3 part comparison.

[apochronaut]

This was stimulated by some questions from a couple of members.
The pictures are in two sets. The first set is of a prepared diatom, with a .33mm sample thickness, similar to what would be encountered with many average grade prepared slides and routine live samples.
The second set is of diatoms again but with a .21mm sample thickness or similar to what many carefully prepared samples are, when using recommended cover slips.

The objectives are most of the 34mm parfocal 4mm(+) ( 40, 45 and 50X) objectives) made by AO for their infinity corrected microscopes, that are cover slip corrected. There four that I know of missing. The objectives are # 1116 45X .66 achromat, # 1078 45X .66 achromat, # 3007 45X .66 achromat, # 1023 40X .66 planachro, # 1309 40X .66 advanced planachro, # 1128 40X .66 advanced planachro, # 1323 40X .80 planapo, # 1028 50X .85 oil achromat( used dry, water immersed and oil immersed) and # 1016 50X .80 oil advanced planachro ( used dry, water immersed and oil immersed). The condenser was a 1.25 N.A. abbe aspheric used unoiled set for kohler illumination.

[apochronaut]

5 more of the same diatom.

[apochronaut]

Last 4 of that diatom.

[apochronaut]

The next set is of the same objectives but used to view a sample diatom mounted with a thinner mount. One more similar to a recommended sample thickness.

[apochronaut]

4 more

[apochronaut]

last 4.

[gekko]

Thank you for this comparison. For one thing, it shows very clearly the attention that we need to pay to sample thickness (and may explain why I've been getting very bad results recently).

[zzffnn]

Thank you, apo.

I find this comparison very informative and helpful.

#1026 indeed looks very good, if not pushed too far. #1016 seems to lack contrast?

Your third to the last image "# 1016 50X .80 oil advanced planachro. .21 sample thickness. water immersed" seems to have some unexpected fog? It should not be worse than thin- dry or thick-water of the same objective?

[apochronaut]

I have a couple of notes of explanation and as well some comments.

I deliberately did not choose diatoms from a test slide, given the fact that test slides are designed to rule out variables for the purpose of testing. In practice, as Gekko notes, with live subjects and even with some prepared slides, uncontrolled variables are included in the sample and can affect results adversely. What I was attempting to do here is see how a series of objectives made over a 25 or so year period in the same factory, corrected for the same microscope system, handle those variables. With dry objectives, the air gap is a fixed variable, which is somewhat affected by sample or coverslip thickness. With immersion objectives, the added influence of the immersion medium becomes an influence that is greatly affected by the sample thickness.
It is clearly recognized that the N.A. of an objective has a great deal to do with it's ability to continue to resolve detail over a broader range of sample thicknesses. This is why for instance, a student objective of 40X magnification, can have an N.A., as low as .45 and 95% of microscopes carry a 40X objective with an N.A. in the mid- 60's. These objectives are very forgiving of sample thickness. However, if you were to break down 100 objectives with an N.A. of .65, you would find a great deal of commonality between them in terms of design. It is not just the N.A., that determines an objective's response to sample thickness or the n of it's immersion medium, it is also determined by factors in the objective's internal design. It would be perfectly possible to make an objective of N.A..50, that had a terrible response to sample thickness variation and it is possible to design an objective , with an N.A. of .85, that has a great capacity to handle various sample thicknesses, as well as various immersion media. The N.A. is just one variable but due to a great deal of homogeneity in many normal objective designs, it has become a standard mathematical measuring stick for an objective's overall performance. Many examples exist that break this rule.
In this test, when I obtained a result that was unusually subpar, such as Fan's observation that the # 1016, 50X .80 oil planachro image was fogged, when using water immersion, I repeated the test with either a different objective of the same type, or I cleaned everything and redid it. The low contrast result with that objective, when using water and a thin sample is accurate.
There would be a natural assumption that if the older # 1026 50X .85 oil achromat gave good results then the newer # 1016 50X .80 advanced planachro should be as good or better, Fan. Well, this would only be within the parameter of the design specifications. The 1016 does have better planarity than the 1026 but plan objectives are very complicated designs incorporating many glass elements of precise refractive indices , precise air gaps , and sophisticated scatter reduction. An achromat, even a high N.A. oil achromat, is relatively much simpler and clearly more forgiving. The third to the last image , which yielded such a muddy picture was dipped into the vary same water as the image that precedes it, and then repeated after cleaning. The result was the same. When the two objectives are immersed in oil, the whole situation changes. Although the 1026, still has slightly better contrast, the resolution is about the same and the planachro oil objective is finally reaching it's potential. The plain fact is that the 1026 is just a better objective, in all ways except planarity, where the 1016 achieves it's target specification quite well. It was designed with fluorescence, histology( where planarity is absolutely necessary) and DF in mind, so a comparison in BF might be a bit unfair. I have never really liked the 1016 and hardly ever use one. When AO redesigned the objective into a D.I.N. barrel, as the # 1758 50X .80 oil neoplan for the 400 series, they dropped the iris and greatly improved the objective, with the benefit of low dispersion glass and computer ray tracing. The original 1016 was designed in the 1960's.
The star of this show, really is the # 1026 50X .85 oil achromat with iris, objective. Even as a dry objective with either a thick or thinner sample , it produced an adequate image, fully usable and in line with the results of the dry .66 achromats. When water immersed, with a thick sample, which is a very nice comfortable situation, it has the lowest chromatic aberration of all the objectives, other than the 40X .80 planapo. It also has exceptional resolution, given that it is immersed in water and is an oil immersion objective. Given the fact that it is the highest N.A. objective in the test, it really does reveal that when the sample preparation is outside of specification, N.A. is not the extreme arbiter of image quality.

[zzffnn]

apo,

I just could not understand why image from #1016, used dry, on 0.21 mm sample would look better than
image from #1016, immersed in water, on 0.21 mm sample.

To an oil objective, water is optically much closer to oil than air does, right? So one would expect more optical error (thus worse inage) when oil objective is used dry, than immersed in water.

[apochronaut]

There are a lot of irregularities with those immersion objectives, not readily explainable. In fact, the dry and water images are better with the thicker sample too. A certain amount of what took place here is counter intuitive but In the coming week, I will think it through. Clearly, the 50X advanced planachro is persnickety and must have an appropriate settup. I think , what the reason for that is internal. Longitudinal spherical aberration seems high with that objective, when not used to spec.