Cree white R2 bin forward voltage

HumanLumen

Newly Enlightened
Joined
Jul 9, 2007
Messages
89
Location
Reading, Berkshire UK
Anyone been able to assess the behaviour of these items in terms of forward voltage? I would imagine that they are similar to the Q4/5 bins (i.e. same process, cream of the crop devices).

Cheers
HL
 
R2's have the same forward voltage distribution as all other XR-E whites.
 
R2's have the same forward voltage distribution as all other XR-E whites.

It was noted by some forum members conducting extensive tests
that Q5s seemed to have significantly higher Vfs than their predecessors.

This was observed across quite a wide sample and wasn't just a one off.

See this thread and post #13

http://candlepowerforums.com/vb/showthread.php?t=178378

So the above comment would appear to be a sweeping generalisation.
 
Last edited:
The LEDs are binned strictly on brightness at 350mA. They come off of the same line after going through the same process. Cree does not alter their process to intentionally raise the Vf of the die in order to fudge brightness data. The first production R2 parts may happen to have a higher average Vf than dimmer bins, but one should not expect that to be the case as time goes on. When designing a circuit for any LED, one should take into account the full range of possible Vf's. The maximum and minimum possible Vf for R2 is no different than previous bins, so one can expect lm/W values indicated by the entire range.
 
The LEDs are binned strictly on brightness at 350mA. They come off of the same line after going through the same process. Cree does not alter their process to intentionally raise the Vf of the die in order to fudge brightness data. The first production R2 parts may happen to have a higher average Vf than dimmer bins, but one should not expect that to be the case as time goes on. When designing a circuit for any LED, one should take into account the full range of possible Vf's. The maximum and minimum possible Vf for R2 is no different than previous bins, so one can expect lm/W values indicated by the entire range.

Seeing is believing.

Many saw the Q5s had abnormally high Vfs.

The only way to be sure about anything is to measure it.

In view of the findings on CPF of the higher Vfs of the Q5s, its a perfectly
justifiable question of the OP to ask whether this may be the case for
the R2s. The answer is not a foregone conclusion.
 
The Vf spec for the XR-E is measured as the forward voltage when 350mA of current is flowing through the device. Any XR-E that you buy could have a forward voltage anywhere from 2.9V or so to 3.9V. Cree does not release their percentage distribution, nor will they bin for forward voltage. If people are independently measuring the Vf for the R2 bin, or any other bin, and finding that on average, one bin has a higher Vf than another, that may be the current case. That may continue to be the case, or the average Vf could migrate over time. Also, it is possible that the parts tested all happened to be higher than average.

I always recommend that people assume worst case forward voltage with all LEDs they buy, regardless of manufacturer or bin, because there is always a chance you'll get the highest Vf.
 
Thanks Steve,.....and this question is to everyone... is there any way of achieving this simply and quickly? :thinking:
..... like if you wanted to measure the resistance of something you would simply measure it with a multimeter!! .....
feel free to jump in here, if you measure your forward voltage on a regular basis lol :)
 
Last edited:
Oh, you certainly can measure the Vf of the LEDs that you have in hand. Just get a constant current driver at 350mA, verify the current output with an ammeter connected in series with your LED, and then measure the voltage by putting a voltmeter in parallel with the LED.

But really, when you're concerned about the Vf, it is often important to know what you're going to get BEFORE you've bought it. For that, you can really only look at the datasheet and plan for the entire range possible.
 
Scenario: Cree, Lumileds, Seoul, etc. releases a new higher brightness LED. All the typical hobbyist vendors, i.e. DealX, etc. rush and get parts as soon as they can so they can make the most money on the new product release (which I do not blame them for). At the same time others are rushing through OEM channels to get their first samples...... people do tests, then results end up being published here.

