Optics positioning for Köhler illumination and transmitted light

[Kyle Michael Douglass]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Hi everyone,

I have a question concerning trans-illumination and commercial
microscope designs that I am hoping someone on the list could answer.

On a "standard" inverted microscope, the condenser lens is free to move
in the vertical direction. This allows one to image the field stop onto
the sample and is how one places the microscope into Köhler
illumination. Additionally, and if I am not mistaken, the primary
requirement of Köhler illumination is that an image of the light source
should be formed in the back focal plane of the condenser.

My question is: how does moving only the condenser lens ensure that the
field stop is imaged onto the sample and that the light source is imaged
onto the condenser's back focal plane at the same time? If the sample
were to change axial positions, then it would seem impossible to ensure
that both conditions are met when only the condenser is free to move.

My guess at the moment is that the depth of field of the condenser is
large enough to accommodate a range of axial sample positions, but this
is only a guess.

Thanks!
Kyle

--
Kyle M. Douglass, PhD
Post-doctoral researcher
The Laboratory of Experimental Biophysics
EPFL, Lausanne, Switzerland
http://kmdouglass.github.io
http://leb.epfl.ch

[Doube, Michael]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

On 08/08/17 17:10, Kyle Douglass wrote:
> If the sample were to change axial positions

Then it will be out of focus with respect to the objective. Placing your specimen in focus under a low-mag objective is step 1 of Köhler illumination.

You can think of Köhler as a 5-part system: light source, condenser, sample, objective, final image plane. If any of the parts moves axially without a correcting adjustment in the others, you will lose either (or both) focus and/or even illumination.

Michael

--
Michael Doube, PhD
Lecturer, Comparative Biomedical Sciences
The Royal Veterinary College, University of London
Royal College Street
London NW1 0TU
United Kingdom

[phone] (Internal: 5503)
@mdoube

<http://www.rvc.ac.uk>

This message, together with any attachments, is intended for the stated addressee(s) only and may contain privileged or confidential information. Any views or opinions presented are solely those of the author and do not necessarily represent those of the Royal Veterinary College.

[Zdenek Svindrych-2]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Hi Kyle,
you're right, usually the field iris position is fixed, and only the
condensed lens (with aperture iris, DIC, phase, etc elements) can move.
There are three parts to the answer:

1) you're right again, the depth of focus of the lamp filament image is
quite large (few cm), because the filament image is magnified considerably,
maybe 10 times. But it's not the depth of focus of the condenser, it's depth
of focus of the optics between the lamp and the condenser.

2) on some systems (Olympus, Zeiss) you can adjust the filament position so
it's in focus in the BFP of the condenser.

3) usually there is a diffuser somewhere, making the "filament image in the
BFP of the condenser" idea less of a concern.

best, zdenek


--
Zdenek Svindrych, Ph.D.
W.M. Keck Center for Cellular Imaging (PLSB 003)
Department of Biology,University of Virginia
409 McCormick Rd, Charlottesville, VA-22904
http://www.kcci.virginia.edu/
tel: [phone]

---------- Původní e-mail ----------
Od: Kyle Douglass <[hidden email]>
Komu: [hidden email]
Datum: 8. 8. 2017 12:17:08
Předmět: Optics positioning for Köhler illumination and transmitted light
"*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Hi everyone,

I have a question concerning trans-illumination and commercial
microscope designs that I am hoping someone on the list could answer.

On a "standard" inverted microscope, the condenser lens is free to move
in the vertical direction. This allows one to image the field stop onto
the sample and is how one places the microscope into Köhler
illumination. Additionally, and if I am not mistaken, the primary
requirement of Köhler illumination is that an image of the light source
should be formed in the back focal plane of the condenser.

My question is: how does moving only the condenser lens ensure that the
field stop is imaged onto the sample and that the light source is imaged
onto the condenser's back focal plane at the same time? If the sample
were to change axial positions, then it would seem impossible to ensure
that both conditions are met when only the condenser is free to move.

My guess at the moment is that the depth of field of the condenser is
large enough to accommodate a range of axial sample positions, but this
is only a guess.

Thanks!
Kyle

--
Kyle M. Douglass, PhD
Post-doctoral researcher
The Laboratory of Experimental Biophysics
EPFL, Lausanne, Switzerland
http://kmdouglass.github.io
http://leb.epfl.ch
"

[Craig Brideau]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

As a related aside, with LEDs the emitter is very small compared to a lamp
filament, so it can actually be much easier to accidentally focus an LED
onto your sample compared to a filament source. Most LED systems
incorporate some form of lens or diffuser-based system to prevent this from
happening, but it is still a possibility depending on your particular
optical train.

