Green light tests

[Cadmium]

Comparing two LED flood lights for use as Visual-Green induced IR fluorescence (LUM).

Ideally the light should be isolated between 400nm and 600nm, depending on what longpass filter is desired on the lens.

This filter stack that I am using on the small 'white' flood lamp at the right in these tests is: Schott VG9 1mm + S8612 2.5mm.

This is a 'thin weak' version of this stack, and it can be modified to cut lower, around 620nm. I have made it thin to test leakage into the UV also.

This stack cuts at 400nm (OD5), peaks at about 520nm, and cuts red/IR at about 640nm (OD3) - to - 680nm (OD5).

Many LED's can have very intense light, which in some cases can penetrate through even very strong OD suppression,

so these tests are to illustrate what amount of Red and IR light is getting through to my IR fluorescence shots.

Again, ideally for my uses I want only visual green.

 

The light at the left is a 60W green LED flood lamp. However, as you can see from the tests, these 'green' LED's would still need to be filtered to block light above 700nm. Not much heat.

The light on the right is a 10W visual (RGB) lamp, which when filtered works quite well. It stays amazingly cool. Very inexpensive lamp.

 

The top row are all shot in visual.

The second row is shot with RG610, RG665, and RG9 filters on the lens, to show how high the flood lights are transmitting into the red/IR range.

The bottom row is to test/illustrate how much UV is given off by these lamps, and how much UV is transmitted through the filter stack I am using in these tests.

Ideally, I don't want any UV light transmitted from the filtered light Visual-Green light, because IR light can induce IR fluorescence also, I don't want to mix the UV light with the Visual-Green light.

In the bottom row, right hand shot, I compare the UV output of the MTE 303 to the UV output of the other two lamps.

 

Just to double check, Baader U + S8612 2.5mm on the lens, shows that the green LED lamp is transmitting some UV. The green lamp needs to be filtered.

[colinbm]

Yes it sure looks like some wide-band LED lights Steve.

Realising that these LEDs are pretty powerful, how does the test go with just reflected light off some paper or PTFE target a metre or so away from the lights, please ?

Col

[Andrea B.]

Steve, that BaaderU + S8612 combo could be passing Violet light and not UV?? Test that with your U360+S8612 combo which I think passes much less Violet.

[Cadmium]

Col, Nice idea, but this demonstration shows me that the filter stack I used on the right hand lamp works to block IR as well as UV. It works the same on the green LED lamp. I may try something else that cuts off below 600nm at some point.

Andrea, U-360 stacks typically transmit a slight bit more above 400nm than the Baader U.

 

I also tested U-340 2mm which bloks most of the light from the green lamp, so it is mostly violet (400nm+) that is seen in the above test with the green lamp,

but the U-340 doesn't block most of the light from the white lamp, which obviously has a little more UV light.

 

Conclusion:

1) the lamps need to be filtered, even if they are only green LEDs.

2) the filter stack I am using works to block both IR and UV and even Violet (just not as much red as I want).

 

Here is the test stack used for the light filter. Either glass can be used thicker, but 620nm is about the lowest this can block by using thicker S8612.

[Andrea B.]

Andrea, U-360 stacks typically transmit a slight bit more above 400nm than the Baader U.

 

Well, I'm not so sure? Recently I got the following. But it could just be high red.

 

 

 

Raw histogram for BaaderU UV-Pass + Baader UV/IR-Cut

 

Raw histogram for U-360(2.0) + S8612(2.0) UV-Pass Stack + Baader UV/IR-Cut

There is at least 1/2 stop less leakage in the U-360 stack.

 

BaaderU + Baader UV/IR Cut after white balance and reset of white point (extreme push).

The photo shows violet, blue and green leak but very little red.

 

U-360 + S8612 + Baader UV/IR Cut after white balance and reset of white point (extreme push).

The photo shows some violet leak, but nothing else. Great combo!

[Cadmium]

Andrea, Nice test. A little off topic. I agree with you that my test of the green LED flood lamp is mostly violet.

So your test with the the Baader U + Baader UV/IR-Cut stack transmits more visual rang light than the U-360 2mm + S8612 2mm stack does? Would you agree with that based on your above test?

[Cadmium]

Andrea, I just did these tests. You may be onto something, but I don't know what it is exactly right now.

The first pic just shows the lamp, and the lamp lit up, both shot in visual.

 

This second set shows:

Row 1) Baader and U-360 2mmm + S8612 2mm stack on the lens (each alone).

Row 2) First row + Baader UV/IR-Cut filter.

Row 3) First row + Schott KV418.

UV/IR-Cut cuts just slightly below 400nm (with Red/IR blocking at about 700nm).

KV418 cuts above 400nm, almost identical profile to GG420 (but with no Red/IR blocking like the UV/IR-Cut has).

(note the circular VG9 + S8612 filter stack in each shot on the left)

[Andrea B.]

Very good test, Steve!

 

Somewhere in the back o' my mind.....I remember something about a bit of a green leak in some UV-pass filtration. This could be what you've captured. Later today I'll look for a reference to this. I have to find my hard drive with all the old UV photos on it. I know it's around here somewhere! :D

[Cadmium]

My test shows that the Baader is the one leaking more blue than the U-360 stack, as you were showing above also.

Are you are referring to the U-340 1mm thick 550nm+ leak?

The test above doesn't show that the stack is the one that is leaking visual.

My test seems to agree with your test.

[Andrea B.]

No, I was recalling either a test I did or something I read about a manufactured UV-pass filter. Not about the raw U glass. But I have not had a chance to look for anything yet.

 

May we please know the maker of the green LED array? :)

[Cadmium]

Here are the two lights I tried above, both from eBay.

272251509947 - 10W LED White Flood Light

361505836982 - 60W Green Flood Light

 

Both of these lights need to be filtered, and If they are filtered It doesn't matter if you get white or green.

I think I would get white, because it is more versatile.

 

LED's are not a great way to test filters. Colin's idea above is better for real life.

This topic was a test of the filter I am using on the lights, in combination with the filters I am using on the lens, for UVIIRF and VIIRF (LUM),

not to test UV filters. I think UV filters are best tested with real life reflected scenarios.

[Cadmium]

If anyone is interested, I just realized that Schott BG7 would cut at 600nm if about 3.5mm to 4mm thick.

It transmits some UV, so it could be stacked with VG9 1mm or GG420, depending on the situation.