Testing high NA darkfield condenser

Here you can discuss different microscopic techniques and illumination methods, such as Brightfield, Darkfield, Phase Contrast, DIC, Oblique illumination, etc.
Message
Author
User avatar
patta
Posts: 402
Joined: Sun May 10, 2020 6:01 am
Location: Stavanger Norway
Contact:

Testing high NA darkfield condenser

#1 Post by patta » Fri Feb 11, 2022 11:09 pm

Here follow the thread on the unsold "ultracondenser", darkfield for oil immersion.
viewtopic.php?f=16&t=14790&sid=99ac660a ... 641bf528d1

New/refurbished tools to check out how the illumination is coming out:
- slide with pinhole 0.3mm (aluminium foil with an hole made with a pin, mounted under a coverslip);
- test slide, UNI-Posca pencil ink under coverslip; it is composed of nice fine pigment grains;
- Ground screen, a BK7 glass block with one side covered with milky grease; to visualize the illumination;
- Prism apertometer, a BK7 45 deg prism, with a pinhole in the center of hypotenuse (downward in the photo) and graduated NA scale on the sides;
- Fake immersion oil bottle (refilled with frying oil, new);
.
Attrezzi_controllo_illuminazione.jpg
Attrezzi_controllo_illuminazione.jpg (129.77 KiB) Viewed 10047 times
.
Lenses used (dream team):
Lomo 60x 1.0-0.7 (actually goes to 0.2), good conditions;
CZJ 60x 1.40, clean but miscollimated;
Olympus Plan 100x 1.25 short, collimated but dirty;
Nikon PA 20x 0.75, supposedly good.

Eyepiece Olympus WHK 10x/20 goes well (only) with the Oly 100x;
The condenser to be tested, darkfield manufactured few days ago.
.
Obiettive_test_condensatore_12_02_2022.jpg
Obiettive_test_condensatore_12_02_2022.jpg (101.98 KiB) Viewed 10047 times
Last edited by patta on Sat Feb 12, 2022 8:49 am, edited 2 times in total.

User avatar
patta
Posts: 402
Joined: Sun May 10, 2020 6:01 am
Location: Stavanger Norway
Contact:

Re: Testing high NA darkfield condenser

#2 Post by patta » Fri Feb 11, 2022 11:46 pm

According to the prism apertometer, the condenser emits a narrow cone with NA around 1.4 (I'd say 1.38-1.45), over a 1.2 mm slide and like 0.2mm of condenser-slide spacing; oil at all interfaces. The illumination does not pass though water (total internal reflection). This only for the center.
No photos for this, not much interesting.

Below the photo of the illumination on the ground screen; it is the hollow cone that hits the diffuser plane, after passing through the pinhole placed where the specimen should be.
It is a relatively neat stripe, with no color fringes up or down. The bright bands are from the condenser ring mirror, that is a thin aluminum foil glued and pressed over 3D printed (relief-banded) surface.
.
Prisma_Tyndall_Ultracondensatore_5_9mm.jpg
Prisma_Tyndall_Ultracondensatore_5_9mm.jpg (74.51 KiB) Viewed 10033 times
.
As comparison, below a refractive Abbe condenser (not achromatic), with a "darkfield patch" in the center; about 0.7-0.8; color fringes are visible
.
Prisma_Tyndall_Abbe_07_patch_8mm.jpg
Prisma_Tyndall_Abbe_07_patch_8mm.jpg (60.27 KiB) Viewed 10033 times
.

Edit Saturday morning
A drawing that *should* illustrate the above photos; the "smile" in the photos is a vertical plane section of the illumination cone:
.
Illustrazione_blocco_tyndall.jpg
Illustrazione_blocco_tyndall.jpg (90.73 KiB) Viewed 9996 times
.
And managed to take a similar photo with the 45 prism apertometer.
.
Apertometro_su_condensatore_5.jpg
Apertometro_su_condensatore_5.jpg (115.15 KiB) Viewed 9996 times

User avatar
patta
Posts: 402
Joined: Sun May 10, 2020 6:01 am
Location: Stavanger Norway
Contact:

Re: Testing high NA darkfield condenser

#3 Post by patta » Sat Feb 12, 2022 11:04 am

Now some photos.

