I was thinking about doing a lit warm white tail cap light. I was also thinking about putting maybe a red transparent sheet between the rubber boot and the emitters to create an imperfect incandescent color light without a cheesy looking tail cap. You could probably do that with any color like incandescent green for that look. Just thinking aloud. Anyways, really cool build. I wish we could see it in person as pictures can’t do it justice. Like with the 2000K E21A. I thought that emitter would put out an unpleasant flat orange light, but it turned out to be a lovely orange/warm tint. It looks like SON-Deluxe/Deluxe HPS, but better. You can see some HPS lights being overdriven, and they may exhibit a more white-ish color like the 2000K E21A/SON-Deluxe.
Luminus has some 1800K SMD LEDs from Mouser like the XD16 that might be interesting candidates for future mods. There’s a 95 CRI varient to.
Yes I know, I have looked into the Luminus Cube leds, but last time I checked the 1800K 95CRI variant was nowhere for sale in small quantities. And Mouser is always troublesome for us Europeans because of the high shipping costs.
And of course it was fun to achieve it this way, I found it surprisingly easy to turn 2700K 80CRI into 95CRI by simply adding the right amount of red.
Btw, I ordered some 2700K 80 CRI Nichia NSSLT02A leds from leds.de that I just found yesterday, much cheaper than the XD16 for lighted tail use, 10 cent a piece.
I wouldn’t call this totally pointless. It’s a rare glimpse into spectral power distribution tuning and improving CRI/R9. :sunglasses: Time you enjoy wasting isn’t wasted time.
Supposedly Cree did something like this years ago in their light fixtures to help with how inefficient warm white emitters were in the reds, they used separate red emitters and warm white in the same street light fixture. I haven’t been able to find much more information on this though.
Djozz, I had a separate question. Did you get your 2000K E21A’s from Clemence?
Those NSSLT02’s look interesting. Hopefully with better deep reds making them an even better candidate for this R9 enhancement treatment. 98/99 R9 anyone?
Yes, they came from Clemence, I would not know another source for these rare leds, although leds.de (Lumitronix) has a good assortment of current Nichia leds.
Ah, for some reason my 2000K E21A’s had showed a higher color temperature of 2350K in my tests. Even with a faceted reflector and a frosted lens blending the light, I couldn’t get the CCT lower. It makes me wonder as the E21A looks like overdriven SON/HPS and PC Amber looks like under-driven HPS/SON, I wonder if the Luminus 1800K cube will look like stock HPS but with high CRI.
Just checked: in my Tool AA on a 14500 battery, the E21A is driven at 1.1A, reads 2100K in the hotspot and 1970K in the spill just outside the hotspot, both slightly under the BBL. The Tool has a (quality looking) AR-coating on the lens though that can mess a littlebit with the tint.
I have my setup in a quad with a 2.8 amp driver. Maybe, I’m not pushing the needed amps into the four emitters to get the color temperature down? Is it one of the cool white sst range that color temperature shifts on it’s drive currents, I think? It could be the optics at play like you said as well, and even metering equipment inaccuracy.
I hope it not metering inaccuracy because that is difficult to check, you mostly need to trust the manufacturer for that. The only check that I could do on my spectrometer myself is that I recorded a Helium lamp spectrum and at least all peaks were within the nanometer accurate. But that is not all that can be wrong.
As an addition to the OP, I wanted to reduce the red peak a bit more to see how that affected things, and at the same time get a bit more light overall, so I swapped the 8.9 kOhm resistor of the XD16 string to 4.7 kOhm. Result is 0.0016 lumen, so indeed a bit more , but what happened to the tint?:
Again a bit higher CCT as expected, now 2210K, CRI similar with R9 a bit down (but still great) and R11 and R12 went up, but still just a bit under the BBL. I’m ok with that for a firefly illumination so I keep it this way.
Nice mod Djozz, now you just started a DIY CRI/CCT making!
You should try mixing those colored E17A you have. I guess mixing any warm white LEDs with Amber and Red E17A will create extremely warm CCT with excellent CRI. They have VERY wide spectrum band compared to normal colored LEDs. I tested them and surprised to see the Amber looks much better than LPS.
I never did anything with those colour E17A’s I’m afraid, but I will at some point, they are excellent for colour mixing.
Another day, another idea for a project. I remember that sometime 2013 I decided that I had made my perfect EDC flashlight (with 219A 92 CRI) and that my hobby was finished
That was back in 2013, Fasttech did not exist at the time, I got them from Illumination Supply which is now Illumn.com. They were 4500K and it was then the only high CRI led that was used in flashlights.