Highest power LED suitable for microscope?

[PeteM]

I'm hoping someone knows the best option for a high power and well color corrected LED?

I recently bought a RetroDiode lamp for a spare Olympus BX40. Compared to the regular 30 watt halogen lamp, and using DIC, it seems to provide better color correction and sharper definition, but not a much brighter image. This is a 10 watt LED - far as I know about the brightest commonly available.

What I'd like is a well-corrected higher power led in a single die. Looks like maybe 5mm x 5mm would be ideal in terms of the luminous area.

Does such a LED exist that's both brighter and still with a fairly even distribution of energy across the visible spectrum?

[MicroBob]

Hi Pete,
the Cree XHP 50 and 70 of 18 and 32 W could be an option. White LEDs generally have a dent in the spectrum around 500nm. This is less severe on warm white models and they can be used with a light blue filter. A 4000K-5000K LED gives a nice neutral tone in my eyes.

Bob

[Saul]

Hi Pete,
the Cree XHP 50 and 70 of 18 and 32 W could be an option. White LEDs generally have a dent in the spectrum around 500nm. This is less severe on warm white models and they can be used with a light blue filter. A 4000K-5000K LED gives a nice neutral tone in my eyes.

Bob

Original heat sink is too small & power supply is to weak for these LEDs. Even using this type 2A Chinese power supply with 10w LED, voltage drops to 11v and less.

[Glot]

Are you looking for a direct replacement or something you can play around and build yourself? Quality LED chips produce about 100 lumens per watt now. You can get chips with a CRI of 97. For quality LED chips, you will need to look outside normal suppliers.

[PeteM]

Are you looking for a direct replacement or something you can play around and build yourself? Quality LED chips produce about 100 lumens per watt now. You can get chips with a CRI of 97. For quality LED chips, you will need to look outside normal suppliers.

I'm looking to build myself. Appreciate any suggestions you might have. Ideal, for me, would be equivalent brightness to a 100 watt halogen bulb, decent color rendition across the spectrum, a die about the size of a filament (say, 5mm x 5mm), and already mounted. I'd build my own heat sink, attach with thermal paste, etc.

[Rorschach]

This is an interesting topic that I will also be following. With some luck, at some point we might be able to compile a list of high-quality parts for all the DIY people.

I am still interested in the ready made LED products for the Orthoplan but apparently the finish quality of the 3D printed parts, mainly the bayonet fitting that attaches to the scopes light, isn't brilliant. Also some of the older models seem to produce banding in microphotography.

[Chris Dee]

Of the high power LED's I've tried some had issues with uneven luminosity due to patchy phosphorous distribution, this is visible through the eyepiece socket when focusing the lamp optics. The best I've found so far is the Cree XHP35, this is a 12v 13W monolithic die. The die size is 3.5x3.5mm and very close to original halogen filament sizes. There are 2 variants both available in high CRI bins, the High Intensity version (Cree XHP35 HI) doesn't have a dome lens and is designed for use behind secondary optics for maximum throw.

I don't have experience of DIC illumination requirements, but suspect there is something wrong with the drive of your 10w LED if its not providing more lumens than a 30w halogen. That would be the first thing I'd look into, even if driven at a conservative 6w it should exceed your old set up.

[DrPhoxinus]

Saul,

Can you tell me what a Type 2a power supply is?

I looked on the web and haven’t found a description.

I have been fooling around and ran into Ohm’s law with the current limits and voltage drop in small conductors.

Gerard

[daruosha]

Saul,

Can you tell me what a Type 2a power supply is?

I looked on the web and haven’t found a description.

I have been fooling around and ran into Ohm’s law with the current limits and voltage drop in small conductors.

Gerard

He meant 2 amps capable power supply.

[Saul]

Are you looking for a direct replacement or something you can play around and build yourself? Quality LED chips produce about 100 lumens per watt now. You can get chips with a CRI of 97. For quality LED chips, you will need to look outside normal suppliers.

I'm looking to build myself. Appreciate any suggestions you might have. Ideal, for me, would be equivalent brightness to a 100 watt halogen bulb, decent color rendition across the spectrum, a die about the size of a filament (say, 5mm x 5mm), and already mounted. I'd build my own heat sink, attach with thermal paste, etc.

Adapting original, or building from the scratch ?

[PeteM]

Are you looking for a direct replacement or something you can play around and build yourself? Quality LED chips produce about 100 lumens per watt now. You can get chips with a CRI of 97. For quality LED chips, you will need to look outside normal suppliers.

I'm looking to build myself. Appreciate any suggestions you might have. Ideal, for me, would be equivalent brightness to a 100 watt halogen bulb, decent color rendition across the spectrum, a die about the size of a filament (say, 5mm x 5mm), and already mounted. I'd build my own heat sink, attach with thermal paste, etc.

Adapting original, or building from the scratch ?

Given a mounted LED with the proper specs, I can take it from there either adapting an old housing or building one from scratch. In the original "spec" I should have mentioned a sufficiently wide and even illumination. Not clear to me if the dome on many of these LEDS is a plus (brightness) or minus (uneven illumination).

[daruosha]

Not clear to me if the dome on many of these LEDS is a plus (brightness) or minus (uneven illumination).

If you can, get a LED without the dome assembly. The dome acts like a lens and disturbs the microscope lighting optics arrangement.

[Glot]

You will need a COB style to get 10 watts. Bigclive.com on you tube has a lot of handy led info. If you are using raw LED then you need current limiting. Sounds like you need a better power supply if the voltage at its output is dropping. I use an iron ore lighting transformer (they are being thrown out regularly) they measure about 12 volts but poor regulation. A full wave rectifier, filter capacitor (2000 my per amp min) and a dc.dc buck circuit to regulate voltage to 12 volts as the COB I use is designed for 12 volts dc direct.

[Saul]

...Adapting original, or building from the scratch ?

Given a mounted LED with the proper specs, I can take it from there either adapting an old housing or building one from scratch...

Heat sink of the old housing is too small for the REAL 10w.
Did you try to measure voltage & current ?

[Glot]

There are basically two types of LED arrays. Constant voltage and constant current. I get the feeling you are trying to compare LED to incandescent directly. They are two different types of light source. Bit like comparing fluroescent to incandescent.

[Glot]

Something along these lines may be whst you are after.

https://www.lumileds.com/products/high-power-leds