The subject is acrylic paint ink from blue "posca" pencils, that is made of pigment grains about 0.5 micron in size. At the border of a line, the grains are few and separate. The black part is black because there are no scattering particles, just transparent mountant. If it's black, it's good, meaning the condenser+objective are working in darkfield mode.
No I don't have nice diatom test slides, just made this "grains slide".

Lamp is a 6W white led; imaging is afocal:
Visual eyepiece Olympus WHK 10x 20;
Objective Canon 50mm f/1.8
Sensor Canon 600D aps-c
It captures a field of about 12mm diagonal (that's my best)

The condenser is always the same.

Objective Lomo 60x at max aperture (NA 1.0), full field:
.
Shoots_FullField_Lomo60x_10.jpg
Shoots_FullField_Lomo60x_10.jpg (95.51 KiB) Viewed 9986 times
.
Lomo 60x with the iris a bit closed (let's say NA 0.50):
.
Shoots_FullField_Lomo60x_050.jpg
Shoots_FullField_Lomo60x_050.jpg (97.42 KiB) Viewed 9986 times
.
Center crop of said Lomo at 1.0
.
Shoots_Centercrop2x_Afocal_Lomo.jpg
Shoots_Centercrop2x_Afocal_Lomo.jpg (105.4 KiB) Viewed 9986 times
Last edited by patta on Sat Feb 12, 2022 1:14 pm, edited 5 times in total.

User avatar
patta
Posts: 402
Joined: Sun May 10, 2020 6:01 am
Location: Stavanger Norway
Contact:

Re: Testing high NA darkfield condenser

#4 Post by patta » Sat Feb 12, 2022 11:15 am

Same afocal setup as above, with the almighty objective Zeiss Jena Apochromat 60x 1.40
This objective is a bit miscollimated because I've opened it for cleaning once. Not really visible in the photos, but visually it shows some issues on-axis.

Full-field, focused in the center:
.
Shoots_FullField_Jena60_center.jpg
Shoots_FullField_Jena60_center.jpg (88.67 KiB) Viewed 9986 times
.
Crop of above photo:
.
Shoots_Centercrop2x_Afocal_CZJ.jpg
Shoots_Centercrop2x_Afocal_CZJ.jpg (91.76 KiB) Viewed 9986 times
.
now this objective isn't Plan, definitely. Let's try to refocus it midway in the field:
.
Shoots_FullField_Jena60_corner.jpg
Shoots_FullField_Jena60_corner.jpg (81.32 KiB) Viewed 9986 times
.
And that's the enlarged mash-up:
.
Shoots_Centercrop2x_Afocal_CZJ_corner.jpg
Shoots_Centercrop2x_Afocal_CZJ_corner.jpg (94.24 KiB) Viewed 9986 times
.
Pretty horrible eh? Anyway, the point here is so check that the CONDENSER is illuminating and is still dark away from the center.
Let's assume that the rainbow CA above arises from a deep misunderstanding between the Deutsch objective and the 日本 eyepiece, and not from my beloved condenser.
Ultra-resolution, I'm not seeing much of that. Let's blame the sunflower oil immersion.
Last edited by patta on Sat Feb 12, 2022 8:33 pm, edited 4 times in total.

User avatar
patta
Posts: 402
Joined: Sun May 10, 2020 6:01 am
Location: Stavanger Norway
Contact:

Re: Testing high NA darkfield condenser

#5 Post by patta » Sat Feb 12, 2022 11:21 am

Here finally a proper Plan with more or less proper eyepiece, Olympus 100x 1.25 HI Plan
.
Shoots_FullField_Oly100_125.jpg
Shoots_FullField_Oly100_125.jpg (100.48 KiB) Viewed 9984 times
.
Shoots_Centercrop2x_Afocal_Oly.jpg
Shoots_Centercrop2x_Afocal_Oly.jpg (96.38 KiB) Viewed 9984 times
Last edited by patta on Sat Feb 12, 2022 11:33 am, edited 1 time in total.

User avatar
patta
Posts: 402
Joined: Sun May 10, 2020 6:01 am
Location: Stavanger Norway
Contact:

Re: Testing high NA darkfield condenser

#6 Post by patta » Sat Feb 12, 2022 11:29 am

The above images were Jpegs, resized straight from the camera - no adjustments. The background appears pitch-black, but visually ain't so, it is slightly bluish.

