Which microscope objectives do not need compensating optics?

[rjlittlefield]

This query is prompted by a recent exchange about what objective I'd recommend to use as a macro lens on a bellows. I posted it to a couple of the Yahoo microscope groups also, and I'll be most interested to hear some answers, either there or here.

The query:

I'm looking for info/guidance about which objectives will work OK with no tube lens and with non-compensating eyepiece or relay optics. This includes using the objective as a macro lens, focusing directly onto a photo sensor.

Up to now, I've been lucky -- the few objectives I've used have worked OK.

However, I'm aware that some objectives are designed with large amounts of lateral chromatic aberration intended to be canceled by compensation in a tube lens, eyepiece, or relay optics. I'd like to avoid such lenses when purchasing, but I don't know which ones they are.

Can you help reduce my ignorance? Where do I go to find info on the amount of compensation required by specific objectives? Even a categorical list of which model lines need high compensation would be very helpful.

Can anybody here shed some light?

Thanks!
--Rik

[Charles Krebs]

Wow... lots going on since I took off last week.

As to this topic....

The Nikon "CF" series objectives are the only "finite" objectives I know of in which chromatic aberrations were corrected "fully" in the objective, so there was no color correction in the eyepieces. (These were the last finite type Nikon made before going to "infinity" optics). However I do believe that the CF series eyepieces did provide correction for field flatness. As mentioned elsewhere, this may not be too serious an issue using extended DOF techniques via image stacking.

Zeiss and Leica generally provided greater chromatic correction in the eyepieces. While I have never seen any data on the objectives themselves, I have seen some info on the amount of correction provided by various eyepieces. (I'll try to find this file and give some hard numbers). As I recall, the Zeiss eyepieces gave about twice the amount of correction as Olympus. There was no standardization between manufacturers, so I think the only way to be sure if results would be satisfactory would be to try each individual objective.

At least "on paper", it would seem that the CF series, 160mm tube length objectives might be the best candidates for use directly on a camera bellows. But it is probably not that simple. I have several CF series objectives. One (which I have no longer) was a 10X LWD (long working distance) that gave color errors I could not live with. I have two other 10X that are OK, so I don't know if it was just that one sample or something more common.

Frankly, I have only made one stack using a microscope objective on bellows. The objective was a Nikon CFN Plan 10x, and it worked well. DOF for each individual image is extremely small due to the fairly high NA.

A couple other things to consider...

Most objectives from the era we are talking about were designed to provide an image circle of about 20mm. This is far smaller that the diagonal of even "reduced frame size" SLRs. If the subject is centered with no detail toward the edges then it might work out OK. If you have the need to record great detail across the full frame, they may not provide it. It is probably reasonable to expect that some objectives will perform far better at the extremes of the image circle (and even provide larger useful image circles) than others. There were also lines of objectives that would provide larger image circles (about 26-30mm) that might be a better choice if you want to record detail as far out to the edges as possible. (Olympus S-Plans and Nikon CFN Plans are two such series)

Another consideration is that objectives are designed to be best at very specific working distances. If you stray too far from this designed distance you will increase spherical aberration. While this is generally a much larger issue with powers 20X and up, it would probably be wise to keep it as close as possible with a highly corrected 10X as well.

... let me see if I can locate the file that had some eyepiece correction data and post that...

[rjlittlefield]

Charlie,

Thanks for the feedback, further info, and clarification of concerns.

What you say has the general flavor of some other responses on the Yahoo Microscope group, although (rats!) there are some contradictions in the details. My query is posted at http://tech.groups.yahoo.com/group/Micr ... sage/34601. Further exchanges should be easily tracked with Yahoo's threaded interface (scroll to bottom of page).

I need to reaffirm for anybody tuning in fresh, that objective-on-bellows is only sort of a peculiar hack that (in my experience) works out surprisingly well. It's not a substitute for a high quality microscope, but for resolution over a limited FOV, it can beat the socks off even top-of-the-line macro lenses, for a lot less money.

--Rik

[Charles Krebs]

OK...

I've provided links to two jpg's that Ted Clarke sent to me when we were discussing this issue a while back. I don't know what book this came from, or have any idea of when it was published. They are a little tough to read, and don't include any Nikon eyepieces... but you can see that Zeiss KPL's corrected -1.4% for CDM (chromatic difference in magnification), while Olympus was generally around -0.7% for CDM.

http://krebsmicro.com/forumpix/CDM4.jpg
http://krebsmicro.com/forumpix/CDM5.jpg

I would not get too obsessive about exact numbers, but it would seem like the Zeiss finite objectives, excellent as they are, might not be a first choice if used as proposed without additional correction for CDM.

