Do I need another objective?

[Free2Fish]

Great point about that objective Pete. I was really only half considering it.
I’m still trying out different techniques to get a bit more resolution out of my objectives. As I write, an LED/flash unit is winging its way to Canada. That may prove to be helpful.
I also sometimes wonder if my immersion oil had an expiry date. The stuff must be over 50 years old. Or perhaps too thick? When focusing with the 100X oil, the specimen keeps moving around. It doesn’t (or didn’t today) happen with the 50X oil. So it may involve contact with the cover glass.
Must pull out my micrometer to measure cover glass thicknesses and use only ones within spec.
Anyway, the whole adventure is an ongoing process that I hope will end up with some decent photomicrographs.
I keep going back to the website of Martin Kreutz where I’ve seen some incredible work. He’s using an older Olympus but it has been updated with the best components available.
Check out his images of a Paramecium caudatum. Scroll down to some of his later photos.
https://realmicrolife.com/paramecium-caudatum/

Harry

[Scarodactyl]

I'd watch for a favorable auction, not a buy it now. I got mine for 450, this one is starting at 600 but it's also VC (a little better than apo) https://www.ebay.com/itm/315246893929

[Free2Fish]

I'd watch for a favorable auction, not a buy it now. I got mine for 450, this one is starting at 600 but it's also VC (a little better than apo) https://www.ebay.com/itm/315246893929

Thanks, great tip! Unfortunately many sellers, including this one, won’t ship to Canada.
I’ll start watching the auctions.

Harry

[Free2Fish]

The money might be better spent on something like a mirrorless camera that can capture every bit of resolution you already have, while eliminating vibration and the like.

Pete,
I've gone back to this comment, realizing I haven't a clue about how mirrorless cameras compare with the old Canon Live View in terms of camera shake. My Canon APS-C sensor with 4.3 um sized pixels will handle all the resolution my objectives can provide so that isn't an issue.
Is there any information out there that suggests that camera shake with a mirrorless camera is less than with an APS-C camera?

Harry

[PeteM]

Given that you're shooting at higher magnifications, you're probably OK. I was imagining all those extra and fat pixels. But as you say; the extra or larger pixels from one of the newer mirrorless cameras won't help much at 400X and above.

I have heard of vibration complaints from some Canon EOS users (70D etc.) over at the photomicrography website, even with silent mode. Not sure why - I've skipped over those posts not having any Canon gear. One would think that with live view, remote actuation, and (soon) flash illumination that vibration shouldn't be much of a problem.

[Scarodactyl]

I've never had problems with canon efcs, it is vibration free during shooting.

[hans]

I’ve wondered about the formula that suggests that when condenser NA is lower than Objective NA the result is an average of the two. Intuition suggests that NA should revert to the lower of the two. Does anyone know if there is a mathematical basis for the averaged formula?

I wondered the same thing. It has been a few years since I looked into it but the averaging thing goes all the way back to Abbe. It was based on some idealized analysis where the object is a grating and resolution in real-world situations tends to be better than predicted by averaging. Analysis for anything more realistic, even idealized pairs of points, was way more complicated. I don't remember details beyond that but I'll see if I can find the papers I downloaded at the time and then never got around to reading carefully.

[apochronaut]

All you need is an iris equipped objective, a .90 corrected condenser and a corrected condenser that allows the objective to work at or above the calculated combined N.A. to determine that.....and probably a good diatom slide, maybe something like p.angulatum.

[Free2Fish]

I have a strong feeling that with a dry condenser one is never going to see the potential available in those 1.4 NA plan apo objectives.
I’m going to engage in some day dreaming now. I’ve got a Zeiss apl 1.40 condenser sitting around doing nothing. What’s to stop me from slipping out of the lab and into the machine shop to fabricate a new dovetail that centers and raises this condenser into my Nikon condenser slot? It’s only glass that focuses light so why wouldn’t that work?
Pete?



