New Cree J Series 3030

Prawn Connery

Well-Known Member
Heh, so, no one actually says that, but you have a whole rant about it anyway? :cool:
So no one actually says that . . .

wietefras said:
For comparing 3000K COBs/SMDs, lumen gives a decent indication of efficacy.
. . . except you? :-?

Which was kinda my point: two identical series diodes, one puts out more lumens, but it may or may not be more efficient, because lumens are weighted in favour of certain wavelengths and any colour shift will affect lumen count regardless of efficiency.

I guess you didn't read my rant . . . but that's OK :wink:
 

Randomblame

Well-Known Member
BTW, the Luxeon Fresh Focus for marbled meat are released earlier this year and they are comparable to Bridgelux Foot Decór Series in terms of effiency.
Lumen is still a pretty good clue to determining the efficiency of an LED, but difficult with such exotic spectra. But I think they can work anywhere between 2.5 and 2.7 μmol/J.
Someone had it counted and converted, I think it was @NoFucks2Give, but no idea where.

I imagine a light using ⅔ 4000°k/CRI80 and ⅓ 1750°k just made of strips instead of COB's.
 

Dave455

Well-Known Member
BTW, the Luxeon Fresh Focus for marbled meat are released earlier this year and they are comparable to Bridgelux Foot Decór Series in terms of effiency.
Lumen is still a pretty good clue to determining the efficiency of an LED, but difficult with such exotic spectra. But I think they can work anywhere between 2.5 and 2.7 μmol/J.
Someone had it counted and converted, I think it was @NoFucks2Give, but no idea where.

I imagine a light using ⅔ 4000°k/CRI80 and ⅓ 1750°k just made of strips instead of COB's.
I saw where Cutter had a Nichia 757 2000k 2 foot strip....Mix this in with some 4000k ?
 

Prawn Connery

Well-Known Member
^ That's the one I want, but they've sold out and I'm just waiting to hear back if they'll have more in time for my build. Helps that I'm also in Australia, so timing and shipping should be a bit more conducive, but I'm not entirely hopeful it will happen. The J Series Crees have been mentioned as a replacement for the Nichias due to them being sold out, which is why I came on here to ask about them.
 

Dave455

Well-Known Member
^ That's the one I want, but they've sold out and I'm just waiting to hear back if they'll have more in time for my build. Helps that I'm also in Australia, so timing and shipping should be a bit more conducive, but I'm not entirely hopeful it will happen. The J Series Crees have been mentioned as a replacement for the Nichias due to them being sold out, which is why I came on here to ask about them.
How would you blend the 2000k with the other strips ?
 

Prawn Connery

Well-Known Member
I wasn't planning to. But they make 3000K and 5000K strips as well if you wanted to blend them. I was considering suplementing them with UVA/B fluoros.

No I didn't.

I was talking about comparing leds of the same color which, if people actually use lumen, would be doing. Which just goes to show your rant is even more baseless.
Semantics aside (or not), your literal reply to a rhetorical quote proves you didn't understand the basis to begin with.

I'll keep it simple and ask again: why do board manufacturers use individual blue and red diodes to supplement deep blue and red spectra instead of the 1750K style diodes highlighted earlier?

And maybe you missed the other point: how can you compare lumen efficiency of same-coloured LEDs when each individual LED may use different spectral intensities to achieve the same overall colour (3000K for example) which can affect its lumen rating?

Screen Shot 2017-11-13 at 14.42.30.png Screen Shot 2017-11-13 at 14.42.55.png


Cree J Series 4000K 80 CRI top
Samsung LM561C 4000K 80 CRI bottom

Do those intensity graphs look the same to you? They don't look the same to me. The Samsung S6 bin puts out a 36 lumens at 25C and the J Series D3 bin puts out 30 lumens at 25C.

Which one is more efficient? Answer that first.

Now you tell me, if the Samsung has 80% relative intensity at 550nm, and Cree 65% relative intensity at 550nm, does it not stand to reason the greener diode (Samsung) will have a higher lumen rating than the blue/red diode?

Or is that baseless argument?
 
Last edited:

wietefras

Well-Known Member
The thing is, you said none of that in your rant. I was just a rant against "So much emphasis seems to be placed on luminous flux - lumens per watt" when it hadn't been brought up even a single time.

But if you want to turn it around now and ask if you can compare efficiency for comparable SPD's, then yes. Yes you can.

