How to calculate PPFD for my build?

CobKits

CobKits

Well-Known Member
sixstring2112 said:
not to be a dick but im starting to wonder why we would take 1 vendors word on numbers from the products they sell (shitizen) lol.these cree numbers were ran by several well known members here many of whom had no ties to sales to cree.
Click to expand...
test em yourself.

i dont post good results for citizen just because i sell them

rather i sell them because they test well and are an excellent value

gen5 1818 matches or beats cxb3590 at all currents and really pulls away above 100W. dont believe me? ask these guys

@robincnn
@Stephenj37826
@Rahz
@RainDan

zero advantage to $35 crees vs citi/luminus/vero starting at half the price
 
CobKits

CobKits

Well-Known Member
sixstring2112 said:
cree 3590 can be had for around 31.00 these days,so how long would it take to cover the extra cost when your gaining 10% eff right away? the citi cobs look to be around the same eff as a cmh/lec setup is right now.40% and better spectrum from the lec is almost a no brainer.
Click to expand...
that simulator is good to compare chips of same manufacturer but not so much mfr vs mfr due to the way each mfr calculates their simulations

i mean we still have people who believe that crees can hit 56% efficiency at chip level in a real-world application when thats been demonstrated to not be the case in several tests
 
Stephenj37826

Stephenj37826

Well-Known Member
CobKits said:
that simulator is good to compare chips of same manufacturer but not so much mfr vs mfr due to the way each mfr calculates their simulations

i mean we still have people who believe that crees can hit 56% efficiency at chip level in a real-world application when thats been demonstrated to not be the case in several tests
Click to expand...
Actually the tools can be somewhat accurate. Cree doesn't include color temp like citizen. So the numbers the tool spits out is actually close to what has been Sphere tested. What most of us don't do is use a realistic Tj which in Crees case can severally impact performance. Samsung's tool thus far has been slightly under exaggerated(which I'm glad to see). The tools are a nice baseline but as always a sphere test at system level is best lol.
 
Stephenj37826

Stephenj37826

Well-Known Member
CobKits said:
can be accurate... but in that case the oft-cited cree 56% efficiency is at an unrealistic 25C Tj
Click to expand...
Also going by data sheets middle of the bin which is not realistic for 3500k Cree honestly. The old numbers where not accurate but moving forward I think we can get a better use with the tools now that we know how to accurately use them.
 
CobKits

CobKits

Well-Known Member
binning makes it fuzzy for sure.

wouldnt 3500k most likely have gone from being mid-upper CD bin to lower-mid DB bin?
 
G

George2324

Well-Known Member
The calculator states that if I run 1 citi 1818 3000k 80 Cri in 1 sq ft of space that my ppfd would be 1811. Which doesn't seem right.

Could anyone help me work out ppfd for what I need?

I'm wanting to get 1500 ppfd at canopy 12 inches away from bulb and penetrate 2 feet into canopy.

This is a sealed co2 grow.

1 cob per square foot total of 72 cobs.

I just can't decide on which type of cob and drive current
 
CobKits

CobKits

Well-Known Member
George2324 said:
I'm wanting to get 1500 ppfd at canopy 12 inches away from bulb and penetrate 2 feet into canopy.
Click to expand...
at one cob per square foot you'll have great penetration. to maximize the intensity 2 foot down at canopy make sure your lights are 24" above your canopy where youre getting 1500 ppfd, and the geometry will be in your favor
 
CobKits

CobKits

Well-Known Member
sixstring2112 said:
.i would love to see citi make a chip that does what cree does @ 36v so we can use our current drivers to run them.
Click to expand...
send me a DB 3500k and ill run it beside the 1212 gen 6.

should be within a few percent below 50W at well under half the price

if theres really a market for replacing cree chips (which doesnt make sense to me this early in their life) i can get luminus CLM22 which is a 36V chip with specs similar to the CXM22 (aka slightly better than CXB3590), for about $17. unfortunately those who are in with constant current drivers are somewhat limited in expansion, where with constant voltage drivers you could just double up the # of cobs on a driver (one cob model per driver to keep the voltage tight)
 
Last edited:
G

George2324

Well-Known Member
CobKits said:
at one cob per square foot you'll have great penetration. to maximize the intensity 2 foot down at canopy make sure your lights are 24" above your canopy where youre getting 1500 ppfd, and the geometry will be in your favor
Click to expand...
If I ran the cobs at 1.4a and had them 24 inch away would I get 1500 ppfd at the canopy though?

For that distance would I not need to run them harder?
 
CobKits

CobKits

Well-Known Member
George2324 said:
If I ran the cobs at 1.4a and had them 24 inch away would I get 1500 ppfd at the canopy though?

For that distance would I not need to run them harder?
Click to expand...
depends on the space, reflectivity, etc. photons dont disappear they just spread out. in a tent you have them hostage.
 
G

George2324

Well-Known Member
I don't run in a tent I have as much reflective as I can but I have a whole room no tent would be this big.

Wish there was a way for me to test these leds in detail before I pay.

I don't suppose anyone who has spare 1818s wants to do a par test with 1 led per square foot? :)
 
Joe34

Joe34

Active Member
@George2324 If you running 1 Cob per Sq Ft @ 40-50w I dont think your going to be dissapointed - Atleast from all the LED grow logs I have read. Although im in the same postion as you...
 
G

George2324

Well-Known Member
The LEDs will be 72w per sq foot If i use the 1818s. Just wanting to optimise as best as I can. I suppose I will just bite the bullet and order them. If i don't hit 1500 par I could add some single 3w diodes around the cob and hope it eventually adds up to 1500 par
 
You must log in or register to reply here.
Top