DHP Presents Citizen Cob 1812 3500k 80CRI Vs 1812 3500k 90 CRI

NoFucks2Give

Well-Known Member
I think what cobby ment is that each graph should be aligned that it peaks at 100% on the y-axis.
Actually this was done because the datasheets do that. The post with the 2700K and 3000K citi began over the datasheets being normalized and it skewed the interpretation. My app has the auto-scaling, I removed the 100% scaling just for this post.

The problem was the datasheets, by normalizing the graphs, made the 2700K appear to have more PAR flux than the 3000K. I already had some of the 2700K and bought the 3000K just to create the posted un-normalized graph. When I bough the citi I also bought the Vero 1750 which came up in the discussion.
 

CobKits

Well-Known Member
looks like vero and luxeon are almost identical

ill try to shoot citi orange cob and the horticultural cobs today if i can. if i send you raw data can you plot em with yours?
 
Last edited:

CobKits

Well-Known Member
The problem was the datasheets, by normalizing the graphs, made the 2700K appear to have more PAR flux than the 3000K.
well what is par flux anyway? anything measurement that cuts off sharply at 700 is tough to quantify

also if i recall your measurements were controlling current with the ldd300, but at different Fv were also talking different wattage from chip to chip
 

CobKits

Well-Known Member
I have seen a number of research papers that refer to white as inefficient, with unnecessary wavelengths. I do not agree with that but it is very clear deep red is the go to wavelength.
i think that interpretation of the mccree curve is what drove people to use blurples with 450, 630, and 660 monos

and it seems in that time weve seen many side by sides that fuller spectrum white lights do better. if you saw @Greengenes707 and @Growmau5 recent tech talk their consensus was yes deep red is the best for photosynthesis but the tech just wasnt there for efficient monos, and the new leaps in red tech make it more viable. but in any case they are still advocating for full spectrum white with augmented red, not straight monos.

after using blurples for many years (osram SSL for last few years, not crappy low-efficiency epistars), i cannot get the robust full growth i get under white cobs. better vigor, larger leaves, just better all around
 
Last edited:

CobKits

Well-Known Member
The Citi have been problematic in that I am always changing the LED driver so use a quick connect/disconnect jumper. That did not work well as the power contacts are breaking off. I received four Citi CLU028-1204C4-273H7K4 on May 11th, and two of the four have a broken contact.
well no manufacturer would recommend wires on a cob without support. when installed correctly they are extremely durable. its good that you can swing luxeons around by the wires but no cob is specifically designed for that
 

NoFucks2Give

Well-Known Member
. if i send you raw data can you plot em with yours?
Yes. All I need is wavelengths and values. It's a very simple program. The graph is an SVG image

that looks like this:

<line x1="368" y1="300" x2="368" y2="228" stroke="#c4ff00" opacity=".3"/>
<line x1="370" y1="300" x2="370" y2="226" stroke="#c8ff00" opacity=".3"/>
<line x1="372" y1="300" x2="372" y2="225" stroke="#ccff00" opacity=".3"/>
<line x1="374" y1="300" x2="374" y2="224" stroke="#cfff00" opacity=".3"/>
<line x1="376" y1="300" x2="376" y2="222" stroke="#d3ff00" opacity=".3"/>
<line x1="378" y1="300" x2="378" y2="221" stroke="#d6ff00" opacity=".3"/>

Where
scale = max(values)/280
Y2 is the value scaled to 280 px. y = 300 - (scale * values[wavelength] )
X1 and X2 = ((wavelength - 380) * 2)
Y1 is the baseline
stroke is the color

Once I see what your data looks like I can set you up with a web app.

well no manufacturer would recommend wires on a cob
I understand that, I just do not conform to the rest of the world. I like taking risks, yet am not very careful.
 

NoFucks2Give

Well-Known Member
osram SSL for last few years,
OSRAM has new SSL's out now. They are looking good.

My theory on monos is has to do with horrible thermal curve on red. It is very difficult to run them at 25°C. I doubt many run them below 100°C.

Now CoBs, are where the research and development budget goes. Deep Blue LED technology.

Where red loses 50% flux due to temperature, deep blue and white lose 10% max.

My focus has been on thermal management.

the mccree curve is what drove people to use blurples
I don't think McCree really supports BR that well. I don't know where they get the Photosynthesis Rate Curve, not McCree.

I took all the McCree numbers and graphed them. This is what I got:

mcCree.jpg

Red and Blue were likley pushed because they are the most efficient LEDs. Green is the least efficient. Who uses green besides traffic lights. Who needs efficient green when our eyes add 10x brightness?

