.

Looking at the NA can help determine what objectives might be useful compared to what you already have. For example, there are 20/0.40 objectives with decent working distances. They will be able to resolve the same size details as a 40/0.40. So a picture taken with the 20/0.40 and then cropped and up-scaled to the same size as the 40X will show about the same amount of detail. Of course this does not take into account some other image characteristics such as digital noise levels, but don't expect to see more smaller details with the 40X.

The 100X you mention has a NA of 0.80. So it will be capable of resolving subject details that are 1/2 the size of the resolvable details of a 0.40 objectives. But it "spreads them out" 100X. So you will be able to discern smaller subject details but the overall image itself would not look really "sharp" unless reproduced at a quite small size (and certainly not at a 100% pixel view.)

Chris S.
Site Admin
Posts: 3718
Joined: Sun Apr 05, 2009 9:55 pm
Location: Ohio, USA

Post by Chris S. »

Keks,

You've already received excellent advice, but let me point out the additional consideration of working distance.

While I'm not familiar with a Nikon 100x/0.80 ELWD objective (presumably of "finite" design, with a tube length of either 160 or 210mm), I'd expect that such a lens would be restricted by an alarmingly limited working distance. A bit of searching turns up an example lens that seems to match your specifications, and is consistent with my conerns: eBay item 192007742686. Unfortunately, the stated working distance of this optic is just 2mm--so short as to be very difficult and constraining for macro work with reflected light.

As an alternative, you might consider the Nikon CF ELWD 60x/0.70 (finite) objective, which has a working distance of 5.9mm. While this objective has modestly less (7/8 ) resolution than the 100x/0.80 you described, it has almost triple the working distance--making it much more practical to use. I have a specimen of this 60x lens, and like it. This said, it is an achromat, and as one would expect with achromats, it shows more axial chromatic aberration (think purple fringes in high-contrast areas) than do apochromats such as the Mitutoyo 100x/0.70. These purple fringes tend to disappear with stacking, but in my laziness, I prefer to avoid them.

--Chris S.

ChrisR
Site Admin
Posts: 8618
Joined: Sat Mar 14, 2009 3:58 am
Location: Near London, UK

Post by ChrisR »

I find that it is indeed very difficult to illuminate a subject, using a 100x NA 0.8 ELWD Nikon, WD 2mm.
At NA 0.8, there's a 106 degree cone of image-forming light going from the subject to the lens, which you have to fill with relevant light, otherwise you aren't utilising the aperture you're paying for.
With a longer WD objective, though the glass is further away from the subject it makes the same angle, so you'd perhaps expect the identical problem. The difference comes from the thickness of the metal surround. It's always going to be at least a couple of millimetres, which is more significant at short WD.
Looking at it simply, NA 0.8, 2mm WD means the glass only needs to be about 5.3mm across.
These being 1mm squares:

Image

Shows how hard it is to illuminate the subject (0.2mm across) evenly from "outside". It's all very much side-light, with much likely to spill straight to the glass.
Far more practical is to light it from inside, ie by epi-lighting, through the objective.
But if you're going to do that, you can use even higher NAs.
Such objectives are available. Some have illumination surrounds for darkfield, which are referred to as BD objectives.

(BD objectives can be useful with the outer barrel removed, to leave a skinny glass support which makes them useful. Above about 60x, though, the WD is extremely short, and they're sprung-ended. If those usefully come apart, I don't know how!)
Chris R

JH
Posts: 1286
Joined: Sat Mar 09, 2013 9:46 am
Location: Vallentuna, Stockholm, Sweden
Contact:

Post by JH »

Hi
The Nikon M plan ELWD 100x 210mm tube 2mm WD is quite nice but I sugest to use it on a microscope. 2mm is enough for diffused LED-light.

Examples of pictures taken with that lens can be found in these posts;

http://www.photomacrography.net/forum/v ... hp?t=22577
http://www.photomacrography.net/forum/v ... hp?t=25028
http://www.photomacrography.net/forum/v ... hp?t=24508
http://www.photomacrography.net/forum/v ... hp?t=27724
http://www.photomacrography.net/forum/v ... hp?t=28112

This is not an APO lens but I have had very little problems with CA. It would be interesting to compare this lens with the Nikon TU plan 100x 0.9 and 0.8 lenses.

Regards Jörgen
Jörgen Hellberg, my webbsite www.hellberg.photo

ChrisR
Site Admin
Posts: 8618
Joined: Sat Mar 14, 2009 3:58 am
Location: Near London, UK

Post by ChrisR »

I tried a ring flash, and table-tennis ball diffuser. I found I could get better resolution from a lower NA objective, because I could light the subject better.
I'm referring to the infinite objective, which is perhaps a little more "blunt".
Chris R

keks
Posts: 44
Joined: Sun Mar 13, 2011 10:41 am
Location: Austria

Post by keks »

Hello Pau/Charles/Chris, thx for your replies, i have clearly mixed up something here, but i got a better picture now. Im not necessarily into going to 100x, but im interested to know about possibilities and limitations. The Nikon 60x ELWD is clearly a good compromise when going to the limits concerning resolution.