Craig

On Tue, Aug 8, 2017 at 10:47 AM, <[hidden email]> wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> Post images on http://www.imgur.com and include the link in your posting.
> *****
>
> Hi Kyle,
> you're right, usually the field iris position is fixed, and only the
> condensed lens (with aperture iris, DIC, phase, etc elements) can move.
> There are three parts to the answer:
>
> 1) you're right again, the depth of focus of the lamp filament image is
> quite large (few cm), because the filament image is magnified considerably,
> maybe 10 times. But it's not the depth of focus of the condenser, it's
> depth
> of focus of the optics between the lamp and the condenser.
>
> 2) on some systems (Olympus, Zeiss) you can adjust the filament position so
> it's in focus in the BFP of the condenser.
>
> 3) usually there is a diffuser somewhere, making the "filament image in the
> BFP of the condenser" idea less of a concern.
>
> best, zdenek
>
>
> --
> Zdenek Svindrych, Ph.D.
> W.M. Keck Center for Cellular Imaging (PLSB 003)
> Department of Biology,University of Virginia
> 409 McCormick Rd, Charlottesville, VA-22904
> http://www.kcci.virginia.edu/
> tel: [phone]
>
> ---------- Původní e-mail ----------
> Od: Kyle Douglass <[hidden email]>
> Komu: [hidden email]
> Datum: 8. 8. 2017 12:17:08
> Předmět: Optics positioning for Köhler illumination and transmitted light
> "*****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> Post images on http://www.imgur.com and include the link in your posting.
> *****
>
> Hi everyone,
>
> I have a question concerning trans-illumination and commercial
> microscope designs that I am hoping someone on the list could answer.
>
> On a "standard" inverted microscope, the condenser lens is free to move
> in the vertical direction. This allows one to image the field stop onto
> the sample and is how one places the microscope into Köhler
> illumination. Additionally, and if I am not mistaken, the primary
> requirement of Köhler illumination is that an image of the light source
> should be formed in the back focal plane of the condenser.
>
> My question is: how does moving only the condenser lens ensure that the
> field stop is imaged onto the sample and that the light source is imaged
> onto the condenser's back focal plane at the same time? If the sample
> were to change axial positions, then it would seem impossible to ensure
> that both conditions are met when only the condenser is free to move.
>
> My guess at the moment is that the depth of field of the condenser is
> large enough to accommodate a range of axial sample positions, but this
> is only a guess.
>
> Thanks!
> Kyle
>
> --
> Kyle M. Douglass, PhD
> Post-doctoral researcher
> The Laboratory of Experimental Biophysics
> EPFL, Lausanne, Switzerland
> http://kmdouglass.github.io
> http://leb.epfl.ch
> "
>

[Kyle Michael Douglass]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Thanks everyone for the replies.

I am specifically thinking of an Olympus IX71 with a IX2-LWUCD condenser; I see no way to adjust the positions of either or the lamp or the collector lens immediately after the lamp and before the field iris. Do the optics in the very top of the microscope housing form a real image of the lamp filament in the open vertical space between the field iris and the condenser lens housing, or is the image of the filament at infinity? (I'm not in the lab to check.)

If it's real, then the image of the lamp filament is fixed in space because the lamp and collector lens are fixed. If it's at infinity, then Mike Model's answer would make some sense: there could be a field lens at the very top of the condenser housing that forms the real image of the filament and moves with the condenser lens, regardless of the position of the full condenser lens housing.

Thanks again,
Kyle

Dr. Kyle M. Douglass
Post-doctoral Researcher
EPFL - The Laboratory of Experimental Biophysics
http://leb.epfl.ch/
http://kmdouglass.github.io

________________________________________
From: Confocal Microscopy List <[hidden email]> on behalf of [hidden email] <[hidden email]>
Sent: Tuesday, August 8, 2017 6:47 PM
To: [hidden email]
Subject: Re: Optics positioning for Köhler illumination and transmitted light

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Hi Kyle,
you're right, usually the field iris position is fixed, and only the
condensed lens (with aperture iris, DIC, phase, etc elements) can move.
There are three parts to the answer:

1) you're right again, the depth of focus of the lamp filament image is
quite large (few cm), because the filament image is magnified considerably,
maybe 10 times. But it's not the depth of focus of the condenser, it's depth
of focus of the optics between the lamp and the condenser.

2) on some systems (Olympus, Zeiss) you can adjust the filament position so
it's in focus in the BFP of the condenser.

3) usually there is a diffuser somewhere, making the "filament image in the
BFP of the condenser" idea less of a concern.

best, zdenek


--
Zdenek Svindrych, Ph.D.
W.M. Keck Center for Cellular Imaging (PLSB 003)
Department of Biology,University of Virginia
409 McCormick Rd, Charlottesville, VA-22904
http://www.kcci.virginia.edu/
tel: [phone]

---------- Původní e-mail ----------
Od: Kyle Douglass <[hidden email]>
Komu: [hidden email]
Datum: 8. 8. 2017 12:17:08
Předmět: Optics positioning for Köhler illumination and transmitted light
"*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Hi everyone,

I have a question concerning trans-illumination and commercial
microscope designs that I am hoping someone on the list could answer.

On a "standard" inverted microscope, the condenser lens is free to move
in the vertical direction. This allows one to image the field stop onto
the sample and is how one places the microscope into Köhler
illumination. Additionally, and if I am not mistaken, the primary
requirement of Köhler illumination is that an image of the light source
should be formed in the back focal plane of the condenser.

My question is: how does moving only the condenser lens ensure that the
field stop is imaged onto the sample and that the light source is imaged
onto the condenser's back focal plane at the same time? If the sample
were to change axial positions, then it would seem impossible to ensure
that both conditions are met when only the condenser is free to move.

