CREE and LumiLEDs binning
- Thread starter NewBie
- Start date
Help Support Candle Power Flashlight Forum
NewBie
*Retired*
evan9162,
Here is your data plotted on the CREE datasheet:
Here is your data plotted on the CREE datasheet:
chimo
Flashlight Enthusiast
What mainly surprised me about the curve is the "concave down" shape as the current went up. :thinking:
Comparing the two XLamp datasheets (1W vs 3W), it looks like the 3W has a slightly flatter curve (delta V as a function of current seems a little higher on the 3W model).
I will plot my 3W when it arrives (Hotbeam is still waiting for the Optics from the GB :shrug: ) and add it to the collected data.
Paul
Comparing the two XLamp datasheets (1W vs 3W), it looks like the 3W has a slightly flatter curve (delta V as a function of current seems a little higher on the 3W model).
I will plot my 3W when it arrives (Hotbeam is still waiting for the Optics from the GB :shrug: ) and add it to the collected data.
Paul
HarryN
Flashlight Enthusiast
I do not have sufficiently good tools to measure Vf directly, but it definitely took a higher voltage to achieve 700ma in the Cree 3 watt than a TWOK.
My very simple setup - battery pack and resistors through the LEDs.
My very simple setup - battery pack and resistors through the LEDs.
When I get home tonight, I'm going to test my third, untouched XL3-7090 - I won't do any soldering - I will simply press it down onto an andonized (hopefully well insulated) heat sink with some thermal goo, and apply the probes directly to the power contacts. My current supply is clipped to the probes, and the probes themselves both apply power and measure voltage - this is how I've done nearly all of my Vf measurement, so I know it's a good method.
Okay, tested the last one. No soldering or anything - I simply slapped some thermal goo on the back and pressed it down onto an andonized aluminium heat sink (for electrical isolation), and touched the probes to the top contacts. It's Vf was even higher than the other two.
Here's the graph of all 3 vs. a Lux III:
Here's the graph of all 3 vs. a Lux III:
i just see this diagramm in the cree datasheet too. and first thought HEY the Cree led´s are better for direktdriven flashlights.
but
i think it´s more.
this U/i curves are also >typical<. I hat this word :-D
so there was/there is and there always will be only one solution driven led´s correct.
current must be fix in the range the led Vf can varie.
this new Lineartech two cell ic seems to be one of this @ 2,4 U(in) I (out) is pretty constant @ 350mA between 2,8 up to 4 Volt(led)...
if we believe the datasheets.
The interessting thing about THAT behaviour is that we cannot easily swap out a led with an other....especially from a different manufactur and high current multilevel flashlights...
For example VB-16 ;-)
Further question:
Would be interessting to see how the U/I curves cree and lumiled would look like at 5 and 40 Degree and cold condition too
@ evan and newbiew.
Are somebody of you able to do this too?
iám very interessted in such stuff.
especially if they change will they also change at normal room temp. 18-25 deggree.....
so far i never see a posting about a Lux1 vf bin with nearly 4 volt @ 350mA
but
i think it´s more.
this U/i curves are also >typical<. I hat this word :-D
so there was/there is and there always will be only one solution driven led´s correct.
current must be fix in the range the led Vf can varie.
this new Lineartech two cell ic seems to be one of this @ 2,4 U(in) I (out) is pretty constant @ 350mA between 2,8 up to 4 Volt(led)...
if we believe the datasheets.
The interessting thing about THAT behaviour is that we cannot easily swap out a led with an other....especially from a different manufactur and high current multilevel flashlights...
For example VB-16 ;-)
Further question:
Would be interessting to see how the U/I curves cree and lumiled would look like at 5 and 40 Degree and cold condition too
@ evan and newbiew.
Are somebody of you able to do this too?
iám very interessted in such stuff.
especially if they change will they also change at normal room temp. 18-25 deggree.....
so far i never see a posting about a Lux1 vf bin with nearly 4 volt @ 350mA
These are all very nice but when is someone gunna make a list real newbies can read
Like bright brighter brightest and the color and cost boom~
Its like Star Bucks, just give me a freakin large cup of coffee black not a latolato lata latey grande mocha
Feel like im doing a cross word puzzle everytime i try and figure out a bin code
But when everyone was finding decent flashlights i was still rubbing sticks together too
Like bright brighter brightest and the color and cost boom~
Its like Star Bucks, just give me a freakin large cup of coffee black not a latolato lata latey grande mocha
Feel like im doing a cross word puzzle everytime i try and figure out a bin code
But when everyone was finding decent flashlights i was still rubbing sticks together too
HarryN
Flashlight Enthusiast
Starbucks - that is a good analogy - and you have definitely entered "lots of options" territory on this forum.