Harry

[zondar]

A dry (air-gapped) condenser will strictly limit the effective NA of any subsequent objective to a theoretical maximum of 1 (meaning somewhat less than 1 in practical terms).

I've heard "an average of the condenser and the objective's NA" here more than once, but that's nonsense. The net, in fact, can be no greater than the lesser of the two.

If you use a NA>1 objective with oil, you need a proper NA>1 oiled condenser, or you waste your objective's potential and might as well use a dry objective.

[PeteM]

A dry (air-gapped) condenser will strictly limit the effective NA of any subsequent objective to a theoretical maximum of 1 (meaning somewhat less than 1 in practical terms). . . .

As Hans noted above, it's likely a bit more complicated. The formula here (Thorlabs site) suggests an additive effect, such that a 1.4na objective and a 0.9 condenser might yield an effective system numerical aperture a bit above 1.0. They note (again, as Hans suggested) this is a somewhat simplified approximation. https://www.thorlabs.com/newgrouppage9. ... p_id=14352

In actual use (comparing my own 60x 1.4 na Apo oil objectives and a couple of .95 na 60x dry objectives -- the latter require significant fiddling with a correction collar to get its best results. With oil immersion and a relatively homogenous illumination column (save the last bit to the not-yet-oiled condenser) a sharp image is easier to get - at least for me and despite attempts to get the coverslip correction just right.

Then, there's the question of what's readily available. Cheaper 60x dry plan achro objectives, without a correction collar, have generally proved a disappointment to me. Cheaper 100x plan achro oil immersion objectives, with no need of a correction collar, are often quite good.

[zondar]

The result of those equations is the minimum distance between two points before they can be individually resolved - the "resolution." So if the NAcd is less than NAobj, then from the second equation the result is a greater distance (a worse resolution) than if the condenser's NA matched that of the objective (as in the first equation).

That's to be expected, right? Yes, indeed. But here's the problem: The equation doesn't take the refractive index of air into account, so it's really useless to the question of dry vs. oiled optics.

Once the cone of light exits the condenser and hits air, you are stuck with a theoretical maximum net NA of 1 regardless of how high an NA the objective has.

Edit: The Thor labs diagram is superficially a bit misleading, as the orange cone on the right is not the exit cone of light (that's the "from sample" dotted lines shown inside the orange cone), but the acceptance cone of the objective. If no light fills that periphery of the acceptance cone, the extra NA is wasted. Still, I imagine it's possible for the sample to diffract some light outside the normal exit cone that a higher-NA objective could still collect.

[zondar]

Thinking about the last point I made a bit more: Yes, provided that the sample radiates light broadly enough to fill the acceptance cone of the objective, some improved resolution can be obtained. But of course many or most biological samples requiring high magnification are nearly entirely transparent, so in that case the improvement would be minor, if any is observable at all beyond a placebo effect.

The bottom line: Have a high-NA (>1) oiled objective? Use an high-NA (>1) oiled condenser, ideally with NAcd=NAobj, or you won't get the good results you are expecting.

[PeteM]

On edit, I see you've addressed this, Zondar:

What's not clear to me is if Snell's Law, ray tracing, and the Abbe diffraction limit entirely describe what happens when a specimen is illuminated and then attempts to illuminate the entire angle "seen" by an objective and described by its numerical aperture.

Imagine, for example, that the specimen is self-illuminating (say, with fluorescent micro-spheres at the limit of optical resolution). Then, something like a 1.4 n.a. oil immersion objective should be able to "see" the entire wide angle of its aperture.

If that specimen is illuminated by, say, a .90 n.a. aperture condenser, then ray tracing would say it can't fill the entire aperture of the objective. As you say, the condenser itself and the 1.0 refractive index of air would predominate. There aren't any "rays" available to fill the full aperture of the 1.4 na objective.

