HELP! Cree MC-E 5A loosing output...

LEDBE
Ogg Vorbis

Ogg Vorbis

Newly Enlightened
Joined
Jan 25, 2007
Messages
147
Location
West Yorkshire, England
Hi, i've just finished a mag 2C mod using a KD super driver, and a cree MC-E in a 5A tint and when i fire it up it starts off with the normal neutral tint but soon goes blue-ish with fading output but then becomes stable...

This only happens on full output... and EVERYTHING is well heatsinked

So can anyone explain what is happening? Will it be ok? Should i only run it at full for a short while?

Thanks

Dan
 
Last edited:
It seems like the LED isn't fully seated on the heat sink.. Perhaps you put too much thermal grease? If that's the reason, the thermal grease should be as thin as possible between the led and heat sink...
 
It will not be OK. Improve the thermal path, please! :wave:

Microinches of thermal paste or epoxy may be your enemy. Get as close to metal-metal as you can.
 
Well, one thing to remember: Thermal grease doesnt conduct heat well. IN fact, it sucks as a heat conductor compared to most metals.

Its just better than air...

So you really want _just a tiny little bit_ of that stuff between your led and your metal base. Just so that where you dont get a metal-on-metal fit, the gap is filled with it instead of air.

If you use thermal epoxy, let it harden while you have at least a couple of kilos pressure on the led (if its a star).
 
What was the drive current? You can cook it even well heatsinked if you pumped too much current through it. The 5A is rated for less than a cool white bin.
 
It was 2.8A... but it wasn't destoryed by electrical means... more mechanical... to be more precise... an impact... that's all i'm saying :aaa:

Lets just say i've learnt to get it right the first time!
 
Marduke said:
What was the drive current? You can cook it even well heatsinked if you pumped too much current through it. The 5A is rated for less than a cool white bin.
Click to expand...

Are you saying Cree specs neutral white emitters for less current rating than the cool white emitters?
 
They do rate the neutral white XR-E lower than the cool whites with a maximum of 700 mA, but I believe the neutral white MC-E is rated 700 mA per die just like the cool whites. The data sheet contains no information that would lead me to believe otherwise, and it makes sense that each die should have the same maximum drive current since the MC-E package essentially contains 4 of the dies used to make the XR-E (EZ1000 die).
 
For the sake of arguement, I've found neutral white and warm white Crees (XR-Es)*far* more sensitive to high current loads (+700mA) than cool whites. I'm assuming the trend carries over onto the MC-E.
 
EngrPaul said:
Are you saying Cree specs neutral white emitters for less current rating than the cool white emitters?
Click to expand...

Yes. In the XR-E package, it's 700mA vs 1000mA max current.

MC-E's are already speced at 700mA per die, but that is more for having four dice in such close proximity. If you carry the same 30% debit for warm white, perhaps it should be 490mA?

In any event, neutral and warm white LED's are quite sensitive to high current, which equal high heat. Heat quickly degrades the warm phosphor.
 
Leave out the wild guessing the unfounded speculation.

The failure description the OP gave fits the symptoms of a plain overheating perfectly, so there is no need to.
 
IMSabbel said:
Leave out the wild guessing the unfounded speculation.

The failure description the OP gave fits the symptoms of a plain overheating perfectly, so there is no need to.
Click to expand...

1) They are not wild or unfounded.

2) Yes, overheating, but why? It may not be considered "overheating" for a cool white LED, but it may be overheating for a warm white LED, hence the above discussion.
 
Overheating was occuring because the OP (me :grin2:) put way too much arctic alumina on!

No idea about 'neutral' white vs. 'cool' white and their respective drive currents, but i think there would be much more information and warnings out there if there was such an issue...

I've now got a Mag C heat sink for a SSC P7 on the way, along with a DSWOI P7! Just a bit upset about the $30 MC-E i destroyed :mecry:

Dan
 
Well, I'm still interested... does anybody know the max safe drive level for a 5A J-bin MC-E with good heatsinking?
 
concept0 said:
Well, I'm still interested... does anybody know the max safe drive level for a 5A J-bin MC-E with good heatsinking?
Click to expand...
I have done a ton of experimenting with warm and neutral white LEDs at high drive levels. I did a test the other day with a 7a or 7b MC-E and was able to get stable output from it at about 4A with it attached to a small heatsink. Also I have not seen an appreciable difference in the ability to handle over driving of warm white XR-Es. Just make sure the heatsinking is very good.
 
well, a few might have read my opinion to arctic when it comes to heatsinking, but there is another thing I am wondering:

why use a thermal glue for connection Star <-> sink?
here it makes way more sense to use screws
(and thermal paste, of course)

placing, fine tuning/change placing, changing emitter with another one ...
everything made better


... would also ease the use of arctic glue, as one can easily put force on the star while the glue cures
 
You must log in or register to reply here.

Similar threads

T
COB LED project help
Replies
0
Views
1K
tobyyost
T
D
Long low-level LED light output
Replies
4
Views
1K
Dave_H
D
A
High Output Cree Street Lamp LED Spec Help
Replies
16
Views
3K
Dave_H
D
P
Can you help me?
Replies
5
Views
1K
Dave_H
D
R
Number of colored Cree XP-Es to match a 12V Cree MKR or another solution?
Replies
3
Views
1K
LEDphile
L

Latest posts

Top