There is more to it than BR. Research using the scientific method, takes way too long. Eventually they will find that elusive wavelength(s) between blue and red. Until then, white will have to do.

It may be yellow amber. Amber the worst thermal curve. Can easily lose 80-90% at 100°C. Although there is a Luxeon phosphor amber (i.e. very warm white). I would never use Luxeon Sun Plus purples. I do like the Color C Line which is the same technology as Sun Plus.

My approach is Red White and Blue with adjustable LED drivers.

What I see the CoB crowd needs is an improved way to power the CoBs. A single 36-48V CC supply with a load balancer that will accurately divide the current to each CoB as desired. Example, nine 1 Amp CoBs, a 10 Amp supply and a circuit board with nine 1 Amp outputs. That would be a cheap and easy product to make.

The chips on this board can handle 1.5 Amp each.


Schematic.jpg
 
Last edited:

The Dawg

Well-Known Member
Whats Up Cobbers And Cobits . First Let Me Say That I Will Have To Rerun This Test. The Reason Is With Defoliation Of The 90 Cri Plant. You See I Defoliated The 90 Cri At 42 Days Into Flowering And that Biotch Shut Down Which Is Why I Didn't Defoliate The 80 Cri. :hump: The 90 Cri Produced Exactly 179.7 grams And The 80 Cri Produced 213.2 For A Difference Of 33.9 Grams.:weed:


IMG_5716.JPG IMG_5717.JPG IMG_5718.JPG IMG_5719.JPG IMG_5720.JPG IMG_5721.JPG IMG_5722.JPG IMG_5723.JPG IMG_5724.JPG IMG_5713.JPG
 

wietefras

Well-Known Member
i think that interpretation of the mccree curve is what drove people to use blurples with 450, 630, and 660 monos
I don't think McCree really supports BR that well. I don't know where they get the Photosynthesis Rate Curve, not McCree.
It comes from looking only at the chlorophyll absorption spectrum.

Citizen keeps quoting chlorophyll absorption charts in their horticulture oriented documentation. Based on that chart they actually claim that plants do nothing with wavelengths between 510nm and 610nm.
 

nogod_

Well-Known Member
Which looks like better smoke?

Hard to find the goods after wading thru pages of lettuce growers regaling a potgrowing community with a whole lot of useless spectrum nitpicking!

Thanks for sharing dawg!

Whats Up Cobbers And Cobits . First Let Me Say That I Will Have To Rerun This Test. The Reason Is With Defoliation Of The 90 Cri Plant. You See I Defoliated The 90 Cri At 42 Days Into Flowering And that Biotch Shut Down Which Is Why I Didn't Defoliate The 80 Cri. :hump: The 90 Cri Produced Exactly 179.7 grams And The 80 Cri Produced 213.2 For A Difference Of 33.9 Grams.:weed:


View attachment 3946852 View attachment 3946853 View attachment 3946854 View attachment 3946855 View attachment 3946856 View attachment 3946857 View attachment 3946858 View attachment 3946859 View attachment 3946860 View attachment 3946861
 

NoFucks2Give

Well-Known Member
do you think these BXRC-27H1000-B-23 (2700k 97cri) would be worth it at about $5 a chip or save and get the luxeon red meat?
You should do fine with the Vero. Spreading out five CoBs is much better than one with 5x flux for uniformity and thermal management. With improved uniformity you can get closer to the canopy and the PPFD increases exponentially with less height.

How much current are you going to push through them? And how will you cool them?
 

BuddyColas

Well-Known Member
Whats Up Cobbers And Cobits . First Let Me Say That I Will Have To Rerun This Test. The Reason Is With Defoliation Of The 90 Cri Plant. You See I Defoliated The 90 Cri At 42 Days Into Flowering And that Biotch Shut Down Which Is Why I Didn't Defoliate The 80 Cri. :hump: The 90 Cri Produced Exactly 179.7 grams And The 80 Cri Produced 213.2 For A Difference Of 33.9 Grams.:weed:


View attachment 3946852 View attachment 3946853 View attachment 3946854 View attachment 3946855 View attachment 3946856 View attachment 3946857 View attachment 3946858 View attachment 3946859 View attachment 3946860 View attachment 3946861
So it was an upset victory by cri 80. Now the CRI 90 camp is demanding a rematch. Glad you're hosting the rematch! One take away thus far...don't be pitching your old CRI 80 LEDs just yet.:mrgreen:
 
Top