It's possible for the field iris and condenser to entirely fill the field of the objectives field of view (at the specimen plane). The rays of a 0.90 n.a. condenser won't trace into the full aperture of the objective, but the specimen itself is fully illuminated (at the specimen plane) to fill the angle required to meet its n.a. and sine-angle spec.

The question is -- to what extent Is it then possible for the specimen itself to scatter and reflect some of that light so it effectively becomes somewhat "self-illuminating" and fills, to some extent, the objective's wider potential aperture??

To add - I surely agree that a 1.4 n.a. objective is best used with an oiled condenser. But in my experience (often with stained specimens and DIC) that same objective performs significantly better than a 0.95 n.a. dry objective, even with a dry 0.90 plan achromat condenser.

[zondar]

Adding in the characteristics of a subject confounds the situation beyond the usual equations and into the realm of ruminations, but sure.

Anyway, I'd suggest that the improvement some people find when using very expensive elite-quality objectives in the discussed situation has more to do with them being just plain better than typical achromats. After all, shouldn't we expect a superbly-corrected >=1.25 NA APO objective to beat most any lower-quality dry objective even if the APO is theoretically restricted by a dry condenser to an NA of 0.95?

[PeteM]

Fair point about objective quality.

In my experience, the cheaper Chinese 60x dry plan achromats with no correction collar are commonly a disappointment. I'd rather add a drop of oil to a 50-60-63-100x plan achromat oil objective. The 50x oil immersion finite objectives from Nikon and Olympus with an iris can often be a relative bargain - and also ideal for darkfield.

In the comparisons above, I've used multiple 60x plan apo oil objectives (Leica, Nikon, and Olympus infinite and Nikon and Olympus finite) and found them better than otherwise very good 60x Plan Fluor Olympus and Nikon dry objectives with correction collars. These better dry 60x objectives are good, but not to the level of their oil-immersed brothers even without oiling the condenser. This isn't usually with living (and thus mostly transparent) tissue.

[hans]

A dry (air-gapped) condenser will strictly limit the effective NA of any subsequent objective to a theoretical maximum of 1 (meaning somewhat less than 1 in practical terms).

Where did you read this? The average thing is everywhere in guides and textbooks, it's just rare that they attempt to explain any sort of physical/mathematical basis for it which is what Harry asked about.

I've heard "an average of the condenser and the objective's NA" here more than once, but that's nonsense. The net, in fact, can be no greater than the lesser of the two.

The idea to use the average NA to predict resolution doesn't come from internet forums. The oldest reference I saw was Ernst Abbe himself in 1873. (I didn't try to find it or read it because it's in German, only read what some newer papers summarized about it.)

That's to be expected, right? Yes, indeed. But here's the problem: The equation doesn't take the refractive index of air into account, so it's really useless to the question of dry vs. oiled optics.

I don't follow this part. Are you saying that it matters how the illumination NA is limited? As in, the behavior of an immersion condenser wide open but dry is fundamentally different than the behavior of the same condenser oiled but with the aperture set a little below 1? (These are not exactly equivalent, of course, since the intensity falloff as the angle approaches total internal reflection is more gradual than with an aperture stop, but I don't see a fundamental difference.)

Once the cone of light exits the condenser and hits air, you are stuck with a theoretical maximum net NA of 1...

Air does limit the illumination NA to 1 or less, but what is "net NA"?

...regardless of how high an NA the objective has.

Assuming "net NA" means something like "NA of an objective that would give equivalent resolution if the condenser weren't limiting" then the papers I looked at disagree, unless I am misunderstand something.

Three papers I had saved are:
1950, H. H. HOPKINS AND P. M. BARHAM, The Influence of the Condenser on Microscopic Resolution
1952, JOHN R. BAKER, Remarks on the Effect of the Aperture of the Condenser on Resolution by the Microscope
1991, D. J. GOLDSTEIN, Resolution in light microscopy studied by computer simulation

[zondar]

Oil is commonly used with microscopes at optical interfaces because its refractive index (RI) is higher than air (~1.5 vs. ~1). This allows light to be bent at a sharper angle than at glass-air interfaces (otherwise the rest of the light would be reflected internally). The sharper angle allows a broader cone of light to exit the condenser.

