LED Lighting and optical resolution questions

Here you can discuss DIY adaptations to the microscope.
Message
Author
LouiseScot
Posts: 1167
Joined: Tue Jul 07, 2020 1:51 pm
Location: Scotland

LED Lighting and optical resolution questions

#1 Post by LouiseScot » Tue Nov 10, 2020 12:50 pm

Hi all

My DIY focus rail scope currently uses a 5W LED with a mirror for illumination. I've just added a much-needed condenser but I'm wondering whether the LED source also needs a diffuser. I have a ground glass one from my old scope but that was used with a 20W halogen bulb. The LED already seems very spread out and even. If I put the glass diffuser in front of the LED it significantly cuts the light intensity. My Swift 380T has a similar glass one with its LED. Would not having a diffuser negatively affect the optical performance?
At the same time, I'm wondering if there's an easy (and cheap!) way to gauge the resolution of various objectives (10x, 20x and 40x) + camera. I want to know if my illumination and objectives are optimised. I've seen some eye-wateringly expensive microscope checkers on Edmunds but they are out of the question! Of course I can easily calculate the theoretical resolution for a given wavelength of light but I'd really like to measure the resolution obtained by the system. Perhaps there are some specimen slides I might be able to get that would help me ascertain the resolutions? Or some other means?

Thanks for any help / suggestions
Louise
A Nikon CF plan 20x; A Swift 380T; A DIY infinity corrected focus rail system with a 40x/0.65 Olympus Plan, a 10x/0.30 Amscope Plan Fluor, and a 20x/0.75 Nikon Plan Apo

BramHuntingNematodes
Posts: 1538
Joined: Tue Jan 21, 2020 1:29 am
Location: Georgia, USA

Re: LED Lighting and optical resolution questions

#2 Post by BramHuntingNematodes » Tue Nov 10, 2020 1:28 pm

I usually start with a field lens with point source lighting. I needed a diffuser for an led bulb once that had a large array of leds that cast uneven shadows, but if you don't see those than a diffuser probably won't help..

I think a test slide is often employed when trying to determine resolution. Sometimes diatoms with fine features. You may also want a way to measure your field of view for absolute rather than relative resolution.
1942 Bausch and Lomb Series T Dynoptic, Custom Illumination

LouiseScot
Posts: 1167
Joined: Tue Jul 07, 2020 1:51 pm
Location: Scotland

Re: LED Lighting and optical resolution questions

#3 Post by LouiseScot » Tue Nov 10, 2020 2:00 pm

BramHuntingNematodes wrote:
Tue Nov 10, 2020 1:28 pm
I usually start with a field lens with point source lighting. I needed a diffuser for an led bulb once that had a large array of leds that cast uneven shadows, but if you don't see those than a diffuser probably won't help..

I think a test slide is often employed when trying to determine resolution. Sometimes diatoms with fine features. You may also want a way to measure your field of view for absolute rather than relative resolution.
I'd like to be able to somehow measure the resolution I'm getting so I can can compare the measured value to the theoretical one. I intuitively feel that must be possible somehow - without great expense!
Thanks
Louise
A Nikon CF plan 20x; A Swift 380T; A DIY infinity corrected focus rail system with a 40x/0.65 Olympus Plan, a 10x/0.30 Amscope Plan Fluor, and a 20x/0.75 Nikon Plan Apo

viktor j nilsson
Posts: 760
Joined: Thu Aug 23, 2018 10:12 pm
Location: Lund, Sweden

Re: LED Lighting and optical resolution questions

#4 Post by viktor j nilsson » Tue Nov 10, 2020 2:03 pm

Well, building a Kohler capable illumination train is doable and can be achieved in many ways. It's a very fun project which you'll learn a lot from. I have done so on my optical bench (a fancy name for holding various optical bits and pieces with Play-Doh that I've stolen from my kids) using, for example, a 10x eyepiece (~27mm focal length) as collector lens.