My guess at the moment is that the depth of field of the condenser is
large enough to accommodate a range of axial sample positions, but this
is only a guess.

Thanks!
Kyle

--
Kyle M. Douglass, PhD
Post-doctoral researcher
The Laboratory of Experimental Biophysics
EPFL, Lausanne, Switzerland
http://kmdouglass.github.io
http://leb.epfl.ch
"

[Zdenek Svindrych-2]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****


Hi Kyle,
ideally, you would use telecentric (or 4-f) arrangement of lenses, that is,
each pair of lenses is separated by the sum of focal lengths of that two
lenses. This leads to alternating field and aperture conjugate planes. But
for size/cost/flexibility/other reasons microsocopy companies use more
general lens arrangement. Moreover, many lenses are far from thin...
For example, in the Olympus L-shaped fluorescence illuminator the field and
aperture irises are next to each other, with no lens in between. And it
works :-).

Indeed, there is an additional lens in front of the condenser lens in the
Olympus LWUCD, but it's not in an 4-f arrangement (the focal length of this
additional lens is maybe twice the distance to the BFP of the condenser). So
while this additional lens helps to form a real image of the filament in the
BFP of the condenser lens, you can say the open space you're referring to is
infinity space. And even if it was, it would not solve the problem. To get
"exact Kohler" the whole illumination train has to move together, including
the lamp.

The standard approach works OK. So, what's your application? Why do you need
to be "exact"?

Your infinity scenario would help you fix the filament image to the BFP of
the condenser (and objective) lens, but that's not the goal. The goal is to
achieve homogeneous illumination of your sample at wide range of conditions.
You can't even treat the illumination optics as imaging optics - the high NA
(0.55), the large field of view (6 mm for the 4x lens), and size/price
constrains leads to huge chromatic, spherical and other aberrations...

Btw, the adjustable Olympus lamp housing is U-ULS100H, it looks identical to
the mercury lamp housing U-ULS100HG (note the difference!), but uses 100 W
halogen lamp instead of mercury short arc lamp.


Best, zdenek


---------- Původní e-mail ----------
Od: Kyle Michael Douglass <[hidden email]>
Komu: [hidden email]
Datum: 8. 8. 2017 13:50:04
Předmět: Re: Optics positioning for Köhler illumination and transmitted
light
"*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Thanks everyone for the replies.

I am specifically thinking of an Olympus IX71 with a IX2-LWUCD condenser; I
see no way to adjust the positions of either or the lamp or the collector
lens immediately after the lamp and before the field iris. Do the optics in
the very top of the microscope housing form a real image of the lamp
filament in the open vertical space between the field iris and the condenser
lens housing, or is the image of the filament at infinity? (I'm not in the
lab to check.)

If it's real, then the image of the lamp filament is fixed in space because
the lamp and collector lens are fixed. If it's at infinity, then Mike Model'
s answer would make some sense: there could be a field lens at the very top
of the condenser housing that forms the real image of the filament and moves
with the condenser lens, regardless of the position of the full condenser
lens housing.

Thanks again,
Kyle

Dr. Kyle M. Douglass
Post-doctoral Researcher
EPFL - The Laboratory of Experimental Biophysics
http://leb.epfl.ch/
http://kmdouglass.github.io

________________________________________
From: Confocal Microscopy List <[hidden email]> on behalf
of [hidden email] <[hidden email]>
Sent: Tuesday, August 8, 2017 6:47 PM
To: [hidden email]
Subject: Re: Optics positioning for Köhler illumination and transmitted
light

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Hi Kyle,
you're right, usually the field iris position is fixed, and only the
condensed lens (with aperture iris, DIC, phase, etc elements) can move.
There are three parts to the answer:

1) you're right again, the depth of focus of the lamp filament image is
quite large (few cm), because the filament image is magnified considerably,
maybe 10 times. But it's not the depth of focus of the condenser, it's depth
of focus of the optics between the lamp and the condenser.

2) on some systems (Olympus, Zeiss) you can adjust the filament position so
it's in focus in the BFP of the condenser.

3) usually there is a diffuser somewhere, making the "filament image in the
BFP of the condenser" idea less of a concern.

best, zdenek


--
Zdenek Svindrych, Ph.D.
W.M. Keck Center for Cellular Imaging (PLSB 003)
Department of Biology,University of Virginia
409 McCormick Rd, Charlottesville, VA-22904
http://www.kcci.virginia.edu/
tel: [phone]

---------- Původní e-mail ----------
Od: Kyle Douglass <[hidden email]>
Komu: [hidden email]
Datum: 8. 8. 2017 12:17:08
Předmět: Optics positioning for Köhler illumination and transmitted light
"*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Hi everyone,

I have a question concerning trans-illumination and commercial
microscope designs that I am hoping someone on the list could answer.

On a "standard" inverted microscope, the condenser lens is free to move
in the vertical direction. This allows one to image the field stop onto
the sample and is how one places the microscope into Köhler
illumination. Additionally, and if I am not mistaken, the primary
requirement of Köhler illumination is that an image of the light source
should be formed in the back focal plane of the condenser.