With a glass-air interface, the angle of the cone of light exiting the condenser is limited. This then limits the cone of light entering the objective to the same angle (barring shenanigans caused by the subject being viewed), regardless of whether the objective has a larger angle of acceptance. The rest of the objective's acceptance angle would be dark and hence part of its potential resolution would be wasted.

So if there's a single glass-air interface in the condenser+objective path, whether at the condenser's output or the objective's input, then the net NA of the condenser+objective is also limited to a theoretical maximum of 1, and the common formula estimating the resolution of the combination wouldn't work anymore.

Note: I'm NOT an expert in the physics of optical systems. Someone with real knowledge will come along and correct me soon enough if need be.

[Scarodactyl]

Note: I'm NOT an expert in the physics of optical systems. Someone with real knowledge will come along and correct me soon enough if need be.

No offense but it sounds like Abbe did back in the late 1800s.

[Free2Fish]

Sounds like we should be able to test this empirically by checking out the circle of light hitting the objective back with a phase telescope. Anybody have any idea how NA might relate to diameter or area of illumination?
Never mind, tried a couple of things but nothing definitive. Back to having fun.

[PeteM]

. . . I've heard (from Stephen, on this site) that the PF ELWD 60x is about as good as it gets for this objective and expect to have one fairly soon. . . .

Going back a bit in this thread -- I can confirm that the PF ELWD 60C prism works very well with Nikon's dry 0.85na Plan Fluor objective - far better than the other prisms (noted above) that provided a DIC effect, but with banding or a failure to reach a sufficiently dark background.

[hans]

Oil is commonly used with microscopes at optical interfaces because its refractive index (RI) is higher than air (~1.5 vs. ~1). This allows light to be bent at a sharper angle than at glass-air interfaces (otherwise the rest of the light would be reflected internally). The sharper angle allows a broader cone of light to exit the condenser.

With a glass-air interface, the angle of the cone of light exiting the condenser is limited.

As I said in the previous reply, no disagreement that an air gap limits the illumination-side NA to 1 or less, that is well understood. I think Pete is asking the right question:

The question is -- to what extent Is it then possible for the specimen itself to scatter and reflect some of that light so it effectively becomes somewhat "self-illuminating" and fills, to some extent, the objective's wider potential aperture??

Unfortunately it seems like there is no easy/intuitive explanation of how exactly light exiting the specimen at angles exceeding the illumination NA contributes to image resolution. The analyses I saw used a lot of math (Fourier optics) or brute-force numerical simulation.

I don't understand your response to Pete's question:

Adding in the characteristics of a subject confounds the situation beyond the usual equations and into the realm of ruminations, but sure.

This then limits the cone of light entering the objective to the same angle (barring shenanigans caused by the subject being viewed), regardless of whether the objective has a larger angle of acceptance.

Isn't the point of this sort of analysis to predict resolution in the image formed by the light after passing some objects of interest? Yes angles stays the same if illumination passes undisturbed to the objective with no specimen present, but how is that fact relevant to the question of whether or not some intervening objects are resolved?

So if there's a single glass-air interface in the condenser+objective path, whether at the condenser's output or the objective's input, then the net NA of the condenser+objective is also limited to a theoretical maximum of 1, and the common formula estimating the resolution of the combination wouldn't work anymore.

Confused what you mean by "anymore", when are you saying it would work? Only if there is no air gap present? If so then I still have the same question as in previous reply, how is having illumination NA limited by an air gap different than simply setting the condenser diaphragm NA slightly less than one?