Here's a really good introduction with examples that you could replicate quite easily:

https://sites.google.com/site/experimen ... microscope

I've bought some cheap M42 iris aperture diaphragms to play with, but you could also just cut fixed-size diaphragms from cardboard.

Regarding resolution, I'd say that you should inspect the rear focal plane of your objective to see if you are illuminating the whole back aperture.

LouiseScot
Posts: 1167
Joined: Tue Jul 07, 2020 1:51 pm
Location: Scotland

Re: LED Lighting and optical resolution questions

#5 Post by LouiseScot » Tue Nov 10, 2020 2:12 pm

viktor j nilsson wrote:
Tue Nov 10, 2020 2:03 pm
Well, building a Kohler capable illumination train is doable and can be achieved in many ways. It's a very fun project which you'll learn a lot from. I have done so on my optical bench (a fancy name for holding various optical bits and pieces with Play-Doh that I've stolen from my kids) using, for example, a 10x eyepiece (~27mm focal length) as collector lens.

Here's a really good introduction with examples that you could replicate quite easily:

https://sites.google.com/site/experimen ... microscope

I've bought some cheap M42 iris aperture diaphragms to play with, but you could also just cut fixed-size diaphragms from cardboard.

Regarding resolution, I'd say that you should inspect the rear focal plane of your objective to see if you are illuminating the whole back aperture.
Yeah I've made Kohler setups with iris diaphragms before on my old Brunel microscope / halogen bulb, but I don't think it's necessary with LED lighting. I wanted to be able to actually measure the resolution I'm getting.

Thanks

Louise
A Nikon CF plan 20x; A Swift 380T; A DIY infinity corrected focus rail system with a 40x/0.65 Olympus Plan, a 10x/0.30 Amscope Plan Fluor, and a 20x/0.75 Nikon Plan Apo

viktor j nilsson
Posts: 760
Joined: Thu Aug 23, 2018 10:12 pm
Location: Lund, Sweden

Re: LED Lighting and optical resolution questions

#6 Post by viktor j nilsson » Tue Nov 10, 2020 2:14 pm

Image

Here's an example where i use a Wild Heerbrugg 10x eyepiece as collimator lens, no field aperture, an iris aperture scavenged from a Lomo condenser as the condenser aperture, and a 44mm led collimator lens as condenser.

[Edit: this was built as a DIY epi illumination system, so my goal was to fill the back aperture of the objective.]
Last edited by viktor j nilsson on Tue Nov 10, 2020 2:18 pm, edited 1 time in total.

viktor j nilsson
Posts: 760
Joined: Thu Aug 23, 2018 10:12 pm
Location: Lund, Sweden

Re: LED Lighting and optical resolution questions

#7 Post by viktor j nilsson » Tue Nov 10, 2020 2:17 pm

LouiseScot wrote: Yeah I've made Kohler setups with iris diaphragms before on my old Brunel microscope / halogen bulb, but I don't think it's necessary with LED lighting. I wanted to be able to actually measure the resolution I'm getting.

Thanks

Louise
Why wouldn't you need Kohler? I mean, you don't need iris apertures. But if you want a bright, even illumination and maximize resolution the easiest way to achieve this would to build a Kohler system, no?

Greg Howald
Posts: 1185
Joined: Tue Oct 20, 2020 6:44 am

Re: LED Lighting and optical resolution questions

#8 Post by Greg Howald » Tue Nov 10, 2020 2:19 pm

A simple diffuser.
I went to the hardware store not long ago looking for some kind of plastic cutting board that I could put at my desk for when I am using a knife or razor blade.
In the hardware store I found a thin plastic cutting board of a couple mm thick and bought it. When I took it home and removed the plastic wrapping I found it very smooth on one side, gradient on the other and translucent.
I put it over the open light source of the scope and bingo. Instant diffusion. Cost less than ten bucks.
Greg

viktor j nilsson
Posts: 760
Joined: Thu Aug 23, 2018 10:12 pm
Location: Lund, Sweden

Re: LED Lighting and optical resolution questions

#9 Post by viktor j nilsson » Tue Nov 10, 2020 2:21 pm

viktor j nilsson wrote:
Tue Nov 10, 2020 2:17 pm
LouiseScot wrote: Yeah I've made Kohler setups with iris diaphragms before on my old Brunel microscope / halogen bulb, but I don't think it's necessary with LED lighting. I wanted to be able to actually measure the resolution I'm getting.