Perhaps your needs are easier to meet than you might think (and I am serious, not trying to be difficult here at all).
- ~ $ 10 to spend - visit Target. They have some nice small LED lights
- ~ $ 50 to spend - Pelican or similar
- ~ $ 100 to spend - Surefire / Streamlight / Higher end Pelican
- ~ $ 250 to spend - Lots of options, including custom made
- ~ $ 1,000 to spend - collection of nice lights
You don't need to be an expert on bins codes and Vf to buy a nice LED light. You do need to be willing to learn them if you want to run on the extreme edge of performance at the best price.
Perhaps your needs are easier to meet than you might think (and I am serious, not trying to be difficult here at all).
- ~ $ 10 to spend - visit Target. They have some nice small LED lights
- ~ $ 50 to spend - Pelican or similar
- ~ $ 100 to spend - Surefire / Streamlight / Higher end Pelican
- ~ $ 250 to spend - Lots of options, including custom made
- ~ $ 1,000 to spend - collection of nice lights
You don't need to be an expert on bins codes and Vf to buy a nice LED light. You do need to be willing to learn them if you want to run on the extreme edge of performance at the best price.
NewBie
*Retired*
evan9162 said:Okay, tested the last one. No soldering or anything - I simply slapped some thermal goo on the back and pressed it down onto an andonized aluminium heat sink (for electrical isolation), and touched the probes to the top contacts. It's Vf was even higher than the other two.
Here's the graph of all 3 vs. a Lux III:
You do have something wierd going on in your setup, maybe it was heat.
I just purchased another batch of these, arrived today, this time the 3W parts, and they do not show what you keep mentioning.
Data:
Code:
mA Vf
1 2.512
2 2.538
5 2.581
10 2.628
20 2.679
30 2.719
40 2.752
50 2.782
60 2.808
70 2.833
80 2.858
90 2.881
100 2.903
150 3.009
200 3.103
250 3.193
300 3.279
350 3.367
400 3.453
450 3.539
500 3.623
550 3.714
600 3.801
650 3.904
700 4.02I did notice you plotted beyond 3W for the CREE part, which occurs at 700mA. You are exceeding the rating on the device above 700mA which is near the point were you saw something funny. Once you exceed the design limits, who knows whats going to happen to a part....
Last edited:
Newbie,
There's nothing about your data that conflicts with what I reported - they are showing exactly what I'm reporting. Your curve almost matches mine perfectly - Test yours at 1000mA, and you'll see it jump up to about 4.7V
Just match up your data to mine:
What part of the behavior of my XL3s is different from yours? Your XL3s have the same I/V curve as mine.
There's nothing about your data that conflicts with what I reported - they are showing exactly what I'm reporting. Your curve almost matches mine perfectly - Test yours at 1000mA, and you'll see it jump up to about 4.7V
Just match up your data to mine:
Code:
current Mine Yours
100 2.95 2.903
200 3.14 3.103
300 3.32 3.279
400 3.48 3.453
500 3.66 3.623
600 3.84 3.801
700 4.03 4.02
800 4.23 ?
900 4.49 ?
1000 4.79 ?What part of the behavior of my XL3s is different from yours? Your XL3s have the same I/V curve as mine.
NewBie
*Retired*
Interesting, so the 3W and 1W CREE parts do behave very differently.
Here is my data plotted on yours:
CREE 1W vs. the Luxeon III:
Here is my data plotted on yours:
CREE 1W vs. the Luxeon III:
Last edited:
In another thread, McGizmo had a chart with a bunch of different power LEDs, their Vf at different current levels, and output measured in his integrating sphere. It showed the XL3 scaling better than the 1W XL7090, but also showed the higher Vf. Granted, there was only one sample of each - we'd need more direct comparisons with the two Cree parts to be sure.
I wonder if whatever improvement Cree made to improve output scaling with current also caused the Vf to be higher - it seems strange to trade off a slight improvement in output scaling with a Vf that much higher - We're talking about an 0.5V increase in Vf at 700mA (~15% from the XL7090 to the XL3) for a little better scaling. Maybe it was an inevitable tradeoff due to some process change or materials change?
