Sun@Home LEDVANCE "SunLight" purple led + Tri-R from Seoul Semiconductors

Other Types of Lights LED Light Bulbs
1

Hi,

I’m new here. Looking for some good human centric lighting.
Couple of years ago I’ve found after many tests the budget Arena Lighting high CRI Ra97, high R9 97, low blue, it seems good quality treated blue leds…
BTW, I’ve found here that @maukka has tested them (5th position :wink: ). My test has been my eye and spectrum curve from the supplier. As the bulb is GREAT with pleasant lighting we have bought n°24 of those. Not available any more at 3200K.
They have today the replacement one, moreless at 2900K (too warm for us)

Waiting for the “Purple” leds + Tri-R at normal budget to equipe all the home; I’ve seen and bough n°20 of Sun@Home E27 bulb from LEDVANCE, high CRI, low blue, they claim SUNLIKE full spectrum; so it should be the purple led + tri-r treatment from Toshiba etc.
Dimmable, 2200 - 5000K…with tri-R…at 9.7eur per unit/bulb at 5 years warranty, at discount, it is THE DREAM
Even wanted to order n°10 more but as usual I double check things and something has not be so clear, at least for me as I don’t have a instrument to be objective and measure (and not technical knowledge for this speciality)

As the light of the bulbs is pleasant, Colors are well rendered, I’ve ordered also the Sun@Home PANON panel for my son desk, also the Sun@Home desk lamp to test etc.
All of this up to the constructor is the 2200-5000K wifi dimmable, full spectrum SUNLIKE. So ALL WITH THE SAME LEDS (should be 3030 from seoul semiconductors)

BUT, I’ve been surprised when I’ve discovered that the spectrum curve supplied from the EU certification site IS NOT the same for on, for on one side “bulbs” (e27, e14, par16) and on other side for panels, integrated led lamps etc???
The panels like PANON have the spectral curve for CCT 5000K that looks like the real SUNLIKE purple led tri-r model (as in their marketing presentations)
The bulb range Sun@Home has, up to their EU certification docs, the spectral curve similar to the “true light” blue leds or a little better than the best treated blue leds, but with huge blue light part, it is ok, but with the hole in the turquoise part of the spectrum. And this curve is not the case for the true SunLIKE purple led + tri-r…at least up to my understanding or I’ve missed something?!

Before I’ve found it and before my order, I’ve asked LEDVANCE support to send me the “bulbs” test curves.
They have sent me the curve for 5000K, thanks to them, so like it is for the E27 bulb.
But I’ve found that it is not for bulbe , it is for the PANON product range (so real sunlike).
With the QR code from the packaging, I’ve discovered the link to the docs where it is completely different for the “bulbs” (more similar to improved blue leds???)

If someone can help to understand this.
Maybe #Maukka

here the images that explains my doubts and the claim that I’ve sent to LEDVANCE, but no answer from them.
we’ll see as if not I’ll not miss the return window for n°20 bulbs e27.
Maybe this is the error, we are 25th November 2022 today, that Ledvance will correct in their certification documents for CE EU, who knows, real strange, or they might mixed some different leds inside for bulbs (I’ve seen 100.000 on/off cycles guaranteed for bulbs, and only 10.000 to 15.000 cycles for panels…maybe related to the leds, maybe power supply)

The GOOD point is that it looks fine for the panel and lamp model, but I don’t have the spectrophotometer to compare to bulbs.

Cheers

2

They are RGBW bulbs, which means that the white light is provided by a SunLike LED with a CTM of, say, 3,500K; so that in order to emit 5,000K, they need to mix white and blue, resulting in a graph that shows the SunLike spectrum mixed with a spike of blue; that’s why you see an apparent gap at ~480 nm (turquoise) that you would not see in true 5,000K SunLike.

Similarly, to get white light with a CTM of 2,200K, they would mix the SunLike LED and a spike of red.

I don’t know if this is true, but I believe it could be, and would explain your confusion. Let me know.

3

Ignore the peak and dip for a moment and look the wavelength at which meaningful light starts being emitted, violet pump starts around 400nm whereas blue pump start at 420nm or later.

Although it starts a little later than i would expect from a violet pump LED the meaningful light output is still earlier than i would expect from a blue pump LED so i think the first image is the spectrum of a violet pump LED, but it doesn’t have the spectrum balance you’d expect from a Toshiba-SSC Sunlike 5000K LED.

Assuming the incorrect spectrum is not caused by a poor choice of diffusion material you’d be right in thinking these are not 5000K 3030 Sunlike LEDs.

‘Sunlike’ covers a wider range than the 3030 LEDs though, i know they COBs as well but i don’t think the spectrograph for those would look like your first image.

In fact i just looked at this datasheet:
http://www.seoulsemicon.com/en/product/document/4975
and this COB at least doesn’t have the spectrograph of your bulb.

Checked another couple:
http://www.seoulsemicon.com/en/product/document/4974
http://www.seoulsemicon.com/en/product/document/4724
and they’re the same as the first.

It’s possible there’s something else in the Sunlike range that’s in your bulb, but i doubt it.

I didn’t know about the Sun@Home range, thanks for the info. I use Sunlike LED strips for main lighting (and some Soraa spotlights) but these are no good for giving to other people so this ready made solution would be perfect if i can get them.

4

Ohhh, i completely missed that they have a variable colour temperature :person_facepalming: .
Your explanation makes sense and makes my post largely redundant.

5

There is still something strange. If my previous explanation is correct, the green part of the spectrum should be lower in height than the red part, which is a characteristic of a color temperature lower than 5,000K. However, the green and red parts of the spectrum have the same height! It could be that they are mixing both blue and green with a warm white light to create the appearance of a 5,000K color temperature (for example, the brand LIFX does this). However, we do not see a (narrow) spike in the green part of the spectrum. That makes me not entirely sure.

6

Anyone tested flicker in Sun@Home bulb’s (E14,E27, other) ?

I found only this test report, with spectre & Ra measurements, but nothing about flicker.
Test-Report LEDVANCE Sun@HOME 9W Classic A40 E27

warm led’s is 2200K (2150-2200K), probably not Sunlike (450nm peak without 415-420nm peak), but also HiCRI, Ra>90,R9>50.
cold led’s is 4800K, probably Sunlike ( 415-420nm peak + 450nm peak, typical spectre), Ra>95,R9>80.

As I know, Seoul Semi not produce Sunlike LED’s with 2200K, at least I didn’t find any mention of them in official documentation.