Thanks

Louise
Why wouldn't you need Kohler? I mean, you don't need iris apertures. But if you want a bright, even illumination and maximize resolution the easiest way to achieve this would to build a Kohler system, no?
Or are you saying that your LED is so bright that you can afford to waste a lot of light? In that case diffusion should be fine.

LouiseScot
Posts: 1167
Joined: Tue Jul 07, 2020 1:51 pm
Location: Scotland

Re: LED Lighting and optical resolution questions

#10 Post by LouiseScot » Tue Nov 10, 2020 2:53 pm

viktor j nilsson wrote:
Tue Nov 10, 2020 2:21 pm
viktor j nilsson wrote:
Tue Nov 10, 2020 2:17 pm
LouiseScot wrote: Yeah I've made Kohler setups with iris diaphragms before on my old Brunel microscope / halogen bulb, but I don't think it's necessary with LED lighting. I wanted to be able to actually measure the resolution I'm getting.

Thanks

Louise
Why wouldn't you need Kohler? I mean, you don't need iris apertures. But if you want a bright, even illumination and maximize resolution the easiest way to achieve this would to build a Kohler system, no?
Or are you saying that your LED is so bright that you can afford to waste a lot of light? In that case diffusion should be fine.
It is very bright! It's a 5W LED :) It's bright enough to let me do short exposures. It's much brighter than the one on my Swift 380T which I think is only 1W, though is adjustable.

Louise
A Nikon CF plan 20x; A Swift 380T; A DIY infinity corrected focus rail system with a 40x/0.65 Olympus Plan, a 10x/0.30 Amscope Plan Fluor, and a 20x/0.75 Nikon Plan Apo

BramHuntingNematodes
Posts: 1538
Joined: Tue Jan 21, 2020 1:29 am
Location: Georgia, USA

Re: LED Lighting and optical resolution questions

#11 Post by BramHuntingNematodes » Tue Nov 10, 2020 3:06 pm

LouiseScot wrote:
Tue Nov 10, 2020 2:00 pm
BramHuntingNematodes wrote:
Tue Nov 10, 2020 1:28 pm
I usually start with a field lens with point source lighting. I needed a diffuser for an led bulb once that had a large array of leds that cast uneven shadows, but if you don't see those than a diffuser probably won't help..

I think a test slide is often employed when trying to determine resolution. Sometimes diatoms with fine features. You may also want a way to measure your field of view for absolute rather than relative resolution.
I'd like to be able to somehow measure the resolution I'm getting so I can can compare the measured value to the theoretical one. I intuitively feel that must be possible somehow - without great expense!
Thanks
Louise
Ok well then get a micrometer slide and a eyepiece reticle to calibrate your fov and get you a diatoms with some tiny features see the smallest ones you can count between reticle markers and go from therr
1942 Bausch and Lomb Series T Dynoptic, Custom Illumination

Hobbyst46
Posts: 4277
Joined: Mon Aug 21, 2017 9:02 pm

Re: LED Lighting and optical resolution questions

#12 Post by Hobbyst46 » Tue Nov 10, 2020 3:12 pm

IMHO a diffuser is no match for Kohler illumination. It does imporve the uniformity of illumination - at a considerable cost of brightness - yet in photography, the residual non-uniformity is revealed. Also, a diffuser does not yield the best contrast, whereas Kohler does. And the above holds for LEDs as well as for other point sources at a finite distance from the collector.

5W power is a lot for LED, until one tries light-hungry illuminations, for example darkfield.