Anyone with one of each (1W & 3W) non-white Crees able to take close-up photos to see if theres any discernable physical difference between the two?
I wonder if whatever improvement Cree made to improve output scaling with current also caused the Vf to be higher - it seems strange to trade off a slight improvement in output scaling with a Vf that much higher - We're talking about an 0.5V increase in Vf at 700mA (~15% from the XL7090 to the XL3) for a little better scaling. Maybe it was an inevitable tradeoff due to some process change or materials change?
Anyone with one of each (1W & 3W) non-white Crees able to take close-up photos to see if theres any discernable physical difference between the two?
NewBie
*Retired*
balazer said:
Through normal channels.
Nothing special as far as normal CREE parts.
I'm really excited over the newer parts that are much more efficient and make alot more lumens that will be out soon. The nice thing about the more efficient parts is that they will produce less heat than the current ones, since the energy is going out as heat, instead of going into making heat.
Here is the scan of one of the labels:
Last edited:
NewBie
*Retired*
evan9162:
You know how folks have noticed the Vf drop over time on the LumiLEDs parts, such as yourself?
Well, here is one of the CREE parts with a curve from one month ago, then it was ran at 700mA 24 hours a day for one month, and you can see there is very little change in Vf on the CREE parts after all:
You know how folks have noticed the Vf drop over time on the LumiLEDs parts, such as yourself?
Well, here is one of the CREE parts with a curve from one month ago, then it was ran at 700mA 24 hours a day for one month, and you can see there is very little change in Vf on the CREE parts after all:
B@rt
Flashaholic
Nice work Jar, seems to me the Cree is going to be a serious competitor. 
NewBie
*Retired*
They certainly do have an awful lot going for them.
They seem to be going in the right direction.
LumiLEDs seems to have crippled the K2, not increasing efficiency much, and just improving how hard you can drive them.
Once the K2 makes it to market, then they will have SMD machine reflowable devices like CREE, OSRAM, and Nichia. Thats good for mass production volumes.
As soon as the new CREE part makes it to market, it looks like it will produce as much light as the LumiLEDs Luxeon K2, but use only 33% the power.
Thats bigtime!
They seem to be going in the right direction.
LumiLEDs seems to have crippled the K2, not increasing efficiency much, and just improving how hard you can drive them.
Once the K2 makes it to market, then they will have SMD machine reflowable devices like CREE, OSRAM, and Nichia. Thats good for mass production volumes.
As soon as the new CREE part makes it to market, it looks like it will produce as much light as the LumiLEDs Luxeon K2, but use only 33% the power.
Thats bigtime!
Hi All! Im just now joined to the forum. Sorry if my english isnt enough good, the reason only: im living in Hungary.
So i have a few question!
I planed to buy a dose of cree Xlamp 7090 white led , and i saw before at the bining and labeling pdf, the P 0001 bin was available, but now i check again this document /4 see how the lumens growth/but they say the highest available bin is the N 0004 for 1W whithe xlapm. Is it a devolution?:-(
Do anybody know a place where can i buy P bin 1W xlamp leds? The distance is does not matter, i have a friend in Cali! 1 month before i asked the ETG tech but they can give me only N bin:-(and i cant select within the N bins its mean maybe i will buy the lowest N-0001 bin:-(
Other question is, how can i fix it and wireing the 7090xlamp chip to the heatsink? I see the cree xlamp dont have a mounted on MetalPCB version like the luxeon star...or can i find this version somewhere?
What is the recomended optics for this led?I need a nearest angle beam around 5 degree at half angle.
Thx
So i have a few question!
I planed to buy a dose of cree Xlamp 7090 white led , and i saw before at the bining and labeling pdf, the P 0001 bin was available, but now i check again this document /4 see how the lumens growth
/but they say the highest available bin is the N 0004 for 1W whithe xlapm. Is it a devolution?:-(Do anybody know a place where can i buy P bin 1W xlamp leds? The distance is does not matter, i have a friend in Cali!
1 month before i asked the ETG tech but they can give me only N bin:-(and i cant select within the N bins its mean maybe i will buy the lowest N-0001 bin:-( Other question is, how can i fix it and wireing the 7090xlamp chip to the heatsink? I see the cree xlamp dont have a mounted on MetalPCB version like the luxeon star...or can i find this version somewhere?
What is the recomended optics for this led?I need a nearest angle beam around 5 degree at half angle.
Thx