Other than that I still think it will boil down to a "try it and see" situation for each objective. When I get a chance I'll try a few that I have and give an assessment of color problems when used in this manner.

[rjlittlefield]

For the permanent record...

Recall that my question

Which microscope objectives do not need compensating optics?

was prompted by an earlier question asked by George Dingwall:

What would you suggest would be a good X10 lens for me to get?

It now seems that my answer to George's question should be something like this:

First, try whatever you have on hand, being careful to place the shoulder of the objective's mounting threads at least 150mm from the camera's sensor. The exact distance is not critical. Extensions much less than 150mm should be avoided, but longer extensions are OK and may give a better match between image size and the camera's sensor. If you do not already have a 10X objective, or the one that you do have does not work well, then buy any inexpensive 10X NA 0.25 achromat designated to work with 160 mm tube length. If you have a choice, get one designed to work with no cover glass; these are typically designated "160/-". Avoid apochromats and any other more highly corrected objectives that are designed to work with "compensating" eyepieces.

I'll copy below some helpful references and pieces of info that have come to me in the last few weeks.

I offer thanks to everyone for contributing to my education.

--Rik

Your post reminded me of an excert from "Photomicrography: a
comprehensive treatise" by Roger Loveland. He mentions exactly what
you've found with using objectives, but is backed by an exceptional
background in optics and microscopy. The following is found on pages
51-52 of volume 1:

"The compound microscope system usually begins with a 2X objective
which has a focal length of 48 mm, or about 2 in. Therefore there is
an overlapping field of application in which either the simple or
compound microscope may be chosen. The magnification range of the
simple microscope is extended upward (from about 25X) by the use of
reversed cine lenses of short focal length, as discussed in Chapter 1
and illustrated in Figure 1-30. The simple microscope is used when
the size of the field is important for the reasons discussed on p. 32.
On the other hand, it is not generally realized that an objective
made for a compound microscope can be used advantageously as a
simple microscope when high central definition and contrast are more
important than size of field. This is partly becaue there are such
stern warnings in some texts never to use such an objective without an
ocular. Actually, within the central field the ocular can only grade
definition and contrast as it extends the area of acceptable field!
With the 2 and 4X (48 and 32 mm) objectives, this area of markedly
superior quality is not negligibly small. These facts are important
to motion photomicrography, since the fields of the motion picture
frame are quite small. The size of the superior central (paraxial)
field diminishes rapidly with the decrease in focal length ofthe
objective, but that of the 10X (16 mm) achromatic objective is still
probably adequate to cover a 16 mm motion picture frame in a superior
manner. On the whole this does not apply to apochromatic objectives,
except with monochromatic light, since part of its correction is made
with the ocular."

So it seems with achromatic objectives of up to 10x you can use these
objectives with no comprimise in optical quality, in fact they're
likely the best available option. The motion picture photomicrography
is actually very applicable today considering the same sizes of CCD
sensors too. LCA can be generally disregarded in lower mag/NA lenses.

The book I cited earlier (its a two volume set) is one of the best
resources for microscope and photomicrography information circa the
1960's I've ever come across.

Today, if FOV can be limited, most modern research microscopes
project the image directly onto the image sensor even when
Apochromatic objectives are used. The term "simple microscope" lost
some of its meaning over the last decades. Modern Nikon systems use
relay lenses only to adjust FOV to match pixel-size with the
resolution limit of the objective.

If Rik purchased Nikon CF lenses (even APOs) that are made for 160mm,
he should be able to obtain "perfect" results with a setup without
relay lens. If he used CFI60 lenses, he will require a tube-lens to
complete the setup.

Personally, I would go with a Nikon CF PlanAPO 10x connected to a
10MP camera using a Bayer filter (e.g. Nikon D200). Since the lens is
corrected for 25mm FOV, cropping is required with a DX format image
sensor. Using a 1.4x teleconverter (of course a Nikon one!) or a
18.0mm by 11.1mm image sensor (smaller than DX), no cropping will be
necessary.

Such a setup is perfect to produce image stacks. Nothing beats its
resolution at this magnification. But with an APO lens, working
distance (WD) is very limited (around 4mm or less).