There are slides with very fine patterns for checking resolution. "Resolution test target". Not a stage micrometer by itself, where line spacings are 0.01mm at best. The price is ... well ... crazy.
Just remember that the theoretical resolution, based on the Abbe or Rayleigh equations, is accurate under "ideal" functioning of all the parts of the microscope.

LouiseScot
Posts: 1167
Joined: Tue Jul 07, 2020 1:51 pm
Location: Scotland

Re: LED Lighting and optical resolution questions

#13 Post by LouiseScot » Tue Nov 10, 2020 3:35 pm

Hobbyst46 wrote:
Tue Nov 10, 2020 3:12 pm
IMHO a diffuser is no match for Kohler illumination. It does imporve the uniformity of illumination - at a considerable cost of brightness - yet in photography, the residual non-uniformity is revealed. Also, a diffuser does not yield the best contrast, whereas Kohler does. And the above holds for LEDs as well as for other point sources at a finite distance from the collector.

5W power is a lot for LED, until one tries light-hungry illuminations, for example darkfield.

There are slides with very fine patterns for checking resolution. Not a stage micrometer by itself, where line spacings are 0.01mm at best.
GU10 LED spot lamps aren't point sources? They don't seem to be - there seems to a matrix of LED elements. Higher wattage bulbs are available e.g. 6W, 7W, 8W bulbs.

Having calculated the theoretical resolutions of my 20x and 40x objectives their mid-ranges are around 340nm and 380nm. I looked at latex microbeads but they are very expensive. Maybe I'll give up on resolution measuring for now...

Louise
A Nikon CF plan 20x; A Swift 380T; A DIY infinity corrected focus rail system with a 40x/0.65 Olympus Plan, a 10x/0.30 Amscope Plan Fluor, and a 20x/0.75 Nikon Plan Apo

BramHuntingNematodes
Posts: 1538
Joined: Tue Jan 21, 2020 1:29 am
Location: Georgia, USA

Re: LED Lighting and optical resolution questions

#14 Post by BramHuntingNematodes » Tue Nov 10, 2020 3:56 pm

Yeah bulbs and arrays can have as bad of artifacts as filaments, but that's what kohler was trying to mitigate in the first place. A diffuser might be a good idea for a bulb, but then also kohler using the diffuser surface as your light source
1942 Bausch and Lomb Series T Dynoptic, Custom Illumination

Hobbyst46
Posts: 4277
Joined: Mon Aug 21, 2017 9:02 pm

Re: LED Lighting and optical resolution questions

#15 Post by Hobbyst46 » Tue Nov 10, 2020 4:26 pm

LouiseScot wrote:
Tue Nov 10, 2020 3:35 pm
... I looked at latex microbeads but they are very expensive. Maybe I'll give up on resolution measuring for now...
Microbeads, especially those under a diameter of 1um, tend to aggregate in suspension. And quite stable aggregates they are.
Diatoms are freely available resolution calibration means. Some of them, that are apparently quite common in nature, look like long narrow needles, and the stria make them resemble rulers. Some are finer-divided than others. Cleaning those forms is relatively easy. Can be mounted dry for objectives that do not require immersion.

So, assuming that the diameter of the field of view is known (by means of a stage micrometer). One aligns the long diatom needle along the diameter of the field of view, so the length of the diatom ruler is found. Counting the stria yields at least an initial estimate of the resolution. No need to identify the diatom for this purpose.

