Fake Cree XML T6?

[midori]

Hi

I bought a Cree XML T6 (10W 3.3V 3A) from a Chinese ebay seller.
The problem is, I cannot get more that ~0.85A-1A of current on it.

I tried driving it with a buck-converter (3.2V), 3.3V PSU rail, and directly from a Li-Po battery (3.5A).

According to the CREE specs, 3.3V should be enough to get to 3A, but the max i got was 1.02A when directly hooked onto Li-Po battery.

In my knowledge PSU and Li-Po should act like a constant current source.

Can anyone advice me or confirm my suspicion about the fake CREE.

Kind regards.

[brted]

Please post a picture or at least get a magnifying lens and see if the LED looks like a real Cree XM-L. If it does, it is probably real because it would be hard to fake.

Also, are you able to measure high currents with your DMM? A lot of DMM's come with leads that have very thin wire. They work fine for voltage, but not so great for amps, especially higher amps. Also make sure you are using the jack 10A or 20A jack (most have DMM's have a different + jack for measuring currents).

Welcome to CPF!

[idleprocess]

Seems like you could confirm the LED Vf by using the +5V rail on a computer PSU with a ~10W resistor (0.75 Ohm ought suffice - should limit current to <3A). This will rule out the Vf limitations on the driver & 3.3V PSU rail and any current limitations on your Li-Po battery and account for the possibility that the Vf is closer to the 3.6V that older-generation white LED's used to see.

[SemiMan]

Vf on the spec sheet is up to 3.5V at 700mA. You just got a part at the higher end of the Vf in all likelihood.


Semiman

[midori]

Please post a picture or at least get a magnifying lens and see if the LED looks like a real Cree XM-L. If it does, it is probably real because it would be hard to fake.

Also, are you able to measure high currents with your DMM? A lot of DMM's come with leads that have very thin wire. They work fine for voltage, but not so great for amps, especially higher amps. Also make sure you are using the jack 10A or 20A jack (most have DMM's have a different + jack for measuring currents).

Welcome to CPF!

Hi! Thanks for the welcome!
I've measured ~4A of current with 30W led without any problems, so It's not DMM's fault.
I'll disassemble the build and take pictures as you've advised.

[midori]

Seems like you could confirm the LED Vf by using the +5V rail on a computer PSU with a ~10W resistor (0.75 Ohm ought suffice - should limit current to <3A). This will rule out the Vf limitations on the driver & 3.3V PSU rail and any current limitations on your Li-Po battery and account for the possibility that the Vf is closer to the 3.6V that older-generation white LED's used to see.

Hi! Thanks for your reply. I need to buy 10W resistor to further test your idea so I'll get back to you tomorrow

[midori]

Vf on the spec sheet is up to 3.5V at 700mA. You just got a part at the higher end of the Vf in all likelihood.


Semiman

Can you please explain further? Does that mean I should drive the led with higher voltage than noted in the specs sheet?
It's also worth noting I'm not using any driver for the LED, but just raw connection to the battery or SMPS PSU, should that matter?

Kind regards.

PS: I'm sorry for multi posting, I'll use multi quote from now on.

[mattheww50]

Vf on LED's 'wanders' from device to device and production run to production run, if the spec sheet it is up to 3.5 volts, it means just that, that the forward voltage required can be as high as 3.5Volts. In addition the resistance of the diode is non-linear. It is quite high until you get very close to Vf, and then falls to a very small value, so small changes in applied voltage result in large changes in current. your 3.2 and 3.3 volt sources may be below Vf for full power output.

As result direct drive for relatively high power LED's tends to be problematic. The direct consequence of that 'problem' is that most drive high power LED's with constant current drivers, i.e. the current is regulated, not the voltage. So the drivers provides whatever voltage is required to attain the desired current. Plan B is to use a higher voltage such as 5V and a resistor that is large enough to insure that voltage drop across the resistor at the maximum permissible drive current will keep the voltage across the LED below the lowest likely Vf. In this case 3 amps would result in a voltage drop of 3 x .75 of 2.25 Volates, leaving the 2.75 volts across the LED. This makes it impossible to exceed the maximum rating of the XM-L LED. The likely current in this configuration is about 2 amps.

[midori]

Vf on LED's 'wanders' from device to device and production run to production run, if the spec sheet it is up to 3.5 volts, it means just that, that the forward voltage required can be as high as 3.5Volts. In addition the resistance of the diode is non-linear. It is quite high until you get very close to Vf, and then falls to a very small value, so small changes in applied voltage result in large changes in current. your 3.2 and 3.3 volt sources may be below Vf for full power output.

As result direct drive for relatively high power LED's tends to be problematic. The direct consequence of that 'problem' is that most drive high power LED's with constant current drivers, i.e. the current is regulated, not the voltage. So the drivers provides whatever voltage is required to attain the desired current. Plan B is to use a higher voltage such as 5V and a resistor that is large enough to insure that voltage drop across the resistor at the maximum permissible drive current will keep the voltage across the LED below the lowest likely Vf. In this case 3 amps would result in a voltage drop of 3 x .75 of 2.25 Volates, leaving the 2.75 volts across the LED. This makes it impossible to exceed the maximum rating of the XM-L LED. The likely current in this configuration is about 2 amps.

Thank you! This was a really comprehensive explanation. I don't have a lot of experience driving LEDs so I wasn't aware of this input (in)flexibility.
I'll try playing with 5V input and ~0.5 ohm resistor. I really need that 3A output. I've also ordered a dedicated LED driver from DX, sku 57779. It *should* solve my LED current problem.

[midori]

Here's an update. I've managed to successfully drive the LED@3A with 3.85V of voltage.
I've hacked a MP2307 buck-module to give variable Vout and this is the result.
I find it quite weird that such a high voltage is required.

[SemiMan]

And perhaps your meter is not accurate either.

[flashlion]

And perhaps your meter is not accurate either.

I too think so.I had a similar problem with XP-G R5.I measured 1.2 Amp direct drive and I thought WOW how bright is R5.After a time I read that is very important the cables on the multimeter to be as short and thick as possible.I measured with short and thick wires and then the current was 2 Amp.After that I use 4AMC 7135 to drive XP-G LED.
It's worth it to try to use thicker wires with your multimeter.