LED lighting....why the hate?

Here you can discuss different microscopic techniques and illumination methods, such as Brightfield, Darkfield, Phase Contrast, DIC, Oblique illumination, etc.
Message
Author
LouiseScot
Posts: 1167
Joined: Tue Jul 07, 2020 1:51 pm
Location: Scotland

Re: LED lighting....why the hate?

#31 Post by LouiseScot » Tue May 24, 2022 7:56 pm

MicroBob wrote:
Tue May 24, 2022 7:46 pm
In my personal view a good Cree LED gives a very nice illumination for the average variety of uses. Louise posted an LED spectrum that is quite typical. White LEDs are actually blue emitters with a fluorescence layer. Apparently they can't reach "white" without the dip at 500nm at the moment with this technique. This dip at blue-green leads to a limited color vision at the opposite end of the spectrum, ca. orange. So the poor pathologist who stares at haematoxilin-eosin stained sections all day will not be happy with the common LEDs.
The CRI values don't really show this limitation so it is always worth to look at the spectrum in the datasheet.

I really like my self built LEDs conversions, they consume little energy, they last, never wore one out so far, they can be adjusted over a wide range of power (which you shouldn't do with a halogen lamp), they are easy to build for me and I have no spares problem with exotic and expensive bulbs. But of cause a good working halogen setup is good too...
Yeah, the dip in the blue-green part of the spectrum does seem to be a limitation in all such LEDs. I suppose there has to be a gap between the stimulation and the phosphor emission wavelengths. Have to live with it!
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

MicroBob
Posts: 3154
Joined: Sun Dec 25, 2016 9:11 am
Location: Northern Germany

Re: LED lighting....why the hate?

#32 Post by MicroBob » Wed May 25, 2022 9:54 pm

I think fluorescence in the blue-green range would be easy to achieve. But it probably would be too intensive for an overall white light. There are separate blue-green leds that could fill this gap in a controllable way and the white balance could be corrected afterwards. Im no pathologist so on my personal priority list such a system for microscope illumination is not near the top. But for room lighting I actually think about it.
I know a guy who made LED ring lights for TV cameras. They were assembled from LEDs of several colours for a good spectrum. Combined with the soft illumination effect of the ring light this helped to make female persons look younger than they actually were. The wrinkels were illuminated so well that they became invisible. According to the guy a most successful product! :lol:

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

Re: LED lighting....why the hate?

#33 Post by LouiseScot » Thu May 26, 2022 8:47 am

MicroBob wrote:
Wed May 25, 2022 9:54 pm
I think fluorescence in the blue-green range would be easy to achieve. But it probably would be too intensive for an overall white light. There are separate blue-green leds that could fill this gap in a controllable way and the white balance could be corrected afterwards. Im no pathologist so on my personal priority list such a system for microscope illumination is not near the top. But for room lighting I actually think about it.
I know a guy who made LED ring lights for TV cameras. They were assembled from LEDs of several colours for a good spectrum. Combined with the soft illumination effect of the ring light this helped to make female persons look younger than they actually were. The wrinkels were illuminated so well that they became invisible. According to the guy a most successful product! :lol:
You can certainly get tuneable LEDs for room lighting these days! Not something I'd bother with myself, but still.

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

jjtr1
Posts: 33
Joined: Sat May 07, 2022 8:51 am

Re: LED lighting....why the hate?

#34 Post by jjtr1 » Thu May 26, 2022 4:16 pm

apochronaut wrote:
Tue May 24, 2022 5:41 pm
An unfortunate aspect of cheap led illumination systems is the loss of information due to the limited number of spectral lines that they produce. The eye has evolved to utilize the full spectrum of daylight in order to be provided with the maximum amount of information available and while no artificial illumination system can duplicate the spectrum of daylight, filament lamp systems provide a more continuous spectrum rather than one made up of an assemblage of discreet spectral lines, such as simple leds provide. The lack of sufficient spectral data will reduce the overall potential for the optical system to transmit maximum detail to the eye or sensor, the latter modelled on the sensitivity of the eye.
LouiseScot wrote:
Tue May 24, 2022 7:56 pm
Yeah, the dip in the blue-green part of the spectrum does seem to be a limitation in all such LEDs. I suppose there has to be a gap between the stimulation and the phosphor emission wavelengths. Have to live with it!
Louise
In the year 2021, even consumer LED bulbs (at least those sold in the EU) have made big strides in spectrum quality. Here is the spectrum of a cheap warm white bulb (socket E14, 6 W/470 lm) sold here for only about $ 4, and marketed as Ra>97:
Image
(sorry for the giant image)