LouiseScot
Posts: 1167
Joined: Tue Jul 07, 2020 1:51 pm
Location: Scotland

Re: LED Lighting and optical resolution questions

#16 Post by LouiseScot » Tue Nov 10, 2020 4:36 pm

Hobbyst46 wrote:
Tue Nov 10, 2020 4:26 pm
LouiseScot wrote:
Tue Nov 10, 2020 3:35 pm
... I looked at latex microbeads but they are very expensive. Maybe I'll give up on resolution measuring for now...
Microbeads, especially those under a diameter of 1um, tend to aggregate in suspension. And quite stable aggregates they are.
Diatoms are freely available resolution calibration means. Some of them, that are apparently quite common in nature, look like long narrow needles, and the stria make them resemble rulers. Some are finer-divided than others. Cleaning those forms is relatively easy. So, assuming that the field of view is known - by means of a stage micrometer, one aligns the long diatom needle along the diameter of the field of view, so the length of the diatom ruler is found. Counting the stria yields at least an initial estimate of the resolution. No need to identify the diatom for this purpose.
Yes, I have some diatoms but on a prepared slide and don't think I could easily marry up the two... I have a stage micrometer. I (hopefully) have some wild diatoms growing in a jar. Will have a look in the next day or two. I don't know how I'd line them up on the micrometer? Sounds very difficult... The micro beads seemed easier but they are too expensive. It doesn't really matter if I can't measure the resolution easily - it would just have been nice to do :)

Louise
A Nikon CF plan 20x; A Swift 380T; A DIY infinity corrected focus rail system with a 40x/0.65 Olympus Plan, a 10x/0.30 Amscope Plan Fluor, and a 20x/0.75 Nikon Plan Apo

apochronaut
Posts: 6271
Joined: Fri May 15, 2015 12:15 am

Re: LED Lighting and optical resolution questions

#17 Post by apochronaut » Tue Nov 10, 2020 4:59 pm

A couple of points. Contrast reduction caused by diffusion, seems minimal, if at all when the diffuser is placed in advance of the collimating optic.

I'm curious which objectives you are trying to verify the resolution of?

LouiseScot
Posts: 1167
Joined: Tue Jul 07, 2020 1:51 pm
Location: Scotland

Re: LED Lighting and optical resolution questions

#18 Post by LouiseScot » Tue Nov 10, 2020 5:12 pm

apochronaut wrote:
Tue Nov 10, 2020 4:59 pm
A couple of points. Contrast reduction caused by diffusion, seems minimal, if at all when the diffuser is placed in advance of the collimating optic.

I'm curious which objectives you are trying to verify the resolution of?
Oh, just my Olympus Plan N 40x and the Nikon plan Apo 20x. It's really about optimising the illumination to get the best photomicrograph resolution and seeing what can be achieved, possibly with the aid of a narrowband filter. It's not a big deal, I just thought it would be interesting to do :)

Louise
A Nikon CF plan 20x; A Swift 380T; A DIY infinity corrected focus rail system with a 40x/0.65 Olympus Plan, a 10x/0.30 Amscope Plan Fluor, and a 20x/0.75 Nikon Plan Apo

hans
Posts: 986
Joined: Thu May 28, 2020 11:10 pm
Location: Southern California

Re: LED Lighting and optical resolution questions

#19 Post by hans » Tue Nov 10, 2020 5:26 pm

LouiseScot wrote:
Tue Nov 10, 2020 4:36 pm
Yes, I have some diatoms but on a prepared slide and don't think I could easily marry up the two... I have a stage micrometer. I (hopefully) have some wild diatoms growing in a jar. Will have a look in the next day or two. I don't know how I'd line them up on the micrometer? Sounds very difficult...
You can do it in two steps: First photograph the stage micrometer. Many image editing programs (GIMP for example) have a tool to measure in units of pixels. Measure the number of pixels in a known distance on the stage micrometer and divide distance/pixels to get a scale factor. Then photographs your diatoms separately, measure feature sizes in pixels, and multiply by your previously-calculated scale factor to convert to distance.

I was recently trying to do something similar:
viewtopic.php?f=6&t=10873

LouiseScot
Posts: 1167
Joined: Tue Jul 07, 2020 1:51 pm
Location: Scotland

Re: LED Lighting and optical resolution questions

#20 Post by LouiseScot » Tue Nov 10, 2020 5:33 pm

hans wrote:
Tue Nov 10, 2020 5:26 pm
LouiseScot wrote:
Tue Nov 10, 2020 4:36 pm
Yes, I have some diatoms but on a prepared slide and don't think I could easily marry up the two... I have a stage micrometer. I (hopefully) have some wild diatoms growing in a jar. Will have a look in the next day or two. I don't know how I'd line them up on the micrometer? Sounds very difficult...
You can do it in two steps: First photograph the stage micrometer. Many image editing programs (GIMP for example) have a tool to measure in units of pixels. Measure the number of pixels in a known distance on the stage micrometer and divide distance/pixels to get a scale factor. Then photographs your diatoms separately, measure feature sizes in pixels, and multiply by your previously-calculated scale factor to convert to distance.