Here a photo with phone camera over the eyepiece, HDR. That's actually a bit exaggerate in term of exposure, the scattering grains are fully burnt in this image. Anyway, there is some stray light and haze, but not so much. I think was the Oly 100?
.
Oly100x_phone.jpg
Oly100x_phone.jpg (132.77 KiB) Viewed 9980 times
.
The Olympus 100x in particular had, visually, a lot of haze. Looking at it with "phase telescope", revealed that indeed the top lens was "a bit dirty". Believe or not, the photo at the previous post has been taken through those geological deposits.
.
Oly100x_top_lens_full_of_crap.jpg
Oly100x_top_lens_full_of_crap.jpg (69.11 KiB) Viewed 9980 times
Last edited by patta on Sat Feb 12, 2022 12:47 pm, edited 5 times in total.

Hobbyst46
Posts: 4275
Joined: Mon Aug 21, 2017 9:02 pm

Re: Testing high NA darkfield condenser

#7 Post by Hobbyst46 » Sat Feb 12, 2022 11:49 am

Impressive project, patta!!
I would try some monochromatic light with it, perhaps a bandpass filter with the same LED.

User avatar
patta
Posts: 402
Joined: Sun May 10, 2020 6:01 am
Location: Stavanger Norway
Contact:

Re: Testing high NA darkfield condenser

#8 Post by patta » Sat Feb 12, 2022 12:19 pm

Great, thanks. Now I don't have any strong monochromatic source or proper filter; oh, wait, a weak blue one; one of next days.
Last for today, let's check how much large field the condenser is covering, use the Nikon Planapo 20x 0.75; mounted on the newest, exclusive CFI stand. Black enamel is back in fashion! The field is limited by a 23mm eyepiece bore. No eyepiece, direct imaging with a camera objective as tube lens (either 90 or 125mm)
.
This photo is from the day after, with different tube lens. For imaging had to remove the Analyzer and Bertrand lens holders,  as they were vignetting the large field
This photo is from the day after, with different tube lens. For imaging had to remove the Analyzer and Bertrand lens holders, as they were vignetting the large field
Nikon_CFI_on_horseshoe.jpg (75.67 KiB) Viewed 9972 times
.
The tube lens isn't working perfectly, but more or less a 1mm field is observable. The subject are the lil' points of grit on black background.
Here the slide is oiled to the condenser; the objective is dry.
.
Nikon_20x_Tamron_90.jpg
Nikon_20x_Tamron_90.jpg (152.77 KiB) Viewed 9972 times
Last edited by patta on Sat Feb 12, 2022 1:32 pm, edited 12 times in total.

User avatar
patta
Posts: 402
Joined: Sun May 10, 2020 6:01 am
Location: Stavanger Norway
Contact:

Re: Testing high NA darkfield condenser

#9 Post by patta » Sat Feb 12, 2022 12:27 pm

I've put the 3D design of the condenser here:
https://www.thingiverse.com/thing:5241935
Uploaded also the FreeCAD model that is a bit messy, can be adapted by modifying the Master Sketch.

Materials:
3D printed parts. about 15 grams total
Aluminum foil
One half-ball lens 9mm diameter (a disk may also usable?)
One round coverslip 18mm
Canada balsam or cement to glue the coverslip

User avatar
patta
Posts: 402
Joined: Sun May 10, 2020 6:01 am
Location: Stavanger Norway
Contact:

Re: Testing high NA darkfield condenser

#10 Post by patta » Sun Feb 13, 2022 12:02 pm

wait a minute, monochromatic LED.. the old trick of the phone screen; it is composed of OLEDS, if one visualize a green image, well' that's emitting green.

Unfortunately turned out too weak for this inefficient darkfield; I've managed only with the Green color, since the screen is more powerful in green and the camera is also more sensitive there. Anyway not strong enough for convenient live-view focusing. It looks sharper indeed, but the extreme low brightness spoils the result.
Hope the Germans sort that out https://www.mikroskopie-forum.de/index. ... ic=43184.0

Tested only with the Olympus 100x Plan:
.
Shoots_FullField_Oly100_125_green.jpg
Shoots_FullField_Oly100_125_green.jpg (148.26 KiB) Viewed 9841 times
.
Attachments
Shoots_Centercrop2x_Afocal_Oly_Green.jpg
Shoots_Centercrop2x_Afocal_Oly_Green.jpg (148.26 KiB) Viewed 9840 times

Hobbyst46
Posts: 4275
Joined: Mon Aug 21, 2017 9:02 pm

Re: Testing high NA darkfield condenser

#11 Post by Hobbyst46 » Sun Feb 13, 2022 12:22 pm

It is very nice to achieve true DF with an 1.25NA objective , without an iris that lowers the NA.

