X-Ray study of LED reliability

Gunner12

Flashaholic
Joined
Dec 18, 2006
Messages
10,063
Location
Bay Area, CA
Came across this video about LED failures from Creative Electron:
https://youtu.be/UoxexFBmr0E?t=3m48s

Didn't know about solder voids, now I know that solder can sometimes look like a sponge.
So, if your build isn't getting as bright as you expect, or is dying early, blame the voids!
 

DIWdiver

Flashlight Enthusiast
Joined
Jan 27, 2010
Messages
2,725
Location
Connecticut, USA
Wow, that video is all over the place. First, it's 50% voiding reduced to 35%, then it's 40% reduced to 8%. I don't see anything in the "before" pictures that approaches 40% voiding, or anything in the "after" that approaches 8%. The voids shown in the X-rays are nothing like the voids shown in the microscopic cross section (macroscopic bubbles vs. microscopic "sponginess", which have completely different causes). First he's talking about die attach (which is within the LED component), then he's talking about attaching the component to a PCB. It's like he's blaming both the LED component manufacturer and the PCB assembler, but talking interchangeably about them, without making any distinction about what he's doing. But even that doesn't make sense because the LED he shows X-rays of is not the same LED he shows on the PCB.

To me, this sounds like a job well done being oversold to an audience whose technical knowledge is being severely dissed (and deservedly so if they let him get away with it).
 

LedTed

Enlightened
Joined
Mar 7, 2010
Messages
740
Location
Britannia
That video makes me miss the real time x-ray machine I used for discrete electronic component troubleshooting.
 

lucca brassi

Enlightened
Joined
Feb 1, 2008
Messages
889
Location
US
Last edited:

Steve K

Flashlight Enthusiast
Joined
Jun 10, 2002
Messages
2,786
Location
Peoria, IL
C-T scan x-rays are quite handy for checking out a piece of electronics that you can't disassemble. I've got a few images left from a scan on a sensor 7 or 8 years ago.....


In the image below, you can see a few bubbles in the solder fillets ....
3894589570_426fbee2f3_z_d.jpg



you can see a bit of this in this image too....
3893799527_3e2d3eb210_z_d.jpg


It is also handy for looking for broken capacitors (especially in a potted device), cracked traces, etc.
 

SemiMan

Banned
Joined
Jan 13, 2005
Messages
3,899
Remind me to find out who that contract manufacturer is, and NEVER use them. They obviously have severe issues with their soldering process, whether that is old paste, bad paste machine, way too fast heat rise in the reflow oven (bubbles), etc.
 

more_vampires

Flashlight Enthusiast
Joined
Nov 20, 2014
Messages
3,475

Steve K

Flashlight Enthusiast
Joined
Jun 10, 2002
Messages
2,786
Location
Peoria, IL
Remind me to find out who that contract manufacturer is, and NEVER use them. They obviously have severe issues with their soldering process, whether that is old paste, bad paste machine, way too fast heat rise in the reflow oven (bubbles), etc.

well, that was the least of the problems that we had with this supplier. The biggest problem was with potting... it wasn't adhering to the plastic housing and had cracks in the potting. Sooo many problems... but we've recently gotten a competing design and should be rid of this device soon <sigh>.
 

SemiMan

Banned
Joined
Jan 13, 2005
Messages
3,899
That seems strange but you need compatible plastics and need to watch for mould release compound on the plastic.

Cracking? Very strange. Sounds like way too hot of a cure.
 

Gunner12

Flashaholic
Joined
Dec 18, 2006
Messages
10,063
Location
Bay Area, CA
This was a area of reliability I haven't though about before, and thought would be nice to share. Glad to see that many of you have more in depth information and experience with this!

http://www.cree.com/search-results?&lcid=9&q=solder joint reliability

That Cree Solder Joint Reliability Study seemed like an interesting read, with xray pics.

Sure was, Osram has also done a study as well, and I'm sure many other companies as well. http://motiv8forums.com/uploads/3/2/9/8/3298228/osram_workshop_led_rf_sept2014.pdf
 
Top