Introducing CobKits.com - specializing in DIY and Citizen COBs

Evil-Mobo

Well-Known Member
which part is unclear? im my test setup
at 1050 mA cree = ~335 umol @ 36W, 59.4% eff
1825 =~335 umol @ 34.3W, ~63% eff
1818 = ~335 umol @ 34.9W, ~61.2% eff
1212 = ~335 umol @ 37.7W ~ 56.2% eff
1212 = ~225 umol @27W, ~59.4% eff (but you need 20 of them vs 15 to match OG PAR)
What would the numbers look like for the 1812's @1050mah?
 

CobKits

Well-Known Member
a little less efficient than the 1818s, Which makes sense if you understand that an 1812 are 18 dies in series and 12 in parallel. the same current running thru an 1818 (18 in series with 18 in parallel) would be more efficient (same voltage per string, less current per die)

from citi tool (conservative but good for relative comparisons):

upload_2016-9-11_21-59-20.png
 

Evil-Mobo

Well-Known Member
a little less efficient than the 1818s, Which makes sense if you understand that an 1812 are 18 dies in series and 12 in parallel. the same current running thru an 1818 (18 in series with 18 in parallel) would be more efficient (same voltage per string, less current per die)

from citi tool (conservative but good for relative comparisons):

View attachment 3778675
TY for the prompt reply.
 

PhotonFUD

Well-Known Member
You nailed it, man. Couldn't have said it any better. 100% accurate.

Well considering his commercial offering (PLC) is no longer doing business it is understandable that he tries to defend his position. His videos speak volumes about how 'well' he grows. If that is what you are looking for, then by all means follow his path. That is the great thing about hobbies, you get to to pick what you want to do. What is right for you might not be for someone else.

In order to get empirical evidence, as has been asked, there has to enough time given to get meaningful data. And there is nothing wrong with asking for empirical evidence especially if you are a commercial operation. Both the research and hobby areas of horticulture provide the proving grounds for new technology.

So far Cobkits has been more than helpful with providing very useful data. That data is great for planning. Sure it might not be exactly real world, but it is within tolerable error and therefore provides a good guide.

I had done some research into using citizen prior to joining this forum and was looking for the cheapest available. Cobkits with the $13.25 1212s didn't seem realistic so I took an interest to these discussions. Glad I did or else i wouldn't know that there is an expected price drop happening.
 

CobKits

Well-Known Member
well its pretty constantly dropping. crees were north of $50 last year and are now upper 30s/low 40s. were assuming 3590s will drop in price when the new models come out but nobody knows when that is and cree just had a substantial drop in price last month

citis are stable but new ones due in january, hard to say what will happen with that

new veros are coming out at old vero pricing so theres no reason youd buy an old vero but they might blow out old stock at $15 or something
 

CobKits

Well-Known Member
but they do. every test ive done has citi beating cree down to 10W and below. Do you have any data that shows otherwise? theres this perception that cree is superior to the current offerings in some way but i have yet to see a single test of cree beating clu-058 and im not the only one testing them
here was a low-current test i did with more data points when i got my better power supply a month ago

upload_2016-9-12_0-6-33.png
 

PhotonFUD

Well-Known Member
well its pretty constantly dropping. crees were north of $50 last year and are now upper 30s/low 40s. were assuming 3590s will drop in price when the new models come out but nobody knows when that is and cree just had a substantial drop in price last month

citis are stable but new ones due in january, hard to say what will happen with that

new veros are coming out at old vero pricing so theres no reason youd buy an old vero but they might blow out old stock at $15 or something

It was more of a general statement. As in I wouldn't know if there was a price drop forthcoming if it wasn't for the discussions. I started with Digikey at ~24/unit, then yours at 13.25/unit and now you have a 10% discount code making them ~11/unit. For my perspective that is a price drop attributed to the forum.
 

GrowUrOwnDank

Well-Known Member
Cree fans are like apple's fans. Don't get along well with Vero and Citizen fans.
Cant wait for the Citizen 3500k 90 cri grow to finish
https://www.rollitup.org/t/northern-grow-lights-photon-180-clu1825-grow.915159/page-2#post-12949577

@sixstring2112 you need to quit the Cree cult and join Vero 29 Gen 7 or Citizen clan. That's whats going on your SST X :peace:
Their Called CCS Which Stands For Cree Cob Snobs And They Lack Imagination And They Are A Monkey See Monkey Do Type of Grower. They Are Also The Type Of Ppl Likely To Walk Off A Cliff While Taking A Selfie Just Saying If The Truth Hurts Then Don't Fool Yourself Cause Its Winking At You :hump:
Damn this is a tough crowd lol.
 

weed-whacker

Well-Known Member
That chart and @CobKits testing have almost nothing to do with one another, both are "correct".

