• You are viewing the forum as a Guest, please login (you can use your Facebook, Twitter, Google or Microsoft account to login) or register using this link: Log in or Sign Up

Chihiros WRGB2 Pro has a separate 'white' LED channel but...

adriancarr

New Member
Joined
9 Feb 2022
Messages
13
Location
Edinburgh, UK
I received the Chihiros WRGB2 Pro 45cm light today from China, so I thought I would do a quick test to confirm it has a separate 'white' LED unlike the original WRGB2.

The good news, is that it does have a separate white LED. I had feared it might just be the software simulating the presence of a white LED using the RGB channels.
The bad news is that the white LED isn't really full spectrum and the light still has a gappy spectrum, though it does improve on the RGB alone.

The below images were taken with spectroscope jerry-rigged to my mirrorless camera. Camera settings were consistent for all images. I plan to extract more precise spectra from the images, but the current unprocessed images still illustrate the results pretty well. The AI Prime Freshwater spectrum shows what a full spectrum LED aquarium light looks like. I included the CFL spectrum for calibration purposes, as it has bands from mercury at particular wavelengths: 436nm and 546nm.

The gappy spectrum isn't an issue for growing plants, though it will affect how accurately colours appears. Chromatic aberration (CA) can also become an issue when there are discrete wavelengths with gaps between them, especially for those who wear strong glasses (which I do).

I hope people find this useful. It's a beast of a light, with some limitations.

Adrian

Chihiros WRGB Pro comparison.jpg
 
Do you think yellow shrimps and fish are going to look very natural under this light? Doesn't look that much of an improvement over regular RGB.
 
Do you think yellow shrimps and fish are going to look very natural under this light? Doesn't look that much of an improvement over regular RGB.

That depends if the shrimp are yellow because they reflect yellow light (e.g. 580nm) or if they look yellow because they reflect both green and red light. Adding the 'white' LEDs definitely makes the colours look different, and probably better. The white LED fills in the gap a bit between green and red, so it should look better, but it is difficult to assess how much. I unfortunately don't have a planted/stocked aquarium at the moment to see how it looks.
 
Very nice, thank you for sharing your work. I wonder if the gaps for the 'white' leds don't come from the leds also being very peaky? In other words the phosphors on the leds are "full-spectrum" (marketing keywords) but the radiation emitted between the peaks was too low to be captured on your camera. The white really looks less bright compared to others.

This will would still produce very saturated colors but at least provide a smal filler for shades of cyan, green and orange instead of circus colors only. Hope they make for better lights in person but for a majority of the target buyers the colors will anyway end up with saturation and contrast 210% for social media.
 
Very nice, thank you for sharing your work. I wonder if the gaps for the 'white' leds don't come from the leds also being very peaky? In other words the phosphors on the leds are "full-spectrum" (marketing keywords) but the radiation emitted between the peaks was too low to be captured on your camera. The white really looks less bright compared to others.

This will would still produce very saturated colors but at least provide a smal filler for shades of cyan, green and orange instead of circus colors only. Hope they make for better lights in person but for a majority of the target buyers the colors will anyway end up with saturation and contrast 210% for social media.
Photos fairly common for low cri high k whites.
Double band in red is a bit odd.


Depends on the phosphors used.
The " all" spectrum sort of confirms the white only spectrum.
 
Photos fairly common for low cri high k whites.
Double band in red is a bit odd.

The double band in red is something I see from my Macbook screen through a spectroscope, but not other displays or lights I've tested.
There seem to examples in the literature of multi-peaked red phosphors doped with Pr3+ (Sun et al, 2020), so I guess it is something similar.

I have ordered some blue (460-475nm) and yellow (590-595nm) 2-3W LEDs and I'm going to see if I can improve the colour rendering and general look of the light. The blue has a lower wavelength than I wanted, I was looking for more of a blue-green 480-490nm. The light is perfectly nice as it is, I'm just tinkering with it out of curiosity.
 
The double band in red is something I see from my Macbook screen through a spectroscope, but not other displays or lights I've tested.
There seem to examples in the literature of multi-peaked red phosphors doped with Pr3+ (Sun et al, 2020), so I guess it is something similar.