I was recently trying to do something similar:
viewtopic.php?f=6&t=10873
Ahh ok - why didn't I think of that! Mind you, if I start looking at individual pixels it will get very blurry. Presumably, though, I could use a reduced image and make any corrections. I'm sure it will be clearer when I do it...
Thanks
Louise
A Nikon CF plan 20x; A Swift 380T; A DIY infinity corrected focus rail system with a 40x/0.65 Olympus Plan, a 10x/0.30 Amscope Plan Fluor, and a 20x/0.75 Nikon Plan Apo

hans
Posts: 986
Joined: Thu May 28, 2020 11:10 pm
Location: Southern California

Re: LED Lighting and optical resolution questions

#21 Post by hans » Tue Nov 10, 2020 6:24 pm

LouiseScot wrote:
Tue Nov 10, 2020 5:33 pm
Mind you, if I start looking at individual pixels it will get very blurry. Presumably, though, I could use a reduced image and make any corrections.
I don't think this is bad, since with typical sensor resolutions these days there can considerable oversampling relative to the theoretical optical resolution. In my setup at 100x for example the scale factor is 33 nm/pixel. I was sometimes working with downscaled images and making a correction as you say, but purely out of convenience. I don't think it really helps with anything, but it doesn't hurt either as long as you are retaining high enough sampling frequency relative to the optical resolution.

Chris Dee
Posts: 216
Joined: Mon Aug 12, 2019 1:02 pm

Re: LED Lighting and optical resolution questions

#22 Post by Chris Dee » Tue Nov 10, 2020 7:04 pm

I experimented with many LED solutions as a replacement source for the 100w halogen Köhler illumination on my Vickers M17. Inserting a diffuser at the light source does have a small impact on contrast on my scope (white balanced). This might not be the case on other scopes. If you are considering a Köhler illumination system for your resolution tests, and don't want a diffused source, I recommend the Cree XHP35.2 LED. Of the LEDs I tested it has the most even colour distribution and light density. The LED is a 4 die construction but unlike previous generations they are very close together, the seam is difficult to see when focussing the source. HTH.

LouiseScot
Posts: 1167
Joined: Tue Jul 07, 2020 1:51 pm
Location: Scotland

Re: LED Lighting and optical resolution questions

#23 Post by LouiseScot » Tue Nov 10, 2020 7:16 pm

Chris Dee wrote:
Tue Nov 10, 2020 7:04 pm
I experimented with many LED solutions as a replacement source for the 100w halogen Köhler illumination on my Vickers M17. Inserting a diffuser at the light source does have a small impact on contrast on my scope (white balanced). This might not be the case on other scopes. If you are considering a Köhler illumination system for your resolution tests, and don't want a diffused source, I recommend the Cree XHP35.2 LED. Of the LEDs I tested it has the most even colour distribution and light density. The LED is a 4 die construction but unlike previous generations they are very close together, the seam is difficult to see when focussing the source. HTH.
I'll see how I get on with the 5W spot lamp as is. I'm not planning on setting up Kohler with it (but never say never!).
Thanks
Louise
A Nikon CF plan 20x; A Swift 380T; A DIY infinity corrected focus rail system with a 40x/0.65 Olympus Plan, a 10x/0.30 Amscope Plan Fluor, and a 20x/0.75 Nikon Plan Apo