I have two old 100x1.25 Zeiss achromat objectives. One of them is 160/-, the other bears no mark of tube length and coverslip (yes/no/both). Unexpectedly, the first one yields a not-too-bad DF with the oil 1.2-1.4 DF cardiodid condenser. But not nearly as nice as your DF above !
My second 100X does not yield DF at all - as expected.

Chas
Posts: 421
Joined: Tue Jan 19, 2021 3:11 pm

Re: Testing high NA darkfield condenser

#12 Post by Chas » Mon Feb 14, 2022 11:18 am

This is quite stunning (literaly) ... it seems to work well. Before you posted the photos using your mobile phone screen as a light source, I guessed that it had to be a pretty inefficient system.
Soooo the question is; why didnt anyone make a df condenser like this, before ?????
(When one reads the descriptions of old oil immersion df condensers most of them say that a funnel stop must be used with oil objectives to limit the numerical aperture).
Apochronaut mentions chromatic effects, do you think that might be a reason?

BTW was the half-spherical lens exactly half or was it a lens that fell out of the top of a condenser, that might be slightly more than half?

User avatar
patta
Posts: 402
Joined: Sun May 10, 2020 6:01 am
Location: Stavanger Norway
Contact:

Re: Testing high NA darkfield condenser

#13 Post by patta » Mon Feb 14, 2022 12:25 pm

Yes the phone screen it is very dim and inefficient in terms of energy - but quite effective in terms of setup time (search "screen green image" on the phone, and there you have it :lol: )

This high NA could be obtained also (possibly better) with a Cardioid design. I'm sure has been done in the past; the fancy "Colonel Woodward" illuminator in the previous thread, looks like it's capable of doing such high NA oblique darkfield. Today, I think it is not common because of practical issues:
- not many suitable subjects. The subject needs to be very thin, mounted in oil or balsam. No live ciliates in water mount! No live blood analysis!
- Darkfield haze: from the images above, the field is really dark; but near the particles, it gets hazy. my guess is that haze comes from the nearby particles out of focus; since the illumination is a very shallow cone (high NA) a lot of haze is produced. With smaller NA, less haze. See sketch attached below; and in post #3 above, the objective with iris closed, cuts off haze. This issue can get very bad with complicated, packed biological subjects.
- Chromatic aberration, after above testing I'd say it is not so troublesome, and if it is, one could go monochromatic, like in phase contrast. The violent rainbows of the 60x 1.40, I'd say is improper eyepiece compensation rather than condenser itself- but not sure yet.
- Modest increase in resolution: sad truth; going from NA 1.25 to 1.4 is a big hassle; but resolution improvement is just some 10% (while price increase of objective is some 1000%)

.
Attachments
DF_highNA_haze.jpg
DF_highNA_haze.jpg (47.54 KiB) Viewed 9778 times
Last edited by patta on Mon Feb 14, 2022 1:58 pm, edited 4 times in total.

User avatar
patta
Posts: 402
Joined: Sun May 10, 2020 6:01 am
Location: Stavanger Norway
Contact:

Re: Testing high NA darkfield condenser

#14 Post by patta » Mon Feb 14, 2022 12:42 pm

Ah forgot, the half-ball lens is really half-sphere, I bought a bunch for another project from Aliexpress Mowen/Bohr (they were 1.5$ each, but they doubled the price last week :x well, still 1/10th than other suppliers.) I chose it for this condenser because it's a standard piece available in stock, only spec is its radius. There are also made of Silica if one fancies UV.
Ideally it should be a bit less than half sphere, such that spherical surface is concentric with the subject.
That was why initially the condenser was made for use with thin slide (the subject isn't much off-center); Now, over a 1.2mm slide, the subject is pushed up off-center, the light isn't impinging orthogonally on the sphere surface, so there is some refraction and hence there must be some CA.

Last photos for this testing campaign: Radial polarization
Normally polarization is linear; with this condenser it is easy to get "radial" polarization, meaning the polarization is different along the hollow cone.
It seems to be used in lasers, or to improve resolution.
This polarization can be achieved for a straight beam, either by combining slices of linear polarizers, like a recombined cake; or with some swirling liquid crystals.

