Could there be something screwy with the cree numbers that throws this off? If youre normalizing to cree values? This math is a bit beyond me.heres if i apply a factor to the 36V data to make it match the measured data:
View attachment 3780379
not a bad fit...... but if i apply that same factor to the 50 and 72V it looks like
View attachment 3780380
If its another 2 months until vero C hits kingbrite i might go straight to getian to get the 3838 and a few more low K of your light engine to balance out the getian coming in at 4000kmy money is that the 72V vero performs close to the 1825. but the fact that the 1825 tops out at 25% higher max wattage makes me think it will still have an advantage at high currents. but we'll soon know. jerry made it sound like 36V 4000K first, 3000 and 5000K end of september. no word on higher voltages
i used a factor of 6.4 to convert my ppfd/W to % eff, based on DB cree 60% efficient at 35W - an approximation but all chips treated equally. temp droop jsut flattens or steepens the curves but the relative relationships are the sameCould there be something screwy with the cree numbers that throws this off? If youre normalizing to cree values? This math is a bit beyond me.
Vero c (Tc25) from datasheet gives 161 lums/watt (aprox 50%efficiency)@ nominal 118watts. Eyeballing those cree numbers and assuming that the efficiency calculations (56% at 50w about right?) from datasheets are right i get ppfd: 8/w should be about 50% efficient. Vero c 29 seems to come in above that at nominal, but im not sure about tha tmath. Could the cree numbers simply not match the datasheet and thats throwing things off? Omg im about to feel the wrath of the cree klan?
i think it was 21 days or so from paying them to getting samples, they say 7 day lead time on 100 pc orders so 14+ days for me to get them.If its another 2 months until vero C hits kingbrite i might go straight to getian to get the 3838 and a few more low K of your light engine to balance out the getian coming in at 4000k
@CobKits how long did they take for arriving from getian?
V6 and V7 at up to 3A is below. i wouldnt run the V6 to 4.2A but the V7 would do better and would be about as efficient as a 3590 is when run at 3A
View attachment 3768029
Ive got a friend who works for an international stage light company that promised to look into samples for me. Will hit him up for next gen citis for 2017q1i think it was 21 days or so from paying them to getting samples, they say 7 day lead time on 100 pc orders so 14+ days for me to get them.
again they wont do anything but 4000 or 5000k unless you buy 100 pc and pay extra
1825s are a few more bucks but i have those in all colors ready to ship now.. and they are 50V as opposed to 100V and outperform the getians
its hard to get everything you want.
getian = lower price and quick lead time but high minimum order + cost of importing
citizen = better chips, they import for me, can order as few as 20, but takes 4 weeks....
im chip agnostic. i hope the new veros kill it and are cheap, we all win. i dont have that many more citis and am playing it cool with ordering too many more... chip market is too dynamic
hard to believe even better citis in 2017Q1. hopefully i get some samples in 2016Q4
would be cool if they dropped the horticultural chips on us that @rmr2 was talking about
nice. different industry angle but photons be photonsIve got a friend who works for an international stage light company that promised to look into samples for me. Will hit him up for next gen citis for 2017q1
i mean i hope im wrong! but not getting my hopes up. based on the 324 LER of last gen vero at 4000K, if the new simulator was right this $27 72V chip would look something like
W efficiency
10 100%
15 85%
20 77%
30 69%
40 63.6%
low current data def screwy. i mean id love 75% efficiency @ $1/W but not holding my breath
Cobkits why is the citizen spreadsheet giving such low effi% numbers ? I don't get It, your numbers are like 15-20lm/w higher then what Citizen is saying. I know real world data is different but still we are talking here about 10%+ difference and that is huge.
If I read the citizen pdf sheet It says there can be a 10% difference between a chip maybe you this explains it ? And you have the golden nugget pheno of the cobs haha in your tests.
Yeah, 70% eff would be 230lm/W... thats just impossible!look at the getian and the 1212 in the efficiency curve in the other test, theyre way more cost effective out past 70% and pull away
well by datasheet. the 1212s are pushing 70% at 100 mA. but measurements that low are subject to error. 5-10 umol of stray ambient lighting can affect your measurements by +/-10% at those extreme low levels. need to test multiple chips at once and average at low currents like that. for now, take any measurements under 10W with a grain of saltYeah, 70% eff would be 230lm/W... thats just impossible!
A CLU048-1212 at 100mA and 25°C (3,2W Pd!!) does 192lm/W.... thats not even 60% mate.