Help please designing LED alternative lighting?

[dolmadis]

I would like to ask members to help me make the best choice I have for a single emitter led in replacement of a filament bulb replacing the electronics with new PSU, Driver etc.,

I think that the LED options I have seen members use are (but there may be more.........)

bare emitter

beaded emitter

bare emitter with a lens (either built into a single emitter LED bulb or a Carclo type lens mounted on a holder over the single emitter).

Discussion welcomed please of the pros and cons.

Thank you.

BR


John

[Pau]

More info, please:
- are you referring to a microscope illuminator? (I suppose you are)
- If so, what model?, built in the base or at a separated lamp house?, where is the collecting lens placed? (pictures...)

I think you've seen my fluorescence illuminator. I use the same kind of setup for transmitted light, I've commented it at several posts here, like http://www.photomacrography.net/forum/v ... 855#215855

[dolmadis]

Hi Pau

Yes, microscopes and EPI illuminators.

But rather than consider a specific application could the forum establish some generic principles to allow anyone coming to such a project to know where to start with the issues?

I know that the LED emitter should be placed at the same point as the filament from googling.

But does the presence of an instrument collector lens determine the type of emitter to be considered?

Smokedaddy found an issue with the PZO Biolar for example. I cannot find the link at the moment.

Does that principle follow on to other light paths in instruments?

BR

John

[dolmadis]

Found it http://www.photomacrography.net/forum/v ... hlight=pzo

To add to the discussion I think that I read a week or so ago on a Yahoo Group that the bead on an emitter could introduce spherical aberration and another poster suggested filing the bead flat !!

It strikes me that if a microscope illumination system has a collector lens or series of lenses in a separate lamphouse or a microscope base that the introduction of a lens over the emitter might cause problems.

But what of a bead? Is it as I recall reading on the Yahoo Group?

Could I also add to to the discussion what consideration should be given to LED reflectors and lenses of beam angle, frosted or clear and diffusion.

Just trying to understand the issues.

BR

John

[enricosavazzi]

The geometric radiation pattern of LEDs is quite variable, see examples at http://planck.reduaz.mx/~imoreno/Public ... ss2008.pdf . In principle, a collector lens should be designed/chosen to match a specific emission pattern. A batwing or side emission pattern can waste most of the LED emitted light if a collector of low NA is used, for example.

In practice, things may not be that critical, because power LEDs nowadays emit so much light that even wasting a good part of it by choosing the wrong collector leaves more than enough entering the collector. If your goal is to modify an existing illuminator that uses an incandescent or halogen lamp, just replace the lamp with a LED with a 100 degree or narrower emission pattern placed at the right spot, and provide enough cooling for the LED. If you get too little light, then it is time to think about more drastic measures like choosing a LED with a different emission pattern.

Filing the built-in bead lens may have unexpected results. Spherical aberration introduced by the bead lens would be the last of my concerns.

[Pau]

In practice, things may not be that critical, because power LEDs nowadays emit so much light that even wasting a good part of it by choosing the wrong collector leaves more than enough entering the collector. If your goal is to modify an existing illuminator that uses an incandescent or halogen lamp, just replace the lamp with a LED with a 100 degree or narrower emission pattern placed at the right spot, and provide enough cooling for the LED. If you get too little light, then it is time to think about more drastic measures like choosing a LED with a different emission pattern.

Filing the built-in bead lens may have unexpected results. Spherical aberration introduced by the bead lens would be the last of my concerns.

My thoughts and limited experience fully concur with Enrico's.
At certain step of modifying one of my illuminators we damaged the lens of the white LED, so I did cut it flat with a scalpel: no improvement and, being the surface not perfect, illumination was less uniform when checking it for Köhler. A LED with built in lens (I have no experience with single LEDs without lens) has a more uniform light emitting surface than a dedicated microscope bulb square filament.

An important point is to have the LED or collector lens focusable.

The only actual issue I have with my Cree XM-L based setup is the intensity regulation at low power, I need to use a strong ND filter for low to middle magnification BF

[Pau]

Smokedaddy found an issue with the PZO Biolar for example...

Rereading Smokedaddy's post the issue didn't come from the lens on he chip but form the extra hemispheric lens, once removed, it seems that the problem was solved, you can see the small lens on the chip at his last device pictures

[dolmadis]

So am I right in concluding that it is not recommended to use of an LED Bulb with an extra hemispheric lens in front of an LED emitter whether bare or beaded?

But a beaded emitter on its own has the benefit of uniformity of light?

When you say "An important point is to have the LED or collector lens focusable" I am not sure what that means. If the collector lens is fixed then is the capability of moving the LED closer or further away to the collector lens an expression of focus?

Thank you.

John

[Pau]

1- Yes, likely too many lenses. Beware that the illuminator collecting lenses are not designed for that hemispheric lens.

2- Maybe, but it seems it is not working well in this position

3- Yes, both systems do exist. In fact I have one of each for my Zeiss:
-The lamp house 100 has a focusable collecting lens
-The lamp house 60 moves the lamp to focus while the lens position is fixed
Both do work well

[dolmadis]

Thank you all. I have a better understanding now.

BR

John

[Andy Davies]

My friend has just let me try out this LED driver 152224875110 on Ebay.

No flicker on any video mode on my Nikon D810 and can lower the output to a comfortable viewing level. Doesn't break the bank at £3.65!

[dolmadis]

Oh dear. I thought that my knowledge was growing!!

May be not?

So this is a dimmer?

How does it fit in please?

Thanks


John

[Pau]

My friend has just let me try out this LED driver 152224875110 on Ebay.
No flicker on any video mode on my Nikon D810 and can lower the output to a comfortable viewing level. Doesn't break the bank at £3.65!

Hi Andy,
I'm interested in more info:
What LED are you driving?
Could you post the complete setup and parts?

My problem with single chip LEDs is their low voltage range with relatively high intensity, for example the Cree XM-L I use ranges from 2.7 to 3.3V going (theoretical) from 200mA to 3000mA *, so it seems more adequate as advised to drive it controlling current more than voltage.

With multiple LEDs arranged in series (COB LEDs, strips...) it's somewhat different because the voltage is higher and, I think, the voltage regulation could be easier.

With a voltage controller like the linked device it seems easy to burn the LED, although being very ignorant in electronics I would be happy being corrected.

* according with Cree dataset and graphics

[Andy Davies]

Hi Pau

It replaces the LuxDrive. I am using CREE XML which are attached to old CPU aluminium heatsinks with a fan. Have also tried using a Peltier cooler but didn't seem to achieve much in terms of overdriving the LED.

LED starts to trip if volts are too much.

[dolmadis]

My problem with single chip LEDs is their low voltage range with relatively high intensity, for example the Cree XM-L I use ranges from 2.7 to 3.3V going (theoretical) from 200mA to 3000mA *, so it seems more adequate as advised to drive it controlling current more than voltage.

* according with Cree dataset and graphics

Both Pau and I were both interested to know more.

According to a review of this device at http://www.epanorama.net/newepa/2016/04 ... le-tested/

There is no adjustable current limiting.

So sorry to ask but how can the device be set to deliver the voltages and amperage in the ranges that Pau defined?

AND

LED starts to trip if volts are too much.

Can you explain this a little more please?

Many thanks again and sorry for the naive questions but I don't have the knowledge and want to learn.

BR


John