Is there any chart or graph available that shows the temperature behind the XPG emitter at various currents from 100ma to 1000ma. I just want to see the temperature difference of the XP-G between 400ma and 800ma. I am planning on using either 9 at 800ma or 12 at 400ma.
Heat vs Current on XP-G's
- Thread starter videoman
- Start date
Help Support Candle Power Flashlight Forum
gcbryan
Flashlight Enthusiast
Isn't temperature going to be determined by factors apart from the emitter?
I'm thinking it wouldn't be possible to have such a chart unless a lot of other variables were set as well.
There are charts that show lumen vs current draw and you see if there is a point at which they become less efficient which would mean more current is being given off as heat rather than light. That's an indirect indicator of what you are looking for.
I'm thinking it wouldn't be possible to have such a chart unless a lot of other variables were set as well.
There are charts that show lumen vs current draw and you see if there is a point at which they become less efficient which would mean more current is being given off as heat rather than light. That's an indirect indicator of what you are looking for.
Last edited:
I agree, but taking an emitter and placing it on a small heat sink and waiting it to stabilize for a while at 400ma and then raising the current to 800ma to note the percentage of rise at the 800ma mark should give an idea. I just don't have a temperature meter on hand to do this test. I did notice however that 9 of them at 800ma are really too hot to the touch ( the enclosure) and the driver goes into thermal shutdown after 15 minutes. I figure that using 12 at 400ma will reduce the lumens output by around 25% but the heat should reduce enough to avoid the shutdown.The light at 2500 lumens reduced to 1900 is not a big issue but the heat reduction is.
TorchBoy
Flashlight Enthusiast
You could buy an inexpensive heat-measuring multimeter and put a fan on your heatsink for less than the cost of three XP-Gs. Having an extra multimeter is pretty handy in itself.
I was thinking of putting a small dc fan if the heat issue cannot be resolved, but I would rather lose 25% of the output to avoid that, It really is for a video camera light mounted pretty close to the microphone. I think that reducing the led current from 800 to 400ma will do the trick. It really is a balancing act. I will get a thermometer soon. Thanks
vaska
Newly Enlightened
You can have a look at the XP-G light vs heat power ratio here:
http://www.candlepowerforums.com/vb/showpost.php?p=3117321&postcount=366
One can make calculations using data from here.
According to the table there, power cosumtion per one die is 1.22W at 400mA and 2,56W at 800mA with 163.97 and 295.3 lumens output correspondingly. So 8 LEDs at 800mA emit 2362.4lm from 20.48W and 12 LEDs at 400mA produce 1967.64lm out of 14.64W. Heat dissipation will make 13.6W for 8x800mA and 9W for 12x400mA.
If you raise 12x400 current up to 12x500mA you'll get 2400lm producing only 9.6W of heat.
Last edited:
TorchBoy
Flashlight Enthusiast
It would have to be an extremely quiet one, and I wonder if RF noise would be a problem. Agreed that it's better to go with passive cooling.
With a thermometer you'll be able to see just how hard you can push them for the heatsinking you've got. I don't think it needs to be top of the line; I saw one at DX for <$6.
That's excellent. If I run 12 at 500ma to 2400 lumens, only thing is that my driver measured 2500ma going out. It was spec'd at 3000ma. That means that I can only have 5 strings of series at 500ma (all paralelled) for a total of 10 leds. Very hard to find a real 3000ma output driver that can be dimmed (one mode) by an external pot. I am using a shark buck 3A which is built like a tank, as it is the only one I know of that is less than 25mm diameter. The Taskled units are a little too big to fit in my enclosure.The voltage input range must accomodate 7.2 - 16V DC. I think my lamp will fall into the 2000 lumen range considering the heat issue and the available drive current from the driver. Not too bad. Perhaps the S2 bins will up that to closer to 2500 when they become available.
vaska
Newly Enlightened
MaxFlex quite answers your purpouses: boost driver with 3 A maximum output current.
ifor powell
Enlightened
3A max input not output... Max Output is somthing like 1350ma on the latest version up from older versions 1200ma....
vaska
Newly Enlightened
It uses LT3477 with 3,5 A maximum switching current, and output current depends on input/output tension ratio. As a manufacturer I've run a lot of destructive experiments to be sure what this chip is capable of
Irrespective of the foregoing MaxFlex's output voltage permits just two LED strings with only 1 A total output current.
I really must have a buck driver. Any one of them leds in series goes out on me and the whole string goes along with it. Paralell string goes, and the rest will take up the added amperage and still go on shinin'. I can't beleive that there aren't many true 3A or over buck drivers out there that are single mode with dimming capability. Reliability is a key factor in my design, driver must not be larger than 25mm in width/diameter.I think I'll be OK with 10 leds in 5 parallel strings of 2. Thanks
R33E8
Newly Enlightened
As far as I know the amount of heat produce is related to how much energy is being given.. So If 12 watts of energy is being pushed in and the LED is 20% efficient, you will get 9.6 watts of heat.. But of course, as the current rises the LED becomes less efficient and outputs more heat per unit of light.. I think I would go with the 12 leds at 400mA since there will be less total power inputed..