My question is: how does moving only the condenser lens ensure that the
field stop is imaged onto the sample and that the light source is imaged
onto the condenser's back focal plane at the same time? If the sample
were to change axial positions, then it would seem impossible to ensure
that both conditions are met when only the condenser is free to move.

My guess at the moment is that the depth of field of the condenser is
large enough to accommodate a range of axial sample positions, but this
is only a guess.

Thanks!
Kyle

--
Kyle M. Douglass, PhD
Post-doctoral researcher
The Laboratory of Experimental Biophysics
EPFL, Lausanne, Switzerland
http://kmdouglass.github.io
http://leb.epfl.ch
"
"

[James Pawley]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Simple answer: setting the condenser position so that the field diaphragm is in-focus at the same plane that you are focused on is sufficient because the manufacturer has set up the position and focal-length of the other optical components to make Kohler Happen. (And you are right, if you change focus by moving the specimen stage and if the condenser is connected to the stage  (rather directly than to the frame of the scope), then strictly speaking, you need to adjust the condenser to meet Kohler every time you change focus plane.

Oh, and the reason that the manufacturers can’t set things up to have Kohler always correct is that the specimen messes things up. If the slide  is a few µm too thick, the focus plane of the condenser will lo longer match that of the objective.

But actually setting up high-NA Kohler is a bit more complicated.

In the Kohler-ill diagram, the source is diagrammed as though it were a plane which can be focused into the BF-plane of the condenser. In fact most sources (arcs and filaments) are actually 3-dimensional. As the collector NA is about 0.7-0.9, it has very shallow depth of field and the entire filament or arc cannot be “in-focus” at any one setting. Consequently, the (3-d) image of this source cannot be perfectly "In- focus” in the BFP of the condenser (or the objective). i.e., no matter how you try, you really cannot set up “real” Kohler. The closest you can come is if you put a sheet of ground glass in the light path and treat this as the “planar source.” Given the layout of most scopes this is usually not possible (You need another lens to focus the actual source onto the ground glass, and this (much dimmer) source is then focused by the official collector etc.).

The other exception is the special illumination system designed about 25 years ago by Gordon Ellis at Woods Hole so that Bob Allen and Shinya Inoue could get the best possible performance in video-enhanced contrast DIC.

These were fiber-coupled arc sources where the output end of a vibrating, mm-size multimode fiber was used as a fairly-uniform, circular source, followed by beam-expander optics to make a more-or-less gaussian beam of the correct size at the BFP. These systems used optics that had the correct (de)magnification to make the image of the end of this fiber match the field of view of the objective you wanted to use (critical illumination). A zoom-optic was often used to match the output of a single finer to the field-width of different objectives. (See Inoue et al)

So, if you are using a coiled filament or arc source, you can only approximate Kohler. And in particular, the “uniform illumination of the field -of-view” that we often desire can only be met over a small field-of-view, one whose size is set by the extent which the magnification of the optics between the source and the imaged-plane in the specimen. If these optics allow you to image a tiny but fairly flat surface of the filament into the condenser BFP, then this flat surface may be a fairly good substitute for a planar source. The problem is that it is also a fairly weak source. The light from any other part of the filament is wasted. Indeed, at one time one could buy incandescent sources made of a fairly flat sheet of tungsten. The problems were that, such bulbs required quite high current to get even ‘yellow-hot’ and tiny changes in temp caused the filament to expand/contract and this made the surface of the filament wander away from the focal plane of the collector lens.

Good luck,

Jim Pawley
              ****************************************
James and Christine Pawley, 5446 Burley Place (PO Box 2348), Sechelt, BC, Canada, V0N3A0,
Phone [phone], email <[hidden email]<mailto:[hidden email]>>
NEW! NEW! AND DIFFERENT Cell (when I remember to turn it on!) 1-604-989-6146

On Aug 8, 17, at 9:35 AM, Michael Doube <[hidden email]<mailto:[hidden email]>> wrote:

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

On 08/08/17 17:10, Kyle Douglass wrote:
If the sample were to change axial positions

Then it will be out of focus with respect to the objective. Placing your specimen in focus under a low-mag objective is step 1 of Köhler illumination.

You can think of Köhler as a 5-part system: light source, condenser, sample, objective, final image plane. If any of the parts moves axially without a correcting adjustment in the others, you will lose either (or both) focus and/or even illumination.

Michael

--
Michael Doube, PhD
Lecturer, Comparative Biomedical Sciences
The Royal Veterinary College, University of London
Royal College Street
London NW1 0TU
United Kingdom

[phone] (Internal: 5503)
@mdoube

<http://www.rvc.ac.uk>

This message, together with any attachments, is intended for the stated addressee(s) only and may contain privileged or confidential information. Any views or opinions presented are solely those of the author and do not necessarily represent those of the Royal Veterinary College.