Hi again,

Leitz is the lowest of the typical compensating lens designs I'm aware
of, however Nikon's CF or CFI (CF are 160mm, CFI are infinity
corrected) would be better as they essentially correct entirely for
all aberrations within the objective itself.

I think Gregor's choice is then likely the best of a Nikon CF PlanAPO.
As you've mentioned a 20X has worked successfully, so don't limit
yourself to anything lower than that. Once you're into 40X lenses
(NAs of approx. 0.65) the lenses do start showing problems resulting
from misuse more quickly, but direct projection from an objective
without an ocular shouldn't be too much of a problem.

It is best though to focus the objective with the appropriate eyepiece
for the objective (a Nikon eyepiece with a Nikon objective for
example) and then move the position of the CCD so that it is in focus
without changing the objective's focus. Otherwise you may be using a
non-optimal focal position of the objective and may affect your image.

--- In Microscope@yahoogroups.com, "microscopy@..." <microscopy@...>
wrote:
>
> I forgot to add the URL of the figure I was talking about.
> You can find Loveland's plot showing the tolerance to changes
> in optical tube length at
> http://www.science-info.net/docs/etc/Tube-Length-na.gif

An excellent reference. This provides very useful data that
quantifies the usual advice that "tube length is important".

Clearly high NA objectives are quite picky, especially if dry.

But do I read the graph correctly that for an NA=0.40 dry objective,
the tolerance is more than +-25 mm of tube length, and that for
NA=0.25, tube length is essentially irrelevant (+-200 mm)?

Are actual lenses as tolerant as these numbers suggest?

The answer is yes and no to your question.

They are exceptionally tolerant at low NAs to degradation within the
image the lens produces, except with an extended tube length you will
observe a greatly increased magnification in the projected image, and
as the image is larger it will be dimmer as well (only so much light
to spread over a larger image).

So ignoring the risks of adding empty magnification and not having
sufficient illumination they are exceptionally tolerant.

In offline email, someone confirmed to me that

Frequently, I am using a low-cost Leica C Plan 10x and C Plan 4x
lens. I got both on eBay for almost nothing. They were new. The main
reason I got these lenses is the fact that their thread is an
M25x0.75mm, which fits loosely in a C-mount female thread (1.00"-32).
Hence I can mount them easily onto any C-mount setup I have.

They are so called infinity corrected lenses, which means that their
optical tube length is beyond 150mm. I never measured it nor did I
use a tube lens. But by moving the lens a little further away from
the object than its focal point distance, the lens forms a real image
at a convenient distance. The quality of the image is very useful and
compares nicely with the one of a 160mm 10x NA 0.25 Achromat when
used with its "perfect" optical tube length of 150mm. So, this would
in a way proof the inertness of low power objectives to changes in
optical tube length.

And finally (or perhaps I should say "first", considering the publication date!):

We will discard the eyepiece and allow the objective alone to project the image. This was the almost universal practice in former days, and the one used exclusively by Dr. J. J. Woodward, whose work, as a whole, has never been excelled. It will not answer with apochromats, which require the addition of a compensating or projecting ocular to complete their corrections; but the case is different with achromats, which, properly corrected, will give a sharper and flatter field, when used alone for projection, than in combination with the ordinary Huyghenian-eyepiece."

- - - end of post - - -

[georgedingwall]

Hi Rik,

That's some great information you have collated on this subject.

I've been keeping my eye out for 160/- objectives, but there have not been many outside of North America. Usually the postage makes buying them from there uneconomical.

I'm determined to find a microscope objective which will produce the sort of quality I wnat to achieve.

Thanks for the details.

Bye for now.

[rjlittlefield]

Adding recent information to this thread...

Charlie mentioned above having one good experience with a "Nikon CFN Plan 10x".

After further discussions offline, I purchased one of those lenses through eBay, $100 for a Nikon CF N Plan Achro 10X NA 0.30, 160/0.17.

It is superb. No visible CA even on closest examination. 9.22 mm working distance. At nominal extension of 150 mm, 10X onto the sensor, it is crisp almost to the corners of my 22.7 x 15.1 mm sensor (27.3 mm diagonal). This is without using a cover slip.

For sample images, see http://www.photomacrography.net/forum/v ... php?t=4416 and http://www.photomacrography.net/forum/v ... php?t=4464 .

--Rik

[augusthouse]

Yep,
I originally printed this thread and it's in my most important reference material file - which is getting larger and larger....

Craig