What may seem like a wide distribution, i.e. 10 people on here doing tests, may represent a very small portion of an overall production run.....which could be 10's of millions of parts over a year. Suppliers will be expected to have accurate numbers in their data sheets. Their large customers by virtue of their production testing processes can probably get a good idea of what is being turned out and will expect honesty in the data sheet (I certainly do!)

So that said, we probably do not have nearly enough data to say that Q5s or R2s have a higher Vf. It is slightly possible the Vf is higher, as the XRE data sheet Vf spec typically should represent all XREs made, and not just the best bins which are likely not the highest proportion of the bins shipping. However, I have a hard time seeing Cree tuning their process for higher Lumen/mA at the expense of efficiency. That would be a marketing ploy that would quickly catch up to them.

Of course what is possible is that some large customers are getting the best parts and we are seeing the rest.

Semiman
 
Recommended max mA is 1000, thats 3.7 V according to this graph that I found at their page.
q5.jpg
 
Last edited:
Ahhh... fascinating.

At around the 3.7v mark, full direct drive from 3 cells or a rechargeable lithium becomes entirely feasible.

Excellent.
 
I recently purchased and received two R2 bin LEDs. They have a forward voltage of about 3.80V and 3.56V at a current of 905mA. This is measured shortly after turn on, and they are well heatsunk, so these voltages are a little higher than if they were hot.

I'm an electrical engineer, and I work in the semiconductor industry. That said, I don't make LEDs, nor do I work for one of the LED manufacturers. Keep that in mind when reading the following...


As I understand it Indium Gallium Nitride (the semiconductor material of choice used to make todays blue LEDs) is not especially compatible with the dopants used to make P type and N type semiconductor material.

When pure, semiconductor materials (such as silicon, or indium gallium nitride) do not conduct electricity. When they are "doped", they can be made to conduct electricity. The stronger the doping the better they conduct, and the lower the resistivity of the resulting material. The doping process involves forcefully implanting, or diffusing at high temperature, special types of impurity atoms into the semiconductor crystal lattice structure. Phosphorous is usually used as a "P" type dopant, and there are a number of "N" type dopants used, such as boron.

The problem with Indium Gallium Nitride is that when you introduce impurity atoms(the dopants) into it, it distrupts the perfect crystal structure of the semiconductor, thereby creating imperfections in the material. These local sites of imperfection are potentially regions where dark recombination can occur in the LED. In other words, you can get current flow through this region of the junction, but no photons are emitted. The heavier you dope the material, the more imperfections in the crystal, and therefore the "current efficiency" of the LED decreases.

On the other hand, if you dope the LED heavily, you make the semiconductor conduct electricity better, and therefore you end up with an LED that has a lower forward voltage at a given current than a lighter doped LED. If you dope the LED heavily, you get an improvement in the "voltage efficiency", but at the expense of the "current efficiency." It is a constant trade off. All LED manufacturers want to make the most efficient LED overall, so they have to find the best possible doping level for their particular manufacturing process and technology.

In all probability, CREE decreased the average doping level on the Q5 bin LEDs, relative to the P4 or other lower flux, lower voltage devices. This decrease in doping reduced the number of imperfections in the die, thereby improving optical output, but at the expense of increasing forward voltage somewhat. Overall however, the Q5 is most certainly a superior product and produces more light from a given package and at higher overall efficiency. My recommendation is to be happy with the higher flux bins. They are, overall, better products than the lower bin devices.


These R2 bin LEDs are really awesome! Extremely bright, very significantly better than the P4. I can't wait for the R4 and beyond.
 
Am I missing something or is the R2 not mentioned in the XR-E spec and binning sheets yet?
 
Am I missing something or is the R2 not mentioned in the XR-E spec and binning sheets yet?

R2 is in limited release ie Cree are not yielding sufficient qty to be able to offer them in the same volume as say Q5. Cree are very conservative with respect to release information as I am sure you are aware, the upside is that when they talk about a product being available, it is immediately available unlike other vendors who press release their products to death... often literally
 
Top