Notice the very minimal dip in blue-green - just 25% compared to the blue emission peak. Also, notice the red tail that goes far into infrared. Spectrum-wise, it's a great imitation of a halogen bulb, without the drawbacks that Apochronaut mentioned.

The EU datasheet this spectrum comes from (publishing the spectrum is obligatory for bulbs sold in EU now): https://eprel.ec.europa.eu/screen/produ ... ces/527840
Its product page (no English, sorry): https://www.emos.cz/led-zarovka-classic ... -bila-ra97

Time for me to extract the chip(s) from the bulb for conversion into a microscope illuminator :D

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

Re: LED lighting....why the hate?

#35 Post by LouiseScot » Thu May 26, 2022 4:34 pm

jjtr1 wrote:
Thu May 26, 2022 4:16 pm
apochronaut wrote:
Tue May 24, 2022 5:41 pm
An unfortunate aspect of cheap led illumination systems is the loss of information due to the limited number of spectral lines that they produce. The eye has evolved to utilize the full spectrum of daylight in order to be provided with the maximum amount of information available and while no artificial illumination system can duplicate the spectrum of daylight, filament lamp systems provide a more continuous spectrum rather than one made up of an assemblage of discreet spectral lines, such as simple leds provide. The lack of sufficient spectral data will reduce the overall potential for the optical system to transmit maximum detail to the eye or sensor, the latter modelled on the sensitivity of the eye.
LouiseScot wrote:
Tue May 24, 2022 7:56 pm
Yeah, the dip in the blue-green part of the spectrum does seem to be a limitation in all such LEDs. I suppose there has to be a gap between the stimulation and the phosphor emission wavelengths. Have to live with it!
Louise
In the year 2021, even consumer LED bulbs (at least those sold in the EU) have made big strides in spectrum quality. Here is the spectrum of a cheap warm white bulb (socket E14, 6 W/470 lm) sold here for only about $ 4, and marketed as Ra>97:
Image
(sorry for the giant image)

Notice the very minimal dip in blue-green - just 25% compared to the blue emission peak. Also, notice the red tail that goes far into infrared. Spectrum-wise, it's a great imitation of a halogen bulb, without the drawbacks that Apochronaut mentioned.

The EU datasheet this spectrum comes from (publishing the spectrum is obligatory for bulbs sold in EU now): https://eprel.ec.europa.eu/screen/produ ... ces/527840
Its product page (no English, sorry): https://www.emos.cz/led-zarovka-classic ... -bila-ra97

Time for me to extract the chip(s) from the bulb for conversion into a microscope illuminator :D
Yeah, but that's room lighting. I don't think those bulbs will fit inside most microscopes however hard you try! There may be expensive, profession grade LEDs that will be suitable but I don't know.

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....why the hate?

#36 Post by viktor j nilsson » Thu May 26, 2022 4:37 pm

Nice spectrum, but certainly that spectrum is not produced by a single die, or even a single COB led. It is probably produced by a combination of many different LED dies.

For microscopy, we need a single die, or a fairly small COB led. The Yuji and Optosolis COB leds are the closest thing that I've found.

jjtr1
Posts: 33
Joined: Sat May 07, 2022 8:51 am

Re: LED lighting....why the hate?

#37 Post by jjtr1 » Thu May 26, 2022 7:00 pm

viktor j nilsson wrote:
Thu May 26, 2022 4:37 pm
Nice spectrum, but certainly that spectrum is not produced by a single die, or even a single COB led. It is probably produced by a combination of many different LED dies.

