PPFD - What sort of performance for this setup and space?

salmonetin said:

...jejeje Wilson... I told you... ...time to shout my mouth and back to chozo...



saludos

Click to expand...

It sounded to good when you confirmed, I just had to double check it, don't get me wrong…
I am still looking at the numbers they are so beautiful… These Leds perform absurdly well. Obrigado

...no problem at all Vescovi...its normal... i understand your pov...



Saludos

ya bongo said:

my short math

starting with the Cree PCT tool
View attachment 3463093
4X CXA3590/ 72V/ 1.05A - Tj=50°C


P=79.67W
Eff.= 104.5lm/W
LER=~320lm/W

eff: 104.5/320=0.327
Prad: 79.667*0.327=26.02W
PPF: 4*26.02*4.65=484µmol/s

and for PPFD I need Distance, Beamangle and Spread of the Cobs.

Click to expand...

The AD bin is 5 or 6 bins lower than the top bins we are getting from CREE (CD and DB). Is that a high CRi 2700K you are looking at? Also those are minimum lm/W figures and Vero has got us in the habit of using typical figures (halfway between the bins).

Regarding the PCT and PDF disagreement, SDS has had Vero 29s and Cree CXA3070 top bins tested in a university lab and confirmed that they put out above their minimum figures listed in the PDF. The PCT also seems to understate the effect of current droop but it is easy to test for current droop using any luxmeter, which I have done and confirmed that the gains we get by running soft are as significant as the PDF indicates. So my conclusion is that the PDF specs are the more accurate.

...thanks a lot Supra... ...and thanks a lot SDS too... ok trust on actual pdfs... ok oido cocina...

...some spds are missing... on various things ... its incredible in 2015... but hey... im too old ...too poor... and some things never change from times or evolutionate at slowers velocities...

Saludos

SupraSPL said:

The AD bin is 5 or 6 bins lower than the top bins we are getting from CREE (CD and DB). Is that a high CRi 2700K you are looking at? Also those are minimum lm/W figures and Vero has got us in the habit of using typical figures (halfway between the bins).

Regarding the PCT and PDF disagreement, SDS has had Vero 29s and Cree CXA3070 top bins tested in a university lab and confirmed that they put out above their minimum figures listed in the PDF. The PCT also seems to understate the effect of current droop but it is easy to test for current droop using any luxmeter, which I have done and confirmed that the gains we get by running soft are as significant as the PDF indicates. So my conclusion is that the PDF specs are the more accurate.

Click to expand...

I'm more curious of WHY such a discrepancy. What is causing the cx line to show lower performce. Is it the thermal resistance figured in. Or something esle. It just doesn't make sense that the tool and data sheets are perfectly matched for other platforms, but not the cx line.

Very good question. The CXB data on the PCT says the AD bin at test current 1900mA Tj 85C = 8539 lumens yet the minimum lumens for that bin is 9000. So whatever caused the discrepancy with CXA is apparently still happening with the CXB.

In the PDF it says Cree maintains a +/-7% tolerance, maybe the PCT is incorrectly applying the 7% penalty to the minimum figures in effort to make sure people are designing for worst case scenario, when in actuality there is about 7% difference from bin to bin and I believe that is what the disclaimer in the PDF is referring to.

just checking the PCT-tool

from the PDF
Using the 72-V CXB3590 LED as an example, at steady-state operation of Tc = 25 °C, IF = 1400 mA, the relative luminous flux ratio is 120% in the chart below.
A CXB3590 LED that measures 11,000 lm during binning will deliver 13,200 lm (11,000 * 1.2) at steady-state operation of Tc = 25 °C, IF= 1400 mA.

the same nimbers in the PCT-tool gives 13106.6.lm
so the misleading between the PDF/PCT is about 0.7%. that´s nothing in my eyes

Very big difference between Tc (case temp) and Tj (junction temp), especially at that very high level of dissipation (89W). The PDF says 11000 lm at 1.2A with Tj 85C, the PCT says 10246 lm, almost exactly -7%

The problem is that other Cree single dies are represented correctly. The xbd at 85c in the pct will be dead on the data sheets bins figure. So why just the cx line? 7% is a whole bin...think about that. Or a 6% more efficient driver. 7% sounds small...but it's valuable and if there is no reason for the cx specific discrepancy...I would like the most correct possible. Plus when the manufacture is the one putting out this conflicting info...I don't think we can just disregard it...even if we are correct in the end.

Also you can flip between Tc and Tj and there is no difference for the tool/pct...at least with the cx line. Haven't played with single die Tc vs Tj.

That may be the answer (a 30C discrepancy at test current), the fact that the PCT is ignoring the difference between Tsp/Tc and Tj, if the PCT is showing us lumens based on Tsp/Tc, when we select 85C that is actually a Tj of ~115. So if we correct it by selecting 55C (actual Tj of 85C) the PCT returns a figure higher than the minimum lumens.

If so, another way to put it the PCT is based on Tc rather than Tj. You can see from the PDF that at 1.9A a Tc of 55C = Tj of 85C


-------------------------------------------------------------------------------------------------------------------------------------------

Looking at the XML2 which is also binned at Tj 85C, the PCT does show a correct figure for Tj 85C and reduces the figure by only 1% when you switch to Tsp 85C. That actually does make sense because at 700mA (2W) Tsp would not be much higher than Tj.