[mmodel]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Still, for all its imperfections, transmitted illumination can be very uniform, much better than in fluorescence


Mike Model


________________________________
From: Confocal Microscopy List <[hidden email]> on behalf of JAMES B PAWLEY <[hidden email]>
Sent: Tuesday, August 8, 2017 5:06 PM
To: [hidden email]
Subject: Re: Optics positioning for Köhler illumination and transmitted light

*****
To join, leave or search the confocal microscopy listserv, go to:
https://na01.safelinks.protection.outlook.com/?url=http%3A%2F%2Flists.umn.edu%2Fcgi-bin%2Fwa%3FA0%3Dconfocalmicroscopy&data=01%7C01%7Cmmodel%40KENT.EDU%7Cf5c20fe2f88f4a9c10a908d4dea173a8%7Ce5a06f4a1ec44d018f73e7dd15f26134%7C1&sdata=VCRSY2EBW6LiLiAuo7s1wC3lEsKdqifaUOFGblg2CMA%3D&reserved=0
Post images on https://na01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.imgur.com&data=01%7C01%7Cmmodel%40KENT.EDU%7Cf5c20fe2f88f4a9c10a908d4dea173a8%7Ce5a06f4a1ec44d018f73e7dd15f26134%7C1&sdata=DvN1K%2FIlPaphQRTmu0ORLYGiekIWVDKZnKa3bkrXPcc%3D&reserved=0 and include the link in your posting.
*****

Simple answer: setting the condenser position so that the field diaphragm is in-focus at the same plane that you are focused on is sufficient because the manufacturer has set up the position and focal-length of the other optical components to make Kohler Happen. (And you are right, if you change focus by moving the specimen stage and if the condenser is connected to the stage  (rather directly than to the frame of the scope), then strictly speaking, you need to adjust the condenser to meet Kohler every time you change focus plane.

Oh, and the reason that the manufacturers can’t set things up to have Kohler always correct is that the specimen messes things up. If the slide  is a few µm too thick, the focus plane of the condenser will lo longer match that of the objective.

But actually setting up high-NA Kohler is a bit more complicated.

In the Kohler-ill diagram, the source is diagrammed as though it were a plane which can be focused into the BF-plane of the condenser. In fact most sources (arcs and filaments) are actually 3-dimensional. As the collector NA is about 0.7-0.9, it has very shallow depth of field and the entire filament or arc cannot be “in-focus” at any one setting. Consequently, the (3-d) image of this source cannot be perfectly "In- focus” in the BFP of the condenser (or the objective). i.e., no matter how you try, you really cannot set up “real” Kohler. The closest you can come is if you put a sheet of ground glass in the light path and treat this as the “planar source.” Given the layout of most scopes this is usually not possible (You need another lens to focus the actual source onto the ground glass, and this (much dimmer) source is then focused by the official collector etc.).

The other exception is the special illumination system designed about 25 years ago by Gordon Ellis at Woods Hole so that Bob Allen and Shinya Inoue could get the best possible performance in video-enhanced contrast DIC.

These were fiber-coupled arc sources where the output end of a vibrating, mm-size multimode fiber was used as a fairly-uniform, circular source, followed by beam-expander optics to make a more-or-less gaussian beam of the correct size at the BFP. These systems used optics that had the correct (de)magnification to make the image of the end of this fiber match the field of view of the objective you wanted to use (critical illumination). A zoom-optic was often used to match the output of a single finer to the field-width of different objectives. (See Inoue et al)

So, if you are using a coiled filament or arc source, you can only approximate Kohler. And in particular, the “uniform illumination of the field -of-view” that we often desire can only be met over a small field-of-view, one whose size is set by the extent which the magnification of the optics between the source and the imaged-plane in the specimen. If these optics allow you to image a tiny but fairly flat surface of the filament into the condenser BFP, then this flat surface may be a fairly good substitute for a planar source. The problem is that it is also a fairly weak source. The light from any other part of the filament is wasted. Indeed, at one time one could buy incandescent sources made of a fairly flat sheet of tungsten. The problems were that, such bulbs required quite high current to get even ‘yellow-hot’ and tiny changes in temp caused the filament to expand/contract and this made the surface of the filament wander away from the focal plane of the collector lens.

Good luck,

Jim Pawley
              ****************************************
James and Christine Pawley, 5446 Burley Place (PO Box 2348), Sechelt, BC, Canada, V0N3A0,
Phone [phone], email <[hidden email]<mailto:[hidden email]>>
NEW! NEW! AND DIFFERENT Cell (when I remember to turn it on!) 1-604-989-6146

On Aug 8, 17, at 9:35 AM, Michael Doube <[hidden email]<mailto:[hidden email]>> wrote:

*****
To join, leave or search the confocal microscopy listserv, go to:
https://na01.safelinks.protection.outlook.com/?url=http%3A%2F%2Flists.umn.edu%2Fcgi-bin%2Fwa%3FA0%3Dconfocalmicroscopy&data=01%7C01%7Cmmodel%40KENT.EDU%7Cf5c20fe2f88f4a9c10a908d4dea173a8%7Ce5a06f4a1ec44d018f73e7dd15f26134%7C1&sdata=VCRSY2EBW6LiLiAuo7s1wC3lEsKdqifaUOFGblg2CMA%3D&reserved=0
Post images on https://na01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.imgur.com&data=01%7C01%7Cmmodel%40KENT.EDU%7Cf5c20fe2f88f4a9c10a908d4dea173a8%7Ce5a06f4a1ec44d018f73e7dd15f26134%7C1&sdata=DvN1K%2FIlPaphQRTmu0ORLYGiekIWVDKZnKa3bkrXPcc%3D&reserved=0 and include the link in your posting.
*****

On 08/08/17 17:10, Kyle Douglass wrote:
If the sample were to change axial positions

Then it will be out of focus with respect to the objective. Placing your specimen in focus under a low-mag objective is step 1 of Köhler illumination.

