Previous in Forum: Neutral Grounding Resistor   Next in Forum: LED Driver
Close
Close
Close
3 comments
Rate Comments: Nested
Power-User

Join Date: Jun 2008
Posts: 121
Good Answers: 1

High Power LEDs , Junction Temperature

11/29/2011 3:48 AM

Is it practical to do junction temperature testing on high power leds during mass production of assembled units. Or are there any mothods that can be used to measure the Tj indirectly? Thanks

__________________
"Shallow men believe in luck. Strong men believe in cause and effect." -- Ralph Waldo Emerson
Register to Reply
Interested in this topic? By joining CR4 you can "subscribe" to
this discussion and receive notification when new comments are added.
Commentator

Join Date: May 2011
Posts: 84
Good Answers: 1
#1

Re: High power leds , junction temperature..

11/29/2011 5:24 AM

may be this is something useful to you.... temperature gun (laser type)

Register to Reply
Power-User

Join Date: May 2010
Location: Cambridge, MA
Posts: 390
Good Answers: 82
#2

Re: High Power LEDs , Junction Temperature

11/30/2011 12:56 PM

First, I'd say that unlike the junction temperature of power semiconductors, which are operated all the way to 150°C, LED junction temps are normally restricted to much lower levels, because the light output suffers at high temps.

I'd say you can measure it the same way we do with power semiconductors, which is to use the diode's forward voltage drop, at a modest current like 1mA or maybe pulsed at 10mA.

First characterize the parameter's temperature coefficient in an oven. Then operate the LED and let the temperature stabilize. Rapidly remove the operating current and switch to the much lower, calibrated, measurement current. Use a sampling circuit or a very fast voltmeter to grab and measure the forward voltage before the junction has a chance to significantly cool down. A short time like 10us should be good enough. In the old days they used up to 300us, but that's too long.

Motorola published a circuit in the '80s, to accomplish this task, see figures 11 and 12 in application note AN1083. I used Google to find a copy (link). The circuit is rather complex, but a simplified version should serve just fine.

With additional characterization, it's sometimes possible to make real-time dynamic temperature measurements, but that's another story.

__________________
Thanks, Win
Register to Reply
Power-User

Join Date: Jun 2008
Posts: 121
Good Answers: 1
#3
In reply to #2

Re: High Power LEDs , Junction Temperature

12/07/2011 4:39 AM

Sorry I couldnt get back on line earlier, my laptop went massive surgery at one of the hospitals here.

Thank u for ur article on . It was informatic and useful , removed lots of ingnorance i had in testing junction temp.

I found them useful. thanks again.

moomoo

__________________
"Shallow men believe in luck. Strong men believe in cause and effect." -- Ralph Waldo Emerson
Register to Reply
Register to Reply 3 comments
Copy to Clipboard

Users who posted comments:

moomoo (1); muditmah (1); Winfield Hill (1)

Previous in Forum: Neutral Grounding Resistor   Next in Forum: LED Driver

Advertisement