As diffraction spoils the resolution capability of objectives, especially of those with high magnification, i guess there is a way to calculate the actual resolution on the sensor, considering diffraction too? Comparing that numbers would give a much better picture when comparing the possibilities with different lenses. Maybe someone can provide a formula to calculate that? (or a link directing to it)

JH, seems like diffraction takes its toll at 100x, but especially the butterfly scales look better than I had expected.

And a little thought about illumination: LED's (even high CRI's) are available with thicknesses below 1 mm, I could imagine well to squeeze some of these into a rather small gap.

Aussie Phil
Posts: 17
Joined: Mon Dec 19, 2016 2:53 pm

Post by Aussie Phil »

Re lighting. What about illumination through or around the objective? Or fibre optics?

Things have moved on tremendously since I last did any photomicroscopy, so I'm curious to know how newer techniques can be applied at this range of magnification.

keks
Posts: 44
Joined: Sun Mar 13, 2011 10:41 am
Location: Austria

Post by keks »

There are objectives with integrated illumination such as the nikon bd plans, but I have no clue how to get the light inside when mounting them on a camera. I was mentioning the LEDs because there are very tiny and flat ones available that can be place between objective and object even if the gap is just few millimeters.

ChrisR
Site Admin
Posts: 8618
Joined: Sat Mar 14, 2009 3:58 am
Location: Near London, UK

Post by ChrisR »

The higher the magnification, the more you have to diffuse the light. That's hard with narrow spaces and point-source LEDs.
Yes, BD objectives' built in illumination path is hard to use on a camera, as But they are good at say 40x with the shroud removed. Short WD but skinny metal support. Through through-lens lighting is possible, but hard. Easier to find a microscope.

See Charles Krebs recent Butterfly scales - Nigh NA, + epi light + years of experience.
Chris R

keks
Posts: 44
Joined: Sun Mar 13, 2011 10:41 am
Location: Austria

Post by keks »

I was thinking about these flat LED's with larger surface, I have seen some high CRI-models from Yuji, for example with about 10-20 mm2 surface area (didn't check all models yet), thicknesses below 1 mm.

Considering an object size of about 1 mm2 at the discussed magnifications, wouldn't it come close to diffuse light when one would place lets say 4 of these LED's within few millimeters to the object?

ChrisR
Site Admin
Posts: 8618
Joined: Sat Mar 14, 2009 3:58 am
Location: Near London, UK

Post by ChrisR »

They would help, but an ordinary flash and a mirror or two can provide plenty of light.
If you put a small tube of paper around the subject (red), you make the light source as large as the space can allow:
Image
You have to consider the lighting from the bug's view. You need a big evenly lit sky (to begin with).

As in the sketch, imagine you're laying on your back in the middle of a room 6 m across. The walls are bright, but only 2m high. The ceiling is BLACK!

You can only achieve that marvellous situation if you cut a paper tube 6mm diameter, 2mm long, and place it very carefully! :(

Yes, a 60x NA 0.7 is a good option.
Chris R

keks
Posts: 44
Joined: Sun Mar 13, 2011 10:41 am
Location: Austria

Post by keks »

The paper cylinder is a good idea.

Im having a close look on the nikon bd plans (or bd plan apos), when stripped there seems to be much better situation concerning illumination:

An example be seen here:
http://www.photomacrography.net/forum/v ... ective+20x

From what i see on the picture of the stripped bd plan, there is about 1 additional millimeter to place flat LED's closely around the cylinder in the center in a much better angle to the object (at least for that 20x). Sadly i don't have any lens dimensions available to make such a nice drawing;)

If no-one tells me thats nonsense i guess i can't resist to try that.

ChrisR
Site Admin
Posts: 8618
Joined: Sat Mar 14, 2009 3:58 am
Location: Near London, UK

Post by ChrisR »

Yes that's what I said here:
ChrisR wrote:(BD objectives can be useful with the outer barrel removed, to leave a skinny glass support which makes them useful. Above about 60x, though, the WD is extremely short, and they're sprung-ended. If those usefully come apart, I don't know how!)
I don't know if the WD from the glass is actually more, I/we suspect it's the same optics. Note that the shroud design varies even for the same objective.

In the case of the 20x, the ELWD is the same NA so there's less point trying it.
The 40x is a higher NA.

The Apos are all sprung so barrel removal is a query.


If you want you can try this of course. This was 6 years ago. Leds will have more options now.
Image

You'll find pictures of several BDs with shrouds removed on the forum, but they can be hard to search for.

20x ELWD - not the one I was looking for:
Image
Chris R

Post Reply Previous topicNext topic