Dangerous-Just Thinking Again!

[GerryR]

I was wondering if a multi-colored light source might be useful in a top-lighting situation. I know there are filters of different colors for BF, but have not seen similar sources for top lighting. I would think a multi-colored source would be useful in highlighting different characteristics of a specimen. I'm just curious if such a thing exists?

[MichaelG.]

Have a look at the link I posted here, Gerry:

viewtopic.php?f=12&t=19355

Post #8

MichaelG.

[GerryR]

Have a look at the link I posted here, Gerry:

viewtopic.php?f=12&t=19355

Post #8

MichaelG.

As "they" say, "There is nothing new under the sun!"

The system I was thinking about (dangerous!) was using LEDs (4-R ,G, B, Y) pulsed simultaneously, with brightness control for each led. That would allow mixing colors to get the different combinations and effects. The system the OP came up with in the link you posted, only allows two LEDs at one time to be used, so I would imagine a limited number of colors, if the posted schematic is what he ended up with.

[MichaelG.]

.

Sorry, Gerry … I may not have been sufficiently explicit

I was referring you to my post #8 in that discussion
… which recommends looking at the Adafruit NeoPixels

[from the linked page.] …. Red, green and blue LEDs are integrated alongside a driver chip into a tiny surface-mount package controlled through a single wire.

MichaelG.

[GerryR]

.

Sorry, Gerry … I may not have been sufficiently explicit

I was referring you to my post #8 in that discussion
… which recommends looking at the Adafruit NeoPixels

[from the linked page.] …. Red, green and blue LEDs are integrated alongside a driver chip into a tiny surface-mount package controlled through a single wire.

MichaelG.

Michael,
I did see that and visited the site. Very interesting in there possible applications to microscopy! Thank you for the link.

[blekenbleu]

I hacked two RGB light sources for incident illumination, mainly to mitigate chromatic aberrations

https://blekenbleu.github.io/microscope/pinspot/


https://blekenbleu.github.io/microscope/#RGB

[GerryR]

Clean and clever!!

[jorymil]

Remember that you can't actually get yellow light by mixing red and green, nor purple by mixing red and blue. The resulting light appears yellow or purple to our eyes, but the actual wavelengths of the incident light do not change. Colors mix; light does not. You can verify this for yourself with an inexpensive prism or diffraction grating.
For example, if a substance fluoresces with violet/near-UV light at 415 nm, you need either a filtered full-spectrum source such as a halogen lamp or mercury discharge tube containing that wavelength, or you need an LED whose spectrum includes that wavelength. A lot of the "full-spectrum" LEDs you see these days are actually ultraviolet LEDs with a wide-spectrum fluorescent coating, though some just stop at blue.
Another example: you will get slightly better resolving power with true violet light (420 nm or so) than with a combination of red and blue that appears violet to the eye. Likewise, if you put a "violet bandpass" filter in front of a red/blue source, light will not be transmitted. https://www.microscopyu.com/techniques/ ... ilter-sets talks about this in some more detail.

[jorymil]

FWIW, blekenbleu: I've been reading your website a bunch lately, as I'm pretty interested in reflected light stuff, and I have an Epistar as well. It's been super-informative, so thank you!

[GerryR]

Here is my version of the lighting source. Four leds, red, green, blue and yellow placed 90 degrees apart at an angle such that each led points to the same spot approximately four inches ahead of the holder, using four acrylic rods to better aim the leds. Just a few pictures to show a couple of "mixing" combos.
How this will affect oblique lighting of a specimen I have no idea yet, but will find out if it was a worthwhile project or not; it was fun in any case.

[blekenbleu]

The system I was thinking about (dangerous!) was using LEDs (4-R ,G, B, Y) pulsed simultaneously, with brightness control for each led. That would allow mixing colors to get the different combinations and effects.

If you make separate images from each primary, you will discover that peak foci differ, e.g. red and green here by 9 microns:

https://blekenbleu.github.io/microscope/#CAL
Using e.g. Hugin to converge those primary images can yield better resolution with minimal chromatic aberrations.

[GerryR]

@blekenblue
When I first read your post I didn't understand your implications. If I was taking photos, I understand that stitching the primary color photos together would produce a higher resolution image than an image from "mixed" light. But I was more interested in what affects "mixed" light had on physical observation through the microscope, not with photographs. Thanks for your input, though, as it gave me another way of "looking" at things!

[GerryR]

Here's the setup I came up with. I'll have to get some pond water, or the like, to see what the affects are.

[jorymil]

That's a really elegant piece of engineering. If you don't mind my asking, do you have an EE background?

[GerryR]

That's a really elegant piece of engineering. If you don't mind my asking, do you have an EE background?

Yes, I'm a retired automation engineer with a BSEE. By the way, thank you for your input in the above post!