With this modern half-ball condenser, the radial polarizer can be done over the converging cone by rolling a film of linear polarizer:
.
Azimut_polarizzatore_disegno.jpg
Azimut_polarizzatore_disegno.jpg (67.18 KiB) Viewed 9777 times
.
Anelli_polarizzazione_cilindrica.jpg
Anelli_polarizzazione_cilindrica.jpg (104.82 KiB) Viewed 9777 times
.
Photo results: Nothing! :cry: Not even with a linear pol analyzer after the objective. Well It's darkfield, we see only the scattered light; Hopefully will be more interesting with other subjects or with COL.
.
Shoots_FullField_Afocal_Oly_radial_pol.jpg
Shoots_FullField_Afocal_Oly_radial_pol.jpg (104.92 KiB) Viewed 9777 times
.
Shoots_FullField_Afocal_Oly_azimuth_pol.jpg
Shoots_FullField_Afocal_Oly_azimuth_pol.jpg (91.07 KiB) Viewed 9777 times
.
Shoots_FullField_Afocal_Oly_aziumt_analyzer.jpg
Shoots_FullField_Afocal_Oly_aziumt_analyzer.jpg (126.39 KiB) Viewed 9777 times
Last edited by patta on Mon Feb 14, 2022 1:32 pm, edited 1 time in total.

User avatar
patta
Posts: 402
Joined: Sun May 10, 2020 6:01 am
Location: Stavanger Norway
Contact:

Re: Testing high NA darkfield condenser

#15 Post by patta » Mon Feb 14, 2022 1:07 pm

About rainbows, here's how the calibration slide looks like in DF (plated chrome over glass)
The same effect can be seen with naked eye, the 0.01mm ruler of the calibration slide looks colored from some angle, like diffraction grating.
.
Shoots_FullField_Afocal_Oly_Calibration.jpg
Shoots_FullField_Afocal_Oly_Calibration.jpg (88.6 KiB) Viewed 9772 times

Chas
Posts: 421
Joined: Tue Jan 19, 2021 3:11 pm

Re: Testing high NA darkfield condenser

#16 Post by Chas » Mon Feb 14, 2022 8:57 pm

I was looking at the description of Watson's immersion paraboloid:
Watson immersion paraboloid resized.jpg
Watson immersion paraboloid resized.jpg (53.08 KiB) Viewed 9726 times
The text made two interesting comments:
a] That if one shuts down the iris with this condenser head, then the lowest NA light is blocked first
b] This condenser prefers a brilliant source of light (in contrast to the Cassegrain where a pearl-frosted bulb is recommended)

I am sort of wondering whether your condensers behaviour might, in someways, be more similar to the paraboloid's than the cassegrain's.
?

User avatar
josmann
Posts: 340
Joined: Fri Sep 17, 2021 1:23 am
Location: San Jose, CA

Re: Testing high NA darkfield condenser

#17 Post by josmann » Mon Feb 14, 2022 10:13 pm

Would a high index ball lens be helpful here at all? You can get some 2.0 index balls and half balls from Edmund Optics. Pricey, but not insane pricey.
The highest quality live-streamed microscopy in the world.
Sundays around 8PST: https://www.youtube.com/@diettoms/streams
Occasionally (for now): https://www.twitch.tv/diettoms

Join the Discord: https://discord.gg/FgpUUnJaSE

apochronaut
Posts: 6233
Joined: Fri May 15, 2015 12:15 am

Re: Testing high NA darkfield condenser

#18 Post by apochronaut » Tue Feb 15, 2022 2:49 am

Chas wrote:
Mon Feb 14, 2022 8:57 pm
I was looking at the description of Watson's immersion paraboloid:
Watson immersion paraboloid resized.jpg
The text made two interesting comments:
a] That if one shuts down the iris with this condenser head, then the lowest NA light is blocked first
b] This condenser prefers a brilliant source of light (in contrast to the Cassegrain where a pearl-frosted bulb is recommended)

I am sort of wondering whether your condensers behaviour might, in someways, be more similar to the paraboloid's than the cassegrain's.
?
Paraboloids are like this one , single reflecting condensers, so the illumination beam gets reversed with the iris diaphragm first closing on the part of the light cone that ends up being the interior or low N.A. Double reflecting condensers such as Cassegrain or Cardioid or Spherical reverse that again.