For microscopy, we need a single die, or a fairly small COB led. The Yuji and Optosolis COB leds are the closest thing that I've found.
I've just drilled a hole in the bulb's frosted cover to peer inside. It's made of 15 identically looking amber colored LED chips, each about 2x3 mm. When on, each chip appears to be glowing uniformly within its surface and all chips emit the same warm white light (I looked through a CD-R :D ).

Picture:
Image

So the only problem with unsoldering one chip from this bulb to build a microscope illuminator would be the possibly insufficient light flux per chip. Here the flux is only 30 lm (0.4 W) per chip (a 20 W halogen puts out about 200 lm, though only 100 lm per side, and has a filament about 1.5x4 mm). But room lighting LEDs are usually run at a fraction of their max power, because that's where the maximum efficiency (lm/W) is. So I think these chips could be made to output much more then they do now, assuming a proper heatsink.

By the way, the relatively low efficiency of 80 lm/W when compared to today's most efficient warm white bulbs at 150 lm/W is very probably due to the extensive red/infrared spectrum tail. The tail provides for superb rendering of deep red hues, but it's questionable whether a microscopist would make use of that.

MicroBob
Posts: 3154
Joined: Sun Dec 25, 2016 9:11 am
Location: Northern Germany

Re: LED lighting....why the hate?

#38 Post by MicroBob » Thu May 26, 2022 7:40 pm

That is a nice spektrum, jjtr1!
If this is from a single die it would be very interesting to investigate further. Many microscopes don't have a true Köhler setup but a frosted lens in the light path that acts as the light source. In this case multiple LED dies could be used. Colour enlargers used this technique to allow the mixing of the light. One drawback of LEDs with nice spectrum could be a reduced efficiency.

MicroBob
Posts: 3154
Joined: Sun Dec 25, 2016 9:11 am
Location: Northern Germany

Re: LED lighting....why the hate?

#39 Post by MicroBob » Thu May 26, 2022 7:42 pm

OK, this LED bulb is efficiency class "G" only : https://www.emos.cz/led-zarovka-classic ... -bila-ra97

jjtr1
Posts: 33
Joined: Sat May 07, 2022 8:51 am

Re: LED lighting....why the hate?

#40 Post by jjtr1 » Thu May 26, 2022 8:16 pm

MicroBob wrote:
Thu May 26, 2022 7:40 pm
That is a nice spektrum, jjtr1!
If this is from a single die it would be very interesting to investigate further. Many microscopes don't have a true Köhler setup but a frosted lens in the light path that acts as the light source. In this case multiple LED dies could be used. Colour enlargers used this technique to allow the mixing of the light. One drawback of LEDs with nice spectrum could be a reduced efficiency.
I've looked at all 15 dies when the bulb was on (through the hole I've drilled) and they seem to be the same color. My point was to show that such a cheap LED bulb sold in stores all over my country can have such a good spectrum. Just two years before, one would have to reach to specialty suppliers like Yuji for an LED with spectrum this good.
MicroBob wrote:
Thu May 26, 2022 7:42 pm
OK, this LED bulb is efficiency class "G" only : https://www.emos.cz/led-zarovka-classic ... -bila-ra97
Please bear in mind this is the new (September 2021) system of efficiency classes. Class G of 2021 corresponds to class A+ of pre-2021. My bulb's efficiency is about 80 lm/W, while most LED bulbs from the same company (EMOS) are in the E class (about >110 lm/W). The most efficient warm white bulb that is sold widely and the only such inhabitant of the C class is to my knowledge the IKEA Solhetta at 160 lm/W (as of February 2022).

Scarodactyl
Posts: 2775
Joined: Sat Mar 03, 2018 9:09 pm

Re: LED lighting....why the hate?

#41 Post by Scarodactyl » Thu May 26, 2022 8:32 pm

LEDs really have come a long way. I love my cree xml2 neutral white in my gem microscope--nice color rendering, plenty of brightness and the parts were cheap. The 30w halogen it replaced* tends to heat the base excessively after a while too, the LED gets mildly warm if it's left on for a long time.

*I got the scope cheap because the electronics had died so keeping the existing halogen wasn't even an option.

I'm working on a fluorescence system but unfortunately the needed wavelength is around 555 (DiI) and LEDs I can find are still behind mercury in brightness by a ways in this range. If it works I'll write it up.