Although I am unclear how @CobKits is getting percentages out of spot PPFD measurements, unless it's just relative to Cree. Big-assed grain of salt if that is true.

yeah that's what has me suspicious too, I am comparing using 6 of the 12 12 vs 3 of the cxb3590 both at 150w total

your data suggests efficiency of about 50% for the 1212 @ 25w

and @CobKits spread suggests around 59%

vs cxb @ about 56%
 

CobKits

Well-Known Member
Although I am unclear how @CobKits is getting percentages out of spot PPFD measurements, unless it's just relative to Cree. Big-assed grain of salt if that is true.
it is 100% relative to cree which is why i always try to mention it as an "approximate, estimated, or relative" efficiency. There was a discussion recently where someone said that cree PCT was actually accurate to "the spreadsheet" within a fraction of a percent. In reality either cree is overreporting lm/W or citi is underreporting lm/W as their data doesnt jive. people are obsessed with their efficiency numbers but in reality if cob A is consistently brighter than cob B at a given wattage, are you going to use the real-world testing data or what the manufacturer told you?

Those mfr-provided tools are great for comparing various intra-manufacturer models and setups but clearly there is a bias when compared with other manufacturer's data
 
Last edited:

CobKits

Well-Known Member
So far Cobkits has been more than helpful with providing very useful data. That data is great for planning. Sure it might not be exactly real world, but it is within tolerable error and therefore provides a good guide.
the only thing i could do to make it more 'real world' is put a plant under the sensor, its an actual light output test of an actual chip on a real heatsink. open to ideas on how to improve the methodology. not perfect but for example i was able to run 5 different getian models and pick the one or two i might be interested in stocking which is 100% based on their performance and cost relative to the others.
 

CobKits

Well-Known Member
yeah that's what has me suspicious too, I am comparing using 6 of the 12 12 vs 3 of the cxb3590 both at 150w total

your data suggests efficiency of about 50% for the 1212 @ 25w

and @CobKits spread suggests around 59%

vs cxb @ about 56%
PCT says for CXB DB bin at 1.4A (48.77W) Tc 25C you will get 9094.6 lm x 3 = 27283 lm @ 186.5 lm/W
citi tool says for 1212 4000k 80 cri at Tc 25C, 700mA (24.2W) you will get 4208 lm x 6 = 25248 lm@ 174 lm/W

my spot tests say that cree@ 1.4A = 435 umol/49.35W = 8.815 umol/W
1212@ 0.7A = 226 umol/23.89W = 9.460 umol/W

mfr data has CXB@7% more efficient under those conditions
real world testing has 1212 @ 7% more efficient under those conditions

i cant call it a fluke when citi beats cree at 3000K, 4000K, 5000K.... how many more chips should i test? heres an old 3000K test. both are 80 cri here. again lets look at PCT vs citi tool

PCT:CB bin: 1.4A: 8395 lm/W @ 48.77W = 172.1 lm/W
citi tool:3000k 80 cri: 4082 lm @ 24.2W = 169 lm/W
advantage: cree by 1.8%

my measurements
CB bin@ 1.4A = 386 umol @ 47.88W = 8.06 umol/W
1212 3000k 80 cri @ 0.7A = 206.6 @ 23.73W = 8.71 umol/W
advantage: citi by 8.1%

upload_2016-9-12_3-11-58.png
 

CobKits

Well-Known Member
i could also calc out bridgelux vs cree or citi using their tool vs online measurements etc but as you can see number crunching gets very intensive. id rather just take the measurements i have and present them with that big ass grain of salt....ill provide raw data to anybody who wants to do whatever they wish with it
 

CobKits

Well-Known Member
its been discussed. every one of these chips has a nearly identical polar radiation pattern per datasheet. i dont have time to do 25 flat-plane measurements per chip (@Malocan has with similar results btw)
 

PhotonFUD

Well-Known Member
2700 and 3000 I have more interest in for the red. And if you decide to stock those red strips from cutter or have something similar, I would be interested in the data from those as well.
 

Rocket Soul

Well-Known Member
Hey Cobby. In another thread u posted tests of a few Getian led cobs, gt3838, ct2828 and gt2828. What where the forward voltage on these units cause they look very interesting pricewise? And what was ur general impresion? Will either of these chips fit with easyto source holders and optics?
 
Top