Hobbyst46
Posts: 4277
Joined: Mon Aug 21, 2017 9:02 pm

Re: LED Lighting and optical resolution questions

#24 Post by Hobbyst46 » Tue Nov 10, 2020 7:17 pm

LouiseScot wrote:
Tue Nov 10, 2020 4:36 pm
Yes, I have some diatoms but on a prepared slide and don't think I could easily marry up the two... I have a stage micrometer. I (hopefully) have some wild diatoms growing in a jar. Will have a look in the next day or two. I don't know how I'd line them up on the micrometer? Sounds very difficult... The micro beads seemed easier but they are too expensive. It doesn't really matter if I can't measure the resolution easily - it would just have been nice to do :)
Sorry that I was misunderstood. No need to marry anything. I place the stage micrometer on the stage, position the objective whose resolution should be measured, and calculate the diameter of the field of view from the scale of the stage micrometer. So the diameter of the FOV is known from now on.
I then replace the stage micrometer with a slide that carries diatoms. If the diatom is elongated and stria are visible, I can align (I believe) it along the diameter of the FOV. And estimate its length. Then count the stria and calculate. It is an estimation. An eyepiece graticule will improve the accuracy of the process but is not essential.

LouiseScot
Posts: 1167
Joined: Tue Jul 07, 2020 1:51 pm
Location: Scotland

Re: LED Lighting and optical resolution questions

#25 Post by LouiseScot » Tue Nov 10, 2020 7:24 pm

Hobbyst46 wrote:
Tue Nov 10, 2020 7:17 pm
LouiseScot wrote:
Tue Nov 10, 2020 4:36 pm
Yes, I have some diatoms but on a prepared slide and don't think I could easily marry up the two... I have a stage micrometer. I (hopefully) have some wild diatoms growing in a jar. Will have a look in the next day or two. I don't know how I'd line them up on the micrometer? Sounds very difficult... The micro beads seemed easier but they are too expensive. It doesn't really matter if I can't measure the resolution easily - it would just have been nice to do :)
Sorry that I was misunderstood. No need to marry anything. I place the stage micrometer on the stage, position the objective whose resolution should be measured, and calculate the diameter of the field of view from the scale of the stage micrometer. So the diameter of the FOV is known from now on.
I then replace the stage micrometer with a slide that carries diatoms. If the diatom is elongated and stria are visible, I can align (I believe) it along the diameter of the FOV. And estimate its length. Then count the stria and calculate. It is an estimation. An eyepiece graticule will improve the accuracy of the process but is not essential.
Yes, I got that, thanks
A Nikon CF plan 20x; A Swift 380T; A DIY infinity corrected focus rail system with a 40x/0.65 Olympus Plan, a 10x/0.30 Amscope Plan Fluor, and a 20x/0.75 Nikon Plan Apo

LouiseScot
Posts: 1167
Joined: Tue Jul 07, 2020 1:51 pm
Location: Scotland

Re: LED Lighting and optical resolution questions

#26 Post by LouiseScot » Tue Nov 10, 2020 7:25 pm

hans wrote:
Tue Nov 10, 2020 5:26 pm
LouiseScot wrote:
Tue Nov 10, 2020 4:36 pm
Yes, I have some diatoms but on a prepared slide and don't think I could easily marry up the two... I have a stage micrometer. I (hopefully) have some wild diatoms growing in a jar. Will have a look in the next day or two. I don't know how I'd line them up on the micrometer? Sounds very difficult...
You can do it in two steps: First photograph the stage micrometer. Many image editing programs (GIMP for example) have a tool to measure in units of pixels. Measure the number of pixels in a known distance on the stage micrometer and divide distance/pixels to get a scale factor. Then photographs your diatoms separately, measure feature sizes in pixels, and multiply by your previously-calculated scale factor to convert to distance.