User avatar
patta
Posts: 402
Joined: Sun May 10, 2020 6:01 am
Location: Stavanger Norway
Contact:

Re: Testing high NA darkfield condenser

#19 Post by patta » Tue Feb 15, 2022 2:22 pm

Yes the condenser here is in practice like a Paraboloid. Below the examples, the first one is the half-ball of this thread.
It would be nice also to remake an old style dry paraboloid, like the dry Wenham.

Cassegrain, I think there there are 2 different:
- Cass - "Schwarzchild" like a telescope, with two mirrors; also like the "reflective" microscope objectives. It is still in use, with the reflective objectives; don't know if it is used also for darkfield.
- Cass - "Nelson", I've seen it only on Watson's catalogue, with two pieces of glasses and a nasty double internal reflection; in the tiny characters below write NA 1.37 - 1.47, oh yeah.

I cant understand the why of the high NA Nelson-Cassegrain; a normal Paraboloid or Cardioid should be capable as well of high NA. Maybe at the time the glass available had too many air bubbles in it, so DF was impaired; while the thin top of the Cassegrain could be blown or stamped to remove bubbles? or the second reflection further reduces stray light?
.
Vintage_DF_condensers.jpg
Vintage_DF_condensers.jpg (142.73 KiB) Viewed 9638 times
.
Vintage_DF_condensers_cassegrains.jpg
Vintage_DF_condensers_cassegrains.jpg (143.49 KiB) Viewed 9630 times
.
Images from microscope-antique, Quekett, Archive.org etc.

User avatar
josmann
Posts: 340
Joined: Fri Sep 17, 2021 1:23 am
Location: San Jose, CA

Re: Testing high NA darkfield condenser

#20 Post by josmann » Tue Feb 15, 2022 7:16 pm

patta wrote:
Tue Feb 15, 2022 2:22 pm
or the second reflection further reduces stray light?
That seems like a good explanation. The way it’s drawn, it looks like that second reflection is designed to be right near the critical angle. So any light which would tend to go into a lower NA cone would not TIR?
The highest quality live-streamed microscopy in the world.
Sundays around 8PST: https://www.youtube.com/@diettoms/streams
Occasionally (for now): https://www.twitch.tv/diettoms

Join the Discord: https://discord.gg/FgpUUnJaSE

Chas
Posts: 421
Joined: Tue Jan 19, 2021 3:11 pm

Re: Testing high NA darkfield condenser

#21 Post by Chas » Tue Feb 15, 2022 8:47 pm

"a normal Paraboloid or Cardioid should be capable as well of high NA"
Here is a table of the performances of presumably modern implementations of the designs from the Olympus site:
DF condensers from olympus site.jpg
DF condensers from olympus site.jpg (41.25 KiB) Viewed 9585 times
https://www.olympus-lifescience.com/en/ ... darkfield/

Or another one, attributed to Needham, ('The microscope a practical guide'??):
DF Table orginaly from Needham.jpg
DF Table orginaly from Needham.jpg (46.14 KiB) Viewed 9583 times
Taken from:
https://www.chem.uci.edu/~dmitryf/manua ... oscopy.pdf
Which also contains some drawings of the Spot Ring Bicentric et al

Chas
Posts: 421
Joined: Tue Jan 19, 2021 3:11 pm

Re: Testing high NA darkfield condenser

#22 Post by Chas » Tue Feb 15, 2022 9:25 pm

And..as a little distraction; attached is E.M.Nelson's obituary:
Nelson obit.pdf
(229.73 KiB) Downloaded 169 times
And just in case you are wondering about 'Norbert's 19th band' here is a further distracting link:
http://www.microscopist.co.uk/about/fav ... opy-paper/
:-)

apochronaut
Posts: 6233
Joined: Fri May 15, 2015 12:15 am

Re: Testing high NA darkfield condenser

#23 Post by apochronaut » Wed Feb 16, 2022 1:41 am

Chas wrote:
Tue Feb 15, 2022 8:47 pm
"a normal Paraboloid or Cardioid should be capable as well of high NA"
Here is a table of the performances of presumably modern implementations of the designs from the Olympus site:

DF condensers from olympus site.jpg

https://www.olympus-lifescience.com/en/ ... darkfield/

Or another one, attributed to Needham, ('The microscope a practical guide'??):