I shows a 4% reduction at 2A (6.3W)

SupraSPL said:

That may be the answer (a 30C discrepancy at test current), the fact that the PCT is ignoring the difference between Tsp/Tc and Tj, if the PCT is showing us lumens based on Tsp/Tc, when we select 85C that is actually a Tj of ~115. So if we correct it by selecting 55C (actual Tj of 85C) the PCT returns a figure higher than the minimum lumens.

If so, another way to put it the PCT is based on Tc rather than Tj. You can see from the PDF that at 1.9A a Tc of 55C = Tj of 85C


-------------------------------------------------------------------------------------------------------------------------------------------
View attachment 3463681

Click to expand...

Now that wasn't too hard to figure out eh. I have been curious about the Tj vs Tsp in the tool...it stuck out to me, but couldn't figure out why. Thanks for clearing up why. That is definitely it. Now we can use the tool with some solid confidence.

A bit off topic…

I am guessing I will need a lot of spread since I am using fewer COBs, I probably should do without lens and reflectors for a better light spread right?

In the two situations. 3.25x3.25 (8 COBs 600W) and 3x2 (4COBs 300W).

I have little head space… We are talking in height, 6 feet in the 600w cabinet. 3.8 feet @ the 300W double cabinet (7.6 feet total divided in two spaces vertically stacked).

Forgot to say everything is lined with white special reflective polyester. Walls, floor, the lot.

what about 8 @350ma in a 2x3?

coolj said:

what about 8 @350ma in a 2x3?

Click to expand...

Spares…

ya bongo said:

my short math

starting with the Cree PCT tool
View attachment 3463093
4X CXA3590/ 72V/ 1.05A - Tj=50°C


P=79.67W
Eff.= 104.5lm/W
LER=~320lm/W

eff: 104.5/320=0.327
Prad: 79.667*0.327=26.02W
PPF: 4*26.02*4.65=484µmol/s

and for PPFD I need Distance, Beamangle and Spread of the Cobs.

Click to expand...

...@ya bongo please need your help or opinion or thoughs...

...PPF: 4 (numbers of cobs) * 26.02 (PAR W or Prad) * 4.65 (but these conversion factor value its for w / m2 to umols*s*m2.....not?...maybe im wrong....) = 484 µmol/s ...measured at what distance?.....

....then from my pov... Prad are PAR W / m2 not PAR W only .........26.02 W / m2....????...

...on other hand when you said ....for PPFD I need Distance, Beamangle and Spread of the Cobs.....

...only one averaged ppfd at # Distance at ...None Reflective... ...Moderately Reflective... ...Highly Reflective....Walls...?... or some ppfd values at # Distance at ...None Reflective... ...Moderately Reflective... ...Highly Reflective....Walls...for zones?....

...what are the normal or tipical values for distance from cob on tipical test ... for example every 5 cm ...or 10 cm ...etc...??...

...what are the normal or tipical values for beamangle from cob on tipical test ... ?

...what are the normal or tipical values for Spread of the Cobs on tipical test ... ?.... ???????????????

i saw... other tool... for take ideas... we need similar tool for our cobs...
http://www.illumitex.com/files/ppfd-calc.php

Walls Surrounding The Area.. ...None Reflective... ...Moderately Reflective... ...Highly Reflective....
Units... meter or feet...
Target Area Length.... help on calcs umols for lengths?...?
Target Area Width.... help on calcs umols for widths?...?
Distance to Target Area.... meter or feet
# of LEDs Along Task Length...for my pov helps on calcs for ppfd length...?
# of LEDs Along Task Width...for my pov helps on calcs for ppfd width...?

....PPFD at # Distance? at ...None Reflective?... ...Moderately Reflective?... ...Highly Reflective?....Walls...

....bonus info,,, i saw the conversion factor on the table 1 page 8 ....for sunlight... but for cob led??......

http://www.licor.com/env/pdf/light/Rad_Meas.pdf

...i saw other info on pdf...

http://ag.arizona.edu/ceac/sites/ag.arizona.edu.ceac/files/Photmetric Radiometric and Quantum Light Units.pdf

...your thoughts... or others thoughts are welcomed too..

...thanks in advance...



Saludos

SupraSPL said:

3.27' X 3.27' (1m²)
75.5W X 8 = 604 Dissipation W
604 * 49.7% efficiency = 300 PAR W
300 * .9 reflector/lens/wall losses = 270 PAR W
270/10.7 = 25.23 PAR W/ft²
25.23 * 4.65 umol/s * 10.7 = 1256 PPFD

3.5' X 3.5'
75.5W X 8 = 604 Dissipation W
604 * 49.7% efficiency = 300 PAR W
300 * .9 reflector/lens/wall losses = 270 PAR W
270/12.25 = 22 PAR W/f²
22 * 4.65 umol/s * 10.7 = 1095 PPFD

Very high intensity, maybe too high in the hot spots? I have not tested above 1000 PPFD with COBs.

Click to expand...

Supra, what would you think of the idea of overhauling the pontoon on an inda-gro 420 so it housed 4 of these 3500K 3590's on each side of the pontoon? (8 total) For coverage of a 4x4