I was recently trying to do something similar:
viewtopic.php?f=6&t=10873
Interesting thread!
A Nikon CF plan 20x; A Swift 380T; A DIY infinity corrected focus rail system with a 40x/0.65 Olympus Plan, a 10x/0.30 Amscope Plan Fluor, and a 20x/0.75 Nikon Plan Apo

hans
Posts: 986
Joined: Thu May 28, 2020 11:10 pm
Location: Southern California

Re: LED Lighting and optical resolution questions

#27 Post by hans » Tue Nov 10, 2020 7:39 pm

Probably also worth keeping in mind, when trying to make accurate measurements, if the objective has some geometric distortion so you will get slightly different results measuring in the center vs. edge of the FOV. The distortion is pretty small in most cases, I think, I see differences of a few percent, definitely less than 10%, across my plan achromatic objectives.

LouiseScot
Posts: 1167
Joined: Tue Jul 07, 2020 1:51 pm
Location: Scotland

Re: LED Lighting and optical resolution questions

#28 Post by LouiseScot » Thu Nov 12, 2020 11:08 pm

I've had a go at imaging the pores on a Pleurosigma formosa, comparing the 40x Olympus objective on the rail and the 40x on the Swift 380T. Not much difference, really. By incorporating the image from a calibration slide (0.01mm) I was able to measure the distance across a stria/line of pores and compute the apparent distance between them. My calculations came up with a value of 910-918nm which compares with an example published image of ~880nm. I calculated the theoretical resolution of a 40x/0.65 objective to be 423nm @ wavelength of 550nm. I imagine I could probably distinguish dots that are around twice as closely spaced... Really, I'd need a something which does demonstrate closer spacing to prove it, but not sure what. Still, it's been interesting to do :)
Anyway, here are a couple of sample images from the Rail microscope/Olympus 40x (Canon 550d) and the Swift (Canon 1100d - slightly bigger pixels). They were set to 100% image size and cropped. I think there are probably some diffraction effects taking place:

Swift 40x:
Swift_Pleurosigma1_+scale.jpg
Swift_Pleurosigma1_+scale.jpg (66 KiB) Viewed 10634 times
Published SEM images of pores (my added text in black):
Pleurosigma_pores.JPG
Pleurosigma_pores.JPG (109.6 KiB) Viewed 10634 times
Rail with Olympus 40x Plan N
Pleurosigma_4+scale.jpg
Pleurosigma_4+scale.jpg (58.25 KiB) Viewed 10634 times
Louise
A Nikon CF plan 20x; A Swift 380T; A DIY infinity corrected focus rail system with a 40x/0.65 Olympus Plan, a 10x/0.30 Amscope Plan Fluor, and a 20x/0.75 Nikon Plan Apo

User avatar
75RR
Posts: 8207
Joined: Sun Oct 12, 2014 2:34 am
Location: Estepona, Spain

Re: LED Lighting and optical resolution questions

#29 Post by 75RR » Fri Nov 13, 2020 7:11 am

.
Interesting comparison

Took the liberty of placing the images side by side and reducing the rail photo to the same size as that of the Swift

I would say the Swift has the edge here
.
Attachments
10µm-.jpg
10µm-.jpg (112.4 KiB) Viewed 10612 times
Zeiss Standard WL (somewhat fashion challenged) & Wild M8
Olympus E-P2 (Micro Four Thirds Camera)

Hobbyst46
Posts: 4277
Joined: Mon Aug 21, 2017 9:02 pm

Re: LED Lighting and optical resolution questions

#30 Post by Hobbyst46 » Fri Nov 13, 2020 8:49 am

LouiseScot wrote:
Thu Nov 12, 2020 11:08 pm
...My calculations came up with a value of 910-918nm which compares with an example published image of ~880nm. I calculated the theoretical resolution of a 40x/0.65 objective to be 423nm @ wavelength of 550nm. I imagine I could probably distinguish dots that are around twice as closely spaced... Really, I'd need a something which does demonstrate closer spacing to prove it, but not sure what...
Maybe a diatom of more closely spaced dots. However, the 423nm is based on the Abbe formula, which assumes that the NA of the condenser is the same as that of the objective. The Raleigh formula would give a distance of: 550/{0.65+NA_condenser}.

Post Reply