[Scarodactyl]

Oh, the normal plan 10x is a really good objective for its price, excellent corner to corner performance even direct on full frame and it also works surprisingly well with DIC, though I'd need to retest the best slider/condenser prism combo.

[zondar]

The common equation for resolution of a condenser+objective combination, as seen in the Thor Labs link, is highly idealized, containing only 3 variables (wavelength and the NA's), and hence seems more like a simple rule-of-thumb than one that can deal with real-life complexities (such as the subject that the light is passing through, changes in the RI of the medium(s) through which the light passes, etc.).

What I still wonder is how the formula (essentially averaging the two NA's) can ever be considered correct? That is, I still think that the net resolution should always be limited by the lesser of the two NA's, at least assuming there's no subject to confound things. (And yes, the situation isn't much different than stopping down a diaphragm, which reduces resolution too.)

I did take a look at the Hopkins and Barham paper noted above, and there's no way I'd have the patience to work through all that math! Even their conclusion was basically "it's complicated!"

Anyway, I don't feel confident enough in my knowledge to contribute to this conversation much further. If he's not busy, I call on Abbe visit us with his ghostly apparition to set everything right!

[Scarodactyl]

I call on Abbe visit us with his ghostly apparition to set everything right!

That would make him an Abberration.

[Free2Fish]

Thanks to everyone who contributed, it was a fun ride. Scarodactyl, you'll be pleased to know that I'm now following a few items on eBay, including a plan apo 20X.
I think I still want a 60X but it will be a dry, either plan apo or plan fluor, whichever falls off the truck first.

Harry

[charlie g]

Harry, thanks for another terrific thread, and especially thank you for the tip, and the link to Dr.Martin Kruetz's "Real Micro Life" website. charlie g

[Free2Fish]

You're more than welcome. In my search for good photomicrography techniques I came across another webste: https://www.plingfactory.de/pling.html
I believe he and Kreutz collaborate on some things.

Harry

[Scoper]

Great point about that objective Pete. I was really only half considering it.
I’m still trying out different techniques to get a bit more resolution out of my objectives. As I write, an LED/flash unit is winging its way to Canada. That may prove to be helpful.
I also sometimes wonder if my immersion oil had an expiry date. The stuff must be over 50 years old. Or perhaps too thick? When focusing with the 100X oil, the specimen keeps moving around. It doesn’t (or didn’t today) happen with the 50X oil. So it may involve contact with the cover glass.
Must pull out my micrometer to measure cover glass thicknesses and use only ones within spec.
Anyway, the whole adventure is an ongoing process that I hope will end up with some decent photomicrographs.
I keep going back to the website of Martin Kreutz where I’ve seen some incredible work. He’s using an older Olympus but it has been updated with the best components available.
Check out his images of a Paramecium caudatum. Scroll down to some of his later photos.
https://realmicrolife.com/paramecium-caudatum/

Harry

Wow..excellent site! Thanks for posting it.

[hans]

The common equation for resolution of a condenser+objective combination, as seen in the Thor Labs link, is highly idealized, containing only 3 variables (wavelength and the NA's), and hence seems more like a simple rule-of-thumb than one that can deal with real-life complexities (such as the subject that the light is passing through, changes in the RI of the medium(s) through which the light passes, etc.).

What I still wonder is how the formula (essentially averaging the two NA's) can ever be considered correct? That is, I still think that the net resolution should always be limited by the lesser of the two NA's, at least assuming there's no subject to confound things. (And yes, the situation isn't much different than stopping down a diaphragm, which reduces resolution too.)

I did take a look at the Hopkins and Barham paper noted above, and there's no way I'd have the patience to work through all that math! Even their conclusion was basically "it's complicated!"

Anyway, I don't feel confident enough in my knowledge to contribute to this conversation much further. If he's not busy, I call on Abbe visit us with his ghostly apparition to set everything right!

I made a separate thread for the NA averaging stuff:
Averaging condenser and objective numerical aperture when predicting resolution