DF Table orginaly from Needham.jpg

Taken from:
https://www.chem.uci.edu/~dmitryf/manua ... oscopy.pdf
Which also contains some drawings of the Spot Ring Bicentric et al
Those sites, such as the Olympus Lifescience .com appear to be prepared by several authors under contract. The same information appears on a Nikon page as well as others, slightly specific to the host corporation's product line but highly generalized to explain various principles of microscopy. The chart of DF types is also highly generalized. Several of the types, Paraboloid and Cardioid for instance, have many permutations from many manufacturers and the specifications are different, depending on which condenser is being considered. A Cassegrain type design on the other hand does not seem to have been embraced by too many companies, possibly only a couple of designs exist, with each originating with the same inventor.
The array on that site has not much to do with Olympus. It is a historical compendium.

apochronaut
Posts: 6233
Joined: Fri May 15, 2015 12:15 am

Re: Testing high NA darkfield condenser

#24 Post by apochronaut » Wed Feb 16, 2022 3:58 pm

patta wrote:
Tue Feb 15, 2022 2:22 pm
Yes the condenser here is in practice like a Paraboloid. Below the examples, the first one is the half-ball of this thread.
It would be nice also to remake an old style dry paraboloid, like the dry Wenham.

Cassegrain, I think there there are 2 different:
- Cass - "Schwarzchild" like a telescope, with two mirrors; also like the "reflective" microscope objectives. It is still in use, with the reflective objectives; don't know if it is used also for darkfield.
- Cass - "Nelson", I've seen it only on Watson's catalogue, with two pieces of glasses and a nasty double internal reflection; in the tiny characters below write NA 1.37 - 1.47, oh yeah.

I cant understand the why of the high NA Nelson-Cassegrain; a normal Paraboloid or Cardioid should be capable as well of high NA. Maybe at the time the glass available had too many air bubbles in it, so DF was impaired; while the thin top of the Cassegrain could be blown or stamped to remove bubbles? or the second reflection further reduces stray light?
.
Vintage_DF_condensers.jpg
.
Vintage_DF_condensers_cassegrains.jpg
.
Images from microscope-antique, Quekett, Archive.org etc.
Optical glass was worked and worked to remove air bubbles and then shattered and inspected for imperfections. I can't see that glass with air bubbles would have been used to make anything in microscope optics, after a certain point in time : mid.19th century or before.
One thing worth considering Patta, is the effect that sunflower oil is having on the N.A. of your entire optical system. With a maximum refractive index of 1.473 but likely less , your objective for starters is no longer a 1.4 N.A., plus the glass to glass homogeneity between the upper condenser element and slide is being broken, reducing the angle and N.A. You are working with lower angles than if homogeneity was intact. Maybe you should calculate this out and see what the angles actually are? Maybe closer to 1.25 for the objective?
I have a Reichert glycerin immersion 63X objective, which I occasionally used to use for DF along with an oil 100X. Sometimes, I have absent mindedly dipped the oil objective into the glycerin before I cleaned off the slide with a resultant dramatic, reduction in resolution. Glycerin has pretty close to the same N.A. as sunflower oil.

Chas
Posts: 421
Joined: Tue Jan 19, 2021 3:11 pm

Re: Testing high NA darkfield condenser

#25 Post by Chas » Thu Feb 17, 2022 8:33 pm

I have only just noticed this... but Watson's immersion Paraboloid was sold with a "supply of suitable slips"
And the thickness of the slides "...must be within 20% of thickness engraved on each of each of the paraboloid mountings"

And they werent very generous with the slides ! :
https://collection.sciencemuseumgroup.o ... condensers
(that one says 1.10 mm on the side)

This one says 1.6 mm on the side:
https://www.microscope-antiques.com/HIP.html

apochronaut
Posts: 6233
Joined: Fri May 15, 2015 12:15 am

Re: Testing high NA darkfield condenser

#26 Post by apochronaut » Thu Feb 17, 2022 9:09 pm

The principle of DF requires that the rays entering the slide do not pass through the upper surface, they reflect off of it and undergo TIR in the slide. Essentially the slide becomes like a fiber optic. If the slide thickness is too thick or thin, then one gets a partial DF with an uneven background and if the thickness is too far off the field brightens. Most DF condensers are quite specitic regarding slide thickness , with some stamping the specifications right on the condenser.

