I bought CREE xhp70.3 MCPCB from KD, installed it on a host and at 11,9v(at the LED!) and it draws only 0.88A instead of about 2.3A according to the datasheet
Before even looking at the power supply, is it even possible for a genuine xhp70.3 to be at 11.9v-12v and not draw 2.3a~ ? with a decent heatsink and wiring
my 12v dc can do 3A, and at 0.88A it didn’t even get warm
At least compare with original led pic
can original cree run at 12v but draw only .88A ?
We don’t know your setup and how you do measuring. With proper supply, wires original led should act according datasheet
If the dc supply can push only 1A at 12V, wouldn’t the xhp70.3 LED voltage drop drastically below 12v?
Hi drakula romanov,
That datasheet chart shows the electrical characteristics at 85C junction temperature. The forward voltage is higher at lower temperatures such as 25C.
There used to be a tool called the “Cree Product Characterization Tool” or PCT which would characterize an LED in greater detail, but Cree seems to have taken it down.
I performed a quick test of a 6V xhp70.3 Hi 4000k 90CRI (from KD) for you. I did the math to convert it to 12V, by doubling the voltage and dividing the current by 2, which results in the same wattage. These readings are taken right from the power supply, not verified with clamp meter + DMM.
At around 23C ambient temperature, with the MCPCB attached to a fan cooled heatsink:
.1A = 10.5V
.2A = 10.68V
.3A = 10.82V
.4A = 10.94V
.5A = 11.04V
1A = 11.5V
1.5A = 11.9V
1.69A = 12V
2A = 12.26V
2.5A = 12.6V
Then I turned the fan off and let the heatsink heat up to where it was too hot to touch comfortably.
This is what I get at 12V with a hot heatsink:
2A = 12.02V 24.14W
The increase in current draw at 12V from cold to hot was about 18%, ~1.69A when cold and 2A when hot.
The decrease from the hot value to the cold value is found with a multiplier of .85.
Based on my test, the LED should be drawing about 1.7A at 12V when cold, and 2A at 12V when hot. This is less current draw than the datasheet would suggest, but my tests are not super accurate either. I need to build a cable for my DMM so I can splice onto MCPCBs to verify the voltage at that point.
Based on the datasheet chart @ 85C, when cold the led should be drawing ~2.0A which is .85 of the datasheet value, as per my test result.
You should be drawing more current. By what method are you measuring current draw?
thank you very much for measuring everything!!!
My measurement was via a multimeter
My multimeter just recently measured 18w in other headlight which is exactly as listed in its specifications
Maybe there is some voltage drop over the long lengths of wires.
If your LED is actually being provided something a bit less than 11.5V, .88A draw seems realistic if the LED is also cold.
How long are the wires from your power supply, and what gauge are they?
Knowing that, a calculator can be used to get an idea of possible voltage drop.
For example:
But you can’t really just use a calculator to get an accurate picture of the situation. Ideally a voltage measurement should be taken at the MCPCB with a DMM. In order not to blind yourself trying to take this measurement, it is best to solder wires to the MCPCB electrical connector pads, and connect your DMM to these wires. Doing this, you’ll see how much voltage your LED is actually being provided.
no more than 1.5 meter long and about 18-20awg, I have 22 wires and they are definitely thinner (compared the copper, not isolation) !
I’ll measure at the MCPCB as well, but it’s very unlikely to be different, the wires are thick enough
Generally current measurements are advised to be taken with a clamp meter.
Your multimeter introduces another variable into the system. It adds more wire length and several unsoldered electrical connections, in addition to other impedance.
I took another measurement with a short 18awg wire soldered to the mcpcb, about the same result… not surprising
Yes, the multimeter (internally) would be using a very low value resistor to measure the current flow (using the voltage drop across it to calculate the current.) With flashlights, the addition of that minute resistance in circuit reduces the current flow through the LED. A clamp meter is the best way to do it.
Wire your power supply to the LED, soldering the wires in place, just like it is in your final setup.
Then you want to measure the voltage at the MCPCB via DMM. A way to do this without blinding yourself, and when there is poor access to the LED, is to solder some additional wires as probes onto the MCPCB contact pads. They can be thin gauge wires, it doesn’t matter, because they aren’t passing current.
Doing that, you will know how much voltage is being provided to your LED. From this you can estimate the current draw.
Using a clamp meter, you could measure how much current is actually being provided to the LED.
I tried a different power supply, this time with a voltage/current meter, wired to the same 12VDC converter
The power supply shows 10.9W~ output, at the led there’s still 10.5w~
95% conversion efficiency, exactly as advertised in the converter specs…
I think that it’s enough data to say the measurement is pretty accurate
Two different independent meters show significantly less wattage than specified datasheet!
That should be enough