You can think of Köhler as a 5-part system: light source, condenser, sample, objective, final image plane. If any of the parts moves axially without a correcting adjustment in the others, you will lose either (or both) focus and/or even illumination.

Michael

--
Michael Doube, PhD
Lecturer, Comparative Biomedical Sciences
The Royal Veterinary College, University of London
Royal College Street
London NW1 0TU
United Kingdom

[phone] (Internal: 5503)
@mdoube

<https://na01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.rvc.ac.uk&data=01%7C01%7Cmmodel%40KENT.EDU%7Cf5c20fe2f88f4a9c10a908d4dea173a8%7Ce5a06f4a1ec44d018f73e7dd15f26134%7C1&sdata=AhD32e5JZhKtXU6kzjSWR%2BHa0wdj2dYloEhFpdqdr%2FQ%3D&reserved=0>

This message, together with any attachments, is intended for the stated addressee(s) only and may contain privileged or confidential information. Any views or opinions presented are solely those of the author and do not necessarily represent those of the Royal Veterinary College.

[James Pawley]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

(Note: This is a response to a (private) question from a list member. I am sending it out to the list without his name.)



Hi Jim, your comment on the depth of the incandescent source and how it impacts Kohler was very interesting. Given that LEDs can actually be constructed as flat planes, do you think an LED array would get you significantly closer to 'perfect' Kohler?




Good question. Not sure.

The output surface of the LE diode is probably flat, but is that the Source? I guess that the depletion region of the diode is actually the source and the shape and position of this region can vary a lot depending on how the diode is designed. As I remember, some early diodes had depletion regions oriented perpendicular to the output surface (I think that these were often used for solid-state lasers). Other LED depletion regions are quite thick and the process of electron-hole recombination that produces the light can occur anywhere throughout this region..

This is more important than one might think because the refractive index of some of the materials used to make LEDs is quite high. (Si: 3.7, GaAs: 3.5; SiN: 2.04 etc. compared to coverslip: 1.515), so a lot depends on the RI of the mounting medium surrounding the chip. Rays travelling even a few µm through such high-RI material and collected by a high-NA collector will be subject to massive spherical aberration (i.e., the planar object cannot be imaged into a planar image in the BFP.)

Depletion regions can be microns or millimetres in thickness. If the former, and if also oriented perpendicular to the optical axis, and at least a millimetre in diameter, then you would have a planar source and (given better optics. See below), could establish proper Kohler. However, as your source would be one diode, you would only have one “colour”. You could arrange 3-4 such systems, each with a different wavelength and use dichroics to combine them in such a way that all projected the image of their luminous surface collinearly into the condenser BFP as though coming from the same axial location, but that would take a bit of effort and a spectrum with 3-4 peaks doesn’t resemble an incandescent  spectrum. You might make do with a “white LED” but then the red and green would come from the phosphor layer, while the blue would come from the depletion region some µm below it (but perhaps also some scattered by the phosphor?).

It would not suffice to use a side-by-side array of diodes operating at different wavelengths.  If this array was imaged into the condenser BFP, ray bundles approaching the specimen from different angles would have different wavelengths. Not good for any type of phase contrast or DIC microscopy that relies on interference.

(Note: Because the 3D nature of most actual sources makes perfect Kohler impossible, there is little reason to create a collector lens of high optical quality (Why make a perfect image of an imperfect source?). In fact collectors have high aberrations and are designed predominantly so that they collect a lot of light (perhaps into the UV) and not bet damaged by being located so close to a very hot (Hg, Xe) arc source. As LED sources come into more common use, these conditions may change and collector lenses may have to improve. They will no longer have to withstand high heat but, they may be called upon to image sources planar into a planar image on the source. To do this they will need to corrected for spherical and chromatic aberration field curvature etc. This correction will have to take account of the RI and thickness of any material between the depletion region and the collector lens.)

Perhaps one of the LED Source vendors can enlighten us? These problems are less of a problem in fluorescence microscopy where phase is less important.

One more point I should have made clear in the last post: You cannot set up Kohler using for instance using, for instance, a curly W filament source BECAUSE although the image of parts of the filament can be focused at the BFP, other parts of the filament image will be imaged above or below this plane and hence they will begin to be visible as a blurred image of the filament in the image planes. This inhomogeneity in the image plane is more acute at high NA, particularly when using oiled condensers and objectives.

Best,

Jim Pawley
              ****************************************
James and Christine Pawley, 5446 Burley Place (PO Box 2348), Sechelt, BC, Canada, V0N3A0,
Phone [phone], email <[hidden email]<mailto:[hidden email]>>
NEW! NEW! AND DIFFERENT Cell (when I remember to turn it on!) 1-604-989-6146



Hi Jim, your comment on the depth of the incandescent source and how it impacts Kohler was very interesting. Given that LEDs can actually be constructed as flat planes, do you think an LED array would get you significantly closer to 'perfect' Kohler?