User avatar
patta
Posts: 402
Joined: Sun May 10, 2020 6:01 am
Location: Stavanger Norway
Contact:

Re: Testing high NA darkfield condenser

#27 Post by patta » Thu Feb 17, 2022 10:31 pm

On both tables above there is the word "aplanatic"; and also in the Nelson's obituary! So should be important for high NA, even if the illumination is just a thin ring. At NA 1.4 and 1.5 the angles are so shallow, aberrations likely get gigantic. So the condenser needs to be well made and relatively sharp, like a high NA objective - with its shallow focus depth, hence tight slide thickness. Two mirror surfaces are needed to get aplanatic, one isn't enough.

Below an illustration of how I think the current condenser get around this issue, with cheats:
-The mirror ain't conic nor parabolic, but is rough and wobbly, so it actually diffuses the light in many directions, instead of focusing it;
- The baffle stops that determine the aperture are after the mirror, so final NA isn't affected by aberrations (assuming that the glass half-ball has no significant effects)
So in practice this condenser is working as a "diffuse ring light + thin annular slits"; very inefficient in terms of light use, but no worries about aberrations. For illumination it needed 6 Watts LED at full power, which in brightfield would be "scorching sun".

Yes I'll get asap a gallon of proper immersion oil and test again (more rational choice than getting a refractometer for vegetable oil); extra perk, the runny sunflower oil was giving much less bubbles than the thick "B" imm oil.
Working NA, the "apertometer" used (photos in post #1 &2 of this thread; it is a prism from binocular) it's measuring the angle at which light cone comes out, after slide and oil layers; at the time I've had measured its RI, and checked it against objectives, it was working ok; so the aperture shouldn't be too much off; however that's also true that the NA of the condenser came out always higher than designed...?

I've stopped working on this condenser, will just use it and test a bit more. Well, ideas about modifications always welcome.
Thanks everybody for the historical contributions, now I know much more about condensers than two weeks ago. Hopefully some manufacturer will start again making the "Nelson Cassegrain"!
.
Condenser "raytracing"
Condenser "raytracing"
Condensatore_talmente_baracco_che_funziona.jpg (96.66 KiB) Viewed 9416 times
.
Prism apertometer profile
Prism apertometer profile
Prism_apertometer_drawing.jpg (69.22 KiB) Viewed 9414 times

Chas
Posts: 421
Joined: Tue Jan 19, 2021 3:11 pm

Re: Testing high NA darkfield condenser

#28 Post by Chas » Sat Feb 19, 2022 1:27 pm

This was the Mk.I soft drinks can Cobb slide holder:
drink can cobb holder.jpg
drink can cobb holder.jpg (65.51 KiB) Viewed 9318 times
-It has a 3/4 inch punch hole in the middle and and 'half a hole' puched each side of it.
The carboard ends do help some mechical stages to 'get a grip'.
Mistakes (apart from trying to flatten the holder whilst the coverslips were inside) :
1] The slightly jagged edges can break the corners off the slips, so a more continuous scissor cut after getting the size about right, would be an improvement.
2] The can material was just a bit to springy, I think I should have attempted to anneal it first.

User avatar
patta
Posts: 402
Joined: Sun May 10, 2020 6:01 am
Location: Stavanger Norway
Contact:

Re: Testing high NA darkfield condenser

#29 Post by patta » Thu Mar 03, 2022 1:43 pm

Apart coverslip holders annealing (and maybe a compressor..)
I'm working a bit more on the condenser, will make the base to fit other stands.

Could somebody help me with the dimensions? I have only the ancient Swift and the Leitz dovetail, and can't find reliable data on the sizes of other makers.

Need Fit Diameter and the approximate height of condensers, as drawing below;
The more the better; at least the "standard 37mm"; maybe also other more sophisticate fittings like dovetails
.
Condenser_fit_sizes.jpg
Condenser_fit_sizes.jpg (130.61 KiB) Viewed 9094 times

Hobbyst46
Posts: 4275
Joined: Mon Aug 21, 2017 9:02 pm

Re: Testing high NA darkfield condenser

#30 Post by Hobbyst46 » Thu Mar 03, 2022 4:50 pm

Does "Standard" refer to Zeiss? if so, I can provide some dimensions later this evening.

Post Reply