MichaelG.
Posts: 3976
Joined: Mon Apr 10, 2017 8:24 am
Location: North Wales

Re: LED lighting....why the hate?

#42 Post by MichaelG. » Thu May 26, 2022 9:13 pm

jjtr1 wrote:
Thu May 26, 2022 4:16 pm
Its product page (no English, sorry): https://www.emos.cz/led-zarovka-classic ... -bila-ra97
Thanks to DeepL … here is a reasonable translation:
.
Colour Rendering Index (CRI)
It's not just a large fireplace with frames of family photos on the mantelpiece that can make your home warm. The right light is essential.

In terms of luminosity and financial savings, LED bulbs are currently unrivalled. When choosing the right one, the most important thing to look for is the shape, the base and the light temperature. But if you're after the best, it's also worth looking at the so-called colour rendering index (CRI).
CRI (Color Rendering Index), often referred to as Ra (Rendering average), indicates how faithfully a light source can display colors compared to natural daylight. Bulbs with a CRI of around 80, which is the minimum value prescribed by the European Union, are commonly sold. But the best models offer a CRI of more than 95, with 100 (short for 100%) being the absolute maximum. See how much difference this tiny parameter makes to the perception of the surrounding colours in the picture. Although this bulb is almost indistinguishable in appearance from a conventional incandescent bulb, it consumes several times less electricity.
.
MichaelG.
Too many 'projects'

MichaelG.
Posts: 3976
Joined: Mon Apr 10, 2017 8:24 am
Location: North Wales

Re: LED lighting....why the hate?

#43 Post by MichaelG. » Thu May 26, 2022 9:26 pm

viktor j nilsson wrote:
Thu May 26, 2022 4:37 pm
Nice spectrum
Very nice spectrum, for an LED

But it’s still worth comparing it with a Black Body spectrum

Have a look at this paper:
https://www.researchgate.net/publicatio ... aw_at_home
Too many 'projects'

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

Re: LED lighting....why the hate?

#44 Post by viktor j nilsson » Fri May 27, 2022 4:18 am

Well, I was wrong! That's a very impressive spectrum for a single emitter. Would be interesting to know what model it is. I agree that the light flux per chip could be a problem, though. But very cool to see such a high quality spectrum on a cheap consumer led product.

Is your plan to desolder the chips and re-assembly them closer together?

MichaelG.
Posts: 3976
Joined: Mon Apr 10, 2017 8:24 am
Location: North Wales

Re: LED lighting....why the hate?

#45 Post by MichaelG. » Fri May 27, 2022 4:43 am

jjtr1 wrote:
Thu May 26, 2022 7:00 pm
I've just drilled a hole in the bulb's frosted cover to peer inside. It's made of 15 identically looking amber colored LED chips, each about 2x3 mm. When on, each chip appears to be glowing uniformly within its surface and all chips emit the same warm white light (I looked through a CD-R :D ).
For what it’s worth : 3020 is one of the standard sizes for Surface-Mount LEDs

https://en.wikipedia.org/wiki/SMD_LED

… if we can identify the chip manufacturer, the should be available [somewhere] on a reel

The search is on !!

MichaelG.
Last edited by MichaelG. on Fri May 27, 2022 5:10 am, edited 1 time in total.
Too many 'projects'

MicroBob
Posts: 3154
Joined: Sun Dec 25, 2016 9:11 am
Location: Northern Germany

Re: LED lighting....why the hate?