You can compare those two 4000K 80 CRI COB/SMDs perfectly well on lumen. The difference is really only a bit more blue on the Cree. In this case going by the SPD charts you would be 2 percent off (in favor of the Samsung), but it's from datasheets anyway. Cree mentions 7 or 8% deviation in their datasheets and then they bin on 10% ranges. So the actual COB/SMD you hold in your hand can be quite far off from the datasheet.

So if you want to be sure what the actuall efficiency is then your only alternative really is to buy both and have each COB/SMD tested in a sphere (Or hope that CobKits tests them at some point). Barring that you will just have to learn to live with say 10% uncertainty of what you are getting or comparing. That 2% in "SPD lumen to PAR difference" really is insignificant in all this.
 

Prawn Connery

Well-Known Member
The thing is, you said none of that in your rant.
The thing is, I mentioned all those points in my "rant".

Look, I get it. English isn't you native language - you don't do nuance - so you can be forgiven for trying to be clever.

But you should really learn to read those graphs - whether they're accurate or not.
 

Greengenes707

Well-Known Member
Simple answer children... what is the LER?? Neither of you will prove anything till you actually put numbers up and do some calculation. Eye balling one section of a relative graph is not calculations. Nor is the word "can"...use actual numbers and what is going on, not speculation and guesses. LER ÷ LPW= Radiant Efficiency will give you the neutral starting point for any calculation you want to make. Then you can go straight to photons is you wanted to get more photosynthesis specific, or try and show some SDP benefit.


EDIT:
On a side note...that data sheet does have some changes/improvements/edits on it over the preliminary data sheets cree gave me a months few. Was 29lm for D3 3000K 70cri...now is 30lm in the link in post 1. I should have samples within the week and will have some boards tested, and end this bitch fest.
 

Prawn Connery

Well-Known Member
Eye balling one section of a relative graph is not calculations.
Indeed. But even a back-of-the-envelope comparison shows the Cree puts out over 50% more 450nm than 550nm, while the Samsung puts out less than 20% more - that's a big enough difference that the precise figures don't matter in terms of the argument raised.
 

Greengenes707

Well-Known Member
Indeed. But even a back-of-the-envelope comparison shows the Cree puts out over 50% more 450nm than 550nm, while the Samsung puts out less than 20% more - that's a big enough difference that the precise figures don't matter in terms of the argument raised.
Photons, lumens per watt, and energy...come from the WHOLE AREA under the curve...not a single nm or tight group of nm's.
So again...do the whole calculations then compare. If you want to argue...do it right and put him in his place with non disputable figures...not you own thoughts that can be "argued against" Facts are facts.
 

Prawn Connery

Well-Known Member
We know that.

The problem with using lumens as a measurement is that they are based on the photopic curve, which ascribes 0 value to any wavelengths that fall outside that curve but which coincide with photosynthetic response.

I don't mean to be rude, but that is the whole point I'm trying to argue.
 

Prawn Connery

Well-Known Member
I also don't want to go around in circles, because I think most of us here already understand all this - that the most efficient LED for growing is the one that converts the most energy to PAR (depending on stage of growth) and not less useful radiation. So lumens that are weighted toward the human eye are not a true measure of efficiency in this case, even when comparing same-colour diodes (all factors taken into account).

It's just that someone who couldn't read properly decided I was having a rant - which, to be fair, I probably was - and now I appear to be arguing semantics.
 

Greengenes707

Well-Known Member
We know that.

The problem with using lumens as a measurement is that they are based on the photopic curve, which ascribes 0 value to any wavelengths that fall outside that curve but which coincide with photosynthetic response.

I don't mean to be rude, but that is the whole point I'm trying to argue.
You're not getting it. The circles ends with LER. Once you have it...there is no more going in circle like you keep doing. That is why actual calculations matter...so you actually can correct the numbers and not someones personal thoughts. Use the formula.
LER=luminous efficacy of radiation...aka what 1 radiant watt of energy from that SPD would produce in lumens. Since the LER represents 100% efficiency...you then take the lm/w and divide it by the LER...that gives you a percent figure of the radiometric efficiency.
LER will give you the staring point that lumens per watt came from in the first place. Lumens per watt(LPW) is not the starting measurement that a sphere or gonio measures. It takes radiant watts...and then applies the photooptic curve to radiant watts based on the SPD. So you go straight back to radiant watts from LPW. The same radiant watt figure is used to take the SPD and apply photon weighting(nm x planks constant)...instead of lumen weighting(photooptic curve)
 
Top