max current thru a white nichia with out death?

snakebite

Flashlight Enthusiast
Joined
Mar 17, 2001
Messages
2,721
Location
dayton oh
i am running a 5600 with full leads at 225 ma with a penny soldered to the neg lead near the case.would try higher but the limit on the hp721a is 225.just gets a hint of the as craig calls it the pissed off blue color.gonna run it till it pukes in the name of science.whats the highest anyone succesfully ran one for more than a few minutes?started at 1:00 pm here.so 2 hours so far.
 

TrevorNasko

Flashlight Enthusiast
Joined
Jul 23, 2001
Messages
1,500
Location
Atlanta, GA>> The Flashlight that was broken shall
Well I've maxed out alot of leds and have not been killed yet
wink.gif
grin.gif
.
Seriously, I think an led redlines at 4.5v max using normal batteries but just dont go above that. really the most fun can be had by sticking it into a wall socket:D
 

The_LED_Museum

*Retired*
Joined
Aug 12, 2000
Messages
19,414
Location
Federal Way WA. USA
Without the penny soldered to the cathode, the LED should last anywhere between 3 seconds and 3-4 minutes with any current much over 120-150mA. Nichia LEDs tested to destruction have failed in this time frame with similar current levels. The penny soldered to the cathode is the only reason your LED hasn't yet popped.

It won't last much above what you're driving it, penny or not, because you'll start having problems with wirebond breakage, wirebond heating, melting or charring of the epoxy encapsulant in the vicinity of the wirebonds, and chemical changes to the LED die & phosphor.

If the bond wires and/or die get hot enough, carbon atoms present in the encapsulant material will dissociate to free carbon, and could eventually cause a short circuit between the anode and cathode just above the surface of the die.

I don't know how hot the phosphor can get before it goes bad; Nichia has not published that info. But like any complex molecule, there is a definite upper limit on its temperature before dissociation or other chemical changes begin to take place.

Cratering of the ball bonds also begins to occur at these seriously high temperatures, if the bond wires themselves don't melt off or carbonize the surrounding epoxy first. In either case, the LED will become intermittent or simply go out and the current in the circuit will drop to 0 or nearly 0. The wirebond may physically seperate from the surface of the die, usually taking a chunk of die with it. In the case of top bonded sapphire-substrated lamps, only the P- or N- layer will come up with the bond wire (depending on which wire pulls up); as the sapphire itself is so hard it would not break or chip off with a near-zero-force seperation like that. But the effect is the same: the LED flickers or goes out entirely.

The encapsulant itself will probably begin to discolor & darken even before it reaches charring temperatures. It could also crack or fracture. If it does this, you'll hear it pop or see a crack going partway or all the way around the LED about 1/3 up from the base. The LED will usually fail at this point, as the wirebonds are pulled off the chip by the seperation of the LED halves. Only on rare occasions does the epoxy fracture above the tops of the bond wire loops, sparing them.

The LED chip itself can get pretty toasty, but once it gets above a certain temperature, atomic dissociation will begin to enter the picture; as the surface temperature begins to approach those encountered during parts of the MOCVD process originally used to make the thing. Gases released from outgassing epoxy, phosphor slurry, and possibly even the material bonding the sapphire substrate to the cathode anvil could then react or chemically bond with the hot gallium nitride and change its chemical & electrical properties. I think the wirebonds would come off before this point though.

Permanent shifting of the 460nm spectral peak out of the phosphor's optimum response wavelength and significantly reduced output at this peak would be two expected consequences of this. The LED would become dim, possibly unusably so.

Cracking or crazing of the substrate could be possible in extreme conditions; if this happens, the circuit is broken and the LED again becomes intermittent or goes out.
 