On Tue, Aug 8, 2017 at 3:06 PM, JAMES B PAWLEY <[hidden email]<mailto:[hidden email]>> wrote:
*****
So, if you are using a coiled filament or arc source, you can only approximate Kohler. And in particular, the “uniform illumination of the field -of-view” that we often desire can only be met over a small field-of-view, one whose size is set by the extent which the magnification of the optics between the source and the imaged-plane in the specimen. If these optics allow you to image a tiny but fairly flat surface of the filament into the condenser BFP, then this flat surface may be a fairly good substitute for a planar source. The problem is that it is also a fairly weak source. The light from any other part of the filament is wasted. Indeed, at one time one could buy incandescent sources made of a fairly flat sheet of tungsten. The problems were that, such bulbs required quite high current to get even ‘yellow-hot’ and tiny changes in temp caused the filament to expand/contract and this made the surface of the filament wander away from the focal plane of the collector lens.

Good luck,

Jim Pawley
              ****************************************
James and Christine Pawley, 5446 Burley Place (PO Box 2348), Sechelt, BC, Canada, V0N3A0,
Phone [phone], email <[hidden email]<mailto:[hidden email]><mailto:[hidden email]<mailto:[hidden email]>>>
NEW! NEW! AND DIFFERENT Cell (when I remember to turn it on!) 1-604-989-6146

On Aug 8, 17, at 9:35 AM, Michael Doube <[hidden email]<mailto:[hidden email]><mailto:[hidden email]<mailto:[hidden email]>>> wrote:

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com<http://www.imgur.com/> and include the link in your posting.
*****

On 08/08/17 17:10, Kyle Douglass wrote:
If the sample were to change axial positions

Then it will be out of focus with respect to the objective. Placing your specimen in focus under a low-mag objective is step 1 of Köhler illumination.

You can think of Köhler as a 5-part system: light source, condenser, sample, objective, final image plane. If any of the parts moves axially without a correcting adjustment in the others, you will lose either (or both) focus and/or even illumination.

Michael

--
Michael Doube, PhD
Lecturer, Comparative Biomedical Sciences
The Royal Veterinary College, University of London
Royal College Street
London NW1 0TU
United Kingdom

[phone] (Internal: 5503)
@mdoube

<http://www.rvc.ac.uk<http://www.rvc.ac.uk/>>

This message, together with any attachments, is intended for the stated addressee(s) only and may contain privileged or confidential information. Any views or opinions presented are solely those of the author and do not necessarily represent those of the Royal Veterinary College.

[Kyle Michael Douglass]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Thanks everyone for your responses. This helps quite a bit.

The reason I asked is for teaching purposes. I find it's difficult for
students to reconcile the simple, thin lens diagrams that they find for
example on MicroscopyU with what they find on real microscopes. I was
hoping that there might be a better way to connect simple theory to what
they work with in real life, but it appears that perhaps the simple
theory isn't so easily translated to real scopes ;)