I have ordered some blue (460-475nm) and yellow (590-595nm) 2-3W LEDs and I'm going to see if I can improve the colour rendering and general look of the light. The blue has a lower wavelength than I wanted, I was looking for more of a blue-green 480-490nm. The light is perfectly nice as it is, I'm just tinkering with it out of curiosity.
You may find this interesting
 

Attachments

  • ledcolor.pdf
    1.9 MB · Views: 228
You may find this interesting
Thanks for the interesting paper. It seems there is a real lack of phosphors that emit between 460-500nm (based on fig1 in the paper). I'm not clear if they are difficult to produce or if manufacturers don't bother developing them as colour vision (cones) in the human eye are least sensitive in this range. It does seem like most white LEDs (blue LED with phosphors for green/red light) have a dip in this region.

Cone-fundamentals-with-srgb-spectrum.svg.png
 
Thanks for the interesting paper. It seems there is a real lack of phosphors that emit between 460-500nm (based on fig1 in the paper). I'm not clear if they are difficult to produce or if manufacturers don't bother developing them as colour vision (cones) in the human eye are least sensitive in this range. It does seem like most white LEDs (blue LED with phosphors for green/red light) have a dip in this region.

View attachment 184685
Not the 95+ cri whites.
Anyways the current way around it
2 different blue pumps, one phosphor pack.
Bridgelux cob.
Vesta%20Thrive_0.png

Luxeon " fresh fish" one blue and phosphors.
fish.jpg
 
Last edited:
Fun w/ phosphors..

JUST for fun on Friday..
Why there is no "purple" or "violet" only deeper blue..
The more observant among you have noticed that cone cells that respond to the color red also produce a signal for parts of the visible spectrum corresponding to blue. And purple is a mix of red and blue. Coincidence?! No; obviously.

The thing is, while every color you perceive looks real, they’re pretty much all just hallucinations of your brain.
Scientifically, purple is not a color because there is no beam of pure light that looks purple. There is no light wavelength that corresponds to purple. We see purple because the human eye can't tell what's really going on
 
Last edited:
Not the 95+ cri whites.
Anyways the current way around it
2 different blue pumps, one phosphor pack.
Bridgelux cob.
Vesta%20Thrive_0.png

Thanks. I've ordered some of the above Bridgelux Thrive LEDs to try: 3500K, 4000K and 5000K. Unfortunately, I couldn't find the 2700K or 6500K.
 
Thanks. I've ordered some of the above Bridgelux Thrive LEDs to try: 3500K, 4000K and 5000K. Unfortunately, I couldn't find the 2700K or 6500K.
Yes that has been the universal problem.. Availability only on paper.
Not to long ago I believe some were at Digikey or Mouser..
Yea here..
5000k are always available.

You can always go the more expensive route of the "fresh fish" COB's.
Actually a really interesting pairing is Fresh fish/crisp white..
Crisp white has some violets chips inside to "brighten" white (flourescence) but runs warm at about 3500K.

Keep in mind 2 things
1) Chips below like 4500K are judged (CRI) against an incandescent light
not sunlight.
2) Natural doesn't always reflect as "punchy".. just err natural.
 
I asked Chihiros support for the colour spectrum of the WRGB2 pro. They replied after almost 3 weeks with what appears to be a relabelled plot for the WRGB2 (which doesn't have a 'white' LED). A disappointing response from them.

chihiros_1.jpeg
 
Yes that has been the universal problem.. Availability only on paper.
Not to long ago I believe some were at Digikey or Mouser..
Yea here..
5000k are always available.

You can always go the more expensive route of the "fresh fish" COB's.
Actually a really interesting pairing is Fresh fish/crisp white..
Crisp white has some violets chips inside to "brighten" white (flourescence) but runs warm at about 3500K.

Keep in mind 2 things
1) Chips below like 4500K are judged (CRI) against an incandescent light
not sunlight.
2) Natural doesn't always reflect as "punchy".. just err natural.

I went for some strip modules from Digi-Key BXEB-L0280Z-50S1000-C-C3 Bridgelux | Optoelectronics | DigiKey

The Luxeon COBs look interesting, I'll take a closer look.
 
For the past week, I have been emailing chihiros about the pro. In fact, they sent me the spectra and PAR data before it was posted on their ChihirosAquaticStudio.
Yesterday, I asked them to check with their engineers to confirm the labeling of these test results. Also to present the spectrum &PAR: with 0% white and only for white.
 
Any news on this? I'm looking for the best lighting option for my new build. I just want the highest CRI/RA and the most accurate colour reproduction with plenty of power and spectral control via app or controller. Thanks.
 
Back
Top