#46 Post by MicroBob » Fri May 27, 2022 5:09 am

jjtr1 wrote:
Thu May 26, 2022 8:16 pm
MicroBob wrote:
Thu May 26, 2022 7:40 pm
That is a nice spektrum, jjtr1!
If this is from a single die it would be very interesting to investigate further. Many microscopes don't have a true Köhler setup but a frosted lens in the light path that acts as the light source. In this case multiple LED dies could be used. Colour enlargers used this technique to allow the mixing of the light. One drawback of LEDs with nice spectrum could be a reduced efficiency.
I've looked at all 15 dies when the bulb was on (through the hole I've drilled) and they seem to be the same color. My point was to show that such a cheap LED bulb sold in stores all over my country can have such a good spectrum. Just two years before, one would have to reach to specialty suppliers like Yuji for an LED with spectrum this good.
MicroBob wrote:
Thu May 26, 2022 7:42 pm
OK, this LED bulb is efficiency class "G" only : https://www.emos.cz/led-zarovka-classic ... -bila-ra97
Please bear in mind this is the new (September 2021) system of efficiency classes. Class G of 2021 corresponds to class A+ of pre-2021. My bulb's efficiency is about 80 lm/W, while most LED bulbs from the same company (EMOS) are in the E class (about >110 lm/W). The most efficient warm white bulb that is sold widely and the only such inhabitant of the C class is to my knowledge the IKEA Solhetta at 160 lm/W (as of February 2022).
Thank you for pointing this out, I hadn't recognizes this change. 80lm/W instead of 160lm/W sonds quite ok, no reason not to use it for this application. At 2700K it might be a bit yellowish for a nice visual impression, but this is personal taste.

The green LEDs I bought are similar to this one: https://www.ebay.de/itm/132454599362?ha ... Sw~RVaS2TT
I haven't put them into use so I can't comment on how well they mix with typical white LEDs.

MicroBob
Posts: 3154
Joined: Sun Dec 25, 2016 9:11 am
Location: Northern Germany

Re: LED lighting....why the hate?

#47 Post by MicroBob » Fri May 27, 2022 6:22 am

Admin notified...

MichaelG.
Posts: 3976
Joined: Mon Apr 10, 2017 8:24 am
Location: North Wales

Re: LED lighting....why the hate?

#48 Post by MichaelG. » Fri May 27, 2022 6:50 am

Here is a very detailed data-sheet from LUMILEDS
https://lumileds.com/wp-content/uploads ... asheet.pdf

Clearly, the haystack in which we are searching for this beautiful needle is large.
:ugeek:

MichaelG.
Too many 'projects'

MichaelG.
Posts: 3976
Joined: Mon Apr 10, 2017 8:24 am
Location: North Wales

Re: LED lighting....why the hate?

#49 Post by MichaelG. » Fri May 27, 2022 7:13 am

Sorry, I am not clever enough to explain this diagram succinctly
.
5E5452B7-EAD6-49FE-8D28-40FBE176A308.jpeg
5E5452B7-EAD6-49FE-8D28-40FBE176A308.jpeg (139.72 KiB) Viewed 4651 times
.

MichaelG.
Too many 'projects'

Sure Squintsalot
Posts: 393
Joined: Mon May 16, 2022 3:44 pm

Re: LED lighting....why the hate?

#50 Post by Sure Squintsalot » Fri May 27, 2022 2:56 pm

As a tech photographer, I used to routinely use one of these to calibrate color rendition for different lighting set-ups.
Screenshot 2022-05-27 085153.jpg
Screenshot 2022-05-27 085153.jpg (38.68 KiB) Viewed 4619 times
Super useful for one-click calibration in post processing with, say, Lightroom.

Does such a thing exist for microscopy?


Edited to add:


...apparently there is!
Screenshot 2022-05-27 090033.jpg
Screenshot 2022-05-27 090033.jpg (42.74 KiB) Viewed 4614 times
https://www.appliedimage.com/product/ni ... 9c-seccal/

But, could you use one of these, despite the serious limitations?
Screenshot 2022-05-27 091050.jpg
Screenshot 2022-05-27 091050.jpg (71.53 KiB) Viewed 4612 times

jjtr1
Posts: 33
Joined: Sat May 07, 2022 8:51 am

Re: LED lighting....why the hate?

#51 Post by jjtr1 » Fri May 27, 2022 3:18 pm

MichaelG. wrote:
Fri May 27, 2022 4:43 am
… if we can identify the chip manufacturer, the should be available [somewhere] on a reel

The search is on !!
Look, the EMOS bulbs are available on German Amazon! I thought EMOS only sells in Czech Republic.
https://www.amazon.de/Colour-Rendering- ... th=1&psc=1
They have E14 and E27, warm and neutral.
I hope I'm not breaking some rule with the link.