Cheers,

Kyle


On 08/09/2017 05:45 AM, JAMES B PAWLEY wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> Post images on http://www.imgur.com and include the link in your posting.
> *****
>
> (Note: This is a response to a (private) question from a list member. I am sending it out to the list without his name.)
>
>
>
> Hi Jim, your comment on the depth of the incandescent source and how it impacts Kohler was very interesting. Given that LEDs can actually be constructed as flat planes, do you think an LED array would get you significantly closer to 'perfect' Kohler?
>
>
>
>
> Good question. Not sure.
>
> The output surface of the LE diode is probably flat, but is that the Source? I guess that the depletion region of the diode is actually the source and the shape and position of this region can vary a lot depending on how the diode is designed. As I remember, some early diodes had depletion regions oriented perpendicular to the output surface (I think that these were often used for solid-state lasers). Other LED depletion regions are quite thick and the process of electron-hole recombination that produces the light can occur anywhere throughout this region..
>
> This is more important than one might think because the refractive index of some of the materials used to make LEDs is quite high. (Si: 3.7, GaAs: 3.5; SiN: 2.04 etc. compared to coverslip: 1.515), so a lot depends on the RI of the mounting medium surrounding the chip. Rays travelling even a few µm through such high-RI material and collected by a high-NA collector will be subject to massive spherical aberration (i.e., the planar object cannot be imaged into a planar image in the BFP.)
>
> Depletion regions can be microns or millimetres in thickness. If the former, and if also oriented perpendicular to the optical axis, and at least a millimetre in diameter, then you would have a planar source and (given better optics. See below), could establish proper Kohler. However, as your source would be one diode, you would only have one “colour”. You could arrange 3-4 such systems, each with a different wavelength and use dichroics to combine them in such a way that all projected the image of their luminous surface collinearly into the condenser BFP as though coming from the same axial location, but that would take a bit of effort and a spectrum with 3-4 peaks doesn’t resemble an incandescent  spectrum. You might make do with a “white LED” but then the red and green would come from the phosphor layer, while the blue would come from the depletion region some µm below it (but perhaps also some scattered by the phosphor?).
>
> It would not suffice to use a side-by-side array of diodes operating at different wavelengths.  If this array was imaged into the condenser BFP, ray bundles approaching the specimen from different angles would have different wavelengths. Not good for any type of phase contrast or DIC microscopy that relies on interference.
>
> (Note: Because the 3D nature of most actual sources makes perfect Kohler impossible, there is little reason to create a collector lens of high optical quality (Why make a perfect image of an imperfect source?). In fact collectors have high aberrations and are designed predominantly so that they collect a lot of light (perhaps into the UV) and not bet damaged by being located so close to a very hot (Hg, Xe) arc source. As LED sources come into more common use, these conditions may change and collector lenses may have to improve. They will no longer have to withstand high heat but, they may be called upon to image sources planar into a planar image on the source. To do this they will need to corrected for spherical and chromatic aberration field curvature etc. This correction will have to take account of the RI and thickness of any material between the depletion region and the collector lens.)
>
> Perhaps one of the LED Source vendors can enlighten us? These problems are less of a problem in fluorescence microscopy where phase is less important.
>
> One more point I should have made clear in the last post: You cannot set up Kohler using for instance using, for instance, a curly W filament source BECAUSE although the image of parts of the filament can be focused at the BFP, other parts of the filament image will be imaged above or below this plane and hence they will begin to be visible as a blurred image of the filament in the image planes. This inhomogeneity in the image plane is more acute at high NA, particularly when using oiled condensers and objectives.
>
> Best,
>
> Jim Pawley
>                ****************************************
> James and Christine Pawley, 5446 Burley Place (PO Box 2348), Sechelt, BC, Canada, V0N3A0,
> Phone [phone], email <[hidden email]<mailto:[hidden email]>>
> NEW! NEW! AND DIFFERENT Cell (when I remember to turn it on!) 1-604-989-6146
>
>
>
> Hi Jim, your comment on the depth of the incandescent source and how it impacts Kohler was very interesting. Given that LEDs can actually be constructed as flat planes, do you think an LED array would get you significantly closer to 'perfect' Kohler?
>
>
>
>
>
>
>
>
>
>
>
> On Tue, Aug 8, 2017 at 3:06 PM, JAMES B PAWLEY <[hidden email]<mailto:[hidden email]>> wrote:
> *****
> So, if you are using a coiled filament or arc source, you can only approximate Kohler. And in particular, the “uniform illumination of the field -of-view” that we often desire can only be met over a small field-of-view, one whose size is set by the extent which the magnification of the optics between the source and the imaged-plane in the specimen. If these optics allow you to image a tiny but fairly flat surface of the filament into the condenser BFP, then this flat surface may be a fairly good substitute for a planar source. The problem is that it is also a fairly weak source. The light from any other part of the filament is wasted. Indeed, at one time one could buy incandescent sources made of a fairly flat sheet of tungsten. The problems were that, such bulbs required quite high current to get even ‘yellow-hot’ and tiny changes in temp caused the filament to expand/contract and this made the surface of the filament wander away from the focal plane of the collector lens.
>
> Good luck,
>
> Jim Pawley
>                ****************************************
> James and Christine Pawley, 5446 Burley Place (PO Box 2348), Sechelt, BC, Canada, V0N3A0,
> Phone [phone], email <[hidden email]<mailto:[hidden email]><mailto:[hidden email]<mailto:[hidden email]>>>
> NEW! NEW! AND DIFFERENT Cell (when I remember to turn it on!) 1-604-989-6146
>
> On Aug 8, 17, at 9:35 AM, Michael Doube <[hidden email]<mailto:[hidden email]><mailto:[hidden email]<mailto:[hidden email]>>> wrote:
>
> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> Post images on http://www.imgur.com<http://www.imgur.com/> and include the link in your posting.
> *****
>
> On 08/08/17 17:10, Kyle Douglass wrote:
> If the sample were to change axial positions
>
> Then it will be out of focus with respect to the objective. Placing your specimen in focus under a low-mag objective is step 1 of Köhler illumination.
>
> You can think of Köhler as a 5-part system: light source, condenser, sample, objective, final image plane. If any of the parts moves axially without a correcting adjustment in the others, you will lose either (or both) focus and/or even illumination.
>
> Michael
>
> --
> Michael Doube, PhD
> Lecturer, Comparative Biomedical Sciences
> The Royal Veterinary College, University of London
> Royal College Street
> London NW1 0TU
> United Kingdom
>
> [phone] (Internal: 5503)
> @mdoube
>
> <http://www.rvc.ac.uk<http://www.rvc.ac.uk/>>
>
> This message, together with any attachments, is intended for the stated addressee(s) only and may contain privileged or confidential information. Any views or opinions presented are solely those of the author and do not necessarily represent those of the Royal Veterinary College.
>
>
>

--
Kyle M. Douglass, PhD
Post-doctoral researcher
The Laboratory of Experimental Biophysics
EPFL, Lausanne, Switzerland
http://kmdouglass.github.io
http://leb.epfl.ch