LEDs for autofluorescence microscopy. Help needed.

Pau · Post by **Pau** » Fri Sep 04, 2015 4:22 am

I have collected some epifluorescence equipment for my Zeiss Standard and I’m planning to make a multi LED illuminator
I’ve used my 10W white LED with not bad results but the fluorescence is very dim and often I need exposures about 2 – 3 minutes at 400 ISO. It doesn’t provide UV excitation.

My idea is to make a longitudinal slider made of heat sink aluminum profile (if it is available!) with up to 8-10 LEDs ordered according with their wavelengths and a common power source and dimmer, of course with independent switches for each LED. The slider will place the LED at the adequate position of a Zeiss lamphouse collimator lens. If the LED emission is narrow enough I could eliminate the need of the excitation filter like in laser microscopes

The goal is a system where the excitation light source could be easily exchanged, even allowing to make multiwave images. BTW I have no budget to afford a Zeiss Colibri and I want to avoid arc lamps.

My main problem is that I’m absolutely unskilled in electronics and LED equipment, so I’d need much help to select and buy the LEDs (I’ve written another post at the Equipment discussions forum about the suitable wavelengths) and the other electronic parts

Could you help me?

Bushman.K · Post by **Bushman.K** » Fri Sep 04, 2015 9:47 am

It is relatively easy to assemble any LED light fixture, using commercial off-the-shelf components and some soldering.

But you have to start from LEDs selection, because it defines all other requirements for the rest of circuitry.

When you will have wavelengths and average light output, it shouldn't be a problem to pick LEDs and power sources for it.

TheLostVertex · Post by **TheLostVertex** » Fri Sep 04, 2015 10:17 am

I have no experience with autofluorescence microscopy. So I am wondering how strong of a light source will you need to excite the dyes adequately for observation and photography? Do you plan on using the led's as a flash or pseudoflash, or constant source? Constant makes setting up the electronics very simple.

Output, formfactor, and wavelength all are considerations. Setting up the electronics should be a pretty easy task after you have figured out all the design needs.

Ichthyophthirius · Post by **Ichthyophthirius** » Fri Sep 04, 2015 10:40 am

Hi Pau,

You don't need many colours. UV is phototoxic, only worth getting if you really need it.

Most useful are royal blue (455 nm) or blue (470 nm) and maybe a green LED (525 nm). They also have the fit the filters you will be able to get.

The Cree XP-E are popular for blue excitation: https://www.led-tech.de/de/High-Power-L ... 0_138.html

The LED emission is not spectrally clean; the royal blue LED still emits 1% of its light in green (> 500 nm) http://www.mikroskopie-forum.de/index.p ... 7#msg39197 ; unless your objects are very bright you'll still need an excitation filter.

Pau · Post by **Pau** » Fri Sep 04, 2015 10:46 am

Thanks for your tips.

I need the LEDs to be as small and powerfull as posible, big arrays don't match well with the microscope illuminator designed for the small light source of an arc discharge lamp.

Another requeriment for my desing is that all LEDs can be powered with the same source.

I need continous illumination, no need for flashing.

A very desirable feature is a narrow emision spectrum to avoid the need of costly excitation filters and dichroic beamsplitters (I still will need the emission filters and some generic beamsplitters, but I already have some of them) so I need to buy from a reliable source that posts detailed technical data.

About the slider itself, does anyone know a source of heatsink sold in long pieces to cut?. It doesn't need to be very effective as the maximum continous working time will be of few minutes to avoid sample degradation due to photobleaching. No fan coolers possible due to vibration.
EDIT: I've found this one: https://www.led-tech.de/en/High-Power-S ... 6_114.html Could it be adequate?

Itchy, thanks again, we were typing at the same time

zzffnn · Post by **zzffnn** » Fri Sep 04, 2015 11:09 am

Ichthy,

I thought most popular UV LED has spectrum at around 365nm or 395 nm? It depends on the subject of course.

Edit: sorry, I got it wrong. It looks like for chloroplast autofluorescence , excitation wavelength should he between 420-460 nm :

http://www.biotechniques.com/Biotechniq ... 86142.html

Ichthyophthirius · Post by **Ichthyophthirius** » Fri Sep 04, 2015 11:11 am

Hi Pau,

Do you have the epi-fluorescence condenser IV FL? http://www.manualslib.com/manual/586177 ... e=2#manual

It takes the Zeiss light source Leuchte 60, which has a collector lens and plenty of space for LED cooling bodies www.photomacrography.net/forum/viewtopic.php?t=28069

Pau · Post by **Pau** » Fri Sep 04, 2015 11:28 am

Yes, I have the epiilluminator IV FL, a lamphouse 100 and a lamphouse 60 both bought to profit the collimating lenses. I like the 75RR's solution but I want to use the collimation lens (likelly form the lamphouse 100 because it is focusable) and placing the multi LED slider with a DIY adapter.

In the german forum you linked someone made a pretty similar arrangement, i'ts the first time I've seeen this kind of setup:
http://www.mikroskopie-forum.de/index.p ... 7#msg31687

Bushman.K · Post by **Bushman.K** » Fri Sep 04, 2015 12:10 pm

As I said before, there is no problem to find LEDs from the same maker with different wavelengths (Lumileds, Cree...), including ones already mounted on typical "star"-shaped MPCB. They can have similar working current, therefore you can use single constant current driver to feed them.
Typical color led has full spectum width about 100nm, while effective spectrum is, for sure, much more narrow - about 40nm wide. So, pick the LEDs, then it's easy to find power source.

Pau · Post by **Pau** » Sat Sep 05, 2015 8:58 am

Bushman, than you, 100nm spectum width can be adequate in many cases, at least if the emission filter is good engouh

jvmarti · Post by **jvmarti** » Wed Sep 09, 2015 9:06 am

Hi Pau,
I have found an interesting arrangement by Dr. Horst Wörmann for Multi LED using Zeiss HBO-100 lamp.
See: http://www.mikroskopie-bonn.de/biblioth ... index.html
under the heading "Der LED-Revolver - Umbau eines Zeiss HBO-Lampenhauses auf LED-Technik".

Hope it helps for your project.
Best regards.

Pau · Post by **Pau** » Wed Sep 09, 2015 10:24 am

Hola Jordi, benvingut al fòrum!

Thank you very much for the interesting link, I've not found this site (well, I don't search in German, but Google translation makes it very understandable)