At 2.5 EUR per bulb, with about 30 chips each, why buy a reel? :) Assuming you're good with soldering iron, which I'm not.

jjtr1
Posts: 33
Joined: Sat May 07, 2022 8:51 am

Re: LED lighting....why the hate?

#52 Post by jjtr1 » Fri May 27, 2022 4:14 pm

MichaelG. wrote:
Fri May 27, 2022 7:13 am
Sorry, I am not clever enough to explain this diagram succinctly
.
5E5452B7-EAD6-49FE-8D28-40FBE176A308.jpeg
.

MichaelG.
It says that some leds you get will be pinkish, some greenish, and if you pay premium you can get the whiter ones (bins), those near the white curve that goes from left to right. This is different from being warmer or cooler white or having high or low colour rendering index.

Even with great CRI the LED can have such tint errors. Tint error is a lottery, each batch of chips is different. I bought the EMOS high CRI LED that we are discussing here to use it above my bathroom mirror, for assessing colour shades etc. But it had a tint, so I ended up using an older IKEA LED that happenned to be tintless despite having much lower CRI.

jjtr1
Posts: 33
Joined: Sat May 07, 2022 8:51 am

Re: LED lighting....why the hate?

#53 Post by jjtr1 » Fri May 27, 2022 6:06 pm

MicroBob wrote:
Fri May 27, 2022 5:09 am
80lm/W instead of 160lm/W sonds quite ok, no reason not to use it for this application. At 2700K it might be a bit yellowish for a nice visual impression, but this is personal taste.
This might be a good place to mention that the highest theoretical efficiency for making warm white light with a LED-like spectrum is about 320 lm/W. That's 100% electricity converted into light, 0% heat.
https://www.dial.de/en-GB/projects/effi ... white-led/

So we're already at 50% efficiency in mass produced bulbs, which I find incredible. There has been a massive jump in efficiency between 2019 and 2021.

MichaelG.
Posts: 3976
Joined: Mon Apr 10, 2017 8:24 am
Location: North Wales

Re: LED lighting....why the hate?

#54 Post by MichaelG. » Fri May 27, 2022 9:39 pm

jjtr1 wrote:
Fri May 27, 2022 3:18 pm

At 2.5 EUR per bulb, with about 30 chips each, why buy a reel? :) Assuming you're good with soldering iron, which I'm not.
.
I think you have answered your own question

MichaelG.

.

P.S. __ I wasn’t actually suggesting purchase of a whole reel !!
Too many 'projects'

MichaelG.
Posts: 3976
Joined: Mon Apr 10, 2017 8:24 am
Location: North Wales

Re: LED lighting....why the hate?

#55 Post by MichaelG. » Fri May 27, 2022 11:38 pm

I have just found an EMOS catalogue from 2020 : https://www.emos.eu/wp-content/uploads/ ... 020_GB.pdf

See catalogue page 97

MichaelG.
Too many 'projects'

smollerthings
Posts: 457
Joined: Thu Aug 19, 2021 12:10 pm

Re: LED lighting....why the hate?

#56 Post by smollerthings » Sat May 28, 2022 5:37 pm

EYE C U wrote:
Tue May 24, 2022 1:03 am
99% sure I damaged my left eye using a led source that turned out to have high uv.. the only reason i didn't damage both is my right eye was all fogged up with blood from a internal blood leak. I was right eye blind for about 6 months.. now the right is better but the left still has a greyed out area. went back to a reg bulb source
Wait what? How powerful was that thing and how long did you stare at it?

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

Re: LED lighting....why the hate?

#57 Post by LouiseScot » Sat May 28, 2022 6:15 pm

smollerthings wrote:
Sat May 28, 2022 5:37 pm
EYE C U wrote:
Tue May 24, 2022 1:03 am
99% sure I damaged my left eye using a led source that turned out to have high uv.. the only reason i didn't damage both is my right eye was all fogged up with blood from a internal blood leak. I was right eye blind for about 6 months.. now the right is better but the left still has a greyed out area. went back to a reg bulb source
Wait what? How powerful was that thing and how long did you stare at it?
I think normal (non-fluor) optics can't transmit much UV, though visible violet/short wave blue light can be potentially damaging. However, I wouldn't have thought the light intensity transmitted by a typical microscope would be anything to worry about. Just think of the UV and blue light you get exposed to looking at a blue sky or outside on a sunny summer's day (without UV sunglasses)!

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
blekenbleu
Posts: 300
Joined: Tue Feb 25, 2020 5:55 pm
Location: South Carolina low country
Contact:

Re: LED lighting....why the hate?

#58 Post by blekenbleu » Sun May 29, 2022 12:08 am

Sure Squintsalot wrote:
Fri May 27, 2022 2:56 pm
As a tech photographer, I used to routinely use one of these to calibrate color rendition for different lighting set-ups.

Screenshot 2022-05-27 085153.jpg

Super useful for one-click calibration in post processing with, say, Lightroom.

Does such a thing exist for microscopy?

Edited to add:

...apparently there is!

Screenshot 2022-05-27 090033.jpg

https://www.appliedimage.com/product/ni ... 9c-seccal/

But, could you use one of these, despite the serious limitations?
Screenshot 2022-05-27 091050.jpg
That 35mm slide is used to calibrate slide scanner systems for that Fuji film,
while the xrite (originally Macbeth) color checker was specially formulated with unique colorants
to help identify color rendering issues (illuminant metamerism, this LED issue).
Since all colors in that Fuji slide use different combinations of the same 3 colorants,
it cannot provide details about illuminant spectral power distribution uniformity,
only overall balance and nonlinearity of their mixing. In fact, commercial
35mm color film scanning was traditionally done using very narrow-band densitometers.
That NIST IAM-9C-SECCAL is roughly comparable to the color checker,
except for transmission instead of reflectance,
since spectral transmission in 5nm wavelength increments is included.
Having identified that some LED is effectively "blind" at some wavelengths
does not imply that those gaps can be compensated;
achieving overall color balance does not equate to detecting specific spectral responses
e.g. fluorescence. Put another way, calibrating an LED using the Fuji color slide
does not guarantee that any of those slide colors would match,
as perceived by normal human vision in daylight illumination,
the same color on a biological slide sample illuminated by that LED.
Metaphot, Optiphot 1, 66; AO 10, 120, EPIStar, Cycloptic

MichaelG.
Posts: 3976
Joined: Mon Apr 10, 2017 8:24 am
Location: North Wales

Re: LED lighting....why the hate?

#59 Post by MichaelG. » Sun May 29, 2022 7:02 am

blekenbleu wrote:
Sun May 29, 2022 12:08 am
Having identified that some LED is effectively "blind" at some wavelengths
does not imply that those gaps can be compensated;
achieving overall color balance does not equate to detecting specific spectral responses
Thank you for stating that so succinctly
You have described, in just a few words, the essence of the problem.

The page that I linked in Louise’s concurrent thread elaborates on the ‘workaround’
… known as Correlated Colour Temperature
https://www.lrc.rpi.edu/programs/nlpip/ ... tisCCT.asp

MichaelG.

.

Ref. viewtopic.php?p=122563#p122563
Too many 'projects'

User avatar
blekenbleu
Posts: 300
Joined: Tue Feb 25, 2020 5:55 pm
Location: South Carolina low country
Contact:

Re: LED lighting....why the hate?

#60 Post by blekenbleu » Sun May 29, 2022 11:36 am

MichaelG. wrote:
Sun May 29, 2022 7:02 am
Correlated Colour Temperature
https://www.lrc.rpi.edu/programs/nlpip/ ... tisCCT.asp
Perhaps it helps to also consider an analogous issue with audio:
  • most transducers (and nearly all rooms) have frequency response peaks and nulls
  • room nulls cannot be compensated by parametric or graphical equalizers and more power
  • (however, room nulls can often be mitigated by relocating microphones or loudspeakers)
  • even with sharp nulls in low frequency response, a system can still sound, overall, bass-heavy
Even with a substantial droop in spectral power between peak green and blue,
an LED can still produce a perceived overall bluish (high correlated color temperature) cast.
Metaphot, Optiphot 1, 66; AO 10, 120, EPIStar, Cycloptic

Post Reply