Orion

Flashlight Enthusiast
Joined
Jun 27, 2002
Messages
1,613
Location
Missouri
I have put an experimental light together running 4 white Nichia with no resistors with 3AA batteries. It is quite bright! I was thinking I should put in 6 ohms of resistance, but haven't done so yet. I haven't run it for a very long time yet, but when I turn it on, the light seems white rather than the blue-ish color that (if I understand correctly) occurs when a white Nichia is overdriven.
 

snakebite

Flashlight Enthusiast
Joined
Mar 17, 2001
Messages
2,721
Location
dayton oh
thanks craig.
update
after 24 hours the die is a dark square in the phosphor glob.
still working but a bit down on brightness.
i am just curious to see how long i could torture it.the penny gets hot enough that i cannot hold it more tha 5 seconds.and i pull hot vacuum tubes all the time.
most folks that touched the penny cursed and let go fast.
 

snakebite

Flashlight Enthusiast
Joined
Mar 17, 2001
Messages
2,721
Location
dayton oh
yep
thats how i got the nick.
i breed boa constrictors and a not so tame one latched onto my hand.
some friends that were there called me snakebite ever since.
 

The_LED_Museum

*Retired*
Joined
Aug 12, 2000
Messages
19,414
Location
Federal Way WA. USA
Originally posted by snakebite:
thanks craig.
update
after 24 hours the die is a dark square in the phosphor glob.
still working but a bit down on brightness.
<font size="2" face="Verdana, Arial">Looks like you've probably burnt the phosphor and/or thermally discolored the epoxy encapsulant directly above the die.
 

Lux Luthor

Flashlight Enthusiast
Joined
Nov 10, 2000
Messages
1,944
Location
Connecticut
Craig,

That doomsday description you gave above is rather frightening!
shocked.gif
If you changed a few of the words around, it could probably serve as a description of a supernova explosion, or perhaps the end of life itself.
frown.gif


I don't know about anyone else, but I'm sleeping with the lights on tonight for sure.
grin.gif
 

snakebite

Flashlight Enthusiast
Joined
Mar 17, 2001
Messages
2,721
Location
dayton oh
well it died overnight.
still pulling 225 ma but turned into a dark emitting diode.so even at this short of lifespan one could properly heatsink and abuse like this in a flashlight.
a moment of silence before i bury it at sea(craig knows what i mean)
smile.gif
 

Hemingray

Enlightened
Joined
Jul 2, 2002
Messages
380
Location
New Hampshire
I've run lots of Nichia NSPW500BSs, right off 4.5
V (3 AA cells) without flaming any out, though they do get a tad warm if there is no heatsinking, even a little from a PC board's copper. If I want to be a bit more gentle with them, I just use a 1N4004 or similar silicon rectifier in series and it gives a 0.6 to 0.7 V drop so the LED(s) see 3.8V. The slight loss in over-driven brilliance isn't all that noticeable, and the LEDs will be that much happier and long-lived for it. My latest experiment has 12 of the newest Nichias in parallel, a 1N4004 in series,
being fed off 3 AA's. One SPST switch for on/off, another across the diode for bright/dim. Haven't yet had any angry blue LEDs on this lash-up, yet.

/ed brown in tropical-hot NH
cool.gif
grin.gif
 

snakebite

Flashlight Enthusiast
Joined
Mar 17, 2001
Messages
2,721
Location
dayton oh
Originally posted by moraino:
Snakebite,

What's the brightness compares to says 40mA or...
<font size="2" face="Verdana, Arial">not much to be gained over about 80 ma
 

shankus

Flashlight Enthusiast
Joined
Feb 16, 2003
Messages
1,472
Location
Mojave, CA
[ QUOTE ]
Hemingray said:
If I want to be a bit more gentle with them, I just use a 1N4004 or similar silicon rectifier in series and it gives a 0.6 to 0.7 V drop so the LED(s) see 3.8V. The slight loss in over-driven brilliance isn't all that noticeable...One SPST switch for on/off, another across the diode for bright/dim.

[/ QUOTE ]

Is it noticeable enough to actually use as a dim setting?
 

Badbeams3

Flashlight Enthusiast
Joined
Sep 28, 2000
Messages
4,389
[ QUOTE ]
snakebite said:
stay tuned as we are going to destructivly test some luxeon emitters!

[/ QUOTE ]

Oh man...this is sick /ubbthreads/images/graemlins/popcorn.gif How about putting the little feller in a glass of ice water...hook it up to a car batt...just an idea. /ubbthreads/images/graemlins/icon15.gif
 
Top