Testing Timber's prototype=4000k/660nm and my finger on the flower trigger:)

PSUAGRO.

PSUAGRO.

Well-Known Member
The Dawg said:
Hey P Little Birdy Tells Me Theirs Talk Of Going To The Superbowl In Detroit :P
Click to expand...
ttystikk said:
Talk is cheap in October lol
Click to expand...
sixstring2112 said:
cant go to the superbowl if ya cant even beat the lions lol.i will admit that young qb doesnt make nearly as many mistakes as stafford does.wish we woulda drafted the kid .
Click to expand...
Annnnnnnnnnnnnnnnnnnnnnd back to reality, good game..........this is why I stopped betting, except when the patriots play the browns, easy $$$$:P


took a day, so it's update time

Day 9 of 14/10 from seed under timber's cxb 3590 4000k/osram 660/730nm trigger(322w)==== with my 2 1/2 god/bluediesel girls

DSC06438.JPG DSC06439.JPG DSC06440.JPG DSC06441.JPG DSC06442.JPG .......meh..... and I have a straggler, don't know wtf her problem is, but if she continues this way I'm culling the bitch:fire:..A wise old dawg once said " if you don't have your notebook, you have nothing",lol. Same side of the container again, strange.

DSC06443.JPG DSC06444.JPG ...........chilly,warm,chilly,warm............make up your fucking mind FALL


@Greengenes707 where are you at buddy?



So all the cxb's have been sphere tested and will get the specifics today..........can't talk about it until approval, but I can say that Walt Disney would be proud of our flat pane/avg #s:mrgreen:


be safe and happy growing RIU
 
Last edited:
PSUAGRO.

PSUAGRO.

Well-Known Member
Alright got the approval from both parties......i would like to thank Dan at timber for the gear and Nick for the sphere work! They did an awesome thing for us at RIU , finally getting some solid info.


Annnnnnd back to reality........ oh shit , I already used that line:)

"See the attached files for raw data from the sphere. The spectral data is the raw watt output at each nanometer and the other file shows the power consumption and lumen output. The Lamp P(W) in column X is the power at the wall taken from a UL approved and calibrated power supply, it generally reads a little high as it shows the worst case scenario. We started the flat plane testing, but with the frame you provided they were a little close to each other and didn't have a good spread. I didn't want to take a bunch of data points that were skewed due to position of the COBs.

Another thing to note is that the reflectors lower your lumen and PPF readings due to the loss of energy that occurs when the light bounces off the reflector, you can think of this as friction. In this case you are losing 8.2% of your total output by using the reflectors.

If you want to know total efficiency you can sum the entire Spectrum (W/nm) column in the spectral data sheets and divide by the fixture power number, this shows you watts of light energy divided by input wattage.

Your PPF numbers are as follows:
660nm 143 PPF 1.5 PPF/W 95 W 0.27 W/W
3000K 402 PPF 1.89 PPF/W 213 W 0.40 W/W
3 w/ref 369 PPF 1.73 PPF/W 213 W 0.37 W/W
4000K 410 PPF 1.91 PPF/W 214 W 0.42 W/W

My PPF numbers are:
Mega 1401 PPF 2.15 PPF/W 650 W 0.46 W/W
Veg8 365 PPF 1.92 PPF/W 190 W 0.40 W/W
Mini 255 PPF 1.70 PPF/W 150 W 0.38 W/W

Numbers that I have pulled from various sources around the web from verified 3rd party testing:
Power (W) Flux (lm) Radiant Flux (W) Efficacy (lm/W) Radiant efficacy (W/W)
SK400+ 90* 481.5 40280 125.2 83.66 0.26
SK400+ 120* 466.1 40360 127.4 86.59 0.27
SK600+ 645.1 74710 231.4 115.81 0.36
Illumitex Neo Sol DS 411.1 9018 90.58 21.94 0.22
Kind K5 XL750 485.3 10070 94.15 20.75 0.19
T5HO 8 tube 456.8 24000 60.72 52.54 0.133
Timber 660nm 4 94.78 1257 25.9 13.26 0.273
Timber 3000K 4 212.7 27824 86.19 130.81 0.405
Timber 3000K 4 Ref 212.8 25463 79.11 119.66 0.372
Timber 4000K 214.4 29084 89.8 135.65 0.419
Mega 678.6 95459 309.3 140.67 0.46
Veg8 193.2 25176 76.64 130.31 0.40
Mini 150.6 19829 56.7 131.67 0.38

I did not test the 3500K because it will basically be dead between the 3000K and 4000K. If you want to know what you lose with the reflectors on the 4000K just subtract 8.2% from those numbers as the loss will remain the same. You will see in the 660 data that we took three measurements to see how the output dropped as the fixture warmed up, we then let the 3000K and 4000K units warm up the same amount of time as the lowest numbers from the 660 data.

The pin sinks work pretty well for passive cooling and the CREE COBs are really efficient, but if you want to crack the 2.0 PPF/W range you are going to have to go with more COBs at a lower drive current or active cooling. We hit our numbers by spreading 100 watts out over a 12"x12" board, this is the major advantage of doing custom boards. Rest assured that your numbers are light years ahead of our friends at Kind and Illumitex. I didn't test the Osram thing because we have one here and they are a terrible configuration."

Using timber's 200w 4cob kits, so if you want the specs used (meanwell driver,pinsinks, etc) check the website...... it's a common/popular mah, so thats why it was picked

"The first thing I learned is those online calculators from the LED manufacturers don't work."............. cough 2.8umol/j cough, somebody is going to have a rough day when he gets the sphere results back...... lol


"The Zelion thing is using plastic drivers and those optics will knock off a serious amount of the light output. We tested the 2x2 here without the optics and they weren't any better than the Mini, but the Mini coverage was much much better.
"............ kinda agree with this, but he's an osram hater........ still just wanted a number , can't bitch about volunteer work though! ha ha

Anyways I'll get all the files uploaded when I get near a computer again...... phone sucks

Time for some wine
 
Last edited:
CobKits

CobKits

Well-Known Member
PSUAGRO. said:
The Lamp P(W) in column X is the power at the wall taken from a UL approved and calibrated power supply,
Click to expand...
i dont understand this. is this an AC supply feeding a whole timber kit with the driver? so its basically a precision version of a watts up?

or is it a regulated DC power supply to feed the cobs directly?

the latter obviously has issues in regard to overall system efficiency.
 
Last edited:
CobKits

CobKits

Well-Known Member
apart from that the numbers are near what we would expect:

-the popular 3500k at 1400 ma is claimed to be 56% efficient
-so the 3000K CB is say 92% of this or 51.5%
-assuming they used 110V into the meanwell that is 93% efficient so now youre at 47.9%
-they claim the optics cost 8.2% this sounds like a ballpark but lets call it 92% efficient. now youre at 44%
-theyre testing 40% which i could see. This test is as much of a referendum on the driver/heatsink setup as it is on the chips itself. wed need to test another mfrs chips on the exact same setup to confirm the chip output relative to other models and mfrs

in any case 10% is within the range of bin flux/temperature droop as well. cree temp ratings always confuse me between the incongruent PCT and supra data

Another thing to consider is the entire effect of the system. you basically put 4 large high-surface area black objects in a sphere- its not out of line to think that if white reflectors alone are absorbing 8.2% that these could absorb a solid 10%
 
Last edited:
PSUAGRO.

PSUAGRO.

Well-Known Member
sixstring2112 said:
im really confused so ill just wait for the dust to settle haha.seems like it tested higher than most ?
Click to expand...
Yes, cree cobs are good..............just not magic. 1.9ppf/w bare(no reflector/lens) is efficient

his Samsung chips in the mega are avg 140lm/w at system level, which is very good

gavita pro 1000w de did 1.5ppf/w in his sphere, slightly used(less than dozen fires)

CobKits said:
i dont understand this. is this an AC supply feeding a whole timber kit with the driver? so its basically a precision version of a watts up?

or is it a regulated DC power supply to feed the cobs directly?

the latter obviously has issues in regard to overall system efficiency.
Click to expand...
research how a sphere works

CobKits said:
apart from that the numbers are near what we would expect:

-the popular 3500k at 1400 ma is claimed to be 56% efficient
-so the 3000K CB is say 92% of this or 51.5%
-assuming they used 110V into the meanwell that is 93% efficient so now youre at 47.9%
-they claim the optics cost 8.2% this sounds like a ballpark but lets call it 92% efficient. now youre at 44%
-theyre testing 40% which i could see. This test is as much of a referendum on the driver/heatsink setup as it is on the chips itself. wed need to test another mfrs chips on the exact same setup to confirm the chip output relative to other models and mfrs

in any case 10% is within the range of bin flux/temperature droop as well. cree temp ratings always confuse me between the incongruent PCT and supra data

Another thing to consider is the entire effect of the system. you basically put 4 large high-surface area black objects in a sphere- its not out of line to think that if white reflectors alone are absorbing 8.2% that these could absorb a solid 10%
Click to expand...
Fine, you can pay for the next round of tests!lol...................it's a baseline, take from it what you will


The small framing is my fault and wanted to save shipping costs for Dan, didn't think about the flat pane #s...........or cared about them either:)
